Designing-with-plastics-and-composites-a-handbook.pdf

~ PRCAlECT TEAM FEASIBILITY

-

H

STUDY

Dimensions. structural loads. gov'Vlndustry standards. service environments, etc__ 1

Identify specific functions

Target quantity. cost and production schedule

Apply available experience

I Manual approach

l-.{ COnceptual product layout

Apply design creativity

Computer approach

I

r1 Formulate plan'

R

PRELIMINARY DESIGN ANALYSIS

r-y

I

Dimensions & tolerances

Geometric drawings

or graphic analYSis

Engineering analysiS

Solid

4~

Image manipulation

r-

I

Surface

c-j

Structural Integrity

Environment

---I Physical integrity

Mechanical

simulatkm

I

J---@

Dynamic

Finite element modeling

J

Minimum weight/cost

n

Ae_c.etc.1

--1

rl OPTIMIZE DESIGN

r

Mok:Jfdie destgn

Material selection

I

Fabrication analysis

4

J

Shrinkage analysis Cycle timeJ

I

detai.led/individual orthotropic drawings

Set up safety factors to meet product functions

I Nonstandard parts I Standard parts

~ Provtde

-1

I

selection

---i ~

Material

I

----i ProcesSlequipmentltools _I -1

Produce service per10rmance

I

--I Plant personnel capability I

iii

rl H

~ 1l

Release tooling (mOld/die) for manufacture Set up processing specifications

I

I

rl Equipment I H Material handling I

'iii

15

I j

MANUFACTURING~

H

Automationtrobotics

Plant layout

H

n

ANALYSIS

Y

'12l

§

I

Safety procedures

W

Manufacturing processing specifications

I

I

Troubleshooting gUide Testing/quality control

Plant personnel capability evaluation

f rl

OPTIMIZE &

DOCUMENT

PRODUCTION

~

I

I I I

Ensure product quality Document good manufacturing practice (GMP) procedure

H

I

I

I

Preventive maintenance

Machine operations

H

Update plant personnel training

I

H DevelOp production cost models

I

Finalize testing & quality conlrol

I

p od hed I r uct output sc ue

~ ~:~~:ii~: :~=~~: Inventory schedule

Y Documentation for management to ensure meeting delivery schedules and profitability I J

I

PRODUCT RELEASE

Ensure meeting all product functions Set up value analysis (VA)

I

II

I

Cost analysis

rl Matt flow analysis H Tool thermal analysis I - Tool part I

-1

Databases andlor available information

Process selection

After start of production. analyze ccmplate design (again) to change design/production/safety factors in order to reduce costs

I

I

DESIGNING WITH PLASTI CS AN D COMPOSITES: A HANDBOOK

Donald V. Rosato, Ph. D. David P. Di Mattia Industrial and Graphic Designer

Dominick V. Rosato P.E. Rhode Island School of Design

~

SPRINGER SCiENCE+BUSINESS MEDIA, LLC

Copyright © 1991 by Springer Science+Business Media New York Originally published by Van Nostrand Reinhold in 1991 Softcover reprint of the hardcover 1st edition 1991 Library of Congress Catalog Number 90-46378 ISBN 978-1-4615-9725-4 ISBN 978-1-4615-9723-0 (eBook) DOl 10.1007/978-1-4615-9723-0

All rights reserved. No part of this work covered by the copyright hereon may be reproduced or used in any form by any means-graphic, electronic, or mechanical, including photocopying, recording, taping, or information storage and retrieval systems-without written permission of the publisher.

16

15

14

13

12

11

10

9

8

7

6

5

4

3

2

Library of Congress Cataloging-in-Publication Data Rosato, Donald V. Designing with plastics and composites: a handbook by Donald V. Rosato, David P. Di Mattia, and Dominick V. Rosato. p. cm. Includes bibliographical references and index. 1. Plastics. 2. Engineering design. II. Rosato, Dominick V. III. Title. TP1122.R67 1991 668.4'9--dc20

I. Di Mattia, David P.

90-46378 CIP

Contents

Preface / ix

1. FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES / 1 Design Shape / 49 Success by Design / 51 Computers in Design / 52 Design Procedure / 54 Interrelating Product-Resin-Process Performances / 55

2. THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS / 61 Plastic Structures and Morphology / 66 Thermal Properties of Plastics / 83 Thermal Conductivity and Thermal Insulation / 87 Heat Capacity / 88 Thermal Diffusivity / 88 The Coefficient of Linear Thermal Expansion / 89 Deflection Temperature Under Load / 94 Decomposition Temperature / 95 Mechanical Properties / 96 Physical Properties / 98 Rheology and Deformation / 107 Interrelating Properties, Plastics,and Processing / 116 Orientation / 11 8 Shrinkage / 123

3. PLASTICS: DESIGN CRITERIA / 125 Mechanical Properties / 133 Short-Term Behavior / 135 Long-Term Behavior / 153 Short-Duration Rapid and Impact Loads / 201 Electrical Properties / 223 Friction, Wear, and Hardness Properties / 239 iii

iv CONTENTS

4. ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS / 253 Temperature Introduction / 253 Chemical Resistance / 264 Weather Resistance / 272 Sterilization-Irradiation / 274 Permeability and Barrier Resistance 276 Biological and Microbial Degradation / 281 Flammability / 282 The Ocean Environment / 288 The Space Environment / 297

5. STRUCTURAL DESIGN ANALYSIS / 303 Load-Bearing Products / 303 Loads / 305 Support Conditions / 305 Simplifications and Assumptions / 308 Multiaxial Stresses and Mohr's Circle / 308 Safety Factors / 309 Beam Bending Stresses / 312 Beam Bending and Spring Stresses / 321 Shear Stress and Torsion / 323 Shear Stress and Direct Shear / 325 Pressure Vessels / 325 Externally Loaded RP Pipe / 326 Molded-In Inserts / 340 Press Fits / 342 Snap Fits / 345 Hinges / 349 Thread Strength / 355 Pipe Threads / 358 Gears / 359 Gaskets and Seals / 360 Grommets and Noise / 360 Impact Loads / 361 Thermal Stresses / 362 Structural Foams / 365 Structural Sandwiches / 368 Energy and Motion Control / 371 Failure Analysis / 387 Dimensional Tolerances / 387 Plastics / 388 Processing and Tolerances / 391 Product Specification / 398 Combining Variables / 399 Finite Element Analysis / 399 Cantilevered Snap Fits / 402 Laws and Regulations / 402

CONTENTS v

6. THE PROPERTIES OF PLASTICS / 405 Trade Names / 405 Acrylonitrile-Butadiene-Styrene (ABS) / 406 Acetal / 414 Acrylics / 415 Alkyds / 416 Aminos / 416 Cellulosics / 417 Chlorinated Polyethers / 417 Chlorinated Polyethylene / 418 Cross-Linked Polyethylene (XLPE) / 418 Diallyl Phthalate / 418 Epoxies / 419 Ethylene-Vinyl Acetates 419 Fluoroplastics / 420 Furan / 421 lonomers / 421 Ketones / 422 Liquid Crystal Polymers / 422 Melamines / 423 Nylons (Polyamides) / 423 Parylenes / 425 Phenolics / 426 Phenoxy Resins / 427 Polyallomers / 428 Polyamides / 428 Polyamide-imide / 428 Polyarylates / 430 Polyarylethers / 430 Polyaryletherketone / 430 Polyarylsulfone / 431 Polybenzimidazole / 431 Polybutylenes / 432 Polybutylene Terephthalate / 432 Polycarbonates / 432 Polyesters / 433 Polyetherketone / 436 Polyetheretherketone / 438 Polyetherimide / 439 Polyethersulfone / 440 Polyethylene / 440 Polyethylene Terephthalate / 445 Polyimides / 445 Polymethyl Methacrylate / 445 Polymethylpentene / 446 Polyolefins / 446 Polyphenylene Ether / 446 Polyphenylene Sulfide / 447

vi CONTENTS

Polypropylene / 447 Polystyrene / 449 Polysulfones / 449 Polyurethane / 452 Polyvinyl Chloride / 452 Polyvinylidene Fluoride / 455 Silicones / 455 Urea Formaldehydes / 456 Elastomers / 458 Thermoset Elastomers / 467 Thermoplastic Elastomers / 472 Film and Sheeting / 473 Foams / 475 Transparent and Optical Plastics / 491 Reinforced Plastics and Composites / 493 Regrind and Recycling / 524 Guide for Plastics Identification / 525 Computerized Databases / 528 Selection Worksheets / 535 Selecting Materials / 535 Selecting Materials under Dynamic Loading / 540

7. THE PROCESSING OF PLASTICS / 589 Tolerances and Shrinkages / 596 Model Building / 596 Molds and Dies / 599 Drying Hygroscopic Plastics / 602 Heat History, Residence Time, and Recycling / 602 Process Control / 603 Troubleshooting / 603 Inspection / 606 Injection Molding / 610 Extrusion / 624 Basics of Flow / 630 In-Line Postforming / 641 Blow Molding / 646 Extrusion Blow Molding / 651 Forming / 665 Reinforced Plastics/Composites 670 Other Processes / 687 Selecting Processing / 695

8. AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS / 711 Material Handling / 713 Parts Handling / 713 Finishing and Decorating / 714 Joining and Assembling / 714 Machining / 727

CONTENTS vii

9. TESTING AND QUALITY CONTROL / 731 Basics versus Complex Tests / Specifications and Standards / Orientation and Weld Lines / Types of Tests / 735 Thermoanalytical Tests / 735 Nondestructive Testing / 746 Computer Testing / 753 Quality Auditing / 753 Reliability and Quality Control Failure Analysis / 754 Selecting Tests / 755 Quality and Control / 755

732 732 734

/ 754

10. COMPUTER-AIDED DESIGN / 757 Mold Design / 758 CAD/CAM Modeling / 763 Additional CAD/CAM Features Used in Plastic Part and Mold Design / 773 Process-Analysis Tools / 777 Design Databases / 782 Computer-Integrated Manufacturing / 785 Myths and Facts / 785 Capability and Training / 786 11. DESIGN FEATURES THAT INFLUENCE PERFORMANCE / 789 Basic Detractors and Constraints / 789 Injection Molding / 796 Extrusion / 844 Blow Molding / 847 Thermoforming / 854 Reinforced Plastics and Composites / 856 Rotational Molding / 865 Assembly Methods / 869 Mechanical Loading / 870 12. CONCLUSIONS / 877 Product Diversification / 879 Materials Diversification / 883 Equipment Improvements / 887 The Solid-Waste Problem and Product-Design Solutions / 889 Technical Cost Modeling / 898 Success by Design / 900 Design Considerations / 900 Challenge Requires Creativity / 914 The Future / 915

viii CONTENTS

Appendix A. General Information Sources / 919 Appendix B. Conversions / 921 Appendix C. Trade Names / 925 Appendix D. Computerized Software and Databases / 929 References / 937 Index / 967

Preface

For some time there has been a strong need in the plastic and related industries for a detailed, practical book on designing with plastics and composites (reinforced plastics). This one-source book meets this criterion by clearly explaining all aspects of designing with plastics, as can be seen from the Table of Contents and Index. It provides information on what is ahead as well as today's technology. It explains how to interrelate the process of meeting design performance requirements with that of selecting the proper plastic and manufacturing process to make a product at the lowest cost. This book has been prepared with an awareness that its usefulness will depend greatly upon its simplicity. The overall guiding premise has therefore been to provide all essential information. Each chapter is organized to best present a methodology for designing with plastics and composites. This book will prove useful to all types of industrial designers, whether in engineering or involved in products, molds, dies or equipment, and to people in new-product ventures, research and development, marketing, purchasing, and management who are involved with such different products as appliances, the building industry, autos, boats, electronics, furniture, medical, recreation, space vehicles, and others. In this handbook the basic essentials of the properties and processing behaviors of plastics are presented in a single source intended to be one the user will want to keep within easy reach. Once a product's purpose and service requirements have been established, its successful design and manufacture to meet zero-defects production requires knowledge of 1) the plastic materials from which it is to be made, their nature, and the ways in which processing may affect their properties; 2) the processing methods available for its manufacture; and 3) how to evaluate its properties and apply effective quality control. This reference handbook has been designed to be useful to those using plastics as well as those still contemplating their use. To this end the presentations are comprehensive yet simplified, so that the specialist in a specific field will obtain useful information. The cross-comparisons and interrelations of design facts and figures are extensive, to ensure ease in understanding the behavior of plastics and composites. Designing depends on being able to analyze many diverse, already existing products such as those reviewed in this handbook. One important reason for studying these designs is that this shows how many diverse topics cooperate synergistically to enhance designers' skills. Design is interdisciplinary. It calls for the ability to recognize situations in which certain techniques may be used and to develop problem-solving methods to fit specific design situations. Many different examples of problems are thus presented within this handbook, concerning many products. ix

x PREFACE

With plastics, to a greater extent than with other materials, the opportunity exists to optimize design by focusing on a material's composition, its structural orientation during processing, and other factors described throughout this handbook. Analyses are made of problems that can occur and how to eliminate them or how to take corrective action. This book is intended to provide practical guidelines to designers using plastics or composites. Throughout this handbook, examples that relate to basic strengths of materials are given so as to highlight their influence in different designs. The information to the designer includes the behavior of plastics under extreme performance conditions, relates these behaviors to design principles, and provides important information on design parameters as they interrelate with plastic materials, processing characteristics, and the performance of products. As materials to be fabricated, plastics provide practical, unlimited benefits to the design of products. Unfortunately, as with other materials, such as steel, wood, glass, aluminum, and titanium, no one plastic has all the best traits, so that sometimes selecting a material requires compromising. Successfully applying their advantages and understanding their limitations, as reviewed in this handbook, will allow designers to produce useful, profitable products. There is a wide variation in the types of properties among the fifteen thousand materials commercially available worldwide that are classified as plastics or composites. In general, however, most plastics can be processed into different shapes and sizes. If so required, they can have intricate shapes held to tight tolerances and be made by processes suitable for either limited or mass production. The costs of plastics range from relatively low to extremely expensive, enough to make a plastic appear to be too costly for a given product. However, studying the processing method could result in meeting low product-cost requirements. This handbook thus provides the designer with useful information on the different processing methods as they relate to meeting design and cost requirements. Plastics vitally concern almost everyone worldwide. They occupy an important part of the research, development, design, production, sales, and consumer efforts in diverse industries. As reviewed later, for over a century plastics have been used successfully, in such applications as for packaging, housewares, medicine, marine, aerospace, hydrospace, transportation, biological, appliance, building, and recreation. The significant improvements that have been made in plastic materials, processing, and applications thus far will no doubt be overshadowed by future improvements. Because their broad range of properties makes plastics unique, they are adaptable to different products and markets. With plastics, one can decide on practically any requirement and find for it a processable plastic, whereas other materials have comparatively narrow capabilities. It is nevertheless important to recognize that there are tremendous variations in the properties and performance of plastics. This handbook shows that there is a practical, easy approach to designing with plastics. One of the major aims of this book is to help develop the designer's ability to analyze problems, a most important skill. Although engineering mechanics is based on only a few basic understandable principles, these principles are needed to provide a means to solve many problems relating to present-day design and analysis. This book emphasizes both understanding and applying these principles, so that the designer will have a firm basis for utilizing the principles. It is essential to reemphasize the point made in the text that all data presented on plastic properties are to be used only as guides. Obtain the latest, most complete data from material suppliers and data banks from the various sources referenced throughout this handbook.

PREFACE xi

The infonnation presented herein may be covered by United States or foreign patents. No authorization to utilize these patents is either given or implied; they are discussed as infonnation only. Likewise, the use of general descriptive names, proprietary names, trade names, and commercial designations and the like in no way implies that they may be used freely. They are often legally protected by registered trademarks or some other fonnat even if they are not designated as such in this book. Finally, although the information presented is useful data that can be studied or analyzed that are believed to be true and accurate, neither the authors, contributors, nor publisher can accept any legal responsibility for errors, omissions, or similar factors. In preparing this handbook extensive use was made of the personal industrial and teaching experiences of the authors, going back to 1939, as well as worldwide infonnation from industry and trade associations on materials, equipment, and the like, published books, articles, reports, conferences, and so on, as is evident in the references given at the end of the book. In the preparation of this handbook the authors have been assisted and encouraged by many friends and international business associates. Special acknowledgment must be made to the many different authors cited, including many different material suppliers. All have, whether directly or indirectly, contributed to advancing the state of the art in designing with plastics.

Chapter 1

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES

There is a practical, easy approach to designing with plastics and composites (see Figs. 1-1 to 1-3) that is basically no different than designing with other materials: steel, aluminum, titanium, copper, brass, wood, concrete, and so forth. This book provides useful and necessary information on how to comprehend plastics' and composites' extreme range of properties, structural responses, product-performance characteristics, part shapes, manufacturing processes, and their influence on product performance, the simplifying of designs, as guides on selecting plastics and processes as well as on how to keep up-todate on important information and understand the econc:>mics of designing with plastics [1-200]. * Many different products can be designed using plastics and composites. They will take low to extremely high loads and operate in widely different environments, from highly corrosive ones to those involving electrical insulation. They challenge the designer with a combination of often unfamiliar and unique advantages, and limitations. By understanding the many different structures and properties as well as the design and fabrication capabilities, the designer can meet this challenge as demonstrated by the existence of the many different products made from plastics. They exist in all types of applications-underground, underwater, in the atmosphere, in outer space, in the office, and in the home. Although plastics and composites may appear to some observers to be new, because the industry has an unlimited capacity to produce new plastics to meet new performance and processing requirements, plastics and composites have been used in no-load to extremely high-load situations for over a century. The ever-evolving technology does not mean that plastics and composites will automatically replace other materials. Each material (plastics, metals, wood, aluminum, and so forth) will, basically, be used in favorable cost-to-performance situations. As of the early 1980s, more plastics were used worldwide on a volumetric basis than any other materials except wood and concrete. Before the end of this century there will be on a weight basis more plastics used than the others, except wood and concrete.

*All references are listed in the References section in the back of the book. 1

2 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

l Receive and review product I

I

\1

II

IReject approach I /l'

1

Develop design approach

1-

I

I Material}

Study approach

J

I Complete preliminary appraisal : J~

I

Make final evaluation

I

I I

Assign deSign priority

I

I

Produce product

I

I

J/ ,L,

Process tCost l-

Figure 1·1. To be effective, the evaluation of new product ideas should proceed according to a logical step-to-step process, as shown.

With plastics and composites, to a greater extent than with other materials, an opportunity exists to optimize design by focusing on a material's composition and orientation as well as its structural-member geometry. There are also important interrelationships among shape, material selection (including reinforced plastics, elastomers, foams, and so forth), the consolidation of parts, manufacturing selection, and other factors that provide low cost-to-performance products. For the many applications that require only minimal mechanical performance, shaping through processing techniques can help overcome limitations such as low stiffness with commodity (lower cost) plastics. And when extremely high performance is required, reinforced plastics (RP), composites, and other engineering plastics are available. In this book the term plastics also refers to composites. All processes fit into an overall scheme that requires the interaction and proper control of different operations. The Follow All Opportunities (FALLO) approach shown in Figure 1-2 can be used in any process by including the "blocks" that pertain to the fabricated product's requirements. (See Chapters 7 and 8 regarding basic processing and auxiliaryupstream and downstream-equipment.) The FALLO approach has been used by many processors to produce parts at the lowest cost. Computer programs featuring this type of layout are available (see Chapter 10). The FALLO approach makes one aware that many steps are involved in processing, all of which must be coordinated. The specific process (injection, extrusion, blowmolding, thermoforming and so forth) is an important part of the overall scheme and should not be problematic. The process depends on several interrelated factors, such as designing a part to meet performance and manufacturing requirements at the lowest cost, specifying the plastics, and specifying the manufacturing process. To do so basically requires designing a tool (mold, die, and so forth) around the part, putting the proper-performance fabricating process around the tool, then setting up the necessary auxiliary equipment to interface in the complete fabricating line, and, finally, setting up completely integrated

IN

~

-

._n_OPERAnON

ready lor delivery

MOLDED

'--GOOD--MAN--U,;-A-CTUII--NG-PRA-cn-ce-........

of potential "faults."

Set up practical/useful TROUBLESHOOTING GUIDE based on "causes & remeaes"

.

I

t

Reevaluate all parameters used from part design (use less plastic), use lower cost plastic with similar processing costs (or plastic with higher costprocesses faster-resuits in total lower cost), check hardware performance, & other parameters.

produce part 10 meet same requirements but produced at a lower cost. Use FALLO approach.

IMPORTANT STEP - is to

Il1In6dialely after part is in production - next step-

OPERATION

......---I~-

I

PART

Integrate all individual operations thai produce parts

Set up TEST1NG/QUAUTY CONTROLCharacterize properties: mechanical, physical, chemical, thermal, etc.

I

Figure 1.2. The example shown here for the complete blow-molding fabrication operation is also applicable to the other processes. This diagram shows the FALLO approach.

FALLO FOllow ALL Opportunities

requiremems

based on mar1<et

I

• •

~

r------I

PROJECT TEAM FEASIBILITY STUDY

Image manipulation

Dimensions & tolerances

Databases and/or available

Apply design creativity

Dimensions, structural loads. gov'tJindustry standards, service environments, etc.

\11

I PRODUCT L RELEASE

OPTIMIZE & DOCUMENT PRODUCTION

I

ANALYSIS

I

J

Machine operations

Inventory schedule

Ph· h d I ure aSlng se e u e

Sat up value analysis (VA)

Ensure meeting all product functions

I

II

I

Finalize testing & quality control

Product output schedule ~ Accounting schedule

I I

I

Ensure product quality

Testing/quality control

Troubleshooting guide

I

Preventive maintenance

I

I

After start of production. analyze complete design (again) to change design/production/safety factors in order to reduce costs

-i Documentation for management to ensure meeting delivery schedules and profitability I

-i Develop production cost models

Update plant personnel training

I

Automation/robotics

Material handling

-i Safety procedures I

--1

---1

-1 Equ;pmen~

Document good manufacturing practice (GMP) procedure

Plant personnel capability evaluation

Manufacturing processing specifications

1

-1

---1

r

MANUFACTURING~

I Plant layout

~

I

I

Figure 1-3. Flow diagram with feedback loops for product design using the FALLO approach (Follow ALL Opportunities). This flow diagram uses available CAD/CAElCAMICIM/CAT computer software programs. It provides the project team with the information necessary to aid in making final design decisions. All programs, particularly those with databases, should have the capability for continual updating and expansion.

i

i ,s

11

Set up processing specifications

---1

-1 Release tooling (mold/die) for manufacture

6 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

controls to meet the goal of zero defects. The final step in the FALLO process is that of purchasing equipment as well as materials, then warehousing the material. This interrelationship is different from that of most other materials, where the designer is usually limited to using specific prefabricated forms that are bonded, welded, bent, and so on. Designing has never been easy in any material, particularly plastics, because there are so many. Plastics provide more types than all other materials put together-with about 45 families of plastics, many variations are available (see Figs. 1-4 and 1-5). Of the more than fifteen thousand different plastics, only a few hundred are used in large quantities. Unfortunately, many designers view plastics as a single material, because they are not aware of all the types available. Plastics are a family of materials each with its own special advantages. The major consideration for a designer is to analyze what is required as regards performance and develop a logical selection procedure from what is available (see Chapter 6). The range of properties in plastics encompasses all types of environmental conditions, each with its own individual, yet broad, range of properties. These properties can take into consideration wear resistance, integral color, impact resistance, transparency, energy absorption, ductility, thermal and sound insulation, weight, and so forth (see Chapters 2,3,4, and 5). There is unfortunately no one plastic that can meet all maximum properties. Therefore, the designer has different options, such as developing a compromise, because many product requirements provide options, particularly if cost is of prime importance, or combining different plastics. The combination approach permits using plastics that have different properties. They can just be stacked together, but with the available processes they can also be put together so that each material retains its individuality yet has a bond with the adjoining plastics. These processes of coinjection, coextrusion, and so on are reviewed in Chapter 7. Each of the individual plastics can provide such characteristics as wear resistance, being a barrier to water, an electrical conductor, and adding strength (Chapters 3-5). Recycled (solid waste) plastics can be sandwiched between other plastics. Plastics can also be combined with other materials such as aluminum, steel, and wood to provide specific properties (for example, PVC/wood window frames and plastic/aluminum-foil packaging

TOUGH

r

POLYCARBONATE

ACETAL

POLYPROPYLENE DAP

VINYL

POLYSTYRENE

PHENOLIC

ACRYLIC

POLYESTER

BRITTLE Figure 1-4. An example of the range of mechanical properties for plastics.

Strength /I

/

PlasticsiL={1 ~ Composites / Reinforced Plastics Wood ~) tr--J Stee I Aluminum Concrete

hV

1/

o

0 0

-0

3 150 200 x 10 psi

100

50 , /

1 500

1/

/

/

.-;/

1

1000 MPa

1

Modulus of Elasticity 7

/I

~ Plastics 17" Composites / Reinforced Plastics Wood f-(.J Ir-" Stee I Aluminum Concret

V

e~ 1/

o

0'

()

10 /

20 /

.'

I 100 1

20011

/

6

50 x 10 psi

40

30

/.' V I 360 GPa

Specific Gravity /

/I 1/

~

PlasticS Composites / Reinforced Plastics r Woo d~) Stee I Aluminum Glas s Concrete - Ston e

o

-0

f)

~l 1

2

/

I

4

/

I

6

/ 8

V

Figure 1·5. A general comparison of different materials. 7

Thermal Expansion Plastics Composites / Reinforced'-:'tz===:::f) Plastics Wood Ir-:,...,.-..Y Steel and Iron

o Thermal Conductivity W/moK

00

/

05

/

I

Plastic Foams ~ '7" Plastics / Reinforced Plastics Wood Brick Glass Concrete

10 /

I

0

)

16 I

1

o

/

1

4

2

-') /

1

6

1

/

10

8

Btu / sq.ft.?F / in. /hr.

Maximum Continuous Service Temperature

o

Plastics / Reinforced Plastics Wood Chars Aluminum Copper Alloys Steer Concrete

/ 1/

FO 200

/

400

/

/

600

800

/

~

1000 /

O·

o

1600

I

•

/

/

1800

//

I

Q

I

L /

/

100

200

I

I

/

I

300

/ 1

400

/

I

500

CO

8

1400

/

/J

v

Figure 1-5. (Continued)

1200 I

/

I

600

/

I

700

/

I

800

I

900

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 9

material). All combinations require that certain aspects of compatibility such as processing temperature and coefficient of expansion or contraction exist. (For a review of this area, see Chapter 7). The designer can use conventional plastics that are available in sheet form, in I-beams, or other forms, as is common with many other materials. Although this approach with plastics has its place, the real advantage with plastics lies in the ability to process them to fit the design, particularly when it comes to complex shapes. Two or more partsincluding mechanical and electrical connections, living hinges, colors, and snap fitscan be combined into one part (see Chapter 11). Like other materials, all plastics can be destroyed by hot enough fires. Some bum readily, others slowly, others only with difficulty; still others do not support combustion after the removal of the flame. There are certain plastics used to withstand the reentry temperature of 2,500° F (1,370° C) that occurs when spacecraft return to the earth's atmosphere. (The time exposure is parts of a millisecond.) Different industry standards can be used to rate plastics at various degrees of combustibility. Plastics' behavior in fire depends upon the nature and scale of the fire as well as the surrounding conditions and how the products are designed. For example, the virtually all-plastic 35 mm slide projectors use a very hot electric bulb. When designed with a metal light and heat reflector and fan, no fire develops. Fire is a highly complex, variable phenomenon. Therefore, designing in this environment requires understanding all the variables, so that the proper plastics can be used (see Chapters 2-6). The Design Process

The term design has many connotations, but it is essentially the process of devising a product to fulfill as completely as possible all the requirements of the end user, and, at the same time, satisfy the needs of the producer in terms of marketing and cost effectiveness (that is, return on investment). The efficient use of the available materials and production processes, including the all-important factor of tooling, should be the goal of every design effort (see Fig. 1-3). A Changing World

It would be difficult to imagine the modem world without plastics. Today they are an integral part of everyone's life-style, with applications varying from commonplace domestic articles to sophisticated scientific and medical instruments. Nowadays designers and engineers readily turn to plastics. Exceptional progress has been made in this century worldwide in all markets. As a matter of fact, many of the technical wonders we take for granted would be impossible without versatile, economical plastics. Yet some who are not mindful of the many benefits of plastics still carry negative feelings about them. Some examples of their creative use follow. 1. In recreation. Because people everywhere tend to take their fun seriously, they spend freely on sports and recreational activities. The broad range of properties available from plastics has made them part of all types of sports and recreational equipment for land, water, and airborne activities. Roller-skate wheels are now abrasion- and wearresistant polyurethane. Tennis rackets are molded from specially reinforced plastics and composites using glass, aramid, graphite, or other fibers. Skis are laminated

10 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

composites selectively reinforced to eliminate flutter at high speeds. Similarly sophisticated advanced engineering has been applied to canoes, surfboards and sailboards, outboard engine shrouds, hockey sticks, sails for racing boats, and other equipment (see Fig. 1-6).

a.

b. Figure 1-6. Examples of plastics in recreation; a) tennis rackets; b) beach accessories (chairs, bags, sunglasses, suntan lotion containers, toys, etc.); c) all-plastic sailboat; d) inflatable boat, a sailboat, and surfboards.

c.

d. 11

12 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

2. In electronics. Most of the electrical equipment and electronic devices we use and enjoy today would not be practical, economical, or occasionally even possible without plastics (see Fig. 1-7). 3. In packaging. When packaging problems are tough, plastics often are the answersometimes the only answer. They can perform tasks no other materials can and provide consumers with products and services no other materials can (see Fig. 1-8).

a. Figure 1-7. Examples of plastics in the electrical and electronics field; a) snap-in cable set of plugs and adapters, using Amoco's Ardel D-HlO polyarylate; b) plastic parts in a sixty-ft.diameter high-precision, high-frequency antenna; c) schematic of a reinforced plastics/ composite radome that protects a I50-ft.-diameter radar antenna from its Maine environment; view of reinforced plastics sandwich geodesic radome being assembled; the completely assembled radome; d) space-communication antenna. The "hom of plenty" uses an RP sandwich in its reflective panels (glass-fiber-TS polyester skins with a kraft paper-phenolic honeycomb core). It has a two-ply air-inflated Du Pont HypalonlDacron fabric elastomeric radome that will protect the antenna from the outside environment for many decades and uses other plastics. This site in Maine was the world's first ground-to-ground communication satellite.

b.

c. 13

Figure 1·7. (Continued) 14

/

d.

64-m (210-11.) diem Inllatable radome

Upper equipment room

I \ ~

58-m (192-11 ) dlam radome base~==::~

15

a. Figure 1-8. Examples of plastics in packaging; a) Extrusion and injection blow-molded bottles; b) all-plastic fifty-two-gallon electric hot-water heater meeting UL specifications that uses four different processes-blow molding, filament winding, structural foam molding, and injection molding; c) a cross-section of a multilayered, coextruded, blow-molded container to meet different performance requirements; d) a double-walled structural blow-molded design in one piece that replaced an all-plastic part made from different injection-molded parts.

16

corrugated lo r structure

c. st ru ct ur al rib S l2)

Bolt detail lorm ed by compresslo o weldlog sl o\ Is pioche
~~

/

Large detail Is plOChed ou t

Multiple tackS w ith sel/eral weldS tC' reduce part wa ll sh ill

PL

/

compressed lIaoge w ith slots plOChe d out

17

PL

18 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

d. Figure 1-8. (Continued)

4. In building and construction. This market is the second-largest consumer of plastics, after packaging. Durable and easy to install, as well as cost effective, plastics continue to find more and more applications (see Figs. 1-9 to 1-15). 5. In health care. Health-care professionals depend on plastics for everything from intravenous bags to wheelchairs, disposable labware to silicone body parts. The diversity of plastics allows them to serve in many ways, improving and sometimes prolonging lives, such as a braided, corrugated Dacron (Du Pont's polyester) aorta tube [199] (see Figs. 1-16 to 1-17). 6. In transportation. For today's autos, trucks, and vans, plastics offer a wide variety of benefits, including durability, light weight, corrosion resistance, safety glass, and fuel savings (see Figs. 1-18 to 1-20). 7. In aerospace. During the last half-century, aeronautics technology has soared, with plastics playing a major role. Lightweight durable plastics and composites save on fuel while standing up to forms of stress like creep and fatigue, and different environments (see Figs. 1-21 to 1-34).

1.0

-

. ._.C_ "'" ..... ---

Figure 1-9. An example of the 1940s to 1950s concept of plastics in buildings from The House That Plastics Built [14],

..1<" ....

I1IW'I

1Il00_

_..

----_.. _-

_.."... ~~~:2~====~~~:r~~'r::;~;~~i~::::::=~~~

, .,.. ....

Figure 1·10. A typical use of polyurethane foam-in-place insulation material in wall construction.

Figure 1·11. (Facing page) In this practically all-plastic house built in 1957 by Monsanto and designed at MIT, four cantilevered wings extended 16 ft. from a central core to support the living quarters. To provide sufficient stiffness the 8-ft.-wide by 16-ft.-long U-shaped floor and roof sections were monocoque box girders utilizing 0.30-in.-thick shells of heavy woven glassfiber rovings and thermoset polyester resin. The upper surface of the 4.5-in.-thick floor section was a sandwich structure with 0.30-in.-thick upper facing, an O.lO-in.-thick lower one, and a phenolic resin-impregnated kraft paper honeycomb core. To provide the monocoque action the floor panel was epoxy bonded along its edges to the lower shell. The design loads for this Disneyland structure in the earthquake-prone Los Angeles area were a dead load of about 10 psf, live loads of 50 psf (floor) and 30 psf (roof), and 90 mph for windloading. Stiffness rather than strength, except at the connections, was the controlling factor in this design. To avoid a feeling of shakiness, deflection under full live load at the outboard ends of the cantilevered wings was limited to ! in. When this building, a main attraction for almost two decades, pulled down to provide for a different scene, it had suffered almost no change in deflection, loadcarrying ability, and so on and was almost impossible to destroy by conventional techniques. This was true even though the house had endured the equivalent of many centuries of existence, based on the people traffic it had undergone. 20

R Intorced concretl

_--l,l,\._ _

foundation

21

Figure 1-12. Pavilions at the U.S. Exhibition in Moscow (1959) were in an MIT design in a series of umbrella shapes. Canopies 16 ft. across consisted of double-curved h-in.-thick reinforced glass-fiber and TS polyester plastic skins formed integrally with supporting ribs Hn. thick. The supporting columns, 20 ft. high, were i-in. thick.

22

b.

c. Figure 1-13. Dome-shaped buildings in Lafayette, Indiana, were built in 1966 using polystyrene boards by a Dow Chemical spiral-generation technique. Boards were heat bonded (a) in a continuous pattern to produce the dome shape. The self-supporting domes required no internal or external support during or after manufacture, provided their own insulation, and easily allowed for sections to be cut from the dome to make doors, windows, and connecting halls from dome to dome; b) a model of the medical clinic; c) cross-section of central medical clinic dome, which interconnected with a smaller dome and formed connecting rooms. 23

Figure 1-14. During the 1960s and 1970s B. F. Goodrich had on display throughout the country this polyvinyl chloride type of house.

Figure I-IS. (Facing page) On October 23, 1989, General Electric opened this 3,OOO-sq.-ft. Living Environment concept house one half mile from its Plastics world headquarters in Pittsfield, Mass. It featured advanced design and building methods, processes, and materials, to serve as a laboratory for the entire building industry to explore the feasibility of widespread use of engineering plastics in construction. The two interior floors reflect modem life-styles and the daily living environment of a family of four. The basement contains a 2,500-sq.-ft. business center, offices, and a display area for prototype models, such as factories to produce highly automated housing systems. The kitchen features a center island with adjustable-height eating and working surfaces. It also provides hot and cold areas to prepare foods while keeping them at the proper serving temperatures. A prototypical bottle recycling unit has a bar-code reader to separate containers made from engineering and commodity plastics. 24

b. 25

1. Lens implant 2. Contact lens 3. In vivo artificial hearing system 4. Dental structures 5. External prothesis 6. ArtiflCiallarynx 7. ArtifICial skin 8. Heart valves 9. Artificial heart 10. Kidney-dialysis system 11 . ArtifICial blood (synthetic oxygen carriers) 12. Intraaortic balloon 13. Angioplasty catheter 14. Vascular grafts 15. Sutures 16. Postrnastectomy reconstruction 17. ArtifICial hip, knee 18. Artificia.1finger, toe joints 19. Torn ligaments 20. Natural-action Seattle Foot 21 . Aorta

Figure 1-16. Applications of plastics in the human body require perfonnance and biocompatibility, through the proper combination of design and the use of plastics. 26

Nr Inlet en'D Rain wftle ~pter Defroster nozzles

Plugs, aromrnets. pskets. pads UceIlse plate nuts and clops Battery tray drl.n lube Battery case

Radio speaker &rill Dashboard Mirror case blck EIodricil rqul.to< wse

Dome lamp. lens and bue Coat hooks Re.r shelf seal Rur shelf defroster nozzle. Fende, wellJnfl

ElecIricaI tenn.~1

boards

Deck lid emblem pad Plr .nfl hflhl Itns T•• l !i8h! 8askets

Tallhahls

socket F,esh air inlel Housln8 & valve Voltaae ,e&lll.tor wse Insulator EIeclticil connectors EIect,icil insulato<s Conllct post Insullto< Window crank handle burtns ~tes Window ,qul.to< rollers

~mp

Body wirins harness Door stn er wed,e Door h.ndle pads Door handle escutcheon plate

a.

Pump cutlet nut windshield WiSher Pump control valve Pump control 'SS'y Healer dlfIomr ducts

Instrument cluster ----.,..

Speedometer dill ----.,..: SpetdomtIer CUIS - - - _ Instrument dill IlClnss --_~ ~~ InstlUment panel p.ke!

Instrumenl cluster fac:lns Printed circuli blck bu..

IndlcatOt dill

Instrument tel-tale hou. ns ConlrDI knobs --------,-=i Instrument 1..,1 swilch - - - - ---: driver .nd slee... Electllcal insulators Door handl. e:sc.rtcheon Nm rest - - - - Slid. 10< neutral safety 5"lWikh - - -.......... ~Ir selectOt lieu! Tu.m sli~I----...../ t.bIe control houslnfl Steerin, CDlumn psket Door p a n e l - - - - - - - " Plrkins brI e pulley cable Cowl side trim panel _ _ _ _..J Dommer sw~ch rwtchcI and cam Control cable tube and clamp BII e .nd clutch ped.I bush;np Acceltllto< ped.1 _ _ _ _ _J

I r; //

Windshield wiper .1'1 Wiper mgCor bnIsh hoId.r pl.tes Windshield Wisher t-connection Windshield Wisher VlIvI body

k

M PICM Air• duct blower Inlet Heater duct ~ Heater housina: Heater. .frosttr ~ . Ir distributO< Heater harness Body wirins hamas conditioner housl", Air WIll _-huter 1Jr conditioner 101M.

""r

IJr conditioner front

Air conditioner Id;ustitw dills Seat side shield AE:::~=====SeaI adjuster bushinp Seam;", .... 11 Sewinc ",II

b. Figure 1·17. a) Typical plastic parts in an automotive body; b) automotive dashboards use plastics extensively.

27

Figure 1-18. This 1925 table-model electrocardiograph built by the Sanborn Co. looks like a cross between a bandsaw and the laboratory instrument that brought Frankenstein's monster to life. Many changes in its design and material made it much smaller, lighter in weight, attractive, easy to use, and less expensive to make than the previously all-wood model that went first to aluminum in the 1960s and then to plastics.

a. 28

b.

Co

Figure 1-19. The Motor Wheel Corp. of Lansing, Mich., introduced the Fiberide plastic composite wheel (left) that became standard equipment on the 1989 Dodge Shelby CSX (above) . It was the first successful production wheel made of an elevated-temperature vinylester compound from Goodyear Tire & Rubber Co. in a reinforced plastic/composite. Wheel weight was reduced from 30 to 50 percent over that of mild steel wheels and 10 to 20 percent over aluminum wheels . It is designed to maintain its appearance for years without maintenance, with guaranteed resistance to corrosion and chemicals. Its styling flexibility also provides a potential cost reduction to the auto manufacturer. It is compression molded using high-glass-content sheet molding compound (see Chapter 7) that includes continuous fibers. Composite wheels are stronger than wheels of conventional materials. Many have tried unsuccessfully since the mid1950s to produce RP wheels. 29

~

=

TIRES FR 78 - 14 (UNIQUE LIGHTWEIGHT )

DOWNGAGED UPPER

& LOWER CONTROL ARMS

o

GRAPHITE COMPOSITES

a

Figure 1-20. Ford's lightweight concept vehicle of the 1960s made extensive use of high-performance reinforced plastics and composites employing graphite fiber.

2.3L 14 ENGINE C-3 AUTO TRANS .

PRODUCTION QUARTER PANEL EXTENSIONS

GrFRP FRONT SEAT FRAME (BACK ONLY)

GrFRP REAR SUSPENSION ARMS - UPR. l WR .

~

Figure 1-21. Final flight tests of the Air Force AT-6 (all-reinforced plastic) BT-15 airplane occurred in 1953, completing the designing and testing that started in 1944.

Figure 1-22. The RP sandwich wing of the BT-15. Sandwich construction of glass-fiber-TS polyester skins with cellular cellulose plastic foam core was used in different processes including that of the lost-wax technique. 31

Figure 1-23. A section of the 1944 RP monocoque sandwich fuselage of the BT-15.

Figure 1-24. The "Gossamer Albatross," a human-powered all-plastic airplane that conquered the English Channel in 1979, used Du Pont's Mylar bioriented TP polyester film, Kevlar aramid fiber, Delrin acetal resin. Dacron TP polyester fiber, and nylon as key elements in its construction. This innovative use of lightweight engineered plastics used bicycle pedals to provide power to a propeller. 32

Figure 1-25. This solar challenger all-plastic airplane was designed and built by a team headed by Dr. Paul MacCready of Pasadena, California. A variety of Du Pont lightweight engineering plastics used in its construction kept its weight to two hundred pounds. In 1981 the craft made a Paris-London flight of some two hundred miles using only sunshine as power source. Its top speed was 43 mph. It reached heights exceeding 15,000 ft. and cruised between 5,000 and 10,000 feet. No batteries or other energy-storage devices were used.

Figure 1-26. Reinforced plastics are used extensively by Grumman in its E-2A Hawkeye aircraft for the rotodome, vertical stabilizers, belly radome, electrical paneling, and many other applications. 33

~

Rudder [G)

Engme strut,forward/aft fairings [K]

stabilizer fixed trailing edge panels IK/GJ

Inboard ailer panels (G

,,-

Vertical fin tip (G)

Figure 1-27. Extensive use is made of weight-saving plastic composites throughout the secondary structures of such commercial aircraft as this Boeing 767.

G = Graphite K =Kevlar F =Fiberglass

Main landing-gear doors [K/GJ Seal depressors keel beam fairing and tire burst panels

Wmg-to-body fairing [K/G

Fixed trailing edge panels [K/GJ

Inboard and outboard spoilers [G)

Trailing edge flap track support fairing [K/GJ

Vertical fin fixed trailing edge panels [K/G)

f.:-.::::·a

Aluminum

t.;.!#.~

Titanium Composites

CJ

Horizontal stabilizer (full span)

Other

Over wing fairing

Engine access doors

Forward fuselage

Figure 1-28. More than 26 percent of this McDonnell-Douglas AV-SB Harrier combat aircraft's structural weight is carbon-epoxy composite plastics.

AI STRESSED SKIN

1900

1920

1960

1940

1980

2000

VEAR

Figure 1-29. The applications of various materials in aircraft. 35

80

o

ACAP

LEAR FAN 2100

IrBUSINESS JETS

HELICOPTERS 60

A

LANDING GEAR ADV AGHTER

IDEVELOPMENT I

PRODUCTION APPLICATION AV-8B

FUSELAGES WINGS

10

~

EMPENNAG~\~

F-18

J ./

COMMERCIAL TRANSPORTS

FLAPS '" F.15?/ DOORS............ F.14 F.16 o~----~~~--~=-----~------~----~

1970 1980 FISCAL YEAR Figure 1-30. The implementation of composites in aircraft. 1950

1960

2.0

GrIEp 5116 HOLES

"C._____.,,'

B/Ep NO HOLES

STEEL~-J"'_ _ • - -B/Ep 5116 HOLES

SPECIFIC TENSILE 1.0

--

Ti

__ -;: AI

~:::::::::=::Z~~=:::::==-II'":-~..!:-=~- - G LASS/EPOX V

6

F.'p 10 IN.

.5

~-----~-~-~-~ My

KEVLAR/EPOXY

- --- --- O~--

1930

__L -_ _ _ _ 1940

~

_ _ _ _i -_ _ _ _

1950

1960

GLASS/POLYESTER

- - - - - - SPRUCE ~

_ _ _ _- L____- L____

1970

1980

YEAR OF INTRODUCTION

Figure 1·31. The specific working tensile strengths of various materials. 36

2000

, - - - - - GrlEpNO HOLES

1.5

STRENGTH,

1990

1990

~

2000

kg/m 2 m

LBIIN.2

--w.-

300

AI2024·T3

.01

200

PANEL WEIGHT

WIb

B/Ep IAVCO 55051 GrIEp IT300I52081

1000

o

2000

4000

3000

10

20

IiOOO 30

LB/IN.

""'iN.'" MN/m m

STRUCTURAL INDEX N./b

Figure 1·32. Weights of long compression panels.

1200

1000

ADVANCED COMPOSITE AIRFRAME PRODUCTION, KLB/YR FLY AWAY WT

800

NACELLES LEAR AVIA

600 SECONDARY STRUCTURE (MAINLY COMMERCIAL)

400

o~~

__

'80 '81

~~

__

F-14 F-15 F-16 STRUCTURE F-18 (MAINLY MILITARY) AV·8B

~~

__

~~

__

LEAR AVIA

~~~

'82 '83 '84 '85 '86 '87 '88 '89 CALENDAR YEAR

Figure 1·33. U.S. advanced composite airframe production for commercial and military aircraft and helicopters (1980-1990). 37

38 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

WEIGHT

% METAL AIC TOGW

NOTE: VARIATIONS IN STRUCTURES AND MATERIALS ONLY. ALL OTHER ADVANCED TECHNOLOGIES (I.E., AERODYNAMICS AND PROPULSION) REMAIN UNCHANGED

100 r - -

~

~

80 tt-

-

~

60 r-

t-

FUEL

-

40 f-

SYSTEMS

~

f-

20

o

USEFUL LOAD

-

STRUCTURE

METAL

COMPOSITE

AIC

AIC

Figure 1-34. The effect of structures and materials technology on aircraft weight.

8. In appliances. In this market plastics have been exceptionally beneficial (see Figs. 1-35 to 1-39. For example, in the early 1900s doing simple household tasks was a real chore. Washing, drying, and ironing clothes was a rigorous, two-day affair involving the filling of metal tubs, scrubbing by hand, hanging clothes to dry, and heating cast-iron flat irons on a stove. With new technology and plastics laundry rooms and kitchens the world over are operating and looking better than ever before. In the fall of 1987, Milwaukee Electric Tool Corp. found itself on the short end of the age-old supply-and-demand equation. That is, it was unable to keep up with demand for its heavy-duty electric power tools. The problem was that their machining operations could not turn out enough aluminum die-cast motor housings to keep up with market demand. The firm briefly considered what would have been a long-term solution; a state-of-the-art machining center. But a feasibility study showed that capital costs for such a facility would run into hundreds of thousands of dollars, while resulting savings would amount to a few cents per part. Fortunately, there was another option: plastic motor housings. Du Pont agreed to produce plastic prototypes of the housing in Zytel nylon 82G resin (see Fig. 1-39). The prototypes were quickly assembled; then they endured demanding drop tests and other field tests that are standard for Milwaukee Electric tools. When the housings of impact-resistant Zytel passed the tests with no problem, the firm had anew, lowercost solution to its machining problem: a plastic housing, produced from a production mold that required no machining. The redesign presented several additional opportunities. Initial target was to replace aluminum die-casting, and thereby eliminate machining as well as deflashing, trimming, and spadoning (a surface treatment that imparts a matte finish). But they also wanted to eliminate as many parts as possible, simplify the assembly, and use a

a.

b.

Figure 1·35. Three views of electric irons showing their use of different plastics and processes. Color decorative skirts use BMC (bulk-molding compounds) located above steel sole plates that must withstand continuous operating temperatures of 232°C (450°F), high-impact loads, and be durable. Under the sole plates, fluoroplastics permit an easy nonstick slipping motion of the hot iron during use. 39

40 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

c. Figure 1-35. (Continued)

product that worked as well or better than aluminum. Achieving these goals produced some spectacular benefits; part costs dropped by two-thirds, while manufacturing through-put rates increased. Savings in labor, machine, and assembly operations have been augmented by lower capital and maintenance costs for the plastic tooling. As many as one million plastic housings can be molded without major tool repair or replacement, versus 100,000 parts for the die-casting operation. Six parts in the housing were eliminated. Because plastic used is not conductive, designers were able to do away with insulating parts, such as a coil shield that separated the brush holder area from the aluminum, and the cardboard insulating sleeve that went between the copper wiring of the field core and the housing. Removing the sleeve had the added benefit of creating better airflow inside the housing, so the motor now runs cooler under load. Press-fitting a rear ball bearing into the housing and keeping the bearing securely in place proved to be a major obstacle. The solution: eight small ribs inside the rear bearing pocket. The ribs increase the amount of interference that can be overcome when press-fitting the ball bearing, and keep the bearing in its pocket with a strong, uniform force. Another concern was achieving overall perpendicularity of the housing face where it fits with a mating gear case. The molder solved the problem by repeatedly adjusting molded housing dimensions by a few thousands of an inch. The key to this fine-tuning was to establish three adjustment spots; one at each screw hole location. Thus it was much easier to design mating parts so that they sit on the lands, specific points, rather than trying to align a complete surface. Accurately repeating such minute dimensions requires batch-to-batch resin consistency and process control [10, 12]. 9. In other markets. Plastics are literally part of all markets. A few products of the many that exist are shown in Figures 1-40 to 1-45, and others appear throughout this volume.

Figure 1·36. The Singer Sewing Machine Co. used compression-molded glass fiber-TS polyester material in the 1970s to meet the extremely tight tolerances required on different parts so they would fit properly.

Figure 1·37. Being designedfor-assembly slashed the cost of Hoover's all-plastic vacuum cleaner that was also designed for the user in 1989. Clever design resulted in cutting direct labor in half, saving half the manufacturing floor, reducing the number of components from 215 to 96, requiring only 12 fasteners instead of 56 because most components snapped into place. This vacuum weighed less than twelve pounds (a 25 percent reduction), reduced seven electrical interconnections to none, eliminated all internal wiring, and used no painted parts because the color was molded in during fabrication. Only one turnover of the base was required to install components, whereas many were needed previously

41

Figure 1·38. Plastics in different refrigerator parts may include the following: (1) rollers; (2) door gaskets; (3) gasket magnets; (4) breaker strips; (5) shelf supports; (6) light cover; (7) switch plunger; (8) switch case; (9) hinge shims; (10) hinge bearings; (11) meat pan; (12) ice bucket; (13) door pans; (14) pedal bumper; (15) synthetic paints; (16) coil cover; (17) veg. pan cover; (18) egg storage; hidden parts: (19) thermo. knob; (20) thermo. indicator; (21) relay case; (22) wire insulation; (23) foam insulation; (24) fan grill; (25) drain pan; (26) drain elbow; (27) "wonderwall" laminate; (28) insulation bags; (29) timer knob; (30) tube grommets; (31) suction line insulation; (32) compressor mounts; (33) sealing plugs; (34) wall bumpers; (35) fan supports; (36) sealing compound; (37) liner supports; (38) heat breaks; (39) water pan; (40) vinyl tapes; (41) vinyl sleeves; (42) rivets; (43) drain funnel; (44) tube coating; (45) coil cover plate; (46) coil supports; (47) misc. bumpers; (48) misc. gaskets; other models: veg. pan; evap. door; nameplates; shelf slides; butter door; evap. door hinges; chiller tray; ice cube tray; ice tray grids. Courtesy, Society of Plastics Industry.

Figure 1·39. Plastic motor housing by Du Pont for Milwaukee Electric Tool Corp. The housings are molded in Zytel 82G, which has a lustrous, resin-like surface and a glass fiber content of 33 percent, by weight. This supertough resin enabled the tool to pass six-foot drop tests, which are more demanding than those performed by Underwriters Laboratories. The plastic housing exhibits better impact resistance than aluminum. Courtesy: Du Pont Co.

42

Figure 1-40. In other markets, plastics have been used for RP booms and platforms for aerial work, high-voltage lines, and the like. 43

Figure 1-41. This 8,OOO-gallon underground ribbed fuel tank of fiberglass is highly corrosion resistant. Its long life and reliability against leakage have seen it used in major applications around the world.

Figure 1-42. This gigantic filament-winding machine (see Chapter 5), which is 22 ft. high, 60 ft. wide, 125 ft. long, and has a loo-ton metal mandrel, was desigbed and built in 1966 by the Rucker Co. for Aerojet-General to wind a thirty-ton RP rocket-motor case. Its 150,OOO-gallon tank measured 21 ft. long by 156 in. in diameter, contained about 156 million miles of glass fiber, used an eight-ton textile creel containing sixty spools of glass fiber moving up to 4! mph, and took three weeks to produce each epoxy RP case, in the Todd Shipyard in Los Angeles. 44

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 45

Figure 1-43. The Glasshopper, the first hopper rail car using plastics, by Cargill, Inc., Southern Pacific, and ACF Industries. It had two major advantages: lower tare weight (54,000 lb., or 8,000 lb. lighter than steel cars [1]), and corrosion resistance, principally against the contents to be carried, such as fertilizers. The first such car was built and successfully tested in 1973-1976; the second in 1981-1983, the latter successfully meeting or exceeding the guidelines of the American Association of Railroads. The series included static and dynamic tests, coupler tests, ram tests, velocity-impact tests, fully loaded 6,000-mile tests with speeds up to 70 mph, and more. Finite element modeling (FEM) was used throughout the design stages to aid in structural analysis. E-glass rovings with TS polyester resin were used. The car body was filament wound. Other RP or composite parts included outer panels, wide flange beams, stiffeners, a top sill, roof and edge angles, and hatches. The car's capacity was about 5,000 cu. ft., its overall length 52 ft., and its height from the rail 15 ft.

Plastics have made many major contributions to the contemporary scene. For example, a new biodegradable plastic developed at the Massachusetts Institute of Technology may soon be saving lives in the form of a medical implant. This plastic is now being tested nationwide to determine its effectiveness as a drug-releasing implant in brain cancer patients. These implants, roughly the size of a quarter, are being placed in patients' brains to release the chemotherapy drug BCNU (Carmustine). These biocompatible implants have been found to be safer than injections, which can cause the BCNU to enter the bone marrow or lungs, where the drug is toxic. This plastic, known as polyanhydride, was designed so that water would trigger its degradation but would not allow a drug to be released all at once. The implant degrades from the outside, like a bar of soap, releasing the drug as it becomes smaller. The rate at which the drug is released is determined by the surface area of the implant and the rate of polymer degradation, which can be cU,stomized to release drugs at rates varying from one day to many years. This design approach also holds promise for use with different drugs for various other medical problems.

46 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 1-44. A hand-held three-gallon HDPE extrusion blow-molded sprayer tank.

Design Parameters

In contrast to working in conventional materials, such as metals, designing with plastics usually cannot be based on one key property like tensile or shear stress. Another difference is that usually the designer in plastics does not use standard sections, as in structural steel work, or standard metal components, as for mechanical-engineering applications. Figure 1-46 provides a simplified-flow diagram for setting up a design program. It refers to the more predominant practical approach and the engineering approach. With the practical approach most plastic products are required only to withstand static mechanical loads (that is, no dynamic loads). Thus, conventional short-term static tests generally suffice. The engineering approach recognizes that many plastic products have been in use at least since the 1940s and have been exposed to long-term static and/or dynamic loads based on varying environmental conditions such as temperature. Designers thus consider creep, fatigue, stress, temperature, time, and other data (see Chapters 3-5). Designing for plastic products usually starts with the classic approach of using the tensile, compressive, or flexural strength characteristics of the plastics. Quickly, however, the designer must depart from conventional engineering procedures, as the viscoelastic and heat-resistance properties of the material begin to take precedence in evaluating load performance.

Figure 1·45. An RP stack liner being inspected prior to installation in a 682-ft.-high reinforced concrete chimney (background) of the 1,500-megawatt Intermountain Power Project near Delta, Utah (1985). The liner, of fiberglass supplied by PPG Industries, protects the concrete shell from the corrosive gases that occur when sulfur dioxide is produced during coal-fired power generation. Fiberglass-TS polyester composites can provide years of service under these operating conditions. Such liners have been used in this type of application since at least the 1970s. They rapidly became a viable construction material as against steel and brick liners. The liners are in canlike sections 45 ft. long and 28 ft. in diameter. The sections were filament wound using 46 to 50 percent (by weight) resin-impregnated fiberglass rovings. The completed liner, engineered and manufactured by Ershigs, Inc., of Bellingham, Wash., contained about 100 thousand miles, or H million pounds of fiberglass roving strands. 47

48 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Ideal choice/Compromise Figure 1·46. A simplified flow diagram for a design procedure.

The dimensioning of load·bearing products in many areas of design, except such highly sophisticated parts as those used in aircraft construction, is still often the result of tradition and empiricism, rather than being based on fact. This trial·and·error process is the usual practical approach. Without proper evaluation this approach could lead to failure or overengineering, which soon becomes costly. Good engineering design calls for extensive knowledge of materials' properties as expressed in tangible data. Only then is the designer in a position to predict the satisfactory functioning of a product. It is not possible to obtain analytical mathematical solutions for many engineering designs that employ plastics, metals, ceramics, wood, and other materials. An analytical solution is a mathematical expression that provides guidelines or values for a desired unknown quantity at any location in a product. See the section on finite element analysis (FEA) in Chapter 6, for more information on analytical mathematical solutions.

Design Analysis One factor that has done a great deal to harm the reputation of plastics is that in many cases designers and engineers have, after deciding tentatively to try to introduce plastics, then slavishly copied the metal part it was to replace. Too much emphasis cannot be given the general principle that if plastics are to be used for maximum advantage and with minimum risk of failure, it is essential to cast aside all preconceived notions of design in metals and treat plastics on their own merit as one would with any other material. A hard-and-fast rule to be followed by all intending to use plastics is to design/or plastics. As an example, for the same-size cross-section the strength of conventional plastics (not the high-performance reinforced ones) is considerably less than that of most metals. The designer will thus find it necessary to increase thickness, introduce stiffening webs, and

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 49

possibly use design inserts of various types of threads to secure the proposed part. The process will in some instances also require modification to the shape of the equipment used to produce the part, or mean simplifying the fabricating process or reducing its cost (see Chapter 11). It will become obvious that what is considered good design practice insofar as metals are concerned will not necessarily be good practice for processing plastics. It is advisable when in doubt to review this book, the referenced literature on the subject, or consult processing experts who know the limitations of plastics. By working in close cooperation with the plastics fabricator it is usually possible to arrive at a satisfactory compromise. Almost all current methods of design analysis are based on models of material behavior that are relevant to traditional metallic materials, as for example elasticity and plastic yield (see Chapter 3). These principles are embodied in design formulas' design sheets or charts and in more modem techniques, such as computer-aided design (CAD) using finite element analysis. The design analyst is merely required to supply appropriate elastic or plastic constants for the material. Thus, traditional analysts can be expected to have little experience with plastics, a situation that is changing.

DESIGN SHAPE Because the formability of plastics into almost any conceivable shape is one of its design advantages, it is important for designers to understand what can be done in this medium. Shape, which can be almost infinitely varied in the early design stages, allows a given weight of materials to provide a whole spectrum of strength properties, especially in the most desirable areas of stiffness and resistance to bending. In all materials, elementary mechanical theory demonstrates that some shapes resist deformation from external loads or residual stresses in processing better than do others. This phenomenon stems from the basic physical fact that deformation in beam and sheet sections depends upon the product of modulus (E) and the second moment of inertia (I), commonly expressed as EI. The physical part performance can be changed by varying the moment of inertia or the modulus or both (see Chapter 3). Thickening plastic panels to meet stiffness requirements is an expensive use of any plastic (or other material, for that matter) because inefficiently large quantities of the material are used, and increased heating and cooling times raise the cost of fabrication. Adding more material to a plastic structural component does not always make it stronger. Unless the reinforcing ribs are added in the correctly engineered proportions, some of the additional material may be placed so that it creates high stresses, actually decreasing the loads that cause yielding or fracture (see Fig. 1-47). Lengthy equations for the moment of inertia and for deflection and stress are normally required to determine the effect of ribs on stress. However, nondimensional curves have been developed to allow quick determination of proper rib proportions, and a corresponding program for a pocket calculator or computer will allow for obtaining greater precision when required (see Chapters 3-5). With potential limitations of rib reinforcements, the use of sandwich structures made of foam, with or without facings or other cores like honeycombing, corrugation, and so forth can be used to offer high stiffness-to-weight ratios. These proportions will have a wide range of depths for improving the second moment of inertia and torsional stiffness. And integral skin-molding techniques have the additional advantage of leaving the surface of the panel effectively unfoamed.

Secondary rib

SectionB·B

Figure 1-47. Thin-skinned structures with integral ribbing to carry edge loads. 50

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 51

In pnnciple, the sandwich structure is similar to the I-beam shape in which the facings correspond to the flanges and the cores to the webs. The facings resist axial loads and provide the moment-resisting couple; the core stabilizes the facings against buckling or wrinkling under axial compression and provides resistance to shear in bending. To obtain maximum performance, facings are commonly made of reinforced plastics. Cores are usually of foamed plastics, unreinforced or reinforced plastic honeycomb, phenolic impregnated kraft paper, or balsa wood.

SUCCESS BY DESIGN A skilled designer blends a knowledge of materials, an understanding of manufacturing processes, and imagination into successful new designs. Recognizing the limits of design with traditional materials is the first step in exploring the possibilities for innovative design with relatively new materials. What is important when analyzing designs is to incorporate ergonomics and empathy, to result in a product that truly answers the user's needs. With designing there has always been the need to meet engineering, styling, and performance requirements at the lowest cost. To some there may appear to be a new era where ergonomics is concerned, but this is not true. What is always new is that there are continually easier methods on the horizon to simplify and meet all the specific requirements of a design. Some designers operate by creating only the stylish outer appearance, allowing basic engineers to work within that outside envelope. Perhaps this is all that is needed to be successful, but a more in-depth approach will work better. Beginning with a thorough understanding of the user's needs and keeping an eye toward ease of manufacture and repair, designers should also work from the inside out. The envelope that eventually emerges will then be a logical and aesthetic answer to the design challenge (see Fig. 1-48). With new plastics and processes always becoming available, the design challenge becomes easier, even when taking today's solid-waste problem into account. Today's plastics and processes, as reviewed throughout this book, allow designers to incorporate and interrelate all the aspects of success. In products such as electronics, medical devices, transportation controls (as for aircraft, cars, and boats), and many others where userfriendly design is required, it has to be obvious to all that plastics play an important role.

Responsibilities The responsibilities of designers encompass all aspects of design. Although functional design is of paramount importance, a design is not complete if it is functional but cannot easily be manufactured, or functional but not dependable, or if it has a good appearance but poor reliability. Designers have a broad responsibility to produce designs that meet all the objectives of function, durability, appearance, and low cost. They should not contend that something is now designed and it is now the manufacturing engineer's job to figure out how to make it at a reasonable cost. The functional design and the production design are too closely interrelated to be handled separately. Product designers must consider the conditions under which fabrication will take place, because these conditions affect part performance and cost (Chapter 7). Such factors as production quantity, labor, and material cost are vital. Designers should also visualize how each part is to be fabricated. If they do not or cannot, their designs may not be satisfactory or even feasible from a production standpoint. One purpose of this book is

52 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

PERFORMANCE

G1 C/ G1

AI S

COMPETITION

PC .... ~II----I~

DESIGN

.... __-

LEGAL

__ ~~ R&D

LAB

E:NVIRONMENT

SUPPLIERS

Figure 1-48. The overall design challenge.

to give designers sufficient infonnation about manufacturing processes so that they can design intelligently from a productivity standpoint.

Ethics Although there is no substitute for individual action based on a finn philosophical and ethical foundation, designers have developed guidelines for professional conduct based on the experience of many of them who have had to wrestle with troublesome ethical questions and situations previously. These guidelines can be found in the published codes of ethics for designers and engineers of a number of industry and technical societies.

COMPUTERS IN DESIGN The use of computers in design and related fields is widespread and will continue to expand. It is increasingly important for designers to keep up to date continually with the nature and prospects of computer technologies. For example, plastics databases, accessible

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 53

through computers, provide product designers with property data and information on materials and processes. To keep material selection accessible via computer terminal and a modem, the design database maintains graphic data on thermal expansion, specific heat, tensile stress and strain, creep, fatigue, programs for doing fast approximations of the stiffening effects of rib geometry, educational information and design assistance, and more (see Chapters 36, 10, and the Appendix). Computer-Aided Design Computer-aided design (CAD) is the process of solving design problems with the aid of computers. This function includes the computer generation and modification of graphic images on a video display, printing these images as hard copy using a printer or plotter, analyzing the design data, and electronic storage and retrieval of design information. Many CAD systems perform these functions in an integrated fashion that can increase the designer's productivity manyfold. It is important to recognize that the computer does not change the nature of the design process; it is simply a tool to improve efficiency and productivity. It is appropriate to view the designer and the CAD system together as a design team, with the designer providing knowledge, creativity, and control and the computer accurate, easily modifiable graphics and the capacity to perform complex design analysis at great speeds and store and recall design information. Occasionally, the computer can augment or replace many of the designer's other tools, but it is important to remember that this ability does not change the fundamental role of the designer.

Computer-Aided Design Drafting Computer-aided design drafting (CADD), a part of CAD, is the computer-assisted generation of working drawings and other documents. The CADD user generates graphics by interactive communication with the computer. The graphics are displayed on a video terminal and can be converted into hard copy by a printer or plotter. Computer-Aided Manufacturing Computer-aided manufacturing (CAM) describes a system that can take a CAD product, devise its essential production steps, and electronically communicate this information to manufacturing equipment such as robots. A CAD/CAM system offers many potential advantages over traditional manufacturing systems, including the need for less design effort through the use of CAD and CAD databases, more efficient material use, reduced lead time, greater accuracy, and improved inventory functions.

Computer-Integrated Manufacturing Computer-integrated manufacturing (CIM) is the ,coordination of all stages of manufacturing, which en~bles the manufacturers to custom design products efficiently and economically, by a computer or a system of computers.

54 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Computer-Aided Testing

The computer-aiding testing (CAT) addition to computer-aided activities involves the testing that takes place in all stages of product development (see Chapter 7). The advantage of CAT is that the output of sensors measuring the characteristics of the prototype or finished product can manipulate the product model to improve its accuracy or identify design modifications needed. In this way testing integrates design and fabrication into an ongoing, self-correcting development process. DESIGN PROCEDURE

There are certain elements common to all successful designers: knowledge of their particular design field, experience, creativity, a knowledge of the materials and processes of manufacturing, and the ability to represent a design to others. It is in this ability to represent a design that all the elements come together. A design may be represented by hand-drawn renderings or be generated by computer, with instructions provided by the designer. Regardless of the mode of representation, designing is personally satisfying because of the creativity involved in it. All designers must have a thorough training in graphics. Without it a designer must fail, because as the design conception proceeds the designer's thinking must be recorded in the form of sketches and drawings. Design drawings will often be augmented and supported by mathematical data and diagrams, including computer data. The Safety Factor

A factor of safety (FS) or safety factor (SF) is used to provide for the uncertainties associated with any design. In addition to the basic uncertainties of graphic design, a designer may also have to consider additional uncertainties: 1. Variations in material properties. Because no two plastic (or steel, for that matter) melts are exactly alike-some may have inclusions and so on-the strength properties given in materials tables are usually average values. If the value stated is a manufacturer's value, it probably is the minimum value, which can significantly reduce or eliminate its uncertainty. 2. Effect of size in stating material strength properties. Property tables, unless otherwise stated for plastics, metals, and so on, list strength values based on a specified size, yet larger components generally fail at a lower stress than a similar smaller component made of the same material (Chapter 3). 3. Type of loading. A simple static load is relatively easy to recognize, but there are cases that fail between impact and suddenly applied loads. One thus takes into account infrequently applied fatigue loading mixed with some shock loads, as for example cams, links, or feeding devices (Chapter 5). 4. Effect of processes. The fabricating operations for plastics, steel, glass, and so forth may, and usually do, introduce stress concentrations and residual stresses (Chapter 7). 5. Overall concern for human safety. All design must consider safety of the user who may be near or in contact with the product. Unexpected overloads or other situations may cause breakage and considerable bodily harm.

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 55

In order to take uncertainties into account in design, there is the safety factor or the so-called "factor of ignorance." (See Chapter 5 regarding SF.) Many designers do use the SF, but improper use of it may result in needless waste or the cost of extra material, or even physical or operational failure. Thus, one must define what one means when using the SF. In some cases the SF is stipulated by code or contract requirements.

Predicting Performance Avoiding structural failure can depend in part on the ability to predict performance for all types of materials (plastics, metals, glass, and so on). Design engineers have developed sophisticated computer methods for calculating stresses in complex structures using different materials. These computational methods have replaced the oversimplified models of materials behavior relied upon previously. The result is early comprehensive analysis of the effects of temperature, loading rate, environment, and material defects on structural reliability. This information is supported by stress-strain behavior data collected in actual materials evaluations (see Chapters 3-5). With computers the finite element method (PEA) has greatly enhanced the capability of the structural analyst to calculate displacement, strain, and stress values in complicated plastic structures subjected to arbitrary loading conditions. In its most fundamental form, PEA is limited to static, linear elastic analyses. However, there are advanced finite element computer programs that can treat highly nonlinear dynamic problems efficiently. Important features of these programs include their ability to handle sliding interfaces between contacting bodies and the ability to model elastic-plastic material properties. These program features have made possible the analysis of impact problems that only a few years ago had to be handled with very approximate techniques. Finite element techniques have made these analyses much more precise, resulting in better and more optimum designs (see Chapter 5). Nondestructive testing (NDT) is used to assess a component or structure during its operational lifetime. Radiography, ultrasonics, eddy currents, acoustic emissions, and other methods are used to detect and monitor flaws that develop during operation (see Chapter 9). The selection of the evaluation method(s) depends on the specific type of plastic, the type of flaw to be detected, the environment of the evaluation, the effectiveness of the evaluation method, the size of the structure, and the economic consequences of structural failure. Conventional evaluation methods are often adequate for baseline and acceptance inspections. However, there are increasing demands for more accurate characterization of the size and shape of defects that may require advanced techniques and procedures and involve the use of several methods.

INTERRELATING PRODUCT-RESIN·PROCESS PERFORMANCES In order to understand potential problems and solutions of design, it is helpful to consider the relationships of machine capabilities, plastics processing variables, and part performance (see Fig. 1-49). A distinction has to be made here between machine conditions and processing variables. For example, machine conditions include the operating temperature and pressure, mold and die temperature, machine output rate, and so on. Processing variables are more specific, such as the melt condition in the mold or die, the flow rate versus temperature, and so on (see Chapter 7).

56 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Interrelate Product· Resin· Process Products Properties Appearance Cost

,

~

Resin Density Melt Index Mol. WI. Distribution Additives

~~

-'----~).

Process Temperature Pressure Cycle Mold And Process Design

Figure 1·49. Interrelating product-resin-process performance. A distinction between machine conditions and processing variables must be made in order to avoid mistakes in using cause--effect relationships to advantage. It is the processing variables, properly defined and measured, not necessarily the machine settings, that can be correlated with part performance. There was a time when designers took little interest in the processing of the parts they had designed. They simply sent the drawings to the processor in another department or company and expected perfect parts to emerge, but design and processing are now so interrelated that this separation should not exist if products are to be consistently successful. Those familiar with processing can detect and correct visible problems or readily measure factors such as color, surface conditions, and dimensions. However, less-apparent property changes are another matter. These may not show up until the parts are in service, unless extensive testing and quality control are used (see Chapters 9 and 11).

Advantages and Disadvantages of Plastics As a construction material, plastics provide practically unlimited benefits to the design of products, but unfortunately no one specific plastic exhibits all these positive characteristics. The successful application of their strengths and an understanding of their weaknesses will allow designers and engineers to produce useful products. This book reviews the advantages and disadvantages of plastics. There is a wide variation in properties among the fifteen thousand commercially available materials classified as plastics. They now represent an important, highly versatile group of engineering materials. Like steel, wood, and other materials, specific groups of plastics can be characterized as having certain properties. As with other materials, for every advantage cited, a corresponding disadvantage can probably be found (see Fig. 1-50). In general, most plastics can easily be fabricated into all shapes and sizes. As reviewed, if desired they can have highly intricate shapes held to tight tolerances by using processes suitable for either limited or mass production. Many plastics can be worked using common shop techniques. Other generalizations include the fact that many consider plastics to be of low cost. In fact, some are so expensive that their use is limited to the most sophisticated technology and applications. Regardless, if the cost of materials appears to be too high

FUNDAMENTALS OF DESIGNING WITH PLASTICS AND COMPOSITES 57

for an application, a look at the processing method to be used may show that it could be less expensive, based on the material-to-processing costs. Many designers overlook this aspect. Plastics are typically not as strong or as stiff as metals, and they are prone to dimensional changes, especially under load or heat. Successful designs take these conditions into account when they influence design requirements (see Chapter 3). As will be seen, there are plastics that are stronger or stiffer than metals, and there are those that also have exceptional dimensional stability. Highly favorable conditions such as less density, good thermal insulation, a high degree of mechanical dampening, high resistance to corrosion and chemical attack, and exceptional electric resistance exist for certain plastics. There are of course those that will deteriorate when exposed to sunlight, weather, or ultraviolet light, but then there are those that resist such deterioration. No matter what the material may be, there is always room for improvement, whether it be in plastics, metals, wood, design parameters, testing procedures, or any other category. Designers are generally most familiar with metals and wood and their behavior under load and varying conditions of temperature and environment. For those designing in metals and the other materials that have been used for centuries there is extensive literature available, and since very few changes occur one can easily enter the field of designing with these materials and refer to the handbooks that tell one what to do. As an example, for room-temperature applications most metals can be considered to be truly elastic. When stresses beyond the yield point are permitted in the design, permanent deformation is considered to be a function only of applied load and can be determined directly from the stress-strain diagram. The behavior of most plastics is much more dependent on the time of application of the load, the past history of loading, the current and past temperature cycles, and environmental conditions. Ignorance of these complexities has resulted in the appearance on the market of many plastic products that were improperly designed. Fortunately, product performance has been greatly improved as the amount of technical information -on the mechanical properties of plastics has increased in the last century, particularly since the 1940s. More importantly, designers have become more familiar with the behavior of plastics rather than just pronouncing that one cannot design with plastics, something that has never been true (see Chapters

3-5). Become aware that for any gain there could be a loss not originally included in the design performance.

i Gain

c_-~ ____ - - _

~)===-=-=--=--=¥::: --:j-~:;:._~_=_:__;: =:~L?i := _

Ci

~

LOISS

---.] .,J.,..

hen you gain 'something' there will be a loss .....does that loss Inlluence product performance (for any material: plastic, ood, s eel, glass, e c.). Figure 1-50. For any gain, there could be a loss not originally included in design performance.

58 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Energy to Manufacture

To produce and process plastics requires less energy than practically any other material (see Fig. 1-51). In contrast, glass requires much more energy than any of the materials listed in Figure 1-51.

Solid Waste Only about 7 percent, by weight, of the solid waste produced is plastics. Incinerating, recycling, landfill, and other methods are used to handle the worldwide plastics (and other materials) waste problems. Incineration of plastics produces a high energy content. For example, polystyrene has nearly twice the energy content of coal, without its ensuing problems of ash, acid rain pollution, or harmful emissions. Plastics are just one of many materials that produce solid waste, and, as with other materials, there are good and bad disposal solutions. In the United States, about 20 percent or 81,000 t (90,000 tons) of plastic (polyethylene terephthalate; PET) soft-drink bottles are being recycled. The high-density polyethylene (HDPE) milk bottles recycled amount to 36,000 t (40,000 tons) per year. See Chapter 12 for more details on the waste problem. Cost

It is a popular misconception that plastics are cheap materials-they are not. On a weight basis most plastics are more expensive than steel, and only slightly less so than aluminum (see Fig. 1-52). It should be remembered that it is a bad design practice to select materials on the basis of cost per unit weight rather than volume. By far the real advantage to using plastics to produce low-cost products is the low processing cost, as discussed previously. Figures 1-53 to 1-55 show different factors involving costs.

.£

8

o

Fuel

•

Feedstock

::J

U

."

m

6

Figure 1-51. Energy requirements for different materials.

Metals

CI)

E

= Ci

3

> 10il

-

2

u

Thermosets

0

.S! I 'll a:

Figure I-52. A general cost comparison, based on volume, for a general classification of materials.

Premium rates over bulk rail car 6

Size of contamer

SpeCial size containers

v~ ~ LttY

Drum

Truck

Rail car

250 to 350lb

~

Sag

Sox

$20 billion annual sales used by over 22,000 plants

"'"

Processing Number of plants employees 65% 1 to 19 19% 20 to 49 7% 50 to 99 91%

20% of sales

~~

I

1

ro1000lb

mIJlm E

3 r;~~===;'(jj~

40,000 Ib

150,000 to 200,000 Ib

9% of processing plants (100 + employees)

80% of sales

Figure 1-53. Estimated plastics dollar purchases by plant size and size of containers. 59

High Volume Parts Power 37% Labor 3% Water 19% Plant overhead 41% Machine operation 11 %

Taxes 0.5% Overhead 0.5% Precision Parts

Power 14% Water 6% Labor 2% Machine Operation 24%

Labor 4%

Figure 1·54. Share of cost to injection mold high-volume parts and precision parts.

+-'

en

0

U

t

c.

Cl :J '+-

~

Cost variation Target Performance requirement

~

Figure 1·55. The influence of manufacturing costs and perfonnance requirements on product cost. 60

Chapter 2

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS

There are some fairly broad basic guidelines that can be followed when designing a product to be made from plastics. This chapter analyzes the major groupings of plastic materials in terms of their characteristics [201-253] Chapters 3-5 then present a review of basic structural design considerations. Information about and data on plastic materials are given in Chapter 6. Plastics comprise an extraordinarily large, diverse class of materials numbering about fifteen thousand that displays a broad range of properties and processing characteristics. Like other materials, plastics are variously identified, such as plastics, resins, polymers, elastomers, foams, reinforced plastics, and composites. The terms plastics, resins, and polymers are usually taken as synonymous and are so used in this book but there are technical distinctions. A polymer is a pure unadulterated material that is usually taken as the family name for the materials, including rubbers, that have long chainlike atoms or molecules. The chains contain various combinations of oxygen, hydrogen, nitrogen, carbon, silicon, chlorine, fluorine, and sulfur. Although plastics are soft and moldable, even approaching a liquid condition during manufacture, in their finished state they are solid. Pure polymers are seldom used on their own. Technically, it is when additives are present that the terms plastic or resin are used. Elastomers are flexible. Plastics can provide flexible to rigid foams. Reinforced plastics or composites are plastics (polymers or resins) with reinforcing additives, such as fibers and whiskers, added, basically, to increase the product's mechanical properties. Throughout this book these terms are used precisely according to their respective areas of interest. The term plastics is not a definitive one. Metals, for instance, are also permanently deformable and are therefore plastic. How else could roll aluminum be made into foil for kitchen use, or tungsten wire be drawn into a filament for an incandescent light bulb, or a 90 t (100 ton) ingot of steel be forged into a rotor for a generator? Likewise the different glasses, which contain compounds of metals and nonmetals, can be permanently shaped at high temperatures. These cousins to polymers and plastics are not considered plastics within the context of this book. The term plastics became attached to polymeric materials because these materials are basically capable of being molded or formed, as are clay or plaster. Potters use wet clay to create their art, although these objects are not called plastics. Despite this seeming 61

Table 2-1. Types of Plastics Acetal (POM) Acrylics Polyacrylonitrile (PAN) Polymethylmethacrylate (PMMA) Acrylonitrile butadiene styrene (ABS) Alkyd Allyl diglycol carbonate (CR-39) Allyls Diallyl isophthalate (DAIP) Diallyl phthalate (DAP) Aminos Melamine formaldehyde (MF) Urea formaldehyde (UF) Cellulosics Cellulose acetate (CA) Cellulose acetate butyrate (CAB) Cellulose acetate propionate (CAP) Cellulose nitrate Ethyl cellulose (EC) Chlorinated polyether Epoxy (EP) Ethylene vinyl acetate (EVA) Ethylene vinyl alcohol (EVOH) Fluorocarbons Fluorinated ethylene propylene (FEP) Polytetralluoroethylene (PTFE) Polyvinyl fluoride (PVF) Polyvinylidene fluoride (PVDF) Furan lonomer Ketone Liquid crystal polymer (LCP) Aromatic copolyester (TP polyester) Melamine formaldehyde (MF) Nylon (Polyamide) (PA) Parylene Phenolic Phenol formaldehyde (PF) Phenoxy Polyallomer Polyamide (nylon) (PA) Polyamide-imide (PAl) Polyarylethers Polyaryletherketone (PAEK) Polyaryl sulfone (PAS) Polyarylate (PAR) Polybenzimidazole (PBI) Polycarbonate (PC) Polyesters Aromatic polyester (TS polyester) Thermoplastic polyesters Crystallized PET (CPET) Polybutylene terephthalate (PBT) Polyethylene terephthalate (PET) Unsaturated polyester (TS polyester)

62

Polyetherketone (PEK) Polyetheretherketone (PEEK) Polyetherimide (PEl) Polyimide (PI) Thermoplastic PI Thermoset PI Polymethylmethacrylate (acrylic) (PMMA) Polymethylpentene Polyolefins (PO) Chlorinated PE (CPE) Cross-linked PE (XLPE) High-density PE (HDPE) lonomer Linear LDPE (LLDPE) Low-density PE (LDPE) Polyallomer Polybutylene (PB) Polyethylene (PE) Polypropylene (PP) Ultra-high-molecular weight PE (UHMWPE) Polyoxymethylene (POM) Polyphenylene ether (PPE) Polyphenylene oxide (pPO) Polyphenylene sulfide (PPS) Polyurethane (PUR) Silicone (SI) Styrenes Acrylic styrene acrylonitrile (ASA) Acrylonitrile butadiene styrene (ABS) General-purpose PS (GPPS) High-impact PS (HIPS) Polystyrene (PS) Styrene acrylonitrile (SAN) Styrene butadiene (SB) Sulfones Polyether sulfone (PES) Polyphenyl sulfone (PPS) Polysulfone (PSU) Urea formaldehyde (UF) Vinyls Chlorinated PVC (CPVC) Polyvinyl acetate (PVAc) Polyvinyl alcohol (PVA) Polyvinyl butyrate (PVB) Polyvinyl chloride (PVC) Polyvinylidene chloride (PVDC) Polyvinylidene fluoride (PVF)

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 63

contradiction in the use of this term, plastics definitely identifies the materials described in this book and those produced by the worldwide plastics industry. To better understand the properties of plastics it is important to know about the transitions that occur, such as those that have a glass transition temperature of T11' a term explained in this chapter. Nearly all the mechanical properties of plastics are determined primarily by these transitions and the temperatures at which they occur. With a change in temperature different plastics can have either quick or gradual changes in viscosity and as temperatures increase the materials can change from basically rigid solids to liquids either quickly or gradually, depending on their chemical structure and composition.

Overview Each plastic has its own distinct or special properties and advantages. See Tables 2-1 and 2-2 for names and properties typical of plastics. They fall into two groups: thermoplastics (TP) and thermosets (TS) as summarized in Figures 2-1 and 2-2 and Table 23. The dividing line between a TP and a TS is not always distinct. For instance, crosslinked TSs are TPs during their initial heat cycle and prior to chemical cross-linking. Others, such as a cross-linked polyethylene (XLPE), normally are TPs that have been cross-linked either by high-energy radiation or chemically, during processing. In addition to the broad categories of TPs and TSs, TPs can be further classified in terms of their structure, as either crystalline, amorphous, or liquid crystalline. Other classes include elastomers, copolymers, compounds, commodity resins, and engineering resins. Additives, fillers, and reinforcements are other classifications that relate directly to plastics' properties and performance.

Table 2-2. Properties of Some Plastics Propeny Low Temperature Low Cost Low Gravity Thermal Expansion Volume Resistivity Dielectric Strength Elasticity Moisture Absorption Steam Resistance Flame Resistance Water Immersion Stress Craze Resistance High Temperature Gasoline Resistance Impact Cold Flow Chemical Resistance Scratch Resistance Abrasive Wear Colors

Thermoplastics

Thermosets

PP. PE. PVc. PS Polypropylene methylpentene Phenoxy glass TFE PVC EV A. PVc. TPR Chlorotritluorethylene Polysulfone TFE. PI Chlorinated polyether Polypropylene TFE. PPS. Pl. PAS Acetal UHMW PE Polysulfone TFE. FEP. PE. PP Acrylic Polyurethane Acetate, PS

DAP Phenolic Phenolic/nylon Epoxy-glass fiber DAP DAP. polyester Silicone Alkyd-glass fiber DAP Melamine DAP All Silicones Phenolic Epoxy-glass fiber Melamine-glass fiberglass Epoxy Allyl diglycol carbonate (C-39) Phenolic-canvas Urea. melamine

TFE

64 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Thermoplastic: These plashes become soft hen exposed to sufficient heat and harden when cooled, no matter how often the process is repeated

ThennoseHing: The plastics materials belont1ng to this group are set into permanent shape when heat and pressure are applied to them during forming Reheating ill not so! en these materials

Figure 2-1. The characteristics of thermoplastics (TPs) and thermosets (TSs).

Thermoplastics Thermoplastics are resins that repeatedly soften when heated and harden when cooled (see Fig. 2-2). Many are soluble in specific solvents and burn to some degree. Their softening temperatures vary with the polymer type and grade. Care must be taken to avoid degrading, decomposing, or igniting these materials. Generally, no chemical changes take place during processing. An analogy would be a block of ice that can be softened (turned back to a liquid), poured into any shape mold or die, then cooled to become a solid again. TPs generally offer higher impact strength, easier processing, and better adaptability to complex designs than do TSs. Most TP molecular chains can be thought of as independent, intertwined strings resembling spaghetti. When heated, the individual chains slip, causing a plastic flow. Upon cooling, the chains of atoms and molecules are once again held firmly. With subsequent heating the slippage again takes place. There are practical limitations to the number of heating and cooling cycles before appearance or mechanical properties are affected (see Chapters 6 and 7).

Thermosets Thermosets are resins that undergo chemical change during processing to become permanently insoluble and infusible (see Fig. 2-2 and Table 2-2). Such natural and synthetic rubbers (elastomers) as latex, nitrile, millable polyurethanes, silicone butyl, and neoprene, which attain their properties through the process of vulcanization, are also in the TS

THE STRUCTURE AND .BASIC PROPERTIES OF PLASTICS 65

Example of a Thermoset Processing Heat-Time Profile Cycle

11...

~~--~~----~~ I..aov Time High

a. Start of process b. Plastic melted d. Plastic permane~t1y hard

Example of a Thermoplastic Processing Heat·Time Profile Cycle

-.- ----

.s::::

!~

tl-~-! - _b_ - ~

a

--

c

,

a. Start of process b. Plastic melted c. Plastic hard but can be resoftened

~-------Low - - Time -... High Figure 2-2, The melting characteristics of TSs and TPs, based on their heat-time processing profiles. .......

family (see Chapter 4). The best analogy with TSs is that of a hard-boiled egg whose yolk has turned from a liquid· to a solid and cannot be converted back to a liquid (see Fig. 2-1). In general, with their tightly cross-linked structure TSs resist higher temperatures and provide greater dimensional stability than do most TPs. The structure of TSs, as of TPs, is also chainlike. Prior to molding. TSs are similar to TPs. Cross-linking is the principal difference between TSs and TPs. In TSs, during

Table 2-3. Melt-Processing Temperatures for Thermoplastics Processing Temperature Rate Material

·c

"F

ABS Acetal Acrylic Nylon Polycarbonate LOPE HOPE Polypropylene Polystyrene PVC, rigid

180-240 185-225 180-250 260-290 280-310 160-240 200-280 200-300 180-260 160-180

356--464 365-437 356-482 500-554 536-590 320-464 392-536 392-572 356-500 320-365

Not.: Values are typical for injection molding and most eXIIUsion operations. Extrusion coating is done at higher temperalUIes (i.e., about6OO"F for LOPE).

66 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

>-

l-

CONSTANT TEMPERATURE

en

o() en

>

MELTING X-LINKING COMPOUNDING ("8" STAGING) (MOLDING)

TIME

Figure 2-3. Viscosity change during the processing of thennoset plastics. The B stage represents the start of the heating cycle that recycles viscosity and is then followed by a chemical reaction (cross-linking) and solidification of the plastics.

curing or hardening the cross-links are fonned between adjacent molecules, resulting in a complex, interconnected network that can be related to its viscosity and perfonnance (see Figs. 2-3 and 2-4). These cross-bonds prevent the slippage of individual chains, thus preventing plastic flow under the addition of heat. If excessive heat is added after crosslinking has been completed, degradation rather than melting will occur. TSs generally cannot be used alone structurally and must be filled or reinforced with materials such as calcium carbonate, talc, or glass fiber. The most common reinforcement is glass fiber, but others are also used (see Chapters 6 and 7).

PLASTIC STRUCTURES AND MORPHOLOGY In addition to the size of the molecules and their distribution, the shapes or structures of individual polymer molecules also play an important role in detennining the properties and processability of plastics. There are those that are fonned by aligning themselves into long chains of molecules and others with branches or lateral connections to fonn complex structures. All these fonns exist in either two or three dimensions. Because of the geometry, or morphology, of these molecules some can come closer together than others. These are identified as crystalline, all others as amorphous. Morphology influences such properties as mechanical and thennal, swelling and solubility, specific gravity, and chemical and electric properties. This behavior of morphology basically occurs with TP, not TS, plastics. When TSs are processed, their individual chain segments are strongly bonded together during a chemical reaction that is irreversible.

Crystalline and Amorphous Plastics Plastic molecules that can be packed closer together can more easily fonn crystalline structures in which the molecules align themselves in some orderly pattern. During

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 67

-----------

t

I

>-

+-'

......

C1l

Q.

o......

Q.

ro

u

c ro .r:. u

Elastic modulus

- - -_ _ _- Strength

Stress-intensity factor

C1l

~

Distance between cross-links

•

Figure 2-4. The effect of distance between TS cross-linked sites on compressive properties.

processing they tend to develop higher strength in the direction of the molecules. Since commercially perfect crystalline polymers are not produced, they are identified technically as semicrystalline TPs but in this book are called crystalline (as it is called by the plastic industry). The amorphous TPs, which have their molecules going in all different directions, are normally transparent. Compared to crystalline types, they undergo only small volumetric changes when melting or solidifying during processing. Tables 2-4 to 2-8 compare the basic performance of crystalline and amorphous plastics. Exceptions exist, particularly with respect to the plastic compounds that include additives and reinforcements. As symmetrical molecules approach within a critical distance, crystals begin to form in the areas where they are the most densely packed. A crystallized area is stiffer and stronger, a noncrystallized (amorphous) area tougher and more flexible. With increased crystallinity, other effects occur. As an example, with polyethylene there is increased resistance to creep, heat, and stress cracking as well as increased mold shrinkage. In general, crystalline types of plastics are more difficult to process, requiring moreprecise control during fabrication, have higher melting temperatures and melt viscosities,

Table 2-4. The General Morphology of Thermoplastics Crystalline No

Excel No

High High Low Yes Yes *Major exception i5 PC. tTff = Temperatureflime.

Amorphous Transparent Chemical resistance Stress-craze Shrinkage Strength Viscosity Melt temperature Critical TlTt

Yes Poor Yes Low Low* High No No

68 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 2-5. The Distinctive Characteristics of Polymers Crystalline

Amorphous

Sharp melting point Usually opaque High shrinkage Solvent resistant Fatigue/wear resistant

Broad softening range Usually transparent Low shrinkage Solvent sensitive Poor fatigue/wear

and tend to shrink and warp more than amorphous types. They have a relatively sharp melting point. That is, they do not soften gradually with increasing temperature but remain hard until a given quantity of heat has been absorbed, then change rapidly into a low-viscosity liquid. If the amount of heat is not' applied properly during processing, product performance can be drastically reduced or an increase in processing cost occur (see Chapter 7). This is not necessarily a problem, because the qualified processor will know what to do. Amorphous plastics soften gradually as they are heated, but they do not flow as easily during molding as do crystalline materials. Processing conditions influence the performance of plastics. For example, heating a crystalline material above its melting point, then quenching it can produce a polymer that has a far more amorphous structure. Its properties can be significantly different than if it is cooled properly (slowly) and allowed to recrystallize, during which processing it becomes amorphous. The effects of time are similar to those of temperature in the sense that any given plastic has a preferred or equilibrium structure in which it would prefer to arrange itself. However, it is prevented from doing so instantaneously or at least on "short notice." If given enough time, the molecules will rearrange themselves into their preferred pattern. Heating causes this action to occur sooner. During this action severe shrinkage and property changes could occur in all directions in the processed plastics. This characteristic morphology of plastics can be identified by tests (see Chapter 9). It provides excellent control as soon as material is received in the plant, during processing, and after fabrication.

Liquid Crystalline Polymers Liquid crystalline polymers (LCPs) are best thought of as being a separate, unique class of TPs. Their molecules are stiff, rodlike structures organized in large parallel arrays or domains in both the melted and solid states. These large, ordered domains provide LCPs with characteristics that are unique compared to those of the basic crystalline or amorphous plastics (see Table 2-9) [2].

Table 2-6. Examples of Crystalline (Semicrystalline) and Amorphous TPs Crystalline

Amorphous

Acetal (POM) Polyester (PET, PBT) Polyamide (nylon) (PA) Fluorocarbons (PTFE, etc.) Polyethylene (PE) Polypropylene (PP)

Acrylonitrile-butadiene-styrene (ABS) Acrylic (PMMA) Polycarbonate (PC) Modified polyphenylene oxide (PPO) Polystyrene (PS) Polyvinyl chloride (PVC)

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 69

Table 2-7. Examples of Key Properties for Engineering TPs Crystalline

Amorphous

Acetal Best property balance Stiffest unreinforced thermoplastic Low friction

Polycarbonate Good impact resistance Transparent Good electrical properties

Nylon High melting point High elongation Toughest thermoplastic Absorbs moisture

Modified PPO Hydrolytic stability Good impact resistance Good electrical properties

Glass reinforced High strength Stiffness at elevated temperatures Mineral reinforced Most economical Low warpage Polyester (glass reinforced) High stiffness Lowest creep Excellent electrical properties

These LCPs provide the designer with unparalleled combinations of properties, such as resisting most solvents and heat. Unlike many high-temperature plastics, LCPs have a low melt viscosity and are thus more easily processed, and in faster cycle times, than those with a high melt viscosity. They have the lowest warpage and shrinkage of all the TPs. When they are injection molded or extruded, their molecules align into long, rigid chains that in turn align in the direction of flow and thus act like reinforcing fibers, giving LCPs both high strength and stiffness. As the melt solidifies during cooling, the molecular orientation freezes into place. The volume changes only minutely, with virtually no frozenin stresses.

Table 2-8. General Properties of TPs During and After Processing Property Melting or softening Density (for the same material) Heat content Volume change on heating After-molding shrinkage Effect of orientation Compressibility

Crystalline·

Amorphoust

Fairly sharp melting point Increases as crystallinity increases Greater Greater Greater Greater Often greater

Softens over a range of temperature Lower than for crystalline material Lower Lower Lower Lower Sometimes lower

·Typical aystalline plastics are polyelbylene, polypropylene, nylon, acetals, ODd Ibermoplastic polyesters. fTypical 3!DOIphous plastics are polystyn:ne, acrylics, PVC, SAN, ODd ADS.

70 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 2-9. General Properties of Crystalline, Amorphous, and Liquid Crystalline Polymers Property Specific gravity Tensile strength Tensile modulus Ductility, elongation Resistance to creep Max. usage temperature Shrinkage and warpage Flow Chemical resistance

Crystalline Higher Higher Higher Lower Higher Higher Higher Higher Higher

Amorphous

Liquid crystalline

Lower Lower Lower Higher Lower Lower Lower Lower Lower

Higher Highest Highest Lowest High High Lowest Highest Highest

In service, molded parts experience very little shrinkage or warpage. They have high resistance to creep. Their fiberlike molecular chains tend to concentrate near the surface, resulting in parts that are anisotropic, meaning that they have greater strength and modulus in the flow direction, typically on the order of three to six times those of the transverse direction. However, adding fillers or reinforcing fibers to LCPs significantly reduces their anisotropy, more evenly distributing strength and modulus and even boosting them. Most fillers and reinforcements also reduce overall cost and place mold shrinkage to near zero. Consequently, parts can be molded to tight tolerances. These low-melt-viscosity LCPs thus permit the design of parts with long or complex flow paths and thin sections.

Elastomers Plastic elastomers are generally lower-modulus flexible materials that can be stretched repeatedly to at least twice their original length at room temperature, but will return to their approximate original length when the stress is released. Thermoset elastomeric or rubber materials have been around for a long time, the rubber types for over a century. They will always be required to meet certain desired properties, but thermoplastic elastomers (TPEs) are replacing traditional TS natural and synthetic rubbers. TPEs are also widely used to modify the properties of rigid TPs, usually by improving their impact strength (see Chapter 6). TPEs offer a combination of strength and elasticity as well as exceptional processing versatility. They present creative designers with endless new and unusual product opportunities. More than 100 major different groups of TPEs are produced worldwide, with new grades continually being introduced to meet different electrical, chemical, radiation, wear, swell, and other requirements. Quite large elastic strains are possible with minimal stress in TPEs; these are the synthetic rubbers. TPEs have two specific characteristics: their glass transition temperature (Tg) is below that at which they are commonly used, and their molecules are highly kinked, as in natural TS rubber (isoprene). When a stress is applied, the molecular chain uncoils and the end-to-end length can be extended several hundred percent, with minimum stresses. Some TPEs have an initial modulus of elasticity of less than 10 MPa (1,500 psi); once the molecules are extended, the modulus increases. The modulus of metals decreases with an increase in temperature. However, in stretched TPEs the opposite is true, because with them at higher temperatures there is increasingly

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 71

1

10

100

Elastic Limit (Percent) Figure 2·5. The strength and elasticity of different materials.

vigorous thennal agitation in their molecules. Therefore, the molecules resist more strongly the tension forces attempting to uncoil them. To resist requires greater stress per unit of strain, so that the modulus increases with temperature. When stretched into molecular alignment many rubbers can fonn crystals, an impossibility when they are relaxed and "kinked." To date, with the exception of vehicle tires, TPEs have been replacing TS rubbers in virtually all applications. Unlike natural TS rubbers, most TPEs can be reground and reused, thereby reducing overall cost. The need to cure or vulcanize them is eliminated, reducing cycle times, and parts can be molded to tighter tolerances. Most TPEs can be colored, whereas natural rubber is available only in black. TPEs also weigh 10 to 40 percent less than rubbers. TPEs range in hardness from as low as 25 Shore A up to 82 Shore D (ASTM test). They span a temperature of - 34 to 177° C ( - 29 to 350°F), dampen vibration, reduce noise, and absorb shock (see Fig. 2·5). However, designing with TPEs requires care, because unlike TS rubber, which is isotropic, TPEs tend to be anisotropic during processing, as with injection molding. Tensile strengths in TPEs can vary as much as 30 to 40 percent with direction. Copolymers

Polymer properties can be varied during polymerization. The basic chemical process is carried out at the resin company, during which the polymer is fonned under the influence of heat, pressure, a catalyst, or a combination thereof, inside vessels or tubular systems called reactors. One special form of property variation involves the use of two or more different monomers as comonomers, copolymerizing them to produce copolymers (two comonomers) or terpolymers (three monomers). Their properties are usually intennediate between those of homopolymers, which may be made from the individual monomers,

72 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

and sometimes superior or inferior to them. (A polymer such as polyethylene is formed from its monomer ethylene, polyvinyl chloride polymer from its vinyl chloride monomer, and so on.)

Compounds Since the first plastic, cellulosic, was produced in 1868, there has been an ever-growing demand for specially compounded plastics. Using a postreactor technique, resins can be compounded by alloying or blending polymers, using additives such as colorants, flame retardants, heat or light stabilizers, lubricants, and so on, and adding fillers and reinforcements-or a combination thereof. The resulting reinforced compounds are usually referred to as reinforced plastics (RPs) or composites.

Alloys and Blends Alloys are combinations of polymers that are mechanically blended. They do not depend on chemical bonds, but do often require special "compatibilizers" (explained below). Plastic alloys are usually designed to retain the best characteristics of each constituent. Most often, property improvements are in such areas as impact strength, weather resistance, improved low-temperature performance, and flame retardation (see Figs. 2-6 to 2-9 and Tables 2-10 to 2-12). The classic objective of alloying and blending is to find two or more polymers whose mixture will have synergistic property improvements beyond those that are purely additive in effect (see Figs. 2-6 and 2-7). Among the techniques used to combine dissimilar

Synergistic effect

Antisynergistic effect

100% A

50A/50 8

100%8

Figure 2-6. Developing synergistic effects is the most usual objective of compounding plastics to gain significantly in performance.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 73

20

g

..s:::::

15

~ ""j'"

-

==

..s::::: Cl I::

~

Cii

10

-g

Q.

.§

"8 .!::!

5

"0 Q)

..s:::::

o

(5 Z

O~

--

________________________

100/0

50/50

0/100

PVC/ASS ratio Figure 2-7. An example of how alloying affects resin properties. Compounding and alloying technology makes it possible to combine two or more polymers into alloys with their own distinctive, often unique properties. The curves in this graph reflect four different poly blends.

polymers are cross-linking, to form what are called interpenetrating networks (IPNs), and grafting, to improve the compatibility of the resins. Alloys can be classified as either homogeneous or heterogeneous. The former can be depicted as a solution with a single phase or single glass-transition temperature (Tg ). A heterogeneous alloy has both continuous and dispersed phases, each retaining its own distinctive Tg • Until recently, blending and alloying were either restricted to polymers that had an inherent physical affinity for each other or else a third component, called a compatibilizer, was employed. These constraints severely limited the types of polymers that could be blended without sacrificing their good physical properties. As a rule, incompatible polymers produce a heterogeneous alloy with poor physical properties. The advances in polymer blending and alloying technology have come until recently through three routes: similar-rheology polymer pairs, miscible polymers such as polyphenylene oxide and polystyrene, or interpenetrating polymer networks (IPNs). All these systems are limited to specific polymer combinations that have an inherent physical affinity for each other. Now, however, there is another overall approach to producing blends via reactive polymers.

74 DESIGNING WITH PLASTICS AND COMPOSlirES: A HANDBOOK ACRYLONITRILE

A

CHEMICAL RESISTANCE ABRASION RESISTANCE HARDNESS SAN

STRENGTH CHEMICAL RESISTANCE NBR

TOUGHNESS LOW-TEMPERATURE PROPERTY RETENTION IMPACT STRENGTH

LUSTER MOLDABILITY STRENGTH RIGIDITY

B~----------------~------------------~S

BUTADIENE

SBR STRENGTH

STRYENE

Figure 2-8. ABS terpolymer properties are shown here as influenced by individual constituent polymer properties.

Interpenetrating Networks. IPNs consist of an interwoven matrix of two polymers. A typical method for producing these alloys involves cross-linking one of the monomers in the presence of the other. The need for a chemical similarity between the two types of molecules is thus reduced, because cross-linking physically traps one with the other. The result is a structure composed of two different intertwined plastics, each retaining its own physical characteristics.

o

Cost index

Plastic Polypropylene Polystyrene Impact styrene (alloy) ABS ABS/PVC (alloy) ABS/Polycarbonate (alloy) Rigid PVC

D

t::J

9J

/LLI

'///11/). 100

-,

500

1

1/111

I 1 IA / I / / .II

II I I I ,~/ " I

'17/

l

I

'////,1 1

1

Polysulfone Polysulfone/ABS (allov)

r7/77//I

1 '//.,1 11'///1

11

tJ

Impact strength index

h

~

~ '///1

Polvcarbonate

Alloy

Yield strength index

1////1 Polyphenyleneoxide (Noryl) r/ ///1 PVC/acrylic (alloy)

III

Unmodified resin

l

100 200

~W

1111i~711111lll

I /

II~VI ~$

1

II /1 I II~ 100

.

/7 Il

450 1250

3000

Figure 2·9. Different plastics can be combined to provide cost-to-performance improvements.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 75

Table 2-10. Upgrading PVC by Alloying and Blending Blending Polymer

Upgraded Property Impact resistance Tensile strength Low-temperature toughness Dimensional stability Heat-distortion temperature Processability

Moldability Plasticization Transparency ChemicaVoil resistance Toughness Adhesion

ABS, methacyrylate-butadiene-styrene, acrylics, polycaprolactone, polyimide, polyurethanes, PVC-ethyl acrylate ABS, methacyrylate-butadiene-styrene, polyurethanes, ethylene-vinyl acetate Styrene-acrylonitrile, polyurethanes, polyethylene, chlorinated polyethylene, copolyester Styrene-acrylonitrile, methacrylate-butadiene-styrene ABS, methacyrylate-butadiene-styrene, polyimide, polydimethyl siloxane Styrene-acrylonitrile, methacrylate-butadiene-styrene, chlorinated polyethylene, PVC-ethyl acrylate, ethylene-vinyl acetate, chlorinated polyoxymethylenes (acetals) Acrylics, polycaprolactone Polycaprolactone, polyurethanes, nitrile rubber, ethylene-vinyl acetate, copolyester, chlorinated polyoxymethylenes (acetals) Acrylics, polymide Acrylics Nitrile rubber, ethylene-vinyl acetate Ethylene-vinyl acetate

Grafting. Grafting two dissimilar plastics often involves a third plastic whose function is to improve the compatibility of the principal components. This "compatibilizer" material is a grafted copolymer that consists of one of the principal components and is similar to the other component. The mechanism is similar to that of having soap improve the solubility of a greasy substance in water. The soap contains components that are compatible with both the grease and the water.

Plastic Composition

Interplay Between Composite Constituents

Reinforcing Medium

Figure 2-10. The composition of plastics.

The

76 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 2-11. Outstanding Properties of Some Commercial Plastic Alloys Properties

Alloy PVC/acrylic PVC/ABS Polycarbonate/ABS ABS/polysulfone Polypropylene/ethylene-propylene-diene Polyphenylene oxide/polystyrene Styrene acrylonitrile/olefin Nylon/elastomer Polybutylene terephthalate/polyethylene terephthalate Polyphenylene sulfide/nylon Acrylic/polybutylene rubber

Flame, impact, and chemical resistance Flame resistance, impact resistance, processability Notched impact resistance, hardness, heatdistortion temperature Lower cost Low-temperature impact resistance and flexibility Processability, lower cost Weatherability Notched Izod impact resistance Lower cost Lubricity Clarity, impact resistance

Reactive Polymer.

A reactive polymer is simply a device to alloy different materials by changing their molecular structure inside a compounding machine. True reactive alloying induces an interaction between different phases of an incompatible mixture and assures the stability of the mixture's morphology. The concept is not new; this technology is now capable of producing thousands of new compounds to meet specific design requirements. The relatively low capital investment associated with compounding machinery (usually less than $1 million for a line, compared with many millions for a conventional reactor), coupled with a processing need for small amounts of tailored materials, now allows small and mid-sized compounding companies to take advantage of it. There are a variety of reactive alloying techniques available to the compounder today. They typically involve the use of a reactive agent or compatibilizer to bring about a

Table 2-12. Examples of Plastic Alloys Using Trade Names Material

Producers

PPO/PS

GE (Noryl)

ABSIPC

Mobay (Bayblend), Fiberite

PC/PET; PC/PBT

GE (Xenoy)

PET/PBT

GAF (Gafite), Hoechst Celanese (Celanex), GE (Valox) General Tire & Rubber, GE, Cycoloy, Cycovin, various compounders

PVC/ABS

PP/elastomer

Reichhold, Hoechst Celanese Montedison

Properties Polyphenylene oxide (PPO) has high strength and high heat resistance but oxidizes at temperatures required for processing; adding polystyrene (PS) makes it possible to process Acrylonitrile-butadiene-styrene (ABS) improves process ability of polycarbonate; PC contributes toughness and heat resistance PC, though tough and able to withstand very high temperatures, lacks good resistance to chemicals; polyethylene terephthalate (PET) and polybutylene terephthalate (PBT) make up for this lack Alloying with PET lowers PBT's impact resistance but brings down its cost Polyvinyl chloride (PVC) adds flame retardance and rigidity to ABS, a more easily processed resin Polypropylene (PP) contributes good heat resistance and processability; elastomers add impact resistance

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 77

molecular change in one or more of the blend's components, thereby facilitating bonding. They include the grafting process mentioned earlier and copolymerization interactions, whereby a functional material is built into the polymer chain of a blend component as a comonomer, with the resultant copolymer then used as a compatibilizer in ternary bonds, such as a PP-acrylic acid copolymer that bonds PP and AA. Another technique is solventbased interactions, using materials such as polycaprolactone, which is miscible in many materials and exhibits strong polarity, as well as hydrogen bonding, using the simple polarity of alloy components.

Additives, Fillers, and Reinforcements Compounding to change and improve the physical and mechanical properties of plastics makes use of a wide variety of fillers (see Fig. 2-10 and Tables 2-13 to 2-16). In general, mechanical properties are significantly increased by adding reinforcing fibers. Particulate fillers of various types usually increase the modulus, plasticizers generally decrease the modulus but enhance flexibility, and so on. Electrical properties may be affected by many additives, especially those that are conductive. Most plastics, which are poor conductors of current, build up a charge of static electricity. Antistatic agents can be used to attract moisture, reducing the likelihood of a spark or discharge. In most cases, different additives are used to provide lower cost and different characteristics encompassing specific overall properties. As an example, coupling agents are added to improve the bonding of a plastic to its inorganic filler materials, such as glass fibers. A variety of silanes and titanates are used for this purpose. Some extenders (that is, fillers) permit a large volume of a given plastic to be produced with relatively little actual resin. Calcium carbonate, silica, and clay are frequently used extenders. Many plastics, because they are organic, are flammable; thus, flame retardants are used in them. Additives that contain chlorine, bromine, phosphorous, metallic salts, and so forth reduce the likelihood that combustion will occur or spread. Lubricants like wax or calcium stearate reduce the viscosity of molten plastic and improve its forming charac-

Table 2-13. Guide to the Use of Fillers and Reinforcements for Composites Properties Improved

i

.~

00:

01

"i§

Filler or Reinforcement u.l! Alumina, tabular Aluminum powder Aramid Bronze Calcium carbonate Carbon black Carbon fiber Cellulose

i i.s .= '" ~ ~ i -5

01 <)

;;

g

·s

:l:!

~

g=

'"

~

= ~

f'"

.~

.g.'"

01

= .9 ~

~

is

IE"

.'" '"

• • • • • • • • • • • • • • • • • • • • • • • • • •

~

:t

.~

.lj

"'" .:l

.~

.'g"

= 8

§

·s

01

01 <) ~

iil

• • • • • • • • •

U

§

~

• • • •

.1 .:!

.j::;;

j »

• • • •

•

"0

'8~ ~ e Ii o ~

.-

=

6

00:"-

SIP S SIP S SIP SIP S SIP

(conI' d)

78 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 2-13. (Continued) Properties Improved

8

..·sJ" Filler or Reinforcement Alpha cellulose Coal, powdered Cotton Fibrous glass Graphite Jute Kaolin Mica Molybdenum disulfide Nylon OrIon Rayon Silica, amorphous Sisal fibers Fluorocarbon Talc Wood flour *p

= thennoplastic,

S

e

c

.9

~

.~

c:!

~

j

..·s" ~

tc

g

til

~

.§

f I ..9 til

~

~

• • • • •

• • • • • • • • • • • • •

~ is

• • • • • • • • • • • • •

."~ til

~

~

• • • • • • • • • • • • • • • • • • • • • • • • •

.~

·c

~

.~

.".g

..·s" . 8

til

•

• • • • • • • • • • • • • • • • • • • •

.~

." ~

~

i

!i

~

~

J J

u

~

.~

• • • • • • • • • • • • • • • • • • • • • • • • •

i. ~

.:

@" ..:.i! S S S SIP SIP S SIP SIP P SIP SIP S SIP SIP SIP SIP S

= thennoset.

tenstlcs. Plasticizers are low-molecular-weight materials that alter the properties and forming characteristics of plastics. An important application is the production of flexible grades of PVc. Colorants must provide colorfastness under the required exposure conditions of light, temperature, humidity, chemical exposure, and so on, but without reducing other desirable properties such as flow during processing, resistance to chalking and crazing, and impact strength retention. Colorants are usually classed as either pigments or dyes. Pigments are insoluble particles large enough to scatter light but not to provide the high transparency of dyes, which are soluble. But dyes are usually poorer in lightfastness, heat stability, and tendency to bleed and migrate in the plastic system, so that they are much less used than pigments. Pigments may be organic or inorganic. Organic ones usually provide stronger, more transparent colors, are higher priced (although not necessarily more costly to use), and more soluble in plastic systems. Inorganics are denser and usually of a larger particle size. Common inorganic pigments include iron oxides in buff colors, titanium dioxide in white, lead and zinc chromates (in yellows, oranges, and reds), and other metal oxides and saIts. Important organic pigments include monochromes and diazos (in yellow, orange, and red), phthalocyanine (in blues and greens), quinacridone (in gold, maroon, violet, and so on), perylene, and others. Carbon blacks are also widely used, both as a colorant and to protect polymers from thermal and UV degradation as well as a reinforcing filler. The various special colorants include metallics, fluorescents, phosphorescents, and pearle scent colorings.

~

(cont'd)

Amorphous polymers are inherently nonwarping molding resins. Only occasionally are fillers such as milled glass or glass beads added to amorphous materials, because they reduce shrinkage anisotropically.

Cost Cost Ductility, cost, tensile strength

Warpage Resistance

Ductility, cost, tensile strength

When reinforced, crystalline polymers yield much greater increases in HOT than do amorphous resins. As with tensile strength, fibrous minerals increase HOT only slightly. Fillers do not increase HDT.

Ductility, cost Ductility, cost Ductility

Ductility, cost Ductility, cost

Glass fibers Carbon fibers Fibrous minerals

Increased HeatDeflection Temperature

5 to 10% glass fibers 5 to 10% carbon fibers Particulate fillers

FR additives interfere with the mechanical integrity of the polymer and often require reinforcement to salvage strength. They also narrow the molding latitude of the base resin. Some can cause mold corrosion.

Ductility, tensile strength, cost

Ductility, tensile strength, cost

FR additive

Flame Resistance

(HOT)

Any additive more rigid than the base resin produces a more rigid composite. Particulate fillers severely degrade impact strength.

Ductility, cost Ductility, cost Ductility

Ductility, cost Ductility, cost Ductility

Glass fibers Carbon fibers Rigid minerals

Increased Flexural Modulus

Glass fibers are the most cost-effective way of gaining tensile strength. Carbon fibers are more expensive; fibrous minerals are least expensive but only slightly reinforcing. Reinforcement makes brittle resins tougher and embrittles tough resins. Fibrous minerals are not commonly used in amorphous resins.

Comments

Ductility, cost Ductility, cost Ductility

Crystalline

Ductility, cost Ductility, cost

Amorphous

Glass fibers Carbon fibers Fibrous minerals

How Achieved

Sacrifice (from Base Resin)

Increased Tensile Strength

Desired Modification

Table 2-14. Trade-offs in Thermoplastics and Composites

=

CD

PTFE Silicone MoSe Graphite

Glass fibers Carbon fibers Lubricating additives Carbon fibers Carbon powders

Reduced Wear

Electrical Conductivity

}

Glass fibers Carbon fibers Fillers

How Achieved

Reduced Coefficient of Friction

Reduced Mold Shrinkage (Increased moldto-size capability)

Desired Modification

Ductility, cost Tensile strength, ductility, cost

Ductility, cost Tensile strength, ductility, cost

Cost

Ductility, cost Ductility, cost Tensile strength, ductility, cost

Ductility, cost Ductility, cost Tensile strength, ductility, cost

Cost

Crystalline

Amorphous

Sacrifice (from Base Resin)

Table 2-14. (Continued)

Resistivities of I to 100,000 ohm-cm can be achieved and are proportional to cost. Various carbon fibers and powders are available with wide variations in conductivity yields in composites.

The subject of plastic wear is extremely complex and should be discussed with a composite supplier.

These fillers are soft and do not dramatically affect mechanical properties. PTFE loadings commonly range from 5 to 20%; the others are usually 5% or less. Higher loadings can cause mechanical degradation.

Reinforcement reduces shrinkage far more than fillers do. Fillers help balance shrinkage, however, because they replace shrinking polymer. The sharp shrinkage reduction in reinforced crystalline resins can often lead to warpage. The best "mold-tosize" composites are reinforced amorphous composites.

Addition of fibers tends to balance the difference between inftow and crossflow shrinkage usually found in crystalline polymers. When a particulate is used to reduce and balance shrinkage, some fiber is needed to offset degradation.

Comments

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 81

Table 2-15. The Influence of Fillers and Reinforcements on Thermoplastics Resin

Reinforcements

+ Can more than double tensile

Amorphous ABS SAN Amorphous Nylon Polycarbonate Modified PPO Polystyrene Polysulfones

+

+ ±

+ + -

Crystalline Aceals Nylon 6,6/6 6/10, 6/12, 11, 12 Polypropylene Polyphenylene sulfide Thermoplastic Polyesters Polyethylene

+ + + ±

+ + -

strength Can increase flexural modulus fourfold Raise HDT slightly Toughen brittle resins, embrittle tough resins Can provide 1000 ohm-cm resistivity Reduce shrinkage Reduce melt flow Raise cost Can more than triple tensile strength Can raise flexural modulus sevenfold Can nearly triple HDT Toughen brittle resins, embrittle tough resins Can provide 1 ohm-cm resistivity Reduce shrinkage Cause distortion Reduce melt flow Raise cost

Fillers - Lower tensile strength

+ Can more than double flexural

+ -

+

+ -

+

modulus Raise HDT slightly Embrittle resins Can impact special properties such as lubricity, conductivity, flame retardance Reduce and balance shrinkage Reduce melt flow Can lower cost

- Lower tensile strength + Can more than triple flexural modulus + Raise HDT slightly - Embrittle resins + Can impart special properties such as lubricity, conductivity, magnetic properties, flame retardance + Reduce shrinkage + Reduce distortion - Reduce melt flow + Can lower cost

Reinforced Plastics and Composites

Reinforced plastics (RPs) or composites hold a special place in the design and manufacturing industry because they are quite simply unique materials. During the 194Os, reinforced plastics (or low-pressure laminates, as they were then commonly known) were easy to identify. The basic definition then, as now, is simply that of a plastic reinforced with either a fibrous or nonfibrous material. TSs such as polyester and epoxies and glass fiber dominated the field. What essentially characterizes RPs is their ability to be molded into extremely small but also large shapes well beyond the basic capabilities of other processes, at little or no

Table 2-16. Example of the Effect of Carbon Black on Mechanical Properties of an ABS

c, %

Tensile Modulus E, N/mm2 (kips/in. 2)

0 3 5 7.5 10 15 20

2,280 2,500 2,720 2,820 3,010 3,540 4,000

Filler Content

(331) (362) (394) (409) (436) (513) (580)

Breaking Strength (7" N/mm (kips/in. 2) 30.9 44.2 43.2 37.7 35.1 27.8 24.8

(4.48) (6.41) (6.26) (5.47) (5.09) (4.03) (3.60)

Elongation at Break ET, %

Impact Strength a" kI/m2

8.2 3.4 3.1 2.5 2.2 1.9 1.1

208 36 43 41 31 29 26

82 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 2-17. Examples of Different Composite Systems Reinforcement Material

Properties Modified

Metal

Metal, ceramic, carbon, glass fibers

Ceramic

Metallic and ceramic particles and fibers

Elevated temperature strength Electrical resistance Thermal stability Elevated temperature strength

Matrix Material

Glass

Ceramic fibers and particles

Organics, Thermosets, Thermoplastics

Carbon, glass, organic fibers, glass beads, flakes, ceramic particles, metal wires

Chemical resistance Thermal resistance Mechanical strength Temperature resistance Chemical resistance Thermal stability Mechanical strength Elevated temperature strength Chemical resistance Antistatic Electrical resistance EMF shielding Flexibility Wear resistance Energy absorption Thermal stability

pressure. Also, there are instances in which less heat is required. (See Chapter 7 on methods of processing RPs.) Consequently, RPs went by the name low-pressure lami-

nates. In the past, the term high-pressure laminates was reserved for melamine- and phenolicimpregnated papers or fabrics compressed under high pressures (about 13.8 to 34.5 MPa, or 2,000 to 5,000 psi) and heated to form either decorative laminates (for example, Formica and Micarta) or industrial laminates for electrical and other industries. By the early 1960s, the processing of RPs had begun to involve higher pressures, and "low-pressure laminates" was dropped in favor of simply "RPs." But even then, the name referred primarily to reinforced TSs and encompassed specialized RP molding processes. By 1970 major changes had occurred. Reinforcements other than glass fiber were in use and TPs as well as TSs were being reinforced in volume. The application of RTS and RTP methods of processing began to increase, using conventional processing techniques like injection molding and rotational molding. By this time the industry required a more-inclusive term to describe RPs, so composite was added. (For some of the different composites that exist see Table 2-17.) The fiber reinforcements included higher-modulus glasses, carbon, graphite, boron, aramid (Du Pont's Kevlar aramid is the strongest synthetic fiber in the world, five times as strong as steel on an equal-weight basis), and others. Plastics include use of the heat-resistant TPs such as the polimides, polyamideimide, and so on. Chapters 3-7 provide more details.

Commodity and Engineering Plastics About 90 percent of plastics (weightwise) can be classified as commodity resins, the others being engineering resins. The five commodities of LDPE, HDPE, PP, PVC, and PS account for about two thirds of all the resins consumed. The engineering resinsnylon, PC, acetal, and so on-are characterized by improved performance in higher

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 83

mechanical properties, better heat resistance, higher impact strength, and so forth. Thus, they demand a higher price. There are commodity resins with certain reinforcements and/or alloys with other resins that put them into the engineering category. Many TSs are engineering resins. THERMAL PROPERTIES OF PLASTICS

In order to select materials that will maintain acceptable mechanical characteristics and dimensional stability designers must be aware of both the normal and extreme operating environments to which a product will be subject. Plastics' properties and processes are influenced by their thermal characteristics such as melt temperature (Tm), glass-transition temperature (Tg ), dimensional stability, thermal conductivity, thermal diffusivity, heat capacity, coefficient of thermal expansion, and decomposition (Td)' Table 2-18 provides some of these data on different plastics. All these thermal properties relate to how to determine the best useful processing conditions to meet product performance requirements. There is a maximum temperature or, to be more precise, a maximum time-to-temperature relationship for all materials preceding loss of performance or decomposition. This section therefore reviews important thermal properties. More details about ASTM standards are given in Chapter 9. The effects of temperature on plastics are discussed throughout this book, particularly in Chapters 3-6. Melt Temperature

The Tm occurs at a relatively sharp point for crystalline materials. Amorphous materials basically do not have a Tm; they simply start melting as soon as the heat cycle begins. In reality there is no single melt point, but rather a range, which is often taken as the peak of a DSC curve (see Chapter 9). The Tm is dependent on the processing pressure and the time under heat, particularly during a slow temperature change for relatively thick melts during processing. Also, if the Tm is too low, the melt's viscosity will be high and more power be required to process it. If the viscosity is too high, degradation will occur. Glass-Transition Temperature

The glass-transition temperature (Tg) is the point below which plastic behaves as glass does-it is very strong and rigid, but brittle. Above this temperature it is neither as strong or rigid as glass, but neither is it brittle. At Tg the plastic's volume or length increases (see Figs. 2-11 to 2-13). The amorphous TPs have a more definite Tg • A plastic's thermal properties, particularly its Tg , influence its processability in many different ways. The selection of a plastic should take these properties into account. A more expensive plastic could cost less to process because of its shorter processing time, requiring less energy for a particular weight. The Tg is unique to amorphous TPs. It occurs at a specific temperature that depends'! on pressure and specific volume and is lower than the melting point. Designers should" know that above Tg the mechanical properties are reduced.('Most noticeable is a reduction in stiffness by a factor that may be as high as 1,000. Therefore, the operating temperature of an amorphous TP is usually limited to below its Tg • Amorphous TPs generally have several transitions.

~

g/cm3

(167) (549) (493) (28.1) (418)

2.68 8.8 7.9 0.45 6.7

A = Amorphous resin.

• = Crystalline resin.

(A) (A) (A) (A) (A)

(C)

(C) (C)

(C) (C)

Aluminum Copperlbronze Steel Maple wood Zinc alloy

PVC

PC

PP HOPE PTFE PA PET ABS PS PMMA

(56) (60) (137) (71) (84) (66) (66) (75) (75) (84)

(lbJft.3)

0.9 0.96 2.2 1.13 1.35 1.05 1.05 1.20 1.20 1.35

Plastics (morphology)

Density

(334) (273) (626) (500) (490) (221) (212) (203) (510) (390)

1,000 1,800 2,750 400 (bums) 800

168 134 330 260 250 105 100 95 266 199

Melt Temperature Tm, °C ("F)

(41) 5 -110(-166) -115 (-175) 50 (122) 70 (158) 102 (215) 90 (194) 100 (212) 150 (300) 90 (194)

Glass Transition Temperature Tg °C ("F)

3000 4500 800 3 2500

(0.068) (0.290) (0.145) (0.140) (0.087) (0.073) (0.073) (0.145) (0.114) (0.121)

(72.5) (109) (21.3) (0.073) (60.4)

2.8 12 6 5.8 3.6 3 3 6 4.7 5

Thennal Conductivity (1(}' calls . cm 0c) (BTU/lb.°F)

0.23 0.09 0.11 0.25 0.10

0.9 0.9 0.3 0.075 0.45 0.5 0.5 0.56 0.5 0.6

(0.004) (0.004) (0.001) (0.003) (0.002) (0.002) (0.002) (0.002) (0.002) (0.002)

Heat Capacity callg °C (BTU/lb."F)

4900 5700 1000 27 3700

3.5 13.9 9.1 6.8 5.9 3.8 5.7 8.9 7.8 6.2

(1900) (2200) (338) (10.5) (1430)

(1.36) (5.4) (3.53) (2.64) (2.29) (1.47) (2.2) (3.45) (3.0) (2.4)

Thennal Diffusivity 1(}' cm21s 10-3 ft. 21hr.)

19 18 11 60 27

81 59 70 80 65 60 50 50 68 50

(10.6) (10) (6.1) (33) (15)

(45) (33) (39) (44) (36) (33) (28) (28) (38) (128)

Thennal Expansion l~cmlcm °C (1~ inJin. "F)

Table 2-18. Examples of Thermal Properties of TPs (properties of some common materials included for comparison)

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 85

r, Temperature Figure 2·11. The effect of the glass-transition temperature (Tg) on the volume or length of TPs.

The glass transition generally occurs over a relatively narrow temperature span and is similar to the solidification of a liquid to a glassy state; it is not a phased transition. Not only do hardness and brittleness undergo rapid changes in this temperature region, but other properties, such as the coefficient of thermal expansion and specific heat, also change rapidly. This phenomenon has been called second-order transition, rubber transition, and rubbery transition. The word transformation has also been used instead of

Temperature Figure 2·12. Solidification during processing of glassy/amorphous and crystalline thermoplastics.

86 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

450-

> ~ U i= (J)

«...J LU

LL.

0

(J)

::> ...J ::> 0

0

~

I I I I I I I I .;a

Co CI

M

I I I I I I I I

I 0.450

AMORPHOUS

-100

o

100

CROSS-LINKED

200

Figure 2-13. An example of the dynamic and mechanical properties of thennoplastics and thennoset plastics in relation to their glass-transition temperatures (Tg) and melt temperatures (Tm).

transltzon. When more than one amorphous transItIOn occurs in a polymer, the one

associated with segmental motions of the polymer backbone chain, or accompanied by the largest change in properties, is usually considered to be the glass transition. The glass transition temperature can be determined readily only by observing the temperature at which a significant change takes place in a specific electric, mechanical, or other physical property. Moreover, the observed temperature can vary significantly, depending on the specific property chosen for observation and on details of the experimental technique (for example, the rate of heating, or frequency). Therefore, the observed Tg should be considered to be only an estimate. 'the most reliable estimates are normally obtained from the loss peak observed in dynamic mechanical tests or from dilatometric data (ASTM D-20).

Mechanical Properties and the Tg As can be seen from Table 2-18, the value of Tg for a particular plastic is not necessarily a low temperature, which immediately helps explain some of the differences we observe in plastics. Por example, because at room temperature polystyrene and acrylic are below their respective Tg values, we observe these materials in their glassy stage. In contrast, at room temperature natural rubber is above its Tg [Tg = -75°C ( - l03°P); Tm = 30°C (86°P)], with the result that it is very flexible. When it is cooled below its Tg natural rubber becomes hard and brittle.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 87

Dimensional Stability Dimensional stability is an important thermal property for the majority of plastics. It is the temperature above which plastics lose their dimensional stability. For most plastics the main determinant of dimensional stability is their Tg • Only with highly crystalline plastics is Tg not a limitation. Substantially crystalline plastics in the range between Tg and Tm are referred to as leathery, because they are made up of a combination of rubbery noncrystalline regions and stiff crystalline areas. The result is that such plastics as PE and PP are still useful at room temperature and nylon is useful to moderately elevated temperatures even though those temperatures may be above their respective glass-transition temperatures.

THERMAL CONDUCTIVITY AND THERMAL INSULATION Thermal conductivity is the rate at which a material will conduct heat energy along its length or through its thickness. ASTM tests give an indication of how much heat must be added to a unit mass of plastic in order to raise its temperature 1°C. This is an important factor, since plastics are often used as effective heat insulation (see Fig. 2-14) in heatgenerating applications and in structures where heat dissipation is important. The high degree of the molecular order for crystalline TPs makes their values tend to be twice those of the amorphous types. The conductivity of plastics is dependent on a number of variables and cannot be reported as a single factor. It depends mainly on temperature and molecular orientation

..,-_ __ _ __

...

i~i~~i~~i~-

i

~

L.

2Smm Polyurethane 1..0mm Polystyrene 5 I. mm Cork Minerai wool 50mm 65mm Fibreboard

'40 mm Sot I woad

80mm Concrete blocks

86Cmm Bricks

Figure 2-14. An example of the equivalent thickness of common building and insulation materials required to achieve the same degree of thermal insulation.

88 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

r

->>

Oriented, flow direction

o

~

'0 C

o n; o

...E Q)

.c

I

No orientation'--_---

Oriented, cross-flow direction

I-

Temperature Figure 2-1S. An example of the variations in thermal conductivity created by temperature changes and plastics' molecular orientation.

(see Fig. 2-15). Its dependence can be ascertained. However, the molecular orientation may vary within a product, resulting in a variation in thermal conductivity. Thus, it is important for the designer to recognize such a situation. For certain products, skill is required to estimate a part's performance under steady-state heat-flow conditions, especially those made of composites. The method and repeatability of the processing technique can have a significant effect. In general, thermal conductivity is low for plastics and the plastic's structure does not alter its value significantly. To increase it the usual approach is to add metallic fillers, glass fibers (Fig. 2-16), or electrically insulating fillers such as alumina (refer to Table 2-13). Thermal conductivity can also be decreased, by foaming (see Chapter 6).

HEAT CAPACITY The heat capacity or specific heat of a unit mass of material is the amount of energy required to raise its temperature 10. It can be measured either at constant pressure or constant volume. If at constant pressure it is larger than at constant volume, because additional energy is required to bring about a volume change against,external pressure. The specific heat of amorphous plastics increases with temperature in an approximately linear fashion below and above Tg , but a steplike change occurs near the Tg • No such stepping occurs with crystalline types. For plastics, heat capacity is usually reported during constant pressure heating (see Fig. 2-17). Plastics differ from traditional engineering materials because their specific heat is temperature sensitive.

THERMAL DIFFUSIVITY Whereas heat capacity is a measure of energy, thermal diffusivity is a measure of the rate at which energy is transmitted through a given plastic. It relates directly to processability. In contrast, metals have values hundreds of times larger than those of plastics.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS B9 0.63 E

~

056

.I

049

:: 042 .; 0.35 u

.gc 8

028 0.21

r""

/

I

Qj

I

I

I-

o o

~

3.0 u~·

PC

---r-

2.5

10

;:)-

-g'":' O~

2.0 ~"2

1.5

psr

1.0

I

5

.

3.5 ~u.

/."

i

0.07

4.0

Nylon 6/6 _ _ ~

iii E 014 ~

4.5

I

I

'V .-

~;
0.5

15

20 25 Glass, %

30

35

40

o

Figure 2·16. An example of the effects on thermal conductivity of varying its glass fiber content (by weight) in composite plastics.

Thennal diffusivity detennines plastics' rate of change with time. Although this function depends on thennal conductivity, specific heat at constant pressure, and density, all of which vary with temperature, thennal diffusivity is relatively constant.

THE COEFFICIENT OF LINEAR THERMAL EXPANSION Like metals, plastics generally expand when heated and contract when cooled. Usually, for a given temperature change TPs have a greater change than metals. The coefficient

Specific heat. Btu / lb . 0 F

o

0 10

0.20

0.30

0.40

0.50 0.60

o

0.42

0.84

1.26

1.68

2.10 2.52

Polyurethanes Allyis (diglycol carbonate) Polyesters EpOXies Phenolics Polyimides

Specific heat , kJ / kg . K Figure 2·17. The range of specific heat for neat plastics.

90 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

of linear thermal expansion (CLTE) is the ratio between the change of a linear dimension to the original dimension of the material per unit change in temperature (per ASTM standards). It is generally given as cm/cm/°C or in.!infF (see Table 2-19 and Figs. 218 to 2-20). Figure 2-18 provides information on contraction at low temperatures. The CLTE is an important consideration if dissimilar materials like one plastic to another or a plastic to metal and so forth are to be assembled. The CLTE is influenced by the type of plastic (liquid crystal, for example) and composite, particularly the glassfiber content and its orientation. It is especially important if the temperature range includes a thermal transition such as Tg • Normally, all this activity with dimensional changes is available from material suppliers readily enough to let the designer apply a logical approach and understand what could happen. The design of products has to take into account the dimensional changes that can occur during fabrication (see Chapters 3-7) and during its useful service life. With a fIlismatched CLTE there could be destruction of plastics from factors such as cracking or buckling. Expansion and contraction can be controlled in plastic by its orientation, cross-linking, adding fillers or reinforcements (see Fig. 2-20), and so on. With certain additives the CLTE value could be zero or near zero. For example, plastic with a graphite filler contracts rather than expands during a temperature rise (see Fig. 1-5, "Thermal Expansion"). As shown in Table 2-19, composites with only glass-fiber reinforcement can be used to match those of metal and other materials. In fact, TSs are especially compounded to have little or no change. In a TS the ease or difficulty of thermal expansion is dictated for the most part by the degree of cross-linking as well as the overall stiffness of the units between the crosslinks. The less flexible units are also more resistant to thermal expansion. Such influences as secondary bonds have much less effect on the thermal expansion of TSs. Any cross-linking has a substantial effect on TPs. With the amorphous type, expansion is reduced. In a crystalline TP, however, the decreased expansion as a result of crosslinking may be partially offset by a loss of crystallinity.

Thermal Stresses If a plastic part is free to expand and contract, its thermal-expansion property will usually be of little significance. However, if it is attached to another material, one having a lower CLTE, then the movement of the part will be restricted. A temperature change will then result in developing thermal stresses in the part. The magnitude of these stresses will depend on the temperature change, the method of attachment and relative expansion, and the modulus characteristics of the two materials at the point of the exposed heat. For instance, a 304.8-cm (120-in.)-long extruded TP willi a high CLTE is securely fastened to a heavy steel member. It is subjected to a 43.33°C (1lO°F) temperature change, from 21.11 °C (70°F) to - 40°C ( - 40°F). CLTE cm.lcm.fC Steel TP

1.6 X 10-5 15.2 X 10-5

Temperature change (0C)

x x

43.33 43.33

x x

Length (cm)

Contraction (cm)

304.8 304.8

0.21131 2.00746

Comparing the contraction for the two materials shows that the TP part will contract about ten times as much as the steel. Because of the higher-modulus steel, the contraction of the TP will be restrained and thermal stresses will occur. The level of stress developed

Table 2-19. Examples of the Coefficient of Linear Thermal Expansion (CLTE) for Plastics and Other Materials Material

in.lin/OF x lO-s (cm/cm °C x lO-s)

Fused Quartz Liquid Crystal-OR TS Polyester-OR Phenolic-OR Silicone-OR Pine Wood Olass DAP-OR Epoxy-OR Nylon-OR Steel Concrete Copper Bronze Brass PPO-OR Aluminum PC-OR TP Polyester Polyimide Magnesium ABS-OR Zinc PS/HI PP-OR PPS-OR Acetal-OR Zinc PVC/Rigid Acrylic TS Polyester Polysulfone Epoxy Polycarbonate Phenolic ABS Nylon Acetal Polypropylene TP Polyurethane Polyethylene/LD Fluorocarbon Epoxy Polyethylene/HD TPX TP Polyester

0.02 (0.036) 0.3 (0.54) 0.3 (0.54) 0.4 (0.72) 0.4 (0.72) 0.4 (0.72) 0.4 (0.72) 0.5 (0.9) 0.6 (1.08) 0.6 (1.08) 0.6 (1.08) 0.8 (1.44) 0.9 (1.62) 1.0 (1.8) 1.0 (1.8) 1.2 (2.2) 1.2 (2.2) 1.3 (2.3) 1.3 (2.3) 1.3 (2.3) 1.4 (2.5) 1.6 (2.9) 1.7 (3.1) 1.8 (3.2) 1.8 (3.2) 2.0 (3.6) 2.2 (4.0) 2.2 (4.0) 2.7 (4.9) 2.8 (5.0) 3.0 (5.4) 3.0 (5.4) 3.0 (5.4) 3.6 (6.5) 3.8 (6.8) 4.0 (7.2) 4.5 (8.1) 4.8 (8.6) 4.8 (8.6) 5.6 (10.1) 5.6 (10.1) 5.6 (10.1) 6.0 (10.8) 6.1 (11.0) 6.5 (11.7) 6.9 (12.4)

These are only typical values to account for many different grades, molding conditions, product shapes, wall thicknesses, and other variants. The plastics presented are basically unfilled or reinforced. GR refers to glass-fiber-reinforced compounds that usually have 10 to 40 percent, by weight, reinforcement. Other reinforcements, particularly graphite, and different fillers can result in significant different CLTEs. CLTEs on specific plastics or compounds are available from material suppliers. Then apply those data or your derived data to the design.

91

92 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

o -0.21-1 . . .~S!A;E~1~0~20~st:e:el~. . . . . . . . . ." " " ' "

-0.4 -0.6 -0.8 -1.0~~__~--

-1.2

-1.8

.~ u ~

§

u

-2.6

I

I

-273-253 -233 -213 -193 -173 -153 -133 -113 -93

Temperature. 'C.

Figure 2-18. An example at low temperatures of thennal contraction in unfilled TPs. With filled, and particularly certain reinforced TSs (refer back to Fig. 1-5 "Thermal Expansion"), dimensional change can be significantly reduced or even be at zero.

in the TP will be detennined basically by its coefficient of expansion and the modulus of elasticity (see chapters 3-5) as follows:

CLTE

Temperature change eC)

TP-Steel 15.2 x lO-s - 1.6 x lO-s

x

43.33

x

Modulus at -4
Stress developed

19.3 x lOS

11.38 MPa (1,653 psi)

In this example, while the level of stress produced is below the yield stress of the TP, the presence of stress increasers can lead to failure of the product. Stress raisers can magnify the effect of thennally induced stress at the point at which the TP's tensile strength will be exceeded, which will be followed by part failure. Stress raisers may be in the fonn of sharply reduced section thicknesses, notches from poor trimming operations, or fastener holes. In its simplest fonn, thennal stress can be calculated by using the following equation:

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 93

where (J'

=

thermal stress

Ep = elastic modulus of plastic

a 1 = CLTE of material # I a2 = CLTE of material #2

tl.T = temperature change The goal is to eliminate or significantly reduce all sources of thermal stress. This can be achieved by keeping the following factors in mind: 1) when adding material for local reinforcement, select a material with the same or a similar CLTE; 2) where plastic is to be attached to a more-rigid material, use mechanical fasteners with slotted or oversized holes to permit expansion and contraction to occur; 3) do not fasten dissimilar materials tightly; and 4) adhesives that remain ductile, such as urethane and silicone, through the product's expected end-use temperature range can be used without causing stress cracking or other problems (see Chapter 8). In addition to dimensional changes from changes in temperature, other types of dimensional instability are possible in plastics, as in other materials. Water-absorbing plastics, such as certain nylons, may expand and shrink as they gain or lose water, or even as the relative humidity changes. The migration or leaching of plasticizers, as in certain PVCs, can result in slight dimensional change. Traces of an unreacted monomer may be delayed during polymerization (the producing of plastics) and result in the con-

Thermal expansion, in/inrF x 10-5

Thermoplastics

I··..

Vinyls Fluorocarbons Cellulosics Nylons Polypropylenes Acrylics Styrenes Polyallomers

Phenolics Polyesters

~. ~

•• •• ••

•

Acetals Chlorinated polyethers Polycarbonates Polyimides Polysulfones Polyphenylene oxides Ethylene vinyl acetates

Thermosets Urethanes

Polyethylenes

~

Epoxies Allylics Silicones Alkyds Aminos

~ 4

•

• • •

Figure 2-19. Examples of the range of the coefficient of linear thennal expansion for TPs and TSs.

94 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

6 ai '"

Eb III

~

5 '-

oS x

4

~ ~ =,

3

iii u.

'0 .5: E

g 2

:~ .~ :a5 ~ o ><

()

III

4,~BT

"'"'"~~'"

Nylon 6/6

PC

1

o

--- --------....:..

~

o

5

10

15

20

25

30

35

40

Glass, %

Figure 2-20. Examples of the effect on the coefficient of thermal expansion of incorporating in TPs different amounts by weight of glass reinforcements. traction of a plastic part after it is fabricated and has been put into service. And as a final example plastic parts that have been stress oriented (see Chapters 3-5 and 7) as a result of processing operations may stress-relieve themselves in service by warping and so forth. There are procedures to eliminate or control these problems, as well as tests to determine if a potential dimensional instability exists in the parts (see Chapter 9).

DEFLECTION TEMPERATURE UNDER LOAD The deflection temperature under load (DTUL), also called the heat distortion temperature (HDT) of a plastic is a method to guide or assess its load-bearing capacity at an elevated temperature. Details on the method of testing are given in ASTM D648. Basically, a 1. 27 -cm (!-in. )-deep plastic test bar is mounted on supports 10.16 cm (4 in.) apart and loaded as a beam (see Fig. 2-21). A bending stress of either 66 psi or 264 psi (455 g Pa or 1,820 g Pa) is applied at the center of the span. The test is conducted in a bath of oil, with the temperature increased at a constant rate of 2°C per minute. The DTUL is the temperature at which the sample attains a deflection of 0.0254 cm (0.010 in.).

Figure 2-21. An apparatus to test deflection temperature under load (DTUL).

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 95

This test is only a guide. It represents a method that could be correlated to product designs (see Chapters 4-5), but, as with most other tests conducted on test specimens and not on a finished product, it is just a guide. In this test, if the specimen contains internal stresses the value will be lower than a specimen with no stresses. In fact, the test can be used to determine the degree of stress. Since a stress and the deflection for a certain depth of test bar are specified, this test may be thought of as establishing the temperature at which the flexural modulus decreases to particular values: 35,000 psi (240 MPa) at 66 psi load stress, and 140,800 psi (971 MPa) at 264 psi. DECOMPOSITION TEMPERATURE

For applications having only moderate thermal requirements, thermal decomposition may not be an important consideration. However, if the product requires dimensional stability at high temperatures, it is possible that its service temperature or processing temperature may approach its temperature of decomposition (Table 2-20). A plastic's decomposition temperature is largely determined by the elements and their bonding within the molecular structures as well as the characteristics of additives, fillers, and reinforcements that may be in the compounds (see Table 2-13). Aging at Elevated Temperatures

Aging at elevated temperatures typically involves exposing test specimens or products at different temperatures for different extended times (see chapters 3-6). Tests are performed at room or the respective testing temperatures for whatever mechanical, physical, or electrical property is of interest. These tests of aging can be used as a measure of thermal stability in design as is done with other materials. Temperature Index

The Underwriters Laboratories (UL) tests are recognized by various industries to provide continuous temperature ratings, particularly in electrical applications. These ratings include separate listings for electrical properties, mechanical properties (including impact), and mechanical properties without impact. The temperature index is important if the final product has to receive UL recognition (see Chapters 3-5).

Table 2-20. Temperature Decomposition (Td ) Ranges for Various Polymers Material

PP PC

pvc PS

PMMA

ABS PA

PET F1uoropolymer

"F 610--750 645-825 390--570 570--750 355-535 480--750 570--750 535-610 930--1020

Nott: Adding certain IiIIers and Jeinfon:cmcnlS can nlisc decomposition temperatwes.

(321-399) (341-441) (199-299) (299-399) (180--280) (249-399) (299-399) (280--322) (499-549)

96 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

MECHANICAL PROPERTIES Throughout this book all the different types of mechanical properties are presented and reviewed, particularly in Chapters 3-5. These mechanical properties include a tremendous range of different types that can usually be characterized by their stiffness, strength, and toughness.

Stiffness The same factors that influence thermal expansion (see Chapters 3-5) dictate the stiffness of plastics. Thus, in a TS the degree of cross-linking and amount of overall flexibility are important. In a TP its crystallinity and secondary bond's strength control its stiffness.

Strength The subject of strength is much more complex than stiffness, since so many different types exist: short- or long-term, static or dynamic, and torsion or impact strengths (see Chapters 3-5). Some strength aspects are interrelated with those of toughness. This section reviews certain simplified concepts of strength that are important influences on strength based on long- and short-term exposure. The crystallinity of TPs is important for their short-term yield strength. Unless the crystallinity is impeded, increased molecular weight generally also increases the yield strength. However, the cross-linking of TSs increases their yield strength substantially but has an adverse effect upon toughness. Long-term rupture strengths in TPs are increased much more readily by increasing the secondary bonds' strength and crystallinity than by increasing the primary bond strength. Fatigue strength is similarly influenced, and all factors that influence thermal dimensional stability also affect fatigue strength. This is a result of the substantial heating that is often encountered with fatigue, particularly in TPs.

Toughness Toughness is usually the most complex factor to define and understand. Tough plastics are variously described as ones having a high elongation to failure or ones in which a lot of energy must be expended to produce failure. For high toughness a plastic needs both the ability to withstand load and the ability to elongate substantially without failing except in the case of reinforced TSs, which may have high strengths with low elongation [1, 13, 14,29,32,42]. It may appear that factors contributing to high stiffness are required, but this is not true, because there is an inverse relationship between flaw sensitivity and toughness: the higher the stiffness and the yield strength of a TP, the more flaw sensitive it becomes. However, because some load-bearing capacity is required for toughness, high toughness can be achieved by a high trade-off of certain factors [62, 63]. Crystallinity increases both stiffness and yield strength, resulting usually in decreased toughness. This is true below Tg in most noncrystalline (amorphous) plastics, and below or above the Tg in a substantially crystalline plastic. However, above the Tg in a plastic having only moderate crystallinity increased crystallinity improves its toughness. Furthermore, an increase in molecular weight from low values increases toughness, but with continued increases, the toughness begins to drop.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 97

Cross-linking produces some dimensional stability and improves toughness in a noncrystalline type of plastic above the Tg , but high levels of cross-linking lead to embritdement and loss of toughness. This is one of the problems with TSs for which an increase in Tg is desired. Increased cross-linking or stiffening of the chain segments increases the Tg , but it also decreases toughness. As previously discussed, a popular way to increase toughness is to blend, compound, or copolymerize a brittle plastic with a tough one (see Tables 2-10,2-11,2-12 and Figs. 2-6 to 2-9). Although some loss in stiffness is usually encountered, the result is a satisfactory combination of properties. Theoretical versus Actual Values Through the laws of physics, chemistry, and mechanics, theoretical values can be determined for different materials. These are compared to actual values in Table 2-21. With steel, aluminum, and glass the theoretical and actual experimental values are practically the same, whereas for polyethylene, polypropylene, nylon, and other plastics they are far apart, and have the important potential of reaching values that are far superior to those of the other materials [47]. Table 2-21. Comlarison of the Theoretically Possible and Actual Experimental Values for Mo ulus of Elasticity and Tensile Strength of Various Materials· Modulus of Elasticity

Tensile Strength

Experimental

Experimental

Type of Material

Theoretical , N/mrn2 (kpsi)

Fiber, N/mrn2 (kpsi)

Nonnal Polymer, N/mrn2 (kpsi)

Polyethylene

300,0000

100,000 (33%) (14,500)

1,000 (0.33%) (145)

27,000

20,000 (40%) (2,900)

1,600 (3.2%) (232)

16,000

5,000 (3%) (725)

2,000 (1.3%) (290)

27,000

80,000 (100%) (11,600)

70,000 (87.5%) (10,100)

11,000

210,000 (100%) (30,400)

210,000 (100%) (30,400)

21,000

76,000 (100%) (11,000)

76,000 (100%) (11,000)

7,600

(43,500) Polypropylene

50,000 (7,250)

Polyamide 66

160,000 (23,200)

Glass

80,000 (11,600)

Steel

210,000 (30,400)

Aluminum

76,000 (11,000)

Theoretical , N/mrn2 (kpsi)

(3,900)

(2,300)

(3,900)

(1,600)

(3,050)

(1,100)

'For the experimental values the percentage of the theoretically calculated values is given in parenthesis. as (47).

Fiber, N/mrn2 (kpsi)

Nonnal Polymer, N/mrn2 (kpsi)

1,500 (5.5%) (218)

30 (0.1%) (4.4)

1,300 (8.1 %) (189)

38 (0.24%) (5.5)

1,700 (6.3%) (246)

50 (0.18%) (7.2)

4,000 (36%) (580)

55 (0.5%) (8.0)

4,000 (19%) (580)

1,400 (6.67%) (203)

800 (10.5%) (116)

600 (7.89%) (87)

98 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

When polyethylene was first produced in the early 1940s, physicists in England, the United States, and Germany predicted a tremendous potential for it [37]. At that time the properties of PEs were much lower than those presently available. Out of that original general-purpose PE, have been developed such specific PEs as LDPE, HOPE, and UHMWPE (see Chapter 6).

PHYSICAL PROPERTIES An overall review of physical properties is presented here, including some properties that are specific to plastics. For more details, refer to the ASTM standard references and Chapter 9.

Density and Specific Gravity The density of any material is a measure of its mass per unit volume, usually expressed as grams per cubic centimeter (g/cc) or pounds per cubic inch (lbs.lin.3) (see Tables 2-22 and 2-23 and Figs. 2-22 to 2-24). See Fig. 2-23 for determining the specific gravity of filled compounds. It is necessary to know the density of a particular plastic in order to calculate the relationship between the weight and volume of the material in a specific product. Specific gravity is the ratio of the mass of a given volume of plastic compared to the mass of the same volume of water, both being measured at room temperature (23°C/73.4"F); in other words, it is the density of the plastic divided by the density of the water. Since this is a dimensionless quantity, it is a convenient one for comparing different materials. Like density, specific gravity is used extensively in determining parts' cost, weight, and quality control. The ASTM D792 standard provides the relationship of density to specific gravity at 23°C. Density, glcc = Specific Gravity

X

0.9975

Also Specific Gravity x 0.0361 = lb.lcu. in.

Opacity and Transparency Opacity or transparency are important when the amount of light to be transmitted is a consideration. These properties are usually measured as haze and luminous transmittance. Haze is here defined as the percentage of transmitted light through a test specimen that is scattered more than 2.5° from the incident beam. Luminous transmittance is the ratio of transmitted light to incident light. Table 2-24 provides the optical and various other properties of different transparent plastics. Some definitions of key terms used in identifying optical conditions follow: Refractive index. The ratio of the velocity of light in free space to the velocity of light in the medium. Light scattering. The change in direction of a portion of the light transmitted due to refraction or reflection at the surfaces of inclusions in the material.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 99

Table 2-22. Specific Gravity and Density Comparisons of Different Materials Specific Gravity

Density, Ib.lcubic in.

Thermoplastics ABS Acetal Acrylic Cellulose Acetate Cellulose Acetate Butyrate Cellulose Propionate Ethyl Cellulose Methyl Methacrylate Nylon, Glass-Filled Nylon Polycarbonate Polyethylene Polypropylene Polybutylene Polystyrene Polyimides PVC-Rigid Polyester

1.06 1.43 1.19 1.27 1.19 1.21 1.10 1.20 1.40 1.12 1.20 0.94 0.90 0.91 1.07 1.43 1.20 1.31

0.0383 0.0516 0.0430 0.0458 0.0430 0.0437 0.0397 0.0433 0.0505 0.0404 0.0433 0.0339 0.0325 0.0329 0.0386 0.0516 0.0433 0.0473

Thermosets Alkyds, Glass-Filled Phenolic-G.P. Polyester, Glass-Filled

2.10 1.40 2.00

0.0758 0.0505 0.0722

Rubber

1.25

0.0451

Metals Aluminum SAE-309 (360) Brass-Yellow (#403) Steel--CR Alloy (Strip & Bar) Steel-Stainless 304 Magnesium ~91B Iron-Pig, Basic Zinc-SAE-903

2.64 8.50 7.85 7.92 1.81 7.10 6.60

0.0953 0.3070 0.2830 0.2860 0.0653 0.2560 0.2380

Materials

The number of grams per cubic centimeter is the same as the specific gravity. For example, if the specific gravity is 1.47, that substance has a density of 1.47 gmsIcm3 .

Birefringence. The property of anisotropic optical media that causes polarized light with one orientation to travel with a different velocity than polarized light with another orientation. Polarized light. Light that has the electric field vector of all the energy vibrating in the same plane. Looking into the end of a beam of polarized light one would see the electric field vectors as parallel or coincident lines. Dichroism. A property of an optical material that causes light of some wave-lengths to be absorbed when the incident light has its electric field vector in a particular orientation and not absorbed when the electric field vector has other orientations. Light transmissability. The ratio of the light exiting from an optical material to the light entering the material.

100 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 2-22. Weight in grams and volume in cubic inches versus specific gravity.

Haze. The cloudy appearance in a plastic material caused by inclusions that produce light scattering.

Color. The sum effect of the wavelengths of light transmitted by or reflected from a material.

Dispersion. A property of an optical material that causes some wavelengths of light to be transmitted through the material at different velocities and the velocity is a function of the wavelength. This causes each wavelength of light to have a different refractive index.

Table 2-23. Specific Gravity as a Function of Mass per Volume

Specific Gravity

Ounces per Cu. Inch

Grams per Cu. Inch**

0.90 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 l.l0 1.11 l.l2 l.l3 l.l4 l.l5 l.l6 l.l7 l.l8 l.l9 1.20 1.21 1.22 1.23 1.24 1.25 1.26 1.27 1.28 1.29 1.30

0.5220 0.5258 0.5316 0.5374 0.5431 0.5489 0.5447 0.5605 0.5662 0.5720 0.5778 0.5836 0.5894 0.5951 0.6009 0.6067 0.6125 0.6182 0.6240 0.6298 0.6356 0.6414 0.6471 0.6529 0.6587 0.6645 0.6702 0.6760 0.6818 0.6876 0.6934 0.6991 0.7049 0.7107 0.7165 0.7222 0.7280 0.7338 0.7396 0.7454 0.7511

14.748 14.912 15.076 15.240 15.404 15.568 15.732 15.895 16.059 16.223 16.387 16.551 16.715 16.879 17.042 17.206 17.370 17.534 17.698 17.862 18.026 18.189 18.353 18.517 18.681 18.845 19.009 19.173 19.337 19.501 19.664 19.828 19.992 20.156 20.320 20.484 20.648 20.811 20.975 2 l.l 39 21.303

Specific Gravity

Ounces per Cu. Inch

Grams per Cu. Inch

Specific Gravity

Ounces per Cu. Inch

Grams per Cu. Inch

1.31 1.32 1.33 1.34 1.35 1.36 1.37 1.38 1.39 1.40 1.41 1.42 1.43 1.44 1.45 1.46 1.47 1.48 1.49 1.50 1.51 1.52 1.53 1.54 1.55 1.56 1.57 1.58 1.59 1.60 1.61 1.62 1.63 1.64 1.65 1.66 1.67 1.68 1.69 1.70

0.7569 0.7627 0.7685 0.7743 0.7800 0.7858 0.7916 0.7974 0.8031 0.8089 0.8147 0.8205 0.8263 0.8320 0.8378 0.8436 0.8494 0.8551 0.8609 0.8667 0.8725 0.8783 0.8840 0.8898 0.8956 0.9014 0.9071 0.9129 0.9187 0.9245 0.9303 0.9360 0.9418 0.9476 0.9534 0.9591 0.9649 0.9707 0.9765 0.9823

21.467 21.631 21.795 21.959 22.122 22.286 22.450 22.614 22.778 22.942 23.106 23.269 23.433 23.597 23.761 23.925 24.089 24.253 24.417 24.581 24.745 24.908 25.072 25.236 25.400 25.564 25.726 25.891 26.055 26.219 26.383 26.547 26.711 26.875 27.039 27.202 27.366 27.530 27.694 27.858

1.71 1.72 1.73 1.74 1.75 1.76 1.77 1.78 1.79 1.80 1.81 1.82 1.83 1.84 1.85 1.86 1.87 1.88 1.89 1.90 1.91 1.92 1.93 1.94 1.95 1.96 1.97 1.98 1.99 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10

0.9880 0.9938 0.9996 1.0054 1.0112 1.0169 1.0227 1.0285 1.0343 1.0400 1.0458 1.0516 1.0574 1.0632 1.0689 1.0747 1.0805 1.0862 1.0920 1.0978 1.1036 1.1094 1.1152 l.l209 l.l267 1.1325 1.1383 l.l440 l.l498 l.l556 1.1614 1.1672 l.l729 l.l787 1.1845 l.l903 l.l96O 1.2018 1.2076 1.2134

28.022 28.186 28.350 28.513 28.677 28.841 29.005 29.169 29.333 29.497 29.660 29.824 29.988 30.152 30.316 30.480 30.644 30.808 30.971 31.135 31.299 31.463 31.627 31.791 31.955 32.119 32.282 32.446 32.610 32.774 32.938 33.102 33.266 33.429 33.593 33.757 33.921 34.085 34.249 34.413

·'The number of grams per cubic centimeter is the same as the specific gravity. For example. if the specific gravity is 1.47, that substance has a density of 1.47 gms/cm3. Factor used in converting to ounces per cubic inch Factor used in converting to grams per cubic inch

= Specific gravity multiplied by 0.5778. = Specific gravity mUltiplied by 16.387.

To compute: Specific gravity-multiply pounds per cubic foot by .01604. Pounds per cubic foot-multiply specific gravity by 62.4. Pounds per cubic inch-multiply specific gravity by .0361. One ounce equals 28.3495 gms. One gram equals .03527 93 oz.

101

102 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

C Reference line

o

Specific gravity, compound

40

40

30 25

30

20

20

070 060 050

05

25

A Specific gravity, resin or filler 50

C2

40 30 25

B Weiqht fraction

20 15 10 09 08 07 06 05

03 02

10 08 06 05 04 03

040 035 030

04 03

025 020

01

015

C1

010 009 008 007 006 005 004

Figure 2-23. A nomograph for determining the specific gravity of filled compounds by using various fillers and reinforcements.

Elasticity Elasticity is the ability of a material to return to its original size and shape after being deformed. TP elastomers (TPEs) and rubber both have excellent elasticity (see Chapter 6).

Plasticity Plasticity is the inverse condition of elasticity. A plastic that tends to stay in a shape or size to which it has been deformed has high plasticity. Plastics exhibit plasticity when stressed beyond their yield points (see Chapters 3-5). This accounts for the ability of some plastics to be cold formed. When TPs are heated to their softening point they have almost perfect plasticity.

Ductility Ductility is the ability of a material to be stretched, pulled, or rolled into shape without destroying its integrity. The different plastics provide a wide range of ductility, from very little (even zero) to extreme amounts, like the TPEs.

~

-=

Refractive index (nD) Abbe. value (v) dnldt x Io-'I"C Haze (%) Luminous transmittance (0.125-in. thickness) Critical angle (ic) Deflection temperature 3.6 F/min., 264 psi 3.6 F/min., 66 psi Coefficient of linear thermal expansion Recommended max. cont. service temp. Water absorption (immersed 24 hrs. at 73"F) Specific gravity (density) Hardness (0.25-in. sample) Impact strength (Izod Notch) Dielectric strength Dielectric constant 60HZ 1()6 Hz Power factor 60Hz 1()6 Hz Volume resistivity

Properties

ohm-cm

1018

0.05 0.03

D 150 D 257

3.7 22.2

198 214 3.6 198 0.3 1.19 M97 0.3-0.5 500

42.2

92

1.491 57.2 8.5 <2

D 150

ft.-Ib.lin. V/mil

in./in.1"F x 1Q-6 "F %

D696-44 D 570-63 D792 D 785-62 D 256 D 149-64

"F

% % degree

Units

D 648-56

D 1003 D 1003

DS42 D542

ASTM Method

Methyl Methacrylate (Acrylic)

>1016

0.0002 0.0002-O'(XlO4

2.6 2.45

ISO 230 3.5 180 0.2 1.06 M90 0.35 500

1.590 30.9 12.0 <3 88 39.0

Polystyrene (Styrene)

Table 2-24. Properties of Some Transparent Plastics

8 x 1016

0.0007 0.0075

2.90 2.88

280 270 3.8 255 0.15 1.20 M70 12-17 400

1.586 34.7 14.3 <3 89 39.1

Polycarbonate

101'

0.006 0.013

3.40 2.90

450

212 3.6 200 0.15 1.09 M75

1.562 35 14.0 <3 90 39.6

Methyl Methacrylate Styrene Copolymer

104 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Water Absorption Water absorption is the percentage increase in weight of a plastic due to its absorption of water. Standard testing procedures, such as twenty-four hours of immersion, are prescribed in ASTM standards. The tests are conducted for different lengths of time and at varying temperatures, as well as with different solutions. All this seeming redundancy is important with certain plastics, since water absorption can affect mechanical and electrical properties as well as the dimensions of parts. Plastics with very low waterabsorption rates tend to have better dimensional stability.

Water Vapor Transmission There are substantial differences in the rates at which water vapor and other gases can permeate different plastics. For instance, polyethylene is a good barrier for moisture or water vapor, but other gases can permeate it rather readily. Nylon, on the other hand, is a poor barrier to water vapor but a good one to other vapors. The permeability of plastic films is reported in various units, often in grams or cubic centimeters of gas per 100 sq. in. per mil of thickness (0.001 in.) of film per twenty-four hours. The transmission rates are influenced by such different factors, as pressure and temperature differentials on opposite sides of the film (see Chapters 3-6). The effectiveness of a vapor barrier can be rated in a term such as perms. An effective vapor barrier in buildings should have a rating no greater than, say, 0.2 perm. A rating of one perm means that one sq. ft. of the barrier is penetrated by one gram of water vapor per hour under a pressure differential of one in. of mercury. One in. of mercury equals virtually 0.5 psi; one gram is one seven-thousandth of a pound. A similar problem is presented by vehicle tires and certain blow-molded bottles, which must be virtually impermeable to air and other gases. The most impermeable of the rubbers is butyl rubber (see Chapter 6), though the carcass of a tire is not made from this rubber. Because of its impermeability to gases, butyl rubber is also used as a roof

Stiffness. Hardness. TenSile (Yield), Barner Properties, Chemical ReSistance, Abrasion ReSistance. Dlelectnc Constant. Softening Temperature

Elongation Impact Strength

DenSIty, gms Icc

0910

0920

0930

0940

0950

~LOW DenSlty~ ~Med,um Dens'ty~ 1I--. ...- - H I 9 h DenSity Figure 2-24. The effect of density on the properties of polyethylene.

I

0960 .,

I

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 105

coating. With plastic bottles, different layers of both coinjected and coextruded plastics (see Chapter 7) can be used to fabricate the bottle to make it impenneable to different vapors and gases (see chapters 4-5).

Toughness Toughness in regard to physical properties refers to a material's ability to absorb mechanical energy without fracturing. A tough material can absorb mechanical energy and undergo either elastic or plastic defonnation. Generally high-impact unfilled resins have excellent toughness. However, low- or moderate-impact resins can also display considerable toughness if the ultimate strength of a material is high enough, such as reinforced TS composites.

Hardness Hardness is closely related to strength, stiffness, scratch resistance, wear resistance, and brittleness. The opposite characteristic, softness, is associated with ductility. There are different kinds of hardness that measure a number of different properties (see Fig. 2-25). The usual hardness tests are listed in three categories: a) to measure the resistance of a material to indentation by an indentor; some measure indentation with the load applied, some the residual indentation after it is removed, such as tests using Brinell hardness, Vickers and Knoop indentors, Barcol hardness, and Shore durometers; b) to measure the resistance of a material to scratching by another material or by a sharp point, such as the Bierbaum hardness or scratch-resistance test and the Moh one for hardness; and c) to measure rebound efficiency or resilience, such as the various Rockwell hardness tests. The various tests provide different behavior characteristics for plastics, as described by different ASTM standards such as D 785. The ASTM and other sources provide different degrees of comparison for some of these tests, as in Figure 2-26. Some ductile plastics, such as PC and ABS, can be fabricated like metals with punching and cold-fonning techniques. These processing techniques are analogous to the hardness tests in that a rigid "indentor" is pressed into a sheet of a less-rigid plastic.

Hardness Scales

Durometer A

~

Rubber band

I nner tube

Auto tire tread

Figure 2-25. The hardness of different materials using different test methods.

106 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

10,000 Diamond

10

5000

2000 1000

500 110 200

100

Topaz

8

60

Ouartz

7

40

100

20

80

o

Orthoclase

6

Apatite

5

Fluorite

4

Calcite

3

Gypsum

2

Talc

1

Easily machined steels

Rockwell C

60 40 20

50

80

o

Rockwell B

140 120 100

Brasses and aluminum alloys

80 20 10

60 130 120 100-

5 Brmell h,mJness

Most plas! ICS

40 20 Rockwell M

80 60 40 Rockwell R

Mohs hardness

Figure 2-26. An approximate comparison of different hardness scales,

Brittleness Brittleness is simply lack of toughness. Plastics that are brittle frequently have lower impact and higher stiffness properties.

Notch Sensitivity Notch sensitivity, not to be confused with brittleness, is a measure of the ease with which a crack will propagate through a plastic from a preexisting notch, crack, internal void,

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 107

or sharp comer. Certain plastics are not at all notch sensitive, but others can be as sensitive as glass. Different tests are perfonned with variously shaped and sized notches to simulate different characteristics under Izod impact, tensile, and other test specimens. Notches have to be carefully molded in or machined in to ensure the best reproduction of test results. It is most important to control and identify any internal stresses that may exist in test specimens, because the results are directly related to such factors.

Lubricity Lubricity refers to the load-bearing characteristics of a plastic under conditions of relative motion. Those with good lubricity tend to have low coefficients of friction either with themselves or other materials and have no tendency to gall. RHEOLOGY AND DEFORMATION

Rheology is the science that deals with the defonnation and flow of matter under various conditions. The rheology of plastics, particularly of TPs, is complex but manageable. These materials exhibit properties that combine those of an ideal viscous liquid (that is, with pure shear defonnations) with those of an ideal elastic solid (with pure elastic defonnation). Thus, plastics are said to be viscoelastic. The mechanical behavior of plastics is dominated by such viscoelastic phenomena as tensile strength, elongation at breaks, and rupture energy, which are often the controlling factors. The viscous attributes of polymer melt flow are also important considerations in plastics processing and fabrication.

Viscoelasticity The flow of plastics is compared to that of water in Figure 2-27 to show their different behaviors. With plastics there are two types of defonnation or flow; viscous, in which the energy causing the defonnation is dissipated, and elastic, in which that energy is stored. The combination produces viscoelastic plastics. Viscosity is a material's resistance to viscous deformation (flow). Its unit of measure is Pascals· second (Pa·s) or pounds· second/in. 2 (lb·s/in.2). Plastic melt viscosities have a range from 2 to 3,000 Pa·s (glass 1020, water 10--1). The resistance to elastic defonnation is the modulus of elasticity (E), which is measured in Pascals (Pa) or pounds per square inch (psi). Its range for a plastic melt is 1,000 to 7,000 kPa (145 to 1,015 psi), which is called the rubbery range (see Figs. 2-13 and 2-28). Not only are there two classes of defonnation, there are also two modes in which defonnation can be produced: simple shear and simple tension. The actual action during melting, as in a screw plasticator, is extremely complex, with all types of shear-tension combinations. Together with engineering design, defonnation detennines the pumping efficiency of a screw plasticator and controls the relationship between output rate and pressure drop through a die system or into a mold (see Chapter 7).

Shear Rate When a melt moves in a direction parallel to a fixed surface, such as with a screw barrel, mold runner, or die wall, it is subject to a shearing force. As the screw speed increases,

..c:

0)

I

~T ~

0

-l

Low

preSsure) High

Figure 2-27. The rheology and flow properties of plastics differ. The volume of a so-called Newtonian fluid, such as water, when pushed through an opening is directly proportional to the pressure applied (the straight dotted line), the flow rate of a non-Newtonian fluid such as plastics when pushed through an opening increases more rapidly than the applied pressure (the solid curved line). Different plastics generally have their own flow and rheological rates so that their non-Newtonian curves are different.

-15

•

....

o

• • • •

~4

-

~

o

C3

-J

W iL

o

0 2 ~

5

o ~

• •• • ••

• • • • • • • • • • •

o

-J

• • • •

•

• • •

."

~ GLASSy....

• • •

• • • • •

. •

.

:........ :

:

• TRANSITION: . . RUBBERy . . . . . FLOW ....

1 ~--....-~~-.___~--.....-

100

120

......~-.........- - - -

140

TEMPERATURE (Oe) Figure 2-28. An example of the modulus of elasticity of plastics. 108

160

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 109

so does the shear rate, with potential advantages and disadvantages (see Fig. 2-29). The advantages of an increased shear rate are a less viscous melt and easier flow. This shearthinning action is required to "move" plastic (see Chapter 7). When water (a Newtonian liquid) is in an open-ended pipe, pressure can be applied to move it; doubling the water pressure doubles the flow rate of the water. Water does not have a shear-thinning action. However, in a similar situation but using a plastic melt (a non-Newtonian liquid), if the pressure is doubled the melt flow may increase from 2 to 15 times, depending on the plastic used. As an example, linear low-density polyethylene (LLDPE), with a low shearthinning action, experiences a low rate increase, which explains why it can cause more processing problems than other PEs in certain equipment. The higher-flow melts include polyvinyl chloride (PVC) and polystyrene (PS). A disadvantage observed with higher shear rates is that too high a heat increase may occur, potentially causing problems in cooling, as well as degradation and discoloration. A high shear rate can lead to a rough product surface from melt fracture and other causes. For each plastic and every processing condition there is a maximum shear rate beyond which such problems can develop. Shear in the channel of the screw is equal to -rrDN160h (where D = average barrel inside diameter, N = screw RPM, and h = average screw channel depth). This formula does not include the melt slippage between the barrel wall and screw surfaces, but the shear rate obtained is still useful for purposes of comparison. A 2! in. screw with a 0.140 in. channel rotating at 100 RPM results in a shear rate of 93.5 reciprocal seconds (rsec). This value is approximately the desired value in most extrusion processes, with 100 rsec generally being the target. The same formula can be used to determine the shear rate of slippage between the barrel and the screw. With a new barrel, which usually has a small clearance of 0.005 in., a high shear rate of about 2,618 rsec can exist. With this small clearance only a small amount of melt is subject to the higher heat, so that any overheating is overcome by melt mass it encounters (that is, mixes with). As the screw wears, more melt flows through enlarged clearances, but the shear rate is lower. The effect of wear on overheating is usually small and is not the main reason why the complete melt overheats. Shear rates can also be determined in melt flow through mold cavities and particularly in extrusion dies. The formulas applicable to the different-shaped dies usually do not

account for the slippage of melt on die surfaces, but they can be used to compare the process ability of melts and to control melt flow. The formula for a die extruding a rod is 4QI-rrR 3 , for a long slit 6Qlwh 2 , and for an annulus die 6QI-rrRh 2 (where Q = volumetric flow rate, R = radius, W = width, and h = die gap).

Molecular Weight Distribution Plastics are, as discussed, made up of molecules arranged in long, flexible chains. These chains become entangled with each other, entanglements that are largely responsible for high viscosity in melts. Shear can be envisioned as sliding molecules in rotation, which causes the chains to disentangle. At low shear, molecular chains become entangled, but as the shear rate increases they gradually disentangle, and the viscosity is reduced. The result, expressed as a so-called flow curve (see Fig. 2-27), is related to the processability of the plastic material. One method of defining plastics uses their molecular weight (MW), a reference to the plastic molecules' weight and size. Here MW refers to the average weight of a plastic, which is always composed of different-weight molecules. These differences are important

110 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Newtonian fluid t =1)y

o

Shear strain rate.

y

Figure 2-29. The relationship between shear stress and shear rate.

to the processor, who uses the molecular weight distribution (MWD) to evaluate materials. A narrow MWD enhances the performance of plastic products (discussed later). Melt flow rates are dependent on the MWD, as illustrated in Figure 2-30.

Melt Index Test The melt indexer (extrusion plastometer) is the most widely used rheological device for examining and studying plastics in many different fabricating processes. It is not a true viscometer in the sense that a reliable value of viscosity cannot be calculated from the flow index, which is normally measured. However, it does measure isothermal resistance to flow, using an apparatus and test method that are standard throughout the world. The standards used include ASTM D 1238 (U.S.A.), BS 2782-105C (U.K.), DIN 53735 (Germany), 1IS K7210 (Japan), ISO RI133/R292 (international), and others. In this instrument (see Fig. 2-31) the polymer is contained in a barrel equipped with a thermometer and surrounded by an electrical heater and an insulating jacket. A weight drives a plunger that forces the melt through the die opening, using a standard opening of 2.095 mm (0.0824 in.) and a length of 8 mm (0.315 in.). The standard procedure involves the determination of the amount of polymer extruded in 10 min. The flow rate (expressed in g/IO min.) is reported. As the flow rate increases, viscosity decreases. Depending on the flow behavior, changes are made to standard conditions (die opening size, temperature, etc.) to obtain certain repeatable and meaningful data applicable to a specific processing operation. The MI (melt indexer) is easy to operate and relatively low cost; thus, it is widely used for quality control and for distinguishing between members of a single family of polymers. Specifically, this MI makes a single-point test that provides information on

Narrow MWD Material

c ..... Q)"6,

-0'>.Q)

Cii ~ 3: ..c ..... r.... Ecu~ _ II) :J

..... C CJ

Q) Z ·cu-

.r:. 0 CJ

(a)

~

·in 0

()

en

:> (b)

... ".

i

E

/

Low

I

I

/

/

/

Wide or Broad ~ MWD Material

,

_~I

"

"

\

\

\

,,

Molecular Weight Distribution

High

t -

Distribution

>-

·in 0

Wide

()

en

~

~

Shear

:> (c)

Temperature Shear

~

Figure 2·30. Melt-flow rates as a function of molecular weight distribution. a) Molecular weight distribution (MWD) curves; b) viscosity versus shear rate, as related to MWD; and c) factors influencing viscosity.

TION

ORIFICE O.0825·INCH DIAMETE

PISTON, 3fa-INCH DIAMETER

O.315INCH LONG lat!:..~*D~:J

lM ·INCH FLUOROCAR

Figure 2·31. A melt index (MI) test per ASTM D 1238. 111

112 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

resistance to flow at only a single shear rate. Because variations in branching or molecular weight distribution (MWD) can alter the shape of the viscosity curve, the MI may give a false ranking of plastics in terms of their shear rate resistance to flow. To overcome this problem, extrusion rates are sometimes measured for two loads, or other modifications are made. Different companies produce MIs, as listed in different magazines and the literature of test equipment suppliers. In summary, the MI is an indicator of the average molecular weight (MW) of a plastic and is also a rough indicator of processability. Low MW materials have high MIs and are easy to process. High MW materials have low MIs and are more difficult to process, as they have more resistance to flow, but they are processable. End-use physical properties improve as the MI decreases (see Figs. 2-32 and 2-33). Because processability simultaneously decreases, MI selection for a given application is a compromise between properties and processability. Table 2-25 lists typical MI ranges for the more common plastics processes and materials. Materials with other MIs are still processable, but they usually require more sophisticated start-up procedures and process controls.

Elasticity As a melt is SUbjected to a fixed stress or strain, the deformation versus time curve will show an initial rapid deformation followed by a continuous flow (Fig. 2-34). The relative importance of elasticity (deformation) and viscosity (flow) depends on the time scale of the deformation. For a short time elasticity dominates, but over a long time the flow becomes purely viscous. This behavior influences processes: when a part is annealed, it will change its shape; or, with postextrusion (Chapter 7), swelling occurs. Deformation contributes significantly to process-flow defects. Melts with only small deformation have proportional stress-strain behavior. As the stress on a melt is increased, the recoverable

A

I ~

~ ~

8

Y " ~ /'

A. BARRIER PROPEK11ES HARDNESS TENSILE S11tENGt1t 0tEMICAI. RESISTANCE B. FU!XIBII1TY ELONGATlON C. RJGIDm'

CREEP RESISTANCE HEAT RESISTANCE

D. ClARn'Y REDUCED SHRINKAGE

E. SURFACE GLOSS

F. TOUGHNESS STRESS CRACK RESISTANCE

INCREASING MELT INDEX _ _ _ _ _ _ __

Figure 2-32. The effects of density and melt index (MI) changes on the properties of polyethylene (PE), with the properties increasing in the direction of the arrows.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 113

t

Ultimate Tensile Elongation

t

~

2

.

a..

~

:~:~I~~t~eens~!~ance

Chemical Resistance

Brittleness Temperature

Melt VISCOSity

Long Term Load Bearing Properties. Environmental Stress Cracking Resistance Melt Strength

~

10

12

14

16

18

20

Figure 2-33. The effect of the melt index (MI) on the properties of polyethylene.

strain tends to reach a limiting value. It is in the high-stress range, near the elastic limit, that processes operate. Molecular weight, temperature, and pressure have little effect on elasticity; the main controlling factor is MWD. Practical elasticity phenomena often exhibit little concern for the actual values of the modulus and viscosity. Although the modulus is influenced only slightly by MW and temperature, these parameters have a great effect on viscosity and thus can alter the balance of a process.

Flow Performance In any practical deformation there are local stress concentrations. Should the viscosity increase with stress, the deformation at the stress concentration will be less rapid than in the surrounding material; the stress concentration will be smooth and the deformation stable. However, when the viscosity decreases with increased stress, any stress concentration will cause catastrophic failure.

Flow Defects Flow defects, especially as they affect the appearance of a product, play an important role in many processes. Flow defects are not always undesirable, as, for example, in producing a matt finish. Six important types of defects can be identified, which are applied here to extrusion because of its relative simplicity. These flow analyses can be related to other processes and even to the complex flow of injection molding.

Table 2-25. Typical Melt Index Ranges for Common Polymer Processes

Test method: ASTM D 1238.

Process

MI Range

Injection molding Rotational molding Film extrusion Blow molding Profile extrusion

5-100 5-20 0.5-6 0.1-1 0.1-1

114 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Jk? ~

Cit

(8)

A

Time

A·B: Viscoelasticity with slow deformation B: Load removed B·C: Viscoelastic recovery

(d)

A

(b)

t

BEFORE LOADING

LOAD APPLIED

AFTER LOAD RELEASED

D' O-A: Instantaneous loading prOduces Immediate stram. A·B: Viscoelastic deformation (or creep) gradually occurs with sustained load. B-C: Instantaneous e/as/IC teC<M!lYoccurs when load is removed. C·O: Viscoelastic recovery gradually occurs; where no permanent detorma· tion (0') o· 'lith 8 permanent deformation (0"·0'). Any permanent deformation is related to type plastic, amount & rate of loading and fabricating procedure.

(8)

A

(e)

t F • VutabIe Load

Strain" Conetant O-A: Instantaneous loading prOduces immediate strain. A·X: With strain maintained gradual elastic retaxation occurs. X· Y: Instanteous deformation occurs when load is removed. Y·Z: ViscoelaStic deformation gradually occurs as resIdual stresses are relieved. Any permanent deformation IS related to type plastIC, amount & rate of loading and fabricatIng procedure.

Figure 2-34. Elasticity and strain. a) Basic defonnation versus the time curve; b) stress-strain defonnation versus time (the creep effect); c) stress-strain defonnation versus time (the stressrelaxation effect); d) material exhibiting elasticity; and e) material exhibiting plasticity,

Nonlaminar Flow Ideally, a melt flows in a steady, streamlined pattern in and out of a die. Actually, the extrudate is distorted, causing defects called melt fracture or elastic turbulence. To reduce or eliminate this problem, the entrance to the die is tapered or streamlined.

Sharkskin During flow through a die, the melt next to the die tends not to move, whereas that in the center flows rapidly. When the melt leaves the die, its flow profile is abruptly changed to a uniform velocity. This change requires rapid acceleration of the surface layer, resulting in high local stress. If this stress exceeds some critical value the surface breaks, giving the rough appearance called sharkskin. With the rapid acceleration, the deformation is primarily elastic, Thus the highest surface stress, and worst sharkskin, will occur in

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 115

plastics with a high modulus and high viscosity, or in high-molecular-weight plastics of narrow MWD at low temperatures and high extrusion rates. The addition of die lip heating, locally reducing the viscosity, is effective in reducing sharkskin.

Nonplastication This condition produces uneven stress distribution, with consequent lumpiness. The product could appear ugly or have a fine matt finish. With a wide MWD there could be a lack of gloss.

Volatiles Many plastics contain small quantities of material that boil at processing temperatures, or they may be contaminated by water absorbed from the atmosphere. These volatiles may cause bubbles, a scarred surface, and other defects. See Chapters 7 and 8 for methods of removing volatiles by using vented barrels and dryers.

Shrinkages The transition from room temperature to a high processing temperature may decrease a plastic'S density by up to 25 percent. Cooling causes possible shrinkage (up to 3 percent) and may cause surface distortions or voiding with internal frozen strains. As discussed in other chapters, this situation can be reduced or eliminated by special techniques, such as cooling under pressure in the injection-molding process (see Chapter 5).

Melt Structures High shear at a temperature not far above the melting point may cause a melt to take on too much molecular order. In tum, distortion could result. This subject is discussed further in this chapter and in Chapters 3 and 7.

Table 2-26. Performance Influenced by Melt Index and Density of Plastics With Increasing Melt Index Rigidity Heat resistance Stress crack resistance Penneation resistance Abrasion resistance Clarity Flex life Impact strength Gloss Vertical crush resistance Cycle Flow Shrinkage Parison roughness Parison sag Pinch quality Parting line difference

Decreases Decreases

Decreases Decreases Increases Decreases Increases Decreases Decreases Increases Increases

With Increasing Density Increases Increases Decreases Increases Increases Decreases Decreases Decreases Increases Increases_ Decreases Decreases Increases Increases Decreases Increases

116 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Thermodynamics With the heat exchange that occurs during processing, thennodynamics becomes important. It is the high heat content of melts (about 100 cal/g) combined with the low rate of thennal diffusion (10- 3 cm2/s) that limits the cycle time of many processes. Also important are density ,changes, which for crystalline plastics may exceed 25 percent as melts cool. Melts are highly compressible; a 10 percent volume change for a force of 700 kg/cm2 (10,000 psi) is typical. A surface tension of about 20 g/cm may be typical for film and fiber processing when there is a large surface-to-volume ratio.

Chemical Changes The chemical changes that can occur during processing include 1) polymerization and cross-linking, which increases viscosity; 2) depolymerization or damaging of molecules, which reduces viscosity; and 3) complete changes in the chemical structure, which may cause color changes. Already degraded plastics may catalyze further degradation.

Trends Because melts have made different properties and there are many ways to control processes, detailed factual predictions of final output are difficult to arrive at. Research and hands-on operation have been directed mainly at explaining the behavior of melts or plastics like with other materials (steel, glass, and so on). Modern equipment and controls are overcoming some of this unpredictability. Ideally, processes and equipment should be designed to take advantage of the novel properties of plastics rather than to overcome them.

INTERRELATING PROPERTIES, PLASTICS, AND PROCESSING As there are many different plastics, a number of techniques for defining and quantifying their characteristics exist. Important techniques that relate to processing are reviewed in Chapter 7. Molecular weight (MW), mentioned earlier, relates to the size of the molecules that make up a resin. These molecules are not of the same length or weight, and MW has a significant effect on processability and perfonnance. Resins with low MWs are easier to process but weaker and more brittle than those with high MWs. The latter are tougher, more chemically resistant, and so on, and require tighter process controls. Generally, processing the higher-MW plastics requires more energy in the fonn of temperature and pressure. Molecular weight distribution (MWD), also discussed earlier, is an indication of the relative proportions of molecules of different weights and lengths. It shows the breadth of distribution or the ratio of large, medium, and small molecular chains in the resin. If most of the molecules have about the same MW, the MWD is classified as narrow. A wide or broad MWD implies a large variation in MW. Fig. 2-30 compares wide and narrow MWDs. The MWD is independent of both density and the melt index (MI) and must be taken into account in considerations of both processing and product perfonnance. A narrow MWD enables much better, and narrower, process control. Two plastics with the same MI and density will process very differently if their MWDs are dissimilar. Most materials producers provide designers with understandable classification tenninology that is directly useful for designing the best part at the lowest price. Comparisons

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 117

of molecular differences between one plastic and another provide valuable tools for predicting variations in properties and for making judgments regarding the trade-offs involved among competing materials (Figs. 2-22 to 2-24 and 2-30 to 2-33).

Processing and Properties In order to understand potential problems and their solutions, it is helpful to consider the relationships of machine capabilities, plastics processing variables, and part performance (see Chapter 7). A distinction should be made between machine conditions and processing variables. Machine conditions are basically temperature, pressure, and processing time (such as screw rotation/rpm, and so on) in the case of a screw plasticator, die and mold temperature and pressure, machine output rate (lb.Ihr.), and the like. Processing variables are more specific as parameters than are machine conditions, as the melt in the die or the mold temperature, the flow rate, and the pressure used. The distinction between machine conditions and fabricating variables is a necessary one to avoid mistakes in using cause-and-effect relationships to advantage. If the processing variables are properly defined and measured, not necessarily the machine settings, they can be correlated with the parts' properties. For example, if one increases cylinder temperature, melt temperatures do not necessarily also increase. Melt temperature is also influenced by screw design, screw rotation rate, back pressure, and dwell times (see Chapter 7). It is much more accurate to measure melt temperature and correlate it with properties than to correlate cylinder settings with properties. The problem-solving approach that ties the processing variables to parts' properties includes considering melt orientation, polymer degradation, free volume/molecular packing and relaxation, cooling stresses, and other such factors. The most influential of these four conditions is melt orientation, which can be related to molded-in stress or strain [10-12]. Polymer degradation can occur from excessive melt temperatures or abnormally long times at temperature, called the heat history from plasticator to cooling of the part. Excessive shear can result from poor screw design, too much screw flight-to-barrel clearance, cracked or worn-out flights, and such. Orientation in plastics refers simply to the alignment of the melt-processing variables that definitely affect the intensity and performance of orientation.

Plastics with a Memory Thermoplastics can be bent, pulled, or squeezed into various useful shapes. But eventually, especially if heat is added, they return to their original form. This behavior, known as plastic memory, can be annoying. If properly applied, however, plastic memory offers interesting design possibilities for all types of fabricated parts. When most materials are bent, stretched, or compressed, they somehow alter their molecular structure or grain orientation to accomodate the deformation permanently, but this is not so with polymers. Polymers temporarily assume the deformed shape, but they always maintain the internal stresses that keep wanting to force the material back to its original shape. This desire to change shape is what is usually called plastic memory. This memory is often unwelcome. Sometimes we prefer for thermoplastic parts to forget their original shape and stay put, especially when the parts must be coined, formed, machined, or rapidly cooled. Occasionally, however, this memory or instability can be used advantageously.

118 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Most plastic parts can be produced with a built-in memory. That is, their tendency to move into a new shape is included as an integral part of the design. So after the parts are assembled in place a small amount of heat can coax them to change shape. Plastic parts can be deformed during assembly, then allowed to return to their original shape. In this case the parts can be stretched around obstacles or made to conform to unavoidable irregularities without their suffering permanent damage. The time/temperature-dependent change in mechanical properties results from stress relaxation and other viscoelastic phenomena that are typical of polymers. When the change is an unwanted limitation it is called creep. When the change is skillfully adapted to use in the overall design, it is referred to as plastic memory. Potential memory exists in all thermoplastics. Polyolefins, neoprenes, silicones, and other cross-linkable polymers can be given memory either by radiation or by chemically curing. Fluorocarbons, however, need no such curing. When this phenomenon of memory is applied to fluorocarbons such as TFE, FEP, ETFE, ECTFE, CTFE, and PVF z, interesting high-temperature or wear-resistant applications become possible. ORIENTATION

A plastic's molecular orientation can be accidental or deliberate. (Here, accidental refers to orientations that occur in processing plastics that may be acceptable. However, excessive frozen-in stress can be extremely damaging if parts are subject to environmental stress cracking or crazing in the presence of chemicals, heat, and so on). Initially the molecules are relaxed; molecules in amorphous regions are in random coils, those in crystalline regions relatively straight and folded. During processing the molecules tend to be more oriented than relaxed, particularly when sheared, as during injection molding and extrusion. After temperature-time-pressure is applied and the melt goes through restrictions (molds, dies, etc.), the molecules tend to be stretched and aligned in a parallel form. The result is a change in directional properties and dimensions. The amount of change depends on the type of thermoplastic, the amount of restriction, and, most important, its rate of cooling. The faster the rate, the more retention there is of the frozen orientation. After processing, parts could be subject to stress relaxation, with changes in performance and dimensions. With certain plastics and processes there is an insignificant change. If changes are significant, one must take action to change the processing conditions, particularly increasing the cooling rate. By deliberate stretching, the molecular chains of a plastic are drawn in the direction of the stretching, and inherent strengths of the chains are more nearly realized than they are in their naturally relaxed configurations. Stretching can take place with heat during or after processing by blow molding, extruding film, or thermoforming. Products can be drawn in one direction (uniaxially) or in two opposite directions (biaxially), in which case many properties significantly increase uniaxially or biaxially (see Table 2-27, Fig. 2-35 and Chapter 7). Molecular orientation results in increased stiffness, strength, and toughness (Table 228); as well as liquid resistance to liquid and gas permeation. crazing, microcracks, and others in the direction or plane of the orientation. The orientation of fibers in reinforced plastics causes similar positive influences. Orientation in effect provides a means of tailoring and improving the properties of plastics. Considering a fiber or thread of nylon-66, which is an unoriented glassy polymer, its modulus of elasticity IS about 2,000 MPa (300,000 psi). Above the Tg its elastic modulus drops even lower, because small stresses will readily straighten the kinked molecular

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 119

Table 2-27. Effects of Orientation of Polypropylene Films Stretch (%) Properties

None

200

400

600

900

Tensile strength, psi (MPa) Elongation at break, %

5,600 (38.6) 500

8,400 (58.0) 250

14,000 (96.6) 115

22,000 (152.0) 40

23,000 (159.0) 40

Tensile strength, psi (MPa) MD* TDt Modulus of elasticity, psi MD TD Elongation at break, % MD TD *MD

Uniaxial Orientation

Balanced Orientation

8,000 (55.2) 40,000 (276)

26,000 (180) 22,000 (152)

150,000 (1,030) 400,000 (2,760)

340,000 (2,350) 330,000 (2,280)

As Cast

Properties

5,700 (39.3) 3,200 (22.1) 96,000 (660) 98,000 (680) 425 300

300 40

80 65

= Machine direction.

tID = Transverse direction and direction of uniaxial orientation.

chains. However, once it is extended and has its molecules oriented in the direction of the stress, larger stresses are required to produce added strain. The elastic modulus increases. The next stop is to cool the nylon below its Tg without removing the stress, retaining its molecular orientation. The nylon becomes rigid with a much higher elastic modulus in the tension direction (15 to 20 x 103 MPa or 2 to 3 X 106 psi). This is nearly twenty times the elastic modulus of the unoriented nylon-66 glassy polymer. The stress for any elastic extension must work against the rigid backbone of the nylon molecule and not simply unkink molecules. This procedure has been commonly used in the commercial production of man-made fibers since the 1930s.

Table 2-28. The Effect of Molecular Orientation on the Impact Properties of Polypropylene Film ASTM Tensile Impact, ft.-lb./in. 2 Temperature Material Unoriented PP Oriented PP

Room

o

40 above test limit

500

High-Energy Fatigue Impact (55 lb. weight @ 50-in. height) (24.9 kg weight at 127-cm height) Material

Number of drops to failure

Steel Unoriented PP 41 x 103 psi tensile Oriented PP 28 x 103 psi, 32% elongation

12

130

120 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Graph (a)

f

>f-

-7

~ strength

a:::

----

·z

----------------------~

a:::

UJ

Cl.

o

~

<{

I U

UJ

::;:

"~';,mod"'~

elongation to fracture

- - - - - - - - - -..- - energy to break

INCREASING ORIENTATION

_

Graph (c) Graph (b) Glass transition temperature

Flexibility

I

Rubbery : state I

I I

Increasing temperature

l>

Increasing temperature

)0

Figure 2-35. The effect of orientation on the properties of plastics.

Another example of the many oriented products is the heat-shrinkable material found in flat or tubular film or sheets. The orientation in this case is terminated downstream of an extrusion-stretching operation when a cold-enough temperature is achieved. Reversing the operation, or shrinkage, occurs when a sufficiently high temperature is introduced. The reheating and subsequent shrinking of these oriented plastics can result in a useful property. It is used, for example, in heat-shrinkable flame-retardant PP tubular or flat communication cable wrap, heat-shrinkable furniture webbing, pipe fittings, medical devices, and many other products.

Directional Properties The following items are treated here briefly, since they apply to orientation, but they are used extensively throughout this book, particularly in Chapter 3. They are used in identifying all types of plastics, particularly the reinforced plastics and composites.

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 121 Polar Directional Properties

+. mDlthDlfaplc or Unldlrectla,,"1 1m Ildirection.1

o IlIIIlfaplc o. PI...,

o Unnlnloratd Plastics

DiHerent Fiber Orientations

Tensile Fracture Characteristics

Dlthotroplt ar Unldlrectl ... 1 v ~'""S ~I'lo

ft'

I SIr

8 diretllonal

•

~

I'lope:! ts DiStress

~I

I

Stress VS. Strain DIagrams at Various Angles

c

0

O.'lO

~--- Un' ~' reed

5lf1l.

ISIItroplc •• PI p. o.s

'A.

IS',

•

B~ r

Q

Klgh

I

Figure 2-36. Examples of the perfonnance of RP and composites with different orientations of their fiber reinforcements.

Isotropic Materials

In an isotropic material the properties at a given point are the same, independent of the direction in which they are measured. The term isotropic means uniform. As one moves from point-to-point in this type of homogeneous plastic the material's composition remains constant. Also, the smallest sample of material cut from any location has the same properties. A cast, unfilled plastic is a good example of a reasonably homogeneous

122 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 2-37. An example of the three-dimensional or anisotropic directional properties of wood that can also occur in unreinforced or reinforced plastics.

material. With reinforced plastics and other similar materials, isotropic refers only to the plane of the fiber layup; that is, it is only two-directional, rather than a complete isotropic behavior in three planes.

Anisotropic Materials In an anisotropic material the properties vary, depending on the direction in which they are measured. There are various degrees of anisotropy, using different terms such as orthotropic or unidirectional, bidirectional, heterogeneous, and so on (see Fig. 2-36). For example, cast plastics or metals tend to be reasonably isotropic. However, plastics that are extruded, injection molded, and so on and rolled metals tend to develop an orientation in the processing or machined direction. Thus, they have different properties in the machine and transverse directions, particularly in the case of extruded or rolled materials. Wood in anisotropic with distinct properties in three directions. Its highest mechanical properties are in the growth (fiber) direction, with the perpendicular (or second plane) direction having lower properties and the other perpendicular (or third plane) direction having much lower properties (see Fig. 2-37). During World War n, glass fiber-TS polyester resin composites were used in many high-performance, structurally loaded products. The reinforced plastics used many different glass-fiber woven constructions to produce the required directional properties. The design equations and engineering technology approaches used were based on the technology and engineering knowledge of the anisotropic wood performance based on centuries of wood applications in buildings, bridges, and the like. The Wright-Patterson Air Force Laboratories in Dayton, Ohio, and Forest Products Laboratory in Madison, WI, provided the original engineering equations and technical approach to designing with RPs. As covered in Chapters 3-5 three-directional woven fabrics are used in RPs to provide three-dimensional rather than the usual two-dimensional performance [2, 14-18].

THE STRUCTURE AND BASIC PROPERTIES OF PLASTICS 123

BEING MOLDED

(HIGHLY

MO: :V~~

EXAGGERATED) AFTER EJECTIONS & COOLING

I

/ ,/

IT' I

FLOW DIRECTION SHRINKAGE

......... "-

-'\

,

J

~

CROSS FLOW SHRINKAGE

......r-----

Figure 2-38. An example of directional shrinkage in an injection molding that could be related to the anisotropic performance of plastics during processing.

SHRINKAGE This overview has primarily involved mechanical properties, but anisotropy can also be used when referring to the way a material shrinks during processing, such as in injection molding (see Fig. 2-38) and extrusion. Shrinkage is an important consideration when fabricating plastics, particularly crystalline TPs or ones with glass fibers. The flow direction can have more shrinkage than the cross-flow direction. The control of shrinkage is made to meet design requirements by factors such as the design of the mold or die shape, the processing-machine controls, the change of product shape, and the type of plastics (see Chapters 3-6 and 9).

Chapter 3

PLASTICS: DESIGN CRITERIA

Design is essentially an exercise in predicting perfonnance. The designer of plastic parts must therefore be knowledgeable in such behavioral responses of plastics as those to mechanical and environmental stresses. This chapter presents important basic concepts of plastics, mechanical and electrical properties in particular that define their range of design behaviors. Along with the next two chapters, on environmental and structural considerations, it provides the background needed to understand perfonnance analysis and the design methods available to the perfonning designer and engineer, as well as for those less familiar with conventional design and engineering practices. Parts made of plastics can then be designed using the logical appr~ch applied to such other materials as steel, wood, glass, and concrete, which have their own specific techniques of analysis

[1-2,5-14,29-33,40-47, 79-86, 245-347]. Many plastic products seen in everyday life are not required to undergo sophisticated design analysis for their ability to withstand high static and dynamic loads, as forinstance containers, cups, toys, boxes, housings for computers, radios, televisions and the like, and nonstructural products of various kinds like buildings, aircraft, appliances, and electronic devices. For other products designing for production with any material can be complex. Designing is, to a high degree, intuitive and creative, but at the same time empirical and technically mechanical. An inspired idea alone will not result in a successful design; experience plays an important part, but it can easily be developed. An understanding of one's materials and a ready acquaintance with the relevant processing technologies (see Chapter 7) are essential for converting an idea to an actual design. In addition, certain basic tools are needed, such as those for computation and measurement, to ensure that the loads and forces the product is to absorb can be safely withstood. For these reasons design is spoken of as having to be appropriate to the materials of its construction, its methods of manufacture, and the stresses involved. Where all these aspects can be closely interwoven, plastics are able to solve design problems efficiently in ways that are economically advantageous. Here and elsewhere, the equations and calculations used are based on those found in the basic machine design and engineering handbooks [1, 85, 201-3, 245-69, 273-80]. These technical aids have been assembled so as to facilitate the designing of plastic parts to meet the required perfonnance requirements. Observe that the safety factors have been omitted from most calculations, because different designers working on varying products use the appropriate criteria for choosing SFs. In general, an SF initially of 1.5 to 2.5 is used, as is commonly used with metals. This subject is covered in detail in Chapter 5. 125

126 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The equations and calculations presented here are based on the development work of many people over the centuries in simplifying the testing and evaluation of all types of materials. Realistic approaches are thus available for comparing different materials and providing practical and useful guidelines for designing a whole range of products.

Design Analysis Methods Plastics and composites (RPs) have some mechanical characteristics that differ significantly from those of the familiar metals. Consequently, design analysts may have less confidence in these relatively new materials and in their own ability to design with them. Materials selection thus may tend to confine itself to familiar materials, or else products may be overdesigned, and failures may even occur in service, due to faulty design [1]. Also, the statistics available on materials are often presented so as to favor a particular bias, which complicates the process of assessing their relative merits and adds to the confusion. Essentially, what the design analyst requires is relevant and credible design data, together with valid methods for calculating, predicting, and optimizing a part's performance. These methods may involve design formulas and charts such as those included in computer-aided design (see Chapter 10) that provide an opportunity for plastics and composites to be handled on a basis equivalent to that of other materials. Materials selection depends on a wide variety and range of factors, including cost, technical suitability, safety, energy requirements in production and service, quality, the ability to be manufactured in the required quantities and to be satisfactorily assembled and finished, the influence of the service environment, recycling and waste recoverability, its estimated lifetime, and many other factors, which are often interrelated. All these elements must be considered before deciding that the use of a particular material makes good sense. Almost all the current methods for the design analysis of plastics are based on models of material behavior relevant to traditional metals, as for example elasticity and plastic yield. These principles are embodied in design formulas, design sheets and charts, and in the more-modern techniques such as those of computer-aided design (CAD) using finite element analysis (PEA). The design analyst is required only to supply appropriate elastic or plastic constants for the material, and not question the validity of the design methods. Traditional design analysis is thus based on accepted methods and familiar materials, and as a result many designers have little, if any, experience with such other materials as plastics, wood, and glass. Under these circumstances it is both tempting and common practice for designers to treat plastics and composites as though they were traditional materials and to apply familiar design methods with what seem appropriate materials constants. It must be admitted that this pragmatic approach does often yield acceptable results. However, it should also be recognized that the mechanical characteristics of polymers are different from those of metals, and the validity of this pragmatic approach is often fortuitous and usually uncertain. It would be more acceptable for the design analysis to be based on methods developed specifically for the new materials, but this action will require the designer of metals to accept new ideas. Obviously, this acceptance becomes easier to the degree that the new methods are presented as far as possible in the form of limitations or modifications to the existing methods discussed in this book. Caution in using such nonmetallic materials as plastics, composites, wood, and glass is proper in view of the penalties that may be incurred if parts fail in service. However,

PLASTICS: DESIGN CRITERIA 127

there has been in the past a tendency to resist the introduction of plastics and composites for other, less justifiable, reasons. It is thus reasonable to decide not to use them because of a personal lack of the necessary design information or techniques but less so through unWillingness to understand and apply existing information. Obviously, conservation and experience have their own important roles. Thus, materials such as polymers have to combat ignorance and false or misleading information. The latter is usually illustrated when the relative merits of different materials are being considered and different forms of presentation can either be misleading or can cause doubts that lead one to use familiar materials. Table 3-1 gives typical mechanical property data for four materials, the exact values of which are unimportant for this discussion. Aluminum and mild steel have been used as representative metals and polypropylene (PP) and glass fiber-TS polyester reinforced plastics (GRP) as representative plastics. Higher-performance types could have been selected for both the metals and plastics, but those in this table offer a fair comparison for the explanation presented. Also, it appears from the data that these metals are much stiffer and significantly stronger than the plastics. This approach to evaluation could eliminate the use of plastics in many potential applications, but in practice it is recognized that it is the stiffness and strength of the product that is important, not its material properties. To illustrate this correct approach, consider applications in which a material is used in sheet form, as in automotive body panels, and suppose that the service requirements are for stiffness and strength in flexure. First imagine four panels with identical dimensions that were manufactured from the four materials given in Table 3-1. Their flexural stiffnesses and strengths depend directly on the respective material's modulus and strength. All the other factors are shared in common with the other materials, there being no significantly different Poisson ratios. Thus, the relative panel properties are identical with the relative material properties illustrated in Figure 3-1. Obviously, the metal panels will be stiffer and significantly stronger than the plastic ones, based on the identical panel dimensions resulting in the use of equal volumes of materials. Obviously, the lower densities of plastics allow them to be used in thicker sections than metals, which can have a significant influence on panel stiffness and strength. For example, assume that the four panels have equal weights and therefore different thicknesses (t). When the panels are loaded in flexure, their stiffnesses depend on (Er3) and their strength on (
Table 3-1. Examples of the Mechanical Properties of Typical Metal and Plastic Materials

Property

Aluminum

Mild Steel

Polypropylene (PP)

Glass-fiber Reinforced Plastics (GRP)

Tensile modulus (E) l(f GN/m2 (psi) Tensile strength (0-) 103 MN/m2 (psi) Specific gravity (S)

70 (10)

210 (30)

1.5 (0.21)

15 (2.2)

400 (58)

450 (65)

2.7

7.8

GN/m2

= kPa.

40

(5.8) 0.9

280 (40.5) 1.6

128 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

their relative strengths by (cr/?) where s denotes specific gravity. These relative panel properties, illustrated in Figure 3-2, show that the plastics now appear in a much more favorable light. Figures 3-1 and 3-2 present the same basic data from Table 3-1, but in two different forms, and the superficial use of either form can be misleading. In practice, metals and plastics do not have to compete under either of the extreme conditions of equal volume or equal weight, and their positioning between these extremes will depend on the requirements of the particular application. For vehicle body panels, plastics may be used with thicker sections than their metals counterparts (that is, not of equal volume), but the desire to save weight will ensure that they not be used to the other extreme either. Thus the designer has the opportunity to balance out the requirements for stiffness, strength, and weight saving. For the materials data given in Table 3-1 a GRP panel having 2.4 times the thickness of a steel panel has the same flexural stiffness-but 3.6 times its flexural strength and only half its weight. The tensile strength of the GRP panel would be 50 percent greater than that of the steel panel, but its tensile stiffness is only 17 percent that of the steel panel. The designer's interest in this GRP panel would then depend in this context on whether tensile stiffness was what was required. No general conclusions should be drawn on the relative merits of various materials based on this description alone. These examples have been presented merely to illustrate the dangers of superficially interpreting property data and of making dogmatic or generalized statements about the relative merits of various classes of materials. Similar remarks could be made with respect to various materials' costs and energy contents,

~

200

400

~

-

N

N

E

Z

S 100 LU

E

z 200 ~ 0

Figure 3-1. The relative stiffness (open bars) and strength (shaded bars) of sheets made from the materials listed in Table 3-1.

..-

100

-

....

N

E

...~ WI

W 2

50

E

i

z

N

WI

b

Figure 3-2. The relative plate stiffness (open bars) and strength (shaded bars) in flexure for panels of equal weight.

PLASTICS: DESIGN CRITERIA 129

which can also be specified per unit of weight or volume. If these factors are to be treated properly, they too must relate to final product values, including the method of fabrication, expected lifetime, repair record, in-service use, and so on [1]. One important conclusion illustrated by the example given is that plastic products are often stiffness critical, whereas metal products are usually strength critical. Consequently, metal parts are often made stiffer than required by their service conditions, to avoid failure, whereas plastic parts are often made stronger than necessary, for adequate stiffness. Thus, in replacing a component in one material with a similar part in another material it is not usually necessary to have the same part stiffness and strength. It follows that general statements about energy content or cost per unit of stiffness or strength, as well as other factors, should be treated with caution and applied only where relevant. This overview identifies the need for using design analysis methods appropriate for plastics and composites. It also indicates the uncertainty of using with plastics methods derived from metals, and demonstrates the dangers of making generalized statements about the relative merits of different classes of materials. The designer who has basically no familiarity with plastics needs to be receptive to the different methods of handling them. It is necessary to keep an open mind when designing parts with plastics or composites, rather than limiting the design to being an exact replica of the metal part. Let us assume from this point on that this approach is accepted, so a more-detailed examination of the needs of the design analysis methods can follow.

Design Analysis Requirements It should be evident that the full spectrum of the possible materials and applications in load-bearing situations involves many factors that may have to be taken into account. Fortunately, most products involve only a few factors, and others will not be significant or relevant. Regardless, the methods of design analysis must be made available to handle any possible combinations of such factors as the materials' characteristics, the product's shape, the loading mode, the loading type, and other service factors and design criteria.

Materials Characteristics The wide choice available in plastics and composites makes it necessary to select not only between TPs, TSs, reinforced plastics and composites, and molding compounds but also between individual materials within each family of plastic types (see Chapter 6). This selection requires having data suitable for making comparisons which, apart from the availability of data, depends on defining and recognizing the relevant plastics characteristics. There can be, for instance, isotropic (homogeneous) plastics and reinforced plastics that can have different directional properties that run from the isotropic to anisotropic, as seen in Chapter 2. Here RPs can be used advantageously to provide extra stiffness and strength in predesigned directions.

Product Shape Design analysis is required to convert applied loads and other external constraints into stress and strain distributions within a product and calculate the associated deformations. The nature and complexity of these calculations will be strongly influenced by the part's shape. The designing will be simplified if the part approximates a simple engineering

130 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

fonn like a plate or shell, a beam or tube, or some combination of idealized fonns such as a box structure. In such cases standard design fonnulas can be used, with appropriate parameters relating to the factors being reviewed: short- and long-time loadings, creep, fatigue, impact, and so on, using viscoelastic materials. There are of course products whose shape does not approximate a simple standard fonn or where more detailed analysis is required, such as a hole, boss, or attachment point in a section of a product. With such shapes the component's geometry complicates the design analysis for plastics, glass, or metal and may make it necessary to carry out a direct analysis, possibly using finite element analysis (PEA). Loading Mode

Loads applied on products induce tension, compression, flexure, torsion, or shear, as well as distributing the loading modes. The product's particular shape will control the type of materials data required for analyzing it. Its shape will be detennined by the location and magnitude of the applied loads in regard to the position and nature of such other constraints as holes, attachment points, and ribs. Also influencing the design decision will be the method of fabricating the product. The load's magnitude and distribution can be difficult to specify, especially in a system composed of several interacting components of the product, and can control whether the nonlinear effects will be significant. Loading Type

The mechanical behavior of plastics on time-dependent applied loading can cause different important effects on materials viscoelasticity. Loads applied for short times and at nonnal rates, discussed later in detail, causes material response that is essentially elastic in character. However, under sustained loads plastics, particularly TPs, tend to creep, a factor that is usually included in the design analysis (creep data are covered below). Products can also be subjected to intermittent loading involving successive creep and recovery over relatively long time scales. It is not unusual, for instance, for creep defonnation arising during one loading phase to be only partly recovered in the unloading cycle, leading to a progressive accumulation of creep strain (see Figs. 3-3 and 3-4) and

Time~

Figure 3·3. An example of intennittent loading involving successive creep strain and recovery.

PLASTICS: DESIGN CRITERIA 131

Stress levels

---- 2000 p.s.i. -3000p.s.i.

3

;'

",'"

" .....

----,I

I

I I

I

I I \

c:

.~

iii

a.

lE

U

\

00~~~~~----~------~------~--Time, hr.

120

160

Figure 3·4. An example of changes in elasticity for engineering TPs involving one cycle of loading to unloading. The curves show the effects of stress and time under load on strain recovery after unloading.

possibly resulting in creep rupture. An analogue of creep behavior is the stress-relaxation cycle that can occur under constant strain, as seen in Chapter 2. This behavior is particularly relevant with push-fit assemblies and bolted joints that rely on maintaining their load under constant strain. Special design features or analysis may be required to counteract excessive stress-relaxation (see Chapter 11). In many applications, intermittent or dynamic loads arise over much shorter time scales. Examples of such products include chair seats, panels that vibrate and transmit noise, engine mounts and other antivibration parts, and road surface-induced loads carried to wheels and suspension systems. Plastics' relevant properties in this regard are material stiffness and internal damping, the latter of which can often be used to advantage in design. Both properties depend on the frequency of the applied loads or vibrations, a dependence that must be allowed for in the design analysis. The possibility of fatigue damage and failure must also be considered [1].

Other Service Factors Many other factors also discussed in tum influence product design by being directly related to design analysis: an aggressive temperature environment, humidity, ultraviolet light, fire, chemicals, radiation, and so on. Other such factors include thermal expansion and conductivity, electrical conductivity, various friction and wear properties, and the possible effects of processing and aging on materials' properties.

132 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Design Criteria The nature of design analysis obviously depends on having product-performance requirements. These requirements are basically controlled by the product's level of technical sophistication and the consequent level of analysis that can be justified costwise. The analysis also depends on the design criteria for a particular part. If the design is strength limited, to avoid component failure or damage, or to satisfy safety requirements, it is possible to confine the design analysis simply to a stress analysis. However, if a plastic part is stiffness limited, to avoid excessive deformation from buckling, a full stress-strain analysis will likely be required. Even though many potential factors can influence a design analysis, each application fortunately usually involves only a few factors. For example, TPs' properties are dominated by the viscoelasticity relevant to the applied load. Anisotropy usually dominates the behavior of long-fiber composites.

20 A. TP polyester with 30% fibers, E~1.1- x 10 psi (7.6 GPa) B. TP polyester with 15% fibers, E~670,000 psi (4.6 GPa) C. Acrylic, E~O,OOO psi (3.0 GPa) O. PC Dr PPO,E""350,000 psi (2.4 GPa) E. TP polyester~350,000 psi (2.4 GPa) F. Rigid PVC, E~O,OOO psi (3.0 GPa) 1':. HOPE foam (60 pcf, 955 kg/m 3), E:::120,000 psi (827 MPa) H. LOPE foam (57 pef, 915 kg/m 3, E;10,OOO psi (69 MPa) I. PUR foam (7 pcf,112KG/m3)

B 15

.>C

150

100

75

0 E

5

25 H 0

0 0

5

10

15

20

25

Strain, percent Figure 3-5. An example of a range in tensile strength, modulus of elasticity, and elongation of some thennoplastics with and without chopped glass fibers, by weight and type of reinforcement.

PLASTICS: DESIGN CRITERIA 133

Mild steel

Magnesium alloy Modulus Molybdenum

40 millIOn

Steel

30 million 10 million 65 million

Aluminum alloy Magnesium alloy

o

___ J ______ L-

o

0004

0008

0012

I

I

0016

0020

Siram,

In

I _ _____.L 0.024

0.028

0032

lin

Figure 3-6. An example of tensile stress-strain diagrams for some metals.

4 -

Ratio tensile strength (psi) to density (Ibs./cu. in.) x 106

--- Ratio tensile modulus of elasticity (psi) to density (Ibs./cu. in.) Xl0 6

3

/

2 1

1910 1920 1930 1940 1950 1960 1970 1980

1990 2000

Figure 3-7. The growth in and a forecast for the structural properties of reinforced plastics and composites with steel and aluminum.

MECHANICAL PROPERTIES Most plastics are used because they have desirable mechanical properties at an economical cost. For this reason their mechanical properties may be considered the most important of all the physical, chemical, electrical, and other considerations for most applications. Thus, everyone designing with such materials needs at least some elementary knowledge of their mechanical behavior and how it can be modified by the numerous structural factors that can be varied in plastics. Plastics have the widest variety and range of mechanical properties of all materials (see Figs. 3-5 to 3-7). They vary from basically liquids to soft rubbers (elastomers) to hard, rigid solids. A great many structural factors determine the nature of their mechanical behavior, such as whether it occurs over the

134 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

short term (less than a quarter-hour) or the long term. As a rule, design is based on certain minimum strength or minimum deformation criteria [1, 2, 10-22, 62-68, 245347]. Short-term testing is important for quality control, to ensure the constant properties of plastics production. In addition it provides the designer data that permit comparisons of one material with another. However, a true comparison is possible only if both sets of data were determined in exactly the same way. For example, the speed of loading tensile test specimens influences performance factors such as deformation (see Fig. 3-4). Also, comparing the impact resistance of a half-inch specimen with that of an eighthinch specimen will result in a different analysis of the material's properties. Thus, it is necessary to describe the exact testing conditions along with each set of data sheets. Finally, the data from short-term testing give the user an important overall picture of the material. Very short time, impact testing is also covered later (see Fig. 3-8). The long-term testing of certain plastics allows their strength properties to be identified rapidly. Three of the major control-test procedures for long-term testing and predicting product lifetime are creep, fatigue, and impact (see Figs. 3-9 and 3-10).

40r-------~------T.r------~------~

N

..,E 30~------~------+_------~~--~ ~ ~

c;, c

~

Vi U o

Co

§

20r-------~-------+__r~~~----~

10r-------~~~7'+_------~----~

o Notch lip radius (mm)

Figure 3-8. An example of very short-term impact strength, with a notched radius, for several thermoplastics.

PLASTICS: DESIGN CRITERIA 135

10 5

10

Tim. (s)

Figure 3·9. An example of long-term tensile creep curves at 20°C (49Of) for polypropylene (PP) and nylon (N). The numbers in parentheses refer to the stress level, in MPa.

35

5

28

4

21

3

x

2

....

cti

Cl. ~

vi Vl

Q) .... Ci5

Vl

vi Vl

14

Q)

Ci5

7

0 10·'

10 4

10:'

106

0 10-

Cycles to failure Figure 3·10. An example of long-term tensile fatigue curves for dry nylon 6 that is 4.5 nun (0.18 in.) thick, acrylic (PMMA) that is 6.4 nun (0.25 in.) thick, and fluoropolymer (PTFE) that is 6.6 nun (0.26 in.) thick. The test frequency is at 1,800 cpm.

SHORT·TERM BEHAVIOR This section introduces the behavior and response of both unreinforced and reinforced plastics under loads lasting usually only a few seconds or minutes up to a maximum of fifteen minutes. Such short-term tests are used to define the basic or reference designing and engineering properties of conventional materials. Such properties as tensile strength, compressive strength, flexural strength (the modulus of rupture), shear strength, and associated elastic moduli are often shown on the data sheets provided by suppliers of plastic materials and are in computerized data banks. The influence of such factors as time, temperature, additives and reinforcements, and molecular orientation on the basic behavior of these properties is discussed in turn [1, 2, 10-14,62-68,245-87]. For many engineering plastics that are treated as linearly elastic, homogeneous, and

136 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

isotropic, their tensile and compressive properties are considered to be identical. This eliminates the need to measure their properties in compression. Furthermore, if the tension and compression properties are identical, under standard beam-bending theory there is no need to measure their properties in bending. However, in a concession to the nonlinear, anisotropic nature of most plastics, these properties, particularly the flexural ones, are often reported on marketing data sheets. With few exceptions, the stress-strain behavior of plastics can be characterized in terms of the engineering stress (the load divided by the original specimen's cross-sectional area) rather than the true stress (the load divided by the specimen's reduced cross-sectional area that results from Poisson contraction or necking). However, some engineering and scientific investigations do deal with true stress and strain, particularly in regard to the softer plastics that deform significantly before failure. Unless otherwise noted, all references to stress in this book are in terms of engineering stress, which is the most conventional way that tests are conducted, used in design, and reported worldwide.

Tensile Properties The tensile test is the experimental stress-strain test method most widely employed to characterize the mechanical properties of materials like plastics, metals, and wood. From any complete test record one can obtain important information concerning a material's elastic properties, the character and extent of its plastic deformation, and its yield and tensile strengths and toughness. That so much information can be obtained from one test of a material justifies its extensive use. To provide a framework for the varied responses to tensile loading in load-bearing materials that occur, several stress-strain plots, reflecting different deformation characteristics, will be examined. The standard ASTM D 638 explains the internationally accepted method of conducting tensile tests and defines the terms generally used throughout the industry. This standard was used in part to develop the definitions that follow. The standard itself should be referred to for further details. Analyzing stress-strain curves in tension is usually done by constantly measuring the force that develops as a sample is elongated at a uniform rate of extension. Various such curves are shown in Figures 3-11 and 3-12.

Stress Stress is the tensile load applied per unit of the original cross-sectional area at a given moment. The standard unit of measure is in Pa (Pascal) or pounds per square inch (psi).

Strain Strain is the ratio of elongation or deformation to the gauge length of the test specimen, that is, the change in length per unit of original length. It is expressed as a dimensionless ratio; that is, mm1mm (in.lin.). As the strain is increased beyond the material's proportionallimit, the specimen's elastic limit is reached. In this portion of the curve stress is no longer proportional to strain. However, below the elastic limit the material's behavior is elastic; that is, once it is unloaded, its recovery from deformation is essentially complete and instantaneous. Stressing the specimen above its elastic limit results in a degree of permanent set, however, which is dependent on the amount of stressing. This nonrecoverable stressing is called plastic strain. This strain is usually associated with plastics, particularly the unreinforced TPs, but it is also seen in metals and other materials.

PLASTICS: DESIGN CRITERIA 137

A

r--------

I

I I

-..1--

U) U)

II!..... III

A a E· TENSILE STRENGTH AT BREAI< ELONGATION AT BREAI< B· TENSILE STRENGTH AT YIELD ELONGATION AT YIELD C· TENSILE STRESS AT BREAK ELONGATION AT BREAK D· TENSILE STRESS AT YIELD ELONGATION AT YIELD

STRAIN

Figure 3·11. Tensile designations according to ASTM D 638.

Elongation The increase in the length of a test specimen that is expressed as a percentage of its extensometer gauge length is called its percentage of elongation.

Yield The first point on a stress-strain curve at which an increase in strain occurs without any increase in stress is its yield point or yield strength or tensile strength at yield. Some materials may not have a yield point. A yield strength can in such cases be established by picking a stress level beyond the material's elastic limit. The yield strength is generally established by constructing a line to the curve where stress and strain is proportional at a specific offset strain, usually at 0.2 percent. The stress at the point of intersection of the line with the stress-strain curve is its yield strength at 0.2 percent offset.

Proportional Limit A material's proportional limit is the greatest stress at which it is capable of sustaining an applied load without deviating from the proportionality of stress to strain.

138 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 100 90

..J-tj

~~

II I Hard steel

80

'w c. M 0

.. Ui .. 'c

~ ~

~

::::>

70

.,..

60

l..-" ".

.....~ r- Soft steel

50 ~

40

-

/

Ultimate stress

B

/'

r--

~

......

.......

Breaking point_

./

~

30 ~-Elastic limit-beginning of yield strength, I I I I f I 20 lA-Proportional limit

.,..

10 0

i,.-"

'" Polycarbonate 'rielld 0.05

0.10

0.15

Unit Strain

12,500

10,000

.1:!

~

en

€

0.20

Ultimate strength (9500 psi) _____ Yield point (9000 psi)

7,500 5,000

0.25

{in'!in.!

/

Break Proportional limit ---------- (4000 psi)

2,500

0.2

0.4

0.6

0.8

1.0

1.2

Strain (in./in.)

Figure 3-12. Tensile stress-strain diagrams. Top: Hard and soft steels, polycarbonates. Bottom: Polycarbonates, on an extended scale, with specific characteristics usable in il design analysis.

Elastic Limit The elastic limit of a material is the greatest stress at which it is capable of sustaining an applied load without any permanent strain remaining, once stress is completely released.

Tensile Strength The maximum tensile stress sustained by a specimen during a tension test is its tensile strength. Again it is expressed either in Pas (Pascals) or pounds per square in. (psi).

PLASTICS: DESIGN CRITERIA 139

---

Yield point (proportional limit)

/

C

II) II)

.......

Engineering yield strength

V~tress

1'<.

I

Q)

Cf)

/

/

Tensile strength

.1 Strain I

0.2%

Strain, in./in. [em/em)

Figure 3-13. An example of the modulus of elasticity determined on the initial straight portion of the stress-strain diagram.

When a material's maximum stress occurs at its yield point this stress is designated its tensile strength at yield. When the maximum stress occurs at a break:, the designation is its tensile strength at break:. In practice these differences are frequently ignored, often resulting in confusion in designs as to whether or not, for example, work hardening or cold drawing occurs before failure. Modulus of Elasticity

Most materials, including plastics and metals, have deformation proportional to their loads below the proportional limit. Since stress is proportional to load and strain to deformation, this implies that stress is proportional to strain. Hooke's Law, developed in 1676, follows that this straight line (see Fig. 3-13) of proportionality is calculated as Stress Stram

--.- =

Constant

The constant is called the modulus of elasticity (E) or Young's modulus (defined by Thomas Young in 1807), the elastic modulus, or just the modulus. This modulus is the slope of the initial portion of the stress-strain curve, normally expressed in terms such as MPa or GPa (lQ6 psi or Msi). A material not loaded past its proportional limit will return to its original shape once the load is removed. However, some elastic materials do not necessarily obey Hooke's law and simply return to their original shape. In many plastics, particularly the unreinforced TPs, the straight region of the stressstrain curve is not linear or else the straight region of this curve is too difficult to locate. It then becomes necessary to construct a straight line tangent to the initial part of the curve to obtain a modulus called the initial modulus. Designwise, an initial modulus can be misleading, because of the nonlinear elasticity of the material. For this reason, a secant modulus (see below) is usually used to identify the material more accurately. Thus, a modulus could represent Young's modulus of elasticity, an initial modulus, or a secant modulus, each having its own meaning.

140 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Standard ASTM D 638 states that it is correct to apply the term modulus of elasticity to describe the stiffness or rigidity of a plastic where its stress-strain characteristics depend on such factors as the stress or strain rate, the temperature, and its previous history as a specimen. However, D 638 still suggests that the modulus of elasticity can be a useful measure of the stress-strain relationship, if its arbitrary nature and dependence on load duration, temperature, and other factors are taken into account. Secant Modulus

The secant modulus is the ratio of stress to the corresponding strain at any specific point on the stress-strain curve. As shown in Figure 3-14, the secant modulus is the slope of the line joining the origin and a selected point C on the stress-strain curve. This measurement is usually employed in place of a modulus of elasticity for materials where the stress-strain diagram does not demonstrate a linear proportionality of stress to strain or E is difficult to locate. Hysteresis Effect

The hysteresis effect is a retardation of the effect (strain) when a material is subjected to a force or load (see Fig. 3-15). Viscoelasticity

Plastics respond to stress with elastic strain. In this material, strain increases with longer loading times and higher temperatures.

Slope represents tangent, or Young's modulus ' - Slope represents secant modulus at strain C'

III III

CD "-

en

Proportional limit

A'

Strain

C'

Figure 3-14. A diagram describing the tangent modulus and the secant modulus.

PLASTICS: DESIGN CRITERIA 141 15 Proportional

';j

C&.

8

-'" ct

lim~

V/ ;;

10

.;

••..

,,

III

.! ';j

c: ~

5

o

, ,, , ,

~ ..,

,, ,..,

"",

,

",

",

",

",

.., ..,

o

•

..,

..,

..,

~

,,

.., .., ..,

.., ..,

..,

..,

2

,,

, ,,

,

,,3%

",

3

4

Strain, (%)

Figure 3·15. An example of recovery to near zero strain, showing that material can withstand stress beyond its proportional limit for a short time, resulting in different degrees of the hysteresis effect.

Area under the Curve Generally, the area under the stress-strain curve is proportional to the energy required to break the plastic. It is thus sometimes referred to as the toughness of the plastic (Fig. 3-16). However, there are types, particularly among the many fiber-reinforced TSs, that are hard, strong, and tough, even though their area is extremely small.

Poisson's Ratio Poisson's ratio is the proportion of lateral strain to longitudinal strain under conditions of uniform longitudinal stress within the proportional or elastic limit. When the material's deformation is within the elastic range it results in lateral to longitudinal strains that will always be constant. This ratio is designated by the Greek letter v. In mathematical terms, Poisson's ratio is the diameter of the test specimen before and after elongation divided by the length of the specimen before and after elongation. Poisson's ratio will have more than one value if the material is not isotropic. Poisson's ratio always falls within the range of 0 to 0.5. A zero value indicates that the specimen would suffer no reduction in diameter or contraction laterally during elongation but would undergo a reduction in density. A value of 0.5 indicates that the specimen's volume would remain constant during elongation or as the diameter decreases. For most engineering materials the ratio lies between 0.10 and 0.40 (see Table 3-2). Poisson's ratio is a required constant in engineering analysis for determining the stress and deflection properties of plastic, metal, and other structures such as beams, plates, shells, and rotating discs. Temperature, the magnitude of stresses and strains, and the

142 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK T

T

SOFT II< TOUGH

HARD. TOUGH

HARD II< STRONG

tr

HARD II< BRITTLE

SOFT II< WEAK

cr

cr

Figure 3-16. Tensile stress-strain curves for different plastics that relate the area under the curve to their toughness or physical properties.

direction of loading all have their effects on Poisson's ratio. However, these factors usually do not alter the typical range of values enough to affect most practical calculations, where this constant is frequently of only secondary importance. Crazing

When tensile stress is applied to an amorphous (glassy) plastic such as polystyrene, crazing may occur before fracturing. Crazes are like cracks in that they are wedge shaped and form perpendicular to the applied stress. However, they differentiate from cracks by containing plastic that is stretched in a highly oriented manner perpendicular to the plane of the craze, which is to say parallel to the applied stress's direction. Another major distinguishing feature is that unlike cracks, crazes are able to bear stress. Under static loading, the strain at which crazes start to form decreases as the applied stress decreases. In constant strain-rate testing crazes always start to form at a well-defined stress level. Ex tensometers

An extensometer is an instrument to monitor strain in the linear dimension of a test specimen while a load or force is applied to it. The automatic plotting of load with strain produces stress-strain curves.

PLASTICS: DESIGN CRITERIA 143

Table 3-2. An Example of the Range of Poisson's Ratio Material

Range of Poisson's Ratio

Aluminum Carbon steel Rubber Rigid thermoplastics Neat Filled or reinforced Structural foam Rigid thennosets Neat Filled or reinforced

0.33 0.29 0.50 0.20-0.40 0.10-0.40 0.30-0.40 0.20-0.40 0.20-0.40

Test Rates

The test rate or cross-head rate is the speed at which the movable cross-member of a testing machine moves in relation to the fixed cross-member. The speed of such tests is typically reported in cm/min. (in.lmin.). An increase in strain rate typically results in an increasing yield point and ultimate strength (see Fig. 3-17). For most rigid plastics the modulus (the initial tangent to the stress-strain curve) does not change significantly with the strain rate. For softer TPs, such as polyethylenes, the theoretical elastic or initial tangent modulus is usually independent of the strain rate. The significant time-dependent effects associated with such materials, and the practical difficulties of obtaining a true initial tangent modulus near the origin of a nonlinear stress-strain curve, render it difficult to resolve the true elastic modulus of the softer TPs in respect to actual data. Thus, the observed effect of increasing strain is to increase the slope of the early portions of the stress-strain curve (see Fig. 317c), which differs from that at the origin. The elastic modulus and strength of both the rigid and the softer plastics each decrease with an increase in temperature. While in many respects the effects of a change in temperature are similar to those resulting from a change in the strain rate, the effects of temperature are relatively much greater.

Symbols At this point let us summarize the more commonly used symbols used to describe tensile properties, as follows: Symbol P p a a, ac apm T

e 11

E

Be

Unit Load,lb Pressure, psi Stress, psi Tensile stress, psi Compression stress, psi Permissible stress, psi Shear stress, psi Strain, % Poisson's ratio Modulus of elasticity, psi Creep modulus, apparent modulus, psi

Symbol

Er

G L

R, r D,d s, h A J Jp ~

M. Z

Unit Relaxation modulus, psi Torsional modulus, psi Length, in. Radius, in. Diameter, in. Width, section thickness, in. Cross-sectional area, in. 2 Moment of inertia, in. 4 Polar moment of inertia, in.4 Bending moment, lb.lin. Torque lb.lin. Section modulus, in. 3

II

(a)

A

10

B

9

70

60

C

8 50 ";;; 7

a."

0-

M

~6

40

C

:is UJ

A = 20 in./min (S.5 mm/secl

ti;4

B = 0.2 in./min (.OS5 mm/secl

a::

C

v;-

'"UJ '"

30 ~

C = 0.002 in./min (.000S5 mm/sec)

3

:.:E

20

2 Eo " 350,000 psi (2.41 GPo) 0

3

0

10

5

4

6

STRAIN, E, %

(b)

~l.lU

~c: •

·

Low Speed

Medium Speed

Strain

increasing strain rate or decreasing temperature

(c)

"

~ UJ

a:: ....

'" STRAIN,

£

Figure 3-17. Examples of the influence of different test rates and temperatures on basic stressstrain behaviors of plastics. a) Different testing rates, per ASTM D 638, as shown for a polycarbonate; b) effects of tensile testing speeds on the shapes of stress-strain diagrams; c) a simplified version of the effects on curves of changes in test rates and temperatures. 144

PLASTICS: DESIGN CRITERIA 145

Compression Properties The majority of tests to evaluate the characteristics of plastics are perfonned in tension or flexure; hence, the compressive stress-strain behavior of many plastics is not well described. Generally, the behavior in compression per ASTM D 695 is different from that in tension, but the stress-strain response in compression is usually close enough to that of tension so that possible differences can be neglected (see Fig. 3-18). The compression modulus is not always reported, since defining a stress at a strain is equivalent to reporting a secant modulus. However, if a compression modulus is reported, it will generally be an initial modulus. A general rule is that the compressive strength of plastics is greater than its tensile strength. However, this is not generally true for reinforced TSs. The compression testing of foamed plastics provides the designer with the useful recovery rate (see Fig. 3-19). Many of the procedures in compression stress-strain testing are the same as in tensile testing, but in compression testing particular care must be taken to specify the specimen's dimensions. If a sample is too long and narrow, for instance, buckling may cause premature failure. To avoid this, designers should test a specimen with a square cross-section and a longitudinal dimension twice as long as a side of the cross-section.

Figure 3-18. A comparison of tensile and compression stress-strain behavior for thermoplastics.

Test stopped due to buckling of specimen Rate of strain recovery = 270 psi Density: 3.9 Ibcf Cross-section: 5.0 x 5.4 " Height: 6.75"

0.10

0.15 0.20 Strain (in./in.)

0.25

0.30

Figure 3-19. A compression test for rigid foamed insulating polyurethane (3.9 lb.lft. 3) in which almost one-half its total strain was recovered in one week.

146 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

At higher stress levels, compressive strain is usually less than tensile strain. Unlike tensile loading, which usually results in failure, stressing in compression produces a slow, indefinite yielding that seldom leads to failure. Where a compressive failure does occur catastrophically, the designer should determine the material's strength in the same way as with tensile testing-by dividing the maximum load the sample supported by its initial cross-sectional area in Pa (psi). When the material does not exhibit a distinct maximum load prior to failure, the designer should report the strength at a given level of strain (often 10 percent). Flexural Properties

Like tensile testing, flexural stress-strain testing according to ASTM D 790, determines the load necessary to generate a given level of strain on a specimen, typically using a three-point loading (see Fig. 3-20). The sample is characteristically 0.125 x 0.5 x 5 in. (0.25 x 0.5 x 5 in. for foamed material). The bar is supported across a two-inch span (or 4 inches for structural foam) with a load applied at its center. Testing is performed at a constant rate of cross-head movement, typically 0.05 in.!min. for solids and 0.1 in.! min. for foamed samples. Simple beam equations are used to determine the stresses on specimens at different levels of cross-head displacement. Using traditional beam formulas and section properties, the following relationships can be derived where Y is the deflection at the load point (refer to Fig. 3-19): Bending stress

0' ;;:::

3FL 2bh 2

Bending or flexural modulus

Using these relationships, the flexural strength, also called the modulus of rupture, and the flexural modulus of elasticity can be determined (see Table 3-3). The flexural modulus reported is usually the initial modulus from the load-deflection curve. (The flexural data can be useful in product designs that involve such factors as bending loads.) Significantly, a flexural specimen is not in a state of uniform stress. When a simply supported specimen is loaded, the side of the material opposite the loading undergoes the greatest tensile loading. The side of the material being loaded experiences compressive

Table 3-3. Examples of Polypropylene Thermoplastics as Neat, Fiber Reinforced, and Talc Filled, and Their Effects on the Flexural Modulus of Elasticity Unreinforced (neat)

40% Glass Fiber*

180,000 psi (1,240 MPa)

1,100,000 psi (7,600 MPa)

*GJass fiber and talc content are by weight.

40% Talc* 575,000 psi (3,970 MPa)

PLASTICS: DESIGN CRITERIA 147

stress (see Fig. 3-21). These stresses decrease linearly toward the center of the sample. Theoretically the center is a plane, called the neutral axis, that experiences no stress. The stress-strain behavior of plastics in flexure generally follows from the behavior observed in tension and compression for either unreinforced or reinforced plastics. The flexural modulus of elasticity is nominally the average between the tension and compression moduli, where they differ. The flexural yield point is generally that which is observed in tension, but this is not easily discerned, because the strain gradient in the flexural (RP) sample essentially eliminates any abrupt change in the flexural stress-strain relationship when the extreme "fibers" start to yield. APPLIED LOAD

1F

Figure 3-20. A three-point flexural test specimen bending fixture designed per ASTM D 790.

COMp,RESSIVE STRESS

/

Figure 3-21. An example of a flexural test specimen being subjected to compressive and tensile

stresses.

148 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The flexural strength for most plastics under standard ASTM bending tests is typically somewhat higher than their ultimate tensile strength, but flexural strength itself may be either higher or lower than compressive strength. Since most plastics exhibit some yielding or nonlinearity in their tensile stress-strain curve, there is a shift from triangular stress distribution toward rectangular distribution when the part is subject to flexure (see Fig. 3-22). This behavior is similar to that assumed for plastic design in steel and for ultimate design strength in concrete. Thus, the modulus of rupture reflects in part nonlinearities in stress distribution caused by plastification or viscoelastic nonlinearities in the crosssection. Shifts in the neutral axis resulting from differences in the yield strain, and postyield behavior in tension and compression can also affect the correlation between the modulus of rupture and the uniaxial strength results. Even plastics with fairly linear stress-strain curves to failure-for example, short-fiber reinforced TSs-usually display moduli of rupture values that are higher than the tensile strength obtained in uniaxial tests; wood behaves much like this. Qualitatively, this can be explained from statistically considering flaws and fractures and the fracture energy available in flexural samples under a constant rate of deflection as compared to tensile samples under the same load conditions. These differences become less as the thickness of the bending specimen increases, as would be expected by examining statistical considerations. Other methods of flexural testing that can be used are, for example, the cantilever beam method (see Fig. 3-23), which is used to relate different beam designs. It is used in creep and fatigue testing and for conducting testing in different environments.

tensile yield

A ----

neutral axis

----~::-

i

---

i

-----

neutral axis shift

-----------

l·compressiV: yield stress No Yield Unear Stress

Extreme Fibers Yield

I

Full Plastification of Cross Section

Figure 3-22. The elastic and plastic flexural behavior of unreinforced and reinforced plastics.

PLASTICS: DESIGN CRITERIA 149

o

flection

3 InCh6

C_~~6~~ Steel E . 30 106 P$ I (206 10 J MPa l

Aluminum

E ~ 10 (69

X 10 PSI 10J MPa l

Polystyrene

E- 5

{34

106 P$I 103 MPal

Figure 3-23. An example of the effect of the modulus of elasticity on elastic deflection for different materials, using cantilever test specimens. All the test beams have the same lengths and cross-sections.

Shear Properties Unlike the methods for tensile, compressive, or flexural testing, the typical procedure used for determining shear properties is intended only to determine the shear strength, not the shear modulus, of a material that will be subjected to the usual type of direct loading (see Figs. 3-24 to 3-26). Torsion pendulum and oscillatory rheometer techniques are used to determine the shear modulus. The shear strength values are obtained by such simple tests as those shown in Figure 3-24 for single shear and 3-25 and 3-26 for double shear. In these tests the specimen to be tested is sheared between the hardened edges of the supporting block and the block to which the load is applied. The shearing strength is calculated as the load at separation divided by the total cross-sectional area being sheared.

o 005 In

load

Table

Figure 3-24. A fixture with a test specimen for determining direct single shear.

150 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Load

Table

Figure 3-25. A simplified schematic of a fixture with a specimen being tested for direct doubleshear strength.

LOAD Jl'MALE PUNCH ,.--~I..-~

Figure 3-26. A direct-shear stress test with a circular specimen that is in current use by the industry, per ASTM D 732.

The use of the word direct in these tests might seem to imply that this is the only stress being placed on the specimen. However, an inspection of the test fixtures in these last three figures indicates that bending stresses do in fact· exist and the stress cannot be considered as being purely that of shear. Therefore, the shearing stress calculated above must be regarded as an average stress. This type .of calculation is justified in analyzing bolts, rivets, and any other mechanical member whose bending moments are considered negligible. Also, because strain measurements are difficult if not impossible to measure, few values of yield strength can be determined by testing. It is interesting to note that tests of bolts and rivets have shown that their strength in double shear can at times be as much as 20 percent below that for single shear. As mentioned, the values for the shear yield point (in kPa or psi) are generally not available; however, the values that are listed are usually obtained by the torsional testing of round test specimens. As noted, the data obtained using the test method above should be reported as direct shear strength. Designers are nevertheless cautioned to use the shear strength reported by this method only in similar direct-shear situations, because this is not a pure shear

PLASTICS: DESIGN CRITERIA 151

test. This test cannot be used to develop shear stress-strain curves or detennine a shear modulus, because a considerable portion of the load is transferred by bending or compression rather than pure shear. Also, the test results can depend on the susceptibility of the material to the sharpness of load faces. When analyzing plastics in a pure shear situation or when the maximum shear stress is being calculated in a complex stress environment, a shear strength equal to half the tensile strength or that given above is generally used, whichever is less. Basically, shearing stresses are tangential stresses that act parallel to the planes they stress. For example, the shearing force in a beam provides shearing stresses on both the vertical and horizontal planes within the beam. The two vertical stresses must be equal in magnitude and opposite in direction to ensure vertical equilibrium. However, under the action of those two stresses alone the element would rotate clockwise. Clearly, this pair of stresses must be negated by another couple. If the small element is taken as a differential one, the magnitude of the horizontal stresses must have the value of the two vertical stresses. This principle is sometimes phrased as "cross-shears are equal." In other words, a shearing stress cannot exist on an element without a like stress being located 90 degrees around the comer. The block diagram shown at the top of Figure 3-27 is subjected to a set of equal and opposite shearing forces Q. If the material is imagined as an infinite number of infinitesimally thin layers, as shown at the bottom, then there is a tendency for one layer of the material to slide over another to produce a shear fonn of defonnation or failure if the force is great enough [2]. This shear stress can be arrived at as follows:

'1'=

Shear load Area resisting shear

Q =A

The shear stress will always be tangential to the area upon which it acts. The shearing strain is the angle of defonnation v as measured in radians. For materials that behave according to Hooke's Law, shear strain is proportional to the shear stress producing it. Thus, Shear stress '1' ---- = - = Shear strain

v

C onstant

=

G

The constant G, called the shear modulus, the modulus of rigidity, or the torsion modulus, is directly comparable to the modulus of elasticity used in direct-stress applications. Only two material constants are required to characterize a material if one assumes the material to be linearly elastic, homogeneous, and isotropic. However, three material constants have by now been introduced: the tensile modulus of elasticity (E), Poisson's ratio (v), and the shear modulus (G). An equation relating these three constants, based on engineering's elasticity principles, follows: E

- = G

2 (1

+

v)

This calculation, which holds true for most metals, is generally applicable to injectionmolded TPs. However, the designer is already familiar from previous discussions with the inherently nonlinear, anisotropic nature of most plastics, particularly the fiber-reinforced and liquid-crystal ones.

152 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK (a)

Q ........t - - - - /

Q

SHEARING LOAD AREA

(b)

Q .....f - - - /

~---Q

~~

y (RADIANS;

SHEAR STRAIN

Figure 3-27. A basic analysis of shear stress. a) A material with equal and opposite shearing forces; b) a schematic of infinitesimally thin layers subject to shear stress.

It is important to note that a material such as plastics or wood that is weak in either tension or compression will also be basically weak in shear. For example, concrete is weak in shear because of its lack of strength in tension. Concrete beams like RPs, which are filament wound, are strengthened by reinforced bars specially placed to prevent diagonal tension cracking. Although no one has ever been able to determine accurately the resistance of concrete to pure shearing stress, the matter is not very important, because pure shearing stress is probably never encountered in concrete structures. Furthermore, according to engineering mechanics, if pure shear is produced in one member, a principal tensile stress of an equal magnitude will be produced on another plane. Because the tensile strength of concrete is less than its shearing strength, the concrete will fail in tension before reaching its shearing strength.

Torsion Properties As noted, the shear modulus is usually obtained by using pendulum and oscillatory rheometer techniques [2, 62-68, 93, 262, 279, 293]. The torsional pendulum (ASTM D 2236: "Dynamic Mechanical Properties of Plastics by Means of a Torsional Pendulum Test Procedure") is a popular test, since it is applicable to virtually all plastics and uses a simple specimen readily fabricated by all commercial processes or easily cut from fabricated products. The moduli of elasticity, G for shear and E for tension (in psi), are ratios of stress to strain as measured within the proportional limits of the material. Thus the modulus is really a measure of the rigidity for shear of a material or its stiffness in tension and compression. For shear or torsion, the modulus analogous to that for tension is called the shear modulus or the modulus of rigidity, or sometimes the transverse modulus.

PLASTICS: DESIGN CRITERIA 153

LONG-TERM BEHAVIOR With high-perfonnance plastics, dynamic loads such as creep, fatigue, and impact and related issues are important considerations in many designs (see ASTM D 4092). These materials' behaviors are influenced by many factors, including in particular temperature, time, previous stress history, and the ambient conditions. In order for these influencing factors to be examined separately from one another, test methods have been developed to pennit this separation of individual factors. The failure of a plastic product in the perfonnance of its nonnallong-time function is usually caused by one of two factors: excessive defonnation or fracture. For plastics it is more often than not found that excessive creep defonnation is the limiting factor. However, if fracture occurs, it can have more catastrophic results. Therefore, it is essential that designers recognize the factors that are likely to initiate fracture, so that steps can be taken to avoid them. Fractures are usually classified as either brittle or ductile. Although any type of fracture is serious, brittle fractures are potentially more dangerous, because there is no observable defonnation of the material prior to or during breakage. When the failure is ductile, however, large nonrecoverable defonnations become evident, which serve as a warning that all is not well. Plastics fractures are ductile or brittle depending on such variables as their polymer structure, additives, processing conditions, the strain rate, and the temperature and stress system. The principal external causes of fracture are a prolonged steady stress (creep rupture), the continuous application of a cyclically varying stress (fatigue), and the application of a stress (again called creep rupture). In all these cases the fracture processes can be accelerated if the plastic is in an aggressive environment.

Dynamic Mechanical Behavior Dynamic mechanical tests measure the response or defonnation of a material to periodic or varying forces. Generally, an applied force and its resulting defonnation both vary sinusoidally with time. From such tests it is possible to obtain simultaneously an elastic modulus and mechanical damping, the latter of which gives the amount of energy dissipated as heat during the defonnation of the material. The behavior of materials under dynamic load is of considerable importance and interest in most mechanical analyses of design problems where these loads exist. Unfortunately, most current engineering design is still based on the static loading properties of materials rather than their dynamic properties. Often this means overdesigning at best and incorrect design resulting in failure at worst. This problem has continued to exist because of insufficient basic knowledge and, most important, an understanding of the behavior of different materials in spite of significant advances made since the early 1940s. The complex workings of the dynamic behavior problem can best be appreciated by summarizing the range of interactions of dynamic loads that exist for all the different types of materials. Dynamic loads involve the interactions of creep and relaxation loads, vibratory and transient fatigue loads, low-velocity impacts measurable sometimes in milliseconds, high-velocity impacts measurable in microseconds, and hypervelocity impacts. Ideally, it would be desirable to know what the mechanical response would be to the full range of these dynamic loads for each material. However, certain load-material interactions have relatively more importance for engineering design, so that significant work on them exists. Metals that are uniquely under both static and dynamic loads can be cited as outstanding

154 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

cases. The continuum mechanical engineer and the metallurgical engineer have both found these materials to be most attractive to study. At the same time, metals, as compared to plastics, are easier to handle for analysis. Yet there is a great deal that is still not understood about metals, even in the voluminous scientific literature available. Of course, the importance of plastics and plastic composites has been growing steadily, requiring more dynamic mechanical behavior data to become available. In addition to dynamic loads, there is also materials' behavior under dynamic loads, each behavior having its own primary load environment. Material behavior can be classified as 1) ablation, creep, and relaxation behavior with a primary load environment of high or moderate temperatures; 2) fatigue, viscoelastic, and elastic range vibration or impact; 3) a fluidlike flow, as a solid to a gas, which is a very high velocity or hypervelocity impact; and 4) crack propagation and environmental embrittlement, as well as ductile and brittle fractures. Through all this activity emerges the fact that we have reviewed only one of four physical considerations, that of load. The other three factors are temperature, from low to high; pressure, from a vacuum to a high atmospheric; and atmosphere, whether humid air, a corrosive, or a gas or liquid. Basically, all these factors influence dynamic mechanical behavior. It would be endless to start obtaining more factual information and data, but it is sufficient to recognize that these data are available where designs require them. Unfortunately, most of the data are proprietary to different organizations. The usual approach to complex dynamic mechanical or other behavior is thus based on using available data to at least produce a product that can be tested mechanically under the required conditions. The designer can worry less about general mathematical consistency if it is not available, but should concentrate on applying the available engineering data correctly. Information and data on dynamic mechanical properties are available in the literature and CAD databases worldwide. The importance of these data in any structural design application is well known. Different methods of testing are used, with the necessary sophisticated equipment that is required becoming more available [62-68, 78, 252, 253]. Damping The dynamic mechanical behavior of plastics is of great interest and importance. For one thing, the dynamic modulus, or for that matter the modulus measured by any other technique, is one of the most basic of all mechanical properties, with its importance being well known in any structural application. The role of mechanical damping is, however, not as well known. Damping is often the most sensitive indicator of all kinds of molecular motions going on in a material, even in its solid state. Asid~m the purely scientific interest in understanding the molecular motions that can occur, analyzing these motions is of great practical importance in determining the mechanical behavior of plastics. For this reason, the absolute value of a given damping and the temperature and frequency at which the damping peaks occur can be of considerable interest and use. High damping is sometimes an advantage, sometimes a disadvantage. For instance, in a car tire high damping tends to give better friction with the road surface, but at the same time it causes heat buildup, which makes tires degrade more rapidly. Damping reduces mechanical and acoustical vibrations and prevents resonance vibrations from building up to dangerous amplitudes. However, the existence of high damping is generally an indication of reduced dimensional stability, which can be undesirable in structures carrying loads for long periods of time. Many other mechanical properties, including fatigue life,

PLASTICS: DESIGN CRITERIA 155

toughness and impact, and wear and the coefficient of friction are intimately related to damping [252, 253, 293].

Viscoelasticity The section that follows discusses two important types of long-term viscoelastic behavior: creep and stress relaxation. These forms of creep may occur over the life of a part or structure on a time scale as long as 100,000 hours or more (see Figs. 3-28 to 3-31).

Creep Products subjected to a given load develop a corresponding predictable deformation. If it continues to increase without any increase in load or stress, the material is said to be experiencing creep or cold flow. Creep is defined as increasing strain over time in the presence of a constant stress (see, for instance, Figs. 3-28 through 3-32). The rate of creep for any given plastic, steel, or wood material depends on the basic applied stress, time, and temperature [1, 2, 5, 11-14, 62-68, 268-69, 282-301].

Second stage

Fracture

(secondary creep)

(pnmar~

creep)1 iii

o

I I

I

I I

Initial stram

Figure 3-28. The basic concept for evaluating creep-test data.

---UHIGH LOAD

MEDIUM LOAD

~-----;LO'W LOAD

TIME, hours

Figure 3-29. Creep-test data deformation versus time-based deformation on three different loads and stresses.

156 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 1000

800~

';;;

.e'" ...'"
Vi

I

I

I

I

600 -

-

400 f-

-

200 r-

-

0

I

I

I

I

2000

4000

6000

8000

10,000

Time (hours) Figure 3-30. An example of a stress-relaxation curve with an initial applied stress approximating its yield stress,

O'c

~

c:

::J

...

Q.

'n; 'tit c.

O'B2

11\ 11\

~

C,,)

ca

~

tB

Time Figure 3-31. Typical creep and stress-rupture curves,

Attaining Creep Data

Creep-test specimens may be loaded in tension, compression, or flexure in a constanttemperature environment. With the load kept constant, deflection or strain is recorded at regular intervals of hours, days, weeks, months, or years. Generally, results are obtained at three or more stress levels, as was seen in Figure 3-29. Stress-strain-time data are usually presented as creep curves of strain versus log time. Sets of such curves, seen in Figure 3-32, can be produced by smoothing and interpolating data on a computer. These data may also be presented in other ways, to facilitate the selecting of information to meet specific design requirements. Sections may be taken

PLASTICS: DESIGN CRITERIA 157

through creep curves at constant times to yield isochronous stress versus strain curves or at a constant strain, giving isometric stress versus log-time curves as in Figure 3-32 [2]. Standard ASTM D 2990 provides details for different creep tests. If a designer is faced with decisions concerning creep, the most reliable source of information is a test program run under simulated or actual conditions on the part itself or at least on test specimens. The expected operating life of most products designed to withstand creep is usually ten to twenty years, however. It is apparent that actual longtime testing is not likely to be undertaken, so available creep test-data must be used. The so-called long-time tests are undertaken for at least 1,000 hours, the recommended time specified in the ASTM standard based on extensive data accumulated since 1943. The tests are performed under carefully controlled stress (load), temperature, time, and creep (elongation) conditions. To save time, tests for different constant loads are performed simultaneously on different specimens of the same material. Creep tests may be rather extensively conducted, as for example when developing creep data prior to the design and fabrication of the first all-plastic airplane [17]. The usual procedure is to plot the creep versus time curve, but other combinations are possible.

CREEP CURVES. STRAIN VS. LOG TIME

ISOMETRIC STRESS VS LOG TIME

g~

ISOCHRONOUS STRESS VS STRAIN

~----.:=---

Log time

Figure 3-32. A typical presentation of creep data.

Strain

158 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The theoretically shaped curve in Figure 3-28 provides the three typical stages for evaluation. An initial strain takes place almost immediately, consisting of the elastic strain plus a plastic strain, if the deformation extends beyond the yield point. This initial action in the first stage shows a decreasing rate of elongation because of strain hardening. The action most relevant to the designer concerns the second stage, which begins at a minimum strain rate and remains rather constant, because of the balancing effects of strain hardening and annealing. In the third stage a rapid increase in the creep rate is accompanied by severe necking (that is, thickness reduction) and ultimately rupture. Designers are concerned with the second stage in the sense that their target is not to have the product being designed enter into the third stage. Thus, after plotting the creep versus time data of the I,OOO-hour tests, the second stage can be extrapolated out to the number of hours of desired product life. This process is then followed for each of the creep curves. In making this extrapolation it is assumed that the 1,OOO-hour test has allowed the material to enter into the second stage. The material will have behaved similarly to that shown in the curve in Figure 3-28.

Creep Rupture.

Creep-rupture data are obtained in the same way as creep data except that higher stresses are used and the time is measured to failure (see Figs. 3-33 and 3-34). The strains are sometimes recorded, but this is not necessary for creep rupture. The results are generally plotted as the log stress versus log time to failure [2, 62-68]. In creep-rupture tests it is the material's behavior just prior to the rupture that is of primary interest. In these tests a number of samples are subjected to different levels of constant stress, with the time to failure being determined for each stress level. General technical literature and product data sheets seldom provide a complete description of a material's behavior prior to rupture, which to be considered so should include the development of crazing and stress whitening, its strain-time behavior, and the nature of the fracture process, describing yielding and necking and brittleness. As a result, a description of the rupture behavior of a specific plastic compound is not usually a handout. Figure 3-35 shows the curves of a family of thermoplastics describe a failure process that is fairly typical of the behavior of TPs. The time-dependent strains resulting from several levels of sustained or creep stress are shown, together with the development of

60
~

"-

en

l

1000 hour stress rupture strength iit l000"F [538CI 450

300

40

20

100

represents a test to failure

150

10,000

100,000

1000

TilTlt: to Failure, hours

Figure 3·33. Typical stress-rupture data versus temperature.

."

Q.

:a:

PLASTICS: DESIGN CRITERIA 159 60

50

\

40

-:xz

3)

OIl OIl

~

iii

20

"", --

I~ ~~ . , ....

N E

2% .....

........ 1%

.....

....

~

~ -, _--- ---................

". ................

...

~

~-

10

o

10

10J

~

, ~

~

11f

104

Log time to foilure

'\

lsI

Froc:ture - - - Whitening or c:rozing - - - - - - Isometric: curves

Figure 3-34. Typical creep-rupture ductile-to-brittle behavior of thennoplastics.

crazing and of stress whitening. The features that develop in the failure process follow a particular pattern:

Overall behavior. The time-dependent strain at which crazing, stress whitening, and rupture occur decreases with a decreasing level of sustained stress. The time to development of these defects increases with a decreasing stress level. Crazing. This develops in such amorphous plastics as acrylics, PVSs, and PCs as creep deformation enters the rupture phase. Crazes start sooner under high stress levels. Crazing occurs in semicrystalline plastics, but in those its onset is not readily visible. It also occurs in most fiber-reinforced plastics, at the time-dependent knee in the stress-strain curve. Stress whitening. This occurs in many types of plastics, including the amorphous ones like PVC and ABS, and in the crystalline types such as PE and PP. A stress-whitening zone may be a sign of crazing in some plastics where individual fine crazes may be difficult to detect. Stress whitening occurs fairly late in the rupture stage, just prior to yielding. Rupture. Rupture strain decreases steadily with increases in the duration of stress, as discussed. Alternately, the magnitude of stress needed to cause rupture decreases as the duration of stress increases. Figure 3-36 shows the development first of damage and then of yielding in a PVC compound as a function of its being under sustained stress. The decay at the onset of the first damage and of yield strength with the increasing duration of sustained stress is also

160 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

stress whitening

a I is the highest stress

a5 is the lowest stress

III

Z <{

0::

l-

V)

TIME, t, hrs - Log Scale Figure 3-35. Generalized strain versus time in thermoplastics.

90 ~80

10 ·iii

,....,Co 0

\\"::-C'. ,

..n V)

UJ

0:::

loHu.. ''neck;"9''

~60

--

,., ' ...... --.,-:~~:-.....................

0

l-

~70

5

V)

, ,

'

l·_·-

--.J ---- .... stress

......

W

h·Itenmg .

-~

0.001

0.01

0.1

10

102

103

40 30

0

~

..n V)

UJ

0:::

IV)

20

.-.,

10

craze initiation

v

50

c a.. :E

1{)4

0

TIME, t, hrs Figure 3-36. An example of stress versus time to damage and failure of polyvinyl chloride.

evident. In other words, a decrease in the magnitude of the sustained stress lengthens the time over which crazing, stress whitening, and yielding develop. Yielding is frequently taken as the failure criterion for plastics. However, some common types of standardized creep-rupture tests do not determine yielding, only the sustained stress and time to failure. One example of this is the ASTM D 1598 procedure for determining the time-dependent

PLASTICS: DESIGN CRITERIA 1 &1

Table 3-4. Rate of Strength Decay for Wood and Thermoplastics per Decade of Time Under Sustained Stress Range of Decay Rate, * % per Decade of Time

Material Wood

8

7-19 8-13 12-32

PVC

PE ABS

*Change in strength under sustained stress from beginning to end of decade, or unit change in log time.

burst strength of plastic pressure pipes. Plotting the failure or bursting stress against the time to failure for a given material defines its strength regression relationship. Comparisons have been made of the strength-regression characteristics of plastics with those of wood. The capacity of wood to resist sustained stress has been determined to decay at a rate of 8 percent for each decade of time change, that is, its capacity at the end of each decade is 92 percent of what it was at the start of the decade. The decay rates calculated from published strength-regression information on pressure-rated plastic pipe compounds are shown in Table 3-4. The decay rate for the specific plastics tested varies from 7 to 32 percent per decade, depending on the generic type of plastic and the specific compound within that type. The time-dependent strength behavior of some of these plastics is similar to that of wood.

Apparent Creep Modulus. The concept of an apparent modulus is a convenient method for expressing creep, because it takes into account the initial strain for an applied stress plus the amount of deformation or strain that occurs over time. Thus, the apparent modulus EA is calculated as Stress (psi) Initial strain + Creep Because parts tend to deform in time at a decreasing rate, the acceptable strain based on the desired service life of the part must be determined. The shorter the duration of load, the higher the apparent modulus and thus the higher the allowable stress. The apparent modulus is most easily explained with an example. As long as the stress level is below the elastic limit of the material, its modulus of elasticity E can be obtained from this equation:

E ==

Stress (psi) Strain (in.lin.)

For example, a compressive stress of 10,000 psi (69 MPa) gives a strain of 0.015 in. per in. (0.038 cm/cm) for FEP resin at 63°F (l7°C). Then E

10,000 0.015

667,000 psi (4,600 MPa)

162 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

INCREASING STRESS OR STRAIN

LOG TIME

Figure 3-37. An example of plotting the apparent creep modulus versus log time.

If the same stress level prevails for 200 hours, the total strain will be the sum of the initial strain plus the strain due to time. This total strain can be obtained from a creepdata curve. If, for example, the total deformation under a tension load for 200 hours is 0.02 in. per in., then E

=

10,000 n~

\'r.U15

= 5,000,000

.

pSI

(3,448 MPa)

.o~

Similarly, EA can then be determined for one year. Extrapolating from the creep-data curve, which is in fact a straight line, gives a deformation of 0.025 in. per in. Thus, E

=

10,000 0.025 = 400,000 psi (2,759 MPa)

When plotted against time, these calculated values for the apparent modulus provide an excellent means of predicting creep at various stress levels (see Fig. 3-37) [2]. For all practical purposes, curves of deformation versus time eventually tend to level off. Beyond a certain point, creep is small and may safely be neglected for many applications.

Stress Relaxation In a stress-relaxation test a plastic is deformed by a fixed amount and the stress required to maintain this deformation is measured over a period of time (see Fig. 3-38). The maximum stress occurs as soon as the deformation takes place and decreases gradually with time from this value. From a practical standpoint, creep measurements are generally considered more important than stress-relaxation tests and are also easier to conduct.

PLASTICS: DESIGN CRITERIA 163

Those interested in the theory of viscoelasticity and in the relationship of materials' properties to their molecular structure tend to concentrate more on stress-relaxation than creep measurements. One reason may be that stress-relaxation figures are generally more easily interpreted in terms of viscoelastic theory than are creep data. Stress-relaxation data also provide practical information such as determining the stress needed to hold a metal insert in a plastic product, evaluating the additives needed such as antioxidants, choosing cantiliver-type beams, and so on. The stress-relaxation behavior of plastics is extremely temperature dependent, especially in the region of the glass transition. Most amorphous types of plastics at temperatures well below the Tg have a tensile modulus of elasticity of about 3 X 1010 dynes/cm at the beginning of a stress-relaxation test. The modulus decreases gradually with time, but it may take years for the stress to decrease to a value near zero. Not only is the stressrelaxation behavior of an amorphous plastic most sensitive to temperature in its transition region, but at a given temperature in that region the stress changes rapidly with time. With crystalline plastics, the main effect of the crystallinity is to broaden the distribution of the relaxation times and extend the relaxation stress to much longer periods. This pattern holds true at both the higher and low extremes of crystallinity. With some plastics, their degree of crystallinity can change during the course of a stress-relaxation test. This behavior tends to make the Boltzmann superposition principle, explained below under "Materials and Processing," difficult to apply. Many designs incorporate the phenomenon of stress-relaxation. For example, in many products, when plastics are assembled they are placed into a permanently deflected condition, as for instance press fits, bolted assemblies, and some plastic springs (see Fig. 3-39). In time, with the strain kept constant the stress level will decrease, from the same internal molecular movement that produces creep. This gradual decay in stress at a constant strain, known as stress-relaxation, becomes important in applications such as preloaded bolts and springs where there is concern for retaining the load. The amount of relaxation can be measured by applying a fixed strain to a sample and then measuring the load with time [2].

.-. en oentl ....

..Ja: eLI-

~en

TIME,

TIME, (0 )

TIM , t

t

t

TIME, (b)

t

TIME,

t

b0M' (1'\

__

TIME,t

(e)

Figure 3-38. An example of strain behavior under various intennittent and cyclic loads. a) Recovery after creep; b) a strain increment caused by a stress step function; c) strain with stress applied 1) continuously and 2) intennittently.

164 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~

PRESS FIT LIGHT INTERFERENCE

bJ ---,/U\~

BOL TED ASSEMBL Y LIGHT ASSEMBLED STRESS

[I

W:

11

I

Figure 3-39. Examples of constant-strain loads.

The resulting data can then be presented as a series of curves much like the isometric stress curves in Figure 3-32 [2]. A relaxation modulus similar to the creep modulus can also be derived from the relaxation data. Generally, relaxation data are not as available as creep data. However, it has been shown that the decrease in load from stress relaxation can be approximated by using the creep modulus calculated from creep curves. Plastic parts with excessive fixed strains imposed on them for extended periods of time could fail. One example might be the splitting of a plastic tube press fitted over a steel shaft. Unfortunately, there is no relaxation-rupture corollary to creep rupture. For developing initial design concepts, a strain limit of 20 percent of the strain at the yield point or of the yield strength is suggested for high-elongation plastics. Likewise, using 20 percent of the elongation at the break is suggested for low-elongation brittle materials without a yield point, as only a guideline for initial design. Prototype parts should then be thoroughly tested at end-use conditions to confirm the design, or the available data on specific material of interest can provide more exacting limits.

Intermittent Loading The creep behavior of plastics that has been considered so far has assumed that the level of the applied stress will be constant. However, in service the material may be subjected to a complex pattern of loading and unloading cycles (see Fig. 3-38). This variability can cause design problems in that it would clearly not be feasible to obtain experimental data to cover all possible loading situations, yet to design on the basis of constant loading at maximum stress would not make efficient use of materials or be economical. In such cases it is useful to have methods for predicting the extent of the accumulated strain that will be recovered during the rest periods after a number of cycles of load changes. Recovery is the strain response that occurs upon the removal of a stress or strain. The mechanics of the recovery process are illustrated in Figure 3-40, using an idealized viscoelastic model. The extent of recovery is a function of the load's duration and time after load or strain release. In the example of recovery behavior shown in Figure 3-40 for a polycarbonate, samples were held under sustained stress for 1,000 hours, and then the stress was removed from the same amount of time. The creep and recovery strain measured for the duration of the test provided several significant points.

PLASTICS: DESIGN CRITERIA 165

First, the sample, which was loaded to about 20 percent of its short-tenn yield strength, or 13.8 MPa (2,000 psi), recovered almost completely one hour after the release of the load, the net strain being 0.03 percent. Second, the sample loaded to 66 percent of its short-tenn yield strength, or 41.4 MPa (6,000 psi), retained a strain of 0.8 percent at 1,000 hours after the release of the load. The initial strain was 2.8 percent, the strain from the 1,000-hour creep an additional 1.7 percent. Thus, only about half the creep strain was recovered. Visually extrapolating the recovery curve reveals that even after a year (104 hr.), about one third of the creep strain (0.6 percent) will remain. The first damage developed during creep or relaxation also affects recovery behavior. If the first damage is prevented by limiting the magnitude and duration of the stress, recovery will eventually be substantially complete for all practical purposes. Conversely, at strains above the first damage limit recovery will be incomplete and pennament deformation should be expected and accounted for in the evaluation. This is true not only for plastics in general but also of reinforced plastics. When RPs are stressed beyond the knee in their stress-strain curve, recovery becomes incomplete and hysteresis is clearly evident.

load

10o

c:

~

rtt

I--- I'

r-::: '2obQ psi

7

,

'"

~

\ \

3QQ2 ps.!.- ~ \ \

~

I' ,

5

~

--"

~

-+-r

.~

en

\

~~ ~ ~\

2

Recovery

,

~

3

w

---'

Creep

5

~

Removed

~

, " .....

.......

\

10-'

"'\.

5 3

10-'

100

10'

102

---

1()3

III

r.

...

~ 10-'

"

'"'\.,

~

7

r-...

" ~ ". i'...

\ -, --L

2

........

...........

--1..l.

3

r---.....

10" 10'

102

10' 10'

Time (hours)

Figure 3-40. Tensile creep and recovery during the intermittent loading of MerIon polycarbonate at 23°C (73"F). Courtegy, Mobay Chemical Corp.

166 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Materials and Processing As covered particularly in Chapters 2, 6, and 7, the various material types and compositions, as well as their processing methods, influence their properties, including creep. In general, crystalline materials have lower creep rates than the amorphous types, and RPs as a whole have significantly improved creep resistance. Some examples of creep data are given in Figures 3-41 through 3-54 and Tables 3-5 and 3-6. The data show all kinds of creep behavior, including the effects of time and temperature on amorphous materials that basically have a curve spread over a much wider time scale than that of the crystalline types? If the temperature is well below the TIP only the first part of the curve will be observed, for it might require years or even centuries to observe the complete curve. In the transition region nearly the complete curve can be observed, in a period from a few seconds to a few hours. If the temperature is well above the Tg , only the upswing in the curve (that is, an increase in the creep) will be observed, unless measurements can be made in a fraction of a second. Not only do the creep properties of crystalline polymers change rapidly with temperature, but in some cases at a given temperature a crystalline type will creep more with time than will the rigid amorphous or cross-linked types. However, a crystalline type above its Tg creeps very little, compared to the others. Thus, crystalline types tend to have an even broader distribution of retardation times than do the amorphous types. (Remember that the term crystalline refers to polymers that are actually semicrystalline.) At small loads the compliance of most materials at a given time is independent of stress. For example, doubling the load doubles the deformation. At higher loads, especially those approaching that which is required to break the plastic, compliance at any given time increases with the load. This effect is generally most pronounced with the crystalline types, the tough polyblends, and the amorphous types in the transition region or above it. However, the rigid types like polystyrene and the highly cross-linked phenol-formaldehyde plastics also show creep elongation, which increases at a rate greater than the first power of the stress at high loads. As a result, doubling the stress more than doubles the amount of elongation. The load or stress has another effect on the creep behavior of most plastics. The volume of an isotropic or amorphous plastic increases as it is stretched unless it has a Poisson ratio of 0.50. At least part of this increase in volume manifests itself as an increase in free volume and a simultaneous decrease in viscosity. This decrease in tum shifts the retardation times to being shorter. 2.5,----,----r-----r-------,-----, 2.0

Polycarbonate

0.5

o Time (hours) Figure 3-41. Tensile creep curves for three thermoplastics.

8000

10,000

PLASTICS: DESIGN CRITERIA 167 0.06 0.05 23"C, 105 MPa

E

E 0.04

E E

.~ 0.03 ~

I

~

~.

I

150 ·C, 70 MPa

~

~

23·C,70MPa

0.02 I"'""

0.D1

o

10

20

30

40

50

60

70

80

90

100

Time,h

Figure 3-42. The tensile creep of unreinforced polyamide-imide. Courtesy, Amoco Perfonnance Products, Inc.

Time to rupture, hr

Figure 3-43. Stress-rupture data for rigid 2-in.-diameter PVC pipe as a function of temperature.

A creep test can be carried out with an imposed stress, then after a time have its stress suddenly changed to a new value and have the test continued. This type of change in loading allows the creep curve to be predicted. The simple law referred to earlier as the Boltzmann superposition principle, holds for most materials, so that their creep curves can thus be predicted [1, 2, 14,94,262,293-301]. The first assumption involved in using the Boltzmann superposition principle is that elongation is proportional to stress, that is, compliance is independent of stress. The second assumption is that the elongation created by a given load is independent of the elongation caused by any previous load. Therefore, deformation resulting from a complex loading history is obtained as the sum of the deformations that can be attributed to each separate load.

2.8 r---r----,------,---r-----,r----,----.--,---, 0.40

'"

0.30

2.1

a.. Cl vi

vi

::J

"3

::J

"C

0

E

"C 0

0.

11>

E u

0.

1.4

0.20

~

U

C I!?

C

'"0.0.

I!?

'"

0. 0. oC(

oC(

0.7

0.10

O~~---~--~~-~--~--~---~-~--~O 10 20 100 40 60 200 400 600 103 Time,h

Figure 3-44. The flexural creep modulus for specific thermoplastics at 66°C (l50°F) and 6.9 MPa (one ksi).

10· 9

8

7

6 5 4

3

20

'iii

Q.

en ~ f/l en

en 2 en ~

U5

103

9 8

5

7

6 5

4

10-2

10-'

100

10·

I

lOs 5 x 104

Time (hours) Figure 3-45. The tensile stress-strain-time correlation resulting from creep for polycarbonates at 23°C (73°F). Courtesy, Mobay Chemical Corp.

168

0

~

"3

E

'iii ..,0.

10'

8 .

6

5 iii

0.

-;

'"'"~

4

-- ---:---

9

-.:::..::.: ~

~

----

-- -... -... -

ill

60

I

(73s o F)

--- -..:::..: 23"C~ '---

50

-_ J~3~uF)

°c 114OoF) ~

40

BoaC I ~ --=.:: ~'_, 140°F)

- -......

en

.S? iii c

23"C

r--.

--_

80~

I

I ...... :)...

'00 oC( 2 1"-.... 21 O

3

~

Failure by Fracture - Yield ___

---

2

10 10·'

10·'

100

10' 10' Time (hours)

103

10'

Figure 3-46. Time-to-fracture lines for polycarbonates at different temperatures. Courtesy, Mobay Chemical Corp.

8000r---+---+---~--~---+--~----~~

hrs

10.2 50

7~r---~--+---+---+---1---~--~~~

...;

~6~r---~--~--~----r---~~-+~~~~


S

:;)

!!!.

o ~ 5~b---+---~---4--~~~~~~~~~n2l ~

en

~4~r---t---i-~~~~~~~~t=~~~

in

'f2

30 in (f)

cCI)

~~r---+---~~~~~---;----r---+-~

(f)

0

~

't1 ~

2~ I---+-.I!H

1~r-~~--+---+---+---4---~--~--~

o

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Strain c (%)

Figure 3-47. An example of isochronous stress-strain curves for polycarbonates resulting from stress relaxation. 169

loo0~--~----~----r---~----'-----~~ 8oo~---+----~----r----+----1-----+-~ 6oor----+----~----

4 3

s::

0 2 c.

c:: c:: (J)

...m G)

;R loor----+----1-----+----+-----r----+_~ 80r----+----1-----+----+-----r----+_~

0.5

60r----+-----r----+----+-----r----+_~

40~--~----~----~--~~--~----~~

10-2

10-'

100

10'

102

103

10'

Time (hours)

Figure 3·48. The relaxation modulus of polycarbonates. Courtesy, Mobay Chemical Corp.

10000 8000 I-6000

60

r-- 1-- --=-~

230

50

C(73f2:1-~

30

:=4000 (J) a. o

~

:::J

!!!.

......

(J) (J)

40

e

~

en~ 2000

20 iD

~

0.5% 23°C (73.5 0F)

CD

(J) (J)

a

'iii

c::

lOS::

Q)

;R

I-

1000 800 600 400

1~

1~

l()O

I~

1~

Time (hours)

1~

1~

Figure 3·49. The stress relaxation of the glass-fiber-reinforced polycarbonate Merion 9310. Courtesy, Mobay Chemical Corp. 170

6 min.

6.-------------~--~~----~~--~--~~

rs

4

I.r

z

< a:: ......

II')

2

6 min. 10

O.I[-=~_.....-=::::::::::::::::::"--J 0.1

10

10 2

103

104

105

TIME, t, hrs Figure 3-50. The tensile creep behavior of a PP copolymer. a) Total creep strain versus time on a semilog scale; b) total creep strain versus time on a log-log scale.

171

Creep Modulus (lbs.lin.2) 5.000 ,...-- -- - - - - -- -- - - - - - - , 4.000 3.000 2.000

Figure 3·51. An example of creep resistance perfonnance for TPs and TSs at 23°C (73°F) and 100 hours with an applied stress of about 2,000 psi.

1.00r--------,.--------,:;_ - - - - - - ,

CREEP _ _-1 FLEXURAL 73 °F

O.75 1-------+r---;;;;:::;-::=~==~_;_;;:;:v;_::::i

__

~~-:-"t"------------ 30% glass bead/nyton 6/6. 1.250 psi

OL_______~------~----~ 10

10"

10"

10'

1lme

Figure 3·52. An example where the creep rate is related inversely to the reinforcements and filler content (ICI-LNP).

O.7 5 . - - - - - - -""'T""- - - - - -r-- - - - - - - , FLEXURAL CREEP

73°F

l

i

O.501---- - - - - =:;;i;:;___-::::::;;=:~=:;:-~-;."; -~ -~-~-~ -~ ---~---::=:.:;:~:';; -----------

~LO---------lo"~--------~lo"~------~1~ 1lme(h)

Figure 3·53. An example where the creep rate is also related directly, but not proportionately, to stress (ICI-LNP). 172

(a)

Laminates in °wet condition tes ted at 23 C In water and parallel to warp Values indicated on curves are percentages of short term ultimate . stress -Approximate strai at rupture

'028 '024

c

-

·020

'-

a

'-

CJ)

·012 -008 0-001

0{)1

0·1

1

10

102

Time, hours

(b)

Laminates in wet condition tested at 23°C in water and parallel to warp - - - - - 4 Values indicated on curves are percentages of short term ultimate stress

·012 ·010

.-c0

'-

CJ)

·006 ·00 '002 0'001

Approximate strain at rupture 0·01

01

1 10 Time,hours

Figure 3-54. An example of tensile creep curves in the direction of maximum fiber orientation. a) A TS polyester RP having 56 percent E-glass, by weight; b) glass-fabriclTS polyester RP in 48 percent glass by weight. 173

174 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 3·5. Flexural Creep Data of Mixed Filler and Reinforced Composites at 73°F (ICI·LNP) Apparent Modulus (103 psi)* Hours

Strain (%) Hours

Filler Type and Content (%)

Stress (psi)*

10

100

1,000

10

100

1,000

Nylon 6/6

Glass fiber 15, mineral 25

2,500 5,000

0.555 0.823

0.623 0.967

0.709 1.140

450 607

401 517

353 439

Polyester (PBT)

Glass fiber 15, mineral 25

2,500 5,000

0.452 0.693

0.470 0.742

0.482 0.819

553 721

532 674

519 610

Nylon 6/10

Ferrite 83

2,500 5,000

0.463 0.638

0.507 0.732

0.568 0.952

540 784

493 683

440 525

Polypropylene

Carbon powder

2,500 5,000

1.100 6.230

1.140 6.920

1.970 8.660

114 40

87 36

63 29

Nylon 6/6

Glass fiber 15, carbon powder

2,500 5,000

2.160

2.400

2.510

116

104

100

Nylon 6

Glass beads 30

1,250 5,000

0.140 0.290

0.320 0.650

0.368 0.750

893 862

391 385

340 333

Base Resin

'To convert psi to pascals (Pa), multiply by 6.895 x

UP.

Creep Modeling

Creep is related to plastics' viscoelastic behavior and can be explained with the aid of a Maxwell model such as that shown in Figure 3-55 [12, 287]. When a load is applied to the system, shown diagrammatically, the spring will deform to a certain degree. The dashpot will first remain stationary under the applied load, but if the same load continues to be applied, the viscous fluid in the dashpot will slowly leak past the piston, causing the dashpot to move. Its movement corresponds to the strain or deformation of the plastic material. When the stress is removed, the dashpot will not return to its original position, as the spring will. Thus we can visualize a viscoelastic material as having dual actions: one of an elastic material, like the spring, and the other like the viscous liquid in the dashpot. The properties of the elastic phase are independent of time, but the properties of the viscous phase are very much a function of time, temperature, and stress. This phenomenon is further explained by looking at the dashpot again, where we can visualize that a thinner fluid resulting from increased temperature under a higher pressure (stress) will have a higher rate of leakage around the piston during the time that the conditions described prevail. Translated into plastic creep, this means that at higher use temperature and higher stress levels the strain will be higher, resulting in greater creep. Visualizing the reaction to a load (without time) by such a dual-component interpretation is valuable to our understanding of the creep process but is basically meaningless for design purposes. For this reason the designer is interested in the actual deformation or part failure over a specific time span. Observations of the amount of strain at certain time intervals must be made, which will then make it possible to construct curves that can be extrapolated to longer time periods. The initial readings, at 1, 2, 3, 5, 7, 10, and 20

PLASTICS: DESIGN CRITERIA 175

hours, are followed by readings every 24 hours up to 500 hours, then readings every 48 hours up to 1,000 hours. The time segment for the creep test is common to all materials. Strains are recorded until the specimen ruptures or is no longer useful because it has yielded. In either case, a point of failure of the test specimen has been reached.

Designing with Creep Data The factors that affect being able to design with creep data include a number of considerations. First, the strain readings of a creep test can be more accessible to a desigrier if they are presented as a creep modulus. In a viscoelastic material the strain continues to increase with time while the stress level remains constant. Since the creep modulus equals stress divided by strain, we thus have the appearance of a changing modulus. Second, the creep modulus, also known as the apparent modulus or viscous modulus when graphed on log-log paper, is normally a straight line and lends itself to extrapolation for longer periods of time. The apparent modulus should be differentiated from the modulus given in the data sheets, which is an instantaneous value derived from the testing machine, per ASTM D 638. Third, creep data application is generally limited to the identical material, temperature use, stress level, atmospheric conditions, and type of test (that is, tensile, compressive, flexural) with a tolerance of ± 10 percent. Only rarely do product requirement conditions coincide with those of a test or, for that matter, are creep data available for all the grades of materials that may be selected by a designer. In such cases a creep test of relatively short duration, say 1,000 hours, can be instigated, and the information be extrapolated to long-term needs. In evaluating plastics it should be noted that reinforced thermoplastics and thermosets display a much higher resistance to creep than do unreinforced plastics. Finally, there have been numerous attempts to develop formulas that could be used to predict creep information under varying usage conditions. In practically all cases the suggestions have been made that the calculated data be verified by actual test performance. Furthermore, numerous factors have been introduced to apply such data to reliable predictions of product behavior.

•

/DaShPot

____ Weight

Figure 3·55. A Maxwell model used to illustrate viscoelastic behavior.

Table 3-6. flexural Creep Data of Glass- and Carbon-fiber-Reinforced Thermoplastics at 73°f (lCI-LNP) Fiber Content Base Resin

(%)

Stress (psi)*

Strain (%) Hours 10

100

Apparent Modulus (103 psi)* Hours 1,000

10

100

1,000

Glass-Fiber-Reinforced Composites ABS

20 40

SAN

20

30 40 Polystyrene

20

40

2,500 5,000 5,000 10,000

0.263 0.520 0.290 0.585

0.288 0.607 0.302 0.615

0.325 0.643 0.332 0.660

951 962 1,724 1,709

868 824 1,656 1,626

769 778 1,506 1,515

2,500 5,000 10,000 5,000 10,000

0.277 0.455 0.910 0.367 0.558

0.239 0.478 0.956 0.389 0.600

0.271 0.540 1.086 0.402 0.642

1,101 1,099 1,099 1,362 1,792

1,046 1,046 1,046 1,285 1,667

923 926 921 1,244 1,558

2,500 5,000 10,000 5,000 10,000

0.273 0.519 1.090 0.280 0.570

0.301 0.550 1.205 0.290 0.630

0.338 0.585 1.350 0.300 0.690

916 %3 917 1,786 1,754

831 909 830 1,724 1,587

740 855 741 1,667 1,449

Polycarbonate

20 30 40

5,000 5,000 5,000 10,000

0.618 0.451 0.312 0.620

0.628 0.462 0.319 0.700

0.654 0.466 0.322 0.710

809 1,109 1,603 1,613

796 1,082 1,567 1,429

764 1,073 1,553 1,408

Polyetherimide

20 40

5,000 5,000 10,000

0.512 0.275 0.554

0.551 0.299 0.599

0.580 0.315 0.631

976 1,818 1,805

907 1,672 1,669

862 1,587 1,585

Polyethylene

20

2,500

0.796

0.894

0.936

314

280

267

Polysulfone

30 40

5,000 5,000 10,000

0.362 0.290 0.590

0.439 0.340 0.670

0.453 0.340 0.680

1,381 1,724 1,694

1,139 1,471 1,492

1,104 1,471 1,471

Polyacetai

30

1,250 2,500 5,000 5,000 10,000

0.159 0.278 0.546 0.380 0.640

0.182 0.320 0.629 0.480 0.800

0.190 0.337 0.670 0.520 0.860

786 899 916 1,316 1,562

687 781 795 1,042 1,250

658 742 746 961 1,163

40 Polypropylene

30 40

5,000 5,000

0.410 0.680

0.460 0.940

0.480 1,130

610 735

543 532

421 442

Polyphenylene sulfide

30

2,500 5,000

0.190 0.350

0.190 0.350

0.190 0.350

1,316 1,429

1,316 1,429

1,316 1,429

Nylon 6

20 30

5,000 5,000 10,000

0.890 0.750 1.533

1.070 0.800 1.892

1.090 0.830 1.933

562 667 652

467 625 528

459 602 517

Nylon 6/10

40

5,000 10,000 5,000

0.550 1.320 0.280

0.640 1.450 0.340

0.680 1.490 0.360

909 756 1,785

781 690 1,471

735 671 1,389

60

(cont'd)

176

Table 3-6. (Continued) Fiber Content Base Resin Nylon 6/6

(%)

30 40

60

Strain (%) Hours

Apparent Modulus (103 psi)* Hours

Stress (psi)*

10

100

1,000

10

100

1,000

2,500 5,000 2,500 5,000 10,000 5,000 10,000

0.340 0.434 0.298 0.380 0.800 0.250 0.560

0.470 0.617 0.391 0.514 0.960 0.320 0.630

0.490 0.662 0.391 0.528 0.990 0.350 0.640

735 1,152 839 1,316 1,250 2,000 1,786

532 810 639 973 1,041 1,563 1,587

510 755 639 947 1,010 1,429 1,562

Polyurethane

40

2,500

0.375

0.481

0.500

667

520

500

High-impact nylon

30

1,250 2,500 5,000

0.270 0.482 1.369

0.290 0.534 1.719

0.330 0.679 2.018

463 519 365

431 468 291

379 368 248

Polyester (PBT)

30

2,500 5,000 5,000 10,000

0.210 0.416 0.278 0.590

0.241 0.478 0.284 0.630

0.252 0.502 0.298 0.640

1,190 1,202 1,799 1,695

1,037 1,046 1,761 1,587

992 996 1,678 1,562

2,500 5,000

0.248 0.640

0.275 0.678

0.324 0.757

1,008 781

909 737

772 660

40 Amorphous nylon

30

Polyester elastomer

30

1,250 2,500 5,000

0.365 0.448 1.460

0.397 0.496 1.550

0.411 0.538 1.660

342 558 342

315 504 322

304 465 301

Polyphenylene oxide

30

2,500 5,000

0.255 0.518

0.277 0.548

0.314 0.625

980 965

902 912

796 800

Carbon-Fiber-Reinforced Composites Polycarbonate

30

2,500 5,000

0.120 0.240

0.128 0.251

0.129 0.260

954 1,104

940 1,089

880 1,044

Polyetherimide

20

5,000 10,000

0.367 0.721

0.399 0.779

0.420 0.820

1,362 1,387

1,253 1,284

1,190 1,219

Polysulfone

30

2,500 5,000

0.098 0.224

0.112 0.239

0.126 0.250

2,551 2,232

2,232 2,092

1,984 2,000

Polyphenylene sulfide

30

2,500 5,000

0.070 0.168

0.078 0.170

0.084 0.170

3,571 2,976

3,205 2,941

2,976 2,941

Nylon 6

30

2,500 5,000

0.221 0.443

0.235 0.467

0.245 0.485

1,131 1,128

1,064 1,066

1,020 1,030

Nylon 6/6

30

2,500 5,000 2,500 5,000

0.140 0.334 0.112 0.240

0.168 0.376 0.133 0.254

0.194 0.390 0.140 0.257

1,786 1,497 2,232 2,083

1,488 1,330 1,880 1,968

1,287 1,282 1,785 1,945

2,500 5,000

0.084 0.196

0.110 0.224

0.112 0.243

2,976 2,551

2,273 2,232

2,232 2,058

40 Polyester (PBT)

30

'To convert psi to pascals (Pa), multiply by 6.895 x 10'.

177

178 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Creep data can be very useful to the designer. The data in Figure 3-56 have been plotted from material available from or published by material manufacturers. The first point is the 100-hour time interval. The data for shorter intervals do not as a rule fit the straight-line configuration that exists on log-log charts for the long-term duration beyond the first lOO-hour test period. The circled points are the 100-, 300-, and 1,000-hour test periods, and other observed values, and a straight line is fitted either through the circles or tangent to them to give the line a slope for long-term evaluation. From this line can be estimated the time at which the strain will be such as to cause tolerance problems in product performance or, by using the elongation at yield as the point at which the material has attained the limit of its useful life, we can estimate the time at which this limit will be reached. The formula "modulus (apparent) = stress/strain" enables us to locate the modulus that corresponds to the test stress and strain (the strain being obtained by using the dimensional change or elongation limit) where it intersects the straight line leading to an appropriate time value. The polycarbonate creep line shows that a limit of 0.010 in elongation is reached at the end of lOS hours (apparent modulus = 200,000 psi) and an elongation (yield) of 0.06 is arrived at after 107 hours, or indefinitely if the 0.010 limitation does not exist. In the interest of sound design-procedure, the necessary creep information should be procured on the prospective material, under the conditions of product usage. In addition to the creep data, a stress-strain diagram, also at the conditions of product usage, should be obtained. The combined information will provide the basis for calculating the predictability of material performance.

Allowable Working Stress. The viscoelastic nature of the material requires not merely the use of data sheet information for calculation purposes but also the actual long-term performance experience gained for it, which can be used as a guide. The allowable working stress is important for determining dimensions of the stressed area and for predicting the amount of distortion and strength deterioration that will take place over the life span of the product. The allowable working stress for a constantly loaded part that is expected to perform satisfactorily over many years has to be established, using creep characteristics for a material with enough data to make reliable long-term predictions of short-term test results. Creep test data when plotted on log-log paper usually form a straight line and lend themselves to extrapolation. The slope of the straight line, which indicates a decreasing modulus, depends on the nature of the material (principally its rigidity and temperature of heat deflection), the temperature of the environment in which the part is used, and the amount of stress in relation to tensile strength. Certain conclusions can now be developed, based on creep-data test results: First, for practical design purposes, the data accumulated for up to 100 hours of creep

Figure 3-56. (Facing page) The apparent modulus versus time at 23°C (73"F) for a) MerIon polycarbonate at 13.8 MPa (2,000 psi); b) an extrapolation of a) beyond 107 hrs.; c) Noryl 731 modified PPO at 13.8 MPa (2,000 psi); d) Delrin 500 acetal at 6.9 MPa (1,000 psi); and e) Zytel 109 nylon at 50 percent relative humidity and 6.9 MPa (1,000 psi). The broken lines represent extrapolated values, the circles, actual test-reading points. The log-log graph sheets are 9 in. x 15 in. and contain 3 x 5 cycles. The end of the first time cycle of 103 represents 1,000 hours.

U="'r cr

"'"''"!,,''' rr=~"o'l

"*'" II"

I

1

I, , I

I, I 1 I

I I I

I, I I

1

•0 0 •

..!?

I

,

I I

I

I

I I

I I ,

I I

I I

I 1

I I I 1 I I

I I

I

I

1

I

,

,I ,,I

,

'

\

I

I

I

fl

I

.

I I

, P

I

:

I

I

I

;

I

I I

1

,

_

I' I

, 1

I

1

I

I

,

,

I

I

,,,

1

, ,

.o

, .•

!

, ,, , I

...

~

•

"b

I(

",I

u

w

0

f

.=i

>c

1

e g

1 • '0 '0 _ U...l...L,J.. ··, ~I"...Jrul....... · '--L..J . JL!......L,...l.,J..,J....J-~\)-..J-..L.-----U.........L......L-I...JL....L-.o.-L~---I

! - .... ~

i c

~

III

~

~

C

..J

.=i

-

.. .. ...

..E

£

I

i

..

u

.

,

... l' o·

to

'0

o

o

!? 'I ....----..,v .------'-=--J'I....--------....v.___--------8 PU I .." "U" '0' .aul\:l :I PUI "0 ':> "u., '0' .0\:1 • ..., rOl

x,

In,"90W't , ... ,.oay

179

to

o

,.::

180 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

are of no real benefit. There is usually too much variation during this test period, which is of a relatively short duration. Next, the apparent modulus values, starting with a test period of 100 hours and continuing up to 1,000 hours, form a straight line when plotted on log-log paper. And finally, this line may be continued for longer periods on the same slope for interpolation purposes, provided the stress level is one quarter to one fifth that of the ultimate strength and the test temperature is no greater than two thirds of the difference between room temperature and the heat deflection temperature at 264 psi. When these limitations are exceeded, there is a sharp decrease in the apparent modulus after 1,000 hours, with indications that failure is from creep is approaching (that is, the material has attained the limit of its usefulness). Since the designer will be expected to plot curves to suit requirements, some examples will be cited that can serve as a guide for potential needs (see Fig. 3-57). This first example (an ABS) uses creep data for 1,000 psi stress at 23°C (73°C). When the line is extended to lOS hours, the apparent modulus is 140,000 psi. If the product is designed for the duration of lOS hours and calculations are made for part dimensions, the modulus of 140,000 psi should be inserted into any formula in which the modulus appears as a factor. At lOS hours the total strain is Stress E=-Strain 140000 ,

=

1,000 Strain

1,000 Strain = 140,000 = 0.007 or 0.7% Based on this calculation, if the product can tolerate this type of strain without its affecting performance, then the dimensional requirements are met. The elongation at yield for this particular ABS is 0.0275, which could be considered the end of the useful strength of the material. The apparent modulus corresponding to this strain at 1,000 psi and 23°C (73°F) is

E

1,000

= 0.0275 = 36,364 psi

In the lower part of the graph in Figure 3-57, draw at the point of 56 x 1()3 on the left side a line parallel to the original creep line and find that it intersects the apparent modulus line at a time of 109 x 0.5 hours. The product would fail at this time, owing to its loss of strength, even if dimensional changes permitted satisfactory functioning of the product. Some charts show creep test data beyond the 1,000-hour duration, and in fact under most conditions the straight line between the 100- and 1,000-hour points is continued into the 10,000- and 20,000-hour range. Even in such charts a deviation from the straight line occurs occasionally, which should not be considered unreasonable, because of all the variables that enter into the test data.

PLASTICS: DESIGN CRITERIA 181

.;;; Q.

-.

M

0

2S

103

;,

'5

'0

...... 0

:e c

'"

102

Q. Q.

<{

102 10 7

I

103

104

105

108

109

10 10

106

107

End of useful life Time (hours)

Figure 3·57. Creep data for ABS. Selecting an allowable continuous working stress at the required temperature must be a process that allows for making an estimation of the elongation at the end of the product's life. For example, if a product will be stressed to 1,700 psi at a temperature of 66°C (150OP) , and data are available for 2,000 psi stress at ?loC (160OP) , this information plotted on log-log paper should allow us to extrapolate the long·term behavior of the material.

Isometric and Isochronous Graphs. Creep curves are a common method of displaying the interdependence of stress-strain-time. However, there are other methods that may also be useful in particular applications, specifically isometric and isochronous graphs. An isometric graph is obtained by taking a constant strain section through the creep curves and replotting this as stress versus time (see Figs. 3·33 and 3·34). It is an indication of the relaxation of stress in the material when strain is kept constant. These data are often used as a good approximation of stress relaxation in a plastic. In addition, if the vertical (stress) axis is divided by the strain, one obtains a graph of the modulus against time (see Figs. 3-37 and 3-44). This graph provides a good illustration of the time-dependent variation of the modulus. An isochronous graph may be obtained by taking a constant time section through the creep curves and then plotting stress versus strain as shown in Figure 3-47. It can also be obtained experimentally by performing a series of brief creep and recovery tests on a plastic. In this procedure a stress is applied to a plastic test piece and the strain is recorded after a specified time, typically 100 seconds. The stress is then removed and the plastic allowed to recover, normally for a period of 4 (4 X 100 sec.). A larger stress is then applied to the same specimen, after recording the strain at the l00-sec. time period; then this stress is removed and the material allowed to recover. This procedure is repeated until enough points have been obtained to let an isochronous graph to be plotted. Isochronous data are usually presented on log-log scales. One reason for doing so is

182 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

that on linear scales any slight, but possibly important, nonlinearity between stress and strain may go unnoticed, whereas the use of log-log scales will usually produce a straightline graph the slope of which gives an indication of the linearity of the material. If the material is perfectly linear, the slope will be at 45 degrees, but if it is nonlinear the slope will be less than 45 degrees. Isochronous graphs are particularly valuable when obtained experimentally, because they are less time consuming and require less specimen preparation than creep curves. Such graphs at several time intervals can also be used to build up creep curves and indicate areas where the main experimental creep program could be most profitable. They are also popular as means of evaluating deformational behavior, because their method of data presentation is similar to the previously discussed conventional tensile test data.

Application as a Theory. To illustrate the applications of various industry theories related to creep, here are some examples of design problems [14]. Example 1 Tension (compression) of linearly viscoelastic prismatic bars (see Fig. 358). The bar of viscoelastic material is subjected to a load F that is assumed to produce an initial stress below its short-time yield strength. Hence, the initial modulus is (3-1) where (J is the applied stress and Eo is the initial strain. If the load E is sustained for time t, the strain at t will equal the sum of the initial strain and the additional strain due to viscoelastic flow. Therefore,

Et

=

(J/( Eo

+

(3-2)

Et )

where E, is the creep or viscoelastic strain. With this in mind, assume that the lO-in. bar in Fig. 3-58 is subjected to a stress of 1,000 psi, which induces a total elongation of 0.10 in. or a strain of 0.01 (0.10/10). Then, from Equation 3-1 the initial modulus Eo is 100,000 psi (1,000/0.10). If the load is sustained for 1,000 hours and the sample stretches or creeps an additional 0.15 in., then the total of initial and creep strains is 0.025 (0.10 + 0.15/10) and the modulus at 1,000 hours, E" is now 40,000 psi. Thus, for linearly viscoelastic behavior, by measuring the creep strains it is possible to draw time-modified modulus curves. Having established these curves, it is then possible to use these data to predict the behavior of the plastic under other conditions. Such timemodified modulus curves for several common thermoplastics were shown in Figure 358. It is important to remember that such curves are valid only for a specific temperature and for strains that do not exceed the limits of the validity of the data.

~b=lin.

F

--1f----------l~ If-o
F

13

2

in.

Figure 3-58. A diagram of the variables involved in analyzing the primatic bar in tension.

PLASTICS: DESIGN CRITERIA 183

Table 3-7. Trial Estimate of Variation in Creep Strain with Time Time (hours)

0 10 100 1,000 10,000 100,000

Apparent Modulus psi (MPa)

20,000 19,000 16,500 14,000 12,000 10,000 7,500

Calculated Creep Strain (in.lin.)

(138) (131) (114) (96) (83) (69) (52)

0.0150 0.0158 0.0182 0.0212 0.0250 0.0300 0.0400

For example, suppose that the bar in Figure 3-58 is made of the polyethylene having the modulus curve in Figure 3-59. The initial elastic modulus of this material was determined to be 20,000 psi at 23°C (73°F). Now suppose that the bar is loaded with a 300 psi stress. Then, using Equation 3-1, the strain in the bar is 0.015 in.lin. (300/20,000). In addition, from Equations 3-2 and 3-3, the creep strain E t is (3-3) Using this relationship and the data in Figure 3-59, it is possible to calculate the creep strain over a l00,OOO-hour period, as in Table 3-7. It will be noted that all the strain values thus calculated are in excess of 0.5 percent. Thus, since the data in Figure 3-59 on which these strain values are based are valid only for strains not exceeding 0.5 percent, the creep values thus calculated are inaccurate. Also, since we obtain an unrealistic answer when we use the specified design stress with this material, we must conclude that it is not possible to apply a stress to a bar of this material of 300 psi.

Acrylic plastic

x

'in

:! 300 ::J

"3

'8

':200 ~

~ c ~100

Polyamide plastic [~-----;~~~~~-------=:::::::::::::::d

19,000 psi

Polyethylene plastic

7500 psi

°1t:~oanrrlt=0~~nrn10cO~~Icr1mOO=0=C~~lmO~,00=0~~lnOO~,Ooo Time, hr

Figure 3-59. Time-modified modulus curves for four common thermoplastics. The data were obtained at 23°C (73°F) and are applicable only for strains of 0.5 percent or less.

184 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

To stay below the maximum strain of 0.5 percent, the imposed load cannot exceed 100 psi. Furthermore, if the stress were 50 psi, the value of the initial strain would be 0.0025 in.!in. Knowing this, and that the limiting strain at time t, or E" is 0.0050 in.! in. and that the initial modulus is 20,000 psi, we can use Equation 3-3 to determine the apparent modulus of the bar E after creep at time t. Doing this we find that the apparent modulus at time t is 10,000 psi. Taking this value and referring to Table 3-7 or to Figure 3-59, we find that this value of the modulus occurs at a strain of 0.005 at 10,000 hours. Thus, with a load of 50 psi on the bar, accurate predictions of creep deformation can be made for service up to 10,000 hours for this particular polyethylene.

Example 2 Bending of bars (see Fig. 3-60). To illustrate the design procedure, assume the bar in this figure is made of an acrylic plastic whose modulus-time properties are those shown in Figure 3-59. Assume that the maximum permissible deflection of the middle of the bar during one year's service is 0.15 in. and that the problem is to find the maximum permissible value of bending moment M or load W. It can be shown that the maximum deflection 8max occurs in the center of the bar and that the magnitude of the deflection is given by (3-4) where M is the bending moment, L the length of the bar, E the modulus, and I the moment of inertia of the cross-section. From Figure 3-59, the modulus of the acrylic after one year's time is 300,000 psi and for the conditions imposed Equation 3-4 tells us that the bending moment M is equal to 267 in.-lb. The next step in the process is to determine the tensile strain in the bar associated with the calculated bending moment. It has been shown that for a rectangular cross-sectional bar in pure bending the maximum bending stress O'max is given by O'max

=

(Mh/2J)[(2n

+

1)/3n]

(3-5)

where h is the height of the beam and n is a material constant having a value of 3 for the problem at hand. Using Equation 3-5, O'max is calculated to be 156 psi. Then the tensile strain is 156 psil300,000 or 0.00052 in.!in. Since this strain value does not exceed the applicability of the data in Figure 3-59, the above calculations are consistent with the real behavior of the acrylic plastic being investigated. Turning to the sketch of the bar in Figure 3-60, the maximum deflection occurs at midspan and is given as (3-6) Solving Equation 3-5 for W, using the conditions listed above, gives us a value for maximum load of 56 lb. Remembering that in pure tension it can be shown that M = Wa, Equation 3-5 can be used to calculate the bending stress associated with the load calculated above and thence the strain, using modulus data from Figure 3-59. Doing this, we find that with a load W of 56 lb. the bending strain is 0.000543 in.!in., well within the limitations of Figure 3-59.

PLASTICS: DESIGN CRITERIA 185

M

( ~I_~I ) -v-b=l M

If.!=ot:----,L = 30

I

a = 5 in.

W

l]h=2in.

in.,--...;>~I

End moments

a = 5 in.

in.

W

r- -, rI]

r----'------'---i

b = 1 in.

h = 2 in.

foot---L = 30 in.--->-I Symmetrical

Figure 3-60. Variables involved in an analysis of primatic bars in pure bending. The top drawing applies to the analysis of end moments, the lower drawing to symmetrical loaded bars.

Example 3 Pipeline on supports (Fig. 3-59). A pipeline on supports involves essentially the same problem as that associated with a continuous beam. It is well known that the bending moments at the pipe's supports can be calculated by the theorem of three moments:

where An is the area of the bending-moment diagram for the span and an is the horizontal distance of the centroid of the moment area from supports n - 1, etc. For a system having a large number of equal spans it is easily shown by Equation 3-7 that the maximum moment is equal to WL/12. Therefore, the length between spans L is

L

=

(l20'maxIW) (lIe)

(3-8)

where c is the radius of the pipe as shown in Figure 3-61, and I is the moment of inertia of the cross-section of the pipe.

Thus, by determining the maximum stress to be expected in the pipe, the designer can specify the spacing of the pipe supports. Suppose that the problem involves a length of 2-in.-diameter, Schedule 80 rigid PVC pipe holding water under a pressure of 50 psi at room temperature. The maximum allowable sag or deflection between supports is 0.25 in., which is a realistic value actually used to avoid drainage problems. It is also required that the pipe sag by not more than this amount after 10,000 hours of service. There are no wind loads and the material has an initial elastic modulus of 450,000 psi. It can now be shown that the 2-in.-diameter PVC pipe filled with water has a weight of 0.194 Ib.lft. The various stresses in the pipe can be described by the following relationships: Hoop stress Longitudinal stress Radial stress

= O'h = pdl2t = O'z = (pdI4t) + (MIZ) = O't = -p (for I.D.)

(3-9) (3-10) (3-11)

186 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

o

1

n-l

n

n+l

- - - - L = 30 in.------;~ Figure 3-61. Variables involved in pipe-design problems. The upper-left drawing is used to analyze problems involved in spacing pipe supports; the lower drawing deals with designing thin-wall tubing to contain an internal pressure; the drawing at the right covers the analysis of heavy wall piping under pressure.

where p is the internal water pressure, d is the diameter of the pipe, t is the thickness of the pipe, M is the bending moment, and Z is the section modulus or lie. It can also be shown that the equivalent maximum stress level (1 in the pipe is given by (3-12) Using Equation 3-12, the approximate equivalent wall stresses for several span lengths of the pipe described in the problem are shown in Table 3-8.

It is concluded that stress-rupture failure is not likely even when the span between supports is as much as 15 ft. Even when a conventional safety factor of 4.0 is used to increase the expected design stress to 2,960 (4 x 740) psi, this stress is still below the expected rupture stress at 10,000 hr in Figure 3-62. However, the spacing of the pipe supports must also be such that the deflection between supports does not exceed i in. This requires an analysis of the creep that may be expected with any proposed pipe-support spacing specification. The elastic deflection of the midpoint of continuous pipe between supports is calculated as

8max

= WCI384El

(3-13)

Using this formula, the initial deflection values of the pipe between supports can be calculated as shown in Table 3-8. Using Equation 3-8 and appropriate values of (101(1 from Figure 3-63, the creep deflections after 10,000 hours may also be calculated; the values are also shown in Table 3-8. From Table 3-8 it becomes obvious that although a 15-ft. span between pipe supports would be satisfactory from a pipe-stress standpoint, excessive sag or deflection of the pipe between the supports would take place over the service life with this span length. To avoid exceeding the specified maximum deflection of 0.25 in., Table 3-8 tells us that the span between supports should not exceed 6-8 ft.

PLASTICS: DESIGN CRITERIA 187

Table 3-8. Analysis of Stress and Deflection in a Pipeline on Supports1 Equivalent Stress 3 Span Length 2 ft. (m)

Outer Diameter psi (MPa)

2 4 8 10 15

193 199 279 372 740

(0.61) (1.2) (2.4) (3.0) (4.6)

Inner Diameter psi (MPa) 267 284 366 442 738

(1.33) (1.37) (1.92) (2.56) (5.10)

(1.84) (1.96) (2.52) (3.05) (5.09)

Initital Elastic Deflection4 mils (mm)

Maximum Deflection After 10,000 Hourss mils (mm)

0.43 6.83 109.0 268.0 418.0

0.72 11.47 183.5 450.0 705.0

(0.011) (0.173) (2.769) (6.807) (10.62)

(0.02) (0.2913) (4.661) (11.43) (17.91)

I See Figure 3-61. 2Between pipe supports. 3Difference arises from thickness of pipe wall. 48ased on outer diameter measurements.

5Any

deflection due to internal pressure has been omitted.

Example 4 Expansion of thin-wall pipe under internal pressure (see Fig. 3-61). The stresses in a thin-wall tube are given by Equations 3-9, 3-10, and 3-11 and the strains may be calculated from Eh

== (lIE) [crh - v(crz

+ crt)]

(3-14)

Ez

== (liE) [crz

V(crh

+ crt)]

(3-15)

Et == (lIE) [crt - V(crh

+ crz)]

(3-16)

-

Using Equations 3-14 and 3-9, 3-10, and 3-11, (3-17) and since radial expansion u at the bore of the cylinder (the J.D. of the pipe) is equal to (ET]) (dI2), Equation 3-17 can be rewritten in the form

(3-18) Assuming viscoelastic behavior, all terms in Equation 3-18 containing E and v can be substituted for using the relationships in Equations 3-8 or 3-9 or (3-19) where B is the bulk modulus of the material. Similarly, it can be shown that the longitudinal expansion E is given by (3-20)

188 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

..,

-

I

o

x

2

o~----~~~~~------~----~------~----~

om

Time to rupture, hr

1000

10,000

Figure 3-62. Stress-rupture data for rigid PVC 2-in.-diameter pipe as a function of temperature.

1.0

·0

~0.8 b

A

~0.7

B

::f

'5

1°·6 0.5

Relaxation time, hr Figure 3-63. Stress-relaxation characteristics at 23°C (73°F). a) Rigid PVC, pressure relaxation; b) rigid PVC, tension relaxation; c) PE, pressure relaxation; d) PE, tension relaxation; e) nylon, tension relaxation; f) PTFE, tension relaxation.

Assuming such a thin-wall tube is made from a polyamide material conforming to the behavior in Figure 3-59, determine the deformation that will occur over a five-year service life. The tube has a 2-in. diameter d, wall thickness t of 0.3 in., length L of 30 in., and it is subjected to internal pressure p of 100 psi. The material has a yield strength of about 6,000 psi at room temperature and a modulus of 210,000 psi. From Figure 3-59 the effective modulus of the material after five years is about 80,000 psi. Therefore, crr:/cr = Er:/E = (210,000/80,000) :::::: 2.62. The bulk modulus B may be calculated from

PLASTICS: DESIGN CRITERIA 189 B

=

(3-21)

(E/3)(1 - 2v)

using Eo for E and a value of 0.3 for v. Then B = 66,500 psi. Substituting these values in Equation 3-19, u is found to be 0.00695 in. and, from Equation 3-19, E the total length change is 0.0237 in. In order to compute the external expansion of the pipe (since u is the I.D. change), the material volume per length of pipe is assumed to remain constant during the expansion of the I.D. From this calculation the expanded external diameter of the pipe is found to be 2.61 in. (the original O.D. was 2.6 in.). Therefore, the radial expansion is 0.5 percent or within the limits of the data in Figure 3-59. In addition to estimating the deformation, a check must be made to determine the possible failure of the pipe by yielding. This is done by using the following equation for the equivalent stress in a pipe with capped ends:

P

=

[(3/16)(d/tf

+ (3/d)(d/t) +

1]112

(3-22)

From Equation 3-22 and knowing the yield strength of the material to be 6,000 psi, the maximum allowable pressure is calculated to be 1,590 psi. This is far above the l00-psi design pressure, so there is little chance of the pipe's failing.

Example 5 In the case of a heavy-walled cylinder under internal pressure (see Fig. 361). Using methods similar to those used for the thin-walled cylinder, the creep type of radial expansion for a heavy-walled cylinder can be shown to be

(3-23) where the terms are as before and the dimensions a, r, and c are as detined in Figure 361.

Creep Guidelines. properties:

Here is a summation of the factors to consider when reviewing creep

1. Predictions can be made on creep behavior based on creep and relaxation data. 2. There is generally a less-pronounced curvature when creep and relaxation data are plotted log-log. This facilitates extrapolation and is commonly practiced, particularly with creep modulus and creep-rupture data. 3. Increasing the load on a part increases its creep rate. 4. Increasing the level of reinforcement in a composite increases its .resistance tocreep. 5. Particulate tillers provide better creep resistance than untilled resins but are less effective than tibrous reinforcements. 6. Glass-tiber-reinforced amorphous TP composites generally have greater creep resistance than glass-tiber-reinforced crystalline TP composites containing the same amount of glass tiber.

190 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

7. Carbon-fiber reinforcement is more effective in resisting creep than glass-fiber reinforcement. 8. The effect of a flame-retardant additive on the flexural modulus provides an indication of its effect on long-time creep. 9. Over the past century, many plastic products have been successfully designed everywhere for long-time creep performance, based on the information and test data then available, but much more exists now and will in the future. Fatigue Fatigue is the phenomenon of having materials under cyclic loads at levels of stress below their static yield strength. Fatigue data are used so the designer can predict the performance of a material under cyclic loads. The fatigue test, analogous to static long-term creep tests, provides information on the failure of materials under repeated stresses. This fatigue behavior is by no means a new problem-the term was applied to the failure of a wooden mast by hoisting too many sails too often in the pre-Christian era. As plastics replaced metals and other materials in many critical structural applications, fatigue tests became even more important, since the maximum oscillatory load that a material can sustain is a part of its tensile strength. Thus, the more conventional shortterm tests give little indication about the lifetime of an object subjected to vibrations or repeated deformations. Fatigue tests (see Fig. 3-64) are especially important for the

Figure 3-64. A typical test setup for analyzing bending fatigue.

PLASTICS: DESIGN CRITERIA 191

designer of plastics and composites that are used in load-bearing structures that will be subjected to varying loads [1, 2, 5, 11-14, 16-21,62-68, 150-155,268-70,287-97]. Some examples of products subject to fatigue when they are stressed repeatedly or in some defined cyclic manner are a snap-action plastic latch that is constantly opened and closed, a reciprocating mechanical part on a machine, a gear tooth, a bearing, and any structural component subjected to vibration, such as an aircraft wing or any part that will be subjected to repeated impacts. Such cyclic loading can cause mechanical deterioration and progressive fracturing of the material, leading to its ultimate failure. Basically, under a repeated applied cyclic load, fatigue cracks begin somewhere in the specimen and extend during the cycling. Eventually the crack will expand to such an extent that the remaining material can no longer support the stress, at which point the part will fail suddenly. However, failure for different service conditions may be defined differently than just as the separation of two parts. ASTM D 671 defines failure as occurring also when the elastic modulus has decreased to 70 percent of its original value. Practical failure must also include the melting of any part of a specimen, excessive change of dimensions or the warping of the part, and the crazing, cracking, or formation of internal voids or deformation markings. These types of defects all may seriously affect performance strength. Plastics are susceptible to brittle crack-growth fractures as a result of cyclic stresses in much the same way metals are. In addition, because of their high damping and low thermal conductivity, plastics are prone to thermal softening if the cyclic stress or cyclic rate is high. The thermoplastics with the best fatigue resistance include PP and ethylene-propylene copolymers. Testing

In testing, a specimen is subjected to the periodic varying of stresses by means of a mechanically operated device. The stresses applied generally alternate between equal positive and negative values or from zero to the maximum positive or negative values. For practical purposes, in testing plastics a certain minimum stress, instead of zero, is often used. The test can be performed in alternating bending (called flexural fatigue testing; see Fig. 3-64), and also as tensile, compn~ssion, or torsion testing, or as an alternating tensile-compression test. Fatigue data are normally presented as a plot of the stress (S) versus the number of cycles (N) that cause failure at that stress; the data plotted defined as an S-N curve. Test results are illustrated graphically in the form of curves. Examples of fatigue curves for unreinforced and reinforced plastics are shown in Figure 3-65. The values for stress amplitude and the number of load cycles to failure are plotted on a diagram with logarithmically divided abscissa and English or metrically divided ordinates. The fatigue behavior of a material is normally measured in either a flexural or a tensile mode. Specimens may be cracked or notched prior to testing, to localize fatigue damage and permit measuring the crack-propagation rate. In constant-deflection amplitude testing a specimen is repeatedly bent to a specific outer fiber strain level. The number of cycles to failure is then recorded. In constant flexural load amplitude testing a bending load is repeatedly applied to the specimen. This load causes a specified outer-fiber stress level. The number of cycles to failure is then recorded. Both modes of flexural fatigue testing can be related to the performance of real structures, one to those that are flexed repeatedly to a constant deflection and the other to those that are repeatedly flexed with a constant load. In constant-elongation amplitude testing, a specimen is repeatedly stretched to a specified tensile strain or elongation level.

5r-------,------c~--------._------.

PMMA, 0 250 10 Ihlck

4

3

ci o
2

~

~

Endurance limits

PTFE, 0.260 In Ihlck

O~------~--------~--------~------~ 10l 10' 10' 10'

a,

Cycles 10 failure

8.-~~~----~-----,------._-------,

7

......

6

PheooIic

~

5

::

4

"

~ 2

-.....,

'....

Alkyd

...., Nylon (dfy) .... _____ Polycarbonate =PTFE

.... ~--- ___ ::."'= ....

~ 0'03

b.

Oiatlyl

----phthalate

" , '" ' " ',....

~0.. 3

~

Epoxy '

10'

Polysullone

10'

Cycles to lat ure

8 ISS) iii

no ::!E

li

§ ~

Vi

6

(4 1)

iii

2

~

Acrylic

on

Polye.hylene

(1 4)

0

C_

~

4

1281

I

~rOPY lene

-

E

~

:::l

~

I

III

PTFE

lol

10' Cycles

::E 10~

10'

'0 Failure

10'

d.

200 180 160 140 120 100 80 60 40 20 0

Carbon / epoxy Boron / epoxy ) Ararrold / epoxy

I

~

BO~

aluminum S-glass / epoxy

2024-T3

aluminum

14 12 10 0.8 0.6 04 0.2

11I

CL

C)

10

10

10'

Number of cycles to failure. N

Figure 3-65. Typical room-temperature fatigue characteristics for certain TPs and TSs (a, b, and c); different TS composites and metals (d, e). The stresses are from cantilever bending under constant load, zero mean stress, and a frequency of 1,800 cycles/min. per ASTM D 671. 192

PLASTICS: DESIGN CRITERIA 193 l00~----r-----r-----r-----r-----r---~

~

C)

~ a:

~~----+-----+-----~-----+-----+----_4

t-

!/)

~ OO~--~~~-+--==~~~~----~--~ c(

:i:

~

...J

:::> II..

o tZ

~

a: w

20r-----r-----+-----;------+--~~~~_4

Q.

e.

103

10·

105

CYCLES TO FAILURE

Figure 3-65. (Continued)

Its number of cycles to failure is then recorded. In constant tensile load-amplitude testing, a tensile load is repeatedly applied· to a specimen to produce a specified tensile stress level. Then the number of cycles to failure is recorded. In both tensile tests the minimum stress-strain should remain positive, to prevent the specimen from buckling. Of the four testing methods reviewed, the two flexural techniques are the most commonly used. Since fatigue cracks often start at a random surface imperfection, considerable scatter occurs in fatigue data, increasing with the increasing lifetime wherever crack initiation occupies most of the fatigue life of a specimen. When a line of the best fit is drawn from the available data points on an S-N curve, this represents the mean life expected at any given stress level or the stress that would cause, say, 50 percent of the part failures in a given number of cycles. If sufficient data are available, much more information can be provided when different curves for various percentages of failure are plotted. Where such data are available, reasonable design criteria would be based on some probability for failure, depending on how critical the effects of failure would be. If a large, expensive repair of a complex mechanism would result from the fatigue failure of one part, then a 10 or even 1 percent probability of failure would be a more likely design criterion than the 50 percent suggested above. The fatigue strength of most TPs is about 20 to 30 percent of the ultimate tensile strength determined in the short-term test but higher for RPs. It decreases with increases in temperature and stress-cycle frequency and with the presence of stress concentration peaks, as in notched components. ASTM Special Technical Publication No. 91 discusses in detail the important ramifications to be considered in the various statistical aspects of fatigue testing. Most often, the fatigue curves as well as the tabulated values of endurance strengths and endurance limits are based on the 50 percent probability curve. As a result, designers do not resort to using scatter-band curves unless they are involved with a design that takes a statistical approach. The designer requiring information on the highest order of reliability should always contact the manufacturer or run tests.

194 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Endurance Limit

To develop S-N curves like those in Figures 3-65 and 3-66, the fatigue specimen is loaded until, for example, the maximum stress in the sample is 275 MPa (40 ksi). At this stress level it may fail in only 10 cycles (see Fig. 3-66). These data are recorded and the stress level is then reduced to 206 MPa (30 ksi). This specimen may not break until after 1,000 stress cycles at this rather low stress level. This procedure is repeated until a stress level is determined below which failure does not occur. In this example of a relatively high fatigue performance material that develops a flat portion of the S-N curve, this stress level is found to be 158 MPa (23 ksi). A test duration of 10 million (107) stress cycles is usually considered infinite life. This type of testing is expensive, principally because it involves a large number of samples and much statistical evaluation. The end result, determining the fatigue endurance limit of a material, is an extremely important design property. This property should be used in determining the allowable stresses in products, rather than just the short-term yield strength, any time a part will see cyclic loading in service. Cyclic loading significantly reduces the amount of allowable stress a material can withstand. If data are not available on the endurance limit of a material being considered for use, a percentage of its tensile strength can be used. This percentage varies with the different material systems. For engineering plastics the endurance limit could be about 50 percent of its tensile strength, as with metals. Taking this 50 percent approach requires the designer to become familiar with fatigue-testing results on plastics and other materials, so that the proper evaluation can be applied. However, to design correctly, requires obtaining reliable S-N curves with the required endurance limit, as in Figure 3-66. Plastics are subject to fatigue, with a wide range of performance, and efforts should be made to arrive at endurance limit information if a fail-safe design is desired.

300

Individual f allgue tests

250 Endurance limit 123.000 pSI) [159MPal

Vl

200

0>

C

150

"0 C

Q.I

co

100 100

lOS Number of Cycles to Failure

Figure 3-66. The use of an S-N curve to establish a fatigue endurance limit strength. The curve asymptotically approaches a parallel to the abscissa, thus indicating the endurance limit as the value that will not produce failure. Below this limit the material is much less susceptible to fatigue failure.

."

~

:E

PLASTICS: DESIGN CRITERIA 195

Heat Generation

Since plastics are viscoelastic, there is the potential for having a large amount of internal friction generated within the plastics during mechanical deformation, as in fatigue. This action involves the accumulation of hysteretic energy generated during each loading cycle. Examples of products that behave in this manner include coil or leaf springs and shoe soles. Because this energy is dissipated mainly in the form of heat, the material experiences an associated temperature increase. When heating takes place the dynamic modulus decreases, which results in a greater degree of heat generation under conditions of constant stress. The greater the loss modulus of the material, the greater the amount of heat generated that can be dissipated. Plastics for fatigue applications can therefore have low losses. If the plastic's surface area is insufficient to permit the heat to be dissipated, the specimen will become hot enough to soften and melt. The possibility of adversely affecting its mechanical properties by heat generation during cyclic loading must therefore always be considered. The heat generated during cyclic loading can be calculated from the loss modulus or loss tangent of the plastics. The rate dependence of fatigue strength demands careful consideration of the potential for heat buildup in both the fatigue test and in service. Generally, since the buildup is a function of the viscous component of the material, the materials that tend toward viscous behavior will also display a sensitivity to cyclic load frequency. Thus, thermoplastics, particularly the crystalline polymers like polyethylene that are above their glass-transition temperatures, are expected to be more sensitive to the cyclic load rate, and highly crosslinked plastics or glass-reinforced plastics are less sensitive to the frequency of load. From this discussion it should be obvious that care must be taken in the use of the type of accelerated fatigue testing that is common for metals. For example, a frequency of 30 Hz is not uncommon for metal tests. Figure 3-67 shows the significant change in the fatigue life of a PMMA as measured by excessive thermal softening at frequencies well below 30 Hz. Depending on the type of plastic, testing at frequencies of a few Hz or less is required, to avoid such softening. In contrast, ifthe component is to be subjected to high-frequency loads in service, the test should be performed at similar frequencies. As is evident in Figure 3-67, high-frequency loadings may show no significant heat buildup, provided stresses are small, particularly when the part is to be cooled. Fatigue results in a shift from ductile to brittle failure with the increased number of load cycles. Figure 3-67 also compares the strength-regression behavior obtained under sustained stress with the regression under a 0.5 Hz cyclic stress applied in a square wave form. The curves for an equal duration of tensile stress, as represented by either the time under sustained load or the cumulative time under stress during the square-wave loading of the fatigue test. Compared to the static loading, the fatigue loading results in both a pronounced shift from ductile to brittle fracturing and a marked decrease in the time to failure at a given stress. Reinforced Plastics and Composites

In common with metals and unreinforced plastics, RPs also are susceptible to fatigue. However, they provide high performance when compared to unreinforced plastics and many other materials (see Figs. 3-65d and e, 3-68, 3-69, and Table 3-9). If the matrix is a TP, there is a possibility of thermal softening failures at high stresses or high frequencies. However, in general the presence of fibers reduces the hysteritic heating

196 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

effect, with a reduced tendency toward thermal softening failures. When conditions are chosen to avoid thermal softening, the normal fatigue process takes places as a progressive weakening of the material from crack initiation and propagation (see Fig. 3-69). Plastics reinforced with carbon, graphite, boron, and aramid are stiffer than the glassreinforced plastics (GRPs) and are less vulnerable to fatigue. (E-glass is the most popular type used; S-glass improves both short- and long-term properties.) In short-fiber GRPs cracks tend to develop easily in the matrix, particularly at the interface close to the ends of the fibers. It is not uncommon for cracks to propagate through a TP matrix and destroy the material's integrity before fracturing of the molded product occurs (see Fig. 3-70). With short-fiber composites fatigue life can be prolonged if the fiber aspect ratio of its length to its diameter is large, such as at least a factor of five, with ten or better for maximum performance. In most GRPs debonding can occur after even a small number of cycles, even at modest levels. If the material is translucent, the buildup of fatigue damage can be observed. The first signs (for example, with glass-fiber TS polyester) are that the material becomes opaque each time the load is applied. Subsequently, the opacity becomes permanent and more pronounced, as can occur in corrugated RP roofing panels. EventUally, resin cracks will become visible, but the product will still be capable of bearing the applied load until localized intense damage causes separation in the component. However, the first appearance of matrix cracks may cause sufficient concern, whether for safety or aesthetic reasons, to limit the useful life of the product. Unlike most other materials, GRPs give visual warning of their fatigue failure. Since GRPs can tend not to exhibit a fatigue limit, it is necessary to design for a specific endurance, with safety factors in the region of three to four being commonly used. Higher fatigue performance is achieved when the data are for tensile loading, with zero mean stress. In other modes of loading, such as flexural, compression, or torsion, the fatigue behavior is worse than that in tension. This is generally thought to be caused by the setting up of shear stresses in sections of the matrix that are unprotected by some method such as having properly aligned fibers that can be applied in certain designs. Another technique, which has been used successfully in high-performance RP aircraft wing structures and other applications, incorporates a thin, high-heat-resistant film such as Mylar between layers of glass fibers. With GRPs this construction significantly reduces the self-destructive action of glass-to-glass abrasion and significantly increases the fatigue endurance limit. Fracture Mechanics

The fracture mechanics theory developed for metals is also adaptable for use with plastics. The basic concepts remain the same, but since metals and plastics are different they require different techniques to describe their fatigue-failure behaviors. Some of the comments made about crack and fracture influences on fatigue performance relate to the theory of fracture mechanics. The fracture mechanics theory method, along with readily measured material properties, component geometry, and loading information, can be used to design against fatigue failure. The fracture mechanics model also gives insight into materials' development by showing how their resistance to crack propagation depends on both molecular and structural factors. Service failures in plastics can be caused by fatigue. When time is the critical factor, this type of failure is called static fatigue or creep rupture. If mechanical load reversal

PLASTICS: DESIGN CRITERIA 197

~

o I 75...

5

10

STRESS AMPLITUDE, a, MPa 20 25 30 35 40 45

. . . . . . . .Ioo... . ._

----no

....,j--.. . . ._

~150

15

......._

50

55

60

65

....- -....._ _........!...-. .SO

~

70 ....

cooling - - - -forced cooling

....

~

cP

60

~I~

g ~

....~

50

~IOO

40 ~

~

~

a

2.5 Hz \ "

_ z 75

~~

30

50

.-'

10 UJ

o

2

34567 STRESS AMPLITUDE, a, 103 psi

8

9

a:: u

0

~

".________(b_)_ _ _ _ _ _ _ _ _. . 70

~ 10

60

...

6

5 \',

4 3 2

~

It

9

7

~ a:: In

Z

~

,"

__ - ... -- - ...""'~"-:1"'"

1 ~

a

20 -

a:: u

~ 32

....~

""

----

... brittle fract':re

20

I0 0.0 I O. I 10 102 TIME UNDER CYCLIC OR CONSTANT STRESS, hrs

~

~ ~ ~

..... 10 ~

O~.---~--~--~--~~--~--~--~~O

0.001

~

40 ~ 30

fracture under cyclic load (0.5 ~ '- __ - part ductile, part brittle ........ .... _

I

50

Q

~

....

V)

Figure 3·67. Examples of how fatigue loading in flexure affects thennoplastics. a) Temperature effects in PMMA acrylic with and without the cooling of specimens; b) a comparison of strength regression in rigid PVC under cyclic and sustained loads of equal duration.

or the number of cycles controls failure, the term employed is cyclic fatigue. Interaction between the material and an environment capable of damaging it can lead to stress cracking in the static case and fatigue in the cycle one. An additional failure mode is thermal degradation, in which the temperature increases within the sample from hysteretic energy dissipation.

198 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

10.0

FATIGUE ENDURANCE OF CRYSTALLINE-RESIN-BASE

-------

~SrTES 7.5

.~,.... 0

~

fJ)

5.0

.... -30% glass fiber/acetal

~ ........

X 2.5

..;

300/0 glass fiber/nylon 6

.......3ro glass tiber/nylon 6/10 -30% glass fiber/polypropylene

o

10'

C}des to failure, N

10.0r---..........- - - - , - - - - - - - - , - - - - - - - - - - ,

(f)

li 5.0

!

rn

FATIGUE ENDURANCE OF AMORPHOUS-RESIN.BASE COMPOSITES

0% glass fiber/polycarbonate 30% glass fiber /polysulfone 2.51--- - - - - - + - - - - -- -+-=--.....:..:..--=-- - - \

OL-__________~~----------~~----------~ 1~

1~

Cycles to failure. N

1~

1~

Figure 3-68. Fatigue curves for reinforced plastic composites using different types and amounts of fibers and resins (lCI-LNP).

Traditional fatigue testing produces the familiar S-N diagram. In this type of testing the crack initiation phase usually represents a large fraction of a part's life. However, the crack propagation phase reveals a material's inherent fracture resistance under fracture mechanics testing. The mechanical description of a fracture is usually divided into three stages: crack initiation, stable or incremental fatigue crack propagation, and rapid or catastrophic fracture.

10.0 ~'-----IIE:"""---~-------r--------'

30% carbon fiber/PBT

FATIGUE ENDURANCE 2.5 GLASS VS CARBON REINFORCEMENT

°1~~----------~&~----------~·~--------~7 U 10· 10· 10 Cycles to fa/lure. N

111 111

r----'\--

Fatigue endurance at 23°C I

I

\~rton 50% lo~g gl••• fI~r/ny~ 8/8

Dry, •• molded 114 ~~=8 rndffired • 501 RH .. ~ I

~ 12

j

v.

...... 10

...... "'-

8

" " " 50% _I .hort I gl... fiber/nylon ~ 6/6

--

Dry••• molded I 1. _ _M~tur8 condffioned, 50% RH

5

5

5

10'

CycIa to f. lu.re

18

111

endurance at 23° and 120°C "--- r---Fatigue ~ I I I I f ...............

....

---E

114

~

-- ~r--_

'0

:::. 12

:z

10

~ .............

~ Yerton 50% long gla.. fiberl nylon 6/6 At 23°C

"'-

-....

8

--. ""~% .hort gl ••• fiber/nylon At23°c

... ~~

8,e-

--

r--. 5

5

5

-

10'

Cycle. to f. ure

Figure 3·68. (Continued)

199

N

8

_ _ _ _ _ _--,

Epoxy

Fiber Epoxy "E" Fiber Epoxy US"

Fiber! Epoxy (Thomel 300)

Fiber! Epoxy (Kevlar 49)

~

1---20

1---40

1---60

80

Strength--100

Fiber! Epoxy (HTS)

Percent of Ultimate Static

Figure 3-69. A summary of the high-perfonnance fatigue properties of advanced composites, comparing plastic types with different materials based on their percent of ultimate static tensile strength.

_ _ _ _ _ _ L...-JJ 107

.:J 1()5

Cycles r--,-.

PLASTICS: DESIGN CRITERIA 201

Designing with Fatigue Data The ranking of fatigue behavior among various plastics should be conducted after an analysis is made of the application and the testing method to be used or being considered. It is necessary to also identify whether the product will be subjected to stress or strain loads. Plastics that exhibit considerable damping may possess low fatigue strength under a constant stress amplitude but exhibit a considerably higher ranking in constant deflection amplitude and strain testing. Also needed consideration is the volume of material under stress in the product and its surface area-to-volume ratio. Because plastics are viscoelastic, this ratio is critical in that it influences the temperature that will be reached. At the same stress level, the ratio of stressed volume to area may well be the difference between a thermal short-life failure and a brittle long-life failure, particularly with TPs. Another factor is whether the product will be in an isothermal or adiabatic heat condition or its thermodynamic behavior. This heat condition is strongly dependent on the loading rate and environmental influences such as temperature, water, solvents, ultraviolet light, air speed, and others discussed in Chapter 4. There are different design approaches to eliminating the basically hysteretic heating. For example, using a plastic with a low viscous response to mechanical stress minimizes heat generation because this material is usually very stiff. Heat-transfer conditions can be improved by increasing the flow of air (see Fig. 3-65) or other coolant (water, gases, etc.) across the surface of the part. The part's design can also be altered to decrease mechanical energy input, slow the cyclic loading rate, increase the surface area for dissipating heat through fins and the like, and other alterations. As usual with plastics and other materials, sharp comers or abrupt changes in their cross-sectional geometry or wall thickness should be avoided because they can result in weakened, high-stress areas. The areas of high loading where fatigue requirements are high need more generous radii, combined with optimal material distribution. Radii of ten to twenty times are suggested for extruded parts, and one quarter to one half the wall thickness may be necessary for moldings to distribute stress more uniformly over a large area (see Chapters 7 and 11). In evaluating plastics for a particular cyclic loading condition, the type of material and the fabrication variables are quite important. Remember that the many plastics perform very differently, whether they are TPs, TSs, unreinforced and reinforced composites. The basic rules to providing fatigue endurance can now be summarized. First, fiber reinforcement provides significant improvements in fatigue with carbon fibers and graphite and aramid fibers being higher than glass fibers (see Fig. 3-68). The effects of moisture in the service environment should also be considered, whenever hygroscopic plastics such as nylon, PCs, and others are to be used (see Fig. 3-68 and Chapter 8). For service involving a large number of fatigue cycles in TPs, crystalline-type composites offer the potential of more predictable results than those based on amorphous types, because the crystalline ones usually have definite fatigue endurance (Chapter 2). Finally, for optimum fatigue life in service involving both high-stress and fatigue loading, the reinforced high-temperature performance resins like PEEK, PES, and PI are recommended (see Table 3-9).

SHORT-DURATION RAPID AND IMPACT LOADS This chapter has thus far dealt with the behavior of plastics and composites during shortand long-term loading conditions. As with any material, the properties obtained under

Table 3-9. a) Fatigue Endurance Limit Data of Reinforced Thermoplastics per ASTM D 671 at 1,800 cycles/min. Short-Fiber Molding Compounds Stress at Failure (psi) Cycles to Failure

Fiber Type and Content (%) Base Resin

Glass

Carbon

lQ4

HP

106

107

SAN

30

8,500

7,500

6,500

5,500

Styrene

30 40

8,000 9,500

7,000 7,750

6,000 6,500

5,000 5,500

Polycarbonate

20 30 40

9,000 12,500 14,500

6,000 7,000 8,750

5,200 5,500 6,100

5,000 5,350 6,000

ETFE copolymer

30

4,500 9,000

3,600 6,300

3,500 6,100

3,500 6,100

30

Polysulfone

30 40

14,000 16,000

6,500 7,750

5,000 6,000

4,500 5,500

Polyethersulfone

30 40

16,000 19,000 22,000

7,500 8,500 10,000

6,000 7,600 8,000

5,000 6,200 6,700

30

Acetal copolymer

30

9,000

7,000

7,000

7,000

Polypropylene

30

5,500

4,500

4,500

4,500

13,000

9,700

9,500

9,500

Polyphenylene sulfide

30

Nylon 6*

30

7,000

6,000

5,750

5,750

Nylon 6/10*

30 40

6,800

8,000

5,750 7,000

5,600 7,000

5,500 7,000

6,500 10,500

5,900 9,300

5,300 9,100

5,200 9,100

3,400 8,000 9,000 13,000 15,000 6,400

3,200 6,500 7,300 10,500 10,300 4,400

3,100 6,000 7,000 8,000 8,800 3,900

3,100 5,900 7,000 8,000 8,500 3,700

ll,ooo 13,000

7,200 9,200

5,600 7,400

5,100 6,500

7,200

5,800

4,900

4,750

18,000

17,500

17,500

17,500

Nylon 6/6 40

Nylon 6/6* 30 40 30 40 30t

Polyester (PBT)

30 30

Modified PPO PEEK *Moisture-conditioned to 50% RH. tGlass bead.

202

30 30

PLASTICS: DESIGN CRITERIA 203

Table 3-9. b) Elevated Temperature Property Comparisons of Short- and LongFiber Glass with Nylon 6/6 Composites (lCI-LNP) Long Fiber

Short Fiber

Property

30%

50%

30%

50%

13.8 7.8 14.3 5.27

14.3 5.3 17.4 5.50

19.2 5.6 23.7 8.90

7.3 9.5 7.4 4.80

7.8 6.2 8.9 5.19

8.3 6.8 10.0 7.51

At 300°F Tensile strength (10 3 psi) Elongation (%) flexural strength (10 3 psi) flexural modulus (10s psi)

12.8 9.3 13.8 4.64 At 400°F

Tensile strength (103 psi) Elongation (%) flexural strength (103 psi) flexural modulus (10s psi)

6.3 8.6 6.9 3.96

Data on long-fiber glass-reinforced grades are for Verton compounds.

'To convert psi to pascals (Pa), multiply by 6.895 X

UP.

such conditions provide a basis for understanding and characterizing their basic behavior. For the most part, many of the behavioral characteristics discussed are valid for a wide range of loading rates (see Figs. 3-71 and 3-72). There may be significant shifts in behavior, however, at load or strain durations that are much shorter than those discussed, which usually take about a second or less to perform. This section deals with loading rates that are significantly faster than those covered so far, namely rapid and impact loading [l, 2, 11-14, 62-68, 268, 269, 302-5). Designers with a background in using other materials will recognize both the similarities and the differences in the behavior of the plastics discussed. As an example, impact resistance has also been a continuing issue with other engineering materials, particularly certain metals [268] with similarities to many of the phenomena observed in plastics. The concept of a ductile-to-brittle transition temperature in plastics is likewise well known in metals, notched metal parts being more prone to brittle failure than unnotched specimens. Of course there are major differences, such as the short time moduli of many plastics compared with those in steel, that may be 30 x 106 psi (207 x 106 kPa). Although the ductile metals often undergo local necking during a tensile test, followed by failure in the neck, many ductile plastics exhibit the phenomenon called a propagating neck. These different engineering characteristics also have important effects on certain aspects of impact resistance.

Deformation and Toughness Deformation is an important attribute in most plastics, so much so that it is the very factor that has led them to be called plastic. For designs requiring such traits as toughness or elasticity this characteristic has its advantages, but for other designs it is a disadvantage.

204 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

However, there are plastics, in particular the RPs, that have relatively no defonnation or elasticity and yet are extremely tough (see Fig. 3-73). This type of behavior characterizes the many different plastics available (see Table 3-10). Some, tough at room temperature, are brittle at low temperatures. Others are tough and flexible at temperatures far below freezing but become soft and limp at moderately high temperatures. Still others are hard and rigid at nonnal temperatures but may be made flexible by copolymerization or adding plasticizers . By toughness is meant resistance to fracture. However, there are those materials that are nominally tough but may become embrittled due to processing conditions, chemical attack, prolonged exposure to constant stress, and so on. A high modulus and high strength, with ductility, is the desired combination of attributes. However, the inherent nature of plastics is such that their having a high modulus tends to associate them with low ductility, and the steps taken to improve the one will cause the other to deteriorate.

Crack initiation

l

1

r

Fiber buckles

1

Fiber breaks

Slot/crack

Crack propagation

II

~

I

~

Adjoining fibers buckles

Adjoining fibers break

Crack propagates

Fracture

~

I

~

ROO~l cross section

Open surfaces still support during compression half of loading

t

~

Fails during tensile half of loading

Figure 3-70. A sequence of push-pull fatigue in unidirectional glass-tiber-reinforced plastics (TS polyester) by a microbuckling process.

10 1• 10 12 10 1: c .2

<0 E (;

10·

0

10 2

.J:

...u~

10 C

'"c:

10-·

2 .2 10"t>

a~

more

10·

~

"t>

I day

10· I cia.

I I I I I I

1st( Cr~

stress

and

ruptu~

Stlndard

0

u J! 10-·

stille testing

I I I I I I

10-'

10- 10 10- 12

10- 10

10- 8

10- 6

I I I

I I I I I

Raptd

ioadlnl

I I I

Impact

I

10-· 10- 2 10° Strain rate in.fin./sec.

102

10·

10'

Figure 3·71. The relationship of rapid-loading strain rates to those developed in other methods of testing. VElOCITY. FT.ISEC

1.000

TYPICAL CASES

-FIRED PROJECTILE - BA nED BASEBAll - PITCHED BASEBAll

100

-FOOTBAll HElMET - TEN·FOOT FAll -IZOO IMPACT TEST

10

-REFRIGERATOR OOOR·SLAM

1.0

0.1

-CONVENTIONAL TENSilE STRENGTH TESTS 0.01

Figure 3·72. Some typical velocities that refer to rapid loading. 205

Temperature. ° F

- 150 30

30

-60

120

210

300 20

10 ~--~----~--~----~--~7

HIPS I 3 ~--~----'----' 1 ---.--~2

~~~--~----~---+----i07

03 ~+-+---.---~---+~~02

=

...

13 a. E

01 ~--~--------+---~--~ 007

0 03 ~--~------~~--~---;002

00 L-__ __ - 100 - 50 ~

b.

~

o

____~______~0007

50

100

150

Temperature . oC

Figure 3-73. Some examples of toughness in plastics. a) Toughness related to heat deflection or rigidity; b) toughness or impact related to temperature for polystyrene (PS) and high impact, rubber modified, polystyrene (HIPS).

206

PLASTICS: DESIGN CRITERIA 207

Table 3-10. Examples of Toughness or Fracture Characteristics for Thermoplastics Material PMMA PA SAN

Polymethylmethacrylate Polystyrene Styrene-acrylonitrile copolymer

ABS CA HDPE PA PB PC POM PP PTP PVC

Acrylonitrile butadiene styrene Cellulose acetate High-density polyethylene Polyamide (Nylon) Polybutene Polycarbonate Polyoxymethylene Polypropylene Polyethylene terephthalate Polyvinyl chloride

LDPE PB

Low-density polyethylene Polybutene Polytetraftuoroethylene

TFE

Unnotched

Notched

Brittle

Brittle

Ductile

Brittle

Ductile

Ductile

Stress-Strain Behavior As previously described, the area under short-term stress-strain curves provides a guide to a material's toughness and impact performance (see Fig. 3-74). Soft, weak materials have a low modulus, low tensile strength, and only moderate elongation to break. (According to ASTM standards, the elastic modulus or the modulus or elasticity is the slope of the initial straight-line portion of the curve.) Hard, brittle materials have high moduli and quite high tensile strengths, but they break at small elongations and have no yield point. Hard, strong plastics have high moduli, high tensile strengths, and elongations of about 5 percent before breaking. Their curves often look at though the material broke about where a yield point might have been expected. Soft, tough plastics are characterized by low moduli, yield values or plateaus, high elongations of 20 to 1,000 percent, and moderately high breaking strengths. The hard, tough plastics have high moduli, yield points, high tensile strengths, and large elongations. Most plastics in this category show cold drawing or necking during the stretching operation. From a practical viewpoint toughness is readily understood, but technically there tends to be no scientific method of measuring it. One definition of toughness is simply the energy required to break the plastic. This energy is equal to the area under the stressstrain curve. The toughest plastics should be those with very great elongations to break, accompanied by high tensile strengths; these materials nearly always have yield points. One major exception to this rule is the plastic composites that use reinforcing fibers like glass and graphite. Stress-strain tests may be made in compression as well as tension. A modulus may be calculated from the initial slope of its curve. And materials under compression are much less brittle than when under tension. Thus, many plastics that are brittle when tested in tension become ductile and show yield points under compression, as, for instance, polystyrene. Typical values of ultimate strength in compression for many plastics are

c:

o

ro 01 c:

.2 LJ..J

a

.~ (3

:::J

o

Figure 3-74. Toughness tends to relate to the area under the stress-strain curve. The ability of a therrnplastic to absorb energy is a function of its strength and its ductility, which tend to be inversely related. The total absorbable energy is proportional to the area within the lines drawn to the appropriate point on the "curve" from the axis. The material in Area A is rubberlike and is just as tough (that is, of equal area) as material C, which is metallike. Most plastics, like material B, fall between these extremes, but some fall into both A and C.

Fle.ural modulus, million psi 4

(601 6

3

1651

0 (60)

2 -.-__

0

(401 6 (401 0

Short-fiber nylon, poly.ste,

~o

3

160)

0

(551

0

(401

Long-fiber compounds

6 2

0

4

5

PET

0

PBT

0

Nylon

Gloss shown ,n porenlheses

6

7

8

9

10

notched Izod Impact Strength, (tt.lb/in.)

Figure 3-75. Long glass-fiber RP molding compounds are tougher and more metallike than conventional short-fiber compounds. 208

PLASTICS: DESIGN CRITERIA 209

about twice that of the tensile strength. Flexural strength tests in which part of a specimen is under tension and part under compression generally give values of ultimate strengths that are between the values for ultimate tension and compression.

Processing and Material Behavior The flow patterns resulting from the conditions of a particular fabricating process are very important in affecting impact strength. Specimens with their molecules or fibers oriented perpendicular to the plane of fracture will exhibit higher impact energies than those with their molecules parallel to the fracture plane. Because the molecules tend to align in the direction of flow during processing, the designer should be able to judge which direction in the finished product will be more brittle. Stress concentrations and unfavorable molecular orientations should in any case not be located at the same place in a design. Reinforcing fibers, specifically the glass fibers, are brittle. Thus, when they are used in conjunction with a brittle matrix, as are certain TSs, it might be expected that the composite would have low fracture energy. In fact, this is not true, and the impact strength of most glass-TS-reinforced plastics is many times greater than the impact strengths of either the fibers or the matrix. An impact strength is higher if the bond between the glass fibers and the matrix is relatively weak, because if it is so strong that it cannot be broken, cracks will propagate across the matrix and fibers, and very little energy will be absorbed. Thus, there is a conflict between the requirements for maximum tensile or flexural modulus or strength (long glass fibers and strong interface bonds) and maximum impact strength (see Fig. 3-75).

Short-Duration Loads Two situations need to be considered when evaluating materials for their response to short-duration loads: rapid loading and impact loading. With rapid loading the loading rate may be much higher than that imposed in basic short-term tests (see Fig. 3-76). However, the loading rate is still less than the transit rate of the stress waves that develop under the applied stress. Stress waves usually travel at about the speed of sound in the material. This loading condition may be considered dynamic in nature, but it usually does not involve collision of the load with the test specimen or product, as occurs under impact. With impact loading, the loading is faster than the transit rate of the stress waves. This loading condition involves a collision of the load with the test specimen or product [62-68].

Rapid Loadings Typical standard test rates in basic tests for plastics vary from 0.0017 to 0.25 mmlmmIsec (0.10 to 15 in.!in.!min.) per ASTM D 638. In standard tests the softer plastics and rubbers are tested at the higher rates, and the rigid plastics and reinforced plastics are tested at the slower rates. Certain high-speed testing machines are capable of applying loads at rates that are much higher than those used in determining the basic behavior in standard tests. Behavior can be characterized over a range of stress or strain rates to determine any behavior changes under rapid loadings. Load and sample elongation may be recorded throughout

210 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

the test, thereby providing a full description of the elastic or viscoelastic modulus, the yield point, and post yield behavior (if any), all of which vary with the rate of load application. These data can be most useful in design. An example of the type, significance, and character of the data obtainable under rapid loads is shown in Figure 3-76. The data shown describe behavior over a spectrum of elongation rates up to several orders of magnitude higher than those obtained in standard tests. These data illustrate the following trends for the specific materials examined: a) tensile strength usually increases with higher strain rate, for all plastics; b) the elongation at break decreases with the strain rate; c) energy to failure, as determined from the area under the stress-deformation plot, generally decreases or remains the same with an increasing test rate. Moreover, a different plastics show a markedly different rate of decay of failure energy with increased test speed. A good example of the usefulness of such data is illustrated by comparing the energy to failure of ABS and PVC at low and high rates of test. PVC shows a much higher energy to failure at lower rates. At the higher test rates, PVC's performance is not much better than that of ABS. Note that these are not general conclusions for the two materials. Their relative behavior may shift drastically, depending on the temperature, the extent of plasticization or modification of the PVC, acrylonitrile and butadiene rubber contents of ABS, and their past exposure to aggressive environments. Behavior under rapid loading can be viewed as merely an extension of behavior obtained under short-term loads. The effects of the load rate and the dynamic effects of rapid loading must of course be recognized in designing a particular structural component. Rapid loadings may be imposed on structures under vehicular traffic, wind gusts, water hammer, in forced movements resulting from vibrating mechanical equipment, and the like. Impact Loadings

Whenever a part is loaded rapidly, it can be said to be subjected to impact loading; Any product that is moving has kinetic energy. When this motion is somehow stopped because of collision, its energy must be dissipated. The ability of a plastic part to absorb energy is determined by such factors as its shape, size, thickness, type of material, method of processing, and environmental conditions of temperature, moisture, and so on. Although the impact strengths of plastics are widely reported, these properties have no particular design value. However, they are important, because they can be used to compare the relative responses of materials. Impact strength can pick up a discriminatory response to notch sensitivity. A better value, impact tensile values, is unfortunately not generally reported. With limitations, the impact value of a material can broadly separate those that can withstand shock loading from those that fare poorly in this response. Of great importance is that they can be compared to the impact performance on the fabricated products. The resulting guidelines will be more meaningful and empirical to the designer. To eliminate broad generalizations, the target is to conduct impact tests on the final product or, if possible, at least on its components. In conducting impact tests on products the usual problem that has to be resolved as well as possible is how it should be conducted. The real test is after the product has been in service and field reports are returned for evaluation. Regardless, the usual impact tests conducted on test samples can be useful if they are properly coordinated with product requirements. The typical tests for impact loading are now reviewed.

10

102

TEST RATE, mm/min.

103

104

105

106

10 7

ISO

20

Go

Il.

·iii

~

.:tI.

r..

I-

~ w

a:: l-

16

100

12

2 w

l-

I

I-

~ UJ

a:: I-

VI

W ...J

Tensile Strength

0

In

8

50 elastomer

4 0

10

U.I

500

10

102

102

...J

~ w

I-

103

TEST RATE, in./min.

104

TEST RATE, mm/min.

103

UJ

104

105

105

106

0 106

10 7 b.


~ 400

Tensile Elongation at Break

w

a::

CO

I-

-<

300

Z

0 200 i=

-< ~ 100 0

...J

W

10

10

102

103

TEST RATE, in./min.

104

TEST RA TE, mm/min. 103 104 I 105

105

106

107

Nc!
000------........- -....- - - -....- - - , 2

~ lwo

N

E c.

~ ~

I~~

-

~600

~3

...J

~

< ~

!

0.5 >-

~~

g

0

I-

>- 200 0

Energy to Failure

..

U~•.-I--~----.--10

~

----9----.nh~--I~OS~-I~A~0 W 102 103 lIT'" uv

TEST RA TE. in./min.

Figure 3-76. Examples of type, significance, and the character of data obtained under rapid loading or high-speed tests. 211

212 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Izod and Charpy Testing

The most common impact tests for plastics and metals are Izod and Charpy tests (see Figs. 3-77 and 3-78). A heavy hammer at the end of a pendulum arm swings down and strikes a cantilevered specimen (in the Izod test) or a beam sPecimen supported on both ends (in the Charpy test). The specimens are unnotched unless the data reports that they are to be notched; for information on specimens and test procedures see ASTM D 256. The kinetic energy of the hammer is large enough that its velocity can be considered constant during the impact. The energy required to break the specimen is determined from the maximum height to which the hammer rises after breaking the sample. The impact strength is defined as the energy loss by the pendulum caused by the impact divided by the thickness of the specimen (see Figs. 3-79 through 3-81). The thickness of the specimen along the direction of the notch or unnotched part and perpendicular to the hammer movement at impact can be varied to give a reasonable energy-absorption value. The geometry of notched impact specimens and in the finished part can have profound effects on impact strength. This notch sensitivity is influenced by the different parameters shown in Figures 3-79 through 3-81. In Figure 3-81, rigid PVC shows a high sensitivity to the radius of the notch tip, whereas the normal-impact ABS is almost insensitive to it. For a given impact condition, the effect of decreasing the notch radius is to increase the strain rate at the notch tip. The effects of notch geometry on PVC and ABS materials can also be qualitatively determined where the energy-to-failure of the PVC decreases at higher strain rates, whereas this property is fairly constant for ABS. These observations apply specifically to the test conditions and specific compounds tested.

Striking Edge of Pendulum

Pointer and Pendulum after striking

!

'.

ISN~~~"'''' .• ;: .' .' '~~pporting

%'" x %... x 2 Y.r '"

Clamp

~I---'_ "

-oJ

Figure 3-77, A schematic of an Izod impact tester. A free-swinging pendulum strikes a vertically supported specimen held in a clamp; a notched specimen is located so that its notch is level with the top of the clamp and is facing the pendulum, per ASTM D 256.

r-----0.315 in. (8 mm) rod

30° ± 2°

Striking edge

Specimen

0.039 in.

Anvil

(1 mm/rod)

1.574 in. (40 mm)

Center of strike (W/2) Specimen support

Figure 3-7S. A schematic of a Charpy impact tester. The specimen is supported as a horizontal simple beam, per ASTM D 256. IZOO Impact Strength, Joules/em

1234567891011121314 ABS

Acetal Acrylics

Nylon Polycarbonate Reinforced polyester

Polyethylene Polyimide

Polystyrene PVC Cloth phenolics

2 4

6

8 10 12 14 16 18 20 22 24

IZOO Im~dct Strength. It Ib/ll'.

Figure 3-79. An example of Izod impact strength for various plastics, conducted at room temperature. 213

Radius of notch (mils) 10 14

~:f50

13 12

9 ~ 8 -;:; 7

~ 2.54 mm

~

8

0.500

~pThiCkness

~ 11 ~ 10

'" .~

0.600

0.400

c ~

~

along notch

OJ C

0.300

.'-;:;"

0.200

8

0.100 in

6 5 4 3 2

-'"

'" u .;;

0.100

2.5

X

10- 2

1.25

X

10- 1

6.25

X

10- 1

Radius of notch R (mm)

Figure 3-80. Izod impact strength for polycarbonates at various temperatures and notch radii.

VI :::J -200

0.25

NOTCH TIP RADIUS, mm

0.5

~

1.0

I

n::: 180 U 160 Z

I

standard notch tip radius

b

1 _I

I

E140

I I

)-E

1--

II I- .~

I - unplasticized PVC

10

~~ 100

2- polypropylene homopolymer

3- high density polyethylene 4- polypropylene homopolymer 5- polypropylene copolymer

"10

In::: U<

b~

6- high density polyethylene

Z<

78910 -

l-

V)

I-

<

UJ ::)

..J

< >

ABS (high impoct) polypropylene copolymer toughened polystyrene ABS (normal impact)

~

!

0

0_0 I 0.02

0.04

0.08

NOTCH TIP RADIUS, in. Figure 3-81. Comparative Izod impact strength notch sensitivities for a range of thermoplastics. 214

PLASTICS: DESIGN CRITERIA 215 TENSILE IMPACT

TEST BAR

~c..---

ANVIL

Figure 3·82. A schematic of a tensile impact tester.

Tensile Impact Testing

This test uses a swinging pendulum similar to that in the Izod test, except that this sample specimen is a relatively small tensile bar (compared to the standard static tensile specimen), which is mounted as shown in Figure 3-81, per ASTM D 1822. It involves determining the energy required to fracture a sample under tensile impact loads. One end of the dumbbell-shaped specimen is mounted in a pendulum. The other end is gripped by a cross-head that travels with the pendulum until the instant of impact, when the crosshead is suddenly arrested at the bottom of the pendulum's swing. The strain rate induced during this test is lower than that developed in Izod test(s) but is significantly higher than that of the standard tensile tests for basic properties. Like other impact testers, the tensile impact test has many of the advantages and disadvantages of the Izod test. A principal advantage is that it can be used to evaluate films, thin sheets, and soft materials that cannot be tested by the Izod method. Falling-Weight Testing

In this test a weight or dart is dropped onto a flat disc of the material being tested. The leading edge of the weight where it impacts the specimen has a specific size. The specimen, larger than an Izod one, is supported or clamped to a metal ring, depending on the test procedure being used. Figure 3-83 shows one example of a falling-dart test. In certain applications this test is valuable for ranking materials based on larger-sized specimens. The effects of impacts on actual parts can also be studied. Standard ASTM D 3029 describes this method of testing with a falling weight. Since the extent of the damage cannot be determined on unbroken specimens, each specimen is tested only once, so at least 20 specimens are required. If a specimen is not broken, weights are added to it according to a specified schedule until failure does occur (with certain test procedures the drop height is changed). The method for analyzing data that is specified in ASTM D 1709 is an excellent technique for evaluating the statistical variations of impact behavior. In this method

216 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 3-83. A schematic of a falling-ball tester.

samples are tested at various drop weights between which either none or all of the specimens fail. The percentage of specimens that breaks is then plotted against the impact energy. Doing so yields a straight line on probability paper. Such plots provide information as to the probability of failure of a material under light impacts. The less steep the curve is, the greater is the chance of these light impacts producing some failures. It is possibleeven likely-for two batches to have the same values of impact energy at which 50 percent will fail under a particular energy impact (a common quality control standard) and yet will have quite different slopes and behave differently during other impact loadings.

Gardner Testing This test, often called that of the falling dart, varies slightly from the falling-ball test. In a Gardner test the ball rests directly on the specimen and a weight falls on the ball, whereas in the falling dart one the ball itself falls from a specific height. With the Gardner test failure is defined as a crack on the tensile (that is, nonimpacted) side of the specimen. However, this determination is somewhat subjective. For more-reliable results, at least twenty-four specimens should be used.

Dynatup Testing One of the most complete impact tests from the standpoint of the information it provides is the Dynatup impact test. In this test the impact energy is delivered in much the same way as in the falling-weight test. In both tup is dropped from some height onto a specimen. However, in the Dynatup test the same amount of impact energy is delivered to each sample, an energy level that is high enough to ensure that each sample breaks. Data from each impact are then recorded (stored in a computer). From this information the crackinitiation energy, crack-propagation energy, and total energy absorbed by the material during impact can be calculated and plotted (see Fig. 3-84). Of all the impact tests, the Dynatup is considered the most complete one. However, it is very expensive compared to the others and is more difficult to adapt to nonambient test temperatures.

PLASTICS: DESIGN CRITERIA 217 300 280 260 240 220 200

~ 180

ci 160

140 6 120 ...J

100 80 60

2

4

6

8

150 140 130 120..; 110 ~ 100 ~ 90 m z 80 r'I1 70 ::0 Cl 60 :< 50 ::1 40 i~ 30 20 10 0

TIME. MSEC

Figure 3-84. Plotted information on load and energy versus time that was elicited by using a Dynatup tester.

Other Tests There are other tests used to evaluate the impact behavior of plastics, including many special setups just for evaluating specific products. These special test setups are usually applicable only to one company or association that has found them to be proven useful. Some of these special tests have then become standards in their own industry, such as ASTM and UL guidelines. Among the more popular ASTM standards are its 0 256, 0 1709, 0 1822, 0 2289, 0 3029, 0 3099, 0 3420, and a few others. In these tests including the special ones, a material is impacted by using various devices such as a ball on a pendulum or puncture tests (see Fig. 3-85), air-driven spherical or piercing-type missiles and others. Factors Influencing Impact

The impact results from each of these tests are extremely sensitive to many different variables, including practically all those throughout this book that have any influence on plastics' performance. At this point it might appear that all impact testing is futile, but this is not the case. Just as in running other tests, the results of impact tests all have meaning as long as some logic is used in evaluating them. An important variable in regard to impact testing, as well as other types, involves the fabricating conditions. Figure 386 is an example relating to melt flow during the injection-molding process, and Figure 3-87 shows the effect that gate locations can have on mold cavities (see Chapter 7). Annealing plastics to relieve or stabilize fabricated parts for certain plastics can also have an effect (see Fig. 3-88). Impact behaviors are affected by fabrication defects such as internal voids, and inclusions and additives such as pigments, allof which can cause stress concentrations within the material. In addition, internal welds caused by the fusion of partially cooled melt fronts, even those with single gates, usually tum out to be areas of weakness. The surface finish of the specimen may also affect its impact behavior. Machined surfaces usually have tool marks that act as stress concentrators, whereas molded surfaces have a characteristic "skin" that may offer some protection against crack initiation.

Displacement. in. Q8

1~

12

25.------:.,.:----..,.-----=T=----..:;,.:-------r------.5.63

20

z

15

:; Metal clamps

-'"

.,;

x

£

'"0

.,;

...J

'"0

...J

10

5~---1--~~_+--~~+---~dr---_+---~

~---~---~10----1~5---~2~0---~25~--~3g· Displacement. mm

Figure 3·85. An example of plotting different loads versus their displacement for a puncture test.

_10

z

u

~

......

~ 8 ...J

t-=

u..

Z 6

I(!)

Z

IJJ

a::

ICJ)

4

NOTCH PERPENDICULAR TO FLOW

IU

~

::E

c o

N

2 ··-----.~__~.~____ NOTCH

"'PARALLEL TO FLOW

O~--~~--~~--~----~----~

MELT TEMPERATURE (OF) Figure 3·86. An example of an injection molding melt flow's orientation in a cavity on the lzod impact strength of a given PVC compound. 218

PC

ABS

PVC (1)

SI NGLE GATE, NOTCHED

15.2

2.4

2.6

14.7

DOUBLE GATE, NOTCHED

14.8

2.1

1.2

0.5

3.1

1.5

2.6

DOUBLE GATE, UNNOTCHED

PVC

(2)

Figure 3-87. During injection molding, a double gate can cause weld lines, when two melt fronts meet in a cavity, and reduce impact strengths, as well as other properties. Single-gated and double-gated Izod impact-test specimens were molded using different plastics to provide impact-strength comparisons.

219

220 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 100 ~ae

o .

.. ..'" .... .. '"

80

!!~

o c: .2 c:

c: • ;; ..

60

l-

'--~

,

i I

II

I \~ .............

.. =. 40

aoc-

i

!

140C

I

I

100C

~

I

a:.. .•E

I

120C

20 0

10

5

15

Time of annealing. hr

Figure 3-88. An example of how annealing time affects Izod impact strength retention. Annealing is the process of subjecting fabricated plastic parts to specific heat-time cycles to relieve or stabilize fabricated parts when so required.

ASTM Melt flow rate g/10 min 1000 ~

2.

£ g> ~ 1;;

~ c. E

-g

2.5

800F-__~9________--, 18

I I

I

600 400 200

N

!---VI I

I I I

I

I

Critical thickness ~

I I I

I I

L---

O~~~------~------~----~

Specimen thickness (mml

Figure 3-89. An example of the effect of specimen thicknesses on Izod impact strength for various molecular weights or melt-index values for polycarbonates.

However, if the molded surface becomes scratched, this protection no longer exists. In addition, moldings occasionally have for decorative effect an embossed surface that can cause a considerable reduction in impact strength as compared to a plain surface. Sample thicknesses can also have an effect (see Fig. 3-89). As the thickness increases the impact usually also increases, since there is more material to carry the load. However, some materials have a critical thickness above which the impact drops off sharply.

Designing with Impact Data A real problem with specifying the impact properties on a product is to define what is a failure. The designer has to decide, for example, to what degree a surface condition, dent, bend, breakage, or shatter will be defined as a failure. Thus, the failure of a product can take many forms. There are situations, for instance, in which excessive elastic deformation will constitute failure. In tum, the definition of failure has to be related to impact testing's results.

PLASTICS: DESIGN CRITERIA 221

A plastic automotive bumper is a good example of how to identify failure. A bumper system is required to absorb specified levels of energy and simultaneously protect the rest of the automobile from damage. If a plastic bumper can withstand an impact without being damaged but undergoes such a large displacement that it dents the automobile's front, it has failed in its function. In other applications the criteria defining failure may be linked directly to damage and be influenced by the situations covered in this section on impact and particularly the last portion, on influencing factors, which includes environment. The environment may have significant effects on impact behavior. When certain plastics are exposed to sunlight and weathering for prolonged periods, they tend to become embrittled through degradation. Alternatively, if the plastic is in the vicinity of a fluid that attacks it, the crack initiation energy may be reduced. Some plastics are affected by very simple fluids, such as domestic heating oils, which act as plastizers for conventional PEs. The effect water can have on impact behavior can be either a disadvantage or an advantage. Various plastics will undergo a change in failure mode as their temperature changes. The usual ductile-brittle transition is specific not only to the generic class of the material but also to its specific grades (see Figs. 3-90 through 3-93). Plastics are easily modified (see Chapter 2) so that their performance in respect to temperatures can change. The PC transition can be lowered from at least -18 to - 29°C (0° to - 20°p). With high-impact ABS, temperatures can go down to - 40°C - 40°F). The behavior of PS can be changed (see Fig. 3-73).

1nc.,eaSlng loug

PoIysly,_ PoIy"",'hylmt'lhoc,yICl'. GlCIss-Iolle
.

.... ,

•:0:

..

.. .

ss

..

.'

.

.

"as

PoIyco.bonCIle NyICin I~II FblytetrolluOlWthyl_

Low -denslly poIyelhy_

Figure 3-90. A comparison of impact strengths as measured by the Charpy test at different temperatures.

15r-------~--------~------~r_------~

I

Nylon Iwetl

-

0·25mm notch

10

E -,

:Ie

.c

0. c: ~

iii U 0

Q.

.§

Acetal

5

OL-______ -40

~

______

~L_

-20

_ _ _ _ _ _J __ _ _ _ _ _

0

~

20

Test temperature

40

loe I

Figure 3-91. A variation in Izod impact strength with temperature for several thermoplastics. TEMPERA TURE, °c -10

-30

0

0

14

.. ":1

;J

.0

-;

].

I:

....

~I

15

....a:: ....

I....

!iUJ

....a:: ....

V)

V)

U

U

< Q.

10

~

....

< Q. ~

....

I: I..:l

I: I..:l

W

W

~

5

!i::J

;;i. 2

!i::J .J

< u..

u..

0

~

20

0

20

40

TEMPERA TURE,

or

60

80

Figure 3-92. The effects of temperature on the falling-weight impact strength of different thermoplastics. 222

PLASTICS: DESIGN CRITERIA 223

A statistical analysis of the data and careful interpretation of the results is required in any evaluation of impact strength, because variability or scatter in impact-strength results is the rule rather than the exception. It would not be uncommon, for example, to observe a fourfold difference between the highest and lowest values in a population of 100 TP samples. Impact testing with a product at its expected end-use conditions is always the best way of determining its behavior during use or abuse. However, specimen testing definitely has merit and can provide guidelines, particularly when it is properly analyzed. It must be recognized that conducting impact and other tests can be difficult or impossible on fabricated parts. This is to because many different impact actions can occur that are not repetitive or otherwise easily amenable to testing. ELECTRICAL PROPERTIES

Plastics and composites offer the designer a great degree of freedom in the design and manufacture of products requiring specific electrical properties (see Fig. 3-94). Their combination of mechanical and electrical properties makes them an ideal choice for everything from tiny electronic components to large electrical equipment enclosures. The

3

Thickness (mm) 5 6 7

4

9

8

20r-+_---r--~----+_--_r--~~--~

1.0 18 0.9 16 0.8 c:

"

14

,Q

~ c;,

.<::

12

c:

~ 10

Ci5 U co

~ 8

l

6

0.3

4

0.2

2

0.1

o~--~--~----+----+----~--~~

.100

.140

.180

.220

.260

.300

.340

Thickness (in)

Figure 3-93. Izod impact strength of a PC (MerIon) versus its thickness at various temperatures, per ASTM D 256. Courtesy, Mobay Chemical Corp.

224 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 3-94. This wire system for GE Plastics' Living Environments Concept House seen in Chapter 1 is made of GE's Noryl resin, for high heat resistance and easy processing. The raceway is a wire channel that runs along the wall, much like a baseboard, to contain electrical outlets, receptacles, telephone plugs, and cable connections. This extruded resin prototype raceway, which is featured in every room, as a UL 94 flammability listing of V-D.

most notable electrical property of plastics is their ability as good insulators, but here and throughout this book there are considered many other important electrical properties available to the designer working with different plastics [12, 14, 44, 54, 62-68, 207,

306-8]. The development of many different plastic compounds and basic polymers continues to expand the use of plastics in electrical applications. However, it is important to understand the factors that can affect their long-term performance. Figure 3-95 and Tables 3-11 and 3-12 show that there are different orders of magnitude between plastics and metals. Depending on the application, plastics may be formulated and processed to exhibit a single property or a designed combination of mechanical, electrical, chemical, thermal, optical, and aging properties and others. The chemical structure of polymers and the various additives they may incorporate provide compounds to meet many different performance requirements.

PLASTICS: DESIGN CRITERIA 225

Table 3-11. The Conductivities of Different Materials Conductivity

IT

Density

Material

S/cm

g/cm3

Copper Silver Aluminum Polyacetylene with iodine Polypyrrole with phenyl sulfonate Polystyrene

5.9 X 105 6.3 x 105 3.6 x lOS 2.0 x 10" 1.5 x 102 10- 16

8.9 10.4 2.7 0.8

1.3 1.05

0

ITlo S/(cm2 g)

6.6 x 6.0 x

10" 10"

1.3 x 105 2.5 x 10" 1.2 x 102 9.5 x 10- 17

Performance The electrical properties of plastics vary from their being basically excellent insulators to being quite conductive in different environments. Figure 3-96 illustrates some of the properties used in design. There are several U.S. and foreign test specifications available for characterizing plastics' electrical properties. The major testing organizations that set the conditions and specifications pertaining to electrical properties are the American Society for Testing and Materials (ASTM), the Canadian Standards Association (CSA), the Underwriters Laboratories (UL), the International Electrotechnical Commission (IEC), the International Organization for Standardization (ISO), and the American National Standards Institute (ANSI). Examples of different properties with different plastics according to ASTM test methods are given in Tables 3-13 through 3-17 and Figures 3-97 and 3-98.

Conductive Plastics and EMI The use of electronics has shown large growth in a variety of kinds of equipment, such as for data processing, transportation and industrial controls, automation, and medical devices. As traditional metal housings for plastic housings continue to become more widespread, the issue of electromagnetic compatibility (EMC) has arisen. EMC is the ability of an electrical device to function normally without interference from or interfering with another electrical device. EMC regulations usually emphasize the containment of electromagnetic interference (EMI) to specific levels across the designed frequency ranges.

Table 3-12. Typical Conductivities of Compounds with Different Additives and Fillers Conductivity Fillers

S/cm

Carbon black Aluminum platelets Steel fibers Carbon fibers Mica coated with nickel

0.01 to 0.1 I to 50 I to 50 0.1 to 10 I to 10

IT

10" 10" 10"

Base polymers

10" 10" 10" 10" 10" AnlISI&I IC malenals

10'0

:;;

::>

~

" C ~

~

;;; ;;; ~

...

u

~

:;

on

1()9

lOA 10' 1()6

S1""C dlsslpa"ve malenals

10 10' 10' 100 0 01 001 10 ' 10 •

Figure 3-95. The surface-resistance spectrum of polymers, conductive plastics, and other materials.

R

®

~ 0

+

r ~J,

LEGEND OIEt.ECTRIC STRE GTH DIElECTRIC CONSTANT DI~~IPATlO" FACTOR • VO~U"E R[ ~ISTIVII' ~ SU RFACE RESISTIVITY 6 ARC RES IS lANCE 7 CORONI R[SIS TANCE

0)

"1

......._ _ _--'3 SPECI" EN

Figure 3-96. Schematic illustrations of electrical properties. 226

=

ELECTRODE

I-.J I-.J

'I

10·

Cuo Zno

Vinyl

10 ' •

10 "

~

101/

Figure 3-97. The spectrum of volume resistivity.

Porcelain

10"

Ceramics

10"

Insulators

resin

10"

Epoxy Resins. Polyesters PolYVinyl Butyral Plastics

Polyethylene Ethyl Cellu· lose Nylon

1_

10' 1

Phenol FormaldeCellu lose hyde Acetate Urea Formaldehyde Cast Phenolics

Celluloid

10' 10'

Semi-Conductors

Conductors

Ni

10 •

Teflon

10 •

Polystyrene

10 •

10"

10'·

10' I

228 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

20

40

"10 fill

60

80

100

Figure 3-98. An example of how additives or fillers can be compounded into many plastics to provide a wide range in the dielectric constant.

The nonconductive characteristics of plastics can become a major drawback in certain applications. Because they are electrical insulators, they do not shield electronic impulses generated by outside sources. Nor do they prevent electromagnetic energy from being emitted from equipment housed in a plastic enclosure. Government regulations have already been set up requiring shielding when the operating frequencies are greater than 10 kHz. Every electronic system has some level of electromagnetic radiation associated with it. If this level is strong enough to cause other equipment to malfunction, the radiating device will be considered a noise source and usually be subjected to shielding regulations. This is especially so when EMI occurs within the normal frequencies of communication. When the electronic noise is sufficient to cause malfunctioning in equipment such as medical devices, however, the results could prove life threatening. Reducing the emission of and susceptibility to EMI or radio frequency interference (RFI) is thus the prime reason to shield medical devices in whatever their type of housing, including plastic. Plastics alone lack sufficient conductivity to shield EMI and RFI interference, however. Designers can reduce or eliminate sufficiently electromagnetic emissions from plastic housings like those of medical devices and computers just by shielding the inner emission sources with metal shrouds in the so-called tin can method. They may reach the same effect by designing electronics to keep emissions below standard limits or by incorporating shielding into the plastic housing itself. Designers and engineers will often employ all

<.Q

N

N

10 14 10 16

10 16 109 10 14

10 15 10 16 10 14 1013

10 14 10 14 10 14 10 16 1013

2 x 10 16

ABS Acrylic Cellulose Ester FEP Nylon 6 Polycarbonate Polyethylene Alkyd DAP (SDl5) Phenolic MFE Epoxy

10 18 3 x 10 15 10 18 10 15 10 16 10 19 1013

Surface

Volume

Material

Resistivity

.005/2.9 .062/3.6 .00613.8 .000512.1 .03114.2 .00113.1 .000112.34 .02/6.0 .026/3.8 .01315.4 .004/3.22

100 Hz .006/2.8 .058/3.2 .01ll3.6 .0005/2.1 .024/3.8 .0013/3.1 .000112.34 .0215.8 .020/3.7 .013/5.3 .004/3.25

I kHz .008/2.8 .04513.1 .024/3.3 .0005/2.1 .03113.8 .00713.1 .000112.34 .01515.4 .016/3.6 .033/4.9 .004/3.25

I mHz

.00712.8 .033/2.9 .022/3.2 .000512.1 .020/4.0 .01ll3.1 .000112.34

10 mHz

Dielectric Constant/Dissipation Factor

.014/2.1 .0007/2.05

.000112.34

.015/3.1 .000112.34

.00112.7

1,000 mHz

.020/3.0 .0008/2.09

.005/2.7

100 mHz

Table 3-13. Resistivity of Volume and Surface and the Dielectric Constant at Different Frequencies

230 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 3-14. Dielectric Strength and the Dielectric Constant Material

Dielectric Strength (MV/m)

Dielectric Constant

ABS (high impact) Acetal (homopolymer) Acetal (copolymer) Acrylic Cellulose acetate CAB Epoxy Modified PPO Nylon 66 (33% glass) PEEK PET (crystalline) PET (36% glass) Phenolic (mineral filled) Polyamideimide Polycarbonate Polyetherimide Polyethersulphone Polyimide Polypropylene Polysulphane Polystyrene Polythene (LD) Polythene (HD)

25 20 20 25 11 10 16 22 15 19 17 50 12 23 23 33 16 22 28 16 20 27 22 45 14 30 25 15 12

2.8 3.8 3.8 3.5 6.0 5.0 3.5 2.6 8.0 3.2 3.3 4.0 5.0 3.5 3.0 3.2 3.5 3.5 2.3 3.0 2.5 2.3 2.3 2.1 3.1 6.0 3.0 5.0 5.0

PTFE PVC (rigid) PVC (flexible) SAN BMC (polyester) SMC (polyester)

three strategies in a single design. What is most important is to attempt to locate all the shielding in a relatively small volume within the larger housing and then tin can it, to provide a simplified solution rather than spreading it out. Among the shieldings incorporated into housings, the most popular and useful applied technologies are those for conductive coatings, zinc-arc spray, and electroless plating. Other methods include the use of conductive foils or molded conductive plastics, silver reduction, vacuum metalization, and cathode sputtering. Although zinc-arc spraying once accounted for about half the market, conductive coatings surpassed it and now maintains the longest single market share. The properties of various coatings are described in Figure 3-99 and Tables 3-18 through 3-20. Other conductive coatings are also used. Unlike other shielding methods, conductive coatings are usually applied to the interiors of housings and do not require additional design efforts to achieve external aesthetic goals. All offer trade-offs in shielding performance, the physical properties of the plastics, ease in production, and cost. Often, differences in test measurements and samples' configurations make comparisons difficult. The ASTM has a relatively new standard that defines the methods for stabilizing materials measurement, thus allowing relative measurements to be repeated in any laboratory. These procedures permit relative performance ranking, so that comparisons of materials can also be made. Nonetheless, the designer will still have to confirm the

PLASTICS: DESIGN CRITERIA 231

Table 3-15. Arc Resistance and Critical Tracking Index Material

Arc Resistance, s

ABS ABS-PC DAP POM PAR LCP MF Nylon 6 Nylon 6/6 Nylon 12 PAE PBT PC PBT-PC PEl PESV PET PF PPO (mod) PPS PSU SMA UP

89 91 150 220 78 192 180 60 60 120 125 184 10 99 126 20 125 190 34 60 120

Critical Tracking Index, V 400+ 250+ 600+ 600+ 200 175+ 600+ 600+ 600+ 600+ 175+ 600+ 100+ 260 100+ 150 250+ 100+ 400+ 130 100+ 600+ 600+

suitability of a material's shielding performance for each system through such conventional means as screen-room or open-field testing. Each approach to shielding should also be sUbjected to simulated environmental conditions, to determine the shield's behavior during storage, shipment, and exposure to humidity, which could accelerate the effects of aging of shielding materials. In this way degradation can be observed, along with other problems that might occur in a product's service life. The Underwriters Laboratories utilizes a combination of methods for environmental conditioning and adhesion testing to evaluate various approaches to shielding and to determine the plastic types that are suitable for use in electronic devices. Their concern is primarily safety should a metalized plastic delaminate or chip off, creating an electrical short that could cause a fire. To maximize results, with any product, the designer should reduce the circuit-noise generation and susceptibility of the product to as much as possible. Consider the choice of shielding early on in the design process, before deciding on final packaging, to minimize the amount of external shielding required. Doing so will also alleviate last-minute shielding fixes and, of course, a good deal of exposure and delay in marketing the product.

Design Concepts Many ideas for advancing electrical and electronic systems have been adopted since the early 1940s, which saw the start of high electronic frequency radar systems, but the

232 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

I_ -

.' '. 1 , '. ' .1 , I ! I f-- I- Elec:troct.g COIItings on 1/S-in. thick polycerbonete sh..t

10

I

L

""'- I'....

70 ~

...........

~

60

\\

~~ ..........::

,\ ~ ~

" ~

'-

20

"'-

~

"""-

10

o

1

:::-

-

....... ~ ~

~ ['...

'" "" r--....

0.'2 mI. raP"ite '\ 0 mIl \I ~

I

.............

. m s"eet _ liS-in. solid. alumlnU ~mi' sil"e!.--

-I .1

lu~inum

.ted a .c:\epoSI ----r

.-

\Iacu~ ~

--

............

.

I

""ite ~ (t\i' 9r8 I O'Ief '2.0........ ~ . ppef '2.0 mIl c~r-L..-- I---

I.

'2.0 mil copper I

~

10

--

//~

~

.....................

...........

=

I.~ silver oV'l'

-r-.-

1.5 mil carbon

I I

--

V

f-"""

1.0 mil graphite I

1

100

Frequency, MHz

1000

10,000

Figure 3-99. How conductive coatings compare in their shielding effectiveness on a l-in.-thick polycarbonate sheet.

earliest major use of plastics for electrical insulation early in this century come with the advent of developments in electrical and telephonic installations requiring insulation. Phenolic plastic compounds became the basic material for insulating wires and other related products. Since then many new plastics have been developed, for widely variant applications, some of which have been included elsewhere in this book and others of which are now discussed.

Folded Membrane and Snap Switches In controlling electronic devices such old standbys as pushbutton and thumb-wheel switches have been joined by a whole new generation of advances in materials and manufacturing like the folded-membrane switch (see Fig. 3-1(0), the elastomer/snap switch, and others. Membrane technology is the technique of producing flat, thin, lightweight switch arrays by joining two or more membranes, usually a TP polyester or a PC. Each membrane features etched or screened conductors placed face to face with a thin material to separate the active elements. A second, surface-printed overlay is attached to the top of the switch assembly to provide a graphic indication of the switch's location. Tactile feel is provided by inserting either plastic or metal snap domes inside the switch assembly. In the membrane technique snap domes provide only tactile feel and are not used as an active switch element, as they are in the case of plastic-elastometer snap-dome technology. The advantages of membrane systems are their low cost in high volumes, their moderate tooling charges, and their capability of providing attractive, bright, durable frontal graphics. These attributes account for the surge in the use of these switches in many applications,

PLASTICS: DESIGN CRITERIA 233

Figure 3-100. A folded-membrane switch that uses thin, flexible, plastic film layers.

including home appliances (particularly microwave ovens), computers, commercial and industrial controls, and all kinds of instrumentation. Elastomer-snap dome switches are rugged, watertight assemblies with a single-piece molded elastomer keypad, usually of silicone, over a printed circuitboard. Metal dome switches are assembled over switch conductors on the board that to date have been able to withstand the most severe environments. Their applications range from machine tools to medical electronics to military tactical ground and shipboard use. The tactile snap of the elastomer-snap dome combination is particularly effective in noisy environments where the operation of a switch must be felt through fingers or even gloves, rather than heard. The material in question can be a molded opaque or translucent elastomer, the latter being selected for applications where backlighting is required. Translucent silicone elastomers are good transmitters of light for smooth, diffuse illumination from intemallight sources such as incandescents, light emitting diodes (LEDs), or emitting lights (ELs). Injection-Molded Circuifboards

Injection-molded substrates for printed wiring boards have the potential of providing significant cost savings over standard epoxy or glass substrates. Such three-dimensional features as spacers, stand-offs, soldering sites, and slots for through-hole connections can be molded in initially rather than added later in costly handling operations (see Fig. 3-101). These features can all be incorporated with the high precision characteristic of the injection-molding process (see Chapter 7) [12]. The use of high-temperature TPs has made this technology feasible, since conventional commodity types of injection-molding materials cannot withstand the high temperatures encountered during soldering operations and actual use. Molded printed circuitboards (MPCBs) must be manufactured of high-strength, high-impact materials, since they are expected to withstand dropping and other abuse without breakage. Circuitboard materials must be chemical resistant to tolerate the various cleaning and processing steps as well as being fire-retardant to meet all relevant codes and requirements. Molded boards have the added advantage of possessing superior properties with regard to conductive anodic filament growth (CAP). Recent work by material suppliers, plating shops, and board manufacturers has made important advances in the metalization and processing of MPCBs. This survey provides

234 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Fully additive process

Screen or photoprlnt

Unclad base material

Adhesion promote and deposit electroless copper

Strip plating resist (optional)

Apply solder mask and protective coating

Figure 3-101. An injection-molded production technique for three-dimensional circuitboard packages where the metalizing process creates a conductor path.

a brief overview of MPCB technology, including its materials, molding, and subsequent processing. It highlights the opportunity this emerging technology holds for the electronics industry and for injection molders. Materials for MPCB applications fall into the general category of engineering thermoplastics. There are currently four classes of materials evaluated for this demanding application: polyphenylene sulfide (PPS), polysulfone (PS), polyether sulfone (PES), and polyetherimide (PEl). Various filled, reinforced, and blended formulations of these base resins have also been tested and often show important improvements in properties and metalization. All these materials show high heat distortion temperatures, with attendant high process melt temperature requirements. All require heated mold cavities with temperatures near or in excess of 149°C (300°F). Their high melt viscosities require unusually high clamping pressures. Glass fibers and inert mineral fillers can be used to provide additional rigidity, and it has been shown that reinforcements may be needed to prevent warpage during hightemperature wave soldering. Printed circuitboard substrates are produced in a conventional injection-molding operation. In injection molding, high-temperature polymer melts of the various engineering thermoplastics discussed are injected into precision steel molds in a complex, dynamic process. High molding pressures (often approaching 138,000 kPa (20,000 psi) are required for these materials, whose flow lengths are relatively short. The molding of larger boards

PLASTICS: DESIGN CRITERIA 235

(> 100 in. 2 ) and use of multicavity tools is, though difficult, not impossible, with proper

mold design and multiple gating. The injection of a high-viscosity melt into a narrow cavity can cause significant moldedin stresses from the orientation effects of the polymer flow. Although these residual stresses can be minimized with careful gating and mold design, MPCBs often require annealing to prevent warpage. Even slight imperfections in surface planarity are debilitating for PCB applications, since their plating will be defective. Metalization is also influenced by orientation in a manner unrelated to the planarity of the board.

Radomes A radome (radiation dome) is simply a cover for a microwave antenna used to protect the antenna from the environment (see Fig. 3-102). Such a dome is basically transparent to electromagnetic radiation and structurally strong. The need for being transparent to radiation rules out metals. The earliest radomes (1942) were of a rubber-coated, airsupported fabric, followed later by an RP made of randomly chopped short glass-fiber, mat-reinforced TS polyester. By 1943 the glass-fiber fabric-TS polyester or epoxy was in use that has been the industry standard worldwide ever since [1, 13, 14, 17, 19, 30, 32,44,49,54,67,68, 86, 310]. The shape of a radome, which is an important factor in its design, is normally chosen on the basis of the optimal electrical characteristics (see Fig. 3-103). For aircraft the ideal shape is a spherical surface with the antenna's gimbal point located at the center of the sphere. Since a spherical radome is virtually impossible to obtain, a simpler configuration of a hemisphere together with a right-circular cylinder are often used, resulting in 95 to 98 percent efficiency for a relatively low-loss dielectric construction material. Streamlined radomes usually must be a compromise between electrical and aerodynamic considerations. As a result, many of today's aircraft and missile radomes bear a strong resemblance to an icicle, both in terms of having an awkward appearance and in their optical or electrical properties. The typical all-plastic ground radome is spherical and constructed of a solid RP or an RP in a sandwich construction. All-plastic radomes up to 150 ft. in diameter have been built using sandwich construction (see Figs. 1-7c, d). Space-frame radomes consist of thin, flat panels of solid laminates in triangular, diamond, rectangular or hexagonal shapes. Each panel is bounded by stiff members to form a polyhedron approximating the shape of a truncated sphere. Although this approach was originally developed for 50- to 150ft. radomes, it also has economic advantages for use in smaller space-frame radomes. Ducted radomes are used that are hemispherical domes mounted on a cylindrical base. These consist of several "orange peel" side panels connected by a spherical cap panel. Such an RP structure can include hot-air ducts between the inner and outer skins of the side panels for anti-icing purposes. The widely used 55-ft.-diameter ground spherical radome consists of membranes of 1I16-in.-thick randomly chopped glass fiber in a TS polyester-resin matrix with the membranes formed integrally and having edge ribs about 114 to 3/8 in. thick and 3 to 4 in. deep. After assembly, by field bolting, the sections form a network of ribs supporting the membrane. The design must take into account both the positive wind load on the windward side and the negative load caused by reduced air pressure on the leeward side. Under compressive stress the ribs must not be unstable and allow the shell to collapse inward. The larger radomes, up to 150 ft. in diameter, either have a network of metal

236 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 3-102. A technician adjusts the streamlined nose radome for a McDonnell-Douglas DC10 commercial jet in preparation for a radar beam-pattern test. In the background is the transmitting antenna that serves as the radar signal source. This glass-fiber-TS polyester radome, almost seven feet in diameter, houses a radar antenna capable of radiating a beam of 60,000 watts at peak power over a 300-nautical-mile range and of scanning a ISO-degree sector of sky ahead to give an advance picture of the weather. A black neoprene covering 0.005 in. thick protects the nose from water erosion.

ribs supporting their RP panels (see Fig. 1-7c) or are hexagonal 6-in.-thick sandwiches with 1I16-in.-thick RP facings bonded to a resin-impregnated honeycomb core. This is the usual construction, but others exist. Properly designing a radome requires complete knowledge of its physical, chemical, and electrical requirements, as well as of the properties of the materials involved. Since the ultimate design criteria are high transmissibility with a minimum of distortion, the ideal wall design would be one that provided, 100 percent transmission and a constant insertion phase for all angles of incidence and polarizations, at all frequencies.

PLASTICS: DESIGN CRITERIA 237 Siroami,ned rarfnmo

I

Or lgmof wave-

~--

RofleClod wayo

.J

Radame TwtCe 'efl~clod wnwe

Figure 3·103. Radome configurations showing their effects on radar waves emanating from the radar reflector or antenna so that the waves are properly focused in the required direction. This is basically the same setup as when optical waves are transmitted through a transparent medium so as not to cause visual distortion.

From an electrical standpoint, a radome is concerned primarily with any loss of gain caused by the presence of plastics or other materials in front of the antenna. In certain applications, side-lobe changes, bore-sight shifts, and the rate of change of the bore-sight shift must be known. The wavelength changes in a way that is approximately proportional to the square root of the difference between the dielectric constants of two media. Therefore, the thickness of a radome's wall becomes exceedingly important in radome design. There are three common types of wall configurations. The thin wall has a small fractional physical thickness (and consequently an electrical thickness) that is usually less than one twentieth of a wavelength. The second type is the half-wave, a solid wall within a thickness of 180 electrical degrees. The third conventional type is the sandwich, which consists of two one-quarter-wavelength panels. The normal-incidence radome has most of its area illuminated approximately normal to its beam. The thickness of a radome's wall is usually constant, except where beam bending occurs in high-incident angles. It is then necessary to vary the thickness of the wall over the surface of the radome. The transmission properties are similar to those exhibited by plane or dielectric sheets when they are transferred by plane waves normal to the surface.

Piezoelectric Materials Piezoelectric transducers either convert an electrical signal to physical motion or vice versa. In the past, the sensor material was usually a ceramic based on titanates of barium or lead zirconium. Their rigidity made them especially useful for converting electrical energy to mechanical motion in products like audio speakers and signaling alarm devices. Unfortunately, ceramics are brittle and cannot be made with large surface areas or into complex shapes where plastics are also used.

238 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Piezoelectric plastic materials such as polyvinylidene fluoride (PVDF) overcomes such drawbacks, however, and offers many other design advantages. Large, thin sheets of PVDF can be cut or stamped into nearly any shape. This flexible film provides less transducing power than ceramics, especially at low frequencies, but it is more sensitive to mechanical stress and can withstand a higher voltage, operate over a broader frequency range, and respond to a wider degree of mechanical stress. These typical properties make it superior to ceramics for many sensor applications. The materials are available in different forms, such as thin sheets ranging from 0.4 to 30 mils thick. Generally, the entire upper and lower surfaces of the material are both metalized. For specific applications, such as keyboards, the metalization is removed from all but selected areas. Applying a DC voltage across piezo film makes the material thinner, longer, and wider in proportion to the voltage level. Reversing the voltage makes the film in tum thicker, shorter, and narrower. Strain constants define the relationship between the applied voltage and the induced mechanical strain. The constants have different values for each axis. Similarly, this film generates a proportional voltage when it is compressed or stretched. Stress constants also define the relationship between the applied mechanical stress and the voltage generated. There is one constant for each axis. Piezo film also produces an electrical signal when it is exposed to infrared light. When heat induces mechanical stresses in the material, a pyroelectric constant defines the resulting voltage. This film is more sensitive to stretching than to compression, as a result of its higher mechanical stress. For example, stretching the film can produce peak voltages from about 20 to 30 volts. A similar result occurs by bending the film over a sharp comer, an alternative and often convenient way of stretching the plastic. The film can also act as a microphone, if it is clamped around its edges so that its central portion is free to vibrate. The resulting audio response is good enough to faithfully produce human voices on a tape recorder. Piezo film acts as an impact detector when it is compressed between two plates. Attaching the film's output to an audio amplifier yields the sound of impacts made on the top plate or the resting surface. The response is the same, whether the top plate is heavily loaded or not. The point is that film response is independent of static pressure, because it responds only to pressure changes. The piezo qualities of PVDF are stable at temperatures up to about 70°C (IS8°C). When suitably treated, the film is stable up to 100°C (2l2°F) but is unfit for exposure above that point. Its dynamic sensitivity runs from about 10-8 to 106 N/m, a 286-dB range. With its many performance capabilities, PVDF has found many uses, a few of which are touched upon here. A simple type of PVDF sensor consists of a strip of film securely fastened to a surface of a thin, flat spring. As a force deforms the spring it also stretches the film, producing a signal. In some versions a permanent magnet or electromagnet exerts force in a manner analogous to that in reed switches. Keyboards, for instance, use sheets of film containing multiple sensor elements that under the films' surfaces are completely metalized. The upper surface will be metalized only in the areas corresponding to the keys and their respective conductors. The sheets of film are secured to a supporting surface that has recesses under each key. Pressure at these points bends the film, producing a relatively high level of signal. External circuitry is used to cancel out extraneous signals produced by noise and vibration. In one type of keyboard, depressing a key causes an infrared source to scan across a coded pattern on the film. The pattern for each key will differ from all the others on the

PLASTICS: DESIGN CRITERIA 239

board. With this scheme only two leads are required for the entire keyboard, rather than needing one for each key, as in the arrangement previously described. Surveillance systems also make use of the film's infrared qualities. One simple version contains two closely coupled sensors. Only one is exposed to a target area, but the two sensors will respond identically to noise, vibration, and overall temperature changes. Signals from the two sensors feed a differential operational amplifier that magnifies the IR response and cancels out common mode noise signals. Designers can use sophisticated schemes in sensors to take advantage of their PYDF capabilities of being able to convert an electrical signal to physical motion or reverse this action. For example, a parabolic reflector can direct IR emanations onto a piezo-film sensor. This sensor will have two identical metal patterns on the exposed surface and a ground plane on the rear one. External circuitry will reject identical signals from either pattern that are caused by noise or vibration, but a person moving in a room will produce an easily identifiable series of pulses from both sensors.

Conductive Polymers The field of conducting polymers continues to undergo active research in many laboratories, such as the Allied Corporation, GTE-Waltham, Polaroid British Petroleum, the University of Califqrnia at Santa Barbara, the University of Pennsylvania, MIT, the University of Massachusetts, the University of Texas at Arlington, and Bell Communication Research. The physics of conduction in these materials is still under study and debate, especially with regard to interchain charge transport mechanisms. The potential applications for these materials range from thin-film applications such as solar cells and integrated circuit devices to lightweight conductors for aircraft and radio frequency or antistatic shields for electronic apparatus. Perhaps the most seriously explored application is that of electrodes for lightweight rechargeable storage batteries. Unfortunately, none of these applications has yet been realized, owing to problems of stability and processability. With few exceptions, these materials lose their conductivity over time, especially on exposure to air or moisture. Polypyrrole exhibits better stability than most conductive polymers and gives hope that even better materials will eventually be found. Another major weakness of these polymers is their processing intractability once converted to their conducting form. The advantage of processing ease that most polymers enjoy over metals is thus not realized. A few exceptions to this situation (e.g., the solubility of p-phenylene sulfide in AsF J as it is doped with AsFs) have led researchers to explore innovative solutions for casting paths and other processing routes. In addition, Polaroid hilS recently commercialized a latex wherein a few percent of polypyrrole is polymerized on the latex-particle surface. The formulation can then be cast into a film on a substrate to form a conductive shield for EMI protection, provided that pinholes can be eliminated.

FRICTION, WEAR, AND HARDNESS PROPERTIES Friction is the resistance against change in the relative positions of two bodies touching one another. If the area of contact is a plane, the relative motion will be a sliding one and the resistance will be called sliding o~ kinetic friction. If the material in the area of contact is loaded beyond its strength, abrasion or wear will take place. Both phenomena are affected by numerous factors such as the load, relative velocity, temperature, and

240 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOk

:~ r-...._ o

5

10

15

../

20

%PTFE

Figure 3-104. Relative wear versus percent by weight of fluoropolymer/PTFE.

material. Although plastics may not be as hard as metal products, their resistance to wear and abrasion may still be excellent. Plastic hardware parts such as cams, gears, slides, rollers, and pinions frequently provide outstanding wear resistance and quiet operation. Smooth plastic surfaces result in reduced friction, as they do in pipes and valves [1, 2, 14, 62-68, 252, 253, 319-31]. The frictional properties of thermoplastics, specifically the reinforced and filled composites, vary in a way that is unique from metals. In contrast to metals, even the highly reinforced resins have low modulus values and thus do not behave according to the classic laws of friction, as developed by theories from the ICI-LNP. Metal-to-thermoplastic friction is characterized by adhesion and deformation resulting in frictional forces that are not proportional to load, because friction decreases as load increases, but are proportional to speed (see Tables 3-21 through 3-24) [323]. The wear rate is generally defined as the volumetric loss of material over a given unit of time. Several mechanisms operate simultaneously to remove material from the wear interface. However, the primary mechanism is adhesive wear, which is characterized by having fine particles of polymer removed from the surface. The presence of this powder is a good indication that the rubbing surfaces are wearing properly. Conversely, the presence of melted polymer or large gouges or grooves at the interface normally indicates that the materials are abrading, not wearing, or the pressure velocity limits of the materials may be being exceeded [323]. The ease and economy of manufacturing gears, cams, bearings, slides, ratchets, and so on with injection-moldable TPs have led to a widespread displacement of metals in these types of applications. In addition to their inherent processing advantages, the parts made from these materials are able to dampen shock and vibration, reduce part weight, run with less power, provide corrosion protection, run quietly, and operate with little or no maintenance, while still giving the design engineer tremendous freedom. These characteristics can be further enhanced and their applications widened by fillers, additives, and reinforcements (see Fig. 3-104 [323] and Chapter 2). Compounding properly will yield an almost limitless combination of an increased load-carrying capacity, a reduced coefficient of friction, improved wear resistance, higher mechanical strengths, improved thermal properties, greater fatigue endurance and creep resistance, excellent dimensional stability and reproducibility, and the like. For example, the primary rein-

PLASTICS: DESIGN CRITERIA 241

forcements and lubricants used by ICI-LNP, a worldwide leader in friction- and wearresistant compounds, for its internally lubricated composites, called Lubricomb, are: PTFE. A low molecular weight modified polytetraftuoroethylene.

Silicone. A high-viscosity, low-vapor-pressure polysiloxane. Glass fiber. For use at 0.00045 in. (0.00114 cm) in diameter; an electrical-grade fiber. Carbon fiber. A high-modulus, highly graphitized reinforcing fiber. Aramid fiber. A high strength and modulus, heat- and wear-resistant fiber. Others. Lubricants such as graphite and molybdenum disulfide, used to obtain different performance characteristics.

Tests Different test results are available to the designer wanting friction and wear data as well as the usual mechanical short- and long-term data, corrosion resistance, readings, and so on. The data presented include the load and velocity capabilities of a bearing material as expressed by the product of the unit load P (in psi) based on the projected bearing area and linear shaft velocity V (in ft.lmin.). The symbol PV denotes the important property of the pressure-velocity relationship. A description of the test method used to generate the limiting PV (LPV) of a compound, and other tests, are provided so that the data can be understood. Note that the LPV test is a short-term independent variable of the wear rate. Once the operating parameters of an application exceed approximately one half the LPV, wear begins to accelerate. Therefore, the working PV can be approximated by this test method by using a factor of two [323]. Wear

Wear tests are conducted with a thrust-washer test apparatus. A sample thrust washer is mounted in an antifriction bearing equipped with a torque arm (see Fig. 3-105) [323]. The test-specimen holder is drilled to accept a thermocouple temperature probe. The raised portion of the thrust washer bears against a dry, cold-rolled, carbon-steel wear ring with a 12- to 16-microinch finish at an 18 to 22 Rockwell C scale hardness at room temperature. Each evaluation is conducted with a new wear ring that has been cleaned and weighed on an analytical balance. The bearing temperature and friction torque are continuously monitored. The test duration is dependent upon the period required to achieve a 360-degree contact between the raised portion of the thrust washer and the wear ring. The average wear factor and duration of this break-in period are then reported. The wear factors reported for each compound are based on its equilibrium wear rate independent of break-in wear. Volume wear is calculated as follows:

w = _W_e....;ig:::...h_t_lo_s_s Density where W is the volume wear (cm3) , weight loss is in milligrams, and density is in grams/cm3 . This volume is used to calculate the wear factor, K: W K=-

FVT

242 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 3-16. Dielectric Properties of Kevlar (Aramid) Fiber/Epoxy Composites*

Electrical Property Dielectric constant, K

Dissipation factor, D

Frequency, Hz 10,000 10,000 10,000,000 10,000,000 10,000 10,000 10,000,000 10,000,000

Test Conditionst

Fiber Kevlar 149

Kevlar 49

Kevlar 29

4.14 4.84 3.90 4.35 0.0103 0.0180 0.0142 0.0240

4.44 5.42 4.14 4.77 0.0131 0.0250 0.0170 0.0280

4.51 5.47 4.19 4.79 0.0135 0.0260 0.0171 0.0290

Dry Wet Dry Wet Dry Wet Dry Wet

'Fabric Style 285 (four·harness satin weave) in Fiberite 934 epoxy. Fiberite Corp., Winona, Minn. tWet indicates that sample was tested after 600 hours at 71°C and 75% RH.

where W volume wear (in. 3 ), V time (hr.).

=

velocity (ft.lmin.), F

= force (lb.), and T = elapsed

Friction The coefficient of friction data are obtained with the same thrust-washer test apparatus. The test specimen is run in against the standard wear ring until a 360-degree contact between the raised portion of the thrust washer and the wear ring is achieved. The temperature of the test specimen is then allowed to stabilize at the test conditions (generally 40 psi, 50 ft.lmin., room temperature, and dry). After thermal equilibrium occurs, the dynamic frictional torque generated is measured with the torque arm that is mounted on the antifriction bearing. An average of a minimum of five readings is taken.

0

I '--,r': I Movi ng-Surface

"-, '.m'" "'.., Lower Sample Holder

~

1"'"1'T'!-

-.... ; ""1

Stationary Surface Table LOWN Samplf' Holdt'r Anti Fnctlon Be.lr1ng

Figure 3·105. A washer-test apparatus to determine the wear, friction, and limiting PV for molding compounds.

PLASTICS: DESIGN CRITERIA 243

Table 3-17. Examples of Electrical, Physical, and Mechanical Properties of High-Performance Thermoplastics ASTM Physical Specific gravity, glc 3 Water absorption, % by weight Electrical Dielectric strength, v/mil Arc resistance, sec Volume resistivity, ohm/cm Thermal HDT at 264 psi. of Long-term service temp, UL index, of Oxygen index, % Mechanical Aexural modulus, psi (MPa) Impact strength, notched Izod, ft.llb.lin. Tensile strength, psi (MPa)

PElt 1.27 0.25

D 570

PEst 1.37 0.43

D 149 D495 D 257

710 128 10 18

400 120 10 16

D 648

392 338

397 356

40

38

D 780 D 256 D638

480,000 (3,310) 1.0 15,200 (104.8)

375,000 (2,586) 1.6 12,200 (84.12)

ppst

PSFt

1.30 0.02

PCt

1.55 0.3

1.20 0.26

425 39 10 17

425 115 10 17

-*

275

345 302

265 255

47

38

25

380 34 lOIS

550,000 (3,792) 0.4 9,500 (65.5)

390,000 (2,689) 1.3 10,200 (70.33)

340,000 (2,344) 2.2 9,500 (65.5)

-Not UL listed.

tNote: Glass filler can considerably extend the perfonnance of the above polymers. PEl = polyetherimide; PES = polyether sulfone; PPS = polyphenylene sulfide; PSF = polysulfone; PC = polycarbonate.

Limiting PV To determine the limiting PV for a compound, a sample cylindrical half-bearing, generally with a 1 x 1 x 0.060 in. wall, is installed in an antifriction bearing mounted in the test apparatus. The bearing's holder is equipped with a torque arm. Load is applied through the antifriction bearing to the test bearing. The shaft can be rotated at surface velocities from 10 to 1,000 fUmin. The load (in psi), velocity (fUmin.), friction torque (lb.-ft.), and temperature eF) at the bearing holder, which is 0.125 in. from the rubbing surfaces, are continuously monitored. A minimum of three velocities are selected to cover a practical range, as, for instance, 10, 100, and 1,000 ft.lmin.

Table 3-18. The Electric-Field Shielding Effectiveness of Three Typical Conductive Coatings Shielding Effectiveness (dB)

Emission Frequency (mHz)

Zinc Arc

Nickel

Copper

30 100 300 1,000

67 57 70 73

55 56 69 47

75 63 71 51

244 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

160

4.0

400

3.0

300

BRG. MATERIAL: lYpical I" x I" SHAFT: Cold rolled Carbon Steel FINISH: 16 Micro-Inches (RMS) HARDNESS: 18-22Rc SURFACE VELOCITY: 100 Ft.lMin. "PV" UMIT @ 100 Ft.lMin.=12,OOO (PSI x Fl/Min.

140 120

Z: 0::;

ID ..J

I

]

100 80

~

1

~ 2.0

F

60 40

S ~ I&.

I&.

}

1~

200

1.0

100

0

0

Temperature-OF

20 0

0

2 3 Time-Hrs.

4

5

Figure 3-106. A load-stepping test to determine the limiting PV capability of a bearing material.

At each velocity a load-stepping test is conducted. The friction torque and bearing temperature, which are plotted continuously, are allowed to reach equilibrium at each loading (see Fig. 3-106) [323]. The equilibrium condition is maintained for about 30 min., at which point the load is increased. At an advance load increment the friction torque and temperature will not stabilize. The slope of the curve will increase sharply in the friction torque or temperature plot. An increase in temperature or torque will eventually result in bearing failure. The pressure limits at several velocities provide a curve showing the limiting PV capability of the bearing material [323]. Hardness

Although hardness is a somewhat nebulous term, it can be defined in terms of the tensile modulus of elasticity. From a more practical side, it is usually characterized by a combination of three measurable parameters: 1) scratch resistance; 2) abrasion or mar resistance; 3) indentation under load (see Table 3-23). To measure scratch resistance or hardness, a specimen is moved laterally under a loaded diamond point. The hardness value is expressed as the load divided by the width of the scratch. In other tests, especially in the paint industry, the surface is scratched with lead pencils of different hardnesses. The hardness of the surface is defined by the pencil hardness that first causes a visible scratch. Other tests include a sand-blast spray evaluation. Abrasion resistance is usually measured by the material's loss in weight (see Table 325) or the change in optical transmission and reflectance after a sample has been exposed

PLASTICS: DESIGN CRITERIA 245

to an abrasive surface. This is usually done under load, for a predetermined number of cycles or a time period specified by ASTM methods. Tests for indention under load are performed basically like the ASTM tests used to measure the hardness of other materials, such as metals and ceramics. There are at least four popular hardness scales in use. Shore A and Shore D are for soft to relatively hard plastics and elastomers. Barcol is used from the mid-range of Shore D to above it. Rockwell M is used for very hard plastics. Figure 3-107 shows the relative ranges covered by these durometers. This diagram does not correlate the different systems in Chapter 2.

140

---

90

50

~

90

"

~

80

BARCOL

SHORED

SHOREA

VERY SOFT

Figure 3-107. The range of hardness common to plastics.

VERY HARD

ROCKWELL M

246 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Plastics-to-Plastics Wear The wear characteristics of one plastic as opposed to another vary widely, even among materials that have good natural lubricity . When an application calls for plastic-to-plastic bearings, shafts, gears, or other wear members, the combination of materials must be chosen carefully [323]. Because plastics are not rigid, they do not behave according to the classic laws of friction. It is these deviations that cause some of the unexpected results when plastics are run against metals. Frictional forces are not proportional to loadfriction increases with increasing speed, and the static coefficient of friction is lower than its dynamic one. When two viscoelastic low-modulus materials are run against each other, additional inconsistencies result. Despite these differences, one trend remains clear: the wear factor generated when TP is run against itself is extremely high, unless its operating temperature and pressure are quite low. In applications requiring all-plastic components, the wear rate can be reduced, if crystalline plastics are being used, by running dissimilar plastics against each other. If amorphous plastics are involved, or if environmental or manufacturing procedures require that only a single compound be used, that compound should contain an internal lubricant, like PTFE at loadings of 15 to 20 percent by weight. Wear is often greater on a moving surface when dissimilar neat resins are paired. Similar behavior occurs with pairs consisting of lubricated, unreinforced resins running against themselves or against dissimilar lubricated resins. The addition of a reinforcing fiber generally produces increased wear in a mating, unreinforced resin. The addition of a reinforcing fiber to both surfaces may result in decreased wear, compared to that in unreinforced resins. The wear factors of glass-fiber composites are lower than those with carbon-fiber materials when run against a carbon-reinforced material, because glass fibers are much harder than carbon ones. Lubricating composites with PTFE dramatically reduces the wear factors in both similar or dissimilar mating resins. During the initial break-in period a film of PTFE is transferred to the mating surface, thus creating a PTFE-to-PTFE bearing condition that lowers the wear factors for both the moving and stationary surfaces. The addition of a PTFE lubricant to the mating material reduces the detrimental effects of glass fibers, with respect to wear, on an opposing surface. Adding a silicone fluid to TP composites results in compounds with reduced wear factors as compared to neat resins, but these factors are greater than those of the PTFElubricated composites. In plastic-to-plastic wear applications, composite pairs having similar wear factors when run against each other are preferable to pairs having great differences in their wear factors, if the total wear in the system is acceptably low [323].

Plastics-to-Metals Wear Most studies on the wear and friction characteristics of plastic composites have concentrated on plastic versus plastic or plastic versus steel wear rings in the same finish and hardness. However, the increased use of aluminum in structural and bearing components has resulted in available, reliable wear-property data involving TP composites run against aluminum surfaces. In addition, cost-reduction programs in the business-machine and appliance industries, which have led to the elimination of some parts-finishing operations, have resulted in characterizing the action between rough metal surfaces and plastic composites.

PLASTICS: DESIGN CRITERIA 247

Tests have been run to broaden this database to include the behavior of TP composites against several metals with different surface finishes. Also, wear factors for the metal surfaces, caused by the rubbing of plastic composites, have been studied. Certain metals chosen for stationary counterface materials were cold-rolled 1141 steel, 304 and 440 stainless steel, 70/30 brass, 2024 aluminum alloy, and phosphor bronze (ICI-LNP) [323].

Design Concepts Short Term

In general, the addition of properly sized reinforcing fibers in typical materials like carbon and glass will result in dramatic increases in strength, stiffness, and thermal properties while the solid internal lubricants act as fillers-MoS 2 is an exception-and reduce the properties of an unmodified or reinforced resin by an amount about equal to the percentage volume of the lubricant. Liquid silicone, for instance, is used at relatively low volume contents and has only a modest effect on mechanical and thermal properties. The product's tensile and flexural properties will be slightly reduced but its impact characteristics improved. Its thermal properties will remain unchanged or be lowered [323]. Long-Term Creep

The flexural creep curves for a family of reinforced or lubricated nylon 6/6 compounds are shown in Figure 3-108. At a given stress, the creep curves for a 30 percent glassreinforced nylon 6/6 will equilibrate after an initial increase in strain while the unmodified curve continues to increase, making normal engineering calculations meaningless. When a lubricant such as 15 percent PTFE is added to the glass-reinforced nylon, however, its mechanical properties are reduced. This property reduction causes the creep curve to shift upward but typically remain parallel to that of the curve with no PTFE. Other changes in the type or amount of internal lubricant or reinforcement will cause the curve to shift up or down on the graph. The magnitude of this shift can be approximated for a given resin family by assuming that the shift will be directly proportional to the change in the short-term flexural modulus (ASTM D 790) value.

Table 3-19. Magnetic-Field Shielding Effectiveness of Three Typical Conductive Coatings Shielding Effectiveness (dB)

Emission Frequency (mHz)

Nickel

Copper

Solid Steel

0.01 0.10 0.50 1.0 10 30

0

0

50

2

2

5

7 13 37 47

8

33 39

Table 3-20. Typical Conductive Coating Systems to Provide EMI/RFI Shielding for Plastics Shielding System Conductive Coatings Silver

Advantages

Disadvantages

Highly conductive (0.1 ohm per square foot or less); applied by conventional spray equipment; easy application; electrically stable (minimal change in resistance with environmental cycling); easily applied to selected area; field repairable.

High cost

Nickel

Low cost (15-30 cents per square foot); good conductivity (less than 1.0 ohm per square foot); applied by conventional spray equipment; easy application; relatively stable (differs with manufacturer); easily applied to selected area; field repairable.

Lesser quality formulations available; some are stable, some are not.

Copper

Highly conductive (less than 0.5 ohm per square foot); easy application; low cost (15-30 cents per square foot).

Oxidation can reduce conductivity (resistance can change to effectively make copper an insulator); some may be alloys-if layered with silver, cost will rise.

Graphite

Low cost (5-15 cents per square foot); easy application; excellent ESD (electrostatic discharge) performance.

Less conductivity (ranging from 2 ohms to the thousands per square foot, depending upon the amount of graphite); modest shielding capability (up to 30-40 dB).

ArclFlame Spray

Highly conductive (less than 0.1 ohm per square foot; hard, dense coating.

Requires grit blasting to promote mechanical bonding to plastic; special applications equipment required; requires special applicator safety procedures for dust and fumes; warps thermoplastics; not suitable for thin-walled designs; not field repairable.

Vacuum Metalizationl Ion Plating

Highly conductive (less than 0.1 ohm per square foot); controllable film thickness; not limited to simple housing designs.

Requires primer coat; entire part must be done, forcing exterior painting; not field repairable; specialized application equipment; vacuum chamber size a limiting factor; requires specialized knowledge; subject to corrosion in humid atmosphere unless protected.

Electrolysis Deposition

Highly conductive (both nickel and copper less than 0.1 ohm per square foot).

Requires specialized equipmentlknowledge; entire part must be coated, forcing exterior painting; if copper is used it must be protected by a nickel coating or some other coating.

Conductive Plastics

Good thermal transfer; elimination of secondary operation for shielding.

Requires a secondary operation for grounding.

248

PLASTICS: DESIGN CRITERIA 249

Long-Term Fatigue

The fatigue endurance curves shown in Figure 3-109 can be shifted upward by improving their mechanical properties through additional reinforcement. If the reinforcement is reduced or internal lubricants are added, the mechanical properties will be lowered and the fatigue endurance curve shift downward. The magnitude of these shifts can be approximated by assuming them to be directly proportional to the changes in short-term flexural strength. Lower Costs

Many examples could be cited to show cost reductions in replacing other materials with plastic wear-resistant compounds. For example, Rutihauser Data AG, of Switzerland, a leading manufacturer of gearing arrangements for single- and dual-sheet loaders and paper feeders, became more productive with plastics. They eliminated metal gears on their computer printers on the basis of the cost of materials and the expense of finishing operations. Their unfilled PBT (Celanese's Celanex 2012 grade) used a UL flammability rating of 94VO. This is essential to meet requirements in the major computer markets. Printers operate in an extremely demanding environment. For one thing, paper dust is highly abrasive, requiring all components to have good wear resistance. Chemical resistance is also essential, to withstand greases and solvent cleaners. An ability to withstand fatigue from long-term vibration is likewise important. To help ensure dimensional repeatability, concentricity, and true running, the gears were molded on a three-plate mold, gated at the top with three points of entry [12]. Thus, in addition to performing efficiently as gears, the PBT gears provided excellent frictional properties, with an inherent lubricity contributing to long-term wear resistance. Torque tests conducted on these gears in operation performed at levels ten times in excess of the normal operating values. In the next chapter we shall see how plastics and composites can be made to hold up well under environmental stresses.

Table 3-21. Examples of the Coefficient of Friction for Plastics Coefficient of Friction Material

Static

Dynamic

Nylon Nylon/glass Nylon/carbon Polycarbonate Polycarbonate/glass Polybutylene terephthalate (PBT) PBT/glass Polyphenylene sulfide (PPS) PPS/glass PPS/carbon Acetal

0.2 0.24 0.1 0.31 0.18 0.19 O.ll 0.3 0.15 0.16 0.2 0.04

0.28 0.31 O.ll 0.38 0.20 0.25 0.12 0.24 0.17 0.15 0.21 0.05

PTFE

1.00 (moisture conditioned to 50% R.H.)

0.75

".

f

J

0.50

--50% G/R Nylon 6/6 (5,000 psi)

0.25

0.00 10

102

103 nme, Hrs.

104

Figure 3·108. The flexural creep of nylon 6/6 and its composites at 23 0 C (73 0 F) (ICI-LNP).

10.0 (moisture conditioned to 50% R.H.) 7.5

40% G/R Nylon 6/6

-----i

...

PI

0

30% G/R Nylon 6/6

)C

1

I

5.0

2.5

20% PTFE Lub. Nylon 6/6

0.0 104

105

106

107

Cycles of Failure

Figure 3·109. The fatigue endurance of nylon 6/6 and its components at 23 0 C (73 0 F) (ICI· LNP). 250

Table 3-22. Wear and Friction Properties of Fiber-Reinforced Plastics and Composites Fiber Type and Content (%)

Short

Coefficient of Friction

Wear Factor (10-10 in. 3min.l ft.llb.lhr. )

Long

Static (40 psi)

Dynamic (40 psi, 50 fpm)

0.20 0.28 0.26 0.21 0.13 0.09

0.28 0.35 0.36 0.24 0.18 0.09

0.16 0.13

0.13 0.13

Nylon 6/6 Matrix 0 50 glass

200 60 30 30 14 8

0 50 glass 20 aramid

40 carbon 40 carbon

Polycarbonate Matrix 40 carbon

26 17

40 carbon

Table 3-23. Guide for the Properties of Some Plastic Coatings

..=

B

] ."

.§

0 ::I

0)

0)

~

G

> e::-

>.

.§ 0)

-< Hardness Adhesion Flexibility Mar resistance Gloss (85 units plus 60°) Fabricability after aging Humidity resistance Grease and oil resistance General chemical resistance General corrosion resislance (industrial atmospheres) Exterior durability (pigmented) Exterior durability (clear films) Ratings: J

= excellent,

2

."

0)

>.

"0 Q..

2 2 3 2

c

;;

2 2

2 2 2 2 2

.S!

>.

S.... =s>. >.S ~ B -< = '" ~

."

>.

"0 '"0

t

g

2

2

1 2 2

1 3 3

til

;

"0 '" '.p ~

il: 3 2 1 3 4

2

4

....

i

.E

.~ J::l

til

2 1

3

.... B

'" Ul

>.

B

·c

fi:

-<

i

>.

fi:

:9

0

0

.5

.5

"&u

."

.S!

Q)

-<

E

§

-<

'"0

g "&u 0)

0)

c

'"0) 0)

c

i

~

i:ii

~

i:ii

2

2

2 2 2

2 1

2 2 3 2

2 2 3 2

2 2 2 2

2

2

3

3

2

1 2

1 2

2 2

2 2

1 2

0)

~

.S!

~

2 2

2

1 2

3

2

3

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

3

2

2

2

4

4

2

3

3

3

3

3

3

4

4

2

4

= poor.

= good,

3

= fair,

.g

·c

g

1l

0)

>. > >.

2 2 1 3 4

.§

.. .. =

u:

....l

."

."

·c 0

0)

·c

B

.~

....l

~

~ .§ u

.5

.;;

>. >. c > >.

b ;£ ;£ cl: 2 2 1 3 4

0)

'OJ

2 2

4

~ Q..

u: .S!

"&u -<

2 1 2 4

1 2 2

2 2

2

2

4

251

Table 3-24. Plastic-to-Plastic Wear and Frictional Properties (ICI-LNP)

Base Resin Polycarbonate Polycarbonate Acetal Copolymer Nylon 6/10 Nylon 6/10 Nylon 6/6

Filler System

Base Resin

10% PTFE

Nylon 6/10

15% PTFE 20% Glass fiber 20% PTFE 15% 30% 15% 30% 20%

PTFE

Coefficient of Friction

Filler System

Wear Factory Stationary Surface 10-10 in. 3min.1 ft.llb.lhr.

Static (40 psi)

Dynamic (40 psi, 50 fpm)

PTFE

35

0.04

0.06

15

0.05

0.07

Moving Surface

Stationary Surface

Nylon 6/6

15% 30% 15% 30% 20%

PTFE

30

0.03

0.04

Polycarbonate

10% PTFE

15

0.04

0.06

Polycarbonate

15% PTFE 20% Glass fiber 20% PTFE

18

0.05

0.07

12

0.03

0.04

Nylon 6/10

Glass fiber

PTFE Glass fiber

Glass fiber

PTFE Glass fiber

PTFE

Acetal Copolymer

Table 3-25. The Abrasion Resistance of Moisture-Curing Polyurethane Coatings Type of Coating

Taber Index*

Moisture-curing urethanes Clear floor coating Clear coating Clear exterior coating Brick red Tile green Gray Other coatings Amine catalyzed epoxy varnish Two-part polyester urethane enamel Polyamide epoxy enamel Clear nitrocellulose lacquer Vinyl enamel Urethane oil varnish Phenolic spar varnish Epoxy ester enamel 'Taber index values indicate weight loss in mgll ,000 revolution of the abrasion tester per (ASTM).

252

8 22 24 31 33 35 38 60 95 96 106 155 172 196

Chapter 4

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS

This chapter reviews different environmental conditions that can affect plastics. (Further data pertaining to these conditions is provided in Chapter 6.) Like other materials with which designers work, different plastics can be sensitive to their environments. Even ordinary exposure from sunlight or to household cleaning agents can change the properties of certain plastics. Whereas rust, corrosion, and loss of its properties can plague metals, the cracking, crazing, and loss of its properties can affect certain plastics in the presence of different environments. Basically, environmental effects on certain plastics can be related to their performance, as summarized in Figure 4-1 [1-22, 62-68, 332-89].

TEMPERATURE INTRODUCTION As mentioned throughout this book, plastics can be affected in different ways by temperature. Among other things, it can influence short- and long-time static and dynamic mechanical properties (see Fig. 4-2), aesthetics, dimensions, electronic properties, and other characteristics. Some plastics cannot take boiling water, others can operate up to 149°C (300°F), and the so-called high-temperature types can take various degrees of continuous use above 149°C (300OP). Then there are the plastic composites used as heatshield ablative materials on the nose cones of space vehicles that reach temperatures of about 1,370°C (2,500°F) for fractions of a second upon reentering the atmosphere. Practically all plastics can take heat up to at least what the human body can endure, which is one important reason they are so extensively used. Thermoplastics soften to varying degrees at elevated temperatures, but thermosets are much less affected (Figs. 4-3 and 4-4). In fact, a few plastics even reach 538°C (1,000 oF) (see Chapter 6). The maximum temperatures under which plastics can be employed are generally higher than the temperatures found in buildings, including walls and roofs, but some such as LOPE are marginal and others cannot carry appreciable stresses at moderately elevated temperatures without undergoing noticeable creep. Many plastics can take shipping conditions that are more severe than their service conditions, as in an automobile trunk or railroad boxcar that might reach 52°C (126°F). The response of a plastic to an applied stress depends on the temperature and the time at that temperature to a much greater extent than does that of a metal or ceramic. The variation of an amorphous TP over an extended temperature range can be exemplified 253

Environment Figure 4-1. Basic elements in designing with plastics: a three-dimensional representation of contour plots,

100, Composite and Engineered Plastics

90

.... =

80

·iii

70

C"

><

CI.

"CI

50

:;:

40

~ en Qi ~

60

'iii

30

I-

20

cQ)

l'

Typical Steel

10 O+--'--.--r~r--r~--'-~~'--'~

-100

0 100 200 300 400 500 600 700 800 900 1000

o

Temperature, F Figure 4-2. A guide to maximum short-term tensile stress versus temperature, 254

Figure 4-3. For more marketable attractive colors and toughness, the Sunbeam Appliance Co. assembles its irons with a skirt molded from a glass-reinforced isopolyester (Amoco's resin) bulk molding compound (BMC). The skirt, which is located just above the sole plate, must be able to withstand a continuous operating temperature of 232°C (450°F).

Figure 4-4. Special thermoset polyester resins sometimes demonstrate maximal resistance to high temperature and corrosion for chemical equipment uses, as for example, this carbon tetrachloride storage glass-fiber-reinforced plastic tank. 255

256 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

by the behavior of its elastic modulus as a function of temperature. Four principal regions of such viscoelastic behavior were identified in Chapter 2. With a temperature change the short-term static strength, the elastic modulus, and the elongation behavior of a material will be similar for its tensile, compressive, flexural, and shear properties. As shown in Figure 3-8, a material's strength and modulus will decrease and its elongation increase with increasing temperature at constant strain. The same set of curves can be generated by testing at a constant temperature and various strain rates. The effects of temperature changes on the tensile properties of TPs are shown in Figures 4-5 through 4-7. Curves for creep isochronous stress and isometric stress are usually produced from measurements at a fixed temperature. Complete sets of these curves are sometimes available at temperatures other than the ambient. It is common, for instance, to find creep rupture or apparent modulus curves plotted against log time, with temperature as a parameter. Figure 4-8 shows time-temperature shifting of apparent modulus curves projecting to times beyond a normal testing range [2]. These curves suggest that it would be reasonable to estimate moduli at somewhat longer times than the data available from the lower temperatures. However, a set of creeprupture curves from various temperatures, as in Figure 4-9, would suggest that projecting the lowest-temperature curves to longer times as a straight line could produce a dangerously high prediction of rupture strength, so this approach is not recommended [2]. One advantage of conducting complete creep-rupture testing at elevated temperatures is that although such testing for endurance requires long times, the strength levels of the plastic at different temperatures can be developed in a relatively short time, usually just 1,000 to 2,000 hours. Such a system has been employed for many years by the Underwriters Laboratories and other such organizations [14, 17, etc.]. Testing different impact properties at various temperatures produces a plot that looks

75 Yield strenRth varies considerab Y WIth temperature

-20°C

60

N

E

[,5

z

20°C

III III

50°C

:::E QJ

L.-

Vi

30

.Strain (%)

Figure 4·5. Examples of the effects of temperature on the stress-strain behavior of basic commodity thermoplastics that influence their ultimate and yield strengths, as well as their moduli.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 257

'+ ----+----i 800 , .

11 II>

Q

700

M

o

E

u

0>

9-

600 ---300,.r---I 500

7

""

en c

...~

II)

At 0 2 In lin in

>-

(Q

51

ern/mill)

400

5

300


c:

Q.I ~

3 --

200

"0 Q.I

>-

!E VI c:

Q.I ~

1~__~____L -_ _ _ ~_ _ _ _~_ _~_ _~100

50

0

50

100

150

200

250

TeITlIJeri:lture . F

Figure 4-6. The effects of temperature on acetal, using semilog coordinates.

very much like an elongation versus temperature curve (see Fig. 4-10) [2]. As temperatures drop significantly below the ambient temperatures, most TPs lose much of their roomtemperature impact strength. A few, however, are on the lower, almost horizontal portion of the curve at room temperature and thus show only a gradual decrease in impact properties with decreases in temperature. One major exception is provided by the glass fiber RPs, which have relatively high Izod impact values, down to at least -40°C (-40°F). In regard to testing, the S-N (fatigue) curves for TPs at various temperatures show a decrease in strength values with increases in temperature. The TSs, specifically the TS RPs, can have very low losses in strength, however. AMORPHOUS

t

UNFILLED REINFORCED

TEMPERA TURE

~

Figure 4-7. The modulus behavior of crystalline and amorphous thermoplastics showing their glass-transition temperatures (Tg), melt temperatures (Tm ), and the effects of reinforcement on their DTUL (see Chapter 2 for explanations of Tg , Tm , and DTUL).

258 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

C/)

::J

:5

INCREASING TEMPERA TURE

~~

~~

ct

__________

ct:

................ .................

-~ ..............

Q..

<:(

~

o" -.J

. . . . . ......

.....

.........

TlME- TEMPERA TURE SHIFTING

-7

0

2

3

4

70

50

5

6

YEARS

7

8

i~

9

LOG TiME (HOURS)

Figure 4-8. Apparent modulus curves at different temperatures showing time-temperature shifting to estimate the extended time values at lower temperatures.

Heat Distortion Temperature

The heat distortion temperature (HDT), also called the deflection temperature under load (DTUL), has a softening temperature that denotes the maximum temperature at which a plastic can be used as a rigid material (Figure 4-11, Table 4-1). It may also be considered the upper limit at which the material can support a load for any appreciable time. Thus, the HOT or OTUL can be a very practical, important property. For amorphous TPs the HDT is near the Tg , but for the highly crystalline types it is closer to the melting point. Most HOTs or softening temperatures are arbitrarily defined as being a single point in some kind of deflection-temperature curve. Such a test is quite dependent on the testing and processing conditions. It tends to have only limited use in actual design, but it can be used as a guide or approach to selection. Often, the high-temperature apparent modulus data for a particular time span gives a INCREASING TEMPERA TURE

Lu

ct

2 Q..

::J

ct

Q..

Lu Lu

ct

()

o" -.J

-1

2

3

LOG TIME (HOURS)

Figure 4-9. Creep-rupture curves indicating the danger of making linear projections to longer times at lower temperatures.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 259

u

co E

Q.

TEMPERA TURE

---+

Figure 4-10. The effects of temperature on tensile elongation.

much better estimation of a material's behavior under load in an actual application. However, the deflection test is relatively simple to run, as well as being quick and relatively inexpensive. The basic test procedure under ASTM D 648 for determining the DTUL is to place a sample in an oil bath, give it a simple three-point support, with either a 264- or 66-psi distributed load, and increase the temperature at a rate of 2°C/min. Deflection temperature under load is defined as the oil temperature at which a sample deflects 0.010 in. Several critical parameters have a significant effect: 1) The test method specifies that the sample have a mininum thickness of 0.125 in, but does not specify an upper limit. Since it takes longer for a thick bar to come to thermal equilibrium than a thinner one, a thicker bar will have a higher DTUL than the thinner one, which must be taken into account when comparing data sheets; 2) in general, the higher the load at 66 or 264 psi, the lower the DTUL; 3) an annealed bar usually has a higher DTUL. Annealing refers to placing a bar in an oven at a prespecified elevated temperature for a certain time to relieve any processedin strains in the plastic; and 4) when processing the same material by different processes, the DTULs can be different, as for example compression-molded bars, which usually give higher values than injection-molded ones. Temperature (OF) 700

-

600 500 r-

400 300

..--

-

-

-

r-

r-

r-

_r-

200 100

ASS

Poly- PPO Acetal Poly- PST carbonate sulfone

PET Nylon Thermo- PPS Phenolic set Polyester

Figure 4-11. An example of the heat-resistance perfonnance of thennoplastics and thennosets, based on HDT data ("F at 264 psi).

260 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-1. An Example of Flexural Strength (X 103 psi) at Various Temperatures Engineering Polymer

73°F 23°C

212°F 100°C

300°F 150°C

400°F 200°C

Nylon 66 Nylon 66 PET PET Phenolic (GP) Phenolic (HR) Phenolic Alkyd DAP

17.4* 35.8 29.2 37.6 13.2 13.2 19.4 14.3 18.7

6.2 18.1 13.3 19.2 8.8 10.2 15.9 9.0 14.2

5.0 14.8 8.7 12.5 8.2 8.4 3.5 5.8 7.8

4.1 12.8 7.0 9.8 5.6 6.3 10.4 4.2 3.4

*To convert psi to pascals (Pa). multiply by 6.895 x

HP.

Thermal Aging Section UL 746B provides a basis for selecting high-temperature plastics and provides a long-term thermal-aging index, the RTI or relative thermal index. The testing procedure calls for test specimens in selected thicknesses to be oven aged at certain elevated temperatures (usually higher than the expected operating temperature, to accelerate the test), then be removed at various intervals and tested at room temperature. Another reason for using higher temperatures is that for an application requiring long-term exposure a candidate plastic is often required to have an RTI value higher than the maximum application temperature. The properties tested can include mechanical strength, impact resistance, and electrical characteristics. A plastic's position in a test's RTI is based on the temperature at which it still retains 50 percent of its original properties. The time required to produce a 50 percent reduction in properties is selected as an arbitrary failure point. These times can be gathered and used to make a linear Arrhenius plot of log time versus the reciprocal of the absolute exposure temperature. An Arrhenius relationship is a rate equation followed by many chemical reactions [62, 63, 262]. A linear Arrhenius plot is extrapolated from this equation to predict the temperature at which failure is to be expected at an arbitrary time that depends on the plastic's heat-aging behavior, which is usually 11,000 hours, with a minimum of 5,000 hours. This value is the RTI. As practiced by the UL, the procedure for selecting an RTI from Arrhenius plots usually involves making comparisons to a control standard material and other such steps to correct for random variations, oven temperature variations, the condition of the specimens, and others. The stress-strain and impact and electrical properties frequently do not degrade at the same rate (see Fig. 4-12), each having their own separate RTIs. Also, since thicker specimens usually take longer to fail, each thickness will require a separate RTI. The UL uses RTIs as a guideline to qualify materials for many of the standard appliances and other electrical products it regulates. This testing is done in a conservative manner qualified by judgments based on long experience with such devices; UL does not apply indexes automatically. In general, these RTIs are very conservative and can be used as safe continuous-use temperatures for low-load mechanical products (see Fig. 4-13).

Other Heat Tests There are different heat tests, some being specific to a product environment. Some examples of routine tests were shown in Figure 4-13 and Table 4-1. There are, for instance,

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 261

,00

o

o

c

o c

cu

~

Izod Impact strength Tens ile strength O~----------------------------j

o

Time at elevated temperature, h

_

Figure 4-12. An example of physical property loss due to long-term heat aging from UL 746B.

tests where a tensile load is applied to a specimen and its length is then measured in the same manner as in a creep test, except that in this case the temperature is increased at a constant rate. Distortion temperature is defined as that temperature at which elongation becomes 2 percent. A typical test is that of ASTM D 1637, for tensile heat distortion temperatures. In this test a load of 50 psi is applied to a strip and the temperature is increased at a rate of 2°C/minute. If a sheet is oriented, it will usually shrink before it starts to elongate rapidly. Another thermal consideration concerns warp temperature, the temperature at which a material begins to distort. A test for it can be performed either on the product itself or test specimens. It is somewhat subjective, since the definition of failure with this test usually varies for each application. However, the test can be useful, since it attempts to duplicate actual conditions of use. In a warp test it is important to note not only the % of Retention at 200°C

80 70 60 50

40 30 20 10

o

U L Temp Index

150"C

140"C

22O"C

13O"C

18O"C

200"C

17O"C

P Sultone

PET

PPS

Nylon 66

PES

PAl

P!lenoloc

Figure 4-13. An example of room-temperature tensile strength retention at 204°C (400°F), based on the UL's relative thermal index (RTI) test.

262 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

temperature but also the humidity. With certain materials, humidity combined with elevated temperatures has a significant effect on the material's behavior. This effect would not be evident in HDT (DTUL) tests. Test specimens can also be used to simulate some degree of warpage. Figure 4-14 compares unreinforced and reinforced glass fiber-TS polyester flexural-type specimens at different temperatures in a droop test (with a center support), sag test (end supports), and an expansion test (bolted at three points). The study for this particular test is conducted at various temperatures. By this point it should be clear that analyzing the thermal limits of the various materials available, starting with the maximum and minimum environmental temperatures under which a product must operate and adding any thermal increase from hysteresis heat that develops from flex or vibration and so on will at least tell the designer which materials cannot be used. The ratings given the designer will also provide some idea of the shortterm stiffness to be expected of various materials at elevated temperatures, as well as their thermal aging resistance with regard to certain properties. Establishing two parameters, ASTM D 648 and UL 746B, for a variety of materials provides the designer with a reasonable starting point for initially assessing materials for high-temperature applications. Most high-performance plastics are filled compounds, since fillers and reinforcements (see Fig. 4-15) generally enhance high-temperature strength and stiffness.

with glass

260 0 without glass

with glass without glass

with glass without glass

with glass Ambient

--

---

Droop Test

without glass Sag Test

Expansion Test

Figure 4-14. An example of droop, sag, and expansion tests with a glass-fiber-TS polyester composite.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 263

500,000

GRAPHITE

400,000 .;;;

a.

GLASS

X ..... 300,000 C)

BORON

... Z

'"

..... V!

~ST'"

200,000

TITANIUM

100,000

~ALUM/NUM 0

0

400

800

1200 1600 TEMPERATURE, of

2000

2400

2800

Figure 4-15. An example of tensile strength versus temperature in high-performance materials. A general definition for a high-temperature plastic is one having a thermal value in terms of ASTM D 648 and UL 746B higher than 149°C (300°F). There are numerous plastics that are both processable and have useful mechanical properties in the 149 to 260°C (300 to 500°F) range. Their costs are usually high, but so is their performance. High-temperature plastics fall into the usual categories of TSs and TPs. The TSs are used principally by the aircraft and aerospace markets but also for automotive, industrial, and electronic products. Epoxies are principally used, with others being polyesters, phenolics, and urethanes. These resins are usually reinforced with the high-strength fibers seen in Figure 4-15, individually or in combination with S-glass, graphite, aramid, and others. (About 85 percent by weight of all composites used in conventional-temperature environments only require E-glass.) High-temperature TPs are available to compete with TSs, metals, ceramics, and other nonplastic materials (see Figs 4-16 through 4-18). The heat-resistant TPs include Victrex polyetherketone and polyethersulfone (ICI-LNP), Torlon polyamideimide (Du Pont), Xydar liquid crystal polymer (Amoco), and others seen in Chapter 6. These TPs have high inherent heat resistance and offer such other advantages over TSs as toughness and ease of processing. Some of these plastics are amorphous, with a high Tg , such as PES, but some like PEEK and the liquid crystal polymers (LCPs) are highly crystalline. Some high-temperature polymers are commercially available in neat form, with others being available only in the filled or reinforced form for such high-performance products as PPS and others. Many of these plastics can be processed on standard or modified TP processing equipment, which requires melting at higher temperatures than commodity resins, but others, like the polyimides, generally have to be machined into shape. The direction of high-temperature TSs appears to be toward more toughness using new processing techniques for the aviation and aerospace markets. The high-temperature TPs have been receiving considerable attention as possible replacements for TSs in advanced composites because of their higher toughness, faster processing, and ease of repair. The TPs are being promoted in both unreinforced and reinforced forms as molding and

264 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Temperature OF ISO

--- ----.............: ----

", ,

'" :; D-

100

c;,

~

PES 4100G - PES 4101GL30 - - -

...........

.!E

SO

40% glass-fiber reinforced poiyphenylene sulphide -----

~-~

20,000

... ...

--c:---~ ~

-

.......

U5 ';;; c: ~

"

~~

J:C

~bo

30e

200

100

.......

....... -<00,

:

"

. " .. .., ' ",

.......................

......

~:-

......... ,

........

......

....

30% glass·fiber reinforced _ _ _ polysulphone

polysulphone······ •

....

,

........ ~ . 150

100

50

10,000-

-....., ',- ' '. . ... . " '

0

IS,OOO _.-'---'---

--

5000-

200

Temperature °C

Figure 4-16. An example of the effect of temperature on tensile strength for some heat-resistant ICI-LNP TPs. extrusion materials. Their high continuous-use temperatures, combined with their good chemical, water, and flame resistance and low smoke generation, are finding new applications, particularly as their price is reduced through higher volume usage. The growth of TP-TS hybrids, offering TPs' ready processing combined with TSs' long-term dimensional stability, is also positive.

CHEMICAL RESISTANCE Part of the wide acceptance of plastics is from their relative compatibility to chemicals, particularly to moisture, as compared to that of other materials_ Because plastics are largely immune to the electrochemical corrosion to which metals are susceptible, they can frequently 'be used profitably to contain water and corrosive chemicals that would attack metals. Plastics are often used in corrosive environments for chemical tanks, water Temperature OF 15

'"

D-

C!>

w

.......................

10

:; " "C

PES 410OG-PES 4101GL30 - - 40% glass-fiber reinforced poiyphenylene sulphide - - - --

30<

.............

"'-::"-='::'::.- -

o :; "iii

=> x

-'~'~. - - - - - - -. ~---.---.-",

~ 5

",

- - - - -

--- ..............

...

'"

~------~--------~----~cr--------+-------~

...... .............. ............. .......

30% glass-fiber reinforced _ _ _ polysulphone

polysulphone······ •

200

100

o

50

100

150

200

Temperature °C

Figure 4-17. An example of the effect of temperature on the flexural modulus for some heatresistant ICI-LNP TPs.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 265 400.000-

2.'5-__ --------------~ -:.::::.:::.,- ------ ----'" 2.0

"-

'"

V>

:; "C

'Victrex' PES: 28 MPa (4000 Ibflin') - -

~

~

MPaP(~~\~~7i~~) ------ ~ 20 ~C~~(~8~~~~) ....... ~

28

nylon 66: ___•

20 MPa (2900 Ibf/in2)

polycarbonate: - - 20.6 MPa (3000 Ibf Iin2)

t'i:s-------- __ '. -. -....... ......... 1.0

.........

0.5 -

Data for polysulphone, polycarbonate. acetal copolymer and nylon 66 from Modern Plastics Encyclopaedia.

0 10

----- ---------~

----~~ ""-.. ----2"00":'000-

:::>

'" ....... .......... .

100.000

1 week

1 Yfar

I

10 2

Time - hours

10 3

10 4

Figure 4-18. An example of time dependence for the tensile-creep modulus for some ICI-LNP TPs.

treatment plants, and piping to handle drainage, sewage, and water supply (see Figs. 419 through 4-23). Structural shapes for use under corrosive conditions often take advantage of the properties of RPs and composites. However, certain plastics are subject to attack by aggressive fluids and chemicals, although not all plastics are attacked by the same media. It is thus most practical to select a plastic to meet a particular condition. For example, some plastics like HDPE are immune to almost all commonly found solvents. Polytetrafluoroethylene (PTFE) in particular is noted principally for its resistance to practically all chemical substances. It includes what has been generally identified as the most inert material known worldwide. It is important to recognize that all materials will have problems in certain environments, whether they are plastics, metals, aluminum, or something else. For example, the corrosion of metal surfaces has a damaging effect on both the static and dynamic strength properties of metals because it ultimately creates a reduced cross-section that can lead to eventual failure. The combined effect of corrosion and stress on strength characteristics is called stress corrosion. When the load is variable, the combination of corrosion and the varying stress is called corrosion fatigue. This problem can be controlled in several ways. One is to select the best material, such as stainless steel, a copper allOY, or titanium. Another is to use a nonmetallic protective coating of plastic or film. Certain systems like plating can reduce fatigue strength. Shot peening, then plating, seems to produce much greater improvement, but shot peening, plating, and then baking can bring the fatigue limit to a point lower even than that of the base metal. The point is that all materials have their limitations and must be critically analyzed if no prior experience exists upon which to draw. For example, RP and composite underground gasoline storage tanks have this "experience." A Chicago service station's May 1963 installation was still leaktight and structurally sound when unearthed in May 1988. The tank was one of sixty developed by Amoco Chemical Co. It was fabricated in two semicylindrical sections of fiberglasswoven roving and chopped strand mat impregnated by an unsaturated isophthalic polyester TS resin selected for its superior resistance to acids, alkalis, aromatics, solvents, and hydrocarbons. The two sections were bonded to each other and to end caps with composite lap joints. Today the cylinder would be a single, unified construction as seen in Figure 4-19. The demand for this type of petroleum storage tank has grown rapidly as environmental regulations have become more stringent [114].

Figure 4-19. This CorBan Industries RP water-filtration tank, of glass-fiber-TS polyester, which is 20 ft. in diameter by 32 ft. high, could be the largest low-pressure molded tank ever built and shipped in one piece.

266

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 267

Figure 4-20. These glass-fiber-TS polyester 4,OOO-gallon tanks installed in a marina permit boat owners to purchase gasoline at the pier. Before they were installed, gasoline either had to be carried to the marina or purchased elsewhere, because of corrosive conditions underground for metal or other tanks.

Today's underground tanks must last thirty or more years without undue maintenance. To meet these criteria they must be able to maintain their structural integrity and resist the corrosive effects of soil and gasoline, including gasoline that has been contaminated with moisture and soil. The tank just mentioned that was removed in 1988 met these requirements, but two steel tanks unearthed from the same site at that time failed to meet them. There was no record of how long the steel tanks had been in service, but one was dusted with white metal oxide and the other showed signs of corrosion at the weld line. Rust had weakened this joint so much that it could be scraped away with a pocketknife. Tests and evaluations were conducted on the tank that had been twenty-five years in the ground and also on similarly constructed tanks unearthed at five and a half and seven and a half years that showed the RP tanks could more than meet the service requirements. Table 4-2 provides factual, useful data from these tests. The chemical resistance of plastics is well known (see Chapter 6). Most materials suppliers have by now developed long-term data for the commonly used and other chemicals as well. Great care must be taken in selecting them, particularly regarding environmental conditions. For instance, two materials that do not attack a plastic when used separately may be troublesome when used in combination or diluted with water. And additives such as fillers, plasticizers, stabilizers, colorants, and catalysts can decrease or

268 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 4-21. These animal pens are made from Xycon hybrid resin RPs from the Amoco Chemical Co. The rods are a TS polyester reinforced with glass fibers.

increase the chemical resistance of unfilled or neat resins. Temperature is also important in all cases; careful tests must be made under the actual conditions of use in making a final selection. Of especial importance to chemical resistance, particularly in the RPs, is the processing method used. If, for example, a chemical and a mechanical component act simultaneously, cracking or fiber debonding can occur in the resin, considerably accelerating the diffusion

Figure 4-22. CorBan Industries of Tampa, Florida, mass produced this corrosion-resistant glassfiber-TS polyester filament-wound pipe in 6O-ft. lengths with diameters up to 12 ft. The 6O-ft. lengths of 54-in.-diameter pipe shown here for pulp and paper mill effluent lines were installed in 1968.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 269

Table 4-2. Data on RP/Composite Underground Gasoline Storage Tanks Unearthed after Different Periods Test Results Age at Testing 5.5 Years

7.5 Years

25.0 Years

Buried-excavated

117/65-8/21170

4/4/64-10/24171

5/15/63-5/11188

Flexural strength: Psi MPa

19,500 134

24,200 167

22,400 154

Property

Flexural modulus: Psi MPa Tensile strength: Psi MPa

725 x 103 4,992 10,700 74

795 X 1Q3 5,482 13,600 94

635 x 1Q3 4,378 10,500 72

Tensile modulus: Psi MPa

1,160 x 1Q3 7,260

1,053 X 103 8,000

1,107 X 103 7,630

Tensile elongation: %

l.ll

1.25

1.13

Notched Izod impact strength: ft.-Ib'/in. J/m

9.7 518

11.0 587

14.1 753

of the aggressive media to the glass fibers. Whereas the diffusion of aggressive media such as acids and alkalis proceeds slowly in resins, these media advance rapidly along glass fibers. The serviceability of these types of plastics in corrosive media can be guaranteed only if proper attention is given to processing variables like voids (see Chapter 7), including the fiber orientation and construction.

Figure 4-23. This 9l-ft.-diameter corrosion-resistant glass-fiber-TS polyester filament-wound stack and breech for a Texas chemical plant incorporates bell and spigot joints for ease of installation.

270 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Stress Cracking and Crazing Environmental stress cracking is the cracking of a plastic part that becomes exposed to a chemical agent while the part is under stress. This effect may be caused by exposure to such agents as cleaners or solvents. The susceptibility of affected plastics to stress cracking by a particular chemical agent varies considerably among plastics, particularly the TPs. If the stress is below a critical value, the plastic may sustain little or no damage during exposure. The resistance of a given plastic to attack may be evaluated by using either constantdeflection or constant-stress tests in which specimens are usually coated with but can also be immersed in the chemical agent. After a specified time the degree of chemical attack is assessed by measuring such properties as those of tensile, flexural, and impact (see Figs. 4-24 through 4-26). The results are then compared to specimens not yet exposed to the chemical. In addition to chemical agents, the environment for testing may also require such other factors as thermal or other energy-intensive conditions. A classic example illustrating the effects of stress cracking is the case of the PE milk bottle from the 1950s. A PE polymer and a process to blow mold the bottles were successfully integrated to the point where the lactic acid in the milk would not cause a premature split in the highly stressed neck area of the bottle. As noted, stress cracking is intensified by an increase in temperature. As an example, the results from testing HOPE pressure-pipe specimens in water at 82°C (180°F) show results in a life span of just a few hundred hours but when the water temperature is at 23°C (74°F) the life expectancy becomes fifty years. In both tests, water was moving through the pipes. It is possible with solvents of a particular composition to determine quantitatively the level of stress existing in certain TP moldings where molded-in stresses exist. The stresses need not be applied ones but can be residual (internal) stresses resulting from the molding or other forming process that was used to shape the plastic part. Solvent mixtures suitable for this type of test are available for materials such as PSs, PCs, and acrylics through resin suppliers, who can provide details.

t>

Control. no oil or previosly applied stress' 0 psi · (0 MPa)

~

,';i

Strain.£

Previously no stress or applied stress IHllng .16 hours with sample eoated with vegelable 011 prior to testing lor the short-term stress-strain ballavlor shown.

opsi . (0 MPa)

1000 psi· (6.9 Mpo)

2000 psi· (13.8 MPo)

<>

3000 psi · (20.7 MPa) <>

~I Strain, £

Strain. E

Strain, E:

Figure 4-24. An example of the influence of tensile stress-strain curves subjected to an environment that influences the ductility of a specific plastic.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 271

•

•

•

Figure 4·25. Two tensile test bars under the same stress were sprayed with acetone . The top one cracked quickly, but the other did not fail. Different plastics were used.

Thermoplastic cracking develops under certain conditions of stress and environment, sometimes on a microscale. Because there are no fibrils to connect surfaces in the fracture plane (except possibly at the crack tip), cracks do not transmit stress across their plane. Cracks result from embrittlement, which is promoted by sustained elevated temperatures and ultraviolet, thermal, and chemical environments existing in the presence of stress or strain. There appears to be no practical definition that can sufficiently distinguish between environmental and other stress cracking, although the micromechanics of the two types of cracking may be quite different.

5_

COW'

.,' f

W I'" Rtlt9tftl

&bow. Tooof

01S' l""'l _ • 00lO"0025 to 31 .... 1200 ...... P,ffl. JUI Tvbt

00.'"112111151

Ho" ttcII. 8Iau

SHC'"'C"

IBI

Hole"

Figure 4·26. In this test a specimen is bent in a container and subjected to a chemical agent. This apparatus is per ASTM D 1693.

272 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

For the designer it is unimportant whether cracking develops upon exposure to a benign or an aggressive medium. The important considerations are the embrittlement itself and the fact that apparently benign environments can cause serious brittle fractures when imposed on a product that is under sustained stress and strain, which is true of certain plastics. Crazing, or stress whiting, is damage that can occur when a TP is stretched near its yield point. The surface takes on a whitish appearance in regions that are under high stress. Crazing is usually associated with yielding. For practical purposes stress whiting is the result of the formation of microcracks or crazes, which is another form of damage. Crazes are not true fractures, because they contain strings of highly oriented plastic that connect the two flat faces of the crack. These fibrils are surrounded by air voids. Because they are filled with highly oriented fibrils, crazes are capable of carrying stress, unlike true fractures. As a result, a heavily crazed part can still carry significant stress, even though the part may appear to be fractured. It is important to note that crazes, microcracking, and stress whiting represent irreversible first damage to a material, which could ultimately cause failure. This damage usually lowers the impact strength and other properties of a material compared to those of undamaged plastics. One reason is that it exposes the interior of the plastic to attack and subsequent deterioration by aggressive fluids. In the total design evaluation, the formation of stress cracking or crazing damage should be a criterion for failure, based on the stress applied.

Testing In addition to testing for stress cracking, other useful tests are available for evaluating the chemical resistance of plastics under different conditions. Typical ASTM standards for such tests include: D 1693. "Environmental Stress Cracking Resistance of Ethylene Plastics (ESCRs)."

D 2552. "Environmental Stress Rupture of Type III Polyethylenes Under Constant Tensile Load." D 543. "Resistance of Plastics to Chemical Reagents-Establishes 50 Standards Reagents." D 570. "Water Absorption of Plastics." D 581. "Chemical Resistance of Thermoset Resins Used in Glass Fiber Reinforced Structure." D 1712. "Resistance of Plastics to Subtile Staining." D 2299. "Staining Tests."

WEATHER RESISTANCE Ultraviolet rays and the heat from solar radiation degrade the natural molecular structure of certain plastics. Acrylics, PCs, PPO (see Fig. 4-27), TFE, silicone, and TS polyester are plastics that have outstanding durability under UV exposure. The resistance to sunlight of those that degrade can be significantly improved by using chemical stabilizers and various fillers that can screen and protect the plastic from radiation. Weather resistant paints and coatings can also protect plastics from UV damage. The effects of UV radiation on degradable plastics are usually confined to the exposed surface layers. The general effect is one of embrittlement. Tensile strength may either

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 273

Figure 4-27. These roofing panels by N.ailite International of Miami are featured on GE Plastics' Living Environments Concept House seen in Chapter 1. The panels are made of GE's Noryl PPO resin, known for its excellent moisture, UV, wear resistance, and flame resistance. Nailite first marketed such panels in the Los Angeles area, where a ban on cedar shakes and shingles was being considered. increase or decrease, but the elongation upon breaking is always reduced. A loss of impact strength is the usual measure of UV degradation. The creep rupture strength will also be reduced dramatically, and the onset of the knee in the stress-strain curve of a PE, for example, will be accelerated. UV degradation is aggravated by stresses or strains, and the element may stress crack or craze after deterioration has occurred. The secondary effect of UV degradation is usually a yellowing or browning of the plastic. Other elements of weather and outdoor exposure can interact with UV radiation to accelerate degradation in degradable types of plastics. They include humidity, salt spray , wind, industrial pollutants, atmospheric impurities such as ozone, biological agents, and temperature. The wavelengths that have the most effect on plastics range from 290 to 400 nm (2,900 to 4,000 A). One of the insidious disadvantages of certain plastics is their tendency to absorb moisture from ambient air and then change their size and properties (see Table 4-3). There are protective measures that can be taken with certain plastics in regard to coatings, chemical treatments, additives, and so on. With it is practical, the best way to circumvent problems of this type is to select a plastic with the lowest possible absorption rate or design the product so that such complications do not develop. The unpredictable scheduling and high dollar costs of all-weather natural testing have brought much of the environmental testing into laboratories or other testing centers.

274 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-3. Water Absorption Values for Some Plastics, per ASTM D 570 Plastic PTFE PE, high density PP PVC PS PC PSU

POM Nylon 11 Polyvinyl butyral Nylon 6 Cellulose acetate

Water Absorption, wt % 0.00 <0.01 <0.01 0.03 0.05 0.15 0.22 0.25 0.25 1.0 1.3 1.7

Artificial conditions are provided there to simulate various environmental phenomena and thereby aid in evaluating the test item before it goes into service in natural environments. This environmental simulation and testing requires extensive preparation and planning. It is generally desirable to obtain generalizations and comparisons from a few basic tests to avoid prolonged testing and retesting. The type and number of tests to be conducted, whether natural or simulated, are as usual dependent on such factors as end-item performance requirements, time and cost limitations, past history, performance safety factors, the shape of specimens, the available testing facilities, and the environment. UV test data are usually obtained through actual outdoor exposure or in special test cabinets. Outdoor exposure tests may be as simple as attaching test samples to a surface at a angle suitable for the latitude where the test is being conducted or as complicated as having mirrors and sun-tracking equipment to accelerate the effective exposure. Test cabinets also are available to accelerate testing. Generally, they employ high-intensity xenon or carbon-arc lamps to generate high levels of UV exposure in a relatively short time.

STERILIZATION-IRRADIATION Fundamentally, radiation is the emission of energy in such forms as light and heat or the transfer of energy through space by electromagnetic waves. Irradiation basically identifies the radiant energy per unit of intercepting area. The effect of these energies on degrading plastics and in changing or improving their properties is measurable (see Table 4-4). Most nontechnical people consider only that radiation results in degradation, but the irradiation of plastics is an important science for packaging sterilized medical products, curing plastics, converting certain TPs to TSs, and so on. Sterilization is an important process that involves a major market for the use of plastics in packaging. The most common methods of sterilization are those using heat, steam (autoclaving), radiation, and gas (EtO-ethylene oxide) (see Tables 4-5 and 4-6). Unfortunately, each of these methods has its limitations. There are, however, plastics that do meet performance requirements based on the various different processes, including radiation.

t.;I

;;:

10

10·

102

103

10"

Dose in Megarads lOs

I I I I I I I I II I l I

o

Nylon - - - - - - - - - - - - - - - - - - - - Methylmethacrylate - - - - - - - Butyl and fluorocarbon plastics - - - Silicone rubber - - - - - - - - - - - Polyester film - - - - - - - - - - - - - - Phenolic-urea - - - - - - - - - - - - - Cellulose acetate - - - - - - - - - - - Vinyls - - - - - - - - - - - - - - - Polyethylene - - - - - - - - - - - - - Natural rubber - - - - - - - - - - - - Melamine (general purpose) - - - - - - - - - Polystyrene - - - - - - - - - - - - - - Unfilled semiflexible epoxy - - - - - - - - - - Filled semiflexible epoxy - - - - - - - - - - Rigid epoxy (general purpose) - - - - - - - - - Silicone glass laminate - - - - - - - - - - - - - Mineral-filled phenolic - - - - - - - - - - - - Heat-resistant filled epoxy - - - - - - - - - - - Mica-glass laminate - - - - - - - - - - - - - - - - - Inorganics: Mica, ceramics, glass - - - - - - - - - - - - - -

PTFE - - - - - - - - - - -

Material

Loses 40% of its tensile strength, crumbles at higher doses Loses elongation properties rapidly, tensile strength remains Serious deterioration Deterioration beginning Loses its elasticity Threshold mechanical damage: becomes weak and brittle Little effect Loses 25% mechanical properties: electrical properties unaffected Loses up to half its tensile strength and elongation Loses 75% elongation, 15% tensile strength Loses 25% of its properties Loses 25% or more of its strength Resistivity decreases, mechanical properties remain Little evidence of change Produces stiffening and some blisters May crack, become weak and brittle Little or no effect Little effect No deterioration Virtually unaffected Resistivity temporarily changed, but few other effects noted

Comments

Table 4-4. Radiation Damage to Common Insulating and Packaging Materials

276 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

PERMEABILITY AND BARRIER RESISTANCE In the past, the usual materials used to contain food, gasoline, chemicals, perfumes, medication, and many other items and keep them from permeating or being contaminated were metal and glass. For a century now, however, plastic containers have been entering the arena of packaging. At first only certain plastics could be used, which were usually rather thick or heavy compared with what is used today. There have been various plastics that could provide permeability protection (see Table 4-7). With the growth of plastics use in containers and packages, requirements to make them more compatible or useful developed. The two major approaches for providing permeability resistance in plastic containers involve chemically modifying the plastics' surfaces and, more important from a marketing standpoint, the use of barrier plastics with nonbarrier types to meet cost-toperformance requirements. This is achieved through coextrusion, coinjection, and other such processes [1-15, 62-68, 356-58]. Chemically modifying a plastic's surface during or after fabrication permits controlling the permeation behavior of such parts as diaphragms, film, and containers. These techniques are becoming increasingly important. There is now a search on for better barrier materials for packaging applications, in particular to produce blow-molded gasoline containers. The amount of gasoline permeation through the presently used HDPE tanks, even though it is low, is still excessive, thus requiring some type of barrier. Such a barrier can be created by a layer of functionalized PE formed on the inside of the container wall by a chemical reaction, mostly sulfonation or fluorination [12]. Oxifluorination is a new process in which fluorine gas is thinned with nitrogen to which several percent of oxygen by volume have been added [357]. Subjecting PE to fluorine and oxygen at the same time leads to functionalization of the PE, making it impermeable. This technique permits substantially reducing the required amount of fluorine, resulting in a cost-to-performance improvement. Barrier plastics using oxifluorination are widely used for foods. With these, barriers are needed to protect them against spoilage from oxidation, moisture loss or gain, and changes or losses in favor, aroma, or color. Most plastics can be considered barrier types to some degree, but as barrier properties are maximized in one area (as the gases such as O2 , N2 , or CO 2), such other properties as permeability and moisture resistance diminish (see Table 4-8 through 4-12). To design a multilayer package or container that must have barrier properties, the usual approach could consume time and costly shelf-life testing. One technique that promises to reduce much of this time and cost is called CABD, for Computer-Aided Barrier Design, a database developed by the EV AL Company of America (EV ALCA). This method allows the designer to select only those barrier structures that are most likely to pass the shelf-life testing. Mathematical models can now predict the performance of constituent materials through the full range of processing, sealing, storage, and retail sale conditions. Diffusion/T ransport Properties

The ability of a plastic to protect and preserve products in storage and distribution depends in part upon the diffusion (i.e., transport) of gases, vapors, and other low-molecularweight species through the materials. A substance's tendency to diffuse through the polymer bulk phase is its diffusivity or diffusion coefficient (D). The rate of diffusion is related to the resistance, within the polymer wall, to the movement of gases and vapors. Two important aspects of the transport process are permeability and the migration of

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 277

Table 4-5. General Guide for Plastics' Behavior in Sterilization Peformance Plastics

Dry Heat

PE-LD PE-HD ElVR PP TPX

Autoclaving

Boiling Water

+ +

PVC rigid PVC flexible

Radiation

Gas

+ + +

+ + + + +

+ +

+ +

+

+

+ +

0

0

+

0

+ +

0

PS SAN ABS

0 0

+ + +

PMMA cast PMMA granulate

+

0

0

Polyamide Cellulose

0

0 0

+

+

+ +

+ +

Polyimide

+

+

+

+

+

PTFE PCfFE FEP

+ + +

+ + +

+ + +

+ +

+ + +

0

+

+

+

PC PSU PES PPS

+ + +

+ + +

+ + +

+ + +

+ + +

Phenolic resin Carbamic resin Melamine resin

+

+ + +

+ + +

+ + +

+ + +

0

0

+

+

+

+

+

+

0

+

+

Epoxy resin

+

Silicone PUR Key: + suitable; - unsuitable;

0

restricted suitability.

additives. Possible migrants from plastics can include residual monomers, low molecularweight polymers, catalyst residues, plasticizers, antioxidants, antistatic agents, chain transfer agents, light stabilizers, FR (fire resistant) agents, polymerization inhibitors, reaction products, decomposition products, lubricants and slip agents, colorants, blowing agents, residual solvents, and others.

Permeability The driving force for gases and vapors penetrating or diffusing through, for example, penneable packages is the concentration difference between environments inside and

Table 4-6. Mechanical Property Data after Various Steam Sterilization Cycles Using Vernitron model 8020 and 8089 Units Number of Sterilization Cycles Mechanical Property of Plastics Polycarbonate Tensile strength (psi) Tensile impact strength Poly(ester-carbonate) Tensile strength (psi) Tensile impact strength Poly(etherimide) Tensile strength (psi) Tensile impact strength Poly( ether sulfone) Tensile strength (psi) Tensile impact strength Polysulfone Tensile strength (psi) Tensile impact strength

Initial

20

40

60

80

100

(ft.-lb./in.2)

10,400 220

9,300 79

9,500 78

9,180 9

9,070 5

5,290 3

(ft.-lb./in.2)

9,950 88

11,200 70

11,300 20

10,500 15

10,100 12

4,770 2

(ft.-lb./in.2)

15,400 81

15,800 84

16,100 76

16,100 83

16,300 79

17,300 34

(ft.-lb./in.2)

11,600 164

13,400 100

13,500 101

14,000 97

14,400 100

14,600 87

(ft.-lb'/in.2)

10,500 153

12,100 105

12,200 113

12,300 91

12,400 98

13,000 87

Table 4-7. Permeability Resistance of Different Plastics Oxygen Barrier

Resin HDPE, LDPE EVA Polystyrene Ionomers Polypropylene PVC ABS Polyester PVDC Nylon

Moisture Barrier

Grease Resistance

X X X

X X

X

X X X X X X X

Toughness

Heat Sealability

Cost per Cu. In.

X

X

X

X

Low Low Medium Low Low Medium High High High

X X

Key: X identifies penneability resistance.

Table 4-8. Advantages and Disadvantages of Barrier Plastics Used in Food Packaging Material

Advantages

Disadvantages

High-nitrile (Barex)

Good 02/C02 barrier Monolithic or coextrusion Scrap reuse Not moisture sensitive

Moderate moisture barrier Nonbeverage FDA approval Moderate impact resistance Limited grade offering

EVOH

Excellent 02/C02 barrier Scrap reuse Extended grade offering

Moisture sensitive

PVDC

Excellent 02/C02/H20 barrier Coextrusion, lamination or coatings

Difficulty in scrap reuse No monolithic structures

Nylon (Selar PA)

Moderate 02/C021H20 barrier Monolithic or coextrusion

Moderate 02/C021H20 barrier High-cost O2 barrier

Nylon (MXD6)

Excellent 02/C02 barrier Potential low-cost O2 barrier

Moderate moisture barrier No commercial monolithic containers

278

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 279

Table 4-9. EVOH Odor Permeability* Film

Methyl Ethyl Ketone

HDPElEVOHlEVA (1.25 mil) PP/EVOHlPP (1.0 mil)

Methyl Salicylate

B-Pinene

Toluene

0.0011

0.0013

0.027

0.009

0.0036

0.0003

0.65 <0.01

Styrene 0.0610 <0.0002

PP/PET/PP

2.40

2.16

0.0088

1.310

0.0018

(1.0 mil) PVDC-coated oriented PP (1.8 mil)

0.44

0.871

0.0320

0.470

0.0046

'In g/m 2 x 24 hrs. x 100 ppm @ 73"F, 0% RH (tie layers omitted for clarity).

outside the package. A diffusing substance's transmission rate is expressed by mathematical equations commonly called Fick's First and Second Laws of Diffusion: dC

F=-D dX dC dt

(4-1)

= Dd2C

(4-2)

tfX2

where F = flux (the rate of transfer of a diffusing substance per unit area), D = diffusion coefficient, C = concentration of diffusing substance, t = time, andX = space coordinate measured normal to the section. To measure gas and water vapor permeability, a film sample is mounted between two chambers of a permeability cell. One chamber holds the gas or vapor to be used as the permeant. The permeant then diffuses through the film into a second chamber, where a detection method such as infrared spectroscopy, a manometric, gravimetric, or coulometric method; isotopic counting; or gas-liquid chromatography provides a quantitative measurement. The measurement depends on the specific permeant and the sensitivity required. Three general test procedures used to measure the permeability of plastics films are the absolute pressure method, the isostatic method, and the quasi-isostatic method. The absolute pressure method (see ASTM D 1434-66, "Gas Transmission Rate of Plastic Film and Sheeting") is used when no gas other than the permeant in question is

Table 4-10. Permeation Rates of Organic Solvents on Selected Plastics* Solvent

Film

Thickness (mils)

Chloroform

Xylene

MEK

Kerosene

EVAL®EF-E EVAl®EF-F EVAL®EF-XL LOPE Oriented PP Oriented nylon PET

0.8 0.8 0.6 2.0 0.8 1.0 1.0

0.20 0.13 0.01 178.1 241.3 0.87 20.0

0.09 0.07 0.03 20.97 22.58 0.06 0.11

0.31 0.25 0.02 4.77 0.77 0.17 0.10

<0.003 <0.003 <0.003 4.90 3.42 0.02 0.03

'In g/IOO in. 2 x 24 hrs. @ 2O'C, 65% RH (organic solvent @ 2O'C).

280 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-11. Barrier Properties of Selected Commercially Available Plastics

Polymer

O2 Transmission Rate @ 25°C 65% RH cc.milll00 in.2/24 hrs.

Water Vapor Trans. Rate @40°C9O% RH gm.milll00 in. 2124 hrs.

Ethylene vinyl alcohol Nitrile barrier resin High-barrier PVDC Good barrier PVDC Moderate barrier PVDC Oriented PET Oriented nylon Low-density polyethylene High-density polyethylene Polypropylene Rigid PVC Polystyrene

0.05-0.18 0.80 0.15 0.90 5.0 2.60 2.10 420 150 150 5-20 350

1.4-5.4 5.0 0.1 0.2 0.2 1.2 10.2 1.0-1.5 0.3-0.4 0.69 0.9-5.1 7-10

present. Between the two chambers a pressure differential provides the driving force for permeation. Here the change in pressure on the volume of the low-pressure chamber measures the permeation rate. With the isostatic method, the pressure in each chamber is held constant by keeping both chambers at atmospheric pressure. In the case of gas permeability measurement, there must again be a difference in permeant partial pressure or a concentration gradient between the two cell chambers. The gas that has permeated through the film into the lower-concentration chamber is then conveyed to a gas-specific sensor or detector by a

Table 4-12. General Comparison of Metalized Coextruded Polyolefin (PE) and Aluminum-Foil Laminate Metalized Coextruded Polyolefin Tensile strength MD CD Mullen strength Gurley stiffness MD CD WVTR (grn/csll24 hrs.) Oxygen transmission (cc/csIl24 hrs.) Light transmission Seal type Seal range (40 psi, 15 sec.) Deadfold (subjective) (1-10 Scale) Flex crack resistance (subjective) (1-10 Scale)

Foil Laminate*

18-19 12-13 19-20 70-75 42-47 Approx .. 05

18-19 11 17 117-112

Approx. 10

Less than .004

Slightly less than 1% Fin onlyt 350-500°F

Approx.O%

4

7

8

5

*.0003" gauge foil. wax laminated to 12'12# paper, wax laminated to 8'/2# tissue. tWhile a lap seal is technically possible, the bond is too weak to be considered commerical.

72-77 .0006

Fin or Lap 160--350"F

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 281

carrier gas, for quantitation. Commercially available isostatic testing equipment has been used extensively for measuring the oxygen and carbon dioxide permeability of both plastic films and complete packages. The quasi-isostatic method is a variation of the isostatic method. In this case at least one chamber is completely closed, and there is no connection with atmospheric pressure. However, there must be a difference in penetrant partial pressure or a concentration gradient between the two cell chambers. The concentration of permeant gas or vapor that has permeated through into the lower-concentration chamber can be quantified by a technique such as gas chromatography. Three related methods based on the quasi-isostatic method are used to measure permeability. The most commonly used technique allows the permeant gas or vapor to flow continuously through one chamber of the permeability cell. The gas or vapor permeates through the sample and is accumulated in the lower-concentration chamber. At predetermined time intervals, aliquots are withdrawn from the lower cell chamber for analysis. The total quantity of accumulated permeant is then determined and plotted as a function of time. The slope of the linear portion of the transmission-rate profile is related to the sample's permeability. BIOLOGICAL AND MICROBIAL DEGRADATION

Certain plastics such as TP polyesters, polyurethanes, cellulosics, and plasticized PVC can be degraded by microorganisms. It has been observed that enzymes attack noncrystalline regions preferentially. As a result, the resistance of susceptible polymers to microbial degradation is related directly to the degree of crystallinity of these polymers. They remain relatively immune to attack as long as their molecular weight remains high. Most plastics are characteristically durable and inert in the presence of microbes. This stability is important to plastics' long-term performance. However, for some applications only short-term performance is desired before the product is discarded, as in the fast-food and packaging markets. In such cases it is considered advantageous for discarded plastic to degrade when exposed to microbes. There thus exists a requirement to develop or modify plastics possessing the properties required for their service life, but with the capability of degrading in a timely and safe manner, particularly to handle the worldwide waste situation. The amount of degradation of plastics under the action of bacteria and fungi is of interest because of land-shortage problems in solid-waste management and litter accumulation and other environmental problems on land and sea. The agricultural use of plastics in mulch, films, seeding pots, and binding twines has increased significantly, making biodegradation a desirable feature in plastics, to minimize disposal and soilpollution problems. Many areas worldwide have enacted or are considering legislation to require that disposable plastics be degradable (Chapter 12). With plastics designed to include degradability, the designer could have a monumental problem in ensuring that degradation will occur only after a product's useful life is over. The deterioration of plastics by biological agents should be distinguished from other forms of plastics degradation. Many other types of plastics degradation may be classified clearly as chemical in nature. In them a deteriorative agent causes a chemical degradative reaction to occur. Chemical bonds are broken or new ones established. Different molecular species, of a molecular size smaller or larger than the original desirable species, are formed, and these species no longer have the properties for which the original plastic was chosen.

282 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

This generalization is also true for the degradation caused by heat, electromagnetic radiation, oxygen and ozone, and high-energy nuclear radiation. It is also true for the chemical degradation caused by acids, bases, or other strongly reactive chemical agents. The reaction types include oxidation, ozonization, radical formation, crosslinking, chain scission, and others. The symptoms are described as hardening, embrittlement, softening, cracking, crazing, discoloration, or alteration of specialized properties such as dielectric strength. The situation with some forms of biological deterioration is somewhat different. Where the agent is macrobiological, as in the case of rodents, insects, and marine borers, the attack is physical in nature, such as by gnawing or boring. The attack is not at the atomic or molecular level. Any breaking of molecular bonds such as in polymer chain shortening is thus accidental. The attack may be said to be at the material's structural level, not the polymer molecule level. An important item to note is that most commercially used plastics are not singlecomponent pure substances. Practically always, the basic polymer itself, rarely if ever a single molecular species, is compounded with other components such as plasticizers, pigments, antioxidants, and other additives. More often than not, then, biological susceptibility is due to the nonpolymer component. Plastics' deterioration can be classified as either by a microorganism, a macroorganism, or a marine organism (both micro and macro). In the case of microbiological agents, as in fungal and bacterial deterioration, the polymer alterations are caused by chemical attack. This has been demonstrated for the attack on the natural polymer cellulose by fungi through the cellulase enzymes, for many esters, and for many hydrocarbons. It is not yet so clearly proven for the many synthetic polymers, but there is sufficient evidence that may be ascribed to enzyme action as being probably the chief mechanism. Thus, although the medium of attack is biological, the destructive agents are chemical. Fungal and bacterial deterioration are identified as microbiological and have always caused problems to materials. Fungal attack on plastics has received a great deal of attention beginning with the early days of World War II, when the tropical theaters served to focus attention on the overall problem of materials deterioration. Microbial deterioration of plastics is intimately involved with the moisture problem, especially with regard to plastics in electronic equipment. For this reason much of the literature treats the two problems together. Furthermore, there is often confusion between the deterioration of the electrical properties of plastics, more often than not a moisture phenomenon, and actual deterioration of the substance of the polymer. Most investigators agree that in the electronics field moisture accounts for the greater effect. Often, if the moisture problem is solved the fungal aspect is also overcome, because of the dependence of organisms on water. Yet not all of this twin problem may be ascribed to moisture, for there are instances where microorganisms are able to destroy the substance of a polymer or attack the nonpolymer constituents of a plastic formulation. Furthermore, as one shifts attention from plastics in electronic equipment to other items where plastics are used, there are clear-cut cases of destruction by fungi. Examples may be found in films, fibers, and coatings.

FLAMMABILITY When plastics are used, their behavior in fire must be considered. Ease of ignition, the rate of flame spread and of heat release, smoke release, toxicity of products of combustion, and other factors must be taken into account. Some plastics burn readily, others only

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 283

Table 4-13. Heat Values for Different Building Materials* Potential Heat

Material Woods Douglas fir, untreated Douglas fir (retardant treatment "A") Douglas fir (retardant treatment "B") Douglas fir (retardant treatment "C") Maple, soft, untreated Hardboard, untreated Plastics Polystyrene, wall tile Rigid, polyvinyl chloride, retardant treated Phenolic laminate Polycarbonate resin Insulation Glass fiber, semirigid, no vapor barrier Rock wool batting, paper enclosure Roof insulation board Cork (reconstituted cork sheet) Cellulose mineral board Concrete Cinder aggregate Slag aggregate Shale aggregate Calcareous gravel aggregate Siliceous gravel aggregate Miscellaneous Paint "E" (dried paint film) Asphalt shingles (fire retardant) Building paper (asphalt impregnated) Building paper (rosin sized) Linoleum tile Brick, red, face Charcoal, coconut

Thickness (in.) (mm)

Density (lb./ft.3) (g/cm3)

Weight Basis (Btu/lb.) (cal/g)

Volume Basis (Btu/ft. 3)

319 x 103 308.0

0.75 (19) 0.75 (19)

38.0 37.2

(0.609) (0.596)

8,400 8,290

(4,670) (4,600)

0.75 (19)

47.2

(0.756)

7,860

(4,370)

371.0

0.75 (19)

38.8

(0.622)

7,050

(3,920)

274.0

1.0 (25) 0.25 (6.3)

39.5 59.8

(0.633) (0.958)

7,940 8,530

(4,410) (4,740)

314.0 510.0

0.075 (1.9) 0.147 (3.73)

65.4 86.0

(1.05) (1.38)

17,420 9,290

(9,680) (5,160)

1,140.0 799.0

0.063 (1.6) 0.25 (6.3)

76.4 78.7

(1.22) (1.26)

7,740 13,330

(4,300) (7,406)

592.0 1,050.0

1.0

(25)

3.0

(0.048)

3,040

(1,690)

9.1

3.0

(76)

2.4

(0.038)

1,050

(583)

2.5

10.4 14.8 47.8

(0.167) (0.238) (0.766)

3,380 11,110 2,250

(1,880) (6,172) (1,250)

35.1 164.0 108.0

93.0 110.1 80.5 133.1 166.8

(1.49) (1.764) (0.0206) (2.132) (2.672)

3,080 80 10 -250 -40

(1,710) (5.5) (- 77) ( -22)

286.0 8.9 0.5 -33.1 -6.7

0.05 (1.3) 0.25 (0.64) 0.042 (Ll)

70.7 42.8

(1.13) (0.686)

3,640 8,320 13,620

(2,020) (4,620) (7,567)

588.0 583.0

0.018 (0.46) 118 (3.2) 2.25 (57)

23.6 86.0 139.1

(0.378) (1.38) (2.228)

7,650 7,760 20 13,870

(4,250) (4,310) (1Ll) (7,706)

1.0 (25) 0.25 (6.3) 2.0 (5.1)

(44)

181.0 667.0 2.2

• All weights and percentages refer to original air-dry weight.

with difficulty, and still others do not support their own combustion (see Table 4-13). A plastic's behavior in fire depends upon the nature and scale of the fire as well as the surrounding conditions. Fire is a highly complex, variable phenomenon, and the behavior of plastics in a fire is equally complex and variable [1, 14, 15,62-68, 361-72). Early in this century it was thought that the matter of fire hazard was simple enough: does the material bum, or not? Wood bums; steel does not. Although these statements are certainly true, they are almost irrelevant to the relative fire risk of the two materials. Compare fires in two different buildings, one framed of heavy timbers or plastic bonded-

284 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-14. Ignition Temperatures of Various Plastics Self Ignition

Rash Ignition

Material

°C

of

°C

of

Polyethylene Polypropylene Polytetrafluoroethylene Polyvinyl chloride Polyvinyl fluoride Polystyrene SBR (styrene butadiene rubber) ABS (acrylonitrile butadiene styrene) Polymethyl methacrylate PAN (Polyacrylonitrile) Cellulose (paper) Cellulose acetate 66 Nylon cast 66 Nylon spun and drawn Polyester

350 550 580 450 480 490 450 480 430 560 230 470 450 530 480

662 1,022 1,076 842 896 914 842 896 806 1,040 446 878 842 986 896

340 520 560 390 420 350 360 390 300 480 210 340 420 490 440

644 968 1,040 734 788 662 680 734 572 896 410 644 788 914 824

laminated wood arches and the other of steel framing. The steel frame, particularly if it is of light steel, will collapse after only a few minutes of exposure to fire, but it may require a fire of long duration to bring down the timber-framed building. Fire reaches 1,370°C (2,500°F), and steel basically takes only up to about 538°C (I,OOO°F), making it collapse like a pretzel. Wood, like certain plastics, can take the heat, and it takes a rather long time to self-destruct, thus giving time for people to leave the scene of the fire. Fire tests of plastics, like fire tests generally, are frequently highly specific, with the results being specific to the tests. The results of one type of test do not in fact often correlate directly with those of another and may bear little relationship to actual fires. Some tests are intended mainly for screening purposes during research and development, whereas others, such as large-scale tests, are designed to more nearly approximate actual fires. Consequently, such often-used terms as self-extinguishing, nonburning, jfame spread,

Table 4-15. Decomposition Ranges (Td ) for Various Plastics Temperature Material

°C

~

Polyethylene Polypropylene Polyvinyl acetate Polyvinyl chloride Polyvinyl fluoride Polytetrafluoroethylene Polystyrene Polymethyl methacrylate Polyacrylonitrile Cellulose acetate Cellulose 6 Nylon 66 Nylon Polyester

340-440 320-400 215-315 200-300 370-470 500-550 300-400 180-280 250-300 250-310 280-380 300-350 320-400 280-320

645-825 610-750 420--600 390-570 700-880 930-1,020 570-750 355-535 480-570 480-590 535-715 570-660 610-750 535-610

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 285

and toxicity must be understood in the context of the specific tests with which they are used (see Table 4-14 through 4-16). Some materials may burn quite slowly but may propagate a flame rapidly over their surfaces. Thin wood paneling will burn readily, yet a heavy timber post will sustain a fire on its surface until that is charred, then smolder at a remarkably slow rate of burning. Bituminous materials may spread a fire by softening and running down a wall. Steel of course does not burn, but is catastrophically weakened by the elevated temperatures of a fire. PVC does not burn, but it softens at relatively low temperatures and emits irritating hydrogen chloride fumes. Other plastics may not burn readily but still emit copious amounts of smoke. And some flammable plastics, such as polyurethane, may be made flame retardant (FR) by incorporating in them additives such as antimony oxide. Other plastics basically do not burn, such as silicone and fluorine (see Chapter 6). The principles of good design for fire safety are as applicable to plastics as to other materials. The specific design must be carefully considered, the properties of the materials taken into account, and good engineering judgment applied. When evaluating the fire risk that exists with plastic products it is always best to perform appropriate tests on the end items. However, it is often helpful to select plastic materials for specific applications by first evaluating the flammability of the plastics under consideration in laboratory tests. These tests, often used for specifying materials, fall into the category either of smallscale or large-scale tests. Of course, as in evaluating any properties, having prior knowledge or obtaining reliable data applicable to fire or other requirements is the ideal situation.

Small-Scale Burning In small-scale fire tests, as in many laboratory screening tests, several stages are involved. At relatively low temperatures, as from 80 to 100°C (175 to 212"F), slow oxidation occurs. This feature, which is also characteristic of aging, is often enhanced as temperatures increase. As the temperature is raised closer to 100°C (212"F), the process is accelerated. When the temperature becomes high enough, from 200 to 300°C (390 to 570"F), the process becomes exothermic in the presence of air (oxygen), which is to say

Table 4-16. Specific Heats for Various Plastics

·C·

Material

Temperature callg.

Polyethylene Polypropylene Teflon Polyvinyl chloride Polyvinyl fluoride Polystyrene SBR (styrene butadiene rubber) ABS (acrylonitrile butadiene styrene) Cellulose acetate 6 Nylon 66 Nylon Polyester Phenol formaldehyde Epoxy resins Polyimide

0.55 0.46 0.25 0.25 0.30 0.32 0.45 0.35 0.40 0.38 0.40 0.30 0.40 0.25 0.27

*The specific heat expressed as 811J/1b. "F has the same value.

286 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

that heat is evolved, giving off decomposition products and usually volatile flammable products. Thermoplastics soften or melt, whereas thermosets characteristically maintain their shape. If more heat is added, autoignition will occur, at about 400°C (7500 P), resulting in the combustion of many plastics. However, there are plastics that do not melt, even above 538°C (l,OOOOP) (see Chapter 6). Large-Scale Burning

Large-scale tests can evaluate the contribution of plastic components to a full-scale fire external to the application. This type of test is often used to obtain a preliminary indication of the ways plastics might contribute to a fire. The foregoing description of the successive stages of decomposition and ignition of plastics pertains to small-scale fires that are generally conducted in a laboratory. In real fires, as in a room of a building, the same reactions may take place but the scale and the temperatures involved will be much larger and more complex, leading to phenomena not found in small-scale, controlled laboratory burning. The following stages are generally encountered in real fires: ignition, buildup and spread, flashover, a fully developed fire, and ultimately its propagation. Smoke

Toxic smoke and fumes have became generally recognized as the major cause of fire deaths, making the combustion products released by burning plastics and other materials particularly important. Smoke is recognized by firefighters as being in many ways more -800 ;-TEST CONOITIONS

-700

American National Bureau of Standards Smoke Chamber

600

3.2 mm (0.126 in) samples

r-

Flaming condition

r--

-500

r-400 300

r--

f-200

r--

nn n

100 0 !\'Q

~C:J

.~'Ii

'?'

,~~ ,,~

~<j

~~

No" ~

I'::-~

,'Q

<}~

~~

~

I'::-~ -s-<::;

~~

~",\<j ~

!\.~

~'Ii

K)<::;

~

~"'\v

~~~

~

~

~v

1'::-<::;

-S-'Q

~

,~

~~+

,~,v

~

~

Figure 4-28. An example of smoke emission upon the burning of some plastics compared to Victrex PES, from ICI-LNP.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 287

dangerous than actual flames because 1) it obscures vision, making it impossible to find safe means of egress, thus often leading to panic; 2) it makes helping or rescuing victims difficult if not impossible; and 3) it leads to physiological reactions such as choking and tearing. Smoke from plastics, wood, and other materials usually contains toxic gases such as carbon monoxide, which has no odor, often accompanied by noxious gases that may lead to nausea and other debilitating effects as well as panic. Whether a plastic gives off light or heavy smoke and toxic or noxious gases depends on the basic polymer used, its composition of additives and fillers, and the conditions under which its burning occurs. Some plastics bum with a relatively clean flame, but some may give off dense smoke while smoldering. Others are inherently smoke producing (see Fig. 4-28). The composition of the smoke depends upon the composition of the plastic and the burning conditions, as with other organic materials. In a particular application, therefore, careful consideration should be given to the relative importance of smoke and flame, including creating designs favoring the rapid elimination of smoke by venting, for fending off smoke, and other approaches.

Tests Different regulations, such as those of the Federal Aviation Administration, Department of Transportation, and local building codes, mandate that the designs of certain products comply with specific flammability test requirements. Flame-retardancy requirements generally include limits on flame spread, burning time, dripping, and smoke emission. A multitude of flammability tests have been developed, with more than 100 known just in the United States. The most common ASTM tests are given in Table 4-17. By far the most stringent and most widely accepted test is UL 94, concerning electrical devices. This test, which involves burning a specimen in a vertical position, is the one used for most flame-retardant plastics. In this test the best rating is UL 94 V-O, which identifies a flame with a duration of 0 to 5 s, an afterglow of 0 to 25 s, and the presence of no flaming drips to ignite a sample of dry, absorbent cotton located below the specimen. The ratings go from V-0, V-I, V-2, and V-5 to HB, based on specific specimen thicknesses. The flame spread and dripping tendencies of test materials are also characterized in ASTM standard D635. In this a horizontal test specimen provides the results of the average time of burning (ATB) and average extent of burning (AEB). In both the UL and ASTM tests, the presence of glass fibers and other reinforcements or fillers improves flammability ratings and significantly inhibits dripping. A more quantitative measure of a material's resistance to burning can be determined from ASTM D 2863. This standard measures the minimum concentration of oxygen in an oxygen-nitrogen mixture that will support candlelike burning for three minutes or longer. The results are reported as a Limiting Oxygen Index (LOI). Composites with LOIs above 28 percent are usually listed as UL 94 V-O. Obviously, the higher the LOI value (that is, the more oxygen needed), the lower the combustibility. Since air contains about 21 percent oxygen, any rating below 21 will probably support combustion in a normal, open environment. Smoke emission is measured in an air column above a burning specimen in a National Institute of Standards & Technology (previously the National Bureau of Standards) smoke chamber (see Fig. 4-28). In the NIST test a specified area of plastic is exposed to heat under flaming conditions, with smoke measurements being reported as "specific optical density." This is dimensionless value represents the optical density measured over a unit of path length within a chamber of unit volume that is produced from a test specimen of

288 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-17. Common ASTM Flammability Tests Property Ignitability

Hame spread Smoke

Oxygen content

Test ASTM ASTM UL94 ASTM ASTM ASTM ASTM ASTM ASTM

DI929 0635 EI62 E84 E662 D2843 E84 02863

Type Setchkin apparatus Bunsen burner Bunsen burner Radiant-panel test Steiner tunnel test NBS smoke chamber XP-2 smoke chamber Steiner tunnel test Glass column

unit surface area. The optical density measurement (Dmax) is based on the amount of attenuation of a light beam by smoke accumulating within the closed chamber during flaming combustion. As a reference, the Dmax for red oak is 76. Smoke generated during combustion consists of suspended soot particles that form between the pyrolysis zone and the flame's front. These particles are molecules of highly condensed ring structures that are most readily formed by aromatic polymers such as SAN, SMA, and polyphenylene ether. The polymers having aliphatic carbon backbones, such as polypropylene and nylon, tend to generate less smoke, but in the FR compounds this effect is offset by an increase in smoke caused by halogenated flame-retardant additives. Plastics with a higher thermal stability, such as PC, PSF, PES, PEEK, and PPS, produce the least smoke of the available UL 94 V-O TPs. There are seemingly endless programs to better understand fire tests and continually develop more realistic fire tests. Ohio State University has one with specific heat limits. The NIST has a cone calorimeter for heat release that is more sophisticated than the OSU one. And the National Institute of Building Science has an evolutionary version of the NIST smoke and toxicity test [361]. The outcome of fires involving plastics in buildings and transportation vehicles and the odds of survival for the occupants can be predicted by a personal computer program called Hazard I, developed by the Center for Fire Research, which is a part of the NIST, in Gaithersburg, Maryland. Based on a user-constructed scenario, Hazard I draws on its modules and databases to quantify such key fire variables as flame spread, oxygen depletion, and smoke and toxic-gas generation as fire spreads through imaginary premises. Any combination of furniture, furnishings, and building products, with their related plastics and ignition conditions, can be specified. Besides showing the types and amounts of combustion by-products, this program also figures the amount of time available for escape and, based on behavior models, predicts the likely number of fatalities among the occupants and their probable cause of death. Hazard I has obvious applications for writing fire-code standards and could be useful for establishing liability in fires involving fatalities. It can also be used by compounders to predict the performance of developing FR plastics and to compare plastics. This software package is available from the National Fire Protection Association, Batterymarch Park, Quincy, Massachusetts, 02269. THE OCEAN ENVIRONMENT

From ships to submarines to mining the sea floor, certain plastics can survive sea environments, which are considered more hostile than those on earth or in space. For water-

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 289

Figure 4-29. Extensive use has been made of both unreinforced and reinforced plastics in boats such as this U.S. Navy aircraft flattop for structural and nonstructural parts, electrical devices and wiring, electronic scanners and devices like radomes, optically transparent devices, food storage and dispensing devices, medical systems, buoyant devices, temperature insulation, and many more, particularly of plastics that resist damage from saltwater.

surface vehicles many different plastics have been designed and used successfully in both fresh and the more hostile sea water (see Table 4-18 and Fig. 4-29). Boats have been designed and built up to at least 37 x 9 m (120 x 30 ft.) in RP. This overview highlights underwatt;r developments, since the familiarity of plastics in this regard is rather limited. Plastics have already become vital for operating within the sea, even though comparatively little is yet known about the qualities of the sea (see Fig. 4-30). This frontier's practical opportunities were first developed with submarines, which until the nuclear ones were limited to depths of only a few hundred feet. Many thousands of feet can now be navigated. The crushing pressures below the surface, which increase at a rate of about 1 psi per foot of depth, make corrosion a major threat to the operation and durability of

many materials. For example, the life of uncoated magnesium bolts in contact with steel nuts is less than seventy-two hours, and aluminum buoys will corrode and pit after only eleven months at just four hundred feet. Tests on plastics in deep water have been extremely encouraging. Low-carbon steel corroded at a rate one-third greater than in surface waters. Filament-wound reinforced plastic cylinders and PVC buoys retained their strength. PVC washers and the silicone sealing compound used in steel-to-aluminum joints helped prevent corrosion. Black twisted nylon and polypropylene ropes used to rig and retrieve test platforms were unaffected. Grappling lines attached to platforms, made of steel wire jacketed with extruded high-density PE, prevented corrosion of the steel. PE is also used to protect submerged telephone cables. Plastic primers such as epoxy are used to prevent antifouling paints from corroding metals. These paints generally use cuprous oxide to prevent the growth of barnacles, but at the same time can be harmful to metal. Plastics are used successfully in instruments to determine depth, the velocity of currents, temperature, and as echo sounders. Parts operating to depths of 4,500 m (15,000 ft.) include molded polystyrene rotors, neutrally buoyant polyethylene control vanes, PVC

<.C

N

=

Wood or metal

Wood with canvas Wood Painted canvas Bronze; galvanized mild steel Rubber

Metal Copper, steel, or light alloy Wood or metal Brass Metal casing with glass lens

Cockpit canopy

Compass binnacles Deck covering

Deck fittings Fenders

Fuel piping Fuel tanks

Helmsman's seat Hinges

Navigation lamps

Traditional Material

Cabin interiors and internal fitments

Type of Fitting

Polyester/glass Nylon; occasionally polypropylene One-piece acrylic molding

Nylon Polyester/glass

Polyester/glass with PVC coated nylon curtains Polyester/glass PVC coated fabric; nylon sheathing Nylon; acetal PVC

Polyester/glass

Plastics Material

Cheaper; improved styling possible; no maintenance needed

No tarnishing or danger of plating peeling off More expensive, but longer lasting under normal conditions; remain attractive and easier to clean; no tendency to chalk Cheaper to install and noncorrodible Translucent so that contents are visible; no corrosion, and unaffected by fuels Lighter, cheaper, and no maintenance needed Less expensive and quieter; no polishing required

Much skilled labor required if these items are in conventional materials. Prefabrication in polyester/glass quickens and cheapens fitting out, resulting in a more durable product needing little maintenance Lighter in weight, more durable, and capable of better styling Lighter, more durable, and attractive More expensive, but better looking and more durable

Advantages of Plastics over Traditional Materials

Table 4-18. Examples of Typical Yacht Fittings in Plastics

...

N \C

Ventilators Water piping (cold) Window moldings

Polyurethane foam with PVC covering Acrylic Polyethylene PVC

Polyester/glass Molded acrylic Phenolic/cotton laminate

Wood Teak and metal Rubber, white metal, or lignum vitae Kapok or foamed rubber covered in leather or canvas Bronze or galvanized steel Metal Rubber

Steering column Steering wheel Stem bearings

Upholstery

Polyester or nylon fiber Acrylic; CAB

Cotton Glass

Sails Screens

Natural fibers Bronze or galvanized steel

Brass or galvanized steel

Pumps and bailers Polyester fiber, nylon, polypropylene, etc. Nylon or acetal

Nylon or phenolic laminate; acetal Polyethylene

Metal

Pulley sheaves

Ropes for rigging, anchor warps, etc. Rowlocks and sockets

Polyester/glass

Metal

Outboard motor shrouds

Cheaper; no maintenance required Cheaper to install; noncorrodible More durable; improved color range

No drumming, improved sound insulation; better styling possible Preserves life of ropes; noncorrodible and improved appearance Neither dents nor scars the boat; will float if accidentally dropped overboard More durable and attractive; easier on hands and most fittings Quiet and less harsh on the oars; cheaper than bronze, but slightly more expensive than galvanized steel More durable and requires less maintenance More easily formed to curvature required by modem design trends Lighter and cheaper; no maintenance required Less expensive Cheaper and generally superior in performance, though tending to wear in shallow, sandy water More durable and wider color range

292 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

...-

~----

-

WOII( WIIH IUffY. IHi 'OI'OISE

~

~--

'I(SSUIE

~...

IIANSfll CA ""!iIXCUUION Oll/U ,

\RIPPS. CANYON

- ... -

SEAlAIII ' a ~ .

~

~

(

..

~

I

CUllEN I INOICAIOI

Figure 4-30. In 1965 an extensive test was conducted by the U.S. Navy's Sealab II to assay man's ability to live and work in ocean depths for long periods of time. For forty-five days three groups of ten men each lived fifteen-day periods in a 57 ft. x 12 ft. habitat at a depth of 188 m (205 ft.) one half mile off La Jolla, Calif. Plastic parts as well as other materials were used to provide a highly successful experiment.

buoy supports, O-ring seals, polyethylene flotation, and watertight electrical connectors using PVC, polyurethane, and DAP. The design and the materials used are the keys to the most efficient faired and unfaired cables for towing, monitoring, and controlling instruments in the deepest water. Tests show that faired shapes rather than conventional bare wire are required in order to reach greater depths with a given length of cable, to obtain higher speeds while maintaining a certain depth, to have the dead-weight instrument remain nearer a point directly below the stem of the ship, and reduce vibration, prolong cable life, and maintain the orientation of instruments when a ship rolls and pitches. As towed instruments at the end of a line are at the mercy of hydrodynamic and gravitational forces controlling their motion through the water, the towline requires more attention than the instruments. Reducing cable drag is accomplished by a shape that usually has a concave curved side facing into the stream. The curvature is greatest next to the instrument and diminishes rapidly going up the line. This shape permits towed lines to follow a short, downward-curving arc appended to a straight line. Plastics such as polyethylene, polypropylene, and polyurethane are used to develop the shapes and provide different combinations of desirable characteristics such as ease of wrapping around standard winches, resistance to the water environment and abrasion, good electrical insulation in wire-conducting cables, and ease of fabrication and repair.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 293

The materials studied for deep-submergence hulls are generally limited to steel (Hy 170), aluminum, titanium, reinforced plastics and composites, and glass (see Figs. 4-30 to 4-3~). The submergence materials show the variation of the collapse depth of spherical hulls with the weight displacement of these materials. All these materials initially would permit building the hull of a rescue vehicle operating at 1,800 m (6,000 ft.) with a collapse depth of 2,700 m (9,000 ft.). It would appear that the most practical approach is to use steel or titanium. However, steel alone was considered, since titanium's susceptibility to stress-corrosion failure at high stress levels would not make it safe. For a search vehicle operating at 6,000 m (20,000 ft.) with collapse depth of 9,000 m (30,000 ft.), the only materials that appear suitable are solid glass and RP. No metals can be used, because they potentially do not have sufficient strength-to-weight values. One of the drawbacks to using glass in hulls is its lack of toughness. Another serious problem is the difficulty in designing penetrations and hatches in a glass hull. A solution to these problems could be a filament winding around the glass or using a tough plastic skin. These glass problems show that the RP hull is very attractive on a weight-displace-

5,000 10,000

...

IS,OOO ::: z: 20,000 ;;

-MEAN DEPTH

......

----- ~IW:UM

25,000

ATLANTIC ____ I----t-;;;i;----tt---t---t----+---t---t---t---+MARIANAS TRENCH ----- L_~.J......._..J.__.J..__..J__~_~_ __I__....J.._ PlCIFIC 20 30 40 50 60 70 80 o10 PERCENT Of OCEAN lESS THAN INDICATED DEPTH

~

30,000 35,000

_.l._

90

100

Figure 4-31. The depth limitations of various hull materials in near-perfect spheres, superimposed on the familiar distribution curve of ocean depths, are here summarized. To place the materials in their proper perspective, the common factor relating their strength-to-weight characteristics to a geometric configuration for a specified design depth is the ratio showing the weight of the pressure hull to the weight of the seawater displaced by the submerged hull, a factor referred to as the weight displacement (WID) ratio. The portions of the bars above the depth-distribution curve correspond to hulls having a 0.5 WID ratio, the portion beneath showing the depth attainable by heavier hulls with a 0.7 WID. The ratio of 0.5 and 0.7 is not arbitrary, as it may appear, for small vehicles can normally be designed with WID ratios of 0.5 or less, and vehicle displllCC(ments can become quite large as their WID ratios approach 0.7. Using these values permits making meaningful comparisons of the depth potential of various hull materials. An examination of the data reveals that for all the metallic pressure-hull materials taken into consideration, the best results would permit operation to a depth of about 18,288 m (20,000 ft.) only at the expense of increased displacement. The nonmetallic materials of reinforced plastics (those with just glass-fiber-TS polyester) and glass would permit operation to 20,000 ft. or more with minimum-displacement vehicles.

0 STEEL

....

w w

10,000

u. X

lll.

w 20,000

0

w

K?


8

30,000

40,000

0 .5

02

WID Figure 4-32. Examples of materials for deep-submergence spherical vehicles. Depths are shown against the WID (weight of vehicle divided by the weight of its seawater displacement).

~

45

l-

X

(!) w40~---+~~O+~~~~---1-­

~

VEHICLE CHARACTERISTICS ' PRESSURE HU LL VOLUME 382 CU FT /9 FT SPHERE) OPERATING DEPTH 20.000 FT MAXIMUM SPEED 5 KNOTS ENDURANCE 30 HOURS AT 3 KNOTS

~

D '. '

25~---+----+----1--~~----1

200~2--~0~3~~~0~4--~0~5~--~0~6--~0~7

D

HY1 4 0

STEEL HY 220 HYI05 TITANIUM HY· 140 NONMETALUCS nBER REINFOR CED PLASTICS. GLASS . CERAMICS

WID RATIO FOR FLOATATION SYSTEM

Figure 4-33. The effects of a pressure-hull and flotation system on vehicle weight. 294

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 295

ment ratio, strength-weight ratio, and for its fabrication capability. A significant advantage of RP over solid glass is that it is available today and the technology of fabricating large, thick-wall structures already exists. Also, with an increased modulus of elasticity in new reinforcements additional gains can be obtained beyond what is presently available in conventional RPs. RPs have already been used in different structural applications, to replace conventional metal in seawater-compressed air surfacing ballast tanks in the Alvin depth vehicle. This vehicle, a first-generation deep research vehicle, also used RP in its outer hull construction to enclose the pressure tanks and aluminum frame. In the unmanned acoustical research vehicle of the Ordnance Research Laboratory called Divar, an RP cylinder with a 16 in. OD, 3/4 in. wall thickness, 12 112 in. ID with nine ribs, a 60 in. length and weight of 180 pounds went to depths of 950 m (6,500 ft.). In addition to developing solid RP structures, work has been conducted on composite sandwich structures such as filament-wound plastic skins with low-density foamed core or a plastic honeycomb core to develop more efficient strength-to-weight structures. Sandwich structures using a syntactic core have been successfully tested so that failures occurred at prescribed high-hydrostatic pressures of 28 MPa (4,000 psi). The design of a hull is a very complex problem. Under varying submergence depths there can be significant working of the hull structure, resulting in movement of the attached piping and foundations. These deflections, however slight, set up high stresses in the attached members. Hence, the extent of such strain loads must be considered in designing attached components. Buoyancy in some form is employed in nearly all categories of underwater and surface systems to support them above the ocean bottom or to minimize their submerged weight. The buoyant material can assume many different structural forms utilizing a wide variety of densities, as shown in Table 4-19. The choice of materials is severely restricted by operational requirements, since different environmental conditions exist. For example, lighter, buoyant liquids can be more volatile than heavier liquids. This factor can have a deleterious effect on a steel structure by accelerating stress corrosion or increasing permeability in reinforced plastics. The typical syntactic foam used for buoyancy in many vehicles is made of hollow glass, ceramic, or plastic microspheres of 30 to 300 micron size, uniformly dispersed in a resin such as epoxy. The navy, desiring to develop a material to replace the more conventional gasoline flotation one, produced an excellent syntactic foam. Strict processing and quality control in producing the foam can develop a static hydrostatic pressure of 10,000 psig and fatigue testing of 1,000 cycles. In the Woods Hole Oceanographic Institute's three-man 1,800 m (6,000-foot) depth vehicles, approximately 5,000 lb. of syntactic foam are used to provide buoyancy. With a specific gravity of 0.68, it requires three pounds of material to gain one pound of buoyant effect. However, its main attributes are that of being able to tailor it to fit the available space and being useful to at least a 5,000 psi load. Cavitation Erosion With increasing ship speeds, the development of high-speed hydraulic equipment, and the variety of modem fluid-flow applications to which metal materials are being subjected, the problem of cavitation erosion becomes ever more important. Erosion may occur in either internal-flow systems, such as piping, pumps, and turbines, or in external ones like ships' propellers.

~

N

Liquids Hydrocarbon Ammonia Alcohol Solids Polypropylene Lithium Ice Cellular Wood Syntactic foam Plastic foam Rigid shell Steel Aluminum Titanium Reinforced plastic Glass

Type

20-80 9 7-30 20-100 30-120 50-1,500

(56-57) (33) (57-57.4) (12-44) (37-50) (6.2-50) (9-50) (9-50) (16-50) (9-37) (9-25)

.90-.92 .53 .91-.92 .20-.70 .60-.80 .10-.80 .15-.80 .15-.80 .25-.80 .15-.60 .15-.40

40-120 44-60 54-85

(2.0-10.1) (3.0-12.2) (5.1-150)

(2.0-8.1) (0.91) (0.7-3.0)

(4.0-12) (4.4--6.1) (5.5-8.6)

Compressibilities 10-6 atm- 1 (Pa)

(41-53) (37-56) (50-56)

Density (lbS./ft.3)

.65-.85 .60-.90 .80-.90

glcc

0-5,500 0-8,000 5,000-10 ,000 0-25,000 0-25,000

0-700 500-9,000 0-2,000

Surface

0-38,000 2,000-38,000

400-38,000 I ,000-38,000

I ,000-38,000

(0-1,700) (0-2,400) (1500-3,000) (0-7,600) (0-7,600)

(0-210) (150-2,700) (0-160)

(0-11 ,600) (610-11 ,600)

(305-11,600) (120-11 ,6(0) (305-11,600)

Useful Operating Depths, ft. (m)

Table 4-19. Examples of Buoyant Materials

10-50 10-55 10-45 20-50 20-50

20-48 6-20 12-40

7-8 17-25 7-8

10-20 5-18 7-15

(160-800) (160-880) (160-720) (320-800) (320-800)

(320-770) (96-320) (190-640)

(110-130) (270-400) (110-130)

(160-320) (80-290) (110-240)

Net Buoyancy Ibs./cu. ft. (kg/m3)

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 297

The phenomenon of cavitation was identified as early as 1873, by Osborne Reynolds. By the tum of the century it had been called by its present name by R. E. Froude, the director of the British Admiralty Ship Model Testing Laboratories. Cavitation occurs in a rapidly moving fluid when there is a decrease in pressure in the fluid below its vapor pressure and the presence of such nucleating sources as minute foreign particles or definite gas bubbles. As a result, a vapor bubble forms that continues to grow until it reaches a region of pressure higher than its own vapor pressure, when it collapses. When these bubbles collapse near a boundary, the high-intensity shock waves that are produced radiate to the boundary, resulting in mechanical damage to the material. The force of the shock wave or of the impinging may still be sufficient to cause a plastic flow or fatigue failure in a material after a number of cycles, depending on the properties of the material, the existing hydrodynamic conditions, and the foil-design parameters. The behavior of materials, particularly steel, in cavitating fluids results in an erosion mechanism, including mechanical erosion and electrochemical corrosion. The straightforward way to fight cavitation is to use hardened materials, chromium, chrome-nickel compounds, or plastics. Other cures are to reduce the vapor pressure with additives, reduce the turbulence, change the liquid's temperature, or add air to act as a cushion for the collapsing bubbles.

THE SPACE ENVIRONMENT The space environment, seen as beginning in the center of the earth, extends to infinity. In the past few decades outer space has been penetrated. These initial successful steps depended on a number of factors, one of which was the use of plastics. As in terrestrial uses, plastics have their place in space. Plastics will continue to be required in space applications from rockets to vehicles for landing on other planets. The space structures, reentry vehicles, and equipment such as antennas, sensors, and an astronaut's personal communication equipment that must operate outside the confines of a spaceship will encounter bizarre environments (see Table 4-20). Temperature extremes, thermal stresses, micrometeorites, and solar radiation are sample conditions that will be encountered. Perhaps the most striking phenomenon encountered in outer space is the wide variation in temperature that can be experienced on spacecraft surfaces and externally located equipment. Temperatures and temperature gradients not ordinarily encountered in the operation of ground or airborne structures and equipment are ambient conditions for spacecraft equipment. On such hardware, not suitably protected externally or housed deep within the space vehicle in a controlled environment, these temperature extremes can wreak destruction. Designers of earthbound electronics must fight temperatures that will produce system degradation, but spacecraft electronic designers may be fighting temperatures that will cause their equipment to melt. On both ends of the temperature scale, the ground- or airborne-equipment designer has a simpler environment to contend with. In addition, the space designer has a temperature paradox to consider. A black box cannot simply be placed in a superinsulated enclosure anymore than a human being can. All other factors aside, both would rapidly destroy themselves, because of self-generated heat. The equipment must therefore be exposed to its environment in some manner, but it also needs a great deal of protection. The problem is not as simple as putting on or taking off a sweater, depending on whether the temperature is 21 or 70°F. The problem is to put something on and keep it on, regardless of whether

I»

..c

N

·Stagnation point conditions.

Maximum heating rate, * Btufft. 2/sec. (MW/m2) Heating time, sec. Maximum dynamic pressure, * atm (MPa)

(MJ/m2)

Vehicle parameter, WICDA, psf (MPa) Entry flight velocity, fps (m/s) Entry angle, degrees Total heating rate, * Btu/ft. 2

Entry Vehicles

30 40 (4.05)

25 10 (1.01)

23,000 (7,010) -15 35,000 (398)

16,000 (4,880) -20 8,000 (90.9) 2,000 (22.7)

1,200 (8.27)

800 (5.52)

500 (5.67)

5000-mile Ballistic Entry Nose Cone

1500-mile Ballistic Entry Nose Cone

300 I (0.10)

70 (0.79)

24,000 (7,310) -3 14,000 (159)

75 (0.52)

Earth Orbital Ballistic Entry Capsule

450 (5.1) 1,500 1 (0.10)

6,000 0.1 (0.01)

36,000 (10,970) -6 140,000 (1,590)

30 (0.21)

Lunar Vehicle Earth Entry

120 (1.36)

24,000 (7,310) -2 150,000 (1,704)

250 (1.72)

Lifting Orbital Entry Glider

Table 4-20. Typical Environments of Atmospheric Entry Vehicles

45 600 (60.8)

100 (1.13)

44,000 (13,410) -90 4,000 (45.4)

100 (0.690)

Mars Planetary Entry Probe

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 299

the temperature is - 250 or + 250°F. Many factors give rise to the temperature extremes encountered. However, these extremes must be understood before any consideration can be given to a means of alleviating them.

Ablation The most common design approach for handling intense heating and extremely high temperatures is ablation. In this process surface material is physically removed or a temperature-sensitive component of a composite is preferentially removed. The injected vapors alter the chemical composition, transport properties, and temperature profile of the boundary layer, thus reducing heat transfer to the material's surface. At high ablation rates the heat transfer to the surface may be only 15 percent of the thermal flux to a nonablating surface. Tens of thousands of Btu's of heat can be absorbed, dissipated, and blocked per pound of ablative material through the sensible heat capacity, chemical reactions, phase changes, surface radiation, and boundary layer cooling of the ablator (see Fig. 4-34). Ablative systems are not limited by the heating rate or environmental temperature, but rather by the total heat load. In spite of this limitation, however, the versatility of ablation has permitted it to be used on almost every recent hypervelocity atmospheric vehicle. Moreover, it appears that ablation will continue to be favored as the primary thermal protection method for future flight vehicles. No single universally acceptable ablative material has been developed, nor is one likely to be created. Nevertheless, the interdisciplinary efforts of materials scientists and engineers have resulted in obtaining a wide variety of ablative compositions and constructions. These thermally protective materials have been arbitrarily categorized by their matrix composition. Typical materials are given in Table 4-21.

ENERGY EXCHANGES

Glass Droplets Dense Char Nascent Porous Char Resin Volatilization

CONVECTION

':'.=....::...!....:-~.o.: 10-'"

RADIATION

,

GAS-PHASE COMBUSTION SURFACE COMBUSTION RERADIATION

} of

f TRANSPIRATION COOLING CHEMICAL REACTIONS + f - - -

PHYS ICAl CHANGES

~f---

~=

" . .~ .... -o.,.. \,D~~~1'

-~ ::::;t:-~r:~ :~~Ci'o-:-;:..;d•• t-----tl\..

Figure 4-34. The surface-heat balance of an ablating glass-fIber-reinforced phenolic resin composite.

300 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 4-21. Typical Ablative Compositions Plastic Based

Ceramic Based

Elastomer Based

Metal Based

Polytetrafluoroethylene

Porous oxide (silica) matrix infiltrated with phenolic resin

Silicone rubber filled with microspheres and reinforced with a plastic honeycomb

Epoxy-polyamide resin with a powdered oxide filler

Porous filamentwound composite of oxide fibers and an inorganic adhesive impregnated with an organic resin Hot pressed oxide, carbide, or nitride in a metal honeycomb

Polybutadieneacrylonitrile elastomer modified phenolic resin with a subliming powder

Porous refractory (tungsten) infiltrated with a low melting point metal (silver) Hot-pressed refractory metal containing an oxide filler

Phenolic resin with an organic (nylon), inorganic (silica), or refractory (carbon) reinforcement Precharred epoxy impregnated with a noncharring resin

Plastic-based composites, which employ an organic matrix, are the most widely used class of ablative heat-protective materials. They respond to a hyperthermal environment in a variety of ways, such as depolymerization-vaporization (polytetrafluoroethylene), pyrolysis-vaporization (phenolics, epoxy resins), and decomposition-melting-vaporization (nylon fiber-reinforced plastic). The principal advantages of plastic-based ablators are their high heat-shielding capability and low thermal conductivity. Their major limitations are high erosion rates during exposure to high gasdynamic shear forces and a limited capability to accommodate high heat loads.

Rain Erosion One who walks through a gentle spring rain seldom considers that raindrops can be small destructive "bullets" when they strike high-speed aircraft. These bulletlike raindrops can erode paint coatings, plastic parts, and even magnesium or aluminum leading edges to such an extent that the surfaces may appear to have been sandblasted. Even the structural integrity of the aircraft may be affected after several hours of flight through rain. This problem is of special interest to aircraft engaged in all-weather flying. It affects commercial aircraft, missiles, high-speed vehicles on the ground, spacecraft before and after a flight when rain is encountered, and even buildings or structures that undergo high-speed rainstorms. The critical situations exist in flight vehicles, since flight performance can be affected to the extent that a vehicle can be destroyed. Research and development concerning rain erosion on aircraft has been extensive.

ENVIRONMENTAL CONDITIONS AFFECTING PLASTICS 301

Erosion by rain of the exterior of high-speed aircraft during flight was observed during World War II on all-weather fighter airplanes capable then of flying at 400 mph. The aluminum edges of wings and particularly of the glass-fiber-reinforced TP polyester-nose radomes (particularly the Eagle Wing on B-29s flying over the Pacific) were particularly susceptible to this form of degradation. That the problem continues to exist can be seen in Figure 3-102. Actual flight tests to determine the severity of this phenomenon of rain erosion carried out in 1943 established that aluminum and RP leading edges of airfoil shapes exhibited serious erosion after exposure to rainfall of only moderate intensity. Inasmuch as this problem originally arose with military aircraft, the U. S. Air Force initiated research studies at the Wright-Patterson Development Center's Materials Laboratory in Dayton. It resulted in applying an elastomeric neoprene coating adhesively bonded to RP radomes similar to the coating seen in Figure 3-102. The usual 5-mil coating of elastomeric material used literally bounces off raindrops, even from a supersonic airplane traveling through rain. There is a slight loss of radar transmission (see Figure 3-103) of about 1 percent per mil of thickness, but this is better than losing the radome. The next chapter explores how to analyze and design for the loads and other structural considerations that underlie such external coatings.

Chapter 5

STRUCTURAL DESIGN ANALYSIS

Part design incorporates the factors pertaining to functional and appearance requirements, the properties of the material, whether it be plastic, steel, wood, or of something else, and the feasibility of processing in arriving at an acceptable solution in terms of performance and cost. As mentioned at the outset of this book, the designer has choices that sometimes require compromising the product requirements with processing. The choice made will determine to a large extent the types of problems to be solved regarding not only the part's configuration but also the mold or die design, which must likewise be considered in arriving at a useful, economically acceptable part. In the design or analysis of mechanical components, a systematic approach is desirable. Frequently, the product will be one in which there are insignificant loads and no limitations on deflection. In such cases the experience or practical approach of the designer is usually all that is required. This is especially true with small, load-free plastic parts where the processing requirements dictate a minimum wall thickness that is more than adequate for the part's function. Still, even in these designs those new to plastics often neglect the effects of stresses caused by temperature and other environments as discussed in the last two chapters, processing, assembling, handling, decorating, and shipping (see Chapters 7 and 8). In this chapter, simple analysis techniques are presented that will assist the designer in developing new products to handle the anticipated loading, while keeping stress and deflection within acceptable limits. These techniques will also be useful in product improvement, cost reduction, and the failure analysis of existing parts. The application of simplified, classic stress and deflection equations to plastic parts are presented here. As the complexity of a part increases or when particularly accurate results are required, more exact traditional methods or computerized finite element analysis (PEA) may be required

[1,2, 7-14, 33, 40-45, 62-76, 93, 270, 278, 390-417].

LOAD-BEARING PRODUCTS A fundamental concept in structural analysis is that the structure as a whole and each of its elements together are in a state of equilibrium. This means that there are no unbalanced forces of tension, compression, flex, or shear acting on the structure or a part at any point. All the forces counteract one another, which results in eqUilibrium. When all the forces acting on a given element in the same direction are summed up algebraically, the net effect is zero, with no acceleration. The object does respond to the various forces 303

304 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

internally, however. It is pushed or pulled and otherwise deformed, with internal stresses of varying types and magnitudes accompanying these deformations. Basically, designing a load-bearing product with any material involves first selecting a suitable material and then specifying the shape into which it is to be formed or assembled. One important aspect of shape is its effect on internal stress. As the cross-sectional area of a part increases for a given load, the stresses are reduced. Design is concerned with determining the stresses for a given or hypothetical shape and subsequently adjusting the shape until the stresses are neither high enough to risk fracture nor low enough to suggest that material is being wasted. Stress analysis involves using the descriptions of parts' geometry, the applied loads and displacements, and the materials' properties to obtain closed-form or numerical expressions for internal stresses as a function of the stress's position within the part and perhaps as a function of time as well. The term engineering formulas refers primarily to those equations seen previously and given in engineering handbooks by which this stress analysis can be accomplished [33, 62-68, 76, 85, 93, 268-70, 278, 399, 402-4].

The Pseudo-Elastic Design Method Throughout this book as the viscoelastic behavior of plastics has, been described it has been shown that deformations are dependent on such factors as the time under load and the temperature. Therefore, when structural components are to be designed using plastics it must be remembered that the standard equations that are available for designing springs, beams, plates, and cylinders, and so on have all been derived under the assumptions that 1) the strains are small, 2) the modulus is constant, 3) the strains are independent of the loading rate or history and are immediately reversible, 4) the material is isotropic, and 5) the material behaves in the same way in tension and compression. Since these assumptions are not always justifiable when applied to plastics, the classic equations cannot be used indiscriminately. Each case must be considered on its merits, with account being taken of such factors as the mode of deformation, the service temperature, the fabrication method, the environment, and others. In particular, it should be noted that the traditional equations are derived using the relationship that stress equals modulus times strain, where the modulus is a constant. From the discussion in Chapter 3 it should be clear that the modulus of a plastic is generally not a constant. Several approaches have been used to allow for this condition, some of which are quite accurate. The drawback is that these methods can be quite complex, involving numerical techniques that are not attractive to designers. However, one method has been widely accepted, the so-called pseudo-elastic design method. In this method appropriate values of such time-dependent properties as the modulus are selected and substituted into the standard equations. It has been found that this approach is sufficiently accurate in most cases if the value chosen for the modulus takes into account the projected service life of the product and the limiting strain of the plastic, assuming that the limiting strain for the material is known. Unfortunately, this is not just a straightforward value applicable to all plastics or even to one plastic in all its applications. This value is often arbitrarily chosen, although several methods have been suggested for arriving at a suitable value. One is to plot a secant modulus that is 0.85 of the initial tangent modulus and note the strain at which this intersects the stress-strain characteristic. However, for many plastics, particularly the crystalline TPs, this method is too restrictive, so in most practical situations the limiting strain is decided in consultation between the

STRUCTURAL DESIGN ANALYSIS 305

LIFTING LOAD

t

PRESSURE IN A PIPE

o o 000

C=:J PORTABLE TELEVISION HANDLE

Figure 5-1. Directly applied loads.

designer and the plastic material's manufacturer. Once the limiting strain is known, design methods based on its creep curves become rather straightforward. LOADS

In a simplified approach the first step in analyzing any part is to determine the loads to which it will be subjected. These loads will generally fall into one of two categories, directly applied loads and strain-induced loads [2]. Directly Applied Loads

Directly applied loads are usually easy to understand. They are defined loads that are applied to defined areas of the part, whether they are concentrated at a point, line, or boundary or distributed over an area. The magnitude and direction of these loads are known or can easily be determined from the service conditions. In larger plastic parts the weight of the part itself will not present a significant load. Figure 5-1 shows examples of directly applied loads [2]. Strain-Induced Loads

Frequently, a part becomes loaded when it is subjected to a defined deflection. The actual load then is a result of the structural reaction of the part to the applied strain. Unlike directly applied loads, strain-induced loads are dependent on the modulus of elasticity and, with TPs, will generally decrease in magnitude over time. Many assembly and thermal stresses are the result of strain-induced loads. Figure 5-2 shows two common examples [2]. SUPPORT CONDITIONS

When a load is applied to a part, if the part is to remain in eqUilibrium there must be an equal force acting in the opposite direction. These balancing forces are the reactions at the supports. For purposes of structural analysis there are several support conditions that

306 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

0..--.

METAL PIN

. .- - PLASTIC BOSS

PRESS FIT PRODUCES FIXED HOOD STRAIN IN BOSS. METAL SCREW

PLASTIC HOUSING

1 SCREW TOROUE PRODUCES COMPRESSIVE STRAIN IN HOUSING

Figure 5-2. Strain-induced loads.

have been defined (see Fig. 5-3). The free (unsupported), simply supported, and fixed supports are the most frequently encountered [2].

Free (Unsupported) This support condition occurs where the edge of a body is totally free to translate or rotate in any direction.

Guided This condition is similar to the free end except that its edge is prevented from rotating.

Simply Supported In this support condition transverse displacement in one direction is restricted, as illustrated.

SIMPLY SUPPORTED

HELD (PINNED)

FIXED

Figure 5-3. Support conditions.

307

308 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Held (Pinned) This situation is similar to the simply supported one except that here only rotations are allowed.

Fixed (Clamped or Built-in) This support condition at the end of a beam or plate prevents transverse displacement and rotation. The condition can be thought of as an end support firmly embedded into a fixed solid wall. In practice this condition rarely exists in its pure form, especially with plastic parts, since the mounting points of the parts usually have some give.

SIMPLIFICATIONS AND ASSUMPTIONS For purposes of discussion, in this chapter the following simplifications and assumptions are made regarding simple tension, compression, and flexural and shear performance: 1) the part under load can be broken down into one or more simple structures, beams, plates, pressure vessels, and other segments for analysis; 2) the material being analyzed may be considered to be linearly elastic, homogeneous, and isotropic; while this is not necessarily always true for plastics, this assumption is fundamental to the equations that follow; 3) the equations assume that the load is a single concentrated or distributed static load that is gradually applied for a short period and then removed; however, creep, relaxation, or fatigue loads may be analyzed with the same equations, provided the appropriate modulus and rupture (strength) conditions are applied, 4) the part being analyzed has no residual or molded-in stresses, 5) the equations apply to regions that are remote from the point of application of the load and from any shoulder, hole, or other sudden change in dimension of the structure; and 6) the equations may be used at shoulders, holes, or other sudden changes in dimensions, as long as appropriate stress concentration factors are used.

MULT1AXIAL STRESSES AND MOHR'S CIRCLE Sophisticated design engineers unfamiliar with plastics' behavior will be able to apply the information contained in this and other chapters to applicable sophisticated equations that involve such analysis as mUltiple and complex stress concentrations. The various machine-design texts and mechanical engineering handbooks previously discussed and listed in the References section at the end of this book provide detailed analysis of these stress-concentration factors and other load-bearing parameters. Many structural parts are stressed in a manner that is more complex than simple tension, compression, flex, and shear. Because yielding will also occur under complex stress conditions, a yield criterion must be specified that will apply in all stress states. Any complex stress state can be resolved into three normal components acting along three mutually perpendicular axes and into three shear components along the three planes of those axes. Then, by making a proper choice it is possible to find a set of three axes along which the shear stresses will be zero. These are the principal axes, with the normal stresses along them being called the principal stresses. Determining these principal stresses in a complexly loaded member is the responsibility of the designer, a task normally performed by using Mohr's circle and its associated relationships [42, 404].

STRUCTURAL DESIGN ANALYSIS 309

SAFETY FACTORS In order to take uncertainties into account in a product's design, engineers have introduced what is familiarly called the safety factory (SF) or sometimes the "factor of ignorance." Many designers have already used or calculated a safety factor on material, perhaps without recognizing it. For example, dividing yield-point stress by calculated stress results in a safety factor. This process appears to be simple and straightforward, but unfortunately things are never quite this simple. The designer must be fully aware of what one means when one calculates such a factor or bases a design on it. Improper use of a presumed safety factor may in some cases result in a needless waste of material or in other cases even physical or operational failure. Thus, one must define what is meant when using a safety factor. Designers unfamiliar with plastic products can use the suggested safety factor guidelines in Table 5-1. Any product designed with these guidelines in mind should conduct tests on the products themselves to relate the guidelines to actual performance. With more experience, more-appropriate values will be developed. It is important to remember that the process of materials selection can be only as good as the information on which the selection is based. There are no hard-and-fast rules to follow in setting safety factors for any given material unless experience is gained in it. The most important consideration is of course the probable consequences of failure. For example, a little extra deflection in an outside wall or a hairline crack in one of six internal screw bosses might not cause concern, but the failure of a pressure vessel or aircraft wing might have serious safety or product-liability implications. Before putting any product onto the market, tests should be run on its actual parts at their most extreme operating conditions. For instance, the maximum working load should be applied at the maximum temperature and in the presence of any chemicals that might be encountered in the end use. Furthermore, the loads, temperatures, and chemicals to which a product will be exposed prior to reaching its end use must not be overlooked. Impact loading should be applied at the lowest temperature expected, including that which occurs during shipping and assembly. The effects of variations in resin lots and molding conditions must also be considered. The results should be to provide more logical safety factors pertinent to the product and the materials used in it. Many situations discovered during the testing of preproduction parts can be corrected

Table 5-1. Safety Factors: Preliminary Design Guidelines for Materials, Based on Design Requirementsa Type of Load

Factor-

Static short-tenn loads Static long-tenn loads Variable of changing loads Repeated loads Fatigue or load reversal Impact loads

2-4 4-10 4-10 5-15 5-15 10--20

"The material strength detennined is Ibe minimum required, not !he average or maximum. which is what is normally provided on manufacturers' published data sheets. "Low-range values represent situations where failure is not critical; !he higher values are for where failure is critical. Note: These values are intended for preliminary design analysis only and are not to be used in place of Ihorough product design.

310 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

with a selective use of increased thickness in walls, ribs, and gussets or by eliminating stress concentrations. Changing a material to another grade of the same resin or to a different plastic with a suitable mechanical property profile might also be the solution.

Uncertainties In addition to potential basic graphic design uncertainties, a designer has to consider additional uncertainties that may exist with any material, some of which follow: Material property variations. Material variation is an important factor often overlooked by a designer or analyst when evaluating a particular mechanical component. Because no two plastics are exactly alike and some materials, particularly the RPs, may have inclusions and so on, the short- and long-term strength properties given in materials tables are usually, or even maximum, average values. Effect of size in stating material strength properties. Unless otherwise stated, property tables list strength values based upon a specified size, such as 3.75 cm (! in.). Yet larger components generally fail at lower stress than a similar smaller component made of the same material. Type of loading. A simple static load is relatively easy to recognize, but some cases fall between impact loads and suddenly applied loads or infrequently applied fatigue loading mixed with shock loads, as for example cams, links, or feeding devices. Processing variations. Production operations may, and in fact usually do, introduce stress concentrations and residual stresses (see Chapter 7). Overall concern for human safety. All designs must consider the safety of the user who may be near or in contact with the product. Unexpected sudden overloads should not cause breakage or bodily harm.

Composites It can generally be claimed that fiber based composite materials offer good potential for

achieving high structural efficiency coupled with a weight saving in products, fuel efficiency in manufacturing, and cost effectiveness during service life. Conversely, special problems can arise from the use of composites, due to the extreme anisotropy of some of them, the fact that the strength of their constituent fibers is intrinsically variable, and because the test methods for measuring composites' performance need special consideration if they are to provide meaningful values. Some of the advantages, in terms of high strength-to-weight ratios and high stiffnessto-weight ratios, can be seen in Figure 5-4, which shows that some composites can outperform steel and aluminum in their ordinary forms. However, it should be remembered that these values do not include acceptable safety factors, which may require a value of three. If bonding to the matrix is good, then fibers augment mechanical strength by accepting strain transferred from the matrix, which otherwise would break. This occurs until catastrophic debonding occurs. Particularly effective here are combinations of fibers with polymer matrices, which often complement one another's properties, yielding products with acceptable toughness, reduced thermal expansion, low ductility, and a high modulus. Apart from these combinations, most composites are expected to combine the good properties of two or more materials more cheaply than could be achieved with any single material. Composites offer the option of achieving maximum strength in predetermined

STRUCTURAL DESIGN ANALYSIS 311

Stress

Epoxy resin with 30% glass ~

_ _- - Epoxy resin

Strain

Figure 5-4. Examples of the tensile properties of some composites: epoxy with different reinforcing fibers, steel, and aluminum.

axes by aligning the fibers or by using two-dimensional woven cloth reinforcement. On the other hand, particulate solid fillers in polymeric matrices tend to yield isotropic properties in composites, which is useful, because filler particles act as crack propagation stoppers, thus increasing strength under adverse environments. However, at the high rates of strain that occur upon impact, solid fillers in general do not increase strength, although the rubbery particles incorporated on a molecular scale in acrylonitrile-butadiene-styrene polymers do not have this useful property. As a further advantage, composites make effective use of some materials that are otherwise unable to stand alone, such as mineral fibers or wood flour. When incorporated into polymers-in particular those such as unsaturated polyesters or phenolics-particles can reduce manufacturing shrinkage and yield a more usable product. In service, zero thermal expansion coefficients can be achieved by a suitable choice of starting materials.

Design Allowables In analyzing design allowables mathematically, a statistical analysis can be produced that is based on advanced composite design analysis [405]. Such an analysis identifies four kinds of design allowables, called A basis, B basis, S basis, and Typical basis. Of these, the A and B basis design allow abies are used the most frequently. They are associated with statistical assurance and based on normal distribution represented as Design allowable

=X

x Ks

where X = average values of n tests/observations, K = a constant depending upon the number of specimens, and s = standard deviation [11]. For a basis, K = 3 for 34 specimens; for B basis, K = 3 for six specimens. See Figure 5-5 for the A basis and B basis curves. For the design allowables to be maximal the average value of X should be the maximum

312 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

one and should use the standard deviation of s; hence, the coefficient of variation should be the minimum (CV = siX X 1(0). To obtain the maximum value of K, various design, materials, processing, tooling, and quality control parameters have to be optimized. To obtain the minimum coefficient of variation, either the part or the testing process must be reproducible, with minimal deviation and scatter. Note that obtaining a low CV does not automatically mean that the average value X is maximal or optimal. Low values of a CV can be obtained even though a given testing or part-manufacturing technique may not be optimal. It is not uncommon to obtain a low CV while obtaining low to mediocre test values or analyzing medium-quality parts. The design allowables should also consider service-life degradation in properties, including such environmental ones as temperature/humidity cycling as well as fatigue and so on, as shown in Figures 5-6 and 5-7. It should be evident by this point that the selection of an appropriate factor for a product is rather empirical and greatly dependent upon an individual's or industry's accumulated experience. Where a product or device has a long history of use, the factors based on this history are reliable. In fact, one may still depend upon such data even though modifications have been made in design and materials. In some cases the selection of the safety factor will be stipulated by code or contract requirements. Statistical methods have also been employed to establish safety factors. Here account is taken of the variance in the dimensions and strength of a mechanical component as well as the various uncertainties described above. This approach results in having a safety factor that is in general smaller than that which is based purely on judgment. However, the statistical method also requires estimating possible load and strength variations, thereby making the method somewhat less than rational. Nevertheless, the statistical approach should be of more than just passing interest to modem designers, particularly in those areas where experiential data for components have accumulated. Moreover, this method permits using a relatively low safety factor, if a small percentage of failure is acceptable.

BEAM BENDING STRESSES As indicated in Chapter 3, in simple beam-bending theory a number of assumptions must be made, namely that 1) the beam is initially straight, unstressed, and symmetrical; 2) its proportional limit is not exceeded; 3) Young's modulus for the material is the same in both tension and compression; and 4) all deflections are small, so that planar crosssections remain planar before and after bending. The maximum stress occurs at the surface of the beam farthest from the neutral surface, as given by the following equation (see Fig. 5-8) [2]. Me

M

1

Z

a=-=-

where M = the bending moment in in.llbs., e = the distance from the neutral axis to the outer surface where the maximum stress occurs in in., 1 = the moment of inertia in in.4, and Z = lie, the section modulus, in in. 3 Observe that this is a geometric property, not to be confused with the modulus of the material, which is a material property. I, e, Z, and the cross-sectional areas of some common cross-sections are given in Figure 5-8, and the mechanical engineering handbooks provide many more. The maximum

STRUCTURAL DESIGN ANALYSIS 313 1000

8

r 'N BASIS

I

6

4 2

\

(1)100

!ii

8 ~ 6 15 4

ffiell ~

~

\ \

2

J.

99% PROBABILITY O~ SURVlVA(-:-= 95% CONFIDENCE

\

\

\

\.

r\.

10

.....-

8 6

~

.........

...:::::'7B' BASIS

--f---90% PROBABILITY OF SURVIVAL 2 t---9j% CONFIYNCE 4

1.0

2.0

K

3.0

4.0

Figure 5-5. One-side tolerance factors for the nonnal distribution K.

ACCEPTANCE LEVEL

PART REQUIREMENT FOR SERVICE LIFE

TIME, SERVICE LIFE, PROCESSING CONDITIONS

Figure 5-6. Defect propagation in composites during service life.

INITIAL STRENGTH, 'A' BASIS (34 SPECIMENS) SMALL SCATTER OR /

'B' BASIS (6 SPECIMENS)

TIME, SERVICE LIFE, PROCESSING CONDITIONS

Figure 5-7. Degradation in the composite strength property during service life.

stress and defection equations for some common beam-loading and support geometries are given in Figure 5-9 [2]. Note that for these T- and U-shaped sections in Figure 5-8 the distance from the neutral surface is not the same for the top and bottom of the beam. It may occasionally be desirable to determine the maximum stress on the other, nonneutral, surface, particularly if it is in tension. For this reason, Z is provided for these two sections.

TIE RECTANGULAR

~

c

II-!r I-BEAM

A=bd

c=~ 2

~

I=~ 12

I-b-j

c

-s

~b-lT

z=~

A=bd-h(b-I) c=d

2 I = bcJ3 - h3(b - I)

12

z=

6

bcJ3 - h3(b - I) 6d

H-BEAM

CIRCULAR

A=bd-h(b-I)

A=~ 4

c=!!.. 2

c=d 2

I =

2sb"+ht"

12

I=~

z=

64

z=~

2sb"+ht" 6b

32

C-BEAM A=bd.h(b-I)

TUBE

A - n(do' - d!,) 4

c=~

c=~

I = bcJ3 - h3(b - I)

2

12

2

I = n(do4

-

z=

d(4)

64

z=

n(do4 - d(4) 32do

U-BEAM

Tor RIB

T:f[naT

~~b----l

t h

f

I

Lj -ll~

bcJ3 - h"(b - I) 6d

c

= d _

z=~ c z'=

I =

d 21+S2(b - I) 2(bs+hl)

-j~f.-h A=bd- h(b - I) , T c=b 2b s + hl _ b 2A

;;-.nal

I d·c Ie" + b(d • c)3 • (b . IXd· c • s)3 3

2

t

~d~

l

I = 2sb" 3+ hi" -A(b - c)2

z=~ c

Z,=_I_ b·c

Figure 5-8. The properties of some common cross-sections, based on a mechanical engineering analysis (00 = neutral axis), 314

2

SIMPL Y SUPPORTED BEAM CONCENTRA TED LOAD AT CENTER

a=~

(at load)

CANTILEVERED BEAM (ONE END FIXED) CONCENTRA TED LOAD AT FREE END

(at support)

a =~ Z

(at load)

Y =~

4Z

Y=~

(at load)

3EI

48EI

SIMPL Y SUPPORTED BEAM UNIFORML Y DISTRIBUTED LOAD

' 5 T= r-- ----1 I

F (total load)

G!"'''

CANTILEVERED BEAM (ONE END FIXED) UNiFORML Y DISTRIBUTED LOAD

I

L

Y

= FL

(at center)

a

(at center)

Y = 5FL'

8Z

384EI

r- -1T L

(at support)

a = FL 2Z

(at support)

Y=~ 8EI

BOTH ENDS FIXED CONCENTRA TED LOAD AT CENTER

BOTH ENDS FIXED UNIFORML Y DISTRIBUTED LOAD

Y (at supports)

Y

a=~

(at supports)

(at load)

Y=~

(at center)

a=~ 12Z

8Z

192EI

Y=

FL' 384EI

Figure 5-9. Maximum stress and deflection equations for selected beams. 315

316 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

I

.0208

700%

100%

7

Add 1Ja"wx %"HRib

.0048

85%

6.25%

14

AddW'wx %"HRib

.0064

146%

12.5%

12

Add 1Ja"wx W'HRib

,0118

354%

12.5%

28

Add%"wx W'HRib

.0194

646%

25%

26

Change

Shape

1

+---1-

2

f---}-

Double Height

3

£tj3-

4

tt?-

T

5

lj3-

6

Base

Increase In

Ratio

Increase In Weight

Case

Moment Of Inertia

_1_ Wt.

.0026

2" x V4"

Figure 5·10. Examples of ways of using ribs to increase rigidity and reduce weight.

Ribs The moment of inertia, /, can be changed substantially by adding ribs or gussets or some combination of them. As shown in Table 5-2 and Figures 5-1 and 5-11, there is a better way to achieve this result and still keep weight at a minimum by using ribbing, if space exists for it [1, 10-12). The views include sections of equal stiffness. Adding ribs to a part maintains its thin walls and thus allows fast fabricating cycles. It is possible to reduce the cross-sectional area of a part and consequently reduce the amount of material used in it, with a corresponding weight reduction.

Table 5-2. Design Examples to Obtain the Same Part Rigidity for a Section 1 ft. x 2 ft. Property

Steel

Solid Plastic

Structural Foam

Ribbed Solido

Thickness (in.) E (psi) I (in.4) E x I (rigidity) Weight (lbs.)

0.040 3 x 107 0.000064 1,920 3.24

0.182 3.2 X lOS 0.006 1,920 1.98

0.196 2.56 x lOS 0.0075 1,920 1.78

0.125 3.2 x lOS 0.006 1,920 1.60

"Rib height

= 0.270 in., thickness = 0.065 in., rib spacing = 2.0 in.

STRUCTURAL DESIGN ANALYSIS 317

Aluminum E

=

10.3

X

Zinc 106

E

= 2.0

X

I = 0.0049

I = 0.0254

EI = 5.08 x 104

EI

Area

=

0.283 in.

WVin. = 0.446 oz.

= 5.08

Valox 420 plastic 106

E = 1.2

X

106

I = 0.0424 x 104

EI

= 5.08

x 104

Area = 0.489

Area = 0.170

Wtlin. = 2.01 oz.

Wtlin. = 0.149 oz.

Figure 5-11. Different cross-sectional profiles with equivalent stiffnesses in bending, including GE's Valox 420, a TP polyester plastic.

Folded Plates The methods of analysis and design presented for beams and plates may apply also to more-complex products such as folded plate structures, which range from bottles to roofing to outer-space structures. They are basically assemblies of rectangular, triangular, spherical, or other shapes that behave much like beams, portal frames, arches, or shells. The stresses in some folded structures can be determined with acceptable accuracy by applying elementary beam theory to the overall cross-sections of the plate assemblies. When assemblies are plates whose lengths are large relative to their cross-sectional dimensions (i.e., thin-wall beam sections, ribbed panels, and so on) and are in large plates whose fold lines deflect identically, such as the interior bays of roofs, they can be analyzed as beams. More-elaborate procedures must be used to determine the transverse bending stresses in assemblies of large plates and longitudinal stresses in structures with "pinned" connections along folded lines that do not deflect identically. There are also bellows-style collapsible plastic containers such as bottles that are foldable [12, 406]. As shown in Figures 5-12 and 5-13, the technology of foldable containers in contrast to that of the usual "passive" bottles provides advantages and conveniences such as reduced storage, transportation, and disposal space; prolonged

318 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

b.

l\I(I\'ill~

Port in" ...

=..:..::....:.: • h"'H ' Il,pnllg 1\"111

d.

II

I.

01

Figure 5-12. The theory and operation of the patented collapsible bottle. a) An uncollapsed bottle; b) a collapsed bottle; c) top view of the bottle; and d) definitions. product freshness by reducing oxidation and loss of carbon dioxide, and provides continuous surface access to foods like mayonnaise and jams. The bellows of collapsible containers overlap and fold to retain their folded condition without external assistance, thus providing a self-latching feature. This latching is the result of bringing together under pressure two adjacent conical sections of unequal proportions and different angulations to the bottle axis. On a more technical analytical

STRUCTURAL DESIGN ANALYSIS 319

level the latching is created by the swing action of one conical section around a fixed pivot point, from an outer to an inner, resting position. The two symmetrically opposed pivot points and rotating segments keep a near-constant diameter as they travel along the bottle axis. This action explains the bowing action of the smaller, conical section as it approaches the overcentering point (see Figures 5-12 and 5-13). The initial collapsing of such a bottle should occur no later than three to ten hours after manufacture-the sooner the better. Additional pressure is needed for this first-time collapse in order to create permanent fold rings. The subsequent collapsing and expansion of such bottles before filling them can be performed at the recommended ambient temperature of 20°C (68°F) or higher. In most disposable applications these bottles would undergo three changes of volume: 1) an initial collapsing of the container before shipment and storage; 2) expansion of the container at its destination, before or during filling; and 3) finally collapsing the bottle for disposal. The fold rings designed on the bellows for such containers have proven to be durable and sturdy. Prototype bottles made of PETG and 75 durometer PVC were able to withstand dozens of collapses and still pass their stress tests. As shown in Figure 5-13, the two adjacent conical sectors providing the latching should not exceed an angle of 110° to make a sharp fold ring. The size of rotating conical section B should not exceed 80 percent of conical section A, to prevent confusion and wobbling as the bottle is being collapsed. Product labeling can be accomplished by attaching a floating sleeve to the neck or shoulder section of the bottle. This cylindrical sleeve then accommodates the bellows as they fold from the bottom up and contains them within the sleeve as the jar is collapsed. The maximum length of the sleeve is limited by the collapsed dimension of the jar. An extended cap can also be used to hold the label to the side of the bottle.

Shape and Stiffness In most cases, plastic parts can take advantage of a basic beam structure in their design. Much of the conventional designing with other materials is based on single rectangular shapes or box beams, because in timber and steel these are commonly produced as standard shapes. However, their use in plastic components is often accompanied by a wasteful use of material, as in large steel sections. Hollow-channel, 1-, and T-shapes designed with generous radii (and other basic plastic flow considerations) rather than sharp comers are more efficient on a weight basis in plastics because they use less material, thus providing a high moment of inertia. The moments of inertia of such simple sections, and hence their stresses and deflections, can be fairly easily calculated, using simple theories. Such nonrectangular sections are common in many thermoplastics articles. Channels, T-sections, and hollow comer pillars are found in many parts, such as crates and stacking containers, and inverted U-sections and cantilevers are common in items such as streetlamp housings and aircraft. To process any plastics, unreinforced or reinforced, into curved panels is relatively easy and inexpensive. Such panels conform to recognized structural theories that curved shapes can be stiffer in bending than flat shapes of the same weight. Putting it differently, a square-section component built to withstand external pressure will usually be heavier than one of circular section and the same volume. Both single- and double-curvature designs are widely used to make more-effective use of plastics materials.

=

....hl

Figure 5-13. Design ideas for the collapsible bottle. a) A continuous latch bottle; b) a skip-latch bottle; and c) a section through a bellows showing the collapsing and latching mechanism of a collapsible bottle.

a

STRUCTURAL DESIGN ANALYStS 321

BEAM BENDING AND SPRING STRESSES To illustrate how traditional materials such as metals limit the design process, consider a spring. The manufacturing process in metals limits the options available in producing a variety of shapes in this material. As as result, steel springs are produced in only three basic shapes: the torsion bar, the helical coil, and the flat-shaped leaf spring. By comparison, TPs and TSs can be easily fabricated into a variety of shapes. Switching from metal to plastic thus lets the designer overcome configuration barriers to new spring designs. Figure 5-14 is an example of a TP spring action with a different shape [407]. Composite leaf springs constructed of unidirectional fiber-reinforced plastics have come to be recognized as viable replacements for steel springs in truck and automotive suspension applications and have been used in aircraft landing systems since the early 194Os. Because of the material's high specific strain energy storage capability as compared to steel, a direct replacement of multileaf steel springs by monoleaf composite springs can be justified on a weight-saving basis. Such springs have in fact been in use since the 1960s. Further advantages of composite springs accrue from the ability with them to design and fabricate spring leaves having continuously variable widths and thicknesses along their lengths. Such design features will no doubt lead to new suspension arrangements in which composite leaf springs will serve multiple functions, thereby providing a consolidation of parts and simplification of suspension systems. One distinction between steel and plastic is that complete knowledge of shear stresses is not important in a steel part undergoing flexure, whereas with RP design shear stresses, rather than normal stress components, usually control the design. Procedures have thus been developed for evaluating design stresses because of simple flexure as well as secondary loads like axle windup [1].

Figure 5-14. An injection-molded Du Pont Delrin acetal plastic stapler illustrating a type of spring design with the body and curved spring section molded in a single part. This complex shape could not have been achieved in a single operation in steel. The designer has taken advantage of molding'S versatility to reinforce the curved, frequently stressed back section. When the stapler is depressed, the outer curved shape is in tension and the ribbed center section is put into compression. When the pressure is released, the tension and compression forces are in tum released and the molded part returns to its original position. With this type of plastic having these inherently desirable properties, this repeated spring action has a virtually unlimited life span.

322 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~

T y

\\

Figure 5·15. A zigzag configured multiple-cantilever beam spring.

Recent developments with RP leaf springs have highlighted the need to reassess standards for testing and evaluating them. Because of the anisotropic properties of composites, the standards previously developed for steel components in the laboratory and on the proving ground can give misleading test results in plastics. The concept of spring design has been well documented in various SAE and STP design manuals from the 1970s on. These also give the equations for evaluating design parameters, which are simply derived from geometric and material considerations. Further information enables the calculation of windup (that is, accelerating and braking) and roll stiffnesses for springs as a check against the design requirements. (See, for instance, SAE J788A, Oct. 1970, and STP 376, Jan. 1973.) However, none of this currently available literature is directly relevant to the problem of design and design evaluation regarding composite structures. The design of any composite part is difficult, and unique, because the stress conditions within a given structure depend on its manufacturing methods, not just its shape. Programs have therefore been developed on the basis of the strain balance within the spring to enable suitable design criteria to be met. Stress levels were then calculated, after which the design and manufacture of RP springs became feasible [1]. The cantilever spring can be employed to provide a simple format from a design standpoint. Cantilever springs, which absorb energy by bending, may be treated as beams, with their deflections and stresses being calculated as short-term beam-bending stresses. The calculations arrived at for multiple-cantilever springs (that is, two or more beams joined in a zigzag configuration, as in Figure 5-15) are similar to, but may not be as accurate as, those for a single-beam spring [408]. A zigzag configuration may be seen as a number of separate beams each with one end fixed. The top beam is loaded (F) either along its entire length or at a fixed point. This load gives rise to deflection y at its free end and moment M at the fixed end. The second beam is then loaded by moment M (upward) and load F (the effective portion of load F.

STRUCTURAL DESIGN ANALYSIS 323

as detennined by the various angles) at its free end. This moment results in deflection Yz at the free end and moment Mz at the fixed end (that is, the free end of the next beam). The third beam is then loaded by Mz (downward) and force Fz (the effective portion of F 1), and so on. The total deflection, y, is the sum of the deflections of the individual beams. The bending stress, deflection, and moment at each point can be calculated by using standard equations. To reduce stress concentration, all comers should be fully radiused. This type of spring is often favored because of its greater design flexibility over the single-beam spring. The relative lengths, angles, and cross-sectional areas can be varied to give the desired spring rate!:' in the available space. Thus, the total energy stored in a cantilever spring is equal to Y

where F = total load in lbs., y cantilever spring, in in.-Ibs.

=

deflection in in., and Ec

=

energy absorbed by the

SHEAR STRESS AND TORSION A torsional beam spring absorbs energy by twisting through an angle (} (see Fig. 5-16) and may thus be treated as a shaft in torsion [2]. A shaft subject to torque is generally considered to have failed when the strength of the material in shear is exceeded. For a torsional load the shear strength used in design should be the published value or one half the tensile strength, whichever is less. The maximum shear stress on a shaft in torsion is given by the following equation [2]: T

Tc

=-

J

where T = applied torque in in.-Ibs., c = the distance from the center of the shaft to the location on the outer surface of shaft where the maximum shear stress occurs, in in. (see Fig. 5-17), and J = the polar moment of inertia, in in. 4 (see Fig. 5-17). The angular rotation of the shaft is caused by torque is given by TL

(} = -

GJ

where L

G

= =

the length of the shaft in in. shear modulus, in psi,

E 2(1

E v

= =

+ v)

Young's modulus (tensile modulus), in psi Poisson's ratio

324 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 5-16. A shaft with diameter d and length L under torque T undergoing angular deformation {).

CROSS·SECTION

~ ~ ~do~

10 10rrl

hID

POLAR MOMENT OF INERTIA J

LOGA TION OF MAX. SHEAR STRESS

c

rrc14

d

32

2

rr(do 4-d,4)

do

32

2

rrb 3h

-b

32

2

b 3h

b

9

2

h4

h

9

2

~

Figure 5-17. Polar moments of inertia for common cross-sections.

The energy absorbed by a torsional spring deflected through angle 6 equals

where Me = the torque required for deflection 6 at the free end of the spring, in in.-Ihs.

STRUCTURAL DESIGN ANALYSIS 325 SPRING CLIP ASSEMBLY

ASSEMBLY BY STAKING

SPRING CLIP

?ZZliZlr;..

//jill!

BEFORE STAKING

?1li77~

PARTS

TO BE ASSEMBLED

AFTER STAKING

F'" SPRING CLIP IN PLACE

F'"

7ffffIi»fm

•

F

SHEAR LOAD

SHEAR AREA

A

= ncJ2 4

Figure 5·18. Examples of direct shear.

SHEAR STRESS AND DIRECT SHEAR There are many situations in which direct shear will be applied to a plastic part. Figures 5-18 through 5-20 illustrate assembly by staking and by using a spring clip in which a load might be applied in direct shear. Other examples might include spot welds and pinned structures such as hinges and conveyor links. For direct shear, as in Figure 5-18, the shear stress is simply the load applied divided by the shear area [2]: F

T=-

A

Direct-shear situations such as those illustrated are the only times when it is appropriate to use the shear-strength data generally reported on marketing data sheets, as previously examined. However, since in these cases the load is not only transferred by shear but also contains a considerable bending or compressing component, all of which are highly geometry dependent, the actual strength of a part under direct shear can be highly variable. Therefore, safety factors should be significantly increased in direct-shear situations.

PRESSURE VESSELS The most common pressure-vessel application of plastics is as a tube with internal pressure. In selecting the wall thickness of the tube, it is convenient to use the thin-wall-tube hoop-

326 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK. UNIFORM INTERNAL PRESSURE, P

o=P~

do

1-R

WHERE

R=(:} This equation is for the maximum hoop stress which occurs on the surface of the inside wall of the tube. Figure 5-19. A cylindrical pressure vessel of thin-walled tube. UNIFORM INTERNAL PRESSURE, P

HOOP STRESS 0=

Pd

2t

This equation is reasonably accurate for t < d/1 O. As the wall thickness increases the error becomes quite large. Figure 5-20. A cylindrical pressure vessel of thick-walled tube.

stress equation (see Fig. 5-19). It is useful in determining an approximate wall thickness, even when condition (t < d/lO) is not met. After the thin-wall stress equation is applied, the thick-wall stress equation given in Figure 5-20 can be used to verify the design.

EXTERNALLY LOADED RP PIPE The use and acceptance of buried large-diameter glass fiber, plastic reinforced, filamentwound pipe has increased steadily since the 1950s. Such RP was selected for its superior corrosion-resistance characteristics and installation-cost savings. ASTM standards use the term Reinforced Thermoset Resin Pipe (RTRP). Filament-wound pipe with a double helical angle of continuous-glass reinforcement (discussed later) is but one of several types of RTR pipe constructions.

STRUCTURAL DESIGN ANALYSIS 327

Attempts have been made to utilize perfonnance standards based upon internal pressures and pipes' stiffness, but other factors must be carefully considered in designing buried piping systems, especially the longitudinal effects of internal pressure, temperature gradients, and pipe bridging. Failing to recognize these factors incurs the risk of underdesigning a system. Because of its resin-glass construction, the physical characteristics of RTR pipe and therefore the design techniques needed for it differ considerably from those of older, traditional pipe materials. It is true that RTR pipe design does to a degree parallel the design philosophy for steel pipe, but there is a point where the steel and RTR pipe-design approaches part company, even though steel and RTR pipe are by definition both flexible conduit. In other words, both kinds of pipe can bend and deflect after burial, within certain limitations, without suffering structural or functional failure. In this regard they both differ from concrete pipe, which is a rigid conduit that cannot tolerate bending or deflection to the same extent as RTR and steel pipe. Since an appreciation of the differences between flexible and rigid conduit is essential to a better understanding of RTR pipe design, let us look at these differences. The diagrams in Figure 5-21 illustrate the results of actual load testing on both types of conduit by the Roadway Committee of the American Railway Engineering Association at an installation near Farina, Illinois. Both the flexible and the rigid pipes were buried under thirty-five feet of identical fill material. Obviously, specific pressures vary from installation to installation, but the relationship in the way the two kinds of pipe react to the same burial condition generally remains constant. Let us start by examining a rigid pipe. Because of its rigid, inflexible characteristics, surface load intensifies at the crown of a rigid pipe and is transmitted through the pipe directly to the bed of the trench in which the pipe rests. This is not true with flexible conduit. Because a flexible conduit deflects under covering load of earth, this deflection transfers portions of the load to the surrounding envelope of soil. This is true of both

~ 26

=

Flexible Pipe

Concrete Rigid Pipe

26 pM Computed

26 pM Computed ~

54.7",

~ !1!1_~_1!1!!

Z6

• 1"'"

14 psi

17"'~

~

41 pol

Figure 5-21. A load-testing profile of flexible and rigid underground pipes.

328 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

steel and RTR pipe. The result is that the support of the surrounding earth actually increases the strength of the flexible conduit. Therefore, analyzing the type and consolidation of backfill materials must be considered an integral part of the design process. Two additional observations can be made. First, because a rigid pipe transmits almost all the load of the earth cover to the trench bed, someone will occasionally be heard to say that rigid pipe, such as concrete pipe, does not require side support. This is not true. Second, because of the difference in the ways rigid and flexible conduit distribute the load of their earth cover, flexible piping materials are often said to require less bedding bearing strength, because they impose less of a load on the trench bed. This is indeed true. In fact, it is one of several factors that help to reduce the installed cost of RTR large-diameter pipe. Given these differences between rigid and flexible conduit, let us look at the differences between steel and RTR pipe, both of which are, of course, flexible conduits. First, steel pipe is by definition constructed from a material, steel, that for our purposes is a homogeneous isotropic substance. Therefore, steel pipe can be considered to have the same material properties in all directions; that is, it is equally strong in both the hoop and longitudinal directions (see Fig. 5-22). RTR filament-wound pipe is, however, an isotropic material. That is, its material properties, such as its modulus of elasticity and ultimate strength, are different in each of the principal directions of hoop and longitude. It is here where the design approaches for steel and RTR pipe part company (see Fig. 5-23). This behavior is a result of the construction of filament-wound RTR pipe (see Chapter 7). Its manufacturing is done by winding continuous strands of resin-impregnated glass fiber around a steel mandrel at a precisely controlled helix angle, under controlled tension. A cross-sectional view of an RP layup is shown in Figure 5-24. As seen, the structural wall of the pipe is made up of continuous strands of fiberglass embedded in a resin matrix, plus an internal corrosion barrier of liner. The liner can be constructed from a number of different resins and reinforcement materials, depending on what will eventually be put through the pipe. Incidentally, the thickness of the liner is not considered during design analysis, except for calculating buckling and pipe deflection. Broadly speaking, three factors control the physical properties of RTR pipe. These are the amount of continuous-glass filament used to construct the pipe wall, the prescribed

Figure 5-22. The material properties of relatively homogeneous isotropic steel pipe.

STRUCTURAL DESIGN ANALYSIS 329

dual-helix angle at which the glass is wound around the mandrel, and the type and amount of resin matrix used to bind the glass filaments together. Controlling the strength of the pipe in the hoop and longitudinal directions is done by selecting the wind angle and ratio of glass to resin content. The wind angle for the structural wall is usually from 55 to 65 degrees to the horizontal, and the glass-fiber content is not less than 45 percent by weight. The final material composition of the pipe is determined by calculating the longitudinal and hoop strengths needed to meet installation requirements demanded by the project. By now it should be apparent that, while both steel and RTR pipe are by definition flexible conduit, they are also quite different and therefore require different design approaches, even though initially at least their design considerations are identical. As with steel pipe, the RTR pipe designer must concern oneself with both pipe deflection and buckling analysis. Unlike the steel pipe designer, however, the RTR pipe designer must also examine a third area of concern. This third factor is a combined strain analysis in both the hoop and longitudinal directions. This analysis demands a thorough examination of such important considerations as diametrical bending, internal pressure, temperature gradients, and the ability of

Figure 5-23. The material properties of anisotropic RTR pipe.

SCnK1 ...1 PI~

Wan -....--"",-

IlIlema'

U.~r---'"

Figure 5-24. A cross-sectional view of RTR pipe.

SCreItt1I11 Hoop

330 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

the pipe to bridge voids in bedding. In such a design system the pipe is seen as part of a buried pipe system in the ground. In simplest terms the design goal is to select the correct RTR pipe configuration for a specific application. In other words, we want to design a pipe wall structure of sufficient stiffness and strength to meet the comb'ined loads that the pipe will experience over the long term. There are two primary ways to achieve correct pipe stiffness. One is to design a straight-wall pipe in which the wall thickness controls the stiffness of the pipe (see Fig. 5-25). Another way is to design a rib-wall pipe on which reinforcement ribs of a specific shape and dimension are wound around the circumference of the pipe at precisely calculated intervals. The advantage of rib-wall pipe is that the nominal wall thickness of the pipe can be reduced while maintaining or even increasing its overall strength-toweight ratio. Generally, a rib-wall pipe design is selected for applications where burial conditions are extreme or for difficult underwater installations. The ability to increase or maintain pipe stiffness by means of reinforcement ribs also provides the engineer with the ability to design an RTR pipe system to fit the economic as well as mechanical parameters of a project. The next step in design is to determine the pipe deflection requirements, based on the equation shown in Figure 5-26. The accepted maximum allowable pipe deflection should be no more than 5 percent. This value is the basic standard that AWWA M-ll specifies for steel conduit and pipe, as do the ASTM and ASME. As is obvious, there are a number of factors that contribute to pipe deflection. These are the external loads that will be imposed on the pipe, both the dead load of the overburden as well as the live loads of such things as wheel and rail traffic. The factors affecting RTR pipe deflection are as follows: 1. Design pipe deflection

4. Modulus of soil reaction (E')

(dX)

2. Dead load (Wd) Trench shape Overburden weight Depth of cover 3. Live load (Wd Wheel load and spacing Surcharge

5. 6.

7. 8.

Native soils Type of backfill Differential soil stress and consolidation Deflection lag factor (Dd Bedding shape (k) Pipe stiffness (EI) Pipe radius (R)

In terms of dead loads, the shape of the trench in which the pipe will be buried is also a factor. Generally speaking, a narrow trench with vertical side walls will impose less of a load on the pipe than will a wider trench with sloping side walls. It is necessary also to know the modulus of soil reaction (E'), which is dependent on the type or classification of the native soil, the backfill material that is contemplated, and the desired consolidation of the backfill material. Soil consolidation is important, because it contributes to the strength of a flexible conduit in a buried pipe system. If the designer is to do the job properly, it is important to have accurate data on which to base calculations. That is why test borings and proper laboratory analysis to determine the E' value of the soil sample are essential. An arbitrary textbook selection of a soil modulus should always be avoided. However, if a pipe is to be buried deeper than the sampling zone that underwent laboratory testing to determine E' and if the test bore shows the deeper material to be equal or better, then the engineer may increase the E' value proportionally to the square root of the differential soil stress.

STRUCTURAL DESIGN ANALYSIS 331

RTR Pipe Wall Structures

j"..i

_,..,7 Straight WaU

Rib WaU Figure 5-25. Examples of RTR pipe wall structures.

I

,,

",---- ............" , ,

,

/

\

~-d1--.,

1.---

d2 ---.4 Maximum 5% Deflection

(AX max S5%) by A.... A M.ll, ASTM, and ASME. Figure 5-26. How to calculate maximum allowable pipe deflection, per A WWA M = 11, the ASTM, and the ASME.

Assuming that all the necessary data are available, determining the necessary pipe stiffness for the maximum allowable pipe deflection is relatively simple. The SpanglerIowa equation below provides a useful, reasonable determination of what wall structure will be needed. (See soil-engineering Bulletin 153, "Structural Design of Flexible Culverts," by M. G. Spangler, Iowa Engineering Experimentation Station, 1971.) The deflection equation is as follows:

M% =

J.. [[DLWd + Wd kR3 ] 2R

Pipe stiffness

EI

teL

+ O.061E'R3

~il

stiffness term

332 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

This equation can be rearranged to determine the terms needed to solve for the minimum pipe stiffness (El)d required to meet the deflection-design criterion: (EI)d

= R3

(El)d

=

[k[DL

W~+ WLl

- 0.061 E' ]

0.0365 (10)6 Ibs.-in. 2j(lineal inch); approx. 0.50 in. wall

Assume for the purposes of this project that the calculations indicate that this pipe stiffness of 0.0365(10)6 Ibs.-in.2j(lineal inch) is in fact required to meet this deflectiondesign criterion of no more than 5 percent deflection. Theoretically, we could choose a pipe-wall structure to meet this required stiffness-either a straight-wall pipe with a thickness of approximately 1.3 cm (0.50 in.), or a rib-wall pipe that would provide the same stiffness. But would the wall structure selected be of sufficient stiffness to resist the buckling pressures of burial, or superimposed longitudinal loads? At this point we do not really know. To find out, we must know a few more things, one of being the amount of resistance to buckling that is wanted in the pipe. The ASME Section III Standard of a four-to-one safety factor on critical buckling, based on many years of field experience, should be used. To calculate the stiffness or wall thickness capable of meeting that design criterion one must know what anticipated external loads will occur (see Fig. 5-27). This time, in addition to the dead loads one must also consider the effects of possible flooding on both an empty and full pipe, as well as the vacuum load it is expected to carry. The analysis should include the modulus of soil reaction, because in a buried RTR piping system the elastic medium surrounding the pipe helps increase the pipe's resistance to buckling. The formula into which all these factors can be inserted to determine the critical buckling pressure of the pipe is called the Luscher & Hoeg formula. (This is based on work presented to the Highway Research Board by C. V. Chelapati and J. R. Allgood in Feb. 1970 entitled "Buckling of Cylinders in a Confined Media," and on Allgood's paper "Structures in Soil Under High Loads," Journal of Soil Mechanics and Foundations, Proceedings, ASCE, Mar. 1971.) The L & G buckling equation is:

Flood Water

-I 1 I I I I I I I I , I , I I , I .

Ground Level

Earth Load Vacuum Load

Figure 5·27. Buckling analysis.

Confined Soil Media

STRUCTURAL DESIGN ANALYSIS 333

It has been determined that, with burial depths greater than two thirds the radius of the pipe, this equation provides a means of determining the required pipe stiffness for critical buckling. To make the equation easier to use, it can be rewritten by substituting certain values and solving for the required stiffness for buckling (EI)b as follows: 0.248 p~R3 BE'

(E/)b =

where

B

=

1 - (RIRoi 1.3

R

+

0.52(RIR o)2

= Pipe radius

Ro=R+H H

= Depth of cover (H > i R)

Suppose that this Luscher & Hoeg equation says we will require a pipe stiffness of 0.123(10)6 Ibs.-in. 2/(lineal in.) to meet a four-to-one critical buckling pressure safety factor. This is a straight-wall thickness of approximately 1.9 cm (0.75 in.). But remember that we earlier calculated that a 0.50-in.-thick wall would be sufficient to withstand the anticipated deflection pressure. Which of these two wall thicknesses is correct? Quite logically, it is the larger one of the 0.75 in. thickness, or a rib-wall pipe of equivalent stiffness. To put it another way, after carefully completing both a deflection and a buckling analysis, always select the pipe stiffness that is greater. Now if we were designing in steel pipe, the work would be about over. But since the design is a largediameter RTR buried piping system, we are not. From experience it has been learned that the final choice of an RTR pipe configuration cannot be made until the effects of strains in the longitudinal and hoop directions have been carefully investigated. The reason is obvious, the material, continuous glass-reinforced thermosetting resin, is anisotropic. Unlike a homogeneous isotropic material, such as steel, the strength of RTR pipe in its longitudinal and hoop directions is not equal. The effects of this unequal strength in the two directions must therefore be seriously considered during design if an RTR piping system is to meet the long-term operating requirements of the system being designed. Therefore, before a final wall structure can be selected, it is necessary to conduct a combined strain analysis in both the longitudinal and hoop directions. This analysis will consider thermal contraction strains, the internal pressure, and the pipe's ability to bridge soft spots in the trench's bedding. In order to do this we must know more about the inherent properties of the material we are dealing with-a laminate made up of successive layers of continuous filament-wound fiberglass strands embedded within a resin matrix. We must know the modulus of the material in the longitudinal direction and the hoop direction, plus the material's allowable strain.

334 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

These values are determinable through standard ASTM-type tests, including those for hydrostatic testing, parallel plate loading, coupon tests, and accelerated aging tests (see Fig. 5-28). The next step is to examine Figure 5-29, which shows the tensile stressstrain curve for typical steel-pipe materials. In the steel pipe business, designing is based on the curve's yield point. As noted previously, the yield point on a stress-strain curve beyond which steel pipe will enter into the range of plastic deformation that could lead to a total collapse of the pipe. Generally, steel-pipe designers select an allowable design strain of approximately two thirds of the yield point. RTR pipe designers also use a stress-strain curve similar to that used by steel pipe designers. However, instead of a yield point, they use what is called an empirical weep point, or the point of first crack (see Fig. 5-30). It is determined by either coupon or hydrostatic tests. The weep point is the point at which the matrix becomes excessively strained so that minute fractures begin to appear in the structural wall. At this point it is probable that in time even a more elastic liner will be damaged and allow water or whatever else the pipe is carrying to ooze or weep through the wall. As is the case with the yield point of steep pipe, reaching the weep point is not necessarily cataclysmic. The pipe can still continue to withstand quite a bit of additional load before it reaches the point of ultimate strain and failure. The weep point or strain-to-first-crack in a wall for filament-wound pipe constructed using isophthalic resin is currently found to be not less than 0.009 in.!in. This has been repeatedly demonstrated by careful coupon testing and burst testing of pipes with straingauge instrumentation attached. Thus, design values are based on the strain-to-first-crack or the empirical weep point. For normal design conditions a strain of 0.0018 in.!in. is used, which provides a fiveto-one safety factor. For transient design conditions a strain of 0.0030 in.!in. is used, for a safety factor of approximately three to one. To those familiar with the design safety factors of other pipe manufacturers following NBS Voluntary Product Standard PS 1569, these safety factors may seem modest. However, PS 15-69 is based on the ultimate tensile strength of the material. The next step is to proceed with a strain or stress analysis in the longitudinal and hoop

r:===Hydl'08tatic Teeting

"I t

Flexural

-; t

ParaDel Plate Teeting

. . . . J: ..

:[~ Coupon Teetinlll

Ten.ile

Figure 5-28. Examples of tests using strain gauges to develop stress-strain curves.

STRUCTURAL DESIGN ANALYSIS 335

directions. When conducting this analysis the designer has the option to work in terms of either stress or strain. Strain is generally used, since it can be accurately measured, using reliable strain gauges, whereas stresses have to be calculated. From a practical standpoint both the longitudinal and the hoop analysis determine the minimum structural wall thickness of the pipe. However, since the longitudinal strength of RTR pipe is less than it is in the hoop direction, it is wise to approach longitudinal analysis first. In doing so there are three major factors to consider: the effects of internal pressure, the expected temperature gradients, and the ability of the pipe to bridge voids in the bedding. Analyzing these factors requires that several equations be superimposed, one 120 100 90 80

Stress

Ox10' PSI

70 80 50 40 30 20 10 0

Design Value at 2/3rd Yield

-

Modulus of Elasticity

30(10)' PSI

1000

2000

3000 4000

8000 8000

10,000

Strain

&x(10rl.j n./ in . Figure 5-29. An example of a tensile stress-strain curve for mild steel pipe material.

100 80

80 70

Stress eo

Strain to 1,t Cnlck or empirical Weep Point (0.009 In./ln.)

Ox10' PSI 50

«)

30 20

10 Nonnal

Design Strain

10,000 Tnlnalent .Deslgn Stnlin

15,000

20,000

Strain

&x(10rl.

in. Ii n.

Figure 5-30. An example of a tensile stress-strain curve for reinforced thermoset resin (RTR) pipe material.

336 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

on another. Even though from a practical standpoint all these longitudinal design conditions are solved simultaneously, it is interesting to examine each individually. Some engineers tend to disregard the effects of internal pressure in the longitudinal direction on buried pipe because they theorize that the longitudinal load is cancelled out by the earth surrounding the pipe. Or they may assume that in a gasketed joined pipeline the gasket joints will allow the pipe to move freely, so that no longitudinal load will exist. However, this situation is not necessarily true. Poisson's effect, or Poisson's ratio, which was previously described, can have an influence. This effect occurs when an openended cylinder is subjected to internal pressure. As the cylinder expands diametrically, it also attempts to shorten longitudinally. These movements will not be visible to the naked eye in all cases but can be easily measured with strain gauges. Or the movements can be observed in the shortening of pressurized pipe where a test fixture absorbs the pressure thrust. Since in a buried pipe movement is resisted by the surrounding soil, a tensile load is produced within the pipe. The internal longitudinal pressure load in the pipe is independent of the length of the pipe. Thus, Poisson's effect must be considered when designing any length of pipe, whether long or short, that is part of a buried pipe system. Buried pipes are influenced by friction with their surrounding media. Several equations can be used to calculate the result of Poisson's effect on pipe in the longitudinal direction in terms of stress or strain. The following equation provides a solution for a straight run of pipe in terms of strain. Thus, the longitudinal strain in pipe due to internal pressure is

where EL = pressure. However, where there is a change in direction and thrust blocks are eliminated through the use of harness-welded joints, a different analysis is necessary. This is so because, compared to in straight runs of pipe, the longitudinal load imposed on either side of an elbow is greater. This increased load is the result of internal pressure, a temperature gradient, or a change in momentum of the fluid. Because of this increased load, the pipe joint and elbow thickness may have to be increased to avoid overstraining. There is a special equation, shown in Figure 5-31, to calculate the longitudinal strain in pipe at harnessed elbows. For the sake of simplicity the effects of internal pressure, temperature gradient, and change in momentum of the fluid have been combined into one equation. After examining the effects of internal pressure in the longitudinal direction, the next step is to investigate the longitudinal tensile loads generated by a temperature gradient in the piping system. The goal is to determine the extent of the tensile forces imposed on the pipe because of cooling. When an open-ended cylinder cools, it attempts to shorten longitudinally. A tensile load is then imposed by the resistance of the surrounding soil. As a matter of fact, any temperature change in the surrounding soil or medium that the pipe may be carrying can produce tensile load. The effects of temperature gradient on pipe can be written in terms of strain. Thus, the longitudinal tensile strain generated by a temperature gradient is EL(temperature)

= fl:roc

STRUCTURAL DESIGN ANALYSIS 337

EL-

Fd1-Cose)

Where:

ElA F,

= Total thrust

F,

= Pressure +Temperature + Change in momentum

nd 2 P

V

F: =-4- +auA+QPg

A :: ndt,

a" c ~flTElOC Figure 5-31. An example of longitudinal tensile strain in pipe with hamessed elbows.

where flT is the greater of flTa or flT; T(max)

T(min)

flTa

=

~--~--~--~

flT;

=

T(installation) - T(min)

cr =

2

Coefficient of thermal expansion

In this analysis the designer must consider two conditions and base the pipe design on the one that is worse. One condition is where the temperature differential is one half the difference between the maximum temperature and minimum temperature. The second condition considers the temperature differential between the maximum pipeline temperature at installation and the minimum design temperature. The next step in longitudinal analysis is to examine the bridging. Bridging can occur, and if so must be considered, wherever the bedding grade's elevation or the trench bed's bearing strength varies, when a pipe projects from a headwall, or, of course, in all subaqueous installations. It is a good engineering practice to design the pipe to be strong enough to support the weight of its contents, itself, and its overburden while spanning a void of two pipe diameters (see Fig. 5-32).

]i Earth

Figure 5-32. An example of the longitudinal tensile load on a pipe bridging a void of two pipe diameters.

338 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The effects of longitudinal tensile bridging can be calculated using the following strain equation:

6 L(bridging)

where Wt

=

Pipe contents

+

pipe weight

+

overburden

For simplicity, the condition considers the conservative case where the pipe acts simply as support. The normal practice is to solve all these equations simultaneously, then determine the minimum wall thickness that has strains equal to or less than the allowable design strain. Thus, the minimum structural wall thickness dictated as longitudinal tensile strain is:

6L(allowable design strain)

~

L

[6L(pressure)

+

6L(temperature)

+

6L(bridging)]

The importance of combining longitudinal strain analyses is that it often provides the designer with a minimum wall thickness on which to base the ultimate choice of pipe configuration. For instance, assume that the combined longitudinal analysis indicates a minimum of 0.625, or 5/8 in. of wall thickness. However, deflection analysis calls for a 0.50, or I!2-in., wall, and the buckling analysis says we need a 0.75, or 3/4-in., wall. We had already decided that the most likely candidate was the 3/4-in. wall. Now longitudinal analysis says that a 5/8-in. wall is enough to handle the longitudinal strains likely to be encountered. Which wall thickness, or what pipe configuration-straight wall or ribbed wall--do we now settle on? Economic considerations would have to be weighed, but the experienced designer would most likely choose the 3/4-in. straight-wall pipe. This is, of course, if the design analysis is complete, but it is not, since there still remains strain analysis in the hoop direction. The target is to determine if the combined loads of internal pressure and diametrical bending deflection will exceed the allowable design strain (see Fig. 5-33). This entails investigating the effects of rerounding or decreasing the diametrical deflection that occurs because of internal pressure (see Fig. 5-34). There is a method the engineer can use for this analysis, but it is extremely complex and can be most conveniently checked by the pipe supplier, using a computer. In general, equations are generated to determine the moment and thrust created in the invert area of the deflected pipe, where a pressure term is superimposed. This analysis must examine the strains in the outer- and innermost fibers of the pipe to verify that its wall structure is adequate and not overstrained (see Fig. 5-24). During this analysis the pipe must be examined under conditions of no pressure, minimum pressure, and maximum pressure. Although this analysis should be conducted for both straight-walled and rib-walled pipe, it is particularly important in the case of rib walled (see (Fig. 5-25). That is, because the rib is often thicker than the structural wall of the pipe, by several times the wall's

STRUCTURAL DESIGN ANALYSIS 339

'" I,;;:: Figure 5-33. An example of strain analysis in the hoop direction, with an external load only-there is no internal pressure rerounding.

,:

11

Figure 5-34. An example of strain analysis in the hoop direction, with an external load plus internal pressure rerounding.

thickness, strains along the ribs may be higher than along the straight-walled sections, particularly at the top of the rip. For the sake of this discussion, assume that strain analysis in the hoop direction has confirmed that a structural wall thickness of 1.9 cm (0.75 in.) is satisfactory. Does that mean the pipe design has been completed? Not yet. There's more to a piping system than just the pipe walls. We must still design the joints to connect the straight lengths of pipe together. The designing of joints is perhaps one of the most overlooked areas in piping-system design. Since the performance of the whole piping system is directly related to the performance of the joints, this subject deserves serious attention. For example, use a bell-and-spigot joint with an elastomeric seal (see Fig. 5-35). This type of joint permits rapid assembly of a piping system and thus offers an economic advantage in terms of installed cost. It should be used as much as possible for connecting straight runs of pipe, especially at points where the pipe projects from a rigid structure. In terms of flexibility, the joint should be able to rotate at least two degrees without a loss of integrity. Thus, the threat of failure from unanticipated pipe subsidence is substantially reduced, and changes in the grade line during installation can be the more easily accommodated. The joint must also be designed with corresponding bell and spigot stiffnesses. And the spigot should have a special control ridge to ensure proper gasket seal, even when the pipes on either side of the joint are not uniformly supported. The next type of joint is weld overlays, which are often utilized to eliminate the need for costly thrust blocks (see Fig. 5-36). In designing the pipe we conducted an analysis to ensure that it possessed sufficient longitudinal strength. It makes sense, then, to make the weld joints be at least as strong as the longitudinal strength of the pipe rather than just as an internal pressure-seal pipe. This can be done by noting the pipe's structural wall thickness and the strength relationship between the pipe and the overlay weld. These equations show one way of relating the structural wall thickness of a pipe to its longitudinal design's allowable values first in regard to the longitudinal strength of the pipe and its overlay laminate, to determine the proper thickness of the joint, and second to the longitudinal pipe strength and the weld overlay's bond strength. The equation to

340 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

\ Gaskets

Double Ga8ket

in~le GIIMk I

Figure 5·35. An example of RTR pipe using elastomeric bell-and-spigot seals.

Bell

FIberglass & RIISIIl - -

&..

o..-y -

Shear l.ength

..

External Harnt"M- elded Joint

Internal Hameu-Welded Joint

Figure 5·36. An example of RTR pipe joints with a weld overlay. Reinforced glass-fiber-TS polyester on either the outside or inside of the overlap joint provides extra strength.

ensure joint strength is equal to the one for longitudinal strength of the pipe thickness design. Thus, the thickness design can be calculated as

t(overlay) ;;:: t(structural pipe wall)

longitudinal pipe strength ) (I ongl·tUd·Inal overIay strength

Likewise, the shear length can be figured as

L(overlay) ;;:: t(structural overlay)

(

longitudinal overlay strength ) longitudinal overlay bond strength

When weld joints are used to fabricate special items such as mitered fittings, tees, laterals, and so forth, the engineer should always consider using additional weld overlay thicknesses applied to both the internal and the external surfaces, using RP. In this way a safe design can be developed to resist the forces that attempt to distort the special and "peel" loose a weld overlay or rupture the special piece.

MOLDED-IN INSERTS Plastics perform satisfactorily with metal-molded inserts (see Fig. 5-37) and expansiontype inserts (see Fig. 5·38) [2]. To minimize the stresses created at the metal-plastic interface by the differences in thermal expansion rates for molded-in inserts, observe the following safeguards: 1) the design permitting, use plain, smooth inserts; 2) use simple pull-out and torque-retention grooves when high torque and pull-out retention are required; 3) if a knurled insert is used, keep the size of the knurls to a minimum, remove all sharp comers, and round the hidden end of the insert and keep the knurled section away from parts' edges, 4) keep the inserts clean, removing chemicals such as oil from them; 5) use high mold temperatures to reduce thermal stresses, such as for commodity plastics

STRUCTURAL DESIGN ANALYSIS 341

KNURL

A

1B

C

INTERNAL THREADS

D

I

'1

F

E

t~ l

G

H

~i~~~~~AL JB I SEALING SHOULDER

I

I

TT l

Figure 5-37. Molded-in insert design.

at 82 to 105°C (180 to 220°F); and 6) provide sufficient material around the insert. Use the following guidelines for material thicknesses around inserts: with aluminum use 0.8 times the outer radius of the insert, with brass use 0.9 times it, and in steel use a thickness equal to the outer radius. To ensure a proper interface, prototyping is recommended. When metal inserts require hermetic sealing, consider coating them with a flexible elastomer such as an RTV rubber, or a urethane or epoxy system. A second method is to design an annular space or reservoir at one end of the insert from which to dispense the flexible elastomers to effectively create a hermetic seal. Flexible sealants are also used to compensate for differences in the thermal coefficient of expansion between metal and plastic.

342 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~

V 8 - '·"·' '. [O,.A, , •• [AD$

PLAS TIC PART

CO~(

5PP[AO

IN5(IU

="=~-- SCRflN nPA OS ~'I<":':~="-"I:~~ HE INSE RT

Figure S-38. Example of metal expan ion· type lOlled and nons lotted inserts.

PRESS FITS Press fits, which depend on having a mechanical interface, provide a fast, clean, economical assembly. A common usage is to have a plastic hub or boss that accepts either a plastic or metal shaft or pin. The press-fit procedure tends to expand the hub, creating tensile or hoop stress. If the interference is too great, a high strain and stress will develop. The plastic part will do one of several things. It may fail immediately, by developing a crack parallel to the axis of the hub to relieve the stress, which is a typical hoop-stress failure. It could survive the assembly process, but fail prematurely in use, for a variety of reasons related to its high induced-stress levels. Or it might undergo stress relaxation sufficient to reduce the stress to a lower level that can be maintained. Hoop-stress equations for two typical press-fit situations are shown in Figure 5-39. The allowable design stress will depend upon the particular plastic, the temperature, and other environmental considerations. A simpler, though less accurate, method of evaluating press fits is to assume that the shaft will not deform when pressed into the hub. This is reasonably accurate when a metal shaft is used in a plastic hub. The hoop strain developed in the hub is then given by

STRUCTURAL DESIGN ANALYSIS 343 Ep = MODULUS OF ELASTICITY OF PLASTIC Em vp 0.

= MODULUS OF ELASTICITY OF METAL = POISSON'S RATIO OF PLASTIC = ALLOWABLE DESIGN STRESS FOR PLASTIC

....- - d i

i

= d s • d, = DIAMETRAL

INTERFERENCE

i. = ALLOWABLE INTERFERENCE CASE A SHAFT AND HUB ARE BOTH THE SAME OR ESSENTIALL Y SIMILAR MA TERIALS HOOP STRESS GIVEN "i" IS

o=~Ep_r_ ds r+1 OR, THE ALLOWABLE INTERFERENCE IS

r+1

r GEOMETRY FACTOR

r

=

CASE B SHAFT IS METAL, HUB IS PLASTIC HOOP STRESS GIVEN "i" IS

_'_+_(--:-:::....<),-2_

0=_'_" Ep r ds r +Vp OR, THE ALLOWABLE INTERFERENCE IS

1 -

i. = ds ~ r +Vp

(::y

Ep

r

Figure 5-39. Press-fit conditions for two typical situations.

The hoop stress can then be obtained by multiplying the appropriate modulus. For high strains, the secand modulus will give the initial stress; the apparent or creep modulus should be used for longer-term stresses. The main point is that the maximum strain or stress must be below the value that will produce creep rupture in the material. There is usually a weld line in the hub that can significantly affect the creep-rupture strength of most plastics. An additional frequent complication with press fits is that a round hub or boss is often difficult to mold, if strict processing controls are not used to eliminate potential problems. There is a tendency for a round hub to be slightly elliptical in cross-section, increasing the stresses on the part. For critical part performance and in view of what could occur, life testing should be conducted under actual conditions. The consequences of stress occurring will depend upon many factors, such as the temperature during and after the assembly of the press fit, the modulus of the mating material, the type of stress, the usage environment and-probably the most importantthe type of material being used. Some substances will creep or stress relax, but others will fracture or craze if the strain is too high. Except for light press fits, this type of assembly design can be risky enough for the novice, because the boss might already be weakened by a knit line. Figure 5-40 presents alternative methods for using press fits that present a lower risk of failure [2].

0. . .D

METAL PIN

STRAIGHT (INTERFERENCE) PRESS FIT CAN PRODUCE HIGH STRAINS

?r-Z""'V"""ZT"'l7////? STRAIN~ '~:~~::~;: ALTERNATIVE PRESS FIT DESIGNS FOR LOWER STRESS.

ADD METAL "HOOP" RING PREVENTING EXPANSION OF PLASTIC BOSS.

o

USE BARBS OR SPLINES ON THE METAL PIN TO CREATE INTERFERENCE FIT AND RETENTION

CREATE INTERFERENCE PRESS FIT BY ADDING "CRUSH RIBS" TO THE INSIDE DIAMETER OF THE BOSS

Figure 5·40. Alternate press-fit designs for metal pins in a plastic hub. 344

STRUCTURAL DESIGN ANALYSIS 345

SNAP FITS Snap fits are widely used for both temporary and permanent assemblies, principally in injection- and blow-molded parts. Besides being simple and inexpensive, snap fits have been superior qualities (see Figs. 5-41). Snap fits can be applied to any combination of materials, such as metal and plastics, glass and plastics, and others. All types of plastics can be used. The strength of a snap fit comes from its mechanical interlocking, as well as from friction. The pull-out strength in a snap fit can be made hundreds of times larger than its snap-in force. In the assembly process, a snap fit undergoes an energy exchange, with a clicking sound. Once assembled, the components in a snap fit are not under load, unlike the press fit, where the component is constantly under the stress resulting from the assembly process. Therefore, stress relaxation and creep over a long period may cause a press fit to fail, but the strength of a snap fit will not decrease with time. When used as demountable assemblies, snap fits can compete very well with screw joints. The loss of friction under vibration can loosen bolts and screws. A snap fit is vibration proof, however, because its assembled parts are in a low state of potential energy. There are also fewer parts in a snap fit, which means a savings in component and inventory costs. Successfully designing snap fits depends on observing a set of rules governing the shape, dimensions, materials, and interaction of the mating parts. The interference in a snap fit is the total deflection in the two mating members during the assembly process. Too much interference will create difficulty in assembly, but too little will cause low pull-out strength. A snap fit can also fail from permanent deformation or the breakage

Figure 5-41. This snap assembly, used in a patio table, combines the convenience of snap assembly with the durability of a mechanical fastening.

346 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

of its spring components. A drastic change in the amount of friction, created by abrasion or oil contamination, may ruin the snap. A snap can be characterized by the geometry of its spring component. The most common snaps are the cantilever type, the hollow-cylinder type (as in the lids of pill bottles) and the distortion type. These snaps include those in any shape that is deformed or deflected to pass over an interference. The shapes of the mating parts in a hollow-cylinder snap are the same, but the shapes of the mating parts in a distortion snap are different, by definition. This three-part classification is rather nominal, because the cantilever category is used loosely to include any leaf-spring components, and the cylinder type is used also to include noncircular section tubes.

Design Approach to Snap Fits For high-volume production, snap-fit designs provide economic, rapid assembly. In many products, such as inexpensive housewares or hand-held appliances, they are designed for one assembly only, with no nondestructive means for disassembling them. Where servicing them is anticipated, provision is made for the release of the assembly with a tool. Other designs, such as those used in the battery compartment covers for calculators and radios, are designed for easy release and reassembly many hundreds of times. There is always some part of a snap fit that must flex like a spring, usually past a designed-in interference, and quickly return, or at least nearly return, to its unflexed

Figure 5-42. Using a combination of snap fit, living hinge, and mechanical fastener capabilities in molded polypropylene makes possible the storing and shipping of a large chassis grid in a flat form. For assembly it converts to a cagelike inner housing that holds separate units, for cost savings.

STRUCTURAL DESIGN ANALYSIS 347 PROPORTIONALITY CONSTANT, K. FOR TAPERED BEAM 2.3 2.2 2.1

2.0 1.9 1.8

K

1.7 1.6 1.5 1.4 1.3 1.2 1.1

1.0 0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAPERED BEAM STRAIGHT BEAM

1

r-LJl_~ ~ L T r '"-----.--....

T

Y (MAX DEFLECTION)

Y (MAX DEFLECTION)

ho

DYNAMIC STRAIN •

= 3yho 2L2

hL

ho

DYNAMIC STRAIN • = 3yh o 2L2K

Figure 5-43. A basic snap-fit design for a cantilevered beam with a rectangular cross-section.

position, to create the assembly of two or more parts (see Fig. 5-42). The key to successful design is to provide sufficient holding power, without exceeding the elastic limits of the plastic. Figure 5-43 shows a typical design [2]. Using the beam equations, calculate the maximum stress during assembly. If it stays below the yield point of the plastic, the flexing finger will return to its original position. However, for certain designs there will not be enough holding power, because of the low forces or small deflections.

348 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

It has been found that with many plastics the calculated flexing stress can far exceed the yield point stress, if the assembly occurs too rapidly. In other words, the flexing finger will just momentarily pass through its condition of maximum deflection or strain, and the material will not respond as if the yield stress had been greatly exceeded. A common way to evaluate snap fits is to calculate their strain rather than their stress. Then compare this value with the allowable dynamic strain limit for the particular plastic. In designing the finger it is important to avoid having sharp comers or structural discontinuities that can cause stress risers. A tapered finger provides a more uniform stress distribution, which makes it advisable where possible. Snap fits usually require undercuts, so a mold with a side action is usually used, as shown in Figure 5-44 [2, 11]. Another approach when an opening at the base of the flexing finger is permitted is shown in Figure 5-45 [2, 11]. There are times when all that has to be done is just pop it off the mold, taking advantage of the plastic's flexibility. Another type of system is the snap-on or snap-in kind, used primarily in round parts (see Fig. 5-46) [2].

SIDE CORE MOVES DOWN TO MOLD UNCERCUT, AND IS ACTIVATED UP TO ALLOW MOLD TO OPEN

Figure 5-44. Molding a snap-fit flexing finger with a mold, using a side-core action.

PROJECTION FROM CAVITY. THROUGH PART, AND MATING WITH CORE FORMS UNDERCUT WITHOUT NEED FOR SIDE CORE. IT DOES, HOWEVER, LEAVE SMALL "WINDOW" OR OPENING IN MOLDED PART.

Figure 5-45. Molding a snap-fit flexing finger without side-core action.

STRUCTURAL DESIGN ANALYSIS 349 SNAP·ONFIT

t

Y,I/l

,

FULL PERIMETER SNAp·IN

PRONGED SNAP·IN BALL OR CYLINDER SNAP·IN

PRONGS

~

WlZ&

Figure 5-46. Examples of different snap-fit designs.

A snap fit can be rectangular or of a geometrically more complex cross-section, as shown in Figure 5-47 and Table 5-3 [11,409]. The design approach for the finger is that either its thickness, h, or width, b, tapers from the root to the hook. Thus, the loadbearing cross-section at any location relates more to the local load. The result is that the maximum strain on the plastic can be reduced and less material will be needed. With this design approach, the vulnerable cross-section is always at the root [409]. The deflection force analysis applicable to cantilevered snap fittings is made using Mobay engineering TPs whose values for permissible strains are given in Table 5-4 [409]. Using the equations in Figures 5-48 and 5-49 [409], the permissible deflection y can be determined easily, even for cross-sections in complex shapes. For example, using Design 2 in Table 5-3, with the thickness of the arm decreasing it is linearly half its initial value. Thus, it increases the permissible deflection by more than 60 percent, as compared to a snap-fitting arm with a constant cross-section, as shown in Design 1 [409]. The deflection force P required to bend the finger can be calculated by using the equations in the bottom row of Figure 5-47 for cross-sections in various shapes. The values for the secant modulus for various Mobay engineering TPs can be determined from Figure 5-50, but more complete and up-to-date data should be obtained directly from Mobay. The strain value used should always be the one on which the actual dimensioning was based [409].

HINGES Although integral hinges are feasible with a number of thermoplastics (e.g., the polyolefins, acetal, nylon, styrene-butadiene, etc.), the concept is generally associated with polypropylene. This section discusses the integral hinge as it is generally applicable to polypropylene. There are three techniques used to fabricate integral hinges: molded-in (by injection or blow molding), cold worked, extruded, and coining.

"""" =

.

I

h

P

Y

~

T:

Dl

27':)

t

6

0

I

p = hh' E,E

--.

z

y=0086 oEop h

y=1.09 0EOp h

y=0.67 oEOp h

Rectangle

= 0

p

12

=~

0

0

a'+411h",+h' E,E 2a+b

z

2a+b

h

1.64 a+h", EOI' 2a+h h

a+b", EOI' y= _ _ o_ 2a+b h

Trapezoid

y = 1.28 a+h",o EOI'

y

c,

0

Shape or cross seclion

Figure 5-47. The equations for designing geometrically complex cross-sections of snap fits

Denection force

",E

0

All dimensions in direction Zo for example band a o decrease to one-fourth

~ 3~tz

All dimensions in direction Yo for example o h or /lro decrease to one-half

2

Cross section constant over the length h

,

Permissible denection

Type or design

A

B I

r,

0

/

r,

1.28'K", EOI'

P = Z'41 E,E

=

r,

K", EOI'

y = l.64 o

y

r,

c,

...-

Ring segment

6

y = K(2) EOI'

r2

_C2

C---.t

~

h

3

C(~I

I EOI' =o _0-

P

0

/

= Z'4' E,E

(cJ)

)'=0.43 0Eo /'

CLH

y=0.55 0Eop

y

Irregular cross section

~C'

~C2

o

STRUCTURAL DESIGN ANALYSIS 351

Table 5-3. Equations for Dimensioning Cantilevers in Fig. 5-47 Symbols y

(permissible) deflection ( = undercut) E = (permissible) strain in the outer fiber at the root; in formalae: E as absolute value = percentagel 100 I length of arm h thickness at root b width at root c = distance between outer fiber and neutral fiber (center of gravity) Z section modulus Z = [Ie, where I = axial moment of inertia Es secant modulus (see Fig. P

Notes (l) These formulae apply when the tensile stress is in the small

surface area b. If it occurs in the larger surface area a, however, a and b must be interchanged. (2) If the tensile stress occurs in the convex surface, use C 2 in Fig.

5-48; if it occurs in the concave surface, use C 1 accordingly. (3) c is the distance between the outer fiber and the center of gravity

(neutral axis) in the surface subject to tensile stress. (4) The section modulus should be determined for the surface subject to tensile stress. Section moduli for cross section type C

are given in Fig. 5-49. Section moduli for other basic geometrical shapes are to be found in mechanical engineering manuals.

3-14) (permissible) deflection force

Molded In Hinges An integral hinge can be injected molded by conventional techniques, provided certain factors are kept in mind. The desirable molecular orientation runs transverse to the hinge axis. This can best be achieved by a fast flow through the hinge section, using high melt temperatures. Since these requirements are also consistent with good molding practice, optimum production rates can be maintained. The main concern in integral-hinge molding is to avoid conditions that can lead to delaminating in the hinge section. These include filling the mold too slowly, having too low a melt temperature, having a nonuniform flow front through the hinge section, suffering material contamination as from pigment agglomerates, and running excessively high mold temperatures near the hinge area. An integral hinge can also be produced by postmold flexing. In this process the hinge section is molded, then subjected to stresses beyond the yield point immediately after molding, by closing the hinge. This creates a necking down effect. Stretching the oriented polymer molecules on the outer surface of the hinge radius provides the remarkable flex strength of the thinned-down hinge section. Flexing the molded hinge must be done while it is still hot, through an angle sufficient to stress its outer surface. This postmolding step provides a maximum and uniform orientation in the hinge area but a minimum of applied stresses. The thinness of the hinge area requires that pigments be well dispersed so that agglomerates will not provide focal points of weakness in the hinge structure.

Cold Worked Hinges Where parts are heavy or complex, it may be impractical to force the necessary quantity of resin through the hinge sections. In such cases integral hinges can be obtained by cold

""""hl

6

10 8

~

IIII!!!!!!!!i!!

0.5

41

61

108 1

0.6

.-

0.7 '1/ r2

0.8

:;..--~--~

I

0.9

1( 15

90

75

60

1; !O 1315 1: ;0 1! i5 18'10 10

-~

A / boo' I

45

711,,0

*/1

I

__

.

.

.

.

.

.

.

__

.

.

30

-

_

_

_.

165 100

120

75 _ 90

I~::

.

0.5

0.6

0.7

,,/r2

0.8

.0875 0.9

1.0

6+1--~--+-~--~--4---~-+~~--+-~

10 8

.

:111111

10 .

2.67

C2

Figure 5·48. Mobay diagrams for determining C) and C2 for cross-sectional shape C in Table 5-4 where C) is the concave side under tensile load and C2 the convex side under tensile load, using Mobay engineering TPs.

c,

I.o.l In I.o.l

:I

j

~

'2

80 100

60 ~

40 ~

"j

10

8

6

2

,....

-

.......

2

4

050

10-.1

2

060

.........

070

~

080

i\'

090

\

\ 1\

I'\.

~ '\

~ C"\ ~

""",-

rl/ r2

~

"

165 150 135 120

1 00

15

30

45

60

75

90

105

~ ~ 180

.... .:"

...... :...... """- .......

~

.......

.; In·

.... ...... ...... t:::: S t=: ::::r~~

t::--

1'"-0-

--

r- :::::: ~ ::::: ~ ~ ......... r--. ~ .... ...... ...... ..... 6

10- 3

10

10 '

2

Z/r2 3

600

' "

,

80 100

60

40

20

10

,

6

4

2

1

',8

'2

Example " 8.75 '2 10
800 1000

j

2001

60 80 100

"1

40

400

"

r

6 8 10

4

2~

z, ~ '23 . ZI'2 3

~

r-

I"'-

I--..

10- 4 • 050

2

4

6

10 3

2

4'

, ,6 , 2

•

060070

•

"r-..

1'0...

I'

~~

r,

'2

'\.

i\\

75

30

4S

60.

090 0875

1.00

tj c.J.S.

I

~ 1\'

I~

.;:~l\: 080

......

90

120. 105

150 135'

~ ~ 160 165

..; rno

........ ........ ~ N'" ~

.... I"'-. r--.: "~ r-;:

.........

.........

........ :"' ~

...... .........

-- ;: -- ;: ,....

::t: I""'-

:::::: ;:t:2 ~~~ ...;:::: ,, . ..... .... ~~~ , 10, 4

6

10-'

ZI'2 3 2

Figure 5-49. Nomograms for determining the section modulus for cross-sectional shape C in Table 5-4 where ZI is the concave side under tensile stress and Z2 the convex side under tensile stress, using Mobay engineering TPs.

800 1000

600

400

200

100

60 80

40

,oj

8~

6

:1-

Zl - f 23 . Z/r23

354 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 12,000

11,000

~

10,000 - 14 9,000 - 12 8,000

g~

7,000

Z

iii :::I

6,000 t-"3

'U 0

~

"

~

E

ra

5,000

tJ

Q)

en 4,000 ~

2,000

~

- 8

....X

<;

- 6

-4 Typical of Bayblend Resins

...............

-

-2

Merion M-Grades

1,000

0

iJ

Petlon 7530

'"~

3,000

- 10 Petlon 3530, 4530

0

1

2

Strain,

f,

%

3

4

5

Figure 5-50. The secant modulus at room temperature for Mobay engineering TPs.

working the molded parts. With this process, since the molecules are properly oriented during forming, the direction of the polymer flow during molding is not critical. A press, a home-made toggle job, or a hot-stamping machine can be used to perform the cold-working operation. The male forming die should be about 132 to 138°e (270 to 2800 P). Pressure should be maintained for about 10 seconds. This time can be reduced if the part still retains residual molding heat or is preheated. The recommended preheating temperature is from 80 to 1l ooe (175 to 2300 P). The die backings may be either hard or flexible. With a hard backing, such as steel, the softened polypropylene is die formed into the desired hinge contour. Thinner hinges

STRUCTURAL DESIGN ANALYSIS 355

Table 5-4. Mobay Guide Data for Permissible Short-Term Strain in Snap Joints (Single Joining Operation)* Material Merion polycarbonate (M-Grades) Bayblend resin MH6500 MC2500 MJ2500 MD6500 Petlon polyester

Percentage 4

2.5

1.5

3530 4530 7530 *For frequent separation and rejoining, use about 60 percent of these values.

are usually made by using a flexible backing like stiff rubber. The deformation of this type of backing produces the hinge contour by stretching the softened plastic and generally results in thinner cross-sections.

Extruded Hinges Forming the hinge cross-section by using an extruder die results in a hinge with poor flex life. Because hinges are formed in the direction of the polymer flow, they cannot be sufficiently oriented when flexed. However, if an extruded hinge is formed by the takeoff mechanism while the polypropylene still retains internal heat, the hinge will have properties approaching those of cold working.

Cold-Forming or Coining Hinges More recently it has become possible to create hinges in some of the tougher engineering thermoplastics by coining techniques. In this technique a molded or extruded part is placed in a fixture between two coining bars. Pressure is then applied to the bars and the part is compressed to the desired thickness, elongating the plastic. Coining is effective only when the material is elongated beyond its tensile yield point. The process is usually used for such materials as acetal and nylon, which cannot normally be molded in a sufficiently thin section for a strong, durable hinge.

THREAD STRENGTH Threads can be molded or tapped into a plastic (see Fig. 5-51) [2]. Molded internal threads usually require some type of unscrewing or collapsing mechanism [11]. External threads can be molded either by splitting the mold halves or parting the line across the thread (see Fig. 5-51), if parting the line on the threads is permitted. With a split mold it is basically easier to design the mold and remove the threaded part from the mold during processing. The design of the threads requires control, to prevent excessive shear, resulting in stripping the threads when torqued, and also to limit hoop stresses, which can result in tensile failure. Although the mechanics of stress analysis for screw threads are readily available, the equations for them can be rather complicated. A simplified approximate equation is presented here [2].

356 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK INTERNAL THREAD

EXTERNAL THREAD

___1_ SPLIT CAVITY AT

~

THREADED CORE PIN IN TOOL MUST UNSCREW BEFORE MOLD OPENS.

PARTING LINE.

Figure 5-51. Examples of internal and external threads.

T TIGHTENING TORQUE

EFFECTIVE THREAD ENGAGEMENT LENGTH

NOMINAL THREAD OD PITCH DIA. OF THREAD

Figure 5-52. The torque-force relationships of screw threads.

As in Figure 5-52, the relationship between the torque on a screw and the axial force generated there is approximately

where Jl. = the dynamic coefficient of friction between the sliding surfaces [2]. The shear of threads A is approximately half that of the thread-engagement cylinder, or

STRUCTURAL DESIGN ANALYSIS 357

Therefore, the thread-stripping shear stress is

F

2F

A

7rdL

7=-=-

For standard 6O-degree unified threads there is a radial component of force that is about 60 percent of F. This force is spread over the thread-engagement cylinder, producing internal pressure in the boss. Therefore, the internal pressure in the boss is approximately 0.6F p=-7rdL

This pressure can then be used in the simple thin-walled hoop-stress equation

(1

P(d + 2t) to 0 b' . the boss. ---'--2-t--'tam the hoop stress 10 A major problem with thread torque-force equations is that the coefficient of friction varies significantly with the material and the surface finish. In addition, most published data for the coefficient of friction are produced at high speed and at the loads involved in the thread engagement. Based on these situations, the approximate equations are probably adequate. Furthermore, for initial design purposes it is probably worthwhile to pick an average value such as 0.15. Using this value, the previous equations can be further simplified as follows: The torque-force relationship becomes T

= 0.2 d For F = 5dT .

The thread-stripping, shear-stress equation is then

2F dL

10 T d 2L

7=--=-7r

7r

The pressure generated on the inside of the boss is calculated as O.2F T P=--=-

dL

d 2L

The hoop stress generated in the boss is thus 0.2F

T

(1=--=--

Ltd L t

if it is assumed that d + 2 t = 2d. In regard to screw threads, it is important that the following observations be clearly understood. First, the torque values are based on the coefficients of friction of the mating parts and can thus vary significantly. The use of any compatible lubricant that reduces friction will increase the shear and hoop stresses if the torque remains the same. Therefore, with lubricants, reduce the amount of allowable torque. Second, having high assembly torque to prevent vibrational loosening is frequently ineffective, since creep in the plastic will reduce the effective assembly torque even if

358 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

the fastener does not rotate. Using vibrationproof screws, lockwashers, locknuts, and thread-locking adhesives is usually a better alternative when loosening is considered a problem. Finally, self-tapping screws require additional torque to cut or form their threads. This torque can usually be added to the allowable safe-assembly torque, but for the first assembly only. The appropriate hole design for self-tapping screws is quite dependent on the material and screw design.

PIPE THREADS Properly designed plastic pipe threads usually require only a hand-tight assembling to effect a good seal, especially if a compatible sealant tape or compound is used. Many split bosses are the result of improper field installation, not faulty design. Some alternatives to pipe-thread designs are shown in Figure 5-53 [2]. An approximate equation for the hoop stress produced in a plastic boss having internal pipe threads is as follows: (J"

3T dL

= -t

As with the screw-thread equations in the previous section, assumptions need to be made regarding geometrics, so a 0.15 coefficient of friction is used. Once again, a torque reduction should be made if compatible thread lubricants are used during assembly. As usual, it is important to conduct tests on a final assembly, to ensure reliability and to relate the test results with the mathematical analysis. The final result is developing experience. The following recommendations should be incorporated when possible: 1) with metalto-plastic fittings, make the male thread of plastic so that it will be in compression; 2) use fluoroplastic tape and hand-tight fittings with female pipe threads; 3) when torque cannot be controlled in the field, consider employing an external or molded-in hoop ring (see Figure 5-53); and 4) do not encourage overtorquing by putting flats on the female part to provide a tightening area to grip; a textured surface should be sufficient if one is required. For complete, up-to-date information on thread strength and pipe threads, consult such qualified groups as the HCEPD Design/Technical Service of Hoechst Celanese for design recommendations [2].

POOR DESIGN

PRESSED-ON OR MOLDED-IN METAL HOOP RING

BETTER DESIGNS

grll~

METAL FITTING

PLASTIC _MALE PIPE _ _ _ _ _ _ _ _ THREAD

Figure 5·53. Examples of different designs for pipe threads.

""--..<::c...L....L...-L.J.._...L....."-L:...L..L-

STRUCTURAL DESIGN ANALYSIS 359

GEARS Gear design is one of the more complicated areas for designing with plastics, because the bending, shear, rolling, and sliding stresses all act upon a mechanism whose purpose is to transmit uniform motion and power. In this age of light weight and quieter operation, plastic gears have become increasingly important as a means of cutting cost, weight, and noise without significantly reducing performance. Because plastics are not as strong as steel, they must often perform far closer to their design limits than do metal gears. Although many plastic gear designs are derived from metal-gear technology, plastics demand special consideration, for instance to deal with heat buildup from hysteresis. The basic difference between metal and plastic in gear design is that designs for metal are based on the strength of a single tooth, whereas plastic shares the load among the various gear teeth to spread it out. Thus, in plastics the allowable stress for a specific number of cycles to failure increases as the tooth size decreases, to a pitch of about 48. Very little increase is seen above a 48 pitch, because of the effects of size and other considerations. The following guidelines for good gear design with TPs should be observed: I) determine the gears' conditions of service, such as temperature, load, velocity, space, and environment; 2) establish the short-term plastic properties as against the initial performance requirements; 3) compare the long-term property retention factor as opposed to the life of the gear; 4) using physical property data, calculate the stress levels caused by the various loads and speeds; and 5) then compare these calculated values with the allowable stress levels and redesign as needed to provide an adequate safety factor. Plastic gears fail for many of the same reasons as metal gears, including wear, scoring, plastic flow, pitting, fracture, creep, and fatigue. The causes of these failures are essentially the same. If a gear is lubricated, bending stress will be the most important parameter. Because nonlubricated gears may wear out before a tooth fails, contact stress is the prime factor in their design. Plastic gears usually have a full fillet radius at the tooth root, so they are not as prone to stress concentrations as are metal gears. The bending stress in engineering TPs is based on fatigue tests run at specific pitch-line velocities. A velocity factor should be used if the operating pitch-line velocity exceeds the test speed. Continuous lubrication can increase the allowable bending stress by a factor of at least 1.5. As with bending stresses, calculating surface-contact stress requires using a number of correction factors. For example, a velocity factor is used when the pitch-line velocity exceeds the test velocity. A correction factor is also used to account for changes in operating temperature, gear materials, and the pressure angle. Stall torque, another important factor, could be considerably more than the normal loading torque. At high speeds, plastic gears are also subject to hysteresis heating, which may be severe enough that they actually melt. Avoid this failure by designing the gear drive so that there is a favorable thermal balance between the heat that is generated and that which is removed by the inherent cooling processes. Reducing the rate of heat generation or increasing the rate of heat transfer will stabilize the gears' temperature so that they will run indefinitely until stopped by genuine fatigue failure. In such cases the wear resistance and durability of plastic gears makes them quite useful. Using unfilled engineering plastics usually gives them a fatigue life on an order of magnitude higher than metal gears. Hysteresis heating in plastics can be reduced by several methods, the usual one being to reduce the peak stress by increasing the tooth root area available for torque transmission. Another way to reduce stress on the teeth is by increasing the gear's diameter. Peak stress can also be reduced by geometrically repositioning it using various conventional gear theories.

360 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Using stiffer plastics to reduce hysteresis provides other improvements. For example, the higher-crystalline TPs like acetal and nylon can be further increased by processing techniques that can reinforce their stiffness by 25 to 50 percent. The most effective way to improve stiffness is through the use of fillers and reinforcements, particularly the highstiffness fibers seen in Chapter 6. Fillers and reinforcements are available that will also significantly increase heat transfer. The surrounding fluid, whether liquid or air, can have substantial cooling effects. A fluid like oil is at least ten times better at cooling than air. Agitating these mediums increases their cooling rates, particularly when employing a cooling heat exchanger.

GASKETS AND SEALS Different plastics are used to fabricate gaskets and seals. With many applications, usually the chemical or heat resistance will suggest the choice of the plastic, but often it can be below the optimum in stress relaxation. Polytetrafluoroethylene is used for being virtually inert and having outstanding high-temperature performance. PTFE is extremely vulnerable to creep and stress relaxation, but the many different filled grades let it be extensively used in severe environments. There is a great range of plastics and elastomers to meet widely varying service requirements and all the types of geometric shapes and stress relaxation characteristics. Different tests have been set up to develop industry standards and test evaluations that should be directly useful in their applications. For example, a gasket reI axometer applies a compressive stress to a flat annular specimen, similar to the way many gaskets are stressed in service (see ASTM F 38). This device is simple to operate, inexpensive, and capable of measuring the effects of such pertinent variables as stress relaxation in regard to time and environment. The ability of a gasket to seal against leakage resulting from the pressure of a confined fluid is d~ly related to retained stress. High initial stresses often are required to be able to handle high pressures. In this example, however, high stresses serve to increase the tendency of a gasket to creep, thus requiring stronger and more expensive construction. The usefulness of stress relaxation data to the designer is that they provide a guide in arriving at the usually required suitable design compromise, without overdesigning. These data show that the thinner the gasket, the less stress relaxation occurs. In some material evaluations, stress relaxation can be correlated with geometric variables by means of a shape factor, as follows: Annular area r Shape factor = Total lateral area = 4t where r = OD - ID and t = thickness. The trend of this factor is generally consistent with plastics' characteristics. However, this type of stress-relaxation information has to be interrelated with the individual behavior of the plastics, as derived from the relaxation-test data reviewed in Chapters 2 and 3.

GROMMETS AND NOISE To quiet a noise-generating mechanism, the first impulse is often to enclose it. Sometimes an enclosure is in fact the best solution, but not always. If it can be determined what is

STRUCTURAL DESIGN ANALYSIS 361

Figure 5-54. In addition to reducing noise, the injection-molded polyurethane (PUR) grommet (right) replaces five individual parts and saves time in assembling the lever linkage. During assembly it is snapped into a hole in the steel lever, then a grooved rod is inserted into the grommet. Intended to isolate vibration as well as connect metal parts, such a PUR grommet eliminates the hardening and cracking that used to shorten the life of the old assembly. This design might appear to be mechanically weaker than the cotter pin assembly, but it is at least as strong. The Hn. OD grommets can withstand a 2oo-lb. pull on the rod without undergoing pullout. In addition, the assembly can withstand a loo-lb. cyclic load (about 5,000 cycles at 60 cycles/min.) applied at 60 degrees off the rod axis at 300°F.

causing the noise, appropriate action can be taken to be more specific and provide a costeffective fix (see Fig. 5-54). In some cases the problem is caused by a component such as a stepper motor or gear set that does not produce objectionable noise by itself. The trouble typically develops because a small noise is transmitted to a metal frame or cabinet that then serves to amplify the sound. A cabinet that resonates can be quieted by damping its large flat areas so that they do not act like loudspeakers. Different approaches can be used, such as applying plasticfoam sound insulator or plastic panels, which have low damping characteristics. Various plastics have helped alleviate problems in all types of noise makers, including rotating systems and hammer actions. One popular approach is to use grommets when applicable (see Fig. 5-54). As one example, residential trash compacters that were objectionably noisy have been reduced to acceptable noise levels by redesigning them. Sound-absorbing grommets were used on the motors' bolt attachments and employed with better gears. Testing and all other types of equipment can take advantage of grommets or be redesigned to use plastic, or more of it, in the equipment. Grommets provide their greatest noise reduction through damping in the octave frequency bands above 500 Hz where the ear is most sensitive and sound most annoying [412].

IMPACT LOADS As seen in Chapter 3, loads are often applied abruptly, resulting in significant stress and strain increases. However, the elasticity of most TPs lets recovery usually be complete. Therefore, the steady-state stress and deflection of plastic parts can be considered identical to that of a part that is loaded gradually. However, when impact becomes severe, failure can result from it. Figure 5-55 shows the drop-weight test that can be used to estimate impact stress and impact deflection. This accepted method of testing has its own ster-

362 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Static Stress 0,

Static Deflection

y,

y,

H

Calculate static stress and deflection from placing weight on part.

C

Impact Stress ,--_-:-:-_ 0,

=

0,

y, = y,

DROP HEIGHT

+

V

1 +

~ y,

)

)

NOTES: - This is only an estimate, actual testing is essential. - Calculated impact stress will often exceed ultimate stress listed in data sheets for high impact materials.

Figure 5-55. A standard method for estimating impact stress and deflection by dropping a weight on to a test specimen or part.

eotypical situation where when the drop height is zero the stress and deflection double [2]. Many high-impact plastics can survive large deflections or strains during impact without suffering the permanent deformation or failure one might expect from the stress-strain curves of the plastics as measured at the standard loading rates reviewed in Chapter 3. Therefore, the calculated impact stress of successful parts will often appear to be unreasonably high. Recall that stress-strain behaviors are very different under rapid loading as compared to slow, steady loading conditions. Many plastics tend to have an exceptional capability of dissipating large amounts of mechanical energy when subject to impact.

THERMAL STRESSES When materials with different coefficients of linear thermal expansion (CLTE) are bolted, riveted, bonded, crimped, pressed, welded, or fastened together by any method that prevents relative movement between the parts, there is the potential for thermal stress. Most plastics, such as the unfilled commodity TPs, may have ten times the expansion rates of many nonplastic materials (see Fig. 5-56). Figure 5-57 examines the equations for thermal expansion that apply in various situations [2]. The basic relationship for the CLTE of a part is M.. = oLaT where

M..

the change in length

u

the coefficient of linear thermal expansion

L

the linear dimension under consideration (including hole diameters)

aT

=

the temperature change

STRUCTURAL DESIGN ANALYSIS 363 PLASTIC

CAST IRON PARTS ARE TIGHTLY BOLTED OR BONDED TO METAL PLASTIC ASSEMBLY GLASS WINDOW BONDED TO PLASTIC

ADHESIVE STEEL -~~:t-' BOLT

PLASTIC ----1..-:; FRAME

Figure 5-56. Typical assemblies of plastics to different materials in which thermal stresses can become a problem when proper design is not applied.

If a part is confined so that it cannot expand or contract, the strain induced by the temperature change is figured as ET

M-

= -

L

= culT

Calculating stress is done by multiplying the strain by the tensile modulus of the material at its operating temperature. With a plastic-to-metal attachment, a temperature change causes both to expand. A plastic will usually apply an insignificant load to a metal, but considerable stress will occur in the plastic. (Remember that there are plastics with a CLTE equal to or less than that of steel and aluminum and other such materials.) In this particular metal-to-plastic attachment example, the approximate thermal stress, CIT, in the plastic is CIT = (am - ap)EpaT where

am

=

the coefficient of thermal expansion of the metal

ap

=

coefficient of thermal expansion of the plastic

Ep

=

the tensile modulus of the plastic at the temperature in question

As the temperature increases, many plastics expand more than metals, but their modulus drops. This change produces a compressive load in the plastic part, which then often

FREE EXPANSION

L---I.~; ==i.... r--CHANGE IN ANY LINEAR DIMENSION AL=aLAT

L = LENGTH OF BEAM

Ad=adAT

d = DIAMETER OF HOLE

RESTRICTED EXPANSION

*==-L-~~ STRAIN

A(

STRESS

Or

aAT

=

EEr

ASSEMBL Y WITH METAL

STRESS IN PLASTIC Or = (am - a,) EpA T

RELA TlVE MOTION SHOULDER SCREW

METAL

RELA TlVE MOTION BETWEEN PARTS A T SHOULDER SCREW A rei

= (a p - "m)

LAT

Figure 5·57. Equations for thermal expansion in various situations. 364

STRUCTURAL DESIGN ANALYSIS 365

results in buckling. Conversely, as the temperature drops, plastics shrink more than metals and have an increase in their tensile modulus. The result could be a tensile rupture of the plastic product. In many assemblies the clearances around fasteners, the degrees of failure or yield in adhesives, and warpage or creep will tend to relieve the thermal stress. As with metalto-metal attachments having different CLTEs, proper design allows for such temperature changes, especially with large parts that might be subject to wide temperature variations (see Fig. 5-57). The relative motion, I:J. reI, between two attachment points of joined plastic-to-metal parts in which motion is allowed results in

STRUCTURAL FOAMS A density reduction of up to 40 percent can be obtained in SF parts [12]. The actual density reduction obtained will depend on the parts' thickness, the design's shape, and the flow distance during processing (see Chapter 7). Low-pressure SF parts can have characteristic surface splay patterns; however, the utilization of increased mold temperatures, increased injection rates, or grained mold surfaces will serve to minimize or hide this surface streaking. Finishing systems like sanding, filling, and painting for structural foam area readily available and have proved to be capable of completely eliminating surface splay. Figure 5-58 is an example of an SF part with a good cell structure and finish . High-pressure structural foam parts have generally been found to require little or no

Figure 5-58. Lightweight, durable covers for pickup truck beds are made of structural foam moldings of Prevex resin with a foam concentrate called Spectratech FM1776L, from Quantum Chemical Corp. This chemically activated blowing agent does not attack the injection-molding barrel and screw or the mold's surfaces. Furthermore, since there is very little postblow with the blowing agent, the parts can be painted within four hours of being molded.

366 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

postfinishing. Although high-pressure foam parts may exhibit visual splay, their surface smoothness is maintained and no sanding or filling is required. For structural foam, mold pressures of approximately 4.1 MPa (600 psi) are required, compared to typical pressures of 34.5 MPa (5,000 psi) and greater in injection molding. As a result, large, complicated parts of 22.7 kg (50 lb.) and up can be produced using multinozzle equipment, or up to 15.9 kg (35 lb.) with single-nozzle equipment and hot runner systems. Part size is, in fact, limited only by the size of existing equipment, and part complexity is limited only by tool design and a material's properties. Part cost can be kept in line through such advantages as part consolidation, function integration, and assembly labor savings. When an engineering plastic resin is used with the structural foam process, the material produced exhibits behavior that is predictable over a large range of temperatures. Its stress-strain curve shows a significantly linearly elastic region like other Hookean materials, up to its proportional limit. However, since thermoplastics are viscoelastic in nature, their properties are dependent on time, temperature, and the strain rate. The ratio of stress and strain is linear at low strain levels of 1 to 2 percent, and standard elastic design principles can thus be applied up to the elastic transition point. Large, complicated parts will usually require more critical structural evaluation to allow better prediction of their load-bearing capabilities under both static and dynamic conditions. Thus, predictions require careful analysis of the structural foam's cross-section. The composite cross-section of a structural foam part contains an ideal distribution of material, with a solid skin and a foamed core. The manufacturing process distributes a thick, almost impervious solid skin that is in the range of 25 percent of the overall wall thickness at the extreme locations from the neutral axis (see Fig. 5-59a). These are the regions where the maximum compressive and tensile stresses occur in bending. The simply supported beam has a load applied centrally. The upper skin goes into compression while the lower one goes into tension, and a uniform bending curve will develop (see Fig. 5-59b). However, this happens only if the shear rigidity or shear modulus of the cellular core is sufficiently high. If this is not the case, both skins will deflect as independent members, thus eliminating the load-bearing capability of the composite structure (see Fig. 5-59c). The fact that the cellular core provides resistance against shear and buckling stresses implies an ideal density for a given foam wall thickness. This optimum thickness is critically important in designing complex stressed parts. As a 6.4-mm (i-in.) wall, for example, both modified polyphenylene oxide and polycarbonate resin exhibit the best processing, properties, and cost-in the range of a 25 percent

(0)

(bl

Figure 5-59. A composite cross-section of a structural foam part.

(e)

STRUCTURAL DESIGN ANALYSIS 367

Figure 5-60. A cross-section of a solid material.

~-------b,--------~

Figure 5-61. A cross-section of a sandwich structure.

weight reduction. Laboratory tests show that with thinner walls about 4.0 rom (0.157 in.), this ideal weight reduction decreases to 15 percent. When the wall thickness reaches approximately ~.9 mm (0.350 in.), the weight can be reduced by 30 percent. However the structural foam cross-section is analyzed, its composite nature still results in a twofold increase in rigidity, compared to an equivalent amount of solid plastic, since rigidity is a cubic function of wall thickness. This increased rigidity allows large structural parts to be designed with only minimal distortion and deflection when stressed within the recommended values for a particular foamable resin. Depending on the required analysis, the moment of inertia can be evaluated three ways. In the first approach, the cross-section is considered to be solid material (see Fig. 560). The moment of inertia, Ix, is then equal to Ix = bh3/12 where b = the width and h = the height. This commonly used approach provides acceptable accuracy when the load-bearing requirements are minimal-for example, in the case of simple stressesand when time and cost constraints prevent more exact analysis. The second approach ignores the strength contribution of the core and assumes that the two outer skins provide all the rigidity (Fig. 5-61). The equivalent moment of inertia is then equal to Ix = b(h3 - hI)/12. This formula results in conservative accuracy, since the core does contribute to the stress-absorbing function. It also adds a built-in safety factor to a loaded beam or plate element when safety is a concern. A third method is to convert the structural foam cross-section to an equivalent I-beam section of solid resin material (see Fig. 5-62). The moment of inertia is then formulated as

368 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 5-62. A cross-section of an '-beam. where b l = b(Ec)/(Es), Ec = the modulus of the core, E. = the modulus of the skin, ts = the thickness of the skin, and hI = the height of the equivalent web (core). This approach may be necessary where operating conditions require stringent load-bearing capabilities without resorting to overdesign and thus unnecessary costs. Such an analysis produces maximum accuracy and would thus be suitable for finite-element analysis on complex parts. However, the one difficulty with this method is that the core modulus and the as-molded variations in skin thicknesses cannot be accurately measured.

STRUCTURAL SANDWICHES In the usual construction practice a structural sandwich is a special case of a laminate, whether flat, curved, or otherwise, in which two thin facings of relatively stiff, hard, dense, strong material are bonded to a thick core of foam, honeycomb construction, or the like of a relatively lightweight material that is considerably less dense, stiff, and strong than the facings. Structural sandwiches can be all plastics, all metals, or some combination of plastic and metal. An example of a reinforced plastic with a foam core was shown in Figure 1-12. With this geometry and relationship of mechanical properties, facings are subjected to almost all the stresses in transverse bending or axial loading. The geometry of the arrangement provides for high stiffness combined with lightness, because the stiff facings are at a maximum distance from the neutral axis, similar to the flanges of an I-beam. The continuous core takes the place of the web of an I-beam or box beam, absorbs most of the shear, and stabilizes the thin facings against buckling or wrinkling under compressive stresses. The bond between the core and its facings must resist shear and any transverse tensile stresses set up as the facings tend to wrinkle or pull away from the core [14].

Stiffness For an isotropic material with a modulus of elasticity E, the bending stiffness factor, EI, of a rectangular beam b wide and h deep is

In a rectangular structural sandwich with the same dimensions just given whose facings

STRUCTURAL DESIGN ANALYSIS 369

and core have moduli of elasticity Ef and Eco respectively, and a core thickness C, the bending stiffness factor EI is

This equation is exact if the facings are of equal thickness, and approximate if they are not, but the approximation is close if the facings are thin relative to the core. If, as is usually the case, Ec is much smaller than Ef , the last tenn in the equation can be ignored. For asymmetrical sandwiches with different materials or different thicknesses in their facings or both, the more general equation for I.EI may be used. In many isotropic materials the shear modulus G is high compared to the elastic modulus E, and the shear distortion of a transversely loaded beam is so small that it can be neglected in calculating deflection. In a structural sandwich the core shear modulus Gc is usually so much smaller than Ef of the facings that the shear distortion of the core may be large and therefore contribute significantly to the deflection of a transversely loaded beam. The total deflection of a beam is thus composed of two factors: the deflection caused by the bending moment alone, and the deflection caused by shear, that is, 8 = 8m + 8s where 8 = total deflection, 8m = moment deflection, and 8s = shear deflection. Under transverse loading, bending moment deflection is proportional to the load and the cube of the span and inversely proportional to the stiffness factor, EI. Shear deflection is proportional to the load and span and inversely proportional to shear stiffness factor N, whose value for symmetrical sandwiches is N

=

(h

+

2

c)bG

c

where Gc = the core shear modulus. The total deflection may therefore be written

The values of Km and Ks depend on the type of load. The values for several typical loading conditions are given as shown in Table 5-5.

Stresses in Sandwich Beams The familiar equation for stresses in an isotropic beam subjected to bending, M I

(}'=-1:: y

must be modified for sandwiches to the fonn = (}' y

MEyY EI

370 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

where y = the distance from the neutral axis to the fiber at y, E y = the elastic modulus of the fiber at y, and EI = the stiffness factor. For a symmetrical sandwich the stress in the outennost facing fiber is found by setting

y Ey

= hl2 = Ef

and the stress in the outennost core fiber by setting

y

Ey

= C/2 = Ee

The mean stress in the facings of a symmetrical sandwich can be found from 0'

2M = ---bt(h + c)

where t = the facing thickness. Similarly, the general equation for the shear stresses in a laminate (see the preceding section),

VQ

T=-

bEl

can be used for any sandwich. For a symmetrical sandwich the value of T can be closely approximated by T

2V = ---b(h + c)

Table 5-5. Typical Loading Conditions Loading

Beam Ends

Uniformly distributed Uniformly distributed Concentrated at midspan Concentrated at midspan Concentrated at outer quarter points Concentrated at outer quarter points Uniformly distributed

Both simply supported Both clamped Both simply supported Both clamped Both simply supported

Midspan Midspan Midspan Midspan Midspan

Both simply supported Cantilever, 1 free, 1 clamped Cantilever, 1 free, 1 clamped

Concentrated at free end

Deflections at

Km

K.

5/384 1/384 1148 11192 lln68

118 118 114 114 1/8

Load point

1196

118

Free end

118

1/2

Free end

113

STRUCTURAL DESIGN ANALYSIS 371

Axially Loaded Sandwich Edge-loaded sandwiches such as columns and walls are subject to failure by overstressing the facings or core, or by the buckling of the member as a whole. Direct stresses in the facings and core can be calculated by assuming that their strains are equal, so that

where

(J'f

= the total load = the facing stress

(J'c

= the core stress

Af

=

the cross-sectional area of facings

Ac

=

the cross-sectional area of core

p

Usually the elastic modulus, En of the core is so small that the core carries little of the total load, in which case the equation can be simplified by ignoring its last term, so that for a sandwich b wide with facings t thick, P = 2(J'tht. The column-buckling load of a sandwich L long simply supported at the ends is given by

This variation on the Euler equation takes into account the low shear stiffness of the core. For wall panels held in line along their vertical edges, an approximate buckling formula is

provided the length, L, of the panel is at least as great as its width, b, and that the second term in the bracket of the denominator is not greater than unity [14]. ENERGY AND MOTION CONTROL

Thermoplastic elastomer (TPE) components (see Chapter 6) are frequently subject to dynamic loads where energy and motion controls are required (see Fig. 5-63). The products involved range from sporting goods to home appliances to automobiles to buildings to bridges to boats to aircraft to spacecraft [413-17]. The use of bonded elastomers for energy and motion control in construction is discussed in Chapter 6 to give a technical

372 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Cover

Elastomer

Steel Reinforcing Plates Internal Elastomers Layers Figure 5-63. A section through a high-load elastomeric building isolator showing the elastomer layers to ensure horizontal flexibility. Steel reinforcing plates give rigidity for vertical loads. This isolator resists wind loads elastically without perceptible movement, yields under earthquake loads, and deforms plasticly dampening side-to-side vibrations.

examination of how to analyze and evaluate it, based on work by J. H. Bucksbee, the Manager of Product Design at the Industrial Products Division of the Lord Corporation [413]. Anyone in a building, on the highway, on a rapid transit vehicle, or on a ship has a bonded elastomer working for them. In a building, it controls vibration and noise from motors and engines that is generated to the building itself. For rapid transit it supports the rail and the vehicle, thus reducing noise and vibration to adjacent buildings. For ships, bonded elastomers absorb their berthing energies, with single units being as large as 3 m (9.9 ft.) high and weighing 19 tons. For all these applications, elastomers are used either in shear, compression, tension, torsion, buckling, or a combination of two or more uses, depending on the needs of the specific application. Elastomers are used in the construction industry to control noise and shock and vibration and are used in many applications strictly to accommodate motion. In any application of elastomeric materials, the method of applying the elastomer can either help or hinder the desired end result. Basically, elastomers are used in situations of shear, compression, buckling, and bulk. The particular application will dictate which would be best. Consider, for example, a berthing vessel, a structure that has to be designed to withstand the energy developed by the vessel. The more rigid the system, the higher the reactive forces must be to absorb the vessel's kinetic energy. The area under the structure's load as against its deflection response curve (that is, of the energy absorbed) is typical to that shown in Figure 5-64 [413]. For economic reasons, designers typically reduce the mass of a structure, but doing so reduces the lateral reaction forces that the structure can withstand. This happens as the structure is allowed to deflect more or an energy-absorbing device is applied. An elastomer is ideal in such an environment because it will not corrode. Metal components can thus be totally encapsulated in an elastomer and then bonded to all-metal surfaces. The elastomer can be used in conditions of shear, compression, or buckling. In examining

STRUCTURAL DESIGN ANALYSIS 373

the load-deflection characteristics of these three systems, note that the one that results in the lowest reaction force generally also produces the lowest-cost structure. Figures 5-65, 5-66, and 5-67 show six results that could be obtained, compared to an ideal hydraulic system with 100 percent energy efficiency. Figure 5-65 is of a typical shear curve. The energy capacity is approximately 50 percent efficient, requiring 100 percent more deflection or load, if the deflection or load of a 100percent-efficient curve is required. Figure 5-66 is of a typical compression curve. It shows an energy capacity approximately 35 percent efficient, requiring 300 percent higher deflection or a 350 percent higher load if the deflection or load of a 100-percent-efficient curve is required. Figure 5-67 shows a typical buckling-volume curve. Here the energy capacity is approximately 75 percent efficient, requiring a 25 percent higher deflection or a 20 percent higher load if the deflection or load of a loo-percent-efficient curve is required. The buckling column was selected because it produces the lowest reaction load and the lowest deflection (deflection controls the projection of the berthing system out from the structure). A typical design is shown in Figure 5-68. The energy capacities and weight vary here from 1,900 to 657,000 joules and from 11.3 to 17,690 kg (24.8 to 38,918 lb.). When designing to support structures and allow the horizontal input of an earthquake or to allow the structural movement needed on a structure such as a bridge pier, the vertical and horizontal stiffnesses must be calculated, then a system can be designed. Take, for example, a rectangular elastomeric section with a length of 762 cm, a width of 508 cm, and a thickness of 508 cm (305 x 203 x 203 in.). Table 5-6 lists the formulas used to calculate the respective compression and shear stiffnesses for these data. For the elastomeric section of the example use the following formula:

Kc =

(kc)(762 x 508)(EJ 508

= 762kcEc

= (ks)(762 x

K

508)(Gs)

508

s

= 762ksGs

The calculations for kc and ks adjust for such design parameters as strain, bulk compression, and bending by the elastomer section, as developed over many years of sample testing. A way to forego the pain of getting there is to let ks = .98 and kc = 1.0, using a .69 MPa shear-modulus elastomer as follows:

Ks = 747G s For compression, the shape factor (SF), which is the projected load area of the product divided by the elastomer area that is free to move (known in the industry as the bulge area, or BA), is the major design parameter. For this example the shape factor is calculated thus: Load area (762)(508) = = 0.3 Bulge area (762 + 508)(2)(508) Using Figure 5-69 at a SF = .3 and a Gs = .69 MPa, an Ec Therefore, Kc = 1.8 MN/m and Ks = .5 MN/m.

= 2.42 MPa is obtained.

F

Maximum Lateral Reaction Force

- - - Lateral Deflection

- t - - - - Energy Absorbed

L.....:---~-6

Figure 5·64. Typical load deflection that is characteristic of a marine structure.

F F

c o

;;

Force

u CD ;: CD

C

Energy

Same Energy

CD

E

as

en

~------------~~6

100% Higher Force F

Same Force ------------------

Same Energy 100% Higher Deflection Figure 5·65. Elastomeric shear energy capacity as compared to a 100 percent efficient curve.

374

STRUCTURAL DESIGN ANALYSIS 375

F

I

1

1 1 1 1 1 1

ci 0 :::1 ~I

F

:;:::1 ~I

Force

cpl

EI

cl 0

Energy

:::1 (,)1 :!I

~I

1

I I

~I

c3

c3 350% Higher Load

F

Same Force -----------------------

300% Higher Deflection Figure 5·66. Elastomeric compression energy capacity as compared to a 100 percent efficient

curve.

Applying a 20 percent maximum compression strain to the product results in a maximum compression deflection of 102 cm (41 in.). This allows a maximum compression load of only 19 kg (42 lb.), hardly sufficient to support a building or a bridge. Given that shear stiffness Ks cannot change, the sole remaining option is to change the compression stiffness, by adjusting the shape factor. Designing the product as two units each 762 x 508 x 254 cm (305 x 203 x 102 in.) thick will not change the shear characteristic, but it does change the shape factor to .060. According to Figure 5-69, Ee = 3.03 MPa and Ke becomes 4.6 MN/m per section. With two sections in series the spring rate is 2.3 MN/m, which now allows for 24.3 kg of compression load. Dividing each of these sections into a total of four sections each 762 x 508 x 127 cm (305 x 203 x 51 in.) thick yields a shape factor of 1.2 and an Ee equal to 151.7 MPa. An

F

F

c::

l.g

Force

Same Energy

Energy

~------------~~-6 F

I~ I~ Ie

I ell IE I~ I

20% Higher Force

Same Force

Same Energy

25% High Deflection Figure 5-67. Elastomeric buckling energy capacity as compared to a lOO-percent-efficient curve.

Figure 5-68. A buckling column fendering system.

376

STRUCTURAL DESIGN ANALYSIS 377

!

4

3

8 1

6

3

2

4

8 10

6

1000

800

' L ,,~

~ ' /~

600

I

Ij

400

,

300

J If

I

I

I

compression modulus

100

E<

L

LJ

60

3~

hardness -IRHO

,,~

II""'"

,,'f

_~ 75 -

,

>-- ;;iii 65 ~

6

4

~~

3

-r ~

."",

."",

55

."..

~

~

,..-1

45 -'"

35 ~

I.ttt' II""'"

~

~

"J'/ L

I

L

~

I

'f

//

I

I

I

'-

,

I

~

I

I

,

:/ I I

I

L

I I

~

~~

C0~ «:...;

5

~ --

~~ ~~ ,A~~

~RCt F~£~ 8 I

~~ ~

I

'f ,

1/ / I. If

~~

, I

J J

/

'f

~

20 r- nominal

10

~

/

Ih " ", , r/ I

J '~/~

i/ 1 J. 1(1 / 'I IfIfJ'

/

40

"J' /. i(Lr/

/ ILI/l ''fI

, ", ,

MN/m'

80

-

,,-

~

1

, .,

~ I""" ~ ~ .".;. ",..

/ 1/ "" ~

Shape factor S

200

-

2~ ~........ ~

~

<' I 1I I

shape factor

=

Ie S

=

loaded area force free area LB 2t (L + BJ

10)( shear modulus G I

I

Figure 5-69. A compression modulus (EA ) versus a shape factor (S) for five different elastomer stiffnesses.

individual section Kc will be 46 MN/m, with a series of four being 115 MN/m, allowing a compression load of approximately 120 kg. It always maintains a shear spring rate of approximately .5 MN/m . This is the basic design philosophy for obtaining high-compression loads while maintaining the soft shear stiffnesses needed for seismic considerations or for thermal expansion and contraction. Continued thinning of the individual elastomer sections will drive the compression load-carrying capacity upward.

378 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 5-6. Data Required for Formulas Direction

Compression (Kc)

(kc)(L.A.)(G s)

Formula Variable Variable Variable Variable Variable

Shear (Ks)

k L.A. Ec Gs t

Factor of geometry Load area Compression modulus Elastomer thickness

t Factor of geometry Load area

Shear modulus Elastomer thickness

In the rapid transit industry, wherever there is an elevated structure or subway tunnel, noise and vibration caused by the vehicles can generate unrest among those living along the route. There are three areas that can be adjusted to reduce annoying frequencies: the vehicles' suspension system; the trackbed, including the tie-to-rail interface and the floating slab; and various acoustic barriers. For the vehicle suspension system and trackbed techniques, good elastomeric product design and application are generally sufficient. The product design requires including the design considerations mentioned previously of the compression and shear curve and shape factor, but also introduces new combinations of compression and shear. These occur either by shear and compression in planes 90 degrees to each other or shear with compression, as in seismic and thermo designs, only installed at angles to the horizontal. Figure 5-70 shows a typical installation of a suspension system for a rapid-transit vehicle. The angle of installation (6) is normally 75 to 800 and the chevron angle (0:) is normally 100 to 1200 , depending on the desired vertical to fore-aft to lateral characteristics. Figure 5-71 depicts a floating-slab concept with elastomeric support below it and directfixation elastomeric rail fasteners between the slab and the running rail. Figure 5-72 schematically displays the elastic iterations of a rapid-transit line, not an easy system to analyze dynamically.

Figure 5-70. A typical rapid-transit suspension system using a chevron configuration.

~ .'~'

.. '»

Floating Slab

Resilient Support Pads

Resilient Perimeter Isolation

Figure 5-71. A typical rapid-transit cross-section showing the elastomeric rail fasteners and floatings-slab supports.

Car Body & Bolster Secondary Suspension ~----~------r-~

-

Truck Frame

Wheel/ Rail Interface t"'1J...........- , - - - - - y - - - - '

Rai I Rail Fastener Floating Slab

Slab Suspension

Figure 5-72. A schematic interface showing the elastic components of a rapid-transit system. 379

380 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

During acoustic testing at a major transit authority's elevated structure it came out that the softer the compression and shear stiffnesses of the direct-fixation elastomeric rail fastener, the better the acoustic performance. Figure 5-73 shows the noise-level reduction available from using soft elastomeric rail fasteners instead of stiff ones. The soft fastener has approximately 50 percent of the stiff fastener's spring rate. This figure shows the average wayside noise levels at 30 m (100ft.) from an elevated structure for train operation on soft fasteners compared to stiff ones. For vibration, Figure 5-74 shows the average vibration of the bottom plate of a steel box girder for train operation on soft fasteners as opposed to stiff ones. For each figure the data are plotted for two different car manufacturers. The performance difference is the influence of the cars' suspension systems' parameters on the different elastomeric rail fasteners' stiffnesses. Figure 5-75 shows clearly that there are definitely two populations, two different car manufacturers. Therefore, a system that performs well for one authority may not offer the same performance to another one. We have now looked at the more common types of single-axis load-deflection characteristics possible with elastomeric products, as well as various methods to change the stiffness in one direction while maintaining an initial stiffness in a plane that is 90 degrees from the reference plane. The angular considerations in different directions can now yield an unlimited number of design-configuration options. The next six typical concepts will allow motion through an elastomer in three or more directions. First, Figure 5-76 would be a typical design for seismic concerns and bridgebearing pads. It is capable of supporting high compressive loads while allowing for soft lateral (shear) and torsional characteristics. Although the part shown is a circular unit, square and rectangular products are more common in the construction industry. Figure 5-77 shows another high shape factor design, which in this case takes the elastomer in compression (that is, radially), with the shear modes being axial and torsional.

m

"0

+10

> Q) Q)

0

I/) I/)

c...

-10

0

Q)

> as

=::::

,'"

.... I<:f

y~

"5

:: ::

:: -:

\

~

'VX

........

~.

...

~

...

"X)a-I ""li:r

~~- !E--x",*,;

?"...

-

-20

=-

::

o R Trains >+-----l< B Trains

-

:0=

F

'ii a::

F I-

~

,

I I 16

31. 5

63

I,

, I I I 125 250 sao

11

I l

, I

I

Octave Band Center Frequency -

, ,::-=

I I

2K

1K

:: :: -: :: : :: -: : : :.

-

0

::

-

~:.

::

"0 C ::l (f)

::

::

=-

::l

~

F ::

~

Q)

::

f:

'ii

...J

~

f:

4K

Hz

Figure 5-73. Average wayside noise levels at 100 ft. from an elevated structure for train operations on relatively soft rather than stiff fasteners.

+10 r:D

'tI

I

Qj

0

..J C 0 :0:

-10

> CI) co a..

.Q

:> CI)

> -20 :0: co

Ie

iJ

125

250

R Trains

Qj

a:

8

16

31. 5

63

Octave Band Center Frequency -

500

1K

Hz

Figure 5-74. Average vibration of the bottom plate of a steel box girder for train operations on soft fasteners as against stiff ones.

co

100

~

:::l 0 N

W

a:

90

r:D

'tI

Qj

80

>

CI)

..J CI)

a.. :::l

70

en en CI) a..

~

'tI

c

60

:::l

0

en 'tI

c co

50

-

40

r:D

CI)

> co

U

Soft Fasteners Soft Fasteners -

R Trains B Trains

e----,;] Stiff Fasteners -

R Trains B Trains

0

0

)(

)(

&---.....t!..

Stiff Fasteners -

0

M

16

31. 5

63

125

250

500

lK

Octave Band Center Frequency -

2K

4K

Hz

Figure 5-75. Noise-level readings indicating the results of systems from two different car manufacturers. 381

FAxial

8 Horizontal

Figure 5·76. Axial compression with two shear modes.

F

.0 Radial

¢

8 Torsional

Figure 5·77. Radial compression with two shear modes. 382

Axial

STRUCTURAL DESIGN ANALYSIS 383

--

-~ & Axial

Figure 5·78. Radial compression with three shear modes.

Figure 5·78 is similar to Figure 5-77 except that its metal components have a spherical contour, allowing for torsion about the center line and radial axis. For each of these directions the radial-load deflection characteristic is quite stiff. The axial characteristic is indeed stiffer than the shear in the previous configuration. The torsion about the axial axis is also similar, but the torsion about the radial axial axis is soft. Figure 5·79 shows a product similar to one used in the rapid-transit and railroad industry that is, however, of a conical configuration as opposed to the previously seen chevron configuration from Figure 5-70. This product exhibits high axial and radial characteristics while maintaining soft torsional characteristics. Figure 5-80 shows another variation of Figure 5-79, which produces high radial and axialload·deflection characteristics but maintains a soft torsional nature. This design is basically a combination of the flat, high shape-factor part described in Figure 5·76, and the circular part described in Figure 5·77. Figure 5·81 shows a spherical bearing, a configuration that allows for high compressive loads in the axial direction while maintaining soft torsion about the axial axis and torsion perpendicular to that axis. This configuration is a combination of Figures 5·78 and 5·79. For auxiliary generators and compressors any of these configurations would be viable. However, each individual application has its own design requirements. Figure 5·82 shows four different applications of a mounting system on an auxiliary engine, any of which would be acceptable, depending upon the design analysis.

FAxial

F Radial

Torsional

Figure 5-79. Radial and axial compression with one shear mode.

F Axial

Figure 5-80. Radial and axial compression with one shear mode and high deflection. 384

STRUCTURAL DESIGN ANALYSIS 385

F Axial

~ ~ --

------

¢

Pitch

Figure 5-81. Axial compression with three shear modes.

One of the parameters to consider when applying an elastic suspension system to an energy-producing device is the degree of motion that will be acceptable to the installation. Figure 5-83 illustrates what can happen when the elastic center (that is, the theoretical point which all the elastic mounting characteristics concentrate) is below the center of gravity. The methods used to stabilize such a system are the focalized concept, as seen in Figures 5-82b and 5-83b. For this particular system the elastic center is now projected up through the angle of inclination of the mounts and the center of gravity, decoupling the motion and resulting in translational motion only. Another method is to mount the system so that its mountings are vertical and its elastic center coincides somewhat with the center of gravity but is still above the center of gravity, as in Figure 5-83c. One alternative to Figure 5-83a that would result in high resistance to rotational motion would be to support the system as shown in Figure 5-83d, by installing stabilizer mountings above the center of gravity to reduce lateral motion. The performance of elastomers is of major interest and concern to the design engineer. The readily available data concern the tensile--elongation factor, the compression set, results from durometer tests, and information on oil resistance, heat aging, and the static modulus. In designing for a given environment, certain information makes the designer's job easier and the actual results closer to that predicted: Dynamic modulus at various strains, frequencies, and temperatures. Ozone resistance at different concentration levels. Loss factor at various strains, frequencies, and temperatures. Fatigue of various shape factors and cyclic strains and temperatures. Effects of different ingredients such as carbon black. Drift and set characteristics at various initial strains and temperatures. Electrical resistance.

b

(,1,11"\

r r-U(

t' I

\...!

I

d

c

Figure 5-82. Four different applications of a mounting system on an auxiliary engine.

b

a

+-1-+ -

....

Cenler 01 G"vlty

Cente, of G,avltyand SUlpenllon Elaatle Cente, Coincide MounUngl Located at B...

Mountlngl SUlpenllon Elalt c Cenler

c

d Cente,ol Gravitvand SUlpen. on EtuUc Center Coincide

MounUngs Located In Plane Thru Cente, of G,avlty

Figure 5-83. An elastic center used to stabilize rocking motion. 386

+

Center of G,.vlty

SUlpenlon EluUc Center

STRUCTURAL DESIGN ANALYSIS 387

These types of data are normally generated at the designer's facility with in-housedeveloped test equipment and procedures. FAILURE ANALYSIS

The process of analyzing designs includes the modes of failure analysis that were discussed in Chapters 2, 4, and earlier [1, 2, 10-12]. At an early stage the designer should try to anticipate how and where a design is most likely to fail. The most common conditions of possible failure are elastic deflection, inelastic deformation, and fracture. During elastic deflection a part fails because the loads applied produce too large a deflection. In deformation, if it is too great it may cause other parts of an assembly to become misaligned or overstressed. Dynamic deflection can produce unacceptable vibration and noise. When a stable structure is required, the amount of deflection can set the limit for buckling loads. Because many plastics are relatively flexible, analysis should consider how much deflection might result from the loadings and elevated temperatures any given part might see in service. The equations for predicting such deflections should use the modulus of the material; its tensile strength is not pertinent. Usually, the most effective way to reduce deflection is to stiffen a part's wall by changing its cross-section. Inelastic deformation causes part failure arising out of a massive realignment of the molecular structure. A part undergoing inelastic deformation does not return to its original state when its load is removed. It should be remembered that there are plastics that are sensitive to this situation and others that are not. The existence of an elevated temperature, with or without long-term or continuous loading, would suggest the possibility that a material might exceed its elastic limits. As explained earlier concerning momentary loading, the properties to consider are the proportionallimit and the maximum shear stress. The presence of fracture reflects a load that exceeds the strength of the design. The load may occur suddenly, such as upon impact, or at a low temperature, which will reduce the elongation of the material. A failure may develop slowly, from a steady, high load applied over a long time (creep rupture) or from the gradual growth of a crack from fatigue. If fracture is the expected mode of failure, analysis should examine the greatest principal stresses involved. DIMENSIONAL TOLERANCES

The specific dimensions that can be obtained on a finished, processed plastic product basically depend on the performance and control of the plastic material, the fabrication process and, in many cases, upon properly integrating the materials with the process. In tum, a number of variable characteristics exist with the material itself, as described in Chapters 2 through 6, and the process (Chapters 7 and 8). Unfortunately, many designers tend to consider dimensional tolerances on plastic products to be complex, unpredictable, and not susceptible to control. This is simply not true, though they can be complex. Plastics are no different in this respect than other materials. If steel, aluminum, and ceramics were to be made into complex shapes but no prior history on their behavior during processing existed, a period of trial and error would be required to ensure their meeting the required measurements. If relevant processing information or experience did exist, it would be possible for these metallic products to meet the requirements with the first part produced. This same situation exists with plastics. To be successful with this material requires experience with their melt behavior, melt-flow behavior during pro-

388 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

cessing, and the process controls needed to ensure meeting the dimensions that can be achieved in a complete processing operation. Based on the plastic to be used and the equipment available for processing, certain combinations will make it possible to meet extremely tight tolerances, but others will perform with no tight tolerances or any degree of repeatability [1-14, 40-42, 50-72]. Fortunately, there are many different types of plastics that can provide all kinds of properties, including specific dimensional tolerances. It can thus be said that the real problem is not with the different plastics or processes but rather with the designer, who requires knowledge and experience to create products to meet the desired requirements. The designer with no knowledge or experience has to become familiar with the plasticdesign concepts expressed throughout this book and work with capable people such as the suppliers of plastic materials. Some plastics, such as the TSs and in particular the TS-RP composites, can produce parts with exceptionally tight tolerances. In the compression molding of relatively thin to thick and complex shapes, tolerances can be held to less than 0.001 in. or to even zero, as can also be done using hand layup fabricating techniques. At the other extreme are the unfilled, unreinforced extruded TPs. Generally, unless a very thin uniform wall is to be extruded, it is impossible to hold to such tight tolerances as just given. The thicker and more complex an extruded shape is, the more difficult it becomes to meet tight tolerances. (This situation is also true with most other nonplastic materials.) What is important is to determine the tolerances that can be met and then design around them.

PLASTICS To maximize control in setting tolerances there is usually a minimum and a maximum limit on thickness, based on the process to be used (see Tables 5-7, 5-8, and 5-9). Each resin has its own range that depends on its chemical structure and melt-processing characteristics. Outside these ranges, melts are usually uncontrollable. Any dimensions and tolerances are theoretically possible, but they could result in requiring special processing equipment, which usually becomes expensive. There are of course products that require and use special equipment. One factor in tolerances is shrinkage. Generally, shrinkage is the difference between the dimensions of a fabricated part at room temperature and the cooled part, checked usually twelve to twenty-four hours after fabrication. Having an elapsed time is necessary for many plastics, particularly the commodity TPs, to allow parts to complete their inherent shrinkage behavior. The extent of this postshrinkage can be near zero for certain plastics or may vary considerably. Shrinkage can also be dependent on such climatic conditions as temperature and humidity, under which the part will exist in service, as well as its conditions of storage. Plastics suppliers can provide the initial information on shrinkage that has to be added to the design shape and will influence its processing. The shrinkage and postshrinkage will depend on the types of plastics and fillers. Compared to the TPs, the TSs generally have less filler. The type and amount of filler, such as its reinforcement, can significantly reduce shrinkage and tolerances (see Figs. 5-84 and 5-85). Another influence on dimensions and tolerances involves the coefficient of linear thermal expansion or contraction. This CLTE value usually has to be determined at the part's operating temperature. (See the CLTE information in Chapter 2 for additional details.) So it is important to include in the design specifications the operating temperature conditions, to specify a plastic that will do the job. Plastics can provide all the extremes in

STRUCTURAL DESIGN ANALYSIS 389 3.0

!

Unr" nfo rced IIrad es

!

Q ~~----~----~-4 6 2 WIUth,ckness. mm

1.5

o L.--+____

~---_...--

6

4

2

Unt"nforced nvion 6 (3% mOlstur. 1

Wall thlckn ..... mm

1"7'1 Glau r"nforced

Unrllnlorced PB T

~ nvlon

6 13% mOIIN", '

30% glass

Unr" n lorced POlvcarbOnatl

",nforced paT 30% glau reInforced

polvcarbonate

Figure S-84. How wall thickness affects shrinkage. 200

160 E

E :::I. ,.; co

120

""

~'\

...

= 80

~

1:

IJJ

.0

o

""

"'-PBT

I

""~

NYIOn616~

~'o

5

pc----, '---10

--- -

~ '--

-

15

20

I--

r--

25

30

35

.0

Glass, %

Figure S-8S. How glass reinforcement affects mold shrinkage.

CLTEs, including graphite-filled compounds that could work in reverse. Upon heating, they contract rather than expand, and vice-versa. To assist the designer a Society of the Plastics Industry (SPI) bulletin is available that specifies the limits for certain dimensions. Each material supplier converts these data to suit their specific plastics. Figure 5-86 shows this information. This type of information is intended to give the designer a guide for tolerances that are to be shown on the drawings; these tolerances include variations in part manufacture and some degree of variation in the tooling for TPs and TSs. Figure 5-86 should not be considered a hard-and-fast set of rules for all conditions but

390 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Engineering and Technical Standards

STANDARDS AND PRACTICES OF PLASTICS CUSTOM MOLDERS

PO L Y C AR80NATE

.t

NOTE ~ omnlC'ft:l,,1 " .. Jun ,ho .. n htlnw fcptnrnl comnuWl pwduc:(~.n Iuk:t I'k.t1o II ,he mml cconom 1 he fIne ".lhxs rcp,c"-C'nl cI,~r W\t:'lnce.i Ihtll an be held bur ~I .. ,re:;;u~, COlI

Draw,n.

DImC'nJionl

Code

A

=

I 1

(hochn)

f:::-: -

Dlarn
f-- l f-- '

B - D
-

~

I-

1-

•

(_ Note 11) t -

r- ....

1-' C = Helchl C- ote Il)

.,

1'\ -

-

,-

,-

D- llotlom win be<: Ie .)

F - Ihle lle Oi,meler CKe 01< II)

G = IMe SIl' o.l"h (ICC

OI.,S)

-Dr." AIIowan«:

0

'0011$

012"0020 o 2SO 100.500

-

nillneD

C-

Ok

14)

Thrud Sur

(d.,,)

ConanlrtC"', C," Ole 14) Fille ... RIb$. Cornc .. Cscc Ole 16) Surl... Fan,'"

Color Sllbthly

2

00)

002

at Over

00)

002

025010 O. 00

00)

o JODlo I 000

004

002 00)

.0

1>\0,

10 000 10 0 2SO - -2 r-

0000 10)000

OIlS

106000

001

)

-

00"") 004

2B

EJ.1c.m~1

1.0.

]A

(TI

005

-R) -- --

r CIH

- --)

f-

OIS

'.1 -

-

'-

-

-

r ~

I-

015

-

-

1- -

I-

--

-

I

~IJj

I

1--

-E

./1 --1 I I-F

F""';

n.c.e

- --

A

..,...

REFER I-

r-r-

'-

,-

-

I

1

1- - -

-

002 -

III

hUelnl1

-

1-11- 1-

001 OllIS ,002

pcr .ode C'" "'e IS)

~

001'

- 00) -

rosoo

'- -

f\-

00)

-- ,002 -

I-

Fu,.

--

E = Side win C- Note ")

i- r-

1-

~~ ~

6000.0 11000 Comm fOf" f'Hh addlhem.1 00) inch .dd (Inch..)

---

Plu, or M,nu. In Thou.. nd. at on Inch 6 1 • 9 10 II 12 I) I' IS 16 17 111910 21 111) 2< 2S 2611 21

~~ i":

1 - 1 ...

(..e N e'l)

)

kvcl

I

I

1

D

E NOTES

lo1cr.ncn do "'" '1KII>dc .llowancc 1<><

ch.rK1CnlheJ or mite"'"

C

I

--I" ..

2 - Tolera,,", booed on

• ...11 section 3 - PanlD,lmc mus-11x I.Iktn Into constderJIhon

WIn

4 - Pan desi", should malnilin I th;ckM'U ., Marl, conll.nl OJ pot.ible Com,,$cte uniform II, In IhlS dimcnilon b impouIbic 10 .chlttY!'

- Care mull be .nen .hal the nlio "I lhe cl
hould bC' Increased ... henuer com· de....d cI<·PI .nd aood moIdll1,

7 - Cu lon>er·MoIckr undcrs..ndm, n«
loolln,

Ole 17)

C_ NOlen)

Figure 5·86. A standard tolerance chart for a specific polycarbonate.

should serve as the basis for establishing standards for molded products between the designer and customer and molder. Users will find that two separate sets of values are here represented. Commercial values represent common production tolerances that can be achieved at the most economical level. Fine values represent closer tolerances that can be held, but at a greater cost. The selection of one or the other will depend on the application under consideration and the economics involved. By referring to the hypothetical molded article and its cross-section, illustrated in the table, then using the applicable code number (e.g., A represents the diameter) in the first column of the table and the exact dimensions indicated in the second column, one can find the recommended tolerances either in the chart at the top of the table or in the two columns underneath. (Note that the typical article shown in cross-section in the table may be round or rectangular or some other shape. Thus, dimensions A and B may be either diameters or lengths.)

STRUCTURAL DESIGN ANALYSIS 391

Table 5-7. Guidelines for Wall Thicknesses of TP Molding Materials

Alkyd-glass filled Alkyd-mineral filled Diallyl phthalate Epoxy-glass filled Melamine-<:ellulose filled Urea-<:ellulose filled Phenolic-general purpose Phenolic-flock filled Phenolic-glass filled Phenolic-fabric filled Phenolic-mineral filled Silicone glass Polyester premix

Minimum Thickness in. (mm)

Average Thickness in. (mm)

.040 .040 .040 .030 .035 .035 .050 .050 .030 .062 .125 .050 .040

.125 (3.2) .187 (4.7) .187 (4.7) .125 (3.2) .100 (2.5) .100 (2.5) .125 (3.2) .125 (3.2) .093 (2.4) .187 (4.7) .187 (4.7) .125 (3.2) .070 (1.8)

(1.0) (1.0) (1.0) (0.76) (0.89) (0.89) (1.3)

(1.3) (0.76) (1.6) (3.2) (1.3)

(1.0)

Maximum Thickness in. (mm)

.500 .375 .375 1.000 .187 .187 1.000 1.000 .750 .375 1.000 .250 1.000

(13) (9.5) (9.5) (25.4) (4.7) (4.7) (25.4) (25.4) (19) (9.5) (25.4) (6.4) (25.4)

PROCESSING AND TOLERANCES Processing is extremely important in regard to tolerance control; in certain cases it is the most influential factor. The dimensional accuracy of the finished part relates to the process, the accuracy of mold or die production, and the process controls, as well as the shrinkage behavior of the plastic. A change to a mold or of a die's dimensions can result in wear arising during production and should thus be considered. The mold or die should also be recognized as one of the most important pieces of production equipment in the plant. This controllable, complex device must be an efficient heat exchanger and provide the part's shape. The mold or die designer thus has to have

Table 5-8. Guidelines for Wall Thicknesses of TP Molding Materials Minimum in. (mm) Acetal ABS Acrylic Cellulosics FEP fluoroplastic Nylon Polycarbonate Polyester T.P. Polyethylene (L.D.) Polyethylene (H.D.) Ethylene vinyl acetate Polypropylene Polysulfone Noryl (modified PPO) Polystyrene SAN PVC-Rigid Polyurethane SUrlyn (ionomer)

.015 .030 .025 .025 .010 .015 .040 .025 .020 .035 .020 .025 .040 .030 .030 .030 .040 .025 .025

(0.38) (0.76) (0.63) (0.63) (0.25) (0.38) (1.0) (0.63) (5.1) (0.89) (0.51) (0.63) (1.0) (0.76) (0.76) (0.76) (1.0) (0.63) (0.63)

Average in. (mm)

.062 .090 .093 .075 .035 .062 .093 .062 .062 .062 .062 .080 .100 .080 .062 .062 .093 .500 .062

(1.6) (2.3) (2.4) (1.9) (0.89) (1.6) (2.4) (1.6) (1.6) (1.6) (1.6) (2.0) (2.5) (2.0) (1.6) (1.6) (2.4) (12.7) (1.6)

Maximum in. (mm)

.125 .125 .250 .187 .500 .125 .375 .500 .250 .250 .125 .300 .375 .375 .250 .250 .375 1.500 .750

(3.2) (3.2) (6.4) (4.7) (3.2) (9.5) (12.7)

(6.4) (6.4) (3.2) (9.5) (9.5) (7.6) (7.6) (9.5) (38.1) (19.1)

392 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 5-9. Guide to Tolerances of TP Extrusion Profiles

Wall thickness (%, =) Angles (Deg., = )

pvc

HIPS

PC, ABS

pp

Rigid

Rex.

LOPE

8 2

8 3

8 3

8 2

10 5

10 5

0.010 0.020 0.025 0.027 0.035 0.037 0.050 0.065 0.093 0.125

0.010 0.015 0.020 0.027 0.035 0.037 0.050 0.065 0.093 0.125

0.007 0.010 0.015 0.020 0.025 0.030 0.045 0.060 0.075 0.093

0.010 0.Ql5 0.020 0.030 0.035 0.040 0.065 0.093 0.125 0.150

0.012 0.025 0.030 0.035 0.040 0.045 0.065 0.093 0.125 0.150

Profile dimensions (in., ±) To 0.125 0.125 to .500 .500 to 1 1 to 1.5 1.5 to 2 2 to 3 3 to 4 4 to 5 5 to 7 7 to 10

0.007 0.012 0.017 0.025 0.030 0.035 0.050 0.065 0.093 0.125

the experience or training and knowledge of how to produce the tooling needed for the part and to meet required tolerances [8-13, 32, 45]. Adequate process control and its associated instrumentation are essential to have product quality control. In some cases the goal is precise adherence to a control point. In others it is simply to maintain the temperature within a comparatively narrow range. A knowledge of processing methods can be useful to the designer, to help determine what tolerances can be obtained (see Chapter 7). With such high-pressure methods as injection and compression molding that use 13.8 to 206.9 MPa (2,000 to 30,000 psi) it is possible to develop tighter tolerances, but there is also a tendency to develop undesirable stresses (that is, orientations) in different directions. The low-pressure processes, including contact and casting with no pressure, usually do not permit meeting tight tolerances. There are exceptions, such as certain RPs that are processed at quite low pressures. Regardless of the process used, exercising the required and proper control over it will maximize the obtaining and repeating of tolerances that are achievable. For example, certain injection-molded parts can be molded to extremely close tolerances of less than a thousandth of an inch, or down to 0.0 percent, particularly when TPs are filled with additives or TS compounds are used. To practically eliminate shrinkage and provide a smooth surface, one should use a small amount of a chemical blowing agent «0.5 percent by weight) and a regular packing procedure. For conventional molding, tolerances can be met of ± 5 percent for a part 0.020 in. thick, ± 1 percent for 0.050 in., ±0.5 percent for 1.000 in., ±0.25 percent for 5.000 in., and so on. Thermosets generally are more suitable than TPs for meeting the tightest tolerances. Economical production requires that tolerances not be specified tighter than necessary. However, after a production target is met, one should mold "tighter" if possible, for greater profit by using less material. Table 5-10 reviews factors affecting tolerances. Many plastics change dimensions after molding, principally because their molecular orientations or molecules are not relaxed (see Chapter 2). To ease or eliminate the problem, one can change the processing cycle so that the plastic is "stress relieved," even though that may extend the cycle time, or heat-treat according to the resin supplier's suggestions. An easy method for estimating shrink allowance for injection molding is as follows:

STRUCTURAL DESIGN ANALYSIS 393

Table 5-10. Parameters that Influence Part Tolerance PART DESIGN: MATERIAL: MOLD DESIGN:

MACHINE CAPABILITY:

MOLDING CYCLE:

Part configuration (size/shape). Relate shape to flow of melt in mold to meet performance requirements that should at least include tolerances. Chemical structure, molecular weight, amount and type of fillers/additives, heat history, storage, handling. Number of cavities, layout and size of cavities/runners/gates/cooling lines/side actionslknockout pins/etc. Relate layout to maximize proper performance of melt and cooling flow patterns to meet part performance requirements; preengineer design to minimize wear and deformation of mold (use proper steels); layout cooling lines to meet temperature to time cooling rate of plastics (particularly crystalline types). Accuracy and repeatability of temperatureltime/velocity/pressure controls of injection unit, accuracy and repeatability of clamping force, flatness and parallelism of platens, even distribution of clamping on all tie rods, repeatability of controlling pressure and temperature of oil, oil temperature variation minimized, no oil contamination (by the time you see oil contamination damage to the hydraulic system could have already occurred), machine properly leveled. Set up the complete molding cycle to repeatedly meet performance requirements at the lowest cost by interrelating material/machine/mold controls.

SD = FL (1 + SR), where SD = the mold dimension, SR = the plastic's shrinkage (in in.lin. or mm1mm), and FL = the part dimension. If the parts are small and have thin walls, this estimate is the best guide. If they are large (>25 cm, or 10 in.) or use rather high-shrink plastics, consider using the following method of analysis [418]. The two formulas to use are Current equation listed above: SD 1 = FL (1 Correct equation:

SD 2

= FL

+ SR)

«(1 - SR)

where SD 1 = the mold dimension as determined by the current system and SD 2 = the mold dimension as determined by the correct equation. The error, ER, would simply be the difference between the SD 1 and SD 2 equations. To be more accurate for calculating mold dimensions where the part size and shrink rate increase, this error value should be considered or Table 5-11 be used. This table shows, as one example, that in the low shrink (0.008 millin. or less) materials, the parts must be larger than 15 in. before an error of 0.001 in. will be realized. The allowable error will depend upon each part's particular application. In some cases it will be important to ensure proper mold-size calculations. In others, changing the calculation method will be purely academic. Experience is still a basic requirement for mold design with regard to determining cavity dimensions. The costs for changing mold cavities are high, even when similar moldings are to be produced. Until now, theoretical efforts to forecast linear shrinkage have been limited because of the number of existing variables. One way to solve this problem is to simplify the mathematical relationship, leading to an estimated but still acceptable assessment. This means, however, that the number of necessary processing changes will also be reduced [12]. As a first approximation, a superposition method can be used to predict mold shrinkage (see Fig. 5-87). However, problems arise in measuring the influencing variables, because

~

r..>

..

-0.06 -0.19 -0.32 -0.45 -0.58 -0.71 -0.84 -0.97 -1.10 -1.23 -1.35 -1.48 -1.61 -1.74 -1.87 -2.00 -2.13 -2.26 -2.39 -2.52 -2.65 -2.77 -2.90 -3.03 -3.16

-0.02 -0.05 -0.08 -0.11 -0.14 -0.18 -0.21 -0.24 -0.27 -0.31 -0.34 -0.37 -0.40 -0.43 -0.47 -0.50 -0.53 -0.56 -0.59 -0.63 -0.66 -0.69 -0.72 -0.76 -0.79

1.0 3.0 5.0 7.0 9.0 11.0 13.0 15.0 17.0 19.0 21.0 23.0 25.0 27.0 29.0 31.0 33.0 35.0 37.0 39.0 41.0 43.0 45.0 47.0 49.0

-0.1 -0.4 -0.7 -1.0 -1.3 -1.6 -1.9 -2.2 -2.5 -2.8 -3.1 -3.4 -3.6 -3.9 -4.2 -4.5 -4.8 -5.1 -5.4 -5.7 -6.0 -6.3 -6.6 -6.9 -7.1

0.012 -0.3 -0.8 -1.3 -1.8 -2.3 -2.9 -3.4 -3.9 -4.4 -4.9 -5.5 -6..0 -6.5 -7.0 -7.5 -8.1 -8.6 -9.1 -9.6 -10.1 -10.7 -11.2 -11.7 -12.2 -12.7

0.016

0.030 -0.9 -2.8 -4.6 -6.5 -8.4 -10.2 -12.1 -13.9 -15.8 -17.6 -19.5 -21.3 -23.2 -25.1 -26.9 -28.8 -30.6 -32.5 -34.3 -36.2 -38.0 -39.9 -41.8 -43.6 -45.5

0.020 -0.4 -1.2 -2.0 -2.9 -3.7 -4.5 -5.3 -6.1 -6.9 -7.8 -8.6 -9.4 -10.2 -11.0 -11.8 -12.7 -13.5 -14.3 -15.1 -15.9 -16.7 -17.6 -18.4 -19.2 -20.0 -1.7 -5.0 -8.3 -11.7 -15.0 -18.3 -21.7 -25.0 -28.3 -31.7 -35.0 -38.3 -41.7 -45.0 -48.3 -51.7 -55.0 -58.3 -61.7 -65.0 -68.3 -71.7 -75.0 -78.3 -81.7

0.040 -4 -11 -19 -27 -34 -42 -50 -57 -65 -73 -80 -88 -96 -103

-3 -8 -13 -18 -24 -29 -34 -39 -45 -50 -55 -61 -66 -71 -76 -82 -87 -92 -97 -103 -108 -113 -118 -124 -129 -119 -126 -134 -142 -149 -157 -165 -172 -180 -188

-Ill

0.060

0.050

-121 -132 -142 -153 -163 -174 -184 -195 -205 -216 -227 -237 -248 -258

-Ill

-5 -16 -26 -37 -47 -58 -68 -79 -90 -100

0.070 -7 -21 -35 -49 -63 -77 -90 -100 -118 -132 -146 -160 -174 -188 -202 -216 -230 -243 -257 -271 -285 -299 -313 -327 -341

-300 -322 -344 -367 -389 -411 -433 -456 -478

-240 -258 -276 -294 -312 -329 -347 -365 -383 -401 -418 -436

-522 -544

-500

-11 -33 -56 -78 -100 -122 -144 -167 -189 -211 -233 -256 -278

0.100

-9 -27 -45 -62 -80 -98 -116 -134 -151 -169 -187 -205 -223

0.080 0.090

Plastic Shrink Rate (inches/inch)

'Error values in table are in mil (0.001 incb); thus, for shrink rale of 0.050 inlin and part size of 11 in, the error is 29 mil (0.029 in).

0.008

0.004

Part Size Inches

-50 -150 -250 -350 -450 -550 -650 -750 -850 -950 -1050 -1150 -1250 -1350 -1450 -1550 -1650 -1750 -1850 -1950 -2050 -2150 -2250 -2350 -2450

0.200 -129 -386 -643 -900 -1157 -1414 -1671 -1929 -2186 -2443 -2700 -2957 -3214 -3471 -3729 -3986 -4243 -4500 -4757 -5014 -5271 -5529 -5786 -6043 -6300

0.300

Table 5-11. Error in Mold Size as a Result of Using Incorrect Shrinkage Equation*

-267 -800 -1333 -1867 -2400 -2933 -3467 -4000 -4533 -5067 -5600 -6133 -6667 -7200 -7733 -8267 -8800 -9333 -9867 -10400 -10933 -11467 -12000 -12533 -13067

0.400

-500 -1500 -2500 -3500 -4500 -5500 -6500 -7500 -8500 -9500 -10500 -11500 -12500 -13500 -14500 -15500 -16500 -17500 -18500 -19500 -20500 -21500 -22500 -23500 -24500

0.500

STRUCTURAL DESIGN ANALYSIS 395

they are often interrelated, such as variations in the pressure course in a mold with a varying wall thickness. The parameters of the injection process must be provided. They can either be estimated or, to be more exact, taken from the thermal and rheological layout. The position of a length with respect to flow direction is in practice an important influence. This is so primarily for glass-filled material but also for unfilled thermoplastics, as is shown in Figure 5-88. The difference between a length parallel to (0°) and perpendicular to (9()D) the flow direction depends on the processing parameters. Measurements with unfilled PP and ABS have shown that a linear relationship exists between these points. Regarding this relationship, when designing the mold it is necessary to know the flow direction. To obtain this information, a simple flow pattern construction can be used (see Fig. 5-89). However, the flow direction is not constant. In some cases the flow direction in the filling phase differs from that in the holding phase. Here the question arises of whether this must be considered using superposition. In order to get the flow direction at the end of the filling phase and the beginning of the holding phase (representing the onset of shrinkage), an analogous model was developed that provides the flow direction at the end of the filling phase. For a flow with a Reynolds number less than 10, which is valid regarding the processing of thermoplastics, the following equation can be used: .:1<1> = O. For a two-dimensional geometry with quasistationary conditions, this equation is valid:

Instead of the potential <1>, it is possible to introduce the flow-stream function tis for a two-dimensional flow. The stream lines (tis = constant) and the equipotential lines are perpendicular to each other. To express this, the following Cauchy-Rieman differential equations can be used:

a = atls a = _ atls ax i.Iy i.Iy ax

!lS,

I Mold -temperature

Holding-pressure

ASIIII

------

S

Sn

"'0

'"

Flow-angle

Molding -thickness

S, = SIO + AS, + AS" +

Figure 5-87. Superposition to determine shrinkage.

AS",

2.5 %

2.4

tJf 2.3 2.2

Q)

2.1

'\ \.0 \. "' ~o

0>

'"

... -" c:

2.0

Material: PP ,'}w = 30°C ,'}M = 240°C S = 1.5 mm VF = 140 mm 5

'\ 0

"

\~

L.

CJ)

1.9 PN

1.8

"-

~

f-PN

= 190b~r~

0\ I' "

1.7

'\ \0

1.6 1.5

= 160 bar

0\

10

30

50 Flow-angle

a 70

\..

'\c grd

Figure 5-88. The influence of the flow angle on processing shrinkage.

_"'--_....J S = Figure 5-89. Flow patterns. 396

3 mm

90

STRUCTURAL DESIGN ANALYSIS 397

A differential (two dimensionaVquasi) equation has the same form as is used for a stationary electrical potential field,

as it can be realized with an unmantled molding out of resistance paper and a suitable voltage. To control the theoretically determined flow with respect to the orientation direction, a color study was made. The comparison between flow pattern, color study, and analogous model is shown in Figures 5-90 and 5-91. For a simple geometry the flow pattern method describes the flow direction in the filling phase as well as the holding phase (see Fig. 5-90). This description changes when a core is added and the flow is disturbed (see Fig. 591). In this case the flow at the beginning of the holding phase differs from the flow

Figure 5-90. A comparison between an analogous model, a flow pattern, and color studies.

Figure 5-91. A comparison between an analogous model, a flow pattern, and color studies, with a core added.

398 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

pattern as it is shown in the color study as well as in the analogous model. Even the welding lines are broken in the holding phase so that at this place another flow direction than that in the filling phase is found. With further measurements this influence has to be tested by using more-complex moldings. More information concerning the influence of injection molding on tolerances is contained in Chapter 7. That chapter also describes and provides information that pertains to factors affecting tolerances during extrusion, blow molding, and other processes.

PRODUCT SPECIFICATION Tolerances on dimensions should be specified only where absolutely necessary. Too many drawings show limits on sizes when other means of attaining the desired results would be more constructive. For example, if the outside dimensions of certain drill-housing halves were to have a tolerance of ± 0.008 cm (0.003 in.), this would be a tight limit. Yet if half of the housing were to be on the minimum side and the other on the maximum side, there would be a resulting step that would be uncomfortable to the feel of the hand while gripping the drill. A realistic specification would call for matching of halves so as to provide a smooth joint between them, with the highest step not exceeding 0.002 in. The point is that limits should be specified in such a way that those responsible for the manufacture of a product will understand the goal that is to be attained. Thus we may indicate "dimensions for gear centers," "holes as bearing openings for shafts," "guides for cams," and so on. This type of designation would alert a mold maker as well as the molder to the significance of the tolerances in some areas and the need for matching parts in other places and the clearance needed for assembly in still other locations. Most of the engineering plastics faithfully reproduce the mold configuration, and when the processing parameters are appropriately controlled they will repeat with excellent accuracy. We see plastic gears and other precision parts made of acetal, nylon, polycarbonate, and Noryl whose tooth contour and other precision areas are made with a limit of 0.0002 in. and the spacing of the teeth is uniform to meet the most exacting requirements. The problem with any precise part is to recognize what steps are needed to reach the objective and to follow through every phase of the process in a thorough manner, to safeguard the end product. Throughout this book, shrinkage is discussed based on the material's given characteristics. Different factors can cause variation in shrinkage; indeed, the way processing parameters can influence dimensional variation is very important. Some materials perform better than others in that respect. Generally, if we approach tolerances according to their purposes' a) functional requirements, such as running fit, sliding fit, gear tooth contour, etc.; b) assembly requirements-that is, to accommodate parts with their own tolerances; and c) matching parts for appearance or utility, we should come up with feasible tolerances that will be reasonable and useful. This will be more productive than trying to apply tolerances strictly on a dimensional basis. Tolerances should be indicated only where they are needed, carefully analyzed for their magnitude, and of proven usefulness. It is important to determine if the tolerances shown are realistic for the specified plastic and process. The designer should recognize that extreme accuracy of dimensions is expensive and, in some cases, impossible to hold in processing. Adaptation of metal tolerances to plastics is not advisable. The reaction of plastics to

STRUCTURAL DESIGN ANALYSIS 399

moisture and heat, for example, is drastically different from that of metals, so that pilot testing under extreme use conditions is almost mandatory for establishing adequate tolerance requirements.

COMBINING VARIABLES There are many different factors that could influence the repeatability of meeting tolerances, as well as affecting the production of a product to meet all the other performance requirements. Some products may require only the compliance of one or two processing factors, but others will require many. Computer programs have been developed to provide the capability of integrating all the applicable factors, thus replacing traditional trial-anderror methods [10-12]. Most computer-integrated systems have been developed for injection molding, since a much bigger market exists with it. Other computer systems are available for the other processes (see Appendix 4, Computerized Software and Databases").

FINITE ELEMENT ANALYSIS The opportunity for creative design by viewing many imaginative variations would be blunted if each variation introduced a new set of doubts as to its ability to withstand whatever stress might be applied. From this point of view the development of computer graphics has to be accompanied by an analysis technique capable of determining stress levels, regardless of the shape of the part. This need is met by finite element analysis

[1,2, 10-12,62-68, 72, 381-84, 419-41]. Finite element analysis (PEA) is a computer-based technique for determining the stresses and deflections in a structure. Essentially, this method divides a structure into small elements with defined stress and deflection characteristics. The method is based on manipulating arrays of large matrix equations that can be realistically solved only by computer. Most often, PEA is performed with commercial programs. In many cases these programs require that the user know only how to properly prepare the program input. PEA is applicable in several types of analyses. The most common one is static analysis to solve for deflections, strains, and stresses in a structure that is under a constant set of applied loads. In PEA a material is generally assumed to be linear elastic, but nonlinear behavior such as plastic deformation, creep, and large deflections also are analyzed. The designer must be aware that as the degree of anisotropy increases the number of constants or moduli required to describe the material increases. Uncertainty about a material's properties, along with a questionable applicability of the simple analysis techniques generally used, provides justification for extensive enduse testing of plastic parts before approving them in a particular application. However, it should be noted that as the use of more PEA methods becomes common in plastic design, the ability of PEAs to handle anisotropic materials will demand greater understanding of the anisotropic nature of plastics. PEA does not replace testing; rather, the two are complementary in nature. Testing supplies only one basic answer about a design-either pass or fail. It does not quantify results, because it is not possible to know from testing alone how close to the point of passing or failing a design actually exists. PEA does, however, provide information with which to quantify performance.

400 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Bolt Preload Let us consider an example in which FEA is used to design a way to maintain bolt preloading in plastic assemblies [72]. (The preload is the tensile load on a threaded fastener as a result of assembly torque.) It highlights the problems associated with plasticpart design when the designer is restricted to using an existing nut-and-bolt fastening system originally designed for a metal part that has since been replaced. The problem concerns the loss of the bolt preload in a clamped metal-to-plastic assembly. An actual case history like this is the replacement of aluminum in a die-cast cover using Du Pont's MinIon 22C engineering TP. This mineral-glass reinforced composition of nylon 66 had been specified for the cover because of its excellent stiffness and strength at high temperatures, as well as its ability to be molded into a relatively warpfree part. The cover is bolted to an aluminum housing, with an aluminum carrier gasket interposed between the two (see Fig. 5-92). The fasteners used 6 mm (! in.) bolts with washers torqued to a recommended range of 7.5 to 9.5 N-m (5.5 to 7.0 lb.-ft.). For this example, assume that the cover is exposed to vibration, external loads, and a temperature range of - 40 to 120°C ( - 40 to 248°F). A failure of the bolted joints will most likely occur because of excessive loads under the bolt and washer. The high loads on the cover will eventually cause the molded MinIon to creep under the bolt and make it lose its initial bolt preload. At this point vibration and external loading will cause further loosening of the bolt and a loss of compression between parts. The proper preload on the fastener is required to prevent loosening of the bolt, provide a frictional force between parts to resist bolt shear, and improve the fatigue resistance of the bolt connection. The original fastener in this example performed in the traditional way, a spring. In such a design, as the bolts are tightened the fastener goes into tension. This force causes deflection in the clamped members and bolts. In effect, the force and deflection is the spring rate of the bolted connection. It is commonly accepted that the preload of a bolted fastener during assembly should be 75 to 90 percent of its yield strength, with the lower 75 percent value offering a higher safety factor against the bolt's yielding. The preload is determined by the classical normal tension-compression stress formula (seen in Chapter 3) and uses 75 percent of the bolt's yield strength:


Where: FJ A
FJ A

=-


= Preload tensile load, N (lb.) = =

Bolt stressed area, m2 (in. 2 ) Yield strength stress of bolt, MPa (psi)

Given that the yield strength of a Class 8.8 bolt is 646 MPa (93.7 kpsi), the stressed area of the bolt is 20.1 mm2 (0.031 in. 2) Preload FJ

= = =


x 0.75 x A

646 MPa x 0.75 x 2.01 x 10-5 m2 9740 N (2190 lb.)

STRUCTURAL DESIGN ANALYSIS 401 FEM M6x 1 OOBOit M6Washer

I

COlIer ,n Minion" 22C

I

) AlumInum Carrier Gaske

J \

Alum,num Housong

I

I

Figure 5-92. The assembly of a bolted connection where a finite element analysis stress-contour plot maintained the bolt preload in a plastic assembly.

The following empirical equation can be used to determine the proper torque value for this fastener:

T

= kDFJ

Where: T = Torque, N'm (lb. ·ft.) D k

= =

Normal diameter of bolt, mm (in.) Torsional coefficient

The results of this equation vary depending on the torsional coefficient value used. Testing has verified that an unplated, nonlubricated bolt has a torsional coefficient (k) of 0.20. For plated, lightly oiled bolts, k = 0.15. When other lubricants like greases, oils, and waxes are used, k = 0.12. For analysis, our interest is in the maximum preload value generated during assembly, because of the problem of creep. The maximum preload will occur with a lower k value. The value k = 0.15 will be used, because it is almost impossible to exclude having a light oil lubricant on these parts in a manufacturing environment. The torque value then correlates well with the manufacturer's suggested torque range: Torque T = 0.15 x 6 mm x 9740 N

= 8.77 N'm (6.47 lb. ·ft.) Finite element analysis was then performed on the bolted connection, using a calculated preload of 9,740 N. The analysis showed that the highest principal compressive stress under the washer is 269 MPa (39 kpsi). According to the creep curves for MinIon, in order for the cover not to creep and allow a loss of preload, the stress level would have to be less than 13.8 MPa (2,000 psi) during operation at peak temperatures. The PEA curves show the principal stress distribution in cross-sections for the bolted connection. One solution to this problem is to insert some form of compression-limiting device into the plastic. After molding, a steel collar can be used to carry the preload force. A 2 mm- (0.08 in.) thick steel collar was added to the PEA model and the load case run again. The stress in the MinIon cover now dropped to an acceptable 12 MPa (1,700 psi). In this design arrangement almost all the load is now carried by the steel collar. This

402 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

also means that the load now transfers directly into the aluminum gasket carrier and housing. Increasing the surface area of the collar where it contacts the aluminum parts permits lowering the stress levels to a safe point below the yield strength of the aluminum. The top shape of the collar thus resulted in having lower stresses, as shown by the FEA stresscontour plots for the clamped assemblies [72]. The initial calculations by hand suggested that a collar with a larger surface was needed at the aluminum interface. In this situation, FEA was helpful in fine tuning the design. However, only end-use testing will actually verify results and account for such changing variables as frictional coefficients or the output torque from the bolt driver. There is an easier way to maintain the necessary preload in a plastic part. Largediameter washers will help some, but it would take a diameter greater than 40 mm (1.57 in.) to get low enough stress levels. Using Belleville spring- and lockwashers will only prolong the inevitable if sufficient surface area is not provided. Locknuts, interference threads, and using the prevailing torque nylon patches on bolts and nuts will help only to maintain the relative positions of the nuts to the bolts. None of these features will prevent plastic from creeping. An easier way to maintain a preload is to determine first how much, if any, is needed. In this example there was no option to change the fastening system, but in less-demanding applications, where little force is required to obtain continuity, such alternatives as downsized bolts, Belleville washers, and locking treaded fasteners may be the best way.

CANTILEVERED SNAP FITS The design target for a cantilevered snap fit is to develop a latch geometry for a specific plastic being used that can withstand and recover from the strain developed during deflection. This strain is usually calculated from the standard cantilevered beam equation for a rectangular or tapered cross-section (see Fig. 5-47 and Table 5-3). An important requirement is to prevent the latch from taking a permanent set, or retaining residual deflection, after bending. The amount of residual deflection produced is affected by the ratio of the latch length, L, to its cross-section thickness, t. The smaller the L:t ratio, the greater the strain the latch can withstand without taking a permanent set. The amount of stress that can be tolerated prior to its reaching its yield point depends on the characteristics of the resin and the type and amount of additives that may be present in the plastic compound. As expected, finite element analysis shows that deflection stresses tend to concentrate at the root of a latch where it joins the wall or supporting structure and can produce cracks at that location from repeated use of the latch (see Fig. 5-93). Providing generous root fillets can help, but the better approach is to taper the thickness of the latch-support beam. Its gradual thinning redistributes the stress along its length and can reduce the peak stress level by 25 percent or more. More information on the shapes of beam crosssections is shown in Figures 5-47 through 5-50.

LAWS AND REGULATIONS The consuming public must assume that the producer of a product has shown reasonable consideration for the safety, correct quantity, proper labeling, and other social aspects of the product. Since the 1960s these types of important concerns have expanded and been reinforced by a recognition of the consumer's right to know as well as by concerns

STRUCTURAL DESIGN ANALYSIS 403

Straight (Joss-section

Tapered cross-section

Figure 5-93. Computer-generated graphics for straight and tapered cross-sections of cantilevered snap-fits showing how tapering removes stress from the root area.

for conservation, ecology, antilittering, and the like. Numerous safety-related and socially responsible laws have been enacted and more are on the way. A designer's failure to be aware of and comply with existing regulations can lead to legal entanglements, fines, restrictions, and even jail sentences. In addition, there are also the penalties of costly, damaging publicity, lawsuits, and the loss of consumer goodwill. In the meantime, as for all other industries, the goal of reliable companies and associations is to produce products that eliminate potential problems. Unfortunately, nothing is perfect, so problems can develop, which is simply a fact of life. And there is always more to be done, as in the disposal issue, the subject of Chapter 12. There are many examples of action to eliminate or reduce problems. On the subject of appliance safety the Underwriters Laboratories (UL) have published more than four hundred safety standards to assess the hazards associated with manufacturing appliances. These standards represent basic engineering requirements for various categories of products covered by the organization. For example, under UL's Component Plastics Program a material is tested under standardized, uniform conditions to provide preliminary information as to a material's strong and potentially weak characteristics. The UL plastics program is divided into two phases. The first develops information on a material's long- and short-term properties. The second phase uses these data to screen out and indicate a material's strong and weak characteristics. For example, manufacturers and safety engineers can analyze the possible hazardous effects of potentially weak characteristics, using UL standard 746C. Parts manufactured using concepts in UL Standard 746D provide quick verification of material identification, along with the assurance that acceptable blending or simple compounding operations are used that would not increase the risk of fire, electrical shock, or personal injury. The Standard for Tests for Flammability of Plastic Materials for Parts in Devices and Appliances (UL 94) has methods for determining whether a material will extinguish, or bum and propagate flame.

404 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The UL Standard for Polymeric Materials-Short Term Property Evaluations is a series of small-scale tests used as a basis for comparing the mechanical, electrical, thermal, and resistance-to-ignition characteristics of materials. Most of these tests were extracted from procedures developed by the American Society for Testing and Materials (ASTM) and the International Electrotechnical Commission (1EC), because they are time tested, generally accepted methods to evaluate a given property. Where there was no universally accepted methods the UL developed its own. Despite a significant decrease in reported U.S. fire deaths, the hazards of burning materials remain a major issue for both the country and the plastics industry. The results of two programs being conducted by the National Fire Protection Research Foundation (NFPRF) and the National Institute of Building Sciences (NIBS) should provide new evaluation systems that will be a great advance in consumer safety and product evaluations [432]. The NIBS advisory committee for developing the test protocol are the following: American Association of Retired Persons (AARP) American Furniture Manufacturers Association (AFMA) American Hotel and Motel Association (AHMA) American Society for Testing and Materials (ASTM) Carpet and Rug Institute (CRI) Consumer Product Safety Commission (CPSC) Fire Marshals Association of North America (FMANA) Fire Retardant Chemicals Association (FRCA) General Services Administration (GSA) International Association of Fire Chiefs (IAFC) Man-Made Fiber Producers Association (MMFPA) National Association of Home Builders (NAHB) National Institute of Standards & Technology (NIST) National Conference of States on Building Codes and Standards (NCSBCS) National Electrical Manufacturers Association (NEMA) Underwriters Laboratories (UL) U.S. Fire Administration (USFA) It is the general consensus within the worldwide "fire community" that the only proper way to evaluate the fire safety of products is to conduct full-scale tests or complete firerisk assessments.

Chapter 6

THE PROPERTIES OF PLASTICS

This chapter provides specific infonnation on the range of properties in the many different plastics available to meet different design requirements (see Fig. 6-1). (One example of property ranges was summarized in Fig. 1-4.) The "neat" type represents those made only of a polymer (that is, a plastic) with no filler (reinforcement or other additive). The many filled types are added to plastics, usually in the fonn of small particles, powders, liquids, and fibers to modify its processing, perfonnance, or the cost of the product [433510]. Clearly, the combinations of resins and fillers and the resulting property variations are endless (see Fig. 6-2). The point is that each combination is in fact a new material with its own trade-offs. Some properties will be improved, others unchanged, and still others diminished from those of the basic unfilled plastic. In this chapter there is no relationship, direct or implied, between any plastic in tenns of the space given it and its perfonnance or the size of its market. The largest consumption of these plastics is low-density polyethylene (LDPE) fonnulations, at about 25 percent weightwise, followed by high-density polyethylene (HDPE), then polypropylene, polyvinyl chloride, and polystyrene. These together total about two-thirds of all plastic consumption. All data presented in this chapter, as throughout this book, are per ASTM Standards unless otherwise specified. Also, each plastic discussed here can include only a few advantages and disadvantages; for detailed infonnation see the References section, computerized databases, and, most importantly, the plastic-material suppliers.

TRADE NAMES If the perfonnance requirements for a design provide wide enough limits, materials from different companies can be used in the product. This is so because the manufacturing methods of different companies for producing the same basic polymers, as well as preparing the same basic polymer alloys, blends, and compounds, can produce different perfonnance in properties and processing. The different companies involved can thus produce materials with varying property limits. However, a plastic with specific perfonnance characteristics and limits may be available only from a single company, as is the case with such other materials as steel, aluminum, zinc, ceramic, and glass. Thus, trade names for specific plastics become important. Note 405

406 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Composition Structure

Electrical Thermal Magnetic Gravimetric

I

Strength Ductility Toughness

Service Life

Ri9iry

Size Shape Microtopography

I

I

Figure 6-1. An example of plastics' properties.

that certain plastics can be fabricated by practically any process, but some require only a certain process. Here and throughout this book trade names are used where applicable to highlight specific or special properties. Some examples are shown in Tables 6-1 and 6-2.

ACRYLONITRILE-BUTADIENE-STYRENE (ABS) The ABSs are a family of thermoplastics that contain three basic polymers-acrylonitrile, butadiene, and styrene-in different proportions, resulting in different properties. Acrylonitrile (AN) contributes strength, with heat and chemical resistance; butadiene (B) lends impact strength, toughness, and low-temperature property retention and styrene (S) contributes gloss, processability, and rigidity. The properties are varied principally by adjusting the proportions in which the materials are combined and by altering the molecular weight of the styrene acrylonitrile. With their excellent strength, stiffness, and toughness ABSs can be called engineering plastics. They compare favorably with nylon and acetal in certain applications but are generally less expensive. Parts in ABSs are almost completely unaffected by water, salts, and most inorganic acids, food acids, and alkalis, but much depends on the time, temperature, and especially the stress level. The FDA's acceptance of ABSs depends to some extent on the pigmentation system used in them. Their properties can be further modified by using a fourth copolymer like alphamethyl styrene or blending them with such other polymers as PVC, PC, or the vinyl chlorides and sulfones. Then property trade-offs do occur, as for instance blends with a high modulus usually having low impact strength. In order to increase various mechanical properties significantly, reinforcing grades containing fibers like glass of at least 40 percent by weight are available. Most natural ABSs run from translucent to opaque, but they are also produced as transparent. They can be pigmented to almost any color, with grades available for all fabricating processes. Some grades are designed specifically for electroplating. The general grades may be adequate for some outdoor applications, but prolonged exposure to sunlight will cause color changes and reduce surface gloss, impact strength, and ductility. Tensile strength, hardness, and the elastic modulus are less affected. Pigmenting the ABS black, laminating with an opaque acrylic sheet, and applying certain coating systems will

THE PROPERTIES OF PLASTICS 407

Figure 6·2. Reinforced plastics as composites provide many different combinations of plastics such as this Ford Econoline glass-fiber-TS polyester resin filament-wound driveshaft.

provide weather resistance. For maximum color and gloss retention a compatible coating of opaque, weather-resistant polyurethane can be used on fabricated parts. For weatherable sheet applications, ABSs can be coextruded with a compatible weather-resistant polymer on their outside surfaces. The impact properties of ABS are exceptionally good at room temperature and, with special grades, at temperatures to as low as - 40°C ( - 40°F). Because of its plastic yield at high strain rates, impact failures in ABS are ductile rather than brittle. Also, the skin effect that in most other thermoplastics accounts for a lower impact resistance in thick sections than in thin ones is not pronounced in ABS. A long-term tensile-design stress of 6,900 to 10,340 kPa (1,000 to 1,500 psi) at 23°C (73°F) is recommended for most grades.

i

Ryton Supec Fortron AT-110 Torlon

Victrex Stabar

Ultem Upac

Radel

Victrex Ultrason E

Polyetheretherketone (PEEK)

Polytherimide (PEl)

Polyarylsulfone (PAS)

Polyethersulfone (PES)

Trade Name

Polyphenylene sulfide (PPS) Polyamideimide (PAl)

Resin

ICI BASF

General Electric American Cyanamid Amoco

350 (177)

350 (177)

356 (180)

345 (174)

340 (171)

430 (221)

600 (316)

450 (232)

525 (274)

430 (221)

425 (218°C) 464 (240)

500 (260°C)

Phillips 66 General Electric Hoechst Celanese Amoco Amoco

ICI

°F2

HDTl

Manufacturer Heat resistance, good electrical properties, chemical resistance Superior mechanical properties, wide temperature range (requires postmold treatments) Outstanding thermals, nonflammable, good chemical resistance, radiation resistance High strength and rigidity at elevated temperatures Low melt processing temperatures, hydrolytically stable Transparent, creep resistant at room temperature, high temperature resistance in air and water, low cost

Major Property Advantages

Table 6-1. Trade Names of Sample Unreinforced Thermoplastics

Electrical, electronics, automotive, transportation Electrical appliances, industrial applications

Electrical, industrial, appliances

Wire and cable insulation, automotive, PCBs

Automotive, electrical, industrial , consumer Aerospace motor parts, industrial seals, bearings

Applications

~

Victrex Ultrapek

Xydar Vectra Grantur Victrex SRP

Kapton Vespel Kinel Kermid Matrimid Upilex

Polyetherketone (PEK)

Aromatic copolyesters (based on hydroxybenzoic acid)

Polyimides (PI)

Ou Pont Ou Pont Rhone Poulenc Rhone Poulenc Ciba-Geigy Ciba-Geigy

Amoco Hoechst Celanese Granmont ICI

ICI BASF

Ou Pont Amoco Hoechst Celanese

Du Pont

IHeat dislortion temperature al 820 kPa (264 psi) per ASTM. 2continuous-use temperature per ASTM.

Arylon Bexloy M Ardel Ourel

Polyarylate (PA)

500 (260)

390 (199) 464 (240)

550 (288)

445 (229) 590 (310)

680 (360)

250 (121)

400 (204)

310-345 (154-174)

High temperature resistance, high mechanical properties, very good processability, excellent chemical resistance, extremely low coefficient of expansion Heat resistance, radiation resistance, good dielectrics, low coefficient of !hennal expansion

Thennal expansion close to metal, excellent dimensional stability, flame resistance, warp resistance, low water absorption Good !hennal properties, easy processing, excellent dielectrics

High-temperature films, electrical insulation and parts, mechanical parts, seals, bearings, aircraft, aerospace

Advanced composites, seals, bearings, PCBs, chemical components Electrical, electronics, automotive underhood parts, fiber-optic devices, aerospace, Tupperware

Automotive, appliances, industrial, electrical, electronics

Table 6-2. Design Properties Profile of ICI-LNP Thermocomp® Glass-Reinforced Thermo~lastics and Fluoromelt® Melt-Processi Ie Fluoropolymer Composites Physical

Property

Maximum Glass Fiber Available

Units

Percent

ICI-LNP Thermocomp®

Product Code (30 wt.% Glass Fiber)

Water

Absorption,24 hrs.

Mold Shrinkage

Strength

Flexural Modulus

%

in.fin.

psi

10' psi

D 792

D 570

D 955

D638

D790

Specific Gravity

Tensile

wt.

ASTM

Base Resin

AF-lOO6

40

1.28

0.14

0.001

14,500

1,100

BF-lOO6

40

1.31

0.10

0.005-0.001

17,400

1,500

CF-lOO6 NF-l006 ZF-lOO6

40 40 40

1.28 1.31 1.28

0.05 0.07 0.06

0.005-0.001 0.001 0.001-0.002

13,500 15,000 18,500

1,300 1,300 1,150

Polyethylene (HOPE) Polypropylene Polypropylene

FF-lOO6 MF-lOO6 MFX-l006'

40 40 40

1.17 1.13 1.13

0.Dl5 0.03 0.02

0.0030 0.004 0.0035

10,000 9,800 13,500

900 800 800

Nylon-Type 6 Nylon-Type 6/12 Nylon-Type 6/10 Nylon-Type 6/6 Amorphous Nylon Super-Tough Nylon

PF-lOO6 IF-lOO6 QF-lOO6 RF-lOO6 XF-l006 YF-lOO6

60 60 60 60 40 40

1.37 1.30 1.30 1.37 1.35 1.30

1.1 0.20 0.20 0.9 0.19 0.60

0.0035 0.0035 0.0035 0.004 0.003 0.004

23,000 22,000 21,000 26,000 21,500 17,000

1,200 1,100 1,100 1,300 1,150 900

Acetal Thermoplastic Polyester (PBT) Polyphenylene Sulfide Polyetheretherketone

KFX-l006' WF-l006

40 40

1.63 1.52

0.30 0.06

0.003 0.003

19,500 20,000

1,400 1,200

OF-lOO6 Victrex* PEEK 45OGL30

40 40

1.56 1.49

0.04 0.11

0.002 0.003

20,000 22,800

1,600 1,500

Polycarbonate Polysulfone Polyethersulfone

OF-lOO6 GF-lOO6 Victrex* PES 4 \0 I GL30

40 40 40

1.43 1.45 1.60

0.07 0.20 0.34

0.002 0.002-0.003 0.002

18,500 18,000 20,300

1,200 1,200 1,218

Thermoplastic Polyurethane Polyester Elastomer

TF-IOO6

40

1.46

0.25

0.004

8,200

190

YF-lOO6

40

1.42

0.17

0.004

10,000

320

FEP PFA ETFE

FP-FF-IOO4' FP-PF-10034 FP-EF-lOO6

20 15 30

2.21 2.20 1.89

0.01 0.005 0.02

0.002-0.004 0.012 0.004-0.006

6,000 5,500 14,000

800 430 1,050

Acrylonitrile-ButadieneStyrene (ABS) Styrene-Acrylonitrile (SAN) Polystyrene Styrenic Copolymer Modified Polyphenylene Oxide

·Chemically coupled 2Definition---Air temperature that could cause the 30% glass reinforced resin to lose 50% of its mechanical properties at 100,000 hours. 320% glass fiber reinforced 415% glass fiber reinforced Victrex is a registered trademark of leI. "'UL recognized flame retardant versions available. "''''Values on heat stabilized (H.S.) composites. tUL recognized. :j:Shore D hardness. For additional aid in material comparisons, these resins have been grouped as follows: Styrenics. Olefins, Nylons, Crystallincs. Arylates. and Miscellaneous. For ad(lItlOnal glass-reinforcement levels and other lubricants such as PTFE, MoS; see specific product data sheets, ICI-LNP.

410

Mechanical

Thennal

Izod Impact

Strength Notched! Unnotched

Rockwell Hardness

Electrical

Heat Deflection Temperature @ 264 psi

Long Term2 Service Temperature

Coefficient Linear Thennal Expansion

Flammability

Dielectric Strength (ST)

of

of

10-5 in'/in./°F

UL Subj. 94

volts/mil.

of

Dielectric Constant 60 Hz-IO' Hz

Dissipation Factor 60 Hz10' Hz

D 149

D 150

D 150

ft.-Ibs.lin. D 256

D 785

D 648

1.4/6-7

M99.R124

220

155

1.6

HB*

1.0/3-4

M94.R123

215

140

1.8

HBt*

1.0/2-3 l.l/4-5

M92 M96 M93

215 250 310

120

HB HB

550

2.81-2.81

0.0007-0.0008

195

1.9 1.8 1.4

HBt*

550

2.90-2.90

0.0010-0.0015

1.118-9 1.6/5-6 1.9/10-12

R85 M57.Rlll M57.Rlll

260 295 310

185 220 220

2.7 2.0 2.1

HBt*

475

2.30-2.20

0.001-0.003

2.3120 2.4/20 2.4120 2.0/17 1.2/6--8 4.0/20-22

M92.RI21 M93.R120 M93.R120 M96.R121 R1l9 RI20

420 415 420 490 285 415

215** 210**

I.7 1.5 1.5 1.8 1.8 1.9

HBt*

HBt* HBt*

450 440 440 440

4.20-3.60 4.20-3.50 4.20-3.50 4.20-3.50

0.009-0.018 0.013-0.015 0.013-0.015 0.009-0.0180

HBt

500

1.8/8-10 2.6116--18

M86 M84.R1l9

300 430

220 280

2.2 1.2

HB HBt*

525 510

3.95-3.95 3.60-2.00

0.0035-0.0065 0.002-0.020

1.4/8-9 1.8/15-16

RI21

500 600

355 482

I.3 1.2

vot vot

510 500

3.88-3.78 3.71-3.61

0.003-0.007 0.0019-0.0043

3.7117-18 1.8/14 1.6/10

M95.R1l8 M92.LI08 M98

300 365 420

260 300 374

I.3 1.4 I.3

vtt* vot* vo*

480 480 460

3.50-3.43 3.55-3.49 3.80-3.76

0.0010-0.0075 0.0019-0.0049 0,0030-0,004

9.5/28-29

D65:j:

340

140

2.5

HB

5.0120

D70:j:

340

150

2.5

HB

8.0/17 5.8/17-18 7.5/17-18

D63:j:

350 463 460

390 500 350

2.4 5.5

VO VO VO

475 580 410

2.55-2.52 2.35-2.45 3.5-3.4

0.0020-0.0002 0.002 0.0006-0.005

R74

D 696

240** 185

HB HB

HB HB

411

412 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

An ABS part can usually be bent beyond its elastic limit without its breaking, although it will stress whiten. Although they are not generally considered flexible, ABS parts have enough spring to accommodate the snap-fit assembly requirements of lugs, rings, and buttons. They can use such mechanical fastening methods as holes for rivets and bosses for self-taping screws and can be solvent welded, adhesive bonded, and ultrasonic welded. Molded ABS parts are used in both decorative and protective applications: in communications equipment like telephones (see Fig. 6-3), and for appliances, businessmachine housings, safety helmets, camper tops, pipe fittings, shoe heels, and so on. Chrome-plated ABS has replaced die-cast metals in plumbing hardware and auto grills, wheel covers, and mirror housings. Some typical products vacuum-formed from extruded ABS sheet are refrigerator liners, luggage shells, tote trays, mower shrouds, boat hulls, and recreational vehicle parts. The possible extruded shapes include weather seals, conduit, pipe for drain-waste-vent (DWV) systems, and the like.

Styrene-Acrylonitrile Related to ABS, styrene-acrylonitrile (SAN) is hard, rigid, and transparent. It has no butadiene. It is characterized by excellent chemical resistance, good dimensional stability, and ease of processing. All fabricating processes can be used with SAN, but because it is not toughened, thermoformed shapes may crack during conventional trimming. Special grades of SAN are available that have improved UV stability, vapor-barrier characteristics, and weatherability. The barrier resins, which were designed for the blown-

a Figure 6-3. One way to inject mold an ABS telephone part: a) The complete molded part (bottom), a cross-section (middle), and the two-part metal insert used to shape the interior, which after molding is shaken out; b) the bottom half of the mold including the two-part metal inserts placed on each mold cavity; and c) the molded part containing the two-part metal inserts that are here being removed and shaken to dislodge the metal inserts (other molding methods do not include such removable metal inserts).

THE PROPERTIES OF PLASTICS 413

b

c bottle market, are also tougher and have greater solvent resistance. A few of the typical applications for general-purpose SANs include lenses, vacuum cleaner and humidifier parts, medical syringes, battery cases, food-mixer bowls, and dishwasher-safe houseware products. Because of their compatibility with many higher-priced resins SANs are also used as color-concentrate carriers for some engineering resins.

414 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

ACETAL The superior properties of acetal in terms of its strength, stiffness, and toughness make it an important engineering thermoplastic. It is more dense than nylon, but in many respects their properties are similar and they can be used for the same types of light engineering applications. In some cases a factor that may favor acetal is its relatively low water absorption. The acetal polymer (homopolymer) and copolymer grades offer the designer features that have made them prime contenders for applications based on metal, primarily the die-case metals. Its properties include good fatigue life; exceptional dimensional stability, resiliency, and toughness; good tensile strength and creep resistance under a wide range of temperature and humidity conditions; high solvent resistance and excellent electrical properties, among others. Its abrasion resistance is also generally superior to that of most thermoplastics. Acetal's coefficient of friction on steel is quite low. Its low dissipation factor and dielectric constant are maintained over a wide range of frequencies and up to temperatures of 121°C (250°F), and it is available in colors. Acetal homopolymers when modified can deliver up to seven times greater toughness than when left unmodified in Izod impact tests, and up to thirty times greater toughness as measured by Gardner impact tests. The general-purpose types can be used over a wide range of environmental conditions. For instance, special UV-stabilized grades are used when long-term exposure to weathering is required. Their prolonged exposure to strong acids and bases outside the range of pH 4 to 9 is not recommended. The homopolymers have the highest fatigue endurance of any unfilled commercial thermoplastics. Under completely reversed tensile and compressive stresses, and with 100 percent relative humidity at 23°C (73°F) , their fatigue-endurance limit is 31,000 kPa (4,500 psi) at 106 cycles. They have excellent resistance to creep. Moisture, lubricants, and solventsincluding gasoline and gasohol-have little effect on this property, which is important in parts incorporating self-threading screws and interface fits. The melting points of the homopolymers are higher and they are harder, with more resistance to fatigue, than the copolymers. They are also more rigid and have higher tensile and flexural strength, with a generally lower rate of elongation. Some highmolecular-weight homopolymers are extremely tough and have higher elongation than the copolymers. Homopolymer grades are available that have been modified for improved hydrolysis resistance to 82°C (180°F), similar to the copolymers. Acetal copolymers have an excellent balance of between properties and processing characteristics. Their melt temperature can range from 182 to 232°C (360 to 450°F) with little effect on a part's strength. Acetals are available in translucent-natural white and a wide range of colors and various dimensionally stable, low-warpage grades. They have high tensile and flexural strength, fatigue resistance, and hardness. Among the most creep resistant of the crystalline TPs, they retain much of their toughness through a broad temperature range. Copolymers' strength is only slightly reduced after aging for one year in air at 116°C (240°F). Their impact strength holds constant for the first six months, then falls off by about one-third during the next six months. Aging in air at 82°C (180°F) for two years has little or no effect on their properties, and their immersion in water for one year at that temperature leaves most properties virtually unchanged. Samples tested in boiling water retained nearly their original tensile strength after nine months. Their good electrical and high mechanical properties, notably their UL electrical rating for 100°C (212°F), qualify these plastics for electrical parts requiring long-time stability. Copolymers have excellent resistance to chemicals and solvents. For example, samples

THE PROPERTIES OF PLASTICS 415

immersed for twelve months at room temperature in various inorganic solutions were unaffected except by the strong mineral acids: sulfuric, nitric, and hydrochloric. Most organic reagents tested have no effect on them, nor do mineral oil, motor oil, or brake fluids. Their resistance to strong alkalis is exceptionally good. The copolymers remain stable in long-term, high-temperature service and offer exceptional resistance to the effects of immersion in water at high temperatures. Neither type resists strong acids, and copolymers are virtually unaffected by strong bases. Both types are available in a wide range of melt-flow grades, unreinforced and reinforced grades, and PTFE or silicone-filled grades. Several grades of the homopolymer and copolymer types comply with FDA requirements for repeated contact with food at temperatures to 121°C (250°F).

ACRYLICS

Acrylic thermoplastics (polymethylmethacrylate, or PMMA) are known for their crystal clarity and outstanding weatherability. They are available as cast sheets, rods, and tubes; in extruded sheet and film form, and as compounds for the various fabricating processes. Injection molding and extrusion compounds are available in both standard and highmolecular-weight grades. The property differences between the two formulations are principally in their flow and heat resistance. The higher MWs have lower melt-flow rates and greater strength while hot during processing. The lower MWs are designed for making complex parts in hard-to-fill molds. Also available are high-impact grades that provide the same transparency and weatherability as conventional PMMAs. Clear acrylic is as transparent as the finest optical glass. It has a light transmission capacity of 92percent, an exceptionally low haze level of about 1 percent, and an index of refraction of 1.49, high enough for use in lenses and other optical parts. Colorants can be used to produce a full spectrum of transparent, translucent, or opaque colors. Most colors can be formulated in acrylics for long-term outdoor durability. Acrylics are normally formulated to filter ultraviolet energy in the 360 nm and lower band, but some are opaque to UV light or provide reduced UV transmission. Acrylics' mechanical properties are high for short-term loading, but for long-term service the tensile strength must be limited to 31,000 kPa (1,500 psi) to avoid crazing or surface cracking. Although acrylics are among the most scratch resistant of the TPs, normal maintenance and cleaning operations can scratch and abrade them. A special abrasion-resistant sheet is available that has the same optical and impact properties, even under extreme cold, as standard grades. Acrylics' toughness, as measured by their resistance to crack propagation, can be improved severalfold by including a particular molecular orientation during forming (see Chapter 2). Jet-aircraft cabin windows, for example, which must be rated to last for decades, are made from oriented sheet. Acrylics' transparency, gloss, and dimensional stability are virtually unaffected by as many as thirty to fifty years of exposure to the elements, salt spray, or corrosion atmospheres. They withstand exposure to fluorescent lamps without darkening or deteriorating. They will ultimately discolor, however, when exposed to high-intensity UV light below 265 nm. Special formulations are able to resist UV emissions from such light sources as mercury-vapor and sodium-vapor lamps.

416 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

ALKYDS

Alkyd thermoset compounds are based on unsaturated polyester resins, which are combined with cross-linking monomers, catalysts, reinforcements, lubricants, and fillers. They are similar to TS polyesters, but have lower levels of monomers. They are part of the group that includes the bulk-molding compounds (BMCs) and sheet-molding compounds (SMCs). Their fast molding cycles at low pressure make them easier to process than other TSs by vacuum, compression, transfer, or injection molding. Alkyd compounds are furnished in granular form, extruded rope and logs, as a BMC, and in puttylike grades. Except for the latter, which can be used for encapsulation, these compounds contain fibrous reinforcement. Because the fillers are opaque and the resins amber, translucent colors in alkyds are not possible. However, opaque light shades can be produced in most colors. Low moisture absorption and excellent dimensional stability and electrical properties are the outstanding characteristics of most alkyd compounds. In the electrical grades, moisture absorption can be as low as 0.5 percent. Alkyds are relatively low-loss materials, especially the mineral- or glass-filled grades. Those with cellulose may have higher loss but will drift with temperature and humidity changes. The glass grades as a whole have better heat resistance than the cellulose types. Depending on the type, alkyds can be used continuously up to 177°C (350°F) and for short periods up to 232°C (450°F). Alkyds' dimensional stability in regard to their mechanical and electrical properties are retained over wide temperature ranges. Alkyd parts will resist weak acids, organic solvents, and hydrocarbons such as alcohol and the fatty acids, but they are attacked by alkalis. Halogen- or phosphorus-bearing alkyds with antimony trioxide added will provide improved flame resistance, and other flame-resistant compounds are available that do not contain halogenated additives. Many grades are UL rated at 94V-O in sections under 1116 in. Their flammability ratings, which depend on their specific formulations, can vary from 94HB to V-O, and may vary in section thickness. High-impact alkyds with a high glass content are used in switchgear demanding high performance, electrical terminal strips, and relay and transformer housings and bases. The mineral-filled grades, which can be modified with cellulose to reduce their specific gravity and cost, are used in automobile ignitions, ratio and TV parts, switchgear, and for small-appliance housings. Alkyds with all-mineral fillers have high moisture resistance and are particularly suited for use in electronic components. Certain grades are available that can withstand the temperatures of vapor-phase soldering.

AMI NOS There are two basic types of aminos: urea formaldehydes (UF) and melamine formaldehydes (MF). These thermosets are hard, rigid plastics with good abrasion resistance. Their mechanical properties are sufficiently good for continuous use at moderate temperatures, up to 100°C (212°F). UF is relatively inexpensive, but its propensity for moisture absorption can result in poor dimensional stability. It is generally used for bottle caps, electrical switches, plugs, utensil handles, and trays. MF has lower water absorption and improved temperature and chemical resistance. It is used for tableware, laminated worktops, and electrical fittings. See the melamine and urea sections that follow in this chapter for more details.

THE PROPERTIES OF PLASTICS 417

CELLULOSICS Cellulosic is a family name that applies to a wide group of thermoplastics. They are not synthetic plastics but rather are made from a naturally occurring polymer, cellulose, which is obtained from wood pulp and cotton linters. Cellulose can be made into a film as cellophane or a fiber, rayon, but it must be chemically modified to produce TPs. Because it can be compounded with many different plasticizers in widely varying concentrations, its property range is broad. These plastics are normally specified by their flow, according to ASTM D 569, which is controlled by the plasticizer content. Cellulosics are all processed by conventional TP methods. They include the following types.

Acetates Cellulose acetate (CA), commonly called simply acetate, is noted for its attractive appearance, toughness, and high impact strength. It is used in quality toys, appliance handles, eye-glass frames, pen barrels, caps, and electrical parts. Acetate extruded and cast film and sheet are thermoformed for packaging. Extruded acetate rods have found great popularity as tool handles.

Butyrates Cellulose acetate butyrates (CAB) are tough, transparent, and water resistant. Some typical uses include molded steering-wheel covers, data keyboards and cash register keys, transparent dial covers, tool handles, and street globes. Butyrate sheet is thermoformed for signs and displays, blister packaging, transparent food packaging, and building panels.

Propionates Cellulose acetate propionates (CAP) are abrasion resistant and are used in automobile parts, toothbrush handles, cosmetic containers, face shields, fuel filters, safety goggles, and similar items. Ethyl Cellulosics

Ethyl cellulose (EC) is characterized by toughness over a wide temperature range, dimensional stability, and freedom from odor. Its uses include helmets, gears, slides, flashlight housings, and tool handles.

Nitrates Cellulose nitrate was the first of the plastics to be developed commercially, in 1868, originally to make billiard balls. Because of its flammability its use today is relatively little. It is available in sheet, film, rod, and tube forms that can be fabricated into personal accessories or toilet articles. It is also available as a solution for coatings. CHLORINATED POL YETHERS

Chlorinated polyethers are corrosion- and chemical-resistant thermoplastics whose prime use has been to manufacture products and equipment for the chemical and processing

418 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

industries. It has also found use in molding components for pumps and water meters, pump gears, bearing surfaces, and the like. This plastic resists both organic and inorganic agents, except fuming nitric acid and fuming sulfuric acid, at temperatures up to 121°C (2500 P) or higher. Its heat-insulating characteristics, dimensional stability, and outdoor exposure resistance are also excellent. CHLORINATED POLYETHYLENE

Thermoplastic CPEs are amorphous and crystalline, thus providing a wide range of properties from soft and elastomeric to hard. These materials have an inherent oxygen and ozone resistance, have improved resistance (compared to polyethylenes) to chemical extraction, resist plasticizer volatility and weathering, and have exceptionally high tear strength, heat and aging characteristics, and excellent oil and chemical resistance. Products made from CPEs do not fog at higher use temperatures, as do PVCs, and can be made completely flame retardant. However, they exhibit a chemical instability that is similar to PVC's. They may be used as primary compounding materials or be blended with a PVC, high- or low-density polyethylenes, and other resins to add their benefits. Extruded sheet, supported by foams, is used in certain automobile dashboards and upholstery and door liners. CROSS-LINKED POLYETHYLENE (XLPE)

See the discussion under "Polyethylene." DIALL YL PHTHLATE

Diallyl phthlates (DAP) and diallyl isophthlates (DAIP) are the principal thermosets in the allyl family, with DAP used predominantly. They are used for glass-preimpregnated cloth and paper that must undergo a heat, time, and pressure cycle to produce parts. Molding compounds are reinforced with fibers to improve their mechanical and physical properties. Glass fibers impart mechanical performance, acrylic fibers provide improved electrical performance, polyester fibers enhance impact resistance, and other fibers and fillers can impart different performance traits. In some applications DAPs are competitive with TS polyester compounds. They can, for instance, offer longer shelflife (as the BMCs), less shrinkage during curing, somewhat better chemical or electrical properties, and higher heat resistance. In general, the allyls are more expensive and therefore, find few uses in consumer products. They can, however, be molded at lower pressures and in faster molding cycles. With a triallyl cyanurate formulation, products will withstand temperatures ranging as high as 260°C (5000 P) (used, for example, in high-speed radomes). One major advantage of all allyls over the TS polyesters is their freedom from styrene odor, low toxicity, low evaporation losses during fabricating evacuation cycles, no subsequent oozing or bleed out, and their long-term retention of their electrical-insulating performance. The major use of DAPs is in electrical connectors in communications, computer, and aerospace systems. Their high thermal resistance permits their use in vapor-phase soldering. Their other uses are for arc-track-resistant switchgear and television components, circuitboards, and the like.

THE PROPERTIES OF PLASTICS 419

EPOXIES The family of epoxy thermoset resins (EPs) includes epichlorohydrin with bisphenol-A. These most widely used epoxies range from low-viscosity liquids to high-molecularweight solids. The novolacs are another important class that offer higher thermal properties and improved chemical resistance. The cycloaliphatic types also are important, for applications requiring high resistance to arc tracking and weathering. Epoxies are more expensive than other equivalent plastics, such as the TS polyesters, but they outperform these materials because of their improved performance. Their general properties include toughness, having less shrinkage during curing, good weatherability, low moisture absorption, curing without the evolution of by-products, good wetting and adhesion to a wide variety of surfaces, good mechanical properties and thermal capabilities, excellent fatigue resistance, outstanding electrical properties from low to high temperatures, exceptional water resistance, practically complete resistance to fungus, general corrosion resistance, and other such characteristics. The variety of combinations available in epoxies and reinforcements provides wide latitude in the properties of fabricated parts. Some fiber-reinforced and composite materials can withstand service temperatures even above 260°C (500°F) for brief periods. Their excellent electrical and mechanical performance qualifies them for use in many electrostructural parts. The EPs are used in all the methods of processing plastics. Filled and liquid systems are used for potting and encapsulating electronic and other components, producing excellent adhesion. The casting cycle can be significantly accelerated by using liquid or reaction-injection molding. Another important use for epoxies is in coatings, both as liquids and powders. Such finishes have excellent flexibility, impact and abrasion resistance, are decorative and corrosion resistant, and so on. They adhere to most substrates of plastic, steel, aluminum, and other materials. And they are exceptional adhesives to bond similar or different materials of plastic, steel, aluminum, wood, or glass.

ETHYLENE-VINYL ACETATES Ethylene-vinyl acetates (EVA) copolymers are in the polyolefin family of thermoplastics. They are used in all processes, particularly for extrusion, injection molding, and blow molding. They are used either alone or are coextruded or coinjected and used in compounds to provide unique properties. They approach elastomeric materials in their softness and flexibility. EVA parts have good clarity and gloss, stress-crack resistance, barrier properties, lowtemperature toughness, adhesion, resistance to UV radiation, little or no odor, and retain their flexibility at low temperatures. Their main limitation is their comparatively low resistance to heat and solvents. Chlorinated hydrocarbons, straight-chain paraffinic solvents, and benzene all attack EVAs' resins. However, alcohols, glycols, and weak organic acids do no damage. EVAs are used principally in specialty parts, competing with PVC and rubber. FDA approval exists for their use in direct contact with food. Some EVA products include medical tubing, tubing for beverage vending, milk-packaging and beerdispensing equipment, appliance bumpers, blow-molded bellows for seals, gaskets, and toys, and in hot-melt adhesives.

420 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

FLUOROPLASTICS Auoroplastics (FP) are a class of thennoplastic waxy perfonnance plastics in which all or some of the hydrogen atoms that would typically be bonded to carbon atoms in hydrocarbon plastics (which is to say most plastics) have been replaced by fluorine atoms. Another halogen, chlorine, can also be part of the structure. They are distinguished by a combination of properties (Table 6-3) that includes outstanding resistance to virtually all chemicals in a wide temperature range, superb antisticking characteristics because of their extremely low coefficient of friction, excellent dielectric and nonaging properties, and inherent flame retardancy. Although their mechanical properties are low for engineering plastics, these can be boosted sharply by reinforcements like glass, carbon, and bronze, as well as molybdenum disulfide and other modifiers. These properties have made FPs almost indispensable in the manufacture of products including high-perfonnance sliding mechanisms, corrosion-resistant chemical-process linings, wire and cable core insulation, medical and industrial tubing, and self-lubricating bearings. Not surprisingly, its major area of application is in bearings, particularly if the environment is an aggressive one. FPs are also used widely in areas such as insulating tapes, gaskets, pumps, diaphragms, and popular nonstick coatings on cooking utensils. Despite the impressive perfonnance credentials of the original FP, called polytetrafluoroethylene (PTFE), which was developed in the 1930s and used ever since, its applications were limited by a high melt viscosity that still makes it unsuitable for conventional processing. PTFE in fact never really melts; it just becomes a self-supporting gel above 327°C (620°F). It is processable, but only by using very specific timet temperature/pressure techniques of compression-sintering, ram extrusion, and isostatic molding. Since the 1960s, copolymerization techniques for FPs have produced different FPs with different degrees of ease of processing. In tum, perfonnance trade-offs had to be made, but the significant properties were nevertheless retained, in differing degrees.

Table 6-3. General Properties of Fluoroplastics (arrows show increasing values of the property) and Designations Properties of FrictionCharacter-Thermal Stability-> -Mechanical Strength at High Temp.-> -Softening Temperature-> -Antistick-> ~ohesive Forces--Creep-> ~Dielectric Constant-Chemical ResistanCe-> -Solvent ResistanCe-> ~Mechanical Strength at Ambient Temp.-Permeability-> ~Processing Ease--Oxidative Stability->

Designations PTFE or TFE FEP

~oefficient

~ ~

.... Z 0

u ~

Z

C2

0

:=>

...J ~

~

0

...J

~Adhesive

.... Z

~

Z 0 U

~

~

0

:=>

ti

::z::

~

CTFE or PTFCE PVF PVF2 or PVDF ETFE ECTFE PFA

Poly tetrafluoroethylene Copolymer of hexafluoropropylene and tetrafluoroethylene or fluorinated ethylene propylene Polychlorotrifluoroethylene Polyvinylftuoride Polyvinylidenefluoride Copolymer of ethylene and tetrafluoroethylene Copolymer of ethylene and chlorotrifluoroethylene Polyperftuoroalkoxyethylene

THE PROPERTIES OF PLASTICS 421

Standard injection and extrusion equipment is used with the lower-viscosity polymers fluorinated ethylene propylene (FEP) , ethylene tetrafluoroethylene (ETFE), polyvinylidenefluoride (PVDF), polyperfluoroalkoxyethylene (PFA), ethylene chlorotrifluoroethylene (ECTFE), and others. This substitution, or copolymerization, where fluorine atoms have substitutes, results in property and processing changes.

FURAN Furan resin is a generic term for a thermoset resinous product that contains a heterocyclic unsaturated furan ring in its molecular structures. Pentosans from com cobs and rice hulls are the main sources for the key ingredient, furfural. Commercially, the furfural alcohol polymer is the most important. All furan resins are dark in color and have a reddishblack appearance; when catalyzed to cure they become black. Their biggest use in the corrosion-resistance field is in the manufacture of chemicalresistant cements and equipment. For instance, furan cements have been used for years to bond acid proof brick. The surfaces of the brick may be saturated with alkaline substances and mineral or organic acids as well as many solvents, alone or in combination. Cements can be used in areas where it would be impossible to use other construction materials. They offer corrosion protection to concrete and steel structures, which lengthens their life. As pump base cements, furans can be used in chemical-processing plants, metal-finishing plants, petroleum refineries, fertilizer plants, and pulp and paper mills. The floors and walls of such structures as processing tanks, continuous-strip pickle lines, processing towers, collecting sumps, neutralizing tanks, pits, manholes, and tank cappings can be protected with furan cements. Furans have also found wide use in the manufacture of grinding wheels and foundry molds. A typical furan-based composite possesses good heat and chemical resistance, excellent surface hardness, and is inherently nonflammable. However, their use in the form of fiber-reinforced composites is still comparatively uncommon.

IONOMERS Ionomers are in the polyolefin family. Their interchain ionic bonding distinguishes them from the other polymers. These ionic cross-links occur randomly between long-chain molecules to produce properties usually associated with high-molecular-weight materials. At normal processing temperatures, however, the ionic bonding of these thermoplastics diminishes, allowing them to be processed in conventional extruders and injection-molding machines. Ionomers are extremely tough, with tensile impact strengths as high as 320 J/cm (600 ft.-Ib'/in.) and tensile strengths as high as 35,000 kPa (5,000 psi), with elongation in the range of 300 to 500 percent. In addition, they have excellent abrasion resistance, with an NBS index as high as 640, and optical clarity, a haze rating as low as 40 percent. Compounded ionomers are also available that are stiffer and have better heat resistance than standard grades yet retain their excellent impact resistance. This product, intended for semirigid parts, resists many chemicals, solvents, greases, and oils. The clarity, strength, and good adhesion of ionomer films to metal surfaces are the important properties that have led to its widespread use in food packaging, often as a heat-seal layer in thermoplastic composite structures. Its high impact strength and cut resistance have led to its use in bowling pin and golf ball covers and bumper guards. Its

422 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

automotive uses are based on its impact toughness, light weight, and paintability. Foam injection-molded parts have now replaced heavy rubber and metal bumper guards and license-plate holders. Foams are also used in ski-lift pads, boat bumpers, marine navigation buoys, and wrestling mats. Foamed sheet is used for thermal insulation on pipes and in covers for hot-water storage tanks. Ionomers' footwear uses include box toes, heel counters, and injection-molded athletic soles to which metal cleats can be insert molded or spin welded. In ski boots and ice skates, ionomers provide lightweight, durable shells. KETONES

This broad family of crystalline thermoplastics includes polyetherketone (PEK) and polyetheretherketone (PEEK). See the sections that follow on PEK and PEEK. LIQUID CRYSTAL POLYMERS

Liquid crystal polymers (LCPs), sometimes called super polymers, which were commercially introduced in 1984, are called self-reinforcing plastics because of their densely packed fibrous polymer chains. LCPs are creating new design opportunities for plastics with their exceptional range of properties. They have outstanding strength at extreme temperatures, excellent mechanical-property retention after exposure to weathering and radiation, good dielectric strength as well as arc resistance and dimensional stability, a low coefficient of thermal expansion, excellent flame resistance, and easy processability. Their UL continuous-use rating for electrical properties is as high as 240°C (464°F), and for mechanical properties it is 220°C (428°F). LCPs' high heat deflection value permits LCP molded parts to be exposed to intermittent temperatures as high as 315°C (60(tF) without affecting their properties. Their resistance to high-temperature flexural creep is excellent, as are their fracture-toughness characteristics. LCPs' ease of processing gives them the ability to fill long, narrow molds, which makes them eminently suitable for such high-performance parts as electronic connectors. They may have 138,000 kPas (20,000 psi) or more in tensile strength, with flexural modulus values up to 5 X 106 psi. They are available in grades with heat deflection temperatures (HDT) of about 357°C (675°F) at 1,820 kPas (264 psi). This family of different LCPs resists most chemicals and weathers oxidation and flame, making them excellent replacements for metals, ceramics, and other plastics. LPCs are available in both amorphous and crystalline grades. The amorphous types, with their high strength-to-weight ratios, are particularly useful for weight-sensitive items in aerospace and military parts. Most LCPs can be injection molded, extruded, thermoformed, and blow molded. The crystalline grades, with glass and other fibers, meet the dimensional requirements and stability at high temperatures required of products for the electrical and electronics markets. The LCPs are all exceptionally inert and resist stress cracking in the presence of most chemicals at elevated temperatures, including the aromatic and halogenated hydrocarbons as well as strong acids, bases, ketones, and other aggressive industrial products. Their hydrolytic stability in boiling water is excellent, but high-temperature steam, concentrated sulfuric acid, and boiling caustic materials will deteriorate LCPs. In regard to flammability, LCPs have an oxygen index ranging from 35 to 50 percent. When exposed to open flame they form an intumescent char that prevents dripping and

THE PROPERTIES OF PLASTICS 423

results in an extremely low level of generation of smoke, which contains no toxic byproducts. Its resins have UL 94 V-O and 5V flammability ratings at 1116 in. and an NBS smoke-chamber rating (NBS-Ds-4) of 3 to 5. Its compounds are unaffected by high doses of ionizing or cobalt-60 radiation (up to 10 billion rads), can withstand high levels of ultraviolet radiation, and are transparent to microwaves and other radiation at similar wavelengths. LPCs' molecular structure is attributed to their ease of processing. However, molded LCP parts are highly anisotropic, and weld lines in them tend to be much weaker than would normally be expected. Their properties are not affected by minor variations in processing conditions, and no postcuring is required to obtain their outstanding properties. The major applications of LCPs are in metal and ceramic replacements that require resistance to high temperatures, chemicals, mechanical stress, creep resistance, and so forth. LCP parts include electronic and electrical connectors, sockets and pin-grid arrays exposed to high-temperature manufacturing or service conditions, automotive and aerospace parts that require the ability to withstand high temperatures and flame retardance, and chemical-processing components that exist in aggressive environments.

MELAMINES Melamine formaldehyde (MF) is one of two major thermoset resins in the amino family, the other being ureaformaldehyde. Various kinds of fillers are used to make MF compounds to meet different requirements. MF is rigid and possesses a hard surface capable of withstanding continuous handling and wear with negligible effect. Moreover, its surface is unaffected by common organic solvents, greases, and oils, as well as many weak acids and alkalis. When properly molded into food containers and dishes, an MF does not impart odor or taste to solid or liquid foods. There are MF compounds that are insensitive to heat and are highly flame resistant, depending on the fill used. They are recommended for maximum temperatures ranging from 99 to 121°C (210 to 250°F). Low temperatures produce no observable effects on MFs. MFs are satisfactory for the large majority of electrical applications and are particularly useful where arc resistance is desired. Mineral-filled MFs have one of the highest arc resistances of any plastic plus high dielectric strength and dimensional stability but low moisture absorption. With chopped cotton rags added, an MF has high flexural strength, will absorb considerable shock, and will not support combustion. An alpha-cellulose-filled MF is inherently colorless, light fast, and translucent. By properly choosing pigments and dyes, an unlimited range of stable, unfading colors can be obtained, as well as a wide range of translucencies. MF moldings have good dimensional stability, high dielectric properties, and are little influenced by high humidity and water. Its strength and shock resistance are also good. A major use of MFs with alpha cellulose is in heavy-duty dishware. Decorative dinnerware with printed inlays (that is, designs located below the surface) cannot be washed off, abraded, or damaged in any manner. Surface glazing can be used to eliminate staining or scratching. Properly designed, they are practically unbreakable (see also "Urea," below).

NYLON (POLYAMIDES) Nylon was the first of the so-called thermoplastic engineering plastics in the 1930s. They were originally developed as high-strength textile fibers for stockings. These crystalline

424 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

resins-the new developments include amorphous types-are available for processing by different methods. There are different nylons, but as a family their characteristics of strength, stiffness, and toughness have earned them an important place for the designer when compared to such other materials as the die-casting alloys (see Table 6-4). Nylon 6/6 is the most widely used, followed by nylon 6, with similar properties except that it absorbs moisture more rapidly and its melting point is 21°C (70°F) lower. Also, its lower processing temperature and less-crystalline structure result in lower mold shrinkage. Nylons 11 and 12 have better dimensional stability and electrical properties than the others, because they absorb less moisture. These more expensive types also are compounded with plasticizers, to increase their flexibility and ductility. With nylon toughening the technology advancements, supertough nylons have become available. Their notched Izod impact values are over 10 J/m (20 ft.-Ib.!in.), and they fail in a ductile manner. Other important types of nylons include the castable types, liquid monomers (not the usual solid) that polymerize and become solid at atmospheric pressure. From them complex parts several inches thick and weighing hundreds of pounds can be cast. Another castable liquid monomer is a moldable transparent material. This amorphous type offers better chemical resistance than other thermoplastics that the transparent. Property comparisons among the commercial grades of different nylons vary widely, because so many formulations are available. In general, they all have excellent fatigue resistance, a low coefficient of friction, good toughness (depending on their degree of crystallinity), and resist a wide spectrum of fuels, oils, and chemicals. Nylon 6/6 has the lowest permeability by gasoline and mineral oil of all the nylons. The 6/10 and 6/12 types are used where lower moisture absorption and better dimensional stability are needed. All nylons are inert to biological attack and have electrical properties adequate for most voltages and frequencies. The crystalline structure of nylons, which can be controlled to some degree by processing, affects their stiffness, strength, and heat resistance. Low crystallinity imparts greater toughness, elongation, and impact resistance-but at the sacrifice of tensile strength and stiffness. All nylons absorb moisture, if it is present in the application's environment. An increase in moisture content decreases a material's strength and stiffness and increases its elongation and impact resistance. Type 6/6 nylon usually reaches equilibrium at about 2.5 percent moisture when the relative humidity reaches 50 percent. The eqUilibrium moisture at 50 percent RH in nylon 6 is slightly higher. In general, nylon's dimensions increase by about 0.2 to 0.3 percent for each 1 percent of moisture absorbed. However, dimensional changes caused by moisture absorption can be compensated for by performing moisture conditioning prior to putting parts into service. Such formulations as 6/12, 11, and 12 are considerably less sensitive to moisture than others. When UV stabilizers are compounded in the nylon, they become insensitive to UV light. Carbon black is the most effective stabilizer. UV stabilizers also increase tensile strength and hardness and decrease ductility and toughness slightly. Nylons have good resistance to creep and cold flow, as compared to many of the less rigid thermoplastics. Usually, creep can be accurately calculated, using apparent modulus values, as seen in Chapter 3. They also have outstanding resistance to repeated impact. Nylons can withstand a major portion of a breaking load almost indefinitely. Nylons are used in many different markets, the largest being the automotive. Their performance capabilities make them suitable for different mechanical and electrical hardware, particularly for such under-hood parts as timing sprockets, speedometer gears, cooling fans, wire connectors, windshield-wiper parts, door latches, fender extensions, steering-column-lock housings, brake-fluid reservoirs, and other uses. Their low friction,

THE PROPERTIES OF PLASTICS 425

Table 6-4. Typical Nylon Performance as Compared to Die-cast Alloys Points for Comparison

Die-casting Alloys

Cost of raw materiallton Cost of mold Speed of component production Accuracy of component Postmolding operations

Low High Slower than injection molding of nylon Good Finishing-painting Paint chips off easily

Surface hardness Rigidity

Low-scratches easily Good to brittleness

Elongation

Low

Toughness (flexibility)

Low

Impact Notch sensitivity Young's modulus (E) General mechanical properties

Low Low Consistent Similar to GR grades of 6/6 nylon High Low High Snap fits difficult

Heat conductivity Electrical insulation Weight Component assembly

Nylon High Can be lower-no higher Lower component production costs Good Finishing-not required; painting-not required. Compounded color retention permanent. Much higher. Scratch resistant. Glass-reinforced grades as good or better GR grades comparable; unfilled grades excellent GR grades comparable; unfilled grades excellent All grades good Low Varies with load Higher compressive strength Low High Low Very good

good abrasion resistance, and ability to operate without lubricants qualify nylons for use in many bearing applications, business machines, and appliances. For extra protection, occasional lubrication can be applied. Extruded nylon tubing and hoses are used in hydraulic and other fluid systems, because of their resistance to different fluids. The applications for castings are mostly in industrial equipment: large rollers, bearings, gears, cams, sheaves, guide blocks, wear plates, and the like. Nylon powder, which can be applied either electrostatically or by a fluid bed, provides tough, wear-resistant coatings. PARYLENES

The melting point of these film and coating resins ranges from 290° to 400°C (554 to 752°F), and their glass-transition temperatures range from 60 to 100°C (140 to 212°F).

Parylenes' cryogenic properties are excellent. Their physical properties are unaffected by thermal cycles from 2°K to room temperature. Their thermal endurance in air is as follows: the short-term (1,000 hr.) exposure is 93 to 129°C (200 to 265°F), the long term (ten years) 60 to 100°C (140 to 212°F). In inert atmospheres or in the absence of air, their properties are maintained up to 216 to 279°C (420 to 535°F). These thermoplastics are generally insoluble up to 150°C (302~. At 270°C (518°F) they will dissolve in chlorinated biphenyls, but the solution gels upon cooling below 160°C (320°F). Their weather resistance is poor. Embrittlement is the primary consequence of their exposure to UV radiation.

426 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The first significant commercial application of parylenes was as a dielectric film in high-performance precision electrical capacitors, followed with use in circuitboards and electronic module coatings. These coatings are to protect units from airborne contaminants, moisture, salt spray, and corrosive vapors while maintaining excellent insulator protection. The coatings are also extensively used in the protection of hybrid circuits. Such coatings do not affect parts' dimensions, shapes, or magnetic properties. Free-standing films can be produced of parylene. These ultrathin (250A-3 microns) films, called pellicles, are used as beam splitters in optical instruments, windows for nuclear radiation measuring devices, dielectric supports for planar capacitors, and for extremely fast-responding, low-mass thermistors and thermocouples. Applying parylene requires special, though not complex or bulky, equipment: essentially a vaporizer, a pyrolysis unit, and a deposition chamber. The objects to be coated are placed in the deposition chamber, where the vapor coats them with a polymer. A condensation coating like this does not run off or sag as in conventional coating methods. Neither is it "line of sight," as in vacuum metalizing. In condensation coating the vapor evenly coats edges, points, and internal areas. Although the vapor is all-pervasive, holes can still be coated without bridging. Masking can easily prevent certain areas from being coated, as desired. The objects to be coated can also remain at or near room temperature, thus preventing possible thermal damage. Because of the quantitative nature of this reaction, a coating's thickness can be accurately and simply controlled by manipulating the polymer composition's charge to the vaporizer.

PHENOLICS Since 1909, phenol-formaldehyde has continued to be a low-cost general-purpose thermoset compound to meet a multitude of applications. There are a wide range of fillers for it, each intended to fulfill the needs of particular end-product service environments. Typical among them is wood flour, for general use in electrical wall plates, industrial switchgears, circuit breakers, handles and knobs for small appliances, and such purposes. For slightly better impact strength, cotton flock is used. Mica, glass, and other minerals provide better electrical properties, heat resistance and dimensional stability, as in automotive power brakes, industrial electrical terminal strips, and so on. Compounds are formulated with a one- or two-stage phenolic curing system. In general, one-stage resins are slightly more critical to process. Although phenolics have properties that are somewhat inferior to those of the more expensive TSs, they are usually more easily molded. They can be processed by compression, transfer, and injection molding and to a limited degree by extrusion. The colors of these compounds are limited to black or brown. As is typical of many TSs, they are postcured to obtain maximum performance. Phenolic molding compounds are generally characterized as being low in cost and having superior heat resistance, a high heat-deflection temperature, good electrical properties and flame resistance, excellent moldability and dimensional stability, and good water and chemical resistance. Specialty compounds can provide high-performance heat resistance, impact strength, electrical properties, and creep resistance. The heat-resistant types are used in motor housings for appliances, handles for pots and pans, and other such products (see Fig. 6-4). In general, they can withstand short-term exposure to 204°C (400°F), and some grades can even be subjected to 260°C (500°F). Their long-term exposure ratings range from 149 to 171°C (300 to 340°F). Glass-fiber-reinforced impact grades are used in welding-rod holders, thermostat housings, commutators, and similar parts (see the section below on urea).

THE PROPERTIES OF PLASTICS 427

Figure 6-4. This commercial automotive engine whose major components are made from phenolic composite is a joint R&D venture between Polimotor Research, Inc., and the Rogers Corp., which supplies the phenolic compound. Polimotor has proven the concept of using composites in more than seventy-five racing engines. Moldable phenolic composites enable an engine designer to address the vital concerns of weight, efficiency, noise, and cost. Cast iron offers respectable properties but has a severe weight disadvantage. Aluminum provides that advantage but unfortunately has such high thermal conductivity that efficiency can be reduced by 20 percent. Phenolic composites provide optimal results to meet performance requirements, can be molded into intricate shapes, and are lower in cost to manufacture. For instance, a 2.3-liter four-cylinder, in-line double-overhead-cam engine can produce 175 hp yet weigh only 80 kg (175 lb.). Its 12.7-kg (28-lb.) composite engine block can be molded in just twelve minutes, which may be reducible to four minutes under production conditions.

PHENOXY RESINS Phenoxy resins are an outgrowth of epoxy resin technology. They have moderately good impact resistance, relatively high strength, and good elongation and creep resistance. They are thermoplastics, but they can also be thermosets, through cross-linking chemical reactions. As with most of the aromatic polymers, phenoxy resins' color retention, UV resistance, and weatherability are generally poor. They have limited thermal exposure with a recommended operating temperature range from - 60 to BO°C (-76 to 176°F). Phenoxy resins are resistant to acids and alkalis, have poor solvent resistance (especially in ketones)

428 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

but good resistance to aliphatic hydrocarbons, and resist staining from common household agents rather well. Their permeability, particularly to oxygen and other gases, is the lowest of any melt-processable plastic. Their oxygen permeability is one-twentieth that of HDPE and their moisture-vapor transmission is about the same as that of rigid PVChigher than PE but much lower than the styrenes. The major use of phenoxies is as a vehicle for coating formulations. They are exceptionally useful in primer applications where drying speed, compatibility with various kinds of top coats, and high adhesive strength are required. Phenoxies are used in automotive and marine primers as well as in heavy-duty maintenance primers. Phenoxy molding compounds are limited to applications requiring service temperatures not in excess of about 80°C (176°P). Its combination of good impact strength, clarity, and impermeability makes this resin attractive for molding, including blow molding, of cosmetic, foodstuff, and household-chemical bottles.

POL YALLOMERS Special polymerization can produce ethylene-propylene block copolymers that exhibit crystallinity similar to that seen in the homopolymers. This thermoplastic, which is insoluble in hexane and heptane, has many of the properties of HDPE and PP, along with an annealed density of 0.91, a brittleness temperature as low as -40°C (-400 P), and an Izod impact strength (notched) of 640 Jim (12 ft.-Ib.lin.) with less notch sensitivity. Polyallomers mold as easily as PPs, yet they have better dynamic flexural fatigue resistance, as in living hinges, particularly at low temperatures. Their flow and moldability is better than that of linear PE and has a superior softening point as well as greater hardness, stress-crack resistance, and mold shrinkage. Polyallomers are superior to rubbermodified PPs in color, clarity, moldability, electrical properties, and blush resistance when bent or stretched. This thermoplastic is used for blow-molding bottles, wire coating, extruding film, sheet, and pipe, and thermoforming various packaging applications.

POLYAMIDES See the section above on nylon.

POL YAMI DE-IMI DE Polyamide-imides (PAIs) are engineering thermoplastics characterized by excellent dimensional stability, high strength at high temperatures, and good impact resistance. Molded parts in this material can maintain their structural integrity in continuous use at temperatures of 260°C (5000 P). Different grades are available, such as general purpose, injection moldable, PTFE/graphite wear-resistant compounds, 30 percent graphite-fiberreinforced compounds, 30 percent glass-fiber-reinforced compounds, and so on. The room-temperature tensile strength of an unfilled PAl is about 192 MPa (28,000 psi), its compressive strength about 220 MPa (32,000 psi). At 232°C (4500 P) its tensile strength is about 65 MPa (9,500 psi), or as strong as many engineered plastics at room temperature. Continued exposure at 260°C (500°F) for up to 8,000 hours produces no significant decline in its tensile properties. PAl's flexural modulus of 5,000 MPa (730,000 psi) in an unfilled grade can be increased with graphite fiber reinforcement, to 2.9 x 106 psi. The degree of retention of its modulus

THE PROPERTIES OF PLASTICS 429

at temperatures to 260°C (500°F) is on the order of 80 percent. Its creep resistance, even at high temperatures and under load, is among the best of the thermoplastics, and its dimensional stability is extremely good. The unfilled grade of PAl is rated UL 94V-O at thicknesses as low as 0.008 in. and has an oxygen index of 45 percent. PAIs are extremely resistant to flame and have quite low smoke generation. Some reinforced grades have surpassed the FAA requirements for flammability, smoke density, and toxic gas emission. PAIs' radiation resistance is good, with a tensile strength that drops only about 5 percent after exposure, to 109 rads of gamma radiation. Its chemical resistance is very good, virtually unaffected by the aliphatic and aromatic hydrocarbons as well as halogenated solvents and most acid and base solutions. PAl is attacked, however, by some acids at high temperatures, by steam at high pressure and temperatures, and by strong bases. PAl moldings absorb moisture in humid environments or when immersed in water, but the rate is low and the process reversible. For example, at 50 percent relative humidity and 23°C (73°F) PAIs absorb about 1 percent by weight in 1,000 hours. Parts can be restored to their original dimensions by drying. One important area for the use of PAIs is in structural parts requiring high strength at high temperatures: aerospace products, business equipment, industrial chemical plants, heavy-duty trucks, underground environments, and other such markets. Some of the specific parts for which it is used include electrical connectors, switches, relays, gears, ball bearings, marine winches (see Fig. 6-5), high-load thrust bearings, and so on. Automotive parts using PAl include power and valve trains, piston skirts, tappets, piston rings, valve stems, and timing gears.

Figure 6-5. A high-performance, lightweight racing sailboat winch using a Torlon polyamideimide resin from Amoco Performance Products for both its roller and ball bearings.

430 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

POL YARYLATES

The polyarylate (PAR) resins are a family of TP polyesters with properties that put them within the spectrum of the relatively high-performance polymers. Their cost-performance rating is somewhat between PCs and the high-heat materials. In their mechanical properties PARs resemble the less expensive PCs, but they offer 10 to 23°C (50 to 70°F) higher heat resistance, somewhat better chemical resistance, and nearly comparable optical properties. They also have good toughness and strength retention in long outdoor exposure, good electrical properties, and an inherent flame resistance. PARs can be processed to enforce their various properties (see Chapter 7). Like the more familiar TP polyesters PET and PBT, the PARs are polycondensation products. However, both components are aromatic (ring type) molecules. By using various ratios, different properties become available. Even without glass-tiber-reinforcement, PAR's heat-deflection temperature at 1,820 kPa (264 psi) is about 171°C (340°F). They also have good impact strength over a wide temperature range (down to - 40°C [ - 40°FD, low notch sensitivity, and excellent flammability performance. Most PARs have a UL v-a flame-spread rating of 0.062 or 0.125 in. thickness without the use of flame-retardant additives, have oxygen indexes of about 30 or higher, and generate little smoke. Other useful properties of the PARs include good electrical characteristics as well as high UV and weather resistance without the use of stabilizers. They have at least an 85 percent light-transmission value and a rating as low as 2 percent for surface haze, and remain unaffected by long outdoor exposure. PARs do not have the water-clear transparency of the PCs. Their tint varies from a light straw color to a medium tan, depending on the manufacturing process. Some can be pigmented, and all grades can be supplied opaque. Automotive lighting has recently become a prime market as designers seek unreinforced transparent plastics that can handle the heat load from the higher-intensity bulbs operating in smaller spaces, and other markets also exist. POL YARYLETH ERS

These thermoplastic polymers are generally characterized by their toughness in respect to resistance to heat and UV exposure. Of importance also are their transparency, warp resistance, excellent flexural recovery, high elastic limits, and good electrical and mechanical properties, including outstanding creep resistance. They have as well excellent radiation and oxidation resistance. Their thermal expansion rate is close to that of metal. This amorphous material is subject to environmental stress cracking, however, particularly in the presence of aromatic or aliphatic hydrocarbons. Polyarylether's key properties include a heat-distortion temperature of up to 180°C (355°F), tensile strengths from 69 to 165 MPa (10,000 to 24,000 psi), and flexural modulus ratings from 2,067 to 9,646 MPa (300,000 to 1,400,000 psi). POL YARYLETHERKETONE

Compared with the temperature-stable thermoset plastics and fluoropolymers, clear economic and processing advantages have appeared, particularly for new materials that can be highly stressed thermally and mechanically. Polyaryletherketone (PAEK) is a leading material among the high-temperature stable thermoplastics. This family of plastics allows continuous operating temperature of 250°C (480°F) and-4lepending on the type of shortterm peak load-up to 350°C (660°F). The glass transition and melting temperatures are

THE PROPERTIES OF PLASTICS 431

thermodynamic quantities that depend on the ratio of the ketone to the ether groups. Various complicated configurations can be obtained, such as polyetherketoneetherketoneketone (PEKEKK). The properties of this family of plastics include a tensile strength at break of 85 MPa (12,300 psi), an elongation at break of 56 percent, a tensile modulus of elasticity of 0.6 X 106 psi, a tensile stress at yield at 23°C (74°F) of 104 MPa (15,100 psi) and at 160°C (320°F) of 37 MPa (5,400 psi), an elongation at yield of 6 percent at 23°C (74°F) and of 2 percent at 160°C (320°F), and no break using an unnotched Izod impact test. The processing flow behavior of PAEK does not differ fundamentally from that of other partially crystalline TPs. The shear rate is similar to Nylon 6 and PBT at 25°C (77°F) above the melting point. Besides having good mechanical and rheological properties (see Chapter 2), PAEK is characterized by its favorable behavior in fire. Without additives it has a UL 94V-O rating down to a test-specimen thickness of 0.030 in. The density of PAEK fumes in a fire is the lowest of the TPs, and it has exceptionally low corrosive and toxic fumes. The quantity of heat released upon the outbreak of a fire is quite low, and it meets aviation regulations for interior use. PAEK has high hydrolysis resistance and good resistance to many different chemicals.

POLYARYLSULFONE Polyarylsulfone (PAS) thermoplastics have relatively low melt-processing temperatures, considering their high mechanical properties. For one thing, they are hydrolytically stable. They have a heat-distortion temperature at least 204°C (400°F) at 1,820 kPa (264 psi), tensile strengths to 124 MPa (18,000 psi) or better, and a flexural modulus up to 1.17 X 106 psi. Under stress they are highly resistant to mineral acids and alkali and salt solutions. Their applications are in electrical and electronic items, frozen-food packaging, and aircraft interiors.

POL YBENZIMIDAZOLE PBI has no known melting point and a glass-transition temperature of 427°C (800°F). It has an ultrahigh heat-distortion temperature of 435°C (815°F), retards flame, and will not bum in air. The material can withstand steady temperatures up to 427°C (800°F) and short bursts up to 760°C (I ,400°F). This material is reported to resist steam at 343°C (650°F) and 15 MPa (2,200 psi) pressure. When exposed to saturated steam, PBI absorbs only 0.4 percent moisture. It resists a wide range of chemicals, including harsh ones. This plastic has high mechanical and physical strength properties including a high compression strength. PBI can bear loads for short periods at temperatures up to 650°C (I,200°F). When reinforced with silicon-carbide fibers, it can thwart an attack even from laser weapons. It has a low coefficient both of friction and of thermal expansion (0.000013 in'/in./°F). This wholly aromatic heterocyclic polymer is fabricated by sintering under high pressure. The low-molecular-weight PBI flows better than its high-molecular-weight counterpart. However, HMW PBI outgases less during processing, making it more suitable to mold large parts. PBI has been targeted to replace metals, ceramics, carbon, and other materials where an industry needs materials that are highly resistant to heat and corrosion, as in the chemicals and oil processing, aerospace, and transportation. PBI's main commercial applications include use as valve seats, seals, electrical connections, thrust washers,

432 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

bearings, and other mechanical components. PBI, the clothing pick of the best-dressed astronauts and firefighters, may become the material of choice for parts that need to resist temperatures up to 400°C (750°F) or more.

POLYBUTYLENES The polybutylenes (BPs) are crystalline thermoplastics in the polyolefin family. Compared to other polyolefins, they have superior resistance to creep and stress cracking. PB films have high tear resistance, toughness, and flexibility. Their chemical and electrical properties are similar to those of the PEs, and PPs, but their degree of crystallinity is much lower. This structure results in a rubberlike, elastomeric material with low molded-in stress. The main applications of PBs are in pipe, packaging, hot-melt adhesives, and sealants. Piping for cold-water use out of PBs has a higher burst strength than pipe made from any other polyolefin. Large-diameter pipe has been successfully used in mining and powergeneration systems to convey abrasive materials. PBs can be alloyed with other polyolefins to provide its inherent advantages. Film made into industrial trash bags gives improved resistance to bursting, puncturing, and tearing.

POL YBUTYLENE TEREPHTHALATE See "Thermoplastics" under "Polyesters," below.

POL YCARBONATES Among the engineering thermoplastics, the amorphous PCs are one of the most exceptional materials, distinguished by their highly versatile combination of properties. Most PCs can exist individually in various other materials, but generally not in their entirety. Different grades of pes provide specific properties and processing characteristics. These include the flame retardant and reinforced grades, resistance to weather and UV radiation; EMI and RFI shielding, FDA-approved grades for use in food-contact and medical applications, and grades for different processes such as injection, structural foam and blow molding, the extrusion of film and sheet, and others. PCs are characterized by toughness, heat and flame resistance, and dimensional stability. Thick unreinforced PC resists breakage at temperatures down to - 54°C ( - 65°F). Grades are available to provide high impact strength, based on different thicknesses at room temperature and a notched Izod impact strength of 6.4 to 8.5 J/cm (12 to 16 ft.lb.lin. Even in thick sections, a properly designed PC part has more impact strength at -54°C (-65°F) than most plastics generally do at RT. Many plastics are not tough at 18°C (65°F), but there are plastics that are tough even at much lower temperatures (see Chapter 4). Creep resistance, which is already excellent throughout a broad temperature range, can be further improved by a factor of two to three when PC is reinforced with glass fibers. Polycarbonates' insulating and other electrical properties are excellent and remain almost unchanged by temperature and humidity conditions. One exception is arc resistance, which in PCs is lower than in many other plastics. They are generally unaffected by greases, oils, and acids. Water at RT has no effect on PCs, but continuous exposure in 65°C (150°F) water causes gradual embrittlement. They are soluble in chlorinated hydrocarbons and attacked by most aromatic solvents, esters, and ketones, which cause

THE PROPERTIES OF PLASTICS 433

crazing and cracking in stressed parts. Grades with improved chemical resistance are available, and special coating systems can be applied to provide additional chemical protection. PCs are major additions to the group of polymer blends. Their mechanical and thermal properties make them the natural mainstays of the blends. For instance, some PC grades have notched Izod impacts of up to 960 Jim (18 ft.-Ib'/in.), heat-deflection temperatures of 138°C (280°F), and flexural moduli in the 2,067 MPa (300,000 psi) range. Two families of blends dominate: the thermoplastic polyesters PBT and PET (see Fig. 6-6) and ABS. Many other blends exist .to provide specific desirable characteristics. Blending can eliminate one or more shortcomings by the type of blend and the mixture ratio. Trade-offs exist, as might be expected, but with certain blends there are overall net gains and even synergistic gains (see Chapter 2). The gains include providing for an easier melt flow during processing. In the blending process, the natural water-clearness and transparency of PCs can be reduced or lost. The applications of PCs are extensive, emanating into all types of markets. A sarnple would include electronic connectors, switches, terminal blocks, computer disc packs, storage modules and housings, appliance power-tool housings, vacuum cleaner impellers, fan and air-conditioner grills, automotive instrument panels, indoor and outdoor lighting diffusers, medical kidney dialysers and blood oxygenators, and a host of others.

POLYESTERS There are both thermoplastic and thermoset polyesters, with many of each type each having their own special qualities.

Thermoplastics The two major TPs are polybutylene terephthalates (PBT) and polyethlene terephthalates (PET). These crystalline, high-molecular-weight polymers h~ve an excellent balance of properties and processing characteristics. Unreinforced and glass-fiber-reinforced grades

4.0

3.86

Unrelnlorced

20% Glass filled

Figure 6-6. A PC-PET blend that provides an easier processing flow than does a straight PC.

434 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

are available with UL flammability ratings of 94HB and 5V. They have excellent resistance to a broad range of chemicals at room temperature, including aliphatic hydrocarbons, gasoline, carbon tetrachloride, oils, fats, alcohols, glycols, and esters as well as diluted acids and bases (they are attacked by strong acids and bases). Some examples of applications for PBTs include automotive distributor caps and painted exterior body components, electronic terminal blocks, coil forms, integrated circuit carriers, and other electromechanical parts to replace thermosets. The high-stiffness resins, including the foam injection-molded grades, are well suited for such structural applications as automobile parts. PETs are extensively used in such products as bottles and films. Stretched-injection blow-molded bottles are used for nearly all two-liter carbonated beverage containers. PET is also used for packaging foods, cosmetics, and household chemicals (see Figs. 6-7 and 6-8). With changes in formulation, surface treatment, and processing-from unoriented to bioriented-PET films can produce variations in optical, mechanical, physical, and surface properties. They are used in liquid-crystal displays, metalized photocopier belts, motor and wire insulation, for holographic reproduction, in magnetic tapes and discs, and other uses. Aluminized PET film with nylon fabric set the world's record for altitude in a manned hot-air balloon at 59,500 m (65,000 ft.), in 1988.

Figure 6-7. Different-shaped PET stretched-injection, blow-molded bottles.

THE PROPERTIES OF PLASTICS 435

Figure 6-8. A l.7S-liter (l.8S-quart) whiskey bottle made of a Goodyear PET resin reduces filling-line noise and breakage and is lightweight and shatterproof during shipping and handling.

Thermosets Thermoset polyesters, as the thermoset alkyds, are compounds that are usually based on unsaturated polyester resin systems. The term alkyd refers to thermosets using lower amounts of monomers. When the monomer level is higher, the compound is called a polyester. At room temperature these solutions (the mixtures of the resin and the liquid monomer, which is usually a styrene) are stable. Any of a variety of peroxide catalysts can initiate cross-linking (see curing in Chapter 2) at room temperature or higher. Unlike most other plastics, which are basically constructed around a single ingredient, polyester formulations usually contain substantial amounts of several materials like fillers, reinforcements, additives, and other resins. Special advantages and limitations apply to each compound and their various processing methods. One example of such a trade-off and combination is shown in Table 6-5 and Figure 6-9. A major characteristic of TS polyester and the reason it was developed, in 1942-1944, is that it does not produce byproducts such as moisture or gases during fabrication. Most TSs have by-products that require pressure and bumping to be used during the chemical curing reaction phase. Thus, TS polyesters can be cured at no pressure. However, pressure is applied when compounds have extensive fiber reinforcement and complicated shapes, to ensure the release of trapped air, provide an improved surface, and obtain maximum performance. Fabricating thermoset parts, usually with glass fiber reinforcements, is more varied than with other types of plastic. There are many different methods of processing reinforced TS polyesters, from no pressure to high pressure, employing hand lay-up and spray-up for small to moderate quantities of large parts, using compression molding from bulkmolding compounds, sheet-molding compounds, or glass-fiber preforms for the highvolume production of moderate-sized complex parts, and using pultrusion for basically

436 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 6-5. The Range of Properties Available Using Different Reinforced Thermoset Polyester Resins Thermoset Polyester Compounds

Characteristic

General Purpose

Rigid moldings

Flexible resins and semirigid resins

Tough, good impact resistance, high flexural strength, low flexural modulus

Light, stable, and weather resistant Chemical resistant

Resistant to weather and ultraviolet degradation Highest chemical resistance of polyester group, excellent acid resistance, fair in alkalis Self-extinguishing, rigid

Flame resistant

High heat distortion Hot strength Low exotherm

Extended pot life Air dry Thixotropic

Service up to 500°F (260°C), rigid Fast rate of cure (hot), moldings easily removed from die Void-free thick laminates, low heat generated during cure Void-free and uniform, long flow time in mold before gel Cures tack-free at room temperature Resists flow or drainage when applied to vertical surfaces

Typical Uses Trays, boats, tanks, boxes, luggage, seating Vibration damping: machine covers and guards, safety helmets, electronic part encapsulation, gel coats, patching compounds, auto bodies, boats Structural panels, skylighting, glazing Corrosion resistant applications such as pipe, tanks, ducts, fume stacks Building panels (interior), electrical components, fuel tanks Aircraft parts Containers, trays, housings Encapsulating electronic components, electrical premix parts-switchgear Large complex moldings Pools, boats, tanks Boats, pools, tank linings

constant-section shapes, as well as injection molding (see Fig. 6-10), resin transfer molding, casting, and other techniques (see Chapter 7). The typical strength ranges obtainable in parts fabricated from various forms of compounds and processed by different methods are given in the last section of this chapter. Applications for materials such as these are in boat hulls, architectural panels and structures, vehicle components, swimming pools and filter parts, athletic equipment, housings for products like computers, office-, business-, medical-, testing-, and display equipment, and storage tanks. Further uses include agricultural fertilizer and feed hoppers, pig and other animal stalls, furniture, bathroom components like shower stalls and modular tub-wall segments, aircraft primary and secondary structures, toys, and many products that must exist in all types of corrosive environments. Pultruded parts include electrical insulators, boom arms for aerial-lift trucks, luggage racks, channels, beams, solid rods, and L beams. POL YETHERKETONE

Polyetherketone (PEK) is a partially crystalline high-performance aromatic ketone-based thermoplastic that is heat stable and readily processed. As a member of the ketone family it shares with PEEK such properties as good chemical resistance; exceptional toughness,

Figure 6-9. A Xycon TS polyester-polyurethane hybrid resin from Amoco Chemical Co. was used for the body panels in this futuristic GM XT2 EL Camino. When compared to unsaturated TS polyesters, this Xycon resin with Amoco's Thomel carbon filter offers greater inherent strength, is less brittle, and requires a lower curing temperature and no postcuring.

Figure 6-10. An experimental Chrysler Cordoba front panel injection molded from PPG Industries' fiberglass-reinforced TS polyester compound. This composite structure provides a 30 to 50 percent weight savings over the comparable metal part, in addition to more corrosion resistance. 437

438 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 6-11. This popular practically all-glass-fiber-TS polyester RP leisure boat was injection molded and had hand lay-up of its parts. This battery-powered Sun Cat is easy maneuverable by plastic propellers located within recesses on each RP floating pontoons. strength, rigidity, and load-bearing capabilities; good radiation resistance; the best firesafety characteristics of any thermoplastic, and the ability to be easily melt processed. Super PEKs designed for advanced composites have a continuous-service temperature rating of 260°C (5000 P), a glass-transition temperature (Tg) of 200°C (400 0 P), and a slow crystallization rate that suits them for processes with slow rates of cooling from the melt.

POL YETHERETHERKETONE Polyetheretherketone (PEEK) is a high-temperature crystalline thermoplastic resin suitable for high-performance unreinforced or reinforced parts. This member of the ketone family offers an excellent combination of thermal and combustion characteristics. Its wholly aromatic structure contributes to its high temperature performance, and its crystalline character makes it resistant to organic solvents and dynamic fatigue, and helps it retain ductility in short-term heat aging. PEEK molded compounds absorb much less water than any other thermoplastic. They have good resistance to aqueous reagents, with long-term proven performance at 220°C (430 0 P). Their resistance to attack is over a wide pH range from 60 percent sulfuric acid to 40 percent sodium hydroxide at elevated temperatures. However, PEEK is attacked by some concentrated acids. No solvent attack has yet been observed on molded parts, although some solvents cause the crazing of highly stressed coated wire. This problem can be eliminated by orienting PEEK below its melting point. PEEK compounds are molded or extruded in conventional TP equipment at 350 to 400°C (660 to 7500 P). Their mold shrinkage is about 1 percent for the unreinforced and 0.1 to 1.4 percent for the reinforced grades, depending on their fiber orientation. PEEK is widely used in carbon-fiber composites because of their excellent adhesion to carbon fibers. PEEK's heat-distortion temperature is above 315°C (6000 P) at 1,820 kPa (264 psi), its tensile strengths are up to 215 MPa (31,000 psi), and its flexural modulus is up to 21 X 106 psi. At room temperature PEEK is tough, strong, and rigid, with excellent load-bearing properties over long periods and outstanding resistance to abrasion. On a short-term basis this material is suitable for service temperatures in excess of 300°C (575°P). It also has excellent thermal stability for continuous operation. A lifetime of 50,000 hours can be

THE PROPERTIES OF PLASTICS 439

expected at 260°C (5000 P). With its lack of flame-retardant additives or halogens, PEEK has a limiting oxygen index of 35 percent, meets UL 94V-O requirements, and has an extremely low rate of smoke emission. Tests to date show that PEEK has good resistance to radiation. PEEK moldings do degrade by UV during outdoor weathering, however. When it is natural or pigmented, the effect on PEEK is minimal over a twelve-month period. Paint or other applicable coatings are recommended for use in prolonged or extreme weather conditions. PEEK film can be laminated to itself or to other substrates. Its bond strength depends on the surface preparation and the type of adhesive used. Pilm is available in both a transparent, thermoformable grade and a higher heat-temperature, heat-stabilized version that is more crystalline and less transparent as well as thermoformable. Applications for PEEK range from commercial or industrial to uses in nuclear plants, underground or underwater applications, oil wells, commercial aircraft to military equipment, and for underground or surface-railway equipment. Some typical PEEK products include pump impellers, electrical connectors, valve seals, wires and cable, fire-safety components, and others.

POL YETHERIMIDE Polyetherimide (PEl) is an amorphous engineering TP characterized by high heat resistance, high strength and a high modulus, excellent electrical properties that remain stable over a wide range of temperatures and frequencies, and very good processability. Neat (unmodified) PEl is transparent, with inherent flame resistance and low smoke evolution. Its UL continuous-use listing is 170°C (338°P) and its heat-deflection temperature is 200°C (392°P) at 1,820 kPa (264 psi). It has a Tg of 215°C (419°P), a UL rating of 94V-O at a thickness of 0.016 in. and of 5V at 0.075 in. without additives. PEl has a limited oxygen index of 47 percent (one of the highest among the engineering TPs), its smoke evolution as measured in an NBS chamber test per ASTM E 662 is low, and its dielectric constant remains virtually unchanged between frequencies of 60 to 109 Hz and temperatures of 28 to 82°C (73 to 1800 P). It has a high-volume resistivity of 6.7 x 10 17 ohmcm and a dielectric strength from 830 V/mil at 1116 in. in air to greater than 6,500 VI mil in submil film thicknesses. Its arc resistance exceeds 120 s, meeting one of the UL electrical requirements for the sole support of "live" parts. A key feature of PEl is its ability to maintain properties at elevated temperatures. Por example, at 180°C (356°P) its tensile strength and flexural modulus are 41 and 2,067 MPa (6,000 and 300,000 psi), respectively. The strength and modulus of the glassreinforced grades are higher. The modulus at 1.3 x 106 psi with 30 percent weight of glass to over 80 percent is retained at 180°C (356°P). PEl also has good creep resistance, as indicated by its apparent modulus of 24,115 MPa (350,000 psi) after 1,000 hours at 82°C (1800 P) under an initial load of 35 MPa (5,000 psi). This resin resists a broad range of chemicals under varied conditions of stress and temperature. PEl's resistance to mineral acids, for example, is outstanding. It is, however, attacked by such partially halogenated solvents as methylene chloride, trichloroethane, and strong acids. Its resistance to UV radiation is good with a change in tensile strength after 1,000 hours of xenon-arc exposure that is negligible. PEl's resistance to gamma-ray radiation is also good, there being a strength loss of less than 6 percent after 500 megarads' exposure to cobalt 60 at the rate of one Mrad/hr. Hydrolytic stability tests show that more than 85 percent of PEl's tensile strength is retained after 10,000 hours of immersion in boiling water. This material is suitable for short-term or repeated steam exposure.

440 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

POL YETHERSULFONE This material is a high-temperature engineering TP in the polysulfone family. PES is recommended for load-bearing applications up to 182°C (3600 P). Even without flame retardants it offers low flammability, and it has little change in its dimensions of electrical properties in the temperature range from 0 to 200°C (32 to 3900 P). At room temperature it behaves like a traditional engineering TP, being rough, rigid, and strong, with outstanding long-term load-bearing properties. Of far greater importance are the high-temperature properties that PES has. PES can be used for tens of thousands of hours at 200°C (3900 P) without a significant loss of strength. The UL temperature index is 200°C (3900 P) for a PES compound that is 30 percent glass fiber. Its dimensional changes at 200°C (3900 P) are negligible, although small variations occur in its electrical performance from 0 to 200°C (32 to 3900 P). Compared with most TPs, PES has low flammability without flame-retardant additives and carries a UL 94V-O rating at a thickness of 0.017 in. with glass reinforcement. Its limiting oxygen index is in the range of 34 to 41. Its smoke and gas emission are very low. PES possesses good resistance to X rays, beta rays, and gamma rays. Even though neat it does not have good resistance to outdoor weathering, with a carbon-black filler it is acceptable. Chemical resistance in PES is dependent on the external and molded-in stress levels and the temperature. Its chemical resistance can be improved by annealing it at 200°C (390°F). It has good resistance to aqueous acids, bases, and most inorganic solutions. Most of the common sterilizing solutions and anesthetics can be used with PES safely (many cleaning and degreasing solvents are based on chlorinated and fluorinated hydrocarbons). Unless PES is heavily stressed, it can be cleaned by most of these solvents.

POLYETHYLENE Polyethylenes (PEs) are in the polyolefin family of semicrystalline TPs. The largestvolume plastics used worldwide, PEs are available in many varieties with an equally wide range of properties. Some are flexible, others rigid; some have low impact strength, whereas others are nearly unbreakable; some have good clarity, others are opaque, and so on. The service temperatures for PEs range from -40 to 93°C (-40 to 2000 P). In general they are characterized by toughness, excellent chemical resistance and electrical properties, a low coefficient of friction, near-zero moisture absorption, and good ease of processing. They are basically classified according to their density (see Tables 6-6 through 6-9 and Figs. 6-12 through 6-15). In addition to those PEs listed in the tables, there are others designed to meet different requirements, such as cross-linked PE (XLPE), which by chemical or irradiation treatment becomes essentially a TS with outstanding heat resistance and strength. There is an extra high-gloss HDPE (Fortiflex, by the Soltex Polymer Corp.), another that retains its toughness at very low temperatures and performs at levels between the commodity and engineering resins (Zemid, by Du Pont), and various others. Three basic characteristics of PEs determine their processing and end-use properties: their density, melt index, and molecular weight distribution (see Chapter 2). Their range in density, from 0.890 to above 0.96 g/cm3, is a result of their crystalline structure. This difference accounts for their property variations seen in Chapter 2. As one example, reducing PE's crystallinity increases its impact resistance, cold flow, tackiness, tear

THE PROPERTIES OF PLASTICS 441

4300 TENSILE STRENGTH

STRESS CRACK RESISTANCE'

1400

i

I 0.915

L-L-~~

0.960 DENSITY

o

5

10

15

20

MELT INDEX * Arbitrary values for 0.950 density.

150.000 STIFFNESS

IMPACT STRENGTH

19.000 0.915

DENSITY

0.960

o

5

10

15

20

MELT INDEX ~

_ _ _...,90

GLOSS

HEAT DISTORTION

10 0.915

DENSITY

0.960

Figure 6-12. An example of how density affects PEs.

o

5

10

15

20

MELT INDEX

Figure 6-13. An example of how the melt index affects PEs.

strength, environmental stress-crack resistance, and heat-seal range. As its crystallinity is reduced, decreases occur in stiffness, shrinkage, brittleness temperature, and chemical resistance. Low-Density Polyethylene

Low-density polyethylene (LDPE), the first of the PEs, has good toughness, flexibility, low temperature resistance, clarity in film, and relatively low heat resistance, as well as good resistance to chemical attack. At room temperature LDPE is insoluble in most organic solvents but attacked by strong oxidizing acids. At high temperatures it becomes increasingly susceptible to attack by aromatic, chlorinated, and aliphatic hydrocarbons. The LDPEs are susceptible to environmental and some chemical stress cracking. For example, wetting agents such as detergents accelerate stress cracking. Some copolymers of LDPE are available with an improved stress-cracking resistance.

442 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 100r---------------------------~~----------------~

-0.96

80

BRITTLE WAXES

?fI.

~. 60

z ..J ..J <(

~ 40

>-

Il: U

CONVENTIONAL TYPES

20

O~

LIQUIDS __________

100

~

__________

1,000

~

__________

10,000

~

__________

~

1,000,000

100,000

MOLECULAR WEIGHT

Figure 6-14. The relationships between crystallinity, molecular weight, and the properties of a PE, based on different polymerization reactions.

Linear Low-Density Polyethylene This material is used mainly in film applications but other products using all'types of processes as well. Its properties are different from LDPE and HDPE in that its impact, tear, heat-seal strength, and environmental stress-crack resistance are significantly higher. Its major uses at present are for grocery bags, industrial trash bags, liners, and heavyduty shipping bags.

Table 6-6. Polyethylene Densities (lbs.lft.3)

Density, g/cm

Type

0.910 0.926 0.941 0.960

LDPE MDPE HDPE HMWPE

to 0.925 to 0.940 to 0.959 and above

(56.8-57.7) (57.8-58.7) (58.7-59.9) (59.9)

Table 6-7. Polyethylene Properties Density, g/cmt (lbs.lft. 3) Crystal melt temperature,

LDPE 0.910 to 0.925 (56.8-57.7)

95 to 130 (203-266)

HDPE 0.941 to 0.965 (58.7-60.2)

120 to 140 (248-284)

°C (OF)

Tensile strength, MPa (psi) Tensile modulus, MPa (psi) Elongation to break, % Hardness, Shore D

4.1 96.5 90.0 41

to to to to

15.9 (590-2,300) 262 (14,000-38,000) 800.0 50

21.4 414 20 50

to to to to

37.9 (3,100-5,490) 1250 (60,000-181,200) 1300 70

Table 6-8. The Effects on PEs of Increases in Density, Melt Index, and Molecular Weight PE Property

Density

Melt Index

Tensile strength (at yield) Stiffness Impact strength Low-temperature brittleness Abrasion resistance Hardness Softening point Stress-crack resistance Permeability Chemical resistance Melt strength Gloss Haze Shrinkage

Increases Increases Decreases Increases Increases Increases Increases Decreases Decreases Increases

Decreases Decreases slightly Decreases Increases Decreases Decreases slightly

Increases Decreases Decreases

Molecular Weight

Decreases slightly Decreases Decreases

Increases Decreases Increases slightly Decreases Decreases Increases Decreases Decreases

Increases Decreases Increases

Table 6-9. How Intrinsic Properties Affect PEs Physical Properties

If Density Increases

If Melt Index Increases

IfMWD Broadens

Lower Lower

Melt viscosity Vicat softening point Surface hardness Tensile strength Yield Break Elongation Creep resistance Flexural stiffness Flexibility Toughness

Higher Much higher Higher

Lower Lower Slightly lower

Much higher Slightly higher Lower Higher Much higher Lower Lower

Slightly lower Lower Lower Slightly lower Slightly lower

Lower Lower Higher Higher

Lower

Lower

Low-temperature brittleness Stress crack resistance

Lower Lower

Lower Lower

Higher Higher

Optical properties Transparency Freedom from haze Gloss

Higher* Higher* Higher*

Higher Higher Higher

Barrier properties MVT rate Gas and liquid Transmission Greaseproofness Electrical properties

Lower Much lower Much higher

Slightly lower

Slightly higher

No effect

*Not true in the high-density range.

443

444 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 6,000,000 5,000,000 1900 Ultrahigh molecular weight polymer M.w. range

4,000,000 3,000,000

t

2,000,000

:a

~ "0

HOPE High M.w. range

-

1,000,000

E

~

500,000

~

400,000

.:.

~

I -

I I HOPE High M.W. LV. range

8,

e!

I}I

I

I

1

1900 Ultrahigh molecular weight polymer LV. range

I I I

I I

I II •Variations due to molecular

300,000

I

branching and configuration permit Normal HOPE LV. only approximations of M.W. range I This plot of M.w. versus I. V. is based on an accepted theoretical relationship of I MW = 5.3 X 104 (LV.) 1.37

!

I

I

I

1OO,OOOL-__~__L-~~~~~__~__~~~~ 5

10

15

20 25 30

LV. (I ntrinsic viscosity)

Figure 6-15. The relationship between intrinsic viscosity and molecular weight in an

UHMWPE.

High-Density Polyethylene The rigidity and tensile strength of HOPE are considerably higher than in LOPE and medium OPE (MOPE). HOPE's impact strength is slightly lower, as is to be expected in a stiffer material, but its overall values are high, especially at low temperatures, compared to the other TPs.

High-Molecular Weight HOPE This plastic offers outstanding toughness and durability, particularly at low temperatures. In blow-molding applications HMWHOPE allows drum manufacturers to meet OOT and OSHA specifications. In pipe production it meets the highest strength rating for PE pipe. Many other products are made of it, using rather conventional processing methods.

Ultrahigh Molecular Weight PE The properties of this high-performance plastic are entirely different and much improved, including outstanding abrasion resistance and a low coefficient of friction. The impact strength of UHMWPE is high and its chemical resistance excellent. As with most high-performance polymers, the processing of UHMWPE is not easy. Because of its high melt viscosity-it does not register on a melt-flow index-conventional

THE PROPERTIES OF PLASTICS 445

molding and extrusion would break the long molecular chains that give this plastic its excellent properties. The processing methods used for UHMWPE are basically compression molding, ram extrusion, and warm forming of extruded slugs. The techniques for using screw melting (that is, injection and extrusion processing) are under development to permit maintaining its UHMW structure.

POL YETHYLENE TEREPHTHALATE See the earlier section on polyesters.

POLYIMIDES Polyimides (PIs) were the first so-called high-heat-resistant plastics. They in fact retain a significant portion of their room-temperature mechanical properties from - 240 to 315°C (-400 to +600°F) in air. PIs, which are available in both TPs and TSs, are a family of some of the most heat- and fire-resistant polymers known. As discussed in connection with some of the other polymers, there are others that have heat resistance in the 260°C (500°F) range. Moldings and laminates are generally based on TSs, though some are made from TPs. PIs are available as laminates and in various shapes, as molded parts, stock shapes, and resins (in powders and solutions). PI parts are fabricated by techniques ranging from powder-metallurgy methods to conventional injection, transfer, and compression molding, and extrusion methods. Porous PI parts are also available. Generally, the compounds that are the most difficult to fabricate are also the ones that have the highest heat resistance. The service temperature for the intermittent exposure of PIs can range from cryogenic to as high as 480°C (900°F). Glass-fiber-reinforced PIs retain 70 percent of their flexural strength and modulus at 250°C (480°F). Creep in PIs is almost nonexistent, even at high temperatures. Their deformation under a 28 MPa (4,000 psi) load is less than 0.05 percent at room temperature for twenty-four hours. These materials have good wear resistance and a low coefficient of friction, both of which are factors that can be further improved by including additives like PTFE and MoS 2 . Self-lubricating PI parts containing graphite powders have flexural strengths above 69 MPa ( 10,000 psi). Their electrical properties are also outstanding over wide temperature and humidity ranges. They are unaffected by exposure to dilute acids, aromatic and aliphatic hydrocarbons, esters, alcohols, hydraulic fluids, JP-4 fuel, and kerosene. They are, however, attacked by dilute alkalis and concentrated inorganic acids. PI film has useful mechanical properties, even at cryogenic temperatures. At - 269°C ( - 453°F) this film can be bent around a t in. mandrel without breaking. At 500°C (932°F) its tensile strength is 30 MPa (4,500 psi). An important class of high temperature matrices is the imide compounds. The early compounds met needs for high temperatures but were brittle and difficult to process. Researchers have since overcome these difficulties.

POL YMETHYL METHACRYLATE See the discussion above on acrylics.

446 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

POL YMETHYlPENTENE The major advantages of PMP over other polyolefins are its transparency in thick sections, its short-time heat resistance up to 200°C (400°F), and its lower specific gravity. Unlike the other polyolefins, however, it is transparent, because its crystalline and amorphous phases have the same index of refraction. Almost clear optically, PMP has a light transmission value of 90 percent, which is just slightly less than that of the acrylics. It retains most of its physical properties under brief exposure to heat at 200°C (400°F), but it is not stable at temperatures for an extended time over 150°C (300°F) without an antioxidant. In a clear form it is not recommended where it will have to undergo long-term exposure to UV environments. The chemical resistance and electrical properties of PMP are similar to those of the other polyolefins, except that it retains these properties at higher temperatures than do either PE or PP. In this respect PMP tends to compare well with PTFE up to 150°C (300°F). Molded parts made of this resin are hard and shiny, yet their impact strength is high at temperatures down to - 29°C (- 20°F). Their specific gravity of 0.83 is the lowest of any commercial solid plastic.

POL YOlEFINS The family of polyolefins includes polyethylene, polypropylene, ethylene-vinyl acetate, ionomers, polybutylene, polyisobutylene, and polymethylpentene. Because the chemical and electrical properties of the various polyolefins are basically similar, they often compete for the same applications. However, since the different strength and modulus properties vary greatly with the type and degree of crystallinity, as seen in the preceding tables and figures, the tensile, flexural, and impact strength of each polyolefin may be quite different. Their stress-crack resistance and useful temperature ranges may also vary with their crystalline structure.

POL YPHENYlENE ETHER Alloys, or blends of polyphenylene ether and PSs in various proportions, are available under the tradenames Noryl and Prevex from GE (Noryl is a polyphenylene oxide). They can be processed by conventional equipment that produces either solid or foam products. PPE compounds are characterized by outstanding dimensional stability, low water absorption in the engineering TPs, broad temperature ranges, excellent mechanical and thermal properties from -46 to 121°C (-50 to 250°F), and excellent dielectric characteristics over a wide range of frequencies and temperatures. They are also reasonably easy to process. Several injection molded and extrusion grades are rated UL 94V-I or v-a, including the glass-reinforced compounds. Foamable grades have service-temperature ratings of up to 96°C (205°F) in i-in. sections. Because of their hydrolytic stability, both at room and elevated temperatures, blended parts in PPE can be repeatedly steam sterilized with no significant change in their properties. When exposed to aqueous environments their dimensional changes are low and predictable. PPE's resistance to acid, bases, and detergents is excellent. However, it is attacked by many halogenated or aromatic hydrocarbons.

THE PROPERTIES OF PLASTICS 447

POL YPHENYLENE SULFIDE This crystalline, high-performance engineering TP is characterized by outstanding hightemperature stability, inherent flame resistance, and a broad range of chemical resistance. PPS resins and compounds provide various combinations of high mechanical strength, impact resistance, and electrical insulation, with its high arc resistance and low arc tracking. The pigmented PPS compounds include several grades that are suitable to support current-carrying parts in electrical components. They are essentially transparent to microwave radiation. Unreinforced PPS resins are also available for use in slurry coating and electrostatic spraying. Resin coatings are suitable for food-contact applications as well as chemical processing equipment. PPS is also available with long-fiber glass, carbon, or other reinforced forms. The stampable sheet type contains fiber-mat reinforcement and can be processed by compression molding. Other forms can contain predesigned reinforcement patterns for different processes, such as laminating and thermoforming. These cross-linked types of resins are more crystalline than any of the sulfones, which are generally classified as being more amorphous. They are quite stiff, with a flexural modulus ranging from 1.7 X 106 to 2.5 X 106 psi. Their tensile strengths range from 69 to 172 MPa (10,000 to 25,000 psi). Their HDTs are up to 275°C (525°F) at 1,820 kPa (264 psi). PPS has excellent resistance to a broad range of chemicals, even at high temperatures. In fact, below 200°C (400°F) the resin has no known solvent. PPSs are flame retardant without additives, being rated at UL 94V-O-V5. The oxygen index of the neat resin is 44, with the indexes of the compounds ranging from 47 to 53. Because flame retardance is inherent in it, a regrind will be as flame resistant as a production in the virgin material. More recently developed linear PPSs have a far lower proportion of inorganic impurities than the conventional cross-linked material. They are also characterized by higher ultimate strength and an elongation at break of 4 as compared to I percent, as well as higher flexural strength and notched impact strength. However, the cross-linked product is somewhat more rigid. Linear PPS is partially crystalline, with pronounced thermaltransition ranges similar to those of PET that run from 85 to 100°C (185 to 212°F) for the Tg • Its melting point, Tm , is 280 to 285°C (535 to 545°F).

POL YPROPYLENE Polypropylenes (PPs) are in the polyolefin family of plastics. They are semitranslucent and milky white in color, with excellent colorability. They are produced by a stereoselective catalyst that puts order in their molecular configuration so that the basic resin has a predominantly regular, uniform structure. This means that the molecules crystallize into compact bundles, which makes them stronger than other members of the polyolefin family. PPs are an extremely versatile plastic available in many grades as well as copolymers like ethylene propylene. Neat PP has a low density of 0.90, which, combined with its good balance of moderate cost, strength, and stiffness as well as excellent fatigue, chemical resistance, and thermal and electrical properties, makes PP extremely attractive for many indoor and outdoor applications. The strength, rigidity, heat resistance, and dimensional-stability properties of PP can be increased significantly with glass-fiber reinforcement. Increased toughness is provided in special, high-molecular-weight rubber-modified grades (see Fig. 6-16). The electrical

448 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 6-16. Himont's easily molded HiGlass materials enhance the durability of automobile fender liners. This complex liner mode for GM was injection molded in a two-cavity mold using this glass-fiber-TP polypropylene resin composite that provides excellent structural integrity with good impact resistance to stones and similar objects.

properties of PP moldings are affected to varying degrees by their service temperatures. Its dielectric constant is essentially unchanged, but its dielectric strength increases and its volume resistivity decreases as temperature increases. PPs have limited heat resistance, but heat-stabilized grades are available for applications requiring prolonged use at elevated temperatures. The useful life for parts molded from such grades may be as long as five years at 120°C (250°F), ten years at 130°C (230°F), and twenty years at 99°C (210°F). Specially stabilized grades are UL rated at 120°C (248°F) for continuous service. Basically, PP is classed as a slow-burning material, but it can also be supplied in flame-retardant grades. PPs are unstable in the presence of oxidation conditions and UV radiation. Although all its grades are stabilized to some extent, specific stabilization systems are often used to suit a formulation to a particular environment, such as where it must undergo outdoor weathering. PPs resist chemical attack and staining and are unaffected by aqueous solutions of inorganic salts or mineral acids and bases, even at high temperatures. They are not attacked by most organic chemicals, and there is no solvent for this resin at room temperature. The resins are attacked, however, by halogens, fuming nitric acid and other active oxidizing agents, as well as by aromatic and chlorinated hydrocarbons at high temperatures. Although PPs retain their strength and stiffness at elevated temperatures, their performance at low temperatures leaves much to be desired. However, copolymers of PP that are available offer as much as two to three times the impact strength of general-purpose PP, even at temperatures as low as - 29°C ( - 20°F). PP is widely known for its application in the integral "living hinges" that are used in all types of applications (see Chapter 11). PP's excellent fatigue resistance is utilized in molding these integral hinges.

THE PROPERTIES OF PLASTICS 449

POL YSTYRENE

Polystyrene (PS) is noted for its sparkling clarity, hardness, extreme ease of processing (at least in the case of general purpose PS, or GPPS), excellent colorability, dimensional stability, and relatively low cost. This amorphous TP often competes favorably with higher-priced resins. It is available in a wide range of grades for all types of processes. Modifications available to the basic GPPS include grades for high heat and for various degrees of impact resistance. Clarity and gloss are reduced, however, in the impact grades. Some examples of members in the PS family are compounds of ABS, SAN, and SMA (styrene maleic anhydride). The structural characteristics of these copolymers are similar, but the SMA one has the highest heat resistance. Expandable polystyrene (EPS) is a specialized form of plastics, which usually includes a pentane blowing agent, used to make low densities of 0.75 to 10 lb.lft. 3 of foam shapes and blocks. Different-sized pellets provide changes in density. These materials are well suited for thermal insulation and energy absorption. Various different formulations exist such as ones with improved heat tolerance, solvent resistance, and cushioning abilities. Processing EPS involves two stages. Pirst the pellets are expanded to the density required in the finished part but remain as discrete pellets for a specified time, such as twenty-four hours. The next stage puts the pellets into a well-vented mold cavity in the shape of the final part. The venting consists of having many small holes through the mold cavity through which steam is forced. The steam heats the EPS by direct contact, causing the pellets to expand and fuse together to form void-free, dry foam parts. The pressures used are usually below 50 psi, allowing the use of relatively low-cost aluminum tooling [10]. Solid PS processed parts have low heat resistance, as compared to most TPs. Their maximum recommended continuous service temperature is below 93°C (200 0 P). Their electrical properties, which are good at room temperature, are affected only slightly by higher temperatures and varying humidity (see Pig. 6-17). PS is soluble in most aromatic and chlorinated solvents but insoluble in such alcohols as methanol, ethanol, normal heptane, and acetone. Most fluids in households, as well as drinks and foods, have no effect, but PS is attacked by the oil in citrus-fruit rinds, gasoline, turpentine, and lacquer thinner. PSs are available in FDA-approved grades. POL YSULFONES

Polysulfones (PSUs) are amorphous engineering TPs noted for their high heat-deflection temperatures, outstanding dimensional stability and electrical properties, excellent chemical resistance, and for being biologically inert, rigid, strong, and easily processed by different methods (see Pig. 6-18 and 6-19). It is stable and self-extinguishing in its completely natural, unmodified neat form; in most plastics these qualities must be obtained by using chemical modifiers. PSUs are also heat resistant and maintain their properties in a range from -100°C ( -1500 P) to over 150°C (3000 P). These strong, rigid plastics are the only type that will remain transparent at service temperatures as high as 200°C (4000 P). The name polysulfone has been assigned to polymers with S02 groups in their backbone. The basic types are the standard PSUs, polyaryl, polyether, and polyphenyl. PSUs are available in opaque colors and in mineral-filled and glass- and other reinforced compounds to provide higher strength, stiffness, and thermal stability. Por example, reinforced carbon-fiber PSU is used in human hip joints. The tensile strengths of PSUs go up to 110 MPa (16,000 psi), its flexural modulus to

450 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 6-17. Coaxial electrical cable's PS insulation "buttons" are here being injection molded, using a cold runner system, but it could also use a hot runner one. In this continuous production process, the injection-molding machine (IMM) is on a platform that moves in a rectangular pattern to permit the platens to open and move away from the buttons, as well as to move at the speed of the six-cable copper wire line when the mold is closed and the IMM is injecting the PS. Copper wires are started out with large diameters and are pulled through reduction squeeze rolls to their final thin diameter prior to entering the IMM. The wire-reduction line is to the left of the IMM, with the wire pullers to its left. Automatic devices remove the runn\!rs on-line just after they leave the IMM and the additional cooling station that is shown in this view.

more than 1.0 x 106 psi, and its HDTs to up to 200°C (400°F). A high percentage of its physical, mechanical, and electrical properties is maintained at elevated temperatures. For example, its flexural modulus remains above 0.3 x 106 psi at service temperatures as high as 160°C (320°F). Even after prolonged exposure to such temperatures, its resins do not discolor or degrade. Its thermal stability and oxidation resistance are also excellent at service temperatures well above 150°C (300°F). Heat aging a PSU increases its tensile strength, HDT, and modulus appreciably. However prolonged heat aging for about a year or so decreases its toughness, tensile strength, and elongation. PSU's creep, compared with that of other TPs, is exceptionally low at elevated temperatures and under continuous loads. For example, its creep at 99°C (210°F) is less than that of acetal or heat-resistant ABS at room temperature. The hydrolytic stability of these materials makes them resistant to water absorption in aqueous acidic and alkaline environments. Their combination of hydrolytic stability and heat resistance results in their having exceptional resistance to boiling water and steam, even under autoclave pressures and cyclic exposure to hot-to-cold and wet-to-dry repetitions. The PSUs also share the common drawback of absorbing UV rays, which gives them poor weather resistance. Thus, they are not recommended for outdoor service unless they are protected with paint or are plated or UV stabilized.

Figure 6-18. Bumdy's "surfmate" surface-mounted backplane connector has one of the highest pin densities of any commercial interconnecting device, with its 318 contracts arranged in four rows in a 12.95 cm- (5. I-inch) long connector. It is molded from Amoco's Mindel B-322 resin as standard PSU, to provide a unique combination of the best properties of the crystalline and the amorphous TPs. It permits the precision molding of these devices with less than a .013 cm (0.005 in.) total variation in flatness.

Figure 6-19. Amoco's Udel polysulfone, another standard PSU, molded by the Nalge Co. for a line of reusable filterware, provides user benefits that glass cannot offer. This TP can be injection molded with an undercut for ease in alignment and a positive fit when the top and bottom sections are clamped together. With its lighter weight and break resistance, PSU has FDA, NSF, and 4-A recognition and will withstand repeated autoclave cycles. It is also nontoxic and easy to clean. A comparable glass filter funnel would be priced almost two times that of PSU. A heavier, opaque stainless-steel filter would be the highest cost, about $300. 451

452 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

POLYURETHANE

The polyurethanes (PURs) produced by the reaction of polyisocyantes with polyester- or polyether-based resins can be either TPs or TSs. Extremely wide variations in form and physical or mechanical properties are available in PUR, which exhibit an extraordinary range of toughness, flexibility, and abrasion resistance (see Fig. 6-20). Its grades can range in density from !lb.lft. 3 in its cellular form to 70 lb.lft. 3 in a solid form. PUR's hardness runs from rigid, solid forms at 85 Shore D to soft elastomers. PUR materials are available in three forms: rigid foam, flexible foam, and as an elastomer. They are characterized by high strength and good chemical and abrasion resistance, with superior resistance to ozone, oil, gasoline, and many solvents. The rigid foam type is widely used as an insulation material in buildings, appliances, and such applications (see Fig. 6-21). The flexible foam is an excellent cushioning material for furniture, and the elastomeric type is used in solid tires, shock absorbers, and so on. It has outstanding flex life, cut resistance, and abrasion resistance. Some formulations are as much as twenty times more resistant to abrasion than are metals. The Tg of the flexible foams is well below room temperature, and for rigid foams the Tg is actually higher than room temperature. The cells of a rigid foam are about the same size and uniformity as those of a flexible foam, but rigid foams usually consist 90 percent of closed cells. For this reason their water absorption is low. Compressing the foam beyond its elastic limit will damage its cellular structure. Rigid foams are blown with either carbon dioxide or fluorocarbons, although the fluoros are now being replaced by a nondamaging ozone-blowing agent. Their thermal conductivity is influenced by the blowing agent used and its density. POLYVINYL CHLORIDE

The PVCs comprise a major volume of the plastics consumed worldwide and are the most commercially significant of the different vinyl polymers and copolymers. Although the vinyls differ in having literally thousands of varying compositions and properties, there are certain general characteristics that are common to nearly all these plastics. For one thing, most materials based on vinyls are inherently TP and heat sealable. The exceptions are the products that have been purposely compounded with TSs or crosslinking agents. For example, PVC can be chlorinated (CPVC) and be alloyed with other polymers like ABS, acrylics, polyurethane, and nitrile rubber to improve its impact resistance, heat deflection, and processability. In general, vinyls can be plasticized to give them a wide range of hardness ranging from thin, flexible, free films to rigid molded pieces. Most vinyls are naturally clear, with an unlimited color range for most forms of the materials. They generally have in common excellent water and chemical resistance, strength, abrasion resistance, and selfextinguishability. In their elastomeric form vinyls usually exhibit properties superior to those of natural rubber in their flex life, resistance to acids, alcohols, sunlight, and wear and aging. They are nontoxic, tasteless, odorless, and suitable for use as packaging materials that will come in contact with foods and drugs, as well as for decorative packaging requiring ordinary protection. The vinyl resins can be used in printing inks and be effectively used in coating paper, leather, wood, and, in some cases, plastics. In most forms vinyl can be printed. Rigid PVC, sometimes called the poor man's engineering plastic, has a wide range of properties for use in different products. In addition to the noteworthy properties mentioned,

THE PROPERTIES OF PLASTICS 453

Figure 6-20. A dock box made of a TS polyester-urethane hybrid glass-fiberreinforced material to meet demands for durability and toughness.

Figure 6-21. This 47.2-m- (I55-ft.) tall rocket-motor external fuel tank weighs 720 metric tons (1.6 million lbs .) at launch and must withstand the strain of liftoff. The thermal protective system on the tank is composed of a I-in. layer of a closed-cell polyurethane foam and a plastic charring ablator.

it has high resistance to ignition, good corrosion, and stain resistance, and weatherability. However, it is attacked by aromatic solvents, ketones, aldehydes, naphthalenes, and some chloride, acetate, and acrylate esters. In general, the normal impact grades of PVCs have better chemical resistance than the high-impact grades. Most PVCs are not recommended for continuous use above 60°C (140°F). Chlorination to form CPVC increases its heat resistance, flame retardancy, and density, depending on the amount of chlorination introduced. In regard to flammability, note that the vinyls release hydrochloric acid.

454 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

A popular form of flexible PVC is as a plastisol or organisol. The plastisols are much more popular, since they use plasticizers rather than the solvents used with the organisols, which release organic solvents during processing. There is no solvent problem, thus keeping the air clean, with plastisols. Basically, the PVC plastisols are medium- to highviscosity dispersions of a PVC in liquid plasticizers. Since only the application of heat is required to change such liquids to a solid, they are simple to use in casting parts or in the related processing operations of dip coating, rotational molding, slush molding, spread coating, and pouring. In each of these processes, the liquid in the mold or spread onto a fabric is heated to the fusion temperature, at which point the dispersion gels and then after a short time solidifies. The fusion temperature ofPVCs ranges from 135°C (275°F) to as high as 175°C (350°F). An example of how organisols are processed is the popular dip coating technique in which a metal male tool, such as a glove form or rain boot, is heated to 200 to 230°C (400 to 450°F), then dipped into a cold liquid plastisol. The resin immediately fuses around the form, with the material on the surface of the tool fusing the most completely and the resin on the outside being only partially gelled. The form is then withdrawn and placed in an oven or, if it is on a conveyor belt, is run through a heating tunnel, where the PVC will completely gel. The form is then cooled and the part stripped from the tool. This process can be highly automated to produce gloves, rain boots, drain-pipe seals, corrugated boots for shock absorbers, and many other items requiring the flexible, elastomeric properties of the vinyl plastisols.

Polyvinyl Acetate The PYAs copolymers are odorless, tasteless, nontoxic, slow burning, lightweight, and colorless, with reasonable low water absorption. They are soluble in organic ketones, esters, chlorinated hydrocarbons, aromatic hydrocarbons, and alcohols, but insoluble in water, aliphatic hydrocarbons, fats, and waxes. Water emulsions have extended the use of this resin. Used perhaps most extensively as adhesives, they are also employed as coatings for paper, sizing for textiles, and finishes for leathers, as well as bases for inks and lacquers, for heat-sealing films, and for flashbulb linings.

Polyvinyl Alcohol The major applications of the PYAs are in elastomeric products, adhesives, films, and finishes. Extruded PV A hoses and tubing are excellent for use subjected to contact with oils and other chemicals. PV A is used as a sizing in the manufacture of nylon.

Polyvinyl Butyral The PVBs are soluble in esters, ketones, alcohols, and chlorinated hydrocarbons but insoluble in the aliphatic hydrocarbons. They are stable in dilute alkalis but slowly decompose in dilute acids. PVBs are widely used as safety-glass interlayers and between sheets of acrylic to protect the enclosures of pressurized cabins in aircraft against shattering. Since 1938, PVB film in interlayers from 10 to 40 mils has been an important resource for the glass, automotive, and architectural industries. PVBs are also used as coatings for textiles and paper and as adhesives.

THE PROPERTIES OF PLASTICS 455

Polyvinylidene Chloride Molded parts in polyvinylidene chloride (PVDC) have high strength, abrasion resistance, strong welds, dimensional stability, toughness, and durability. This material is especially suited for injection molding at high speed parts that have heavy cross-sections. Molded PVDC fittings and parts are particularly valuable in industries involving the use of chemicals. For example, pipes of this material are superior to iron pipes to dispose of waste acids. Films produced from PVDC exhibit an extremely low water-vapor transmission rate, as well as flexibility over a wide range of temperatures and heat sealability. They are particularly suitable for various types of packaging, as for medicinal products, metal parts, and food. Food "packaging" for the home refrigerator uses the highly popular Saran wrap from Dow Chemical. POL YVINYLIDENE FLUORIDE Polyvinylidene fluoride (PVF) is a fluorine-containing TP. It is unlike other plastics, being a crystalline, high-molecular-weight polymer of vinylidene fluoride (CH2 = CF2 ). Compounds are available that contain 59 percent fluorine. This nonflammable plastic is mechanically strong and tough, thermally stable, resistant to almost all chemicals and solvents, and is stable to UV and extreme weather conditions. It has a higher strength and abrasion resistance than PTFE. Where unfavorable combinations of chemical, mechanical, and physical environments may preclude the use of other materials, PVF has been successfully used, as for valve and pump parts, heavy wall pipe fittings, gears, cams, bearings, coatings, and electrical insulation. Its limitations include lower service temperatures than the highly fluorinated fluoropolymers, not having antistick qualities, and the fact that it produces toxic products upon thermal decomposition. SILICONES The silicones (SIs) have service temperatures to about 260°C (500°F), good chemical resistance, low water absorption, good electrical properties, and are available in flameretardant grades. They differ from most other plastics in containing only silicon and oxygen (a siloxane bond) and having no carbon in their main polymer chain. This structure gives the SIs a wider temperature capability, with better moisture and oxidation resistance, than most of the carbon-chain polymers. A wide range of flexibility is obtainable in SIs with variations in their side groups and cross-linking, but their tensile strengths are generally inferior to those of the carbon polymers. Different fillers and reinforcements are used to improve SIs, mechanical properties, with the unwanted potential of reducing some of their inherent excellent characteristics. These high-cost molding compounds resulting from using fillers and reinforcements have found widespread use in applications where the retention of physical, electrical, and dimensional properties after long-term high-temperature exposure is important and good corrosion resistance is required. Generally, the SIs are molded by compression and transfer methods. Silicones go into many different products (see Figs. 6-22 and 6-23). The so-called room temperature vulcanizing (RTV) system is an example of a specialty silicone compound that is very popular. All kinds of products are made in SI, such as complex molds in artistically designed shapes. The RTVs come in one- and two-component systems and

456 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

AOHESIVES---- , ....._

....

COMPOUNOS FOIl ~OtNG-----

- . _ -:~.: ~:-

/

-

QAI;Ql'

-

WIItII NCI CA&I I!CIU..AnDfij

. / ___ EXTRUOING ==::;::::::===

~

...OWOICAI. """""

n--.

ELASTOMERS

Figure 6-22. An example of certain silicone products. do not require external heating to cure them. The one-component system gives off acetic acid or methyl alcohol during curing and requires having air-borne moisture to cure. Its by-products can cause shrinkage, offensive odors, and the corrosion of contacting metals. The more popular two-component systems require adding a curing agent. Depending on the compositions of the two components, the pot-life and curing times can be either fast or quite long. Such processing factors as length of time upon mixing (the pot life) determine whether an RTV can be fitted into a mold or form. Depending on the size or, more specifically, the thickness of the part, the thicker ones usually require a slower curing time to ensure a proper curing rate. A complete series of silicone compounds is theoretically possible ranging from linear polymers that are relatively soft and low melting to excessively hard, high-melting types. The properties of these materials can extend from true liquids through the TP stage to a rubbery stage and finally to a complex TS three-dimensional structure characterized by extreme heat resistance, inertness, and hardness.

UREA FORMALDEHYDES Urea formaldehyde TS molding compounds are in the amino family of plastics. The UFs are available in a wide range of colors, from translucent colorless and white through all the colors to a lustrous black. Unlike the colored phenolic compounds, the molded UPs can be made with a considerable degree of translucency, giving them a brightness and depth of color somewhat similar to, although better than, opal glass. Although the UFs' appearance can be duplicated in some TPs, the fact that they are not affected by heat within their range of operating temperatures makes their use necessary in many cases.

THE PROPERTIES OF PLASTICS 457

Figure 6-23. The General Electric Living Environment Concept House has the building's exterior stone surface coating with a water-repellent silicone penetrant (GE's TWR 255) that helps prolong the service life of this stuccolike surface by preventing water and salt intrusion and inhibiting algal growth. It is clear, nonglossy, nonyellowing, and moisture-vapor permeable so that substrates do not crack or spall. These noninflammable (self-extinguishing), odorless, and tasteless materials char at about 200°C (395°F). Temperatures from -21°C (-70°F) to 80°C (175°F) have no effect on them, but higher temperatures over prolonged periods will cause fading and eventual blistering. When used within their temperature limitations, UFs have good electrical properties. They have high dielectric strength, high arc resistance, no tendency to track after arcing, and a low order-power factor. Their electric~ properties are not greatly influenced by high humidity, and they resist static electricity buildup. Under dry conditions, UF moldings are remarkably resistant to corrosive fumes and have no effect on organic solvents in terms of absorption, swelling, or changes in appearance. However, their water absorption is relatively high. For this reason they are not recommended for applications involving continuous or intermittent exposure to water, although occasional exposure will have little effect if the material has been well cured. The UP intermediate water-soluble products are starting materials for the production of adhesives, surface coatings, paper conditioners, and other such special items that require heat and catalysts for their final curing. They are used to bond laminated sheets, sometimes only on the surface, for color effects. The applications ofUFs include sanitary wares such as toilet seats, and knobs, closures, buttons, electrical accessories like housings and switches, laminates, and so on. Compounds of UFs use different additives, fillers, and reinforcements to provide different characteristics and permit processing in different equipment, principally by compression, transfer, and injection molding. Like many other TSs, such as the phenolics and melamines, they are easily preformed and preheated, either by RF preheaters or with screw

458 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

plasticators. The higher bulk-factor grades of the melamine compounds require using special equipment for these operations, because of their lack of easy pourability. Usually the ureas, melamines, and phenolics are compared to each other, as is done throughout this book, since they continue to be old competitors. They each have their place for the designer. Phenolics have by far the largest market, because of their performance and cost advantages. The advantages of the melamines over the ureas include their better retention of electrical properties at elevated temperatures and in the presence of moisture, their superior heat resistance, improved surface hardness and gloss, better resistance to staining, lower water absorption, and being better for products that will be in contact with food and drink. ELASTOMERS

An elastomer may be defined as a natural or synthetic material that exhibits the rubberlike properties of high extensibility and flexibility. Although the term rubber originally meant the TS elastomeric material obtained from the para rubber tree (Hevea braziliensis) it now identifies any thermoset elastomer (TSE) or thermoplastic elastomeric (TPE) material. Such synthetics as neoprene, nitrile, styrene butadiene, and butadiene are now grouped with natural rubber. These TSEs and TPEs serve engineering's needs in fields dealing with shock absorption, noise and vibration control, sealing, corrosion protection, abrasion and friction resistance, electrical and thermal insulation, waterproofing, and all types of load-bearing products [470-81]. These materials are differentiated on the bases of how long such a material if deformed requires to return to its approximately original size after the deforming force is removed, and by its extent of recovery. The standard ASTM D 1566 defines an elastomer as a macromolecular material that is capable at room temperature of recovering substantially in shape and size after the removal of a deforming load. This standard has details on the rates of conducting tests as they relate to quick and forcible deformations. Basically, an elastomer must be capable of retracting within one minute to less than 1.5 times its original length after being stretched at room temperature to twice that length and being held for one minute prior to release. (Consult the standard for the actual definition, the test conditions, and so on.) The ASTM D 2000 and SAE J 200 standards designate rubber or elastomeric materials according to their performance in thermal and oil immersion tests (see Tables 6-10 and 6-11). Thermal tests define the types based on their maximum service temperatures, ranging from 21 to 135°C (70 to 275°F) for rubber and 71° to 274°C (160 to 525°F) for elastomers, using letters A through J. The class designations are based on the maximum volume of swelling upon immersion in the prescribed ASTM #3 oil test and use the letters A through K to designate the ten classes of their volume swell behavior. These type and class designations are then written together. Design Approaches for Elastomers

Selecting an elastomer for an application requires consideration (like for plastics and foams) of many factors, including the mechanical and physical service requirements, the product's life cycle, the material's process ability , and its cost (see Figs. 6-24 and 6-25 and Tables 6-12 and 6-13). A wide range of properties is available, based on the many different compounds that can be produced.

THE PROPERTIES OF PLASTICS 459

Table 6-10. Basic Requirements of Elastomers* Type A

B C D E F 0 H J

Temperature, °C (oF) 70 100 125 150 175 200 225 250 275

Class

(158) (212) (257) (302) (347) (392) (437) (482) (527)

A

B C D E F 0 H J K

Volume Swell, Max. % Not required 140 120 100 80 60 40 30 20 10

OPer ASTM D 2000 and SAE J 200; based on type and class designations.

Stress-Strain Curves The standard stress-strain curve described in Chapter 3 displays important characteristics for rigid or elastomeric materials. Since elastomeric designs usually do not need high strengths, elongation is important and can be related to flexibility and softness. Elastomers are capable of extreme elastic deformation at low levels of stress. This strain is not proportional to stress (see Figs. 6-26 and 6-27). Like solid plastics, the elastomers become brittle below their Tg. which is in the range of - 20°C ( - 4°F) for most rubbers and about - 60°C ( - 75°F) for natural and silicone rubbers.

Table 6-11. Some Popularly Used Elastomers per ASTM D 2000 and SAE J 200 Type ~ ~ Class

AA

AK BA BC BE

BF

BO

BK CA CE

CH DA DE DF

DH

FC FE

FK OE

HK

Typical Rubber Natural rubber, styrene butadiene, butyl, ethylene propylene, polybutadiene, Polyisoprene Polysulfide Ethylene propylene, styrene butadiene (high temperature) Butyl Chloroprene, chlorinated polyethylene Chloroprene, chlorinated polyethylene Nitrile Nitrile, urethane Polysulfide, nitrile Ethylene propylene Chlorosulfonated polyethylene, chlorinated polyethylene Nitrile, epichlorohydrin Ethylene/acrylic Ethylene propylene Chlorinated polyethylene, chlorosulfonated polyethylene Polyacrylate (butyl-acrylate type) Polyacrylate Silicone (high strength) Silicone Fluorinated silicone Silicone Fluorinated rubbers

TSEs High Cost

Fluouroelastomer Acrylate Eplchlorohydrin Nitrile CIIlorosu/fonllted Polyethylene Po/ychloroprene EPDM Butyl Rubber

Low Cost

Natural Rubber SBR Special Purpose

Low Performance Commodity

High Performance Specialties

TPEs High Cost

Polyamides Copolyesters Urethanes EllIstomeric Alloys Olefinic Blends

Low Cost

Styrenics Low Performance Commodity

Special Purpose

High Performance Specialties

Figure 6·24. A general guide for TSEs and TPEs regarding cost as opposed to performance.

RlJBBFR f II XIHII P\ ( 75A

TH~RMOSFT

50A

I (lOA 1 HLRMOPLASTIC flASTOMI:RS

75A

40D

55D

60D

72D

POI'rf THYl ENE lOW DENSITY

HIGH [)( NSITY

POlYPROPYU-NF

Figure 6·25. Durometer-scale relationships and hardness ranges for a few TPEs. The letter designations refer to Shore hardness tests. 460

THE PROPERTIES OF PLASTICS 461

Elastomer parts were usually designed using either closed-form equations or empirical equations developed from previous successful designs. Typically, tooling is built and a few parts molded and tested. Then, based on the results, changes are usually made in the tooling or elastomeric compound to meet the final requirements. This procedure is then repeated as necessary. To help the advanced designing of elastomeric products for the aerospace and petroleum markets, finite element analysis (FEA) techniques may be used to help reduce development problems. The FEA approach is gradually being employed for other marketable products.

Fabrication Techniques

The TSEs tend to be fabricated in compression and transfer-molding equipment. As is typical of TS materials, they produce flash, but with proper mold design, it can be significantly reduced. The TPEs can be processed by all the processes seen in Chapter 7, basically using faster cycle times that are more efficient, with no flash and so on [1012]. TPE processing eliminates the need for the relatively slow, laborious, energy-intensive batch processing associated with TSEs. However, there are TSE-processing techniques that can be used that do provide the efficiency associated with the TPEs.

Table 6-12. Potential Advantages and Disadvantages of Some Natural and Synthetic Elastomers Rubber

Advantages

Disadvantages

Typical Applications

Natural rubber (NR)

Building tack, resilience, and flex resistance Abrasion resistance

Reversion at high molding temperatures Poor ozone resistance

Tires, engine mounts

Good ozone resistance

Poor hot·tear resistance

Good solvent resistance

Poor building tack

Short injection molding cycle (thin parts) Short molding cycle (RIM) and low molding pressure

Poor creep characteristics

Low air permeation in finished parts

Voids caused by trapped air during molding

Moderate solvent resistance

Sticking during processing and premature crosslinking (scorch) with some types

Styrene butadiene rubber (SBR) Ethylene propylene diene monomer (EPDM) Nitrile butadiene rubber (NBR) or nitrile Thermoplastic rubber Polyurethane

Isobutylene isoprene rubber (IIR) or butyl Chloroprene rubber (CR) or neoprene

Adhesion to mold

Tires, general molded goods Door and window seals, wire insulation O-rings and hose

Shoe soles, wire insulation Cushioning, rolls, exterior automotive parts Inner tubes, body mounts for automobiles Hose tubes and covers, V belts

~ N

Oil Resistant (slight swelling in oils, fuels, etc.)

Natural rubber Styrene-Butadiene GRS (thermoset}various Butyl-Exxon Ethylene-Propylene (EPR and EPOM}-various

Non-oil Resistant (swelling or decomposition in presence of oil, fuel, etc.)

Excellent high-temperature, oil, and chemical resistance; very expensive; 65-80 Shore A Very good resistance to chemical deterioration; poor lowtemperature properties; expensive; 60-80 Shore A

-60 to 150

-20 to 550

- 10 to 400

Toys, household items, wire insulation, flexible hose, toys, household items, medical devices Seals, a-rings for special applications Seals, a-rings for hightemperature applications

Ethylene-Ethyl Acetate (EEA), EthyleneVinyl Acetate (EVA}-U.S. Ind. Chern. Perftuoroelastomer: KaJrez-Ou Pont

Fluorocarbon: VitonOu Pont

Tough, rubbery materials; 35 to 95 Shore A; inexpensive; can be blended with polyethylene or polypropylene to increase toughness High flexibility; can be clear; can be filled; moderate price

Very low gas permeability Excellent weather resistance

High resilience, abrasion resistance, tear strength Good abrasion resistance

-80 to 150

-60 to 300

-65 to 220

-65 to 220

-65 to 220

Major Characteristic

Footwear, toys, household items

Inner tubes Automobile weather stripping, seals

Tires

Elastic cord, tires

Typical Applications

Styrene-Butadienes: Kraton-Shell Solprene-Phillips

TPEs

Polymer and Source

Type

Useful Temperature Range, "F

Table 6-13. Properties and Applications of Elastomers

!

Sealing material

Seals, O-rings

Polysulfide-Thiokol Corp.

Fluorosilicone-Dow Corning

Acrylic: Cyanacryl-American Cyanamid Vamac-Du Pont

Seals

Automotive transmission seals

Gasoline hose, cable covers

Neoprene (Chloroprene)Du Pont

Chlorinated Polyethylene (CPE)-Dow Chemical Acrylic, Ethylene

Hydraulic O-rings, aromatic fuel seals

Acrylonitrile (Buna N}-various

-30 to 250

o to 400

-10 to 300

-120 to 350

-65 to 225

-20 to 300

-65 to 250

(cont'd)

Low swell and good strength in oil, fuel, etc.; good abrasion resistance; inexpensive; 40-80 Shore A High strength, good abrasion resistance, excellent weathering, best generalpurpose rubber, 40-95 Shore A, moderate price Good resistance to many solvents, low mechanical and thermal properties, 40-85 Shore A, inexpensive Very good flexibility at low temperatures, good resistance to high temperatures, oils and gasolines, expensive, 50-80 Shore A High temperature and combustion resistant, good weathering, 50-95 Shore A, moderate price Good chemical resistance, poor low-temperature properties, 55 Shore A, moderate price 50 Shore A, moderate price

~

~

a'>

Heat resistant

Type

Fluorocarbon: Viton-Du Pont Fluorel-3M Co.

-20 to 400 Seals, O-rings, etc., not requiring lowtemperature flexibility below -20°F

-20 to 550

Seals, O-rings

Perfluoroelastomer: Kalrez-Du Pont Silicones: Dow Coming, General Electric Co.

Available as curable gum stock, RTV materials, and one-part air-curing; excellent weathering and electrical properties; 30-90 Shore A; moderately expensive Poor low-temperature flexibility, high heat resistance, good weathering and electrical properties, good oil resistance

See above

-60 to 250

Household items, mechanical parts

Olefinic Rubbers: Somel-Du Pont Telca-Goodrich TPR-Uniroyal Vistaflex-Exxon

-100 to 450

Very tough, high-performance material; moderate to highpriced; 40-50 Shore D Medium hardness and performance materials, 50 Shore A to 50 Shore D, inexpensive

-65 to 300

Couplings, gears, mechanical parts

Seals, O-rings, etc., not requiring oil resistance

High strength and excellent abrasion resistance, 75 shore A to 60 Shore D, moderate price

Major Characteristic

-65 to 200

Typical Applications

Useful Temperature Range, OF

Seals, cable jacketing, abrasion-resistant parts

Polyurethane: Estane-Goodrich Pellethane-Upjohn Plastothane-Thiokol Corp. Roylar-Uniroyal Texin-Mobay CyanapreneAmerican Cyanamid Polyester: Hytrel-Du Pont

TPEs

Polymer and Source

Table 6-13. (Continued)

....

(J1

=-

Weather, light, UV, and ozone resistant

TPEs Polyester: Hytrel-Du Pont Olefinics: Somel-Du Pont Telcar-Goodrich TPR-Uniroyal

Neoprene-Du Pont

Chlorosulfonated Polyethylene: Hypalon-Du Pont Silicones and FluorosiliconsDow Comings and G.E. Co. Fluorocarbons: Viton-Du Pont

Auto transmission seals

Acrylic: CyanacryleAmerican Cyanamid TPEs Polyester: Hytrel-Du Pont Ethylene-Propylene: Nordel-Du Pont, Epcar-Goodrich, Vistalon-Exxon

-60 to 250

Seals, noncritical items

See above

See above

(cont'd)

-20 to 300

-65 to 300

See above, moderately expensive

-20 to 400

Fairings, seals, O-rings, etc. ,. requiring good weather resistance and chemical resistance Cable covers, etc., where good all-around properties are required Mechanical parts, seals

See above, expensive

-100 to 450

Fairings, seals, etc. requiring flexibility at low temperatures

Excellent weather resistance, poor oil resistance, good thermal and meachenical properties, resistant to phosphate ester fluids, inexpensive Excellent weather resistance, need not be black, relatively inexpensive See above, expensive

See above

See above, except good lowtemperature properties, fair mechanical properties See above

-50 to 300

-65 to 300

-65 to 300

o to 450

-120 to 350

Flexible coatings, tarps, etc.

Dust seals, fairings exposed to weather

Mechanical parts

Seals, O-rings, etc., in contact with oil

Fluorosilicone: Dow Coming

CI'I CI'I

..

High gas permeability Damping and energy absorbing

Styrene-Butadiene, Ethylene Polymers, Polyesters Urethanes, Olefinic Elastomers

Polynorborene: Norsorex-American Cyanamid

Thermoplastic elastomers TPEs

Extremely low Shore

(RTV)

Room-temperature vulcanizable

Thermally conductive

Silicone filled with electrically conductive particles Silicone and other polymers filled with thermally conductive material Silicones

Fluorocarbons: Viton-Du Pont Silicones and Fluorosilicones Silicone, Fluorosilicone, Chlorobutyl Polyurethane

Electrically conductive

Butyl

Polymer and Source

Specialty Rubbers Low gas permeability

Type

Very soft dampers, etc.

Form-in-place seals, coatings, prototype parts Various parts not to be operated at temperatures over 120140"F (except polyester, 3oo"F)

Conforming gaskets between hot parts and heat sink

RFl shielding

Seals, O-rings, in contact with oil and solvents "Breathable" diaphragms and seals Shock absorbers, mounts, and vibration dampers

Seals, O-rings, not in contact with oil

Typical Applications

Table 6-13. (Continued)

Many types differing mainly in hardness and mechanical properties Thermoplastics, injection moldable at production rates. Used for high production of lightcolored or transparent parts. Polyesters for high-strength harder parts, urethanes for abrasion resistance Very soft to hard, good mechanical properties

-120 to 450

-65 to 160

High thermal conduction and conformance to rough surfaces

-100 to 450

-100 to 400

-65 to 300

Very low resilience, "dead" materials, moderate physicals. Silicones and ftuorosilicones can be used over wider temperature range than chlorbutyl or polyurethane Conductive materials with fair physical properties

See above

-100 to 450

-20 to 400

Best for gas retention, medium physical properties, medium cost See above

Major Characteristic

-65 to 250

Useful Temperature Range, of

THE PROPERTIES OF PLASTICS 467

THERMOSET ELASTOMERS What follows is a general overview of the various types of TSEs [470-79]. Natural Rubber

The commercial base for natural rubber (NR) is latex, a milklike serum produced by the tropical tree Hevea brasiliensis. Naturally occurring latex is the rubber that exudes from these trees in an aqueous serum containing various inorganic and organic substances. The rubber precipitated out of this solution can be characterized as a coherent TS elastic solid. It is against NR that all the other rubbers and elastomers are measured. For centuries it was the only rubber available; it was extensively used even prior to the discovery of vulcanization (TS cross-link curing), in 1883. To date no synthetic material has yet equaled the overall depth of engineering characteristics and consequent wide latitude of applications available in NR. As with the other elastomers, many grades and types of NR are available, as produced by varying impurity levels, collection methods, and fillers and processing techniques. NR is generally considered to be the best of the general-purpose rubbers, meaning those with characteristics suitable for broad applications. With NR, compounds can be produced over a wider stiffness range than with any other material. NR could be the best choice, except where an extreme performance or exposure requirement dictates the use of a special-purpose elastomer, which will often occur at the sacrifice of some other, less critical property. Natural rubber has a great capacity for being deformed. This and its ability to strain crystallize gives it added strength while deformed. Its high resilience, which is responsible for its low heat buildup during flexing, makes it a prime candidate for shock and severe dynamic loads. With its low heat buildup NR is recommended for use in applications where such properties as flexure, cut and abrasion resistance, and general endurance would be adversely affected by heat in less resilient elastomers. The shortcomings of NR include its service temperature, from -18 to 120°C (- 6 to 248°F). Its poor oil, oxidation, and ozone resistance can be minimized either by proper design accommodation or by compounding. Degradation from such environments is essentially a surface effect that can be tolerated or minimized by using thicker crosssections, shielding, or adding antioxidants and antiozonants. Poly isoprene

The synthetic rubber that comes closest to duplicating the chemical composition of NR is synthetic polyisoprene (IR). It shares with NR the properties of good uncured tack, high unreinforced strength, good abrasion resistance, and the characteristics that provide good performance in dynamic applications. However, because of having some inherent impurities, NR is somewhat better overall. An IR is distinguished by its low hysteresis and high tensile strength, but it is readily attacked by solvents, gasoline, and ozone. Its tensile strength is in the range of 24 to 31 MPa (3,500 to 4,500 psi), and its elongation is 550 to 650 percent. One significant disadvantage of IR is its lack of green strength, which is to say during the time period during processing, prior to curing. An IR can be used interchangeably with NR in all but the most demanding parts. The applications are about the same as for NR.

468 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

"iii

.3-

'"'" !:: '" .!:!

400

c:
200


"iii

~

o

200

400

1100

600

Elongation (percent)

Figure 6·26. The stress-strain curve for TSE polyisobutylene.

3000

r-----------...,

20.0

2500 15.0 2000

£a ...

'iii

Co

iii

... Ul

Q)

("j)

1500

10.0

en

!b

3:

en

~

III

1000 5.0 500

o

o

100

200

300

400

0 500

Strain, percent elongation Figure 6·27. Stress-strain curves for TPEs at different durometers of hardness.

Styrene Butadiene This elastomer, SBR, emerged as a high·volume substitute for NR during World War II because of its suitability for use in tires, belts, hoses, rubber floor tiles, and the like. Its tensile strength after compounding it with carbon black and vulcanizing it is 17 to 24 MPa (2,500 to 3,500 psi), which is less than NR's, but it has an elongation of 500 to 600 percent. In abrasion and skid resistance it is superior to NR, but has better resistance to solvents and weathering. This general·purpose rubber continues to be used in many applications where it has replaced NR, even though it does not have NR's overall versatility, because it meets its performance requirements but has a cost advantage over NR. For most uses SBR must be reinforced to have acceptable strength, tear resistance, and general durability. It ~s significantly less resilient than NR, so it has higher heat buildup upon flexing. It also lacks NR's green strength and tack.

THE PROPERTIES OF PLASTICS 469

Polybutadiene This general-purpose, crude-oil-based rubber is more resilient than NR. Polybutadiene (BR) was the material that made the solid golf ball possible. It is superior to NR in its low-temperature flexibility and in having less dynamic heat buildup. However, it lacks NR's toughness, durability, and cut-growth resistance. BR can be used as a blend in NR, SBR, and other materials to improve their low-temperature flexing, but this is achieved at a much higher price and at the sacrifice of other key properties, such as tensile strength, tear resistance, and general durability.

Neoprene Except for BR and JR, neoprene (CR) is perhaps the most rubberlike of all materials, particularly with regard to its dynamic response. CRs are a family of elastomers with a property profile that approaches that of NR, but it has better resistance to oils, ozone, oxidation, and flame. CRs age better and do not soften up on exposure to heat, although its high-temperature tensile strength may be lower than that of NR. Neoprene comes in grades suitable for service at 250°C (4800 P) and has maximal resistance to oils and greases. These materials, like NR, can be used to make soft, relatively high-strength compounds. One important difference is that, in addition to neoprene's being more costly by the pound than NR, its density is about 25 percent greater than NRs. CRs also do not have the low-temperature flexibility of NR, which detracts from their use in low-temperature shock or impact applications.

Nitrile The nitriles (NBRs) are copolymers of butadiene (B) and acrylonitrile (AN) that are used primarily for parts requiring resistance to petroleum oils and gasoline. Their resistance to aromatic hydrocarbons is better than is neoprene's, but not as good as polysulfide's. NBR has excellent resistance to mineral and vegetable oils, but relatively poor resistance to the swelling action of oxygenated solvents like acetone, methyl ethyl ketone, and various other ketones. With its higher AN content its solvent resistance is increased but low-temperature flex decreased. The low-temperature resistance of NBR is inferior to that of NR and, although it can be compounded to improve its performance in this area, the gain is usually at the expense of its oil and solvent resistance. NBR's tear strength is inferior to NR's, and its electrical insulation rating is also lower. NBR is used instead of NR where increased resistance to petroleum oils, gasoline, or aromatic hydrocarbons is required.

Polyurethane These elastomeric combinations of polyesters, or polyethers and diisocyanates, are unusual in that their physical properties do not depend upon the compounding materials. The PURs cross-link and undergo chain extension to produce a wide variety of compounds available both as castable or liquid materials and as solids or mill able gums. The PURS have outstanding abrasion resistance and excellent tensile strength and load-bearing capacity, with an elongation potential that is accompanied by high hardness. Their properties also include low temperature resistance, high tear strength, either a high or a low coefficient of friction, good radiation resistance, and good elasticity, with resilience, even in very hard stock.

470 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Butyl The two types of rubber in this category of butyl (HR) rubber are both based on crude oil. One is polyisobutylene, with an occasional isoprene unit inserted into the polymer chain to improve its vulcanization characteristics. The other butyl is the same, except that chlorine is added to it at about 1.2 percent by weight, resulting in its having greatest vulcanization flexibility and cure compatibility with the other general-purpose rubbers. The HRs have outstanding impermeability to gases and excellent oxidation and ozone resistance. The chemical inertness of HRs is further reflected in their lack of molecular-weight breakdown during processing, which permits the use of hot mixing processes to promote the polymer-filler interaction. Their flexing, tear, and abrasion resistance approach those of NR, and with the HRs' moderate strength (14 MPa, or 2,000 psi) unreinforced compounds can be made at competitive costs. It should be noted that butyls lack the toughness and durability of some of the general-purpose rubbers.

Ethylene Propylene Like the butyls, there are basically two types of ethylene propylene. One is a fully saturated chemically inert copolymer of ethylene and propylene (EPR). The other, called EPDM, has the ethylene and the propylene, plus diene monomer. EPDM is chemically reactive and capable of sulfur vulcanization. The copolymer, EPR, is cured with a peroxide catalyst. The physical properties of EPR and EPDM are not as good as those obtainable with NR. Nevertheless, their property retention is better than that of NR on exposure to heat, oxidation, or ozone. Their bonding is somewhat more difficult, however, especially with EPR. They have a broad resistance to chemicals, but not to oils and other hydrocarbons. Their electrical properties are good.

Chlorinated Polyethylene This family of elastomers is produced by the random chlorination of HOPE (the proprietary Tyrin is from Dow Chemical). The properties of chlorinated polyethylene (CM) include excellent ozone and weather resistance, heat resistance to 149°C (300 0 P) and even 177°C (350 0 P) in many types of oil, dynamic flex resistance, and resistance to abrasion.

Chlorosulfonated Polyethylene This elastomer, Du Pont's Hypalon, has excellent combinations of properties. Among others, they include total resistance to ozone along with excellent resistance to abrasion, weather, heat, flame, oxidizing chemicals, crack growth, and dielectric properties. They also have low moisture absorption and can be made into a wide range of colors, because carbon black is not required as a reinforcement. Their resistance to oil is similar to neoprene's. Their low-temperature flexibility is only fair, at -40°C (-40 0 P).

Acrylates This family of different acrylate polymers is highly resistant to oxygen and ozone, has excellent flex life and permeability resistance, and is resistant to oil swell and deterioration,

THE PROPERTIES OF PLASTICS 471

among other characteristics. Their heat resistance is superior to that of all other commercial rubbers, except for the silicones and the fluorine-containing rubbers. The water resistance of the acrylates (ACMs) is poor; nor is their low-temperature flex good, and these rubbers decompose in alkaline solutions and are swelled by acids. The Ethylene/Acrylics This family of rubbers, Du Pont's Vamac being perhaps best known, are moderately priced, heat and fluid resistant, and surpassed only by the expensive specialty types such as the fluorocarbons and fluorosilicones. A special feature of the ethylene/acrylics (EAMs) is then nearly constant damping characteristic over a broad range of temperatures, frequencies, and amplitudes. They have good resistance to hot oils and hydrocarbon- or glycol-based proprietary lubricants as well as to transmission and power-steering fluids. EAMs are not recommended, however, for use with esters, ketones, or high-pressure steam. Polysulfides These polymers have outstanding resistance to oils, greases, and solvents but have an unpleasant odor, their resilience is poor, and their heat resistance is only fair. The abrasion resistance of polysulfides (PTRs) is half that of NR, and its tensile strength runs only from 8.3 to 9.7 MPa (1,200 to 1,400 psi). However, these values are still retained even after extended immersion in oil. Their increased sulfur content improves their solvent and oil resistance but reduces their permeability to gases. Silicone Silicone rubber (SM) comprises a versatile family of semiorganic synthetics that look and feel like NR, yet have a completely different type of structure and performance from other rubbers. They have no molecular orientation, crystallization, or propensity for stretching and must be strengthened by reinforcements. Although they are on the high end of the cost scale, they can be made to withstand temperatures as high as 315°C (600°F) without deteriorating, and they retain their flexibility down to -129°C (- 200°F). Even though their strength is low (609 MPa or 1,000 psi) compared to other rubbers, SMs have outstanding fatigue and flex resistance. They also do not require high tensile strength to perform in dynamic applications. Their fall-off in tensile properties with extended exposure to high temperatures is much less than for other materials; also under these conditions they resist chemical deterioration, oil, oxygen, and ozone. SMs' chemical inertness makes them well suited for surgical and food-processing equipment. Fluorosilicone This type of silicone, fluorosilicone or FVMQ, provides most of the useful qualities of silicones plus improved resistance to many hydrocarbon fluids, except the ketones and phosphate esters. However, they can be blended with dimethyl silicones, which have good resistance to these fluids at temperatures up to 149°C (3000P) and are most useful where the best in low-temperature flexing is required, in addition to fluid resistance. FVMQ has moderate dielectrics, a low compression set, and excellent resistance to ozone and weathering.

472 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Fluorocarbons The high-perfonnance, high-cost fluorocarbon (FKM) rubbers have outstanding resistance to heat and to many chemicals, oils, and solvents. In air, parts made of FKM retain at least half their original properties after 16 hours' exposure at 315°C (600°F). The same materials offer low-temperature stability down to - 40°C ( - 400 P).

THERMOPLASTIC ELASTOMERS There are different thennoplastic elastomers (TPEs) that meet different requirements (see Table 6-13). As is the case with many TSEs and plastics, their properties can be varied and controlled by varying the ratio of the basic monomers used to compound TPEs, as well as by changing the types and amounts of additives and fillers.

Polyurethanes The first major elastomer that could be processed without vulcanization was the urethanes. The TP polyurethanes (PURs) do not have quite the heat resistance and compression-set resistance of the TS PURs, but most of their other properties are the same. The abrasion resistance among the elastomers is outstanding, their low-temperature flex is good, PURs' oil resistance is excellent at 82°C (180~, and their load-bearing capability ranks them with the best of all the elastomers. Additives can improve their dimensional stability or heat resistance, reduce friction, or increase their flame retardancy, fungus resistance, or weatherability. The excellent abrasion resistance of PURs qualifies them for use in bumpers, gears, rollers, sprockets, solid tires, vibration-damping components, and other such products. The newest, with lower-molecular-weight polymers, have better color stability in regard to UV radiation and hydrolysis than do the conventional grades. The softer grades are used with suitable antioxidants in medical applications and as adhesives. Other new grades have been stabilized for use as wear layers in aircraft wings and the like.

Copolyesters These TPEs are generally tougher over a broader temperature range than are the PURs. They are also easier to work and more forgiving in processing. Several grades, produced by Du Pont (Hytrel), Hoechst-Celanese (Riteflex), and Eastman (Ecdel), range in hardness from 35 to 72 Shore D. These TPEs can be processed by conventional plastics equipment. The copolyesters and urethanes are high priced, but they have excellent dynamic properties, good elongation and tear strength, and good resistance to flex fatigue at both low and high temperatures. Their brittle temperature is below 32°C (90°F), and at -40°C (-400 P) their modulus is only slightly higher than at room temperature. Copolyesters' heat resistance is good to 149°C (3000 P). The copolyesters' resistance to nonoxidizing acids, some aliphatic hydrocarbons, aromatic fuels, hot oils, and hydraulic fluids ranges from good to excellent. Thus, they compete with such rubbers as nitriles, epichlorohydrins, and acrylates. However, hot polar materials, strong mineral acids and bases, and chlorinated solvents and creosols degrade the copolyesters. Their weather resistance is low but can.be improved considerably, with UV stabilizers and carbon-black additives.

THE PROPERTIES OF PLASTICS 473

These TPEs are not direct substitutes for TS rubbers in existing designs. Rather, such parts must be redesigned to use their higher strength and modulus and operate within the elastic limit. Thinner parts, from about! to i in., are usually used. Styrene Copolymer

The styrenics, generally, are the lowest-priced TPEs. They range in hardness from 28 to 95 Shore A. Their tensile strength is lower and their elongation higher than those of SBR or NR, but their weather resistance is about the same. Styrenics' properties can be improved by alloying them with such resins as PP and EVA. They resist water, alcohols, and dilute alkalis and acids. They are soluble in or are swelled by strong acids, chlorinated solvents, and ketones. One type has a service-temperature limit of 65°C (150°F), another to 121°C (250°F). Both have excellent low-temperature flex at - 85°C ( - 120°F). Olefins

The TP olefin elastomers are available in several grades having a room-temperature hardness from 60 Shore A to 60 Shore D. They have the lowest density of all the TPEs, and their cost is in the mid-range of the TPEs. Olefins' flexibility remains down to - 51°C ( - 60°F), and they are not brittle at 32°C (90°F). They are autoclavable and can be used up to 135°C (275°F) in air. These TPOs have good resistance to some acids and most bases. They are attacked by chlorinated hydrocarbon solvents. Olefin compounds rated v-o by UL 94 are available. Polyamide Alloys

These TPEs have high elongation, with good solvent and abrasion resistance and low density. They are designed to replace such TS rubbers as EPDM, nitrile, and neoprene. Elastomeric Alloys

This class of TPEs consists of mixtures using two or more polymers that have received proprietary treatment to give them properties significantly superior to the simple blends of the same constituents. The two basic types are the TP vulcanizates, called TPVs, and the melt-processable rubbers (MPRs). The TPVs are essentially a fine dispersion of highly vulcanized rubber in a continuous phase of a polyolefin. TPVs' cross-linking gives them high tensile strength (7.6 to 26.9 MPa, 1,100 to 3,900 psi), high elongation (375 to 600 percent), resistance to compression and tension set, oil resistance, resistance to flex fatigue, and a maximum service temperature of 135°C (275°F). The specific gravity of TPVs is 0.9 to 1.0, of MPR, 1.2 to 1.3.

FILM AND SHEETING Several secondary forms of plastics have become families of materials in their own right. Film and sheeting, for example, can be used by themselves-package wrapping, tank liners, or industrial structures-or can be shaped into products by various types of thermoforming (see Chapter 7). Virtually all TPs can be made into film or sheeting by extrusion, casting, or calendering (see Table 6-14 and Fig. 6-28).

474 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 6-28. With the touch of a button, interior windows can change from clear to translucent, using liquid-crystal TP polyester film. This window consists of a two-layer laminated GE Lexan sheet to protect an inner polyester film. This polycarbonate product is strong and mar and UV resistant. The film is able to switch electrically between a highly translucent state providing privacy, glare control, and shading to a clear state for good visibility .

There are other plastics used in addition to those listed in Table 6-14. One example is a polyphenylene sulfide film that is biaxially oriented (BOPPS), with 40 percent crystallinity. It is roughly equivalent to a polyester film in its mechanical properties. This self-extinguishing PPS has a long-term heat resistance per UL 746 B of 160°C (320°F) for mechanical and 180°C (356°F) for electrical resistance.

Strapping Extruded strapping tape is another example of a whole family of materials. This basically narrow-width sheeting that is usually oriented to maximize its performance has important uses in packaging applications where only steel strapping had previously been available. Steel and plastic strappings each have their own advantages and disadvantages, so the designer can select the best type based on the product's requirements. For example, when

THE PROPERTIES OF PLASTICS 475

strapping is used where the heat could fluctuate, as in a railroad boxcar or storage room, the steel at high temperature could expand and cause what it contained to become loose. The elastic deformation of plastics lets them retain their tension with temperature change, however. Some typical strapping properties for PP, nylon, polyester (TP) and low-carbon steel are as follows: the breaking strength of a piece! in. wide by 0.020 in. thick is SOO, 640, 600, and 1,200 lbs., respectively; the working-range elongation is 5, 7, 2!, and 1110 percent; the elongation recovery in in. is S.S, 9, 2, and 1110; the ratio ofretained tension at maximum working range to a time after 24 hours is (in psi) 200/S0, 2S0/17S, 300/ 244, and 700/66S; their heat resistance is fair, good, good, and excellent, respectively; and their humidity resistance is excellent, fair, excellent, and excellent.

FOAMS Foamed plastics, whether TPs or TSs, are a special category of the plastics family. They are available with open-celled construction, closed or interconnecting construction, or in combination. Their densities range from 1.6 to over 960 kg/m3 (0.1 to over 60 Ib'/ft.3). They can be rigid, semirigid, or flexible, colored or plain, and the foam can be virtually any TP or TS. The range of properties they offer in terms of their insulating value, rigidity, compressive strength, cushioning and loading, structural characteristics, and others can be very extensive (see Tables 6-1S through 6-19). Their performance depends to a great extent on the type of base plastic used, the type of blowing system, and the method of processing. Each plastic can include fillers or reinforcements to provide certain desirable properties. There are many ways in which foams can be processed and used: as slabs, blocks, boards, sheets, molded shapes, sprayed coatings, in extruded profiles, and "foamed in place" in an existing cavity, in which process the liquid material is poured and allowed to foam. Conventional equipment such as extruders, injection molding machines (see Fig. 6-29), and compressors is used (see Chapter 7) [1, 10-12,40-42, S03, S13, SI4]. The foaming methods vary widely. One is to whip air into suspension or a solution of the plastic, which is then hardened by heat curing. A second is to dissolve a gas in a mix, then expand it when the pressure is reduced. Another is to let a liquid component of a mix be volatilized by heat. Similarly, water produced in an exothermic chemical reaction can be volatilized within the mass by the heat of reaction. A different technique lets carbon-dioxide gas be produced within the mass by chemical reaction. A related way is for a gas such as nitrogen to be liberated within the mass by the thermal decomposition of a chemical blowing agent [12]. Finally, tiny beads of resin or even glass microballoons can be put into a plastic mix or syntactic foam or the like. In syntactic foams, instead of employing a blowing agent to form bubbles in the mass, preformed bubbles of glass, ceramics, or plastic are embedded in a matrix of an unblown resin. In multifoams such preformed bubbles are combined with a foamed resin to provide both kinds of cells. Reducing weight is one obvious objective, but this change may be accompanied by other properties. A mixture of microspheres and the resin can be formulated into a moldable mass that can then be shaped or pressed into cavities and molds much as can molding sand and clay. The properties of the finished hardened or cured mass can then be tailored by a suitable formulation. Synthetic wood, for example, is created by a mixture of polyester resin and small hollow glass spheres. Epoxy foams are commonly syntactic (see Table 6-19).

Table 6-14. Film and Sheeting Property Chart

Specific Gravity

Generic Name ASTM No.

Sq. In. Per Lb. One Mil Thickness

Thickness Range

Manufacturing Method

0-1505

I. Acrylic 2. Acrylonirrile butadiene styrene

1.14-1.26 1.01-1.30

22.000-24.000 25.600-27.400

0.002-0.006 0.0075-2.000

3. Acrylonirrile-styrene copolymer 4. Cellophane

1.08-1.09 1.40-1.50

25.400 II .600-25.000

0.001-0.020 0.0008-0.0017

extrusion calendering. extrusion. press laminated biaxially oriented extrusion

5. Cellulose acetate

1.28-1.43

21.000-22.000

0.0005-0.250

casting. extrusion

6. Cellulose acetate butyrate

1.19-1.22

23.800

0.0011-0.150

casting. extrusion

1.20 1.28-1.31 1.15 1.66-1.68

23.000 21.200-22.000 23.800 16.600

0.003-0.250 0.0005-0.200 0.002-0.015 0.005-0.090

extrusion casting extrusion extrusion

7. 8. 9. 10.

Cellulose propionate Cellulose rriacetate Ethyl cellulose Ethylene chlororrilluoroethylene copolymer (E-CfFE)

II. Ethylene methacrylate 12. Ethylene-tetralluoroethylene copolymer (ETFE) 13. Ethylene-vinylacetate copolymer 14. Fluorinated ethylene propylene copolymer 15. Ionomer

0.942 2.15

29.280 16.300

0.04-0.250

0.924-0.920 2.15

29.000 12.900

0.00075 and up 0.0005-0.060

extrusion extrusion

0.94

29.400

0.001-0.01

extrusion

16. Polyamide (nylon 6/6)

1.14

24.350-25.700

0.0005-0.030

extrusion

17. Polyamide (nylon 6) (biaxially oriented) 18. Polyamide (nylon 11 and 12)

1.16

23.870

0.0005-0.001

extrusion. orientation

1.03

26.630-27.200

0.0005-0.030

extrusion

19. Polycarbonate

1.20

23.000-23.100

0.001-0.250

extrusion

20. Polyester 21. Polyether sulfone

1.380-1.399 20.000-21.000 0.00015-0.014 1.37

20.000

0.001-0.250

22. Polyethylene (low density)

0.910-0.925

30.000

0.0004-0.250

23. Polyethylene (medium density)

0.926-0.940

29.500

0.001-0.250

24. Polyethylene (high density)

0.941-0.965

29.000

0.0004-1.000

476

extrusion. blown. casting extrusion

biaxially oriented. extrusion casting. extrusion calendering. casting. extrusion. biaxially oriented calendering. casting. extrusion. biaxially oriented calendering. extrusion. stress relieving

Maximum-

Heat

minimum Use Clarity

Shelf Life

excellent

indefinite

poor

good

excellent good

good indefinite

excellent

FIammability

Temperature Range of

U.V. Resistance

CS-192

0-759

D-1435

slow burning slow burning to selfextinguishing slow burning

-65 to 215

excellent

-65 to 215

poor to

-80 to 185

good fair

slow burning

o to 350

indefinite

slow burning

good to excellent

indefinite

Sealing Range OP

yes

good

250-300

-IS to 200

good

350-450

slow burning

-30 to 180

good

good

indefinite indefinite

self-extinguishing self-extinguishing

-30 to 200

excellent excellent good

indefinite indefinite

slow burning self-extinguishing

fair

indefinite

slow burning

good

indefinite

self-extinguishing

good

indefinite

burns nonflammable

good

indefinite

o to 400

good

350-450

fair

nonsealing

-75 to 250 -80 to 400

good

nonseaIing 475-500

-80 to 150

fair good

175-350

120 to 180 -415 to 400

fair good

150-300 575-700

good

525

fair to

indefinite

self-extinguishing

-100 to 300

poor to fair

400-450

good good

indefinite

slow burning

-100 to 300

good

400-450

good excellent

indefinite excellent

slow burning self-extinguishing

-150 to 275

fair to good

320-390 350-440

excellent

excellent

-75 to 300

fair to good

325-450

fair

indefinite

poor

500-550

good to excellent excellent

indefinite

slow burning to selfextinguishing slow burning to selfextinguishing slow burning

-70 to 180

fair

230-400

indefinite

slow burning

-70 to 200

fair

250-400

good

indefinite

slow burning

-70 to 250

fair

260-425

-100 to 400-450

(cont'd)

477

Table 6-14. (Continued)

Dimensional Generic Name

Change at High RH%

Water Vapor

Water

Permeability Gas Permeability ccll 00 sq. in. 24 hr.lmil

Gml24 hr.llOO

Absorption 24 hr.

sq. in.lmil@

Immersion

Resistance to

Resistance to

(0" N" CO" SO,)

95° F. 90% RH

%

Acids

Alkalies

D-1434

E-96 (B)

D-570

D-543

D-543

8 (3 mil)

1.4-16

good

0.2-0.3

fair to good

good

good poor

excellent poor

ASTM No. I. 2.

CO, 210, N,9, 0, 36

3. 4.

2.5

CO, 0.4-0.6, H, 1.2-2.2 N, 0.5-0.6, 0, 0.5-0.8

5.

.5

0, 117, N, 40, CO, 1000 CO, 6,000, N, 250, 0,

6. 7.

0.5 0.5

9.

12.0-14.0 0.4-1.34

0.5 45-115

100-200

4-9

poor to good

poor

4.5 - 6.5

0.1-3.4

poor to good

poor

poor

poor

730'

950 8.

High CO, 880, N, IS, 0, 30, 0 CO, 7,800, N, 100

100-200

1.6

50-150 4.8-14.2

2.0-4.5 2.5-7.5

good to poor

poor

fair to poor

excellent

good

good

fair to good

good

good fair

good fair

0.01

excellent

excellent

0.4

good

good

10.

2.5

0,3,500 CO, 110, N, 10, 0, 25

0.6

<0.02

11.

nil

0,335, CO, 4,780

4.1

nil

0.04 2-3

<0.02

12.

CO, 250, N, 30, 0, 10 CO, 6,000, N, 400, 0,

13. 14.

none

good

840 0,3,250, N, 1,500, CO,

<0.01

3,150 0,600

15. 16.

3.4

CO, 160, 0, 25 (50% RH) 0, 1.2-1.4 CO, 153-336, N, 3.4-18,

19.0

9.5

poor

excellent

17.

3.4

10-11 4.0

0.27-0.25

poor poor

good

none

0, 34-92, H, 323 CO, 1,075,0,300, N, 50

11.0

excellent

fair

X I~

CO, 16, 0, 6, N, I

fair to excellent

fair to good

2.1

good

good

22.

none

CO, 2,900, 0, 550, N,

1.2-1.5

0-0/8

excellent

excellent

23.

none

180 CO, 990, 0, 280

0.7-0.9

nil

excellent

excellent

24.

none

CO, 6,200-3,300, 0,

0.3-0.4

nil

excellent

excellent

18. 19. 20.

II

CO, 950, N, 40, 0, 230

21.

800-450, N, 100-50

478

1.7

0.15 <0.8

fair

Resistance to Greases and Oils

Resistance to Organic Solvents

D-722

D-546

good

Resistance to Water and/or Detergents

%

Tearing Strength (Elmendorf GM/mil)

Folding Endurance

D-882

D-882

D-1004

D-2172 6,000+

Tensile Strength Ibs. sq. in @RT

Elongation

good

4,400

130

25

excellent good excellent good

paramatrics, (aliphatics) fair to excellent excellent to good poor poor

good good

1,700-7,800 10,000-15,000 7,000-18,000 8,000-16,400

20-60 10-50 25-35

2-20 2-25

500 250-2,000

good

poor

excellent

4,100-9,700

40-100

3-10

1,200-1,500

good good

poor fair

excellent

5,000 9,000-16,000

80 10-40

25 4-10

1,000-2,000 3,500-4,000

good

poor

8,000-10,000

20-30

7-36

250,000

good

good

good

8,000-10,000

150-250

900-1,300

250,000

poor

poor

excellent

1,500-1,800

600-1,000

good poor

good fair

good good

7,000-8,000 1,000-3,500

300 400-800

60MD 150 TO 600-900

excellent

excellent

excellent

3,000

300

125

4,000

good

good

good

5,000

250-450

30

very high

excellent

excellent

good

9,000-12,000

300-400

50-90

excellent

good

good to poor

32,600-35,000

100-120

good

good

good to poor

7,000-12,000

250-400

good

poor

fair

8,000-10,500

85-105

20-25

250-400

excellent

good to excellent

excellent

25,000-40,000

50-120

12-40

230,000

good poor

good to poor good (6O"C)

good

10 ,000-12 ,000 up to 12,000

20-150 200-800

7-16 50-300

good good

good good (80"C)

fair

up to 17,000 2,400-6,100

50-650 10-650

50-300 15-300

6,000+

very high

>250,000

good

very high very high very high

(cont'd)

479

Table 6-14. (Continued) ~ielectric

Dissipation Factor

Constant

Dielectric Strength,

Resistivity

Volume

Generic Name

103 cps

10" cps

10" cps

103 cps

10" cps

10" cps

V/mil

ohm-em

ASTM No.

0-150

0-150

0-150

0-150

0-150

0-150

0-149

0-257

4.4 2.5-3.0

3.5 2.5-3.0

0.046

0.040

2,400 350-400

10 1" 10 1"

2,000-2,000

lO"

I.

2. 3. 4.

3.2

O.oJ5

(uncoated)

(uncoated)

(uncoated)

5.

3.6

3.2

3.2

0.016

0.33

3,2()(}"5,000

IO lO_IO"

6. 7. 8. 9. 10.

2.9

2.5 3.2 3.3-3.8 3.0 2.6

2.8

0.013

3,400

1O"-10"

0.016 0.002-0.20 0.002

0.03-0.04 O.oJ5 0.033 0.010-0.06 0.013

3,700 1,500 (10 mil) 5,000-6,000

1013-10" 10"

0.008 0.01-0.02 <0.002

0.15 O.oJ5 O.oJ-O.Q2 0.0004

11. 12. 13. 14. 15.

3.2-4.5 3.1 2.6

3.1 2.6 2.6 2.7-2.9 2.7-2.9 2.0-2.50 2.0-2.05

16. 17. 18. 19.

4.0

3.4

3.8 2.99

3.1 2.93

20. 21. 22. 23. 24.

3.2 3.5 2.2 2.2 2.3

3.0 3.5 2.2 2.2 2.3

480

3.2

2.05

0.046

0.0015

3()(}"5oo 3,500 5,000 7,000

850

2.89 2.8 2.2 2.2 2.3

0.05 0.0015

0.03 0.010

0.005 0.0035 0.0003 0.0003 0.0005

0.016 0.006 0.0003 0.0003 0.0005

0.012 0.003-0.008 0.0003 0.0003 0.0005

1,800 1,500 7,500 2,400 (4 mil) 470 500 500

101" lOIS

10 1" 10" 10 1" 4.5 x 1013 1013 101" lOIS

1O I7 _IO IS 101" 10 1" 10 1•

Specific Gravity

Sq. In. Per Lb. One Mil Thickness

1.42 1.18-1.19

19,400 23,400

0.0005-0.005 0.005-5.000

27. Polymethyl pentene 28. Polyparabanic acid 29. Polypropylene, cast

0.83 1.3 0.9

33,360 20,800 30,700

0.003-0.030 0.001-0.005 0.0003-1.000

30. Polypropylene, balanced, oriented 31. Polypropylene, balanced, oriented, coated 32. Polystyrene (casting, extrusion)

0.91 0.94

30,800-30,600 29,500

0.0005-0.0015 0.0009-0.0011

orientation orientation

1.05-1.06

26,100

0.0004 and up

33. Polystyrene (oriented)

1.05-1.06

26,100-26,300

0.0003-0.020

34. Polysulfone

1.24-1.25

22,355

0.0001-0.250

casting, extrusion oriented, biaxially oriented extrusion

35. Polytetraftuoroethylene (Tefton

2.14-2.17

12,700-13,000

Generic Name 25. Polymimide 26. Polymethyl methacrylate

Thickness Range

up to 0.250

TFE)

36. Poly(perftuoroalkoxytetraftuorethylene) (PFA) 37. Polytriftuorochloroethylene (Kel-

2.13-2.16 2.1-2.2

13,000

0.0005-0.020

38. Polyurethane

1.11-1.24

23,000

0.0005 and up

39. Polyvinyl alcohol

1.21-1.31

21,600

0.0005-0.012

40. Polyvinyl chloride (PVC) (rigid)

1.35-1.48

20,000-23,000

0.0001-2.000

41. Polyvinyl chloride (PVC) (ftexible)

1.24-1.45

20,000-24,000

0.0005-0.375

42. Polyvinyl ftuoride (Tedlar) 43. Polyvinylidene ftuoride (Kynar) 44. Rubber hydrochloride

1.38-1.57 1.75-1.78 1.11

20,000 16,000 25,000

0.0005-0.002 0.002-0.020 0.0004-0.0025

45. Vinyl chloride-acetate copolymers (rigid)

1.30-1.39

21,000

0.001-0.030

46. Vinyl chloride-acetate copolymers (ftexible)

1.20-1.35

21,000

0.001-0.030

47. Vinylidene chloride-vinyl chloride copolymer

1.57-1.68

16,300-23,000

0.004-0.010

48. Vinylnitrile rubber

1.18-1.21

22,800-23,700

0.001-0.020

F)

Manufacturing Method casting extrusion, biaxially oriented extrusion casting casting, extrusion

skiving, casting, extrusion casting, extrusion, molding casting, extrusion calendering, casting extrusion casting, extrusion calendering, press, casting, extrusion calendering, press, casting, extrusion extrusion extrusion casting calendering, extrusion, casting calendering, extrusion, casting calendering, biaxially oriented, extrusion, casting casting, extrusion, calendering (cont'd)

481

Table 6-14. (Continued) Maximumminimum Use Generic Name

Shelf Clarity

Life

Flammability

25.

good

excellent

self-extinguishing

26.

excellent

excellent

slow burning

Temperature Range of

-450 to 600 -

to

190

U.V. Resistance good good to excellent

27. 28. 29.

fair

indefinite

bums

good

indefinite

slow burning

excellent

indefinite

slow burning

30. 31. 32. 33.

excellent

indefinite

slow burning

excellent

indefinite

slow burning

excellent

indefinite

34.

good

indefinite

35.

poor

indefinite

36. 37.

good good

38.

good

slow burning

excellent

slow burning

poor

-320 to 500 o to 275 -60 to -60 to -55 to -80 to

240 225 200 175

fair fair fair fair

-100 to 350

poor

nonflammable

-450 to 500

excellent

indefinite

self-extinguishing nonflammable

-450 to 500 -423 to 300

excellent

indefinite indefinite

slow burning

-100 to 190

excellent poor to good

39.

slow burning

-40 to 420

good

excellent

slow burning to

-50 to 200

fair to good

indefinite

self-extinguishing slow burning to self-extinguishing

-50 to 200

poor to good

excellent

40.

good to

41.

excellent good to excellent

42.

good

indefinite

slow burning to

-100 to 225

excellent

self-extinguishing

43. 44.

fair good

45.

excellent

indefinite

self-extinguishing

poor

self-extinguishing

-80 to 275 -20 to 205

slow burning to

-50 to 200

self-extinguishing

46.

excellent

slow burning to self-extinguishing

47. 48.

482

excellent

indefinite

self-extinguishing slow burning

good poor poor to good

-60 to 200

o to 200 (dry) 300 (wet) 32 to 200

poor to good good

Heat Sealing Range of does not heat seal 375-450

Dimensional Change at High RH% 22 x 10-"

Gas Permeability cc/lOO sq. in. 24 hr.lmil (0" N" CO" SO,) CO, 450, 0, 25, N, 6

0.35

Water Vapor Permeability Gml24 hr.llOO sq. in.lmil @ 95 0 F. 90% RH

Water Absorption 24 hr. Immersion %

5.4

2.9 negligible

good

0.005 2.8 0.005 or less

good good excellent

nil nil

excellent excellent

0.04-0.06

good good

none

CO, 800, 0, 240, N, 60

0.7

narrow 200-275

none none

CO, 540, 0, 160, N, 15 CO, 520, 0, 130

280-330

none

AIR 105, CO, 1,080, H,

0.25-0.45 0.40 0.5-0.7, 6.0 7.0-10.0

2,420, N, 49.5,0,310 CO, 405, N, 40, 0, 90

18

He 600-1,150, 0, 1.3-5.8, N, 9.3-19, CO, 11.7-225 CO, 450-1650, N,41-119, 0,75-327

275-375

good to excellent

0.5

285-400 250-350 260-375

does not heat seal 600-700 325-375

Resistance to Acids

0.3

good

<0.01

excellent

0.025

0.03 nil

excellent excellent

40-75

0.55-0.77

poor to good

240-380

none

He 0.2, H, nil, CO, nil, 0" nil (0% RH)

10.0 est

55

340-400

negligible

CO, 970, 0, 150 (50% RH)

4.0

negligible

excellent

275-350

very slight

CO, 5,000, 0, 500 (50% RH)

negligible

400-425

none

CO, 15, 0, 3, N, 0.25, H, 58.1 CO, 5.5, N, 9, O 2 14 CO, 6--13,000, 0, 3.2-3.250

375 225-250 300-350

none

280-350

none

230-315

none

CO, 970, 0, 150 (50% RH) CO, 970, 0, 150 (50% RH) CO, 3.8-150, 0, 8-24, NE, 12-10 low

0.8-20.0

poor

5.2

0.5

good to excellent excellent

2.6 0.008-0.03 (MP)

0.04 5

excellent good

4.4

negligible

0.28

negligible

very good to excellent excellent

0.02-0.6

negligible

excellent

negligible

good

(cont'd)

483

Table 6-14. (Continued)

Generic Name

Resistance to Alkalies

Resistance to Resistance to Greases and Oils Organic Solvents

Resistance to Water and/or Detergents

25. 26. 27. 28. 29.

poor good good poor excellent

excellent good fair good good

excellent fair fair good good

30. 31. 32. 33. 34.

excellent coating attacked excellen! excellent good

good to excellent excellent good good good

good good poor to excellent poor to excellent poor

good good

35. 36. 37. 38. 39.

excellent excellent excellent poor to good poor

excellent excellent excellent good excellent

excellent excellent excellent good excellent

excellent excellent excellent fair to good excellent

40.

42. 43. 44.

excellent good to excellent excellent fair to poor good

poor to excellent excellent good good to excellent poor to good good to very good excellent excellent excellent fair to good good excellent excellent good

45.

excellent

excellent

46. 47.

excellent good to excellent good

41.

48.

484

good good good good

Tensile Strength Ibs. sq. in. @RT

Tearing Strength Elongation (Elmendorf % GMimil)

25,000 8,000-4,000 2,700-4,000 16,000 3,200-10,000

70 4-12 10 10 300-500

25,000-30,000 30,000 7,100-12,100 7,100-12,100 8,400-10,600

45-85 70 3-40 3-40 64-110

1,500-4,000 4,000-4,500 6,300-10,000 5,000-12,000 6,500-12,000

100-350 300-480 50-400 200-700 200-600

8

500 5-1,800 5-7 8 25 9-12 10-100 8-26 220-710 250-800

7,000-14,000 10-175 1,400-17,000 100-510

10-700 60-1,400

7,000-19,000 110-260 5,200-7,400 25-500 3,500-5,000 stretches

12-48 60-1,600

poor to excellent excellent depends on plasticizer fair to excellent poor to excellent very good good to excellent good to excellent excellent

2,000-4,500 250-300 8,000-20,000 35-110

30-1,400 10-7,100

excellent

2,500-4,000

200-960

good

3,000-8,000

depends on 10-30 plasticizer

Dielectric Constant

Dissipation Factor

Folding Endurance

103 cps

10" cps

109 cps

10' cps

Hl" cps

109 cps

3.5

3.4 3.0-3.5

0.003

3.5--4.0

2.58

0.040

0.010 0.030

0.009

25,000

2.1 3.4

2.1 3.4

very high

2.2

2.2

2.2

0.004 0.0003

very high

2.2

2.2

2.2

very high

2.2 2.4-2.7

2.2 2.4-2.7

2.2 2.4-2.7

2.4-2.7

2.4-2.7

3.05

10,000

very high

Dielectric

Volume

Strength,

Resistivity

V/mil

ohm-em

7,000

1018 1015

400

0.003 5,700 0.0003

0.0003

<0.0002

<0.0002

<0.0002

7,000-10,000

<0.0002 0.0005

<0.0002

<0.0002

0.0005

0.0005

7,000-10,000 5,000

2.4-2.7

0.0005

0.0005

0.0005

5,000

3.05

3.00

0.0008

0.0034

0.0041

7,500

2.0-2.1

2.0-2.1

2.0-2.1

0.0002

=2,500 (I mil)

1018

2.06 2.5-2.7

2.06 2.3-2.4

2.06 2.3

0.00115 0.004

2,000

1018 1018

5.2-7.5

5.5-7.1

0.0002 0.000035 0.022-0.24

0.0002 0.000080 0.009-0.017

0.040-0.060

3,000--4,500

10 1" 5 x 101" 3 x 10 15

x x 107 x 107 x 5 x 3

1018

3

1018

1,000-3,700 600-1,300

1-3.0

10 1" 10 16 1016

x

10"

very high very high

3.0-3.3

2.8-3.1

very high

4.0-8.0

3.3--4.5

7.46 3.0

6.1 3.0

2.98

0.019

250,000

0.059 0.006

very high

3.0-3.3

2.8-3.1

2.8

0.0009-0.017

0.006-0.019

0.019

425-1,300

0.070-0.160 0.052-0.063

0.04-0.140 0.050-0.080

0.016

250-1,000 3,000-5,000

2.8

0.009-0.017

0.006-0.017

0.07-0.016

0.04-0.14

0.019

425-1,300 250-1,000

1016 10"_10 14

260-1,280

2

excellent

>500,000

4.0-8.0

3.3--4.5

3.9--4.5

3.0--4.0

2.7

0.110

x 1014 1013 10 16

10"_1014 10 12_1015

very high

485

CD CI'I

.

D 2326

D 696

Coefficient of linear expansion, 1~ in.lin.-"F

D 1621

2.0

(0.29) 40--60

275 (135) 1.0

(0.14) 100

(0.29-0.032) 5

50--60 2-4 (800-960) (32-64) 1,000 1000 (6.89) (6.89) 8,00013,000 (55.1-89.6) 19-21

Syntactic Castable

Continuous at 300 225 (149) 0.20-0.22

(0.15-0.59)

22-85

D 1623

Foamed in Place 2-5 (32-80) 20-54

Test

Thermal conductivity BTU/in.Ihr._ft.2"F (W/mK)

Density, Ib.lft. 3 (kglm3) Tensile strength, psi (MPa) Compression strength at 10% deflection, psi (MPa) Impact strength, ft.-Ib.lin. Maximum service temperature dry, OF ("C)

Property

ASTM

Phenolics

38

25

270 (132)

(51. 7) 45

(37.9) 18

200 (93.3)

50 (800) 5,500 (37.9) 7,500

50 (800) 3,300 (22.7) 5,500

2.0 (32) 42-68 (0.29-0.47) 25-40

Molded

(0.046-0.049)

180-200 (82-93) 0.32-0.34

(0.033) 30-40

165-175 (74-79) 0.23

200-250 (93-121) 0.15-0.21

(0.48-1.90)

4-8 (64-130) 90-290 (0.62-2.00) 70-275

Polyurethane Rigid Closed Cell

(0.024-0.030) (0.022-0.030) 40 30-40

165-175 (74-79) 0.17-0.21

2-5 (32-80) 180-200 (1. 24-1.38) 100-180 at 5% (0.69-1.24)

Extruded

Polystyrene

(0.014-0.12) (0.17-0.28) 0.21

5.5-7.0 (88-112) 110-210 (0.76-1.45) 2-18

Polyvinyl Chloride Phenylene Rigid Oxide Polyethylene MediumClosed Foamable Cell Resin Polycarbonate Density Foam

Table 6-15. Properties of a Few Rigid Plastic Foams

Table 6-16. Properties of a Few Relatively Flexible Foams

Material Polyphenylene oxide Polycarbonate Epoxy resin Isocyanurate Polyether Polystyrene Polystyrene Polyurethane ABS Acetal Nylon 6/6 Polybutylene terephthalate Polyimide Polysulfone Polyvinylchloride

Elongation at Break

Specific Gravity 0.8 0.8 0.78 0.032 0.08 0.17 1.04 0.11 0.86-1.1 1.130 0.97 1.1

(%)

Heat Deflection Temperature °C (OF)

15.0 4.0

96 (205) 128 (262)

3.5

101 (214)

Thennal Conductivity W/mK (BTU-in.lhr. ft.2 . oF) 0.124 0.151 0.7 0.1

(0.860) (1.05) (4.8) (0.69)

0.65 (4.5) 0.3

4.1

1.3

0.87 0.6

3.5 370.0

72 153 255 207 277 177

(2.1)

(162) (307) (491) (405) (531) (351)

Table 6-17. Thermal Properties of a Few Plastic Foams Compared to Wood Property

Polystyrene

Urethane

Polyethylene

Density, Ib./cu. ft. (kg/m3) Insulation Value (K factor, BTU-in.lhr. OF ft.2) (W/m' K) Linear coefficient of thennal expansion, in.lin. OF Maximum temperature for continuous use, OF ("C) Heat of Combustion BTU/lb. (MJ/kg) BTU/cu. ft. BTUlboard ft. Ignition temperature (ASTM D 1929-62T) Flash ignition temp., OF ("C) Self-ignition temp., OF ("C) Surface flame spread (ASTM E 84-61 "Tunnel Test")

1.0-3.0 (16-48)

1.5-2.5 (24-40)

2.0 (32)

0.24-0.30 (0.030-0.043) 4 x 10-5

0.14-0.16 (0.020-0.023) 5 X 10-5

0.35 (0.050) 8 X l(J-5

170---180 (77-82) 16,000 (37.18) 32,000 2,660

650---700 (343-371) 735-915 (391-490) 10---25

250 (121)

40---80

0.7

160 (71)

11,000 (25.56) 16,000 (37.18) 22,000 32,000 1,840 2,660

600 (316) 975 (524)

Wood (red oak)

8,000 (18.59) 320,000 26,600

650 (343) 660 (349)

500 (260) 500 (260)

NonFR

100

487

488 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 6-18. Plastic Foams Reinforced with Glass Fibers Material Property Percent glass, by wt. Specific gravity Tensile strength, psi (MPa) Aexural modulus, 106 psi* Aexura1 strength, psi* Izod impact strength, ft.-Ib'/in. Unnotched impact strength, ft.-Ib'/in. Deflection temperature under load @ 264 psi, OF

eC)

PA

PS

SAN

PC

PC

PP

ABS

30 0.99 15,500 (107) 0.114 24,700 1.2

20 0.84 5,000 (34.5) 0.Q75 8,500 0.6

20 0.85 6,100 (42.1) 0.Q78 9,500 0.6

20 0.94 14,000 (96.53) 0.058 14,200 2.1

40 1.06 19,300 (133) 0.100 17,800 1.8

20 0.67 4,000 (27.6) 0.042 6,000 1.4

20 0.84 7,000 (48.2) 0.075 12,000

6.2

1.5

2.1

7.1

6.1

3.5

3.5

380

190

202

276

272

162

210

(193)

(88)

(94.4)

(135)

(133)

(72.2)

(98.9)

'To convert psi to pascals (Pa). multiply by 6.895 x UP.

Table 6-19. Syntactic Plastic Foams as are Used in Deepwater Flotation Materials

Foam Density Ib./ft. 3 24 27 30 32 34 35 38 42

Glass Microballoons Yes Yes Yes Yes Binary Yes Yes Yes

Epoxy Macroballoons Yes Yes Yes No No No No No

Uniaxial Compressive Yield Strength, psi* 1,700 2,800 2,500 7,300 10,500 10,000 10,000

Hydrostatic Compressive Strength, psi*

Method of Preparation

Resin System

1,800 2,300 3,000 7,000 16,000 14,000 14,000 15,000

Pack-in-place Pack-in-place Infiltration Pack-in-place Infiltration Vacuum cast Cast Cast

Polyester Epoxy Polyester Polyester Epoxy Epoxy Polyester Polyester

'To convert psi to pascals (Pa). multiply by 6.895 x UP.

Cushioning Design When plastic foams are used to cushion products, as they are in packaging, there are specific design approaches to use. It is widely held that the lower density closed-cell foams that are usually priced lower provide superior cushioning performance, but this assumption is usually incorrect. Figure 6-30 illustrates how closed-cell PE foams with differing densities but the same type and thickness behave under the same dynamic cushioning conditions. These curves represent the amount of mechanical shock experienced during an impact. The lower the curve goes, the greater the cushioning efficiency. For densities above 2 pcf ObJft. 3 ) the maximum cushioning efficiency of each material is not significantly different, but the loading at which this maximum efficiency will occur will vary dramatically. If a 40-g package were to be designed according to Figure 6-30 using a 6.0 pef foam, the foam would measure 3 in. thick at a loading of about 1.35 psi. If an identical package were then produced using a 2.2 pcf foam, its shock performance would not go as low as 40 g's but would instead produce about 60 g's, or 50 percent more shock. In order

THE PROPERTIES OF PLASTICS 489

Figure 6-29. A collapsible HDPE (Petrothene L Y 656, from Quantum Chemical) foam shipping container injected molded by the J.I.T. Corp. of Southfield, Michigan. This is 114 cm X 122 cm x 86 cm (45 in. x 48 in. x 34 in.) high. It holds and delivers .85 m3 (30 cu. ft.) of materials, but on the return it takes only .28 m3 (10 cu . ft.), because of its plastic mechanical hinges and latches . Stacked three high, they can hold 1.8 metric tons (4,000 lbs.) in a shipping mode.

to return to 40 g's, the 2.2 pcf package would need to be redesigned. One approach would be to greatly increase the thickness of the pads constructed from the lower-density foam, to provide adequate protection. This approach would, however, increase the package size, impair handling and shipping efficiency, and possibly result in higher costs. The 6.0 pef foam could, however, be reliably used at 1.2 psi in the thickness shown in Figure 6-30. Another approach is to keep the 2.2 pef foam thickness the same, but decrease the loading from 1.35 to 0.87 psi, to get back to the 4O-g level. Although this approach keeps the package size the same, nearly twice as much foam must be used to meet the lower loading. The lower-density foam must therefore cost less than half as much as the high density type on a cost-per-unit volume basis if using the lower-density one is to result in a cost savings. Below a density of about 2.2 pcf the cushioning efficiency can begin to change with the density . This situation is shown in Figure 6-31 where the test results for PE foams in several densities below 3 pef are compared. Thus, lowering the density produces a considerably higher deceleration and reduces cushioning performance. Also significant is the narrower range of usable static loadings at the bottoms of the curves that resulted when the density was reduced. Another important consideration in comparing foams of different densities is their compressive creep resistance (see Chapter 3), their ability to resist undergoing a permanent thickness loss during their time under load. As the density decreases, so does the creep resistance.

80 r----------.----------.---------~

..... 60 -'" .g c

g

!2

~

~

40 6 pcl

~

!

20

1.0 Static

Figure 6-30. How a closed-cell PEs foam density affects both its cushioning and its loading.

180

..... 9

150

•w

~ 120

e I

~

90

~

!

60

30 00

0.3

0.6

0.9

1.2

StatIC

Figure 6-31. A closed-cell PE foam at different densities compared to its cushioning efficiency. 490

THE PROPERTIES OF PLASTICS 491

Although it may seem logical for a lower-density foam to cost less to produce because it contains less plastic, this is not necessarily true. The manufacturing rate, the amount and cost of the blowing agent, and the amount and cost of the base resin all influence the final cost. As a result, very low density foams can actually be more costly to make than others. Thus, it should not be assumed that the cost of a foam will be proportional to its density. (Incidentally, this cost situation can also occur with solid plastics in certain shapes, thicknesses, and types, but in solids the problem is rare.) Cushioning performance is therefore not improved merely by increasing a foam's density. In order to be certain that a material selected is appropriate and efficient, the designer should carefully compare documented performance data.

TRANSPARENT AND OPTICAL PLASTICS The use of plastics in certain transparent or optical applications is marked by selective but significant advantages of plastics over glass. Plastics weigh less and in many cases cost less, yet provide higher performance, such as impact strength (see Figs. 6-32 and 6-33). They also have many more configuration possibilities to simplify assembly. There are expensive plastics with added performance features related to chemical resistance, heat resistance, high tensile and flexural strengths and others that are used in specialty parts. The factor of configuration flexibility is particularly useful in systems that use aspheric or curved surfaces to simplify design and reducing the part count, weight, and cost. Moreover, the light-transmission abilities of plastic optics are comparable to those of high-grade crown glass. And from a safety standpoint, when plastics break they do not splinter, like glass, and are thus less hazardous [11, 12, 14, 62-68, 515, 516].

Figure 6-32. This streamlined headlamp by Ford and GE provides contoured, flush-mounted polycarbonate lenses that reduce air drag and some degree of weight.

492 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 6-33. Complex thermoformed acrylic and PC aircraft canopies being inspected to ensure their having relatively no visual distortion for the pilot. The grid placed on the wall lets the canopy be checked against it to determine if there is any distortion.

Plastic's chief disadvantages are its low scratch resistance and, in some systems, comparative intolerance to severe temperature fluctuations. Even if plastic does have less temperature tolerance than glass, most optical systems do not operate in ambient temperatures beyond the thermal limits of plastics. The processing advantages of plastics that exist, such as injection molding it with multicavity molds, allows low-cost manufacturing to be combined with comparatively inexpensive materials. By carefully sizing a mold for the required production volume, plastics' breakeven cost, compared to glass, can be low. Another advantage of plastics fabrication is that in the mounting and assembly features like brackets, holes, and flanges can be molded integrally with the optical element to result in a single-piece design eliminating mounting hardware and simplifying alignment. Multiple elements can thus easily be combined and molded in unique optical configurations.

THE PROPERTIES OF PLASTICS 493

REINFORCED PLASTICS AND COMPOSITES The terms reinforced plastics (RPs) and composites refer to combinations of plastic and reinforcing materials that usually come in fiber forms, as chopped, continuous, woven fabrics, porous mats, and so on (see Tables 6-20 through 6-26 and Fig. 6-34). Both TS and TP resins are used [417-27]. When the modem RP industry started, in 1940, glassfiber-reinforced unsaturated TS polyesters or low-pressure or contact-pressure curing resins principally were used. Now at least 80 percent by weight of composites are glass fiber and 60 percent are polyester (TS) types. A designer can now produce RP products whose mechanical properties in any direction will be both predictable and controllable. This is done by carefully selecting the resin and the reinforcement in terms of both their composition and their orientation, and following up with the appropriate process (see Chapter 7). All types of shapes can be produced: flat and complex, solid and tubular rods or pipes, molded shapes and housings and other complex configurations, such structural shapes as angles, channels, box and I-beams, and so on. The RPs can in fact produce the strongest materials in the world. The molder has a variety of alternatives to choose from regarding the kind, form, and amount of reinforcement to use (see Tables 6-24 and 6-25 and Figs. 6-35 through 6-41). With the many different types and forms (organics, inorganics, fibers, flakes, and more) available, practically any performance requirement can be met, molded into any shape. Possible shapes range from very small to extremely large, and from the simple to the extremely complex. The reinforcement type and form chosen (woven, braided, chopped, etc.) will depend on the performance requirements and the method of processing the RP. Fibers can be oriented in many different patterns to provide the directional properties desired. Depending on their packing arrangement, different reinforcement-to-resin ratios are obtained. In its simplest presentation, using glass fiber with resin, if the fibers were packed as closely as possible (like stacked pipe), the glass would occupy 90.6 percent of the volume (95.6 percent by weight). With a "square" packing (fibers directly on top of and alongside each other) the glass volume would be 78.5 percent (88.8 percent by weight). Glass fibers and most other reinforcements require special treatment to ensure maximum performance, such as selecting materials compatible with the resins used, protecting individual filaments during handling and processing, and so on.

Strength-to-weight comparison

"'o I

'" I

4

~

Stiffness-to-weight comparison

4

x

X

Ie:

~~

3

-6

2

o

E

...c

~

~

.~

eu

e

:;:::

.~

·u

1i

Vl

1

0 '--''----'---'-----''--'--'---'----''--'---'----

Carbon! epoxy

Glass! epoxy

Wood Aluminum Steel

·u '" ~

O'--C~ar-bo-n~!-'-G-las-s!~~W-OO-d~AI~um-i~nu~m-'-S-te~el~ epoxy

epoxy

Figure 6-34. A comparison of plastic composites with other materials.

494 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The acceptance and use of nonwoven fabrics as reinforcements has led to the development of major products. These reinforcements include felts and paper structures, which usually contain a binder that retains these structures and is compatible with the resin matrix. Combinations of different chopped fibers (glass and aramid, and so on) are also used, including long filaments, woven fabrics, and more. The combinations provide unique properties and, in most cases, permit the molding of different shapes that otherwise would not be possible. The longer fibers are best for optimizing mechanical properties. With short, chopped fiber structures, the fiber length can range from extremely short (0.001 in.) to at least 0.5 in., and on up to 2 in. The length used usually depends on the processing and performance requirements. Basically, to obtain the best mechanical performance with fibers in a properly molded part, it is necessary only for them to have an aspect ratio (length over diameter) of about ten. Practically all TS and TP resins are used in RPs, but a few predominate (see Tables 6-21 through 6-26), with TS polyesters being the major type. The properties typical of polyesters are summarized in Table 6-5. The polyester RPs are used in all processes, but their principal use is in the low-pressure methods (spray-up, hand lay-up, pressure bag molding, casting, pultrusion, rotational molding, filament winding, and compression molding (see Chapter 7). RPs can be processed in different ways. The individual components of reinforcement and resin can be put together by the processor. Although TP resins generally require no additional material, the TSs usually require the addition of different additives and fillers, such as those reviewed in Chapter 2. Most of the TPs used in the RPs are injection molded with compounds prepared by material suppliers. It is estimated that more than half the TSs are prepared by the processor. There are compounds available from material suppliers, as well as in some processing plants, that are ready to be processed, the most popular being the SMCs and BMCs. Sheet Molding Compound An SMC is a reinforced plastic compound in sheet form. Most SMCs combine glass fiber with a polyester (TS) resin. Any combination of reinforcement and resin can be produced (see Fig. 6-39). The reinforcements can have continuous long fibers or any size of chopped fibers laid out in a different orientation from that of the resin (see Fig. 6-40). The different orientation makes it feasible to use SMCs on flat to complex-shaped molds (see Fig. 641). These SMCs will contain various additives and fillers to provide a variety of processing and performance properties. SMCs are made to meet the shelf life required. These B-staged compounds are usually used in a few weeks or months. Some have a shelf life of six months, for example. Suppliers' recommendations should be followed in keeping these compounds at a low temperature, or a curing action will occur (see Chapter 7). TPs are also used in sheet form with different reinforcements and resins. They are called stampable sheets rather than SMCs. These compounds provide unique properties with a quick and easy processing capability. Bulk Molding Compound A BMC is a molding compound that is not produced in sheet form. It basically consists of the mixture used in an SMC, except that it contains only short fibers.

THE PROPERTIES OF PLASTICS 495

Basic Design Theory Fiber-reinforced plastics differ from many other materials because they combine two essentially different materials-fibers and a resin-into a single composite. In this way they are somewhat analogous to reinforced concrete, which combines concrete and steel. However, in the RPs the fibers are generally much more evenly distributed throughout the mass and the ratio of fibers to resin is much higher [1, 14,29,32,86]. In designing fibrous-reinforced plastics it is necessary to take into account the combined actions of the fiber and the resin. At times the combination can be considered homogeneous, but in most cases homogeneity cannot be assumed. Thus, it is necessary to allow for the fact that two widely dissimilar materials have been combined into a single unit. This section sets out the basic elements of design theory. In the basic design approach here, certain fundamental assumptions are made. The first, and most important, assumption is that the two materials act together and that the stretching, compression, and twisting of the fibers and resin under load is the same; that is, the strains in the fiber and resin are equal. This assumption implies that a good bond exists between the resin and the fiber to prevent slippage between them and wrinkling of the fiber. The second major assumption is that the material is elastic, meaning that the strains are directly proportional to the stresses applied and when the load is removed the deformation will disappear. In engineering terms the material is assumed to obey Hooke's Law (see Chapter 2). This assumption is probably a close approximation of the material's actual behavior in direct stress below its proportional limit, particularly in tension, if the fibers are stiff and elastic in the Hookean sense and carry essentially all the stress. This assumption is probably less valid in shear, where the resin carries a substantial portion of the stress. The resin may then undergo plastic flow, leading to creep or relaxation of the stresses, especially when the stresses are high. More or less implicit in the theory of materials of this type is the assumption that all the fibers are straight and unstressed or that the initial stresses in the individual fibers are essentially equal. In practice this is quite unlikely to be true. It is expected, therefore, that as the load is increased some fibers will reach their breaking points first. As they fail, their loads will be transferred to other as yet unbroken fibers, so that failure will be caused by the successive breaking of fibers rather than the simultaneous breaking of all of them. The effect is to reduce the material's overall strength and reduce its allowable working stresses accordingly, but the design theory is otherwise largely unaffected, as long as basically elastic behavior occurs. The development of higher working stresses is thus largely a question of devising fabrication techniques like filament winding to make fibers work together to obtain their maximum strength (see Chapter 7). In the following discussion of design theory the values of a number of elastic constants must be known in addition to the strength properties of the resin, the fibers, and their combination. In the examples used, more-or-Iess arbitrary values for the elastic constants and strength values have been chosen to illustrate the basic theory, but any other values could have been used just as well. The Theory of Combined Action Any material, when stressed, stretches or is otherwise deformed. If the resin and the fiber in reinforced plastics are firmly bonded together, the deformation will be the same in both. For efficient structural behavior high-strength fibers are employed, but these must be more unyielding than the resin. Therefore for a given deformation or strain a higher

Table 6-20. Mechanical, Thermal, and Processing Properties of Glass-FiberReinforced Plastics. Courtesy Owens-Corning Fiberglas Corp.

Material Glass-fiber-reinforced thennosets (RTS)

Glass-fiber-reinforced thennoplastics (RTP)

Unreinforced thennoplastics (TP)

Metals

496

Polyester SMC, compression Polyester SMC, compression Polyester SMC, compression Polyester BMC, compression Polyester BMC, injection Epoxy filament wound Polyester, pultruded Polyurethane, milled fibers (RRIM) Polyurethane, flaked glass (RRIM) Polyester spraying/lay-up Polyester, woven roving, lay-up Acetal resin Nylon 6/6 Polycarbonate Polypropylene Poly(phenylene sulfide) Acrylonitrile-butadiene-styrene terpolymer (ABS) Poly(phenylene oxide) (PPO) Styrene-acrylonitrile copolymer (SAN) Poly(butylene terephthalate) Poly(ethylene terephthalate) Acetal resin Nylon 6/6 Polycarbonate Polypropylene Poly(phenylene sulfide) Acrylonitrile-butadiene-styrene terpolymer (ABS) Poly(phenylene oxide) (PPO) Styrene-acrylonitrile (SAN) Poly(butylene terephthalate) Poly(ethylene terephthalate) ASTM A-606 HSLA steel, cold rolled SAE 1008 low carbon steel, cold rolled AlSI 304 stainless steel TA 2036 aluminum, wrought ASTM B 85 aluminum, die cast ASTM AZ 91B magnesium, die ASTM AG 40A zinc, die cast

Glass Fiber, Wt.%

Specific Gravity

30.0 20.0 50.0 22.0 22.0 80.0 55.0 13.0

D 792 1.85 1.78 2.00 1.82 1.82 2.08 1.69 1.07

23.0

1.17

30.0 50.0 25.0 30.0 10.0 20.0 40.0 20.0

1.37 1.64 1.61 1.48 1.26 1.04 1.64 1.22

20.0 20.0

1.21 1.22

30.0 30.0

1.52 1.56 1.41 1.13

1.20 0.89 1.30 1.03 1.10 1.05 1.31 1.34 7.75 7.86 8.03 2.74 2.82 1.83 6.59

Thermal Coefficient of Expansion D 696

9.4 6.6 6.6 2.0 5.0

Heat Deflection at 1.8 MPa, CO CF) D 648 200+ (392 +) 200+ (392+ ) 200+ (392+ ) (500) 260 (500) 260 200+ (392 +)

78.0

29

(84)

53.1 12.0 4.0 4.7 1.8 1.8 2.4 1.1

200+ 200+ 161 254 141 132 266

(392 +) (392 +) (322) (489) (286) (270) (511)

2.1 2.0

99 143

(210) (289)

2.1 1.4 1.7 4.7 4.5 3.7 3.8

102 213 216 110 75 132 46-60 135

(216) (415) (427) (230) (167) (270) (115-140) (275)

3.2

93-104 (199-219)

68.0 36.0 4.5

100 104 50-85 38-41

(212) (219) (122-185) (100-106)

Thermal Conductivity, (W/m'K)

Specific Heat, J/(kg .

(BTU' in.lhr.ft. 2 OF)

K)b

C 177

8.37 8.37 1.77 6.92

(58.1) (58.1) (12.3) (48.0)

(1,700) (1,700) (2,280) (1,750) (1,520) (4,000) (2,490)

(4,410) (12,500) (37,000) (18,600) (23,100) (12,000) (6,500) (22,000) (11,000)

6.9 15.5 8.6 8.3 5.2 3.7 14.1 6.2

(1,000) (2,250) (1,250) (1,200) (750) (540) (2,050) (900)

0.84-1.67 100

(14,500)

6.3

(910)

100 131 145 81 79 66 34 66 41

(14,500) (19,000) (21,000) (11,700) (11,400) (9,600) (4,900) (9,600) (5,900)

8.6 8.3 9.0 2.6 2.8 2.3 0.7 3.3 2.1

(1,250) (1,200) (1,300) (380)

54 66 57 59

(7,800) (9,600) (8,300) (8,600)

2.6 2.8 1.9 2.8

(380) (400) (280) (400)

(18.0)

1.30

2.60 7.97 14.5 3.47

(18.0) (55.3) (100.6) (24.1)

1.26 1.21

2.42 6.57

(16.8) (45.6)

4.84 12.1 11.2 2.80 2.94 2.34 2.10 2.89

(33.6) (84.0) (77.7) (19.4) (20.4) (16.2) (14.6) (20.1)

1.61

( 11.2)

1.05

0.46 1.46 1.26 1.26 1.88

(11.0) 0.84-1.67 1.59 (8.40) 1.21 1.38 1.76-2.89 (12.2-20.1) (10.5) 1.51 1.42

6.8

43.3

(300)

0.46

6.7 9.6 13.9 11.6 14.0 15.2

60.6 16.3 159 91.8 72.5 113

(420)

0.42 0.50 0.88

(113)

1.05 0.42

D 638 83 36.5 158 41.3 33.5 552 207

(12,000) (5,300) (22,900) (5,990) (4,860) (80,000) (30,000)

19.3

(2,800)

30.4 86.2 255 128 159 83 45 152 76

Tensile Modulus, GPa (kip/in. 2) D 638 11.7 11.7 15.7 12.1 10.5 27.6 17.2

1.26 1.26 1.26 1.26 1.26 0.96 1.17

2.60

(1100) (637) (503) (784)

Tensile Strength, MPa (psi)

(400)

(330) (100) (480) (300)

448

(65,000) 207

(30,000)

331

(48,000) 207

(30,000)

552 338 331 228 283

(80,000) 193 (49,000) 70 (48,000) 71 (33,000) 448 (41,000) 75

(28,000) (10,000) (10,300) (65,000) (11,000) (cont' Ii)

497

Table 6-20. (Continued) Elongation,

%

Material Glass fiber-reinforced thermosets (RTS)

Glass fiber-reinforced thermoplastics (RTP)

Unreinforced thermoplastics (TP)

Metals

498

Polyester SMC, compression Polyester SMC, compression Polyester SMC, compression Polyester BMC, compression Polyester BMC, injection Epoxy filament wound Polyester, pultruded Polyurethane, milled fibers (RRIM) Polyurethane, flaked glass (RRIM) Polyester spraying/lay-up Polyester, woven roving, lay-up Acetal resin Nylon-6/6 Polycarbonate Polypropylene Poly(phenylene sulfide) Acrylonitrile-butadiene-styrene terpolymer (ABS) Poly(phenylene oxide) (PPO) Styrene-acrylonitrile copolymer (SAN) Poly(butylene terephthalate) Poly( ethylene terephthalate) Acetal resin Nylon-6/6 Polycarbonate Polypropylene Poly(phenylene sulfide) Acrylonitrile-butadiene-styrene terpolymer (ABS) Poly(phenylene oxide) (PPO) Styrene-acrylonitrile (SAN) Poly(butylene terephthalate) Poly( ethylene terephthalate) ASTM A-606 HSLA steel, cold rolled SAE 1008 low carbon steel, cold rolled AISI 304 stainless steel TA 2036 aluminum, wrought ASTM B85 aluminum, die cast ASTM AZ91B magnesium, die ASTM A040A zinc, die cast

D 638 <1.0 0.4 1.7 0.5 0.5 1.6 140.0 38.9

1.3 1.6 3.0 1.9 9.0 3.0 3.0 2.0

Flexural Modulus OPa (kip/in2) D 790 11.0 9.7 13.8 10.9 9.9 34.5 11.0 0.26-0.37 1.0 5.2 15.5 7.6 5.5 4.1 3.6 13.1 6.0

(1,600) (1,400) (2,000) (1,580) (1,400) (5,000) (1,600) (38-54) (145) (7,500) (2,250) (1,100) (800) (600) (520) (1,900) (870)

5.0 1.8 4.0 6.6 30.0 60.0 110.0 200.0 1.0 5.0

5.2 7.6 8.1 8.6 2.7 2.9 2.3 0.9-1.4 3.8 2.4-2.8

(750) (1,000) (1,200) (1,250) (390) (420) (330) (130-200) (550) (350-400)

50.0 0.5 50.0 50.0 22.0 37.0

2.3-2.8 3.8 2.3-2.8 2.4-3.1

(330-400) (550) (330-400) (350-450)

40.0 23.0 2.5 3.0 10.0

Compressive Strength, MPa (psi)

Impact Strength Izod at 22°C, JIm

Hardness

D 695 166 (24,100) 159 (23,100) 221 (32,000) 138 (20,000)

D 256 854 438 1036 227 154 2400 1335

D 785 Barcol68 Barcol68 Barcol68 Barcol68 Barcol68 M98 Barcol50 Shore D 65-75

310 (45,000) 207 (30,000)

112 690-800 1760 96 117 107 59 80

152 186 117 183 97 172 145 97

(22,000) (27,000) (17,000) (26,500) (14,000) (24,900) (21,000) (14,000)

121 121 124 172 90 103 86 24 110 69 83

(17,500) (17,500) (18,000) (24,900) (13,000) (14,900) (12,500) (3,500) (16,000) (10,000) (12,000)

96 59 96 96 32 43 854 50-1000 <27 160-320 270

97 59 76 448 331 552

(14,000) (8,500) (11,000) (65,000) (48,000) (80,000)

16-24 43 13-35

338 331 227 283

(49,000) (48,000) (32,900) (41,000)

64

Barcol50 Barcol50 M79 M95 M80 R103 RI23 RI07

Water Absorption in 24 hr. % D 570 0.25 0.10 0.50 0.20 0.20 0.50 0.75

Mold Shrinkage, %

D 955 0.002 0.001 0.004 0.008

1.30 0.50 0.29 0.50 0.14 0.05 0.01 0.30

0.004 0.002 0.005 0.003 0.002 0.002

R107 RI22 RII8 RI20 RII9 RI20, M83 M70 R50-96 RI23 R 107-11 5 RII5

0.24 0.06 0.06 0.05 0.3-1.9 1.0-1.3 0.15 0.03 <.02 0.20-0.45 0.07

0.003 0.002 0.003 0.003 0.005 0.008 0.005-0.007 0.020 0.007 0.004-0.009 0.005-0.007

M80-85 M68-78 M94-101 B80 B34-52 B88

0.20-0.35 0.08-0.09 0.1-0.2

0.005-0.007 0.015-0.020 0.02-0.025

R80 Brinen 85 Brinen 85 Brinen 82

499

500 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 6·21. Properties and Processes for the Major Thermoset Resins Used in Composites Properties

Processes

Polyesters

Simplest, most versatile, economical, and most widely used family of resins; good etectrical properties, good chemical resistance, especially to acids

Epoxies

Excellent mechanical properties, dimensional stability, chemical resistance (especially to alkalis), low water absorption, selfextinguishing (when halogenated), low shrinkage, good abrasion resistance, excellent adhesion properties Good acid resistance, good electrical properties (except arc resistance), high heat resistance Highest heat resistance, low water absorption, excellent dielectric properties, high arc resistance Good heat resistance, high impact strength Good electrical insulation, low water absorption

Compression molding, filament winding, hand lay-up, mat molding, pressure bag molding, continuous pultrusion, injection molding, spray-up, centrifugal casting, cold molding, encapsulation Compression molding, filament winding, hand lay-up, continuous pultrusion, encapsulation, centrifugal casting

Thennosets

Phenolic resins

Silicones

Melamines Diallyl o-phthalate

Compression molding, continuous lamination Compression molding, injection molding, encapsulation Compression molding Compression molding

stress is developed in the fiber than in the resin. If the stress-strain relationships of fiber and resin are known (e.g., from their stress-strain diagrams), the stresses developed in each for a given strain can be computed and their combined action determined. Figure 6-42 shows stress-strain diagrams for glass fiber and two resins. Curve A, typical of glass, shows that stress and strain are very nearly directly proportional to each other to the breaking point. Here stiffness, or the modulus of elasticity as measured by the ratio of stress to strain, is high. Curve B represents a hard resin. Stress here is directly proportional to strain when both are low, but the stress gradually levels off as the strain increases. Its stiffness is much lower than that of glass. It is measured by the tangent to the curve, usually at the origin. Curve C represents a softer resin intermediate between the hard resin and the very soft plastics. Stress and strain here are again directly proportional at low levels, but not when the strains become large. The modulus of elasticity, as measured by the tangent to the curve, is lower than for the hard resin. These stress-strain diagrams may be applied, for example, to the investigation of a rod of which half its total volume is glass fiber and half resin. If the glass fibers are laid parallel to the axis of the rod, at any cross-section, half the total cross-sectional area is glass and half resin. If the rod is stretched 0.5 percent, reference to the stress-strain diagrams in Figure 6-42 will show that the glass is stressed at an intensity of 345 MPa (50,000 psi) and the resin (if resin B) at 52 MPa (7,500 psi) or (if resin C) at 17 MPa (2,500 psi). If, for example, the rod has a total cross-section of one-half square inch, the glass is one-quarter square inch and the total stress in the glass is one quarter times

Continuous strand

Fabric

-

.c

Chopped strand

1:11 C

I!!

fn

Volume of reinforcement --------~) Figure 6-35. The strength-to-volume relationship for reinforcements used in composites. 50

II)

40

...I

2

'iii

8.

8

E

0

u

)(

l'! :> "5 ....

~30

'u

~

'"

Qi

~:>

20

:;

"8

:::!l

10

0

0

Specific gravity

Figure 6-36. The relationship of the modulus of elasticity to the specific gravity of various materials.

g <.l

c;;

11 (279~

"KEVLAR" 29

.,

(2540)

RESIN IMPREGNATED

..-

9 12286) , 8

STRANDS

S' r

c;

•

(2032)

7

~

{I778)


6

tn

(1524)

~

(1270)

"S"-GLASS

5 OTHER ORGANICS

~

{l0

161 ~~~/ • "E"-GLASS

:;:

17

62~

u

150B1

UJ

:;,

----

"KEVLAR" 49

'<

•

•

HT GRAPHITE

•

BORON • HM GRAPHITE

STEEL

• ALUMINUM 12541 °0~~~-----L----~----4~--~5----~6----~

12541 15081 17621 110161 112701 115241 SPECIFIC TENSILE MODULUS, 10 8 in 110 8 em 1

1177B1

Figure 6-37. The relationship of different reinforcing fibers used in composites, comparing their specific tensile strength and their specific tensile modulus of elasticity (the specific value = the mechanical property divided by the specific gravity). 501

N

VI

=

Glass fibers Carbon fibers Rigid minerals FR additive

Glass fibers Carbon fibers Fibrous minerals

5 to 10% glass fibers 5 to 10% carbon fibers Particulate fillers

Increased flexural modulus

Increased heat-deflection temperature (HDT)

Warpage resistance

Flame resistance

Glass fibers Carbon fibers Fibrous minerals

How Achieved

Increased tensile strength

Desired Modification

NA NA Ductility, cost, tensile strength

Ductility, cost Ductility, cost NA

Ductility, cost Ductility, cost Ductility Ductlity, tensile strength, cost

Ductility, cost Ductility, cost NA

Amorphous

Cost Cost Ductility, cost, tensile strength

Ductility, cost Ductility, cost Ductility

Ductility, cost Ductility, cost Ductility Ductility, tensile strength, cost

Ductility, cost Ductility, cost Ductility

Crystalline

Sacrifice (from base resin) Comments Glass fibers are the most cost-effective way of gaining tensile strength. Carbon fibers are more expensive; fibrous minerals are least expensive but only slightly reinforcing. Reinforcement makes brittle resins tougher and embrittles tough resins. Fibrous minerals are not commonly used in amorphous resins. Any additive more rigid than the base resin produces a more rigid composite. Particular fillers severely degrade impact strength. FR additives interfere with the mechanical integrity of the polymer and often require reinforcement to salvage strength. They also narrow the molding latitude of the base resin. Some can cause mold corrosion. When reinforced, crystalline polymers yield much greater increases in HDT than do amorphous resins. As with tensile strength, fibrous minerals increase HDT only slightly. Fillers do not increase HDT. Amorphous polymers are inherently nonwarping molding resins. Only occasionally are fillers such as milled glass or glass beads added to amorphous materials, because they reduce shrinkage anisotropically. Addition of fibers tends to balance the difference between in-flow and cross-flow shrinkage usually found in crystalline polymers. When a particulate is used to reduce and balance shrinkage, some fiber is needed to offset degradation.

Table 6-22. Trade-offs in Thermoplastic Composites*

w

\11

=

Glass fibers Carbon fibers Lubricating additives Carbon fibers Carbon powders

Reduced wear

'Courtesy ICI-LNP Corp.

Electrical conductivity

PTFE Silicone MoS 2 Graphite

Reduced coefficient of friction

}

Glass fibers Carbon fibers Fillers

Reduced mold shrinkage (increased mold-tosize capability)

Ductilty, cost Tensile strength, ductility, cost

Cost

Cost

Ductility, cost Tensile strength, ductility, cost

Ductility, cost Ductility, cost Tensile strength, ductitily, cost

Ductility, cost Ductility, cost Tensile strength, ductility, cost

Resistivities of I to 100,000 ohm-cm can be achieved and are proportional to cost. Various carbon fibers and powders are available with wide variations in conductivity yields in composites.

Reinforcement reduces shrinkage far more than fillers do. Fillers help balance shrinkage, however, because they replace shrinking polymer. The sharp shrinkage reduction in reinforced crystalline resins can often lead to warpage. The best "mold to size" composites are reinforced amorphous composites. These fillers are soft and do not dramatically affect mechanical properties. PTFE loadings commonly range from 5 to 20%; the others are usually 5% or less. Higher loadings can cause mechanical degradation. The subject of plastic wear is extremely complex and should be discussed with a composite supplier.

""~

Polycarbonate

Polyester (PBT)

Nylon 6/6 Nylon 6

Nylon 6/6

Base Resin

1.19 (1.14) 1.29 1.19 (1.14) 1.33 (1.31) 1.23 (1.20)

Specific Gravity D 792 0.008 (0.016) 0.008 0.008 (0.016) 0.013 (0.020) 0.005 (0.006)

Mold Shrinkage (in.lin.) D 955 0.90 (1.50) 0.6 1.0 (1.8) 0.06 (0.08) 0.12 (0.15)

Water Absorption, 24-hr. (%) D 570 14.5 (11.8) (13.5) 13.0 (11.8) 9.5 (8.5) 11.0 (9.0)

100.0 81.4 93.1 89.6 81.4 65 59 75.8 62

Tensile Strength 103 psi (MPa) D 638 0.64 (0.41) 0.55 0.58 (0.40) 0.60 (0.34) 0.54 (0.33)

(4.4) (2.8) (3.8) (4.0) (2.8) (4.1) (2.3) (3.7) (2.3)

Flexural Modulus 106 psi (GPa) D 790 1.0 (0.9) 1.0 1.1 (1.0) 0.8 (1.2) 0.9 (2.7)

Notched D 256

II (60)

9.0

8.5 9.0

6.7

Unnotched D 256

Impact Strength, Izod (ft.-lb.lin.)

2.4 (4.5) 3.1 3.0 (4.6) 3.0 (5.3) 3.0 (3.7)

Thermal Expansion (10-5in.lin.- OF) D 696

Table 6-23. Properties of Unreinforced vs. Aramid Fibers with TP Resin Composites

450 (170) 465 390 (167) 380 (130) 280 (265)

(232) (76.7) (240) (199) (75) (193) (54.4) (138) (129)

Deflection Temperature, 264 psi OF (OC) D 648

Table 6-24. Fiber Reinforcements Used in Reinforced Plastics

Type of Fiber Reinforcement

Density Ib.lin. 3

Specific Gravity

Glass E Monofilament 12-end roving S Monofilament 12-end roving Boron (tungsten substrate) 4 mil or 5.6 mil Graphite High strength High modulus Intermediate Organic Aramid

(g/cm 3 )

2.54 2.54 2.48 2.48

0.092 0.092 0.090 0.090

2.63

(2.5) (2.5) (2.5) (2.5)

Tensile Strength 103 psi (GPa) 500 372 665 550

Specific Strength 106 in.

Tensile Elastic Modulus 106 psi (GPa)

Specific Elastic Modulus 108 in.

(3.45) (2.56) (4.58) (3.79)

5.43 4.04 7.39 6.17

10.5 10.5 12.4 12.4

(72.4) (72.4) (85.5) (85.5)

1.14 1.14 1.38 1.38

0.095 (2.6)

450 (3.10)

4.74

58

(400)

6.11

1.80 1.94 1.74

0.065 (1.8) 0.070 (1.9) 0.063 (1.7)

400 (2.76) 300 (2.07) 360 (2.48)

6.15 4.29 5.71

38 (262) 55* (380) 27 (190)

5.85 7.86 4.29

1.44

0.052 (1.4)

400 (2.76)

7.69

18

(124)

3.46

IJAlso commercially available up to 100 x 106 psi. Note: The principal reinforcement, with respect to quantity, is glass fibers, but many other types are used (cotton, rayon. polyester/TP, nylon,

aluminum, etc.). Of very limited use because of their cost and processing difficulty are "whishers" (single crystals of alumina, silicon carbide, copper, or others), which have superior mechanical properties.

Table 6-25. Thermal Properties of Reinforcing Fibers E Glass

Property Mean fiber diameter, f.L (mils) Therm. Cond., BTU-in.lhr.-ft. 2

10-17 (0.39-0.67)

io

Carbon 7 (0.27)

60 (8.6)

(1.0)

HM Carbon 8 (0.31)

97 (14)

Aramid 12 (0.47)

3.5 (0.50)

(W/m' K)

Specific Heat @ 70°F, BTU/lb.l°F (J/Kg . K) Coefficient of thermal exp., 1~ in.linfF (I~ cm/cm 0c) Longitudinal Transverse Surface energy, ergs/cm2 Note: One micron

= 0.001

One grain of sall

=

0.192 (803)

1.6 4.0 31.0

(2.9) (7.2)

0.17 (710)

0.17 (710)

-0.55 (-0.99) 9.32 (16.8) 53.0

-0.28 (-0.50) (1.8)

0.34 (1400)

-1.1 (-2.0) 33.0 (59.4) 41.0

em or =0.00004 in.

100 microns.

One human hair = 70 microns. The human eye cannot distinguish below 40 microns. Usually the length of short fibers is less than 3.175 mm (0.125 in.), bUI generally is 0.76 to 0.52 mm (0.030 10 0.060 in.), and long fibers are longer than 3.175 mm (0.125 in.).

505

506 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

_N c:> c:> c:> CI

....

ALUMINIU M _ ~ ALLOY TITANIUM AllOY _HIGH TENSI LE _

STEEl

SPECIAL GLASS FIB RE EPOXY FIBREGlAS S SILICA FIB RE EPOXY l1li CARBON FI BRE BORON FIB RE

... CI

CI

ULTIMATE TENS It( STREN GTH (lbJ /sq in _UII I

... ... c:>

'"

CI

CI

C;

CI

.... c:>

CI

CI

CI

CI

~ co

co

!

foI

I

.... c:> '"

W

c:> c:>

CI

CI

I I

I

I

CARBON FI BRE IRON WHIS KER ..."... '" II - SiC WHISK ER ALUMINA WHISKER GRAPHITE WHISKER

II I

s

I I

JI

I I

I I II 10

II

15

30 25 20 a M . ei • •

I

3S

40

SPECIFIC MODULUS (ii' . 10' )

Figure 6-38. Tensile strengths and specific moduli of elasticity of different types of composite plastics including whisker fibers, as compared to aluminum, titanium, and steel.

ROVINGS

RESIN

RESIN

Figure 6-39. A schematic of the method for manufacturing sheet molding compound (SMC) in which continuous glass-filament rovings go through a chopper (where the length of the chopped fibers is present by changing the location of the blades) at a controlled rate and the resin paste compound would be controlled by a doctor blade that provides an opening for the paste to move over the speed-controlled revolving conveyor belt. Plastic carrier films (not shown) eliminate the sticking problem of B-stage TS compounds, permit ease of handling for shipment, cool room storage, and lay-up for fabrication. The films are removed prior to fabrication lay-up.

50,000, or 12,500 Ibs. Similarly, the stress in the resin (if resin B) is 1,875 Ibs., and in resin C is 625 Ibs. The load required to stretch the rod made with resin B is therefore the sum of the stresses in the glass and resin, or 14,375 Ibs. Similarly, for a rod utilizing resin C the load is 13,125 Ibs. The average stress on the one-half square inch crosssection is therefore 28,750 psi or 26,250 psi, respectively. An analogous line of reasoning shows that at a strain of 1.25 percent the stress intensity in the glass is 862 MPa (125,000 psi) and in resins Band C 87 and 31 MPa (12,600 and 4,500 psi), respectively. The corresponding loads on rods made with resins B and Care 34,400 and 32,375 Ibs., respectively.

THE PROPERTIES OF PLASTICS 507

RESIN

ROVINGS

Figure 6-40. This schematic shows the production of SMCs incorporating long, highperformance fiber reinforcements oriented in either the machine direction or positioned in any direction desired, using single or multiple fibers and rovings to obtain the desired orientation.

~

Molding Cavity

~SMCCh

, I i l~! I! I I i , I I'I! 'II , I i ! II III !I ! i I

I

I!

I I I I iI i I ! I I'll I! I I I I I I ' 'I I i I I I I I.I I I

Typical Charge Pattern

Typical Molding

::; ·'windows" 1st ply - pierced 1st ply ::::;"windows" 2nd ply -.- pierced 2nd ply Figure 6-41. This schematic of the off-line production process used when required to cut directional-type SMC to conform to a specific mold contour to significantly reduce or even eliminate unwanted wrinkles during lay-up.

The foregoing can be put into the form of an equation as (6-1)

where mean stress intensity on entire cross-section

ITt

= =

IT r

=

stress intensity in resin total cross-sectional area

At

= =

cross-sectional area of fiber

Ar

=

cross-sectional area of resin

IT

A

stress intensity in fiber

508 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

If the moduli of elasticity, as measured by the tangents to the stress-strain diagrams, are known, the following relationships hold:

or

(6-2)

where E,

= modulus of elasticity of resin

Ef

= modulus of elasticity of fiber

Substituting (6-2) into (6-1), then, we have (6-3) Referring to Figure 6-42, the tangent to the stress-strain curve for glass gives a value of the modulus of elasticity Ef = 10 X 106 psi. The tangents to the two resin curves give values of E, equal to 1.5 x 106 psi and 0.5 x lQ6 psi, respectively. Substituting these values into (6-3) and solving for the stresses in the one-half-square-inch rod in the previous example gives Resin B

Resin C

O"A

0"

1 .0.25 5) = 50,000 ( 0.25 + 10

0"

= 14,375 lb. = 28,750 psi

A

0 .0.25 5) = 50,000 ( 0.25 + 10

= 0"

13,125 lb.

= 26,250 psi

The average values for the modulus of elasticity of the entire cross-section may be computed by dividing 0" by the strain. The strain is 0.5%. Therefore, the two average values for E of the rod, incorporating resins B and C, are 5.75 x lQ6 psi and 5.35 x 106 psi, respectively. For a cross-section made up of a number of different materials, (6-1) may be generalized to i=n

O"A

=

:L O"t Ai

(6-4)

i=1

in which O"t is the tensile strength and Ai the cross-sectional area of any component in the cross-section. This equation can be still further generalized to include tension, compression, and shear as i=n

SA

=

:L SiAi

;=1

(6-5)

THE PROPERTIES OF PLASTICS 509

240

60

220

55

200

50

180

45

160

40 ·iii

'g, 140

c.. 35

§

~ 120

:; 30

.!:

III US

Glass fiber A

'"2!'" 25

2!1oo

US

80

20

60

15

40

10

Tangent to B

I-

I

I

I

I

I

/

20 0

0.5

1.5

1.0

2.0

2.5

3.0

% Strain

% Strain

Figure 6-42. The stress-strain diagrams for a glass fiber A and two resins B and C. Resin B is a hard, high-strength material, resin C of intermediate strength and hardness.

in which Si is the strength property of the cross-sectional area Ai of component i and S is the mean strength property over the entire cross-section A. Similarly, to find the overall modulus of elasticity of a cross-section the equation becomes i=n

EA

(6-6) ;= 1

in which E is the overall modulus of elasticity, A the total cross-section, and Ei the modulus of elasticity corresponding to the partial cross-sectional area Ai. For shear modulus G the equation then becomes i=n

GA

(6-7)

Plain Reinforced Plates

Fibrous reinforced plates, either flat or curved, are commonly made with matt, fabrics, and parallel filaments, alone or in combination. Matt is usually used for good strength at minimum cost, fabrics for high strength, and parallel filaments for maximum strength in a particular direction. Because the fibers in a mat are randomly oriented, matt-reinforced materials have essentially the same strength and elastic properties in all directions in the plane of the

510 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

plate, that is, they are essentially isotropic in the plane. Consequently, the usual engineering theories and design methods employed for isotropic engineering materials may be applied. It is necessary only to know the strength, the modulus of elasticity, the shearing modulus, and Poisson's ratio for the combined matt and resin. These can be obtained from standard stress-strain measurements made on specimens of the particular combination of fiber and resin under consideration. In fabric- and roving-reinforced materials the strength and elastic properties are different in different directions; that is, they are not isotropic, and the usual engineering equations must accordingly be modified. Because fabrics are woven with their yarns at right angles (in the warp and fill directions), a single layer of fabric-reinforced material has two principal directions or natural axes, the longitudinal (warp) and the transverse (fill), at right angles to each other. Such a structure is called orthotropic (in right-angled directions). Having parallel strands of fiber, as in a single layer of roving-reinforced or unidirectional fabric-reinforced plates, also results in orthotropic materials, with one direction parallel and the other at right angles to the fibers. Multilayered plates in which layers of fabric or roving are laid up parallel or perpendicular to each other are also orthotropic. If the same number of strands or yarns is found in each principal direction (balanced construction), the strength and elastic properties will be the same in those directions, but not at intermediate angles; if the number of strands or yarns is different in the two principal directions (unbalanced construction), the strength and elastic properties will be different in those directions as well as at all intermediate angles. In the foregoing discussion the direction perpendicular to the plane of the plate has been neglected because the plate is assumed to be thin and the stresses are assumed to be applied in the plane of the plate rather than perpendicular to it. This assumption, which considerably simplifies the theory, carries through all the following discussion. It is true, of course, that properties perpendicular to the plane of the plate are undoubtedly different than those in the plane of the plate, and in thick plates this difference has to be taken into account, particularly when the stresses are not planar. For isotropic materials, such as mat-reinforced construction, if E is the modulus of

elasticity in any reference direction, the modulus E1 at any angle to this direction is the same, and the ratio E11E is therefore unity. Poisson's ratio v is similarly a constant in all directions, and the shearing modulus is G = E/2(1 + v). If Vi for example, is 0.3, then GIE = 0.385 at all angles. These relationships are shown in Figure 6-43. The following familiar relationships between direct stress (T and strain E and shearing stress T and strain 'Y hold: E

=

'Y

A transverse strain (contraction or dilation) ET

=

(TIE

(6-8)

T/G

(6-9)

ET

is caused by

-VE

(T

equal to (6-10)

For orthotropic materials, such as fabric and roving-reinforced construction, EL and are the elastic moduli in the longitudinal (L) and transverse (T) directions, GLT is the shearing modulus associated with these directions, VLT is Poisson's ratio giving the transverse strain caused by a stress in the longitudinal direction, and VTL is Poisson's ratio giving the longitudinal strain caused by a stress in the transverse direction. The modulus

ET

THE PROPERTIES OF PLASTICS 511 I

I

I

I

I

I

I

I

I

I

1.ooI--------=E~1/:=E------

-

0.80 f-

-

0.70 !-

-

0.60 ,.---

-

0.90 f-

0.301--------------

-

0.20 -

-

0.10 -

-

0.50 -

1

0.40 1-_ _ _ _ _ _ _G_/_E_=_2(..:...l_+....;"')~_ __

"

I

I

I

I

I

I

I

I

I

I

°0~~1~0~2~0~3~0--4~0--5~0~~60~~70~~OO7-~9~0~1700~ Degrees

Figure 6-43. The modulus of elasticity, the shear modulus, and Poisson's ratio for an isotropic material such as fiber matt-reinforced plastics. Because constants do not vary with the angle of load, the ratio of modulus E 1 at any angle to E at any reference direction is unity. The shear modulus is a constant proportion of E, and Poisson's ratio is constant.

at any intennediate angle is E 1 • If 0'1 is a stress applied in the 1 direction at an angle a with the longitudinal direction (see Figure 6-44, top), the stress al causes strain £1 (6-11)

in which E I may be found from (6-12)

This relationship is plotted as EIfEL in Figure 6-44, in which 0° corresponds to the longitudinal direction and 90° to the transverse direction. A transverse strain £2 is caused by 0'1 (6-13)

in which (see Figure 6-44) V12 =

;~

{VLT -

~ (1

+

2VLT

+

;~ - ~:J sin

2

2a}

(6-14)

Unlike isotropic materials, when stress 0'1 is applied at any angle except 0° and 90°, it causes shear distortion. The shear strain 'V12 is found from (6-15)

512 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

in which (see Figure 6-44)

=

ml

sin 2 a{VLT

A shearing stress

1T12

+ EL ET

!

EL - cos 2 a(1 + 2VLT + EL _ EL)} ET GLT

2 GLT

applied in the 1-2 directions causes shear strain

(6-16)

"112

(6-17) in which (see Fig. 6-44)

GLT = GLT G I2 EL

{(I +

2VLT + EL) _ ET

(1 +

2vLT + EL - EL) cos 2 2a} ET GLT

(6-18)

This relationship is plotted as G I2 /G LT in Figure 6-44. Unlike isotropic materials, stress 1'12 causes strain 101 in the 1 direction: (6-19)

2

1.00 0.90 0.80 0.70 eo!

I:

0.50 0.40 0.30 0.20 0.10 0 -0.10

e.. ~

--rS.,.

0.60

~

'C

... 0.50

~

J; 0.40

0.30

--~ ~

0.20 0.10 0 Angle a

Figure 6·44. The elastic constants of an unbalanced orthotropic material. E\, G l2 , and 'Y12 are all functions of the angle between the direction of stress and the longitudinal axis (warp direction) of the material. Factors ml and m2 account for the direct and shear strains caused by direct and shear stresses, respectively. Angle 0° is the longitudinal direction, angle 90° the transverse direction.

THE PROPERTIES OF PLASTICS 513

and strain

E2

in the 2 direction: (6-20)

in which (see Fig. 6-44) m2

=

sin 2a{VLT

+

EL - -21 EL - sin2 ET GLT

a(

1

+

2VLT

+

EL _ EL)} ET GLT

(6-21)

The two values of Poisson's ratio are related: (6-22) In plotting Figure 6-44 the following values were used: EL

=

5,000,000 psi

ET

=

500,000 psi

GLT

=

550,000 psi

VLT

=

Voo

VTL =

=

V90°

0.450

= 0.045

These values might correspond, for example, to a parallel glass-filament-reinforced panel employing an intermediate polyester resin. When an orthotropic material is balanced, its longitudinal and transverse properties are the same; that is, EL = ET and VLT = VLT = VTL. The properties are symmetrical about the 45° angle, as shown in Figure 6-45, in which the following values were used: EL = ET = 3,000,000 psi

=

VLT

GLT

=

VTL

= 0.20

500,000 psi

These values might correspond, for example, to a square-weave or symmetrical satinweave fabric-reinforced construction. As an example of an application of the foregoing equations, tensile stress ITI acting on the small plate at the top of Figure 6-44 is 68 MPa (10,000 psi), its shear stress 'T12 is 28 MPa (4,000 psi) and angle a is 30°. Then, from Figure 6-44 we can determine that EI

or

G 12/G LT

=

0.81 or G I2

V12 = - 0.0286

Then strains caused by EI

=

IT 1

=

=

0.367

x 5,000,000 = 1,830,000 psi

0.81 x 550,000

ml = 4.66

m2

=

445,000 psi

= 4.98

are

10,000/1,830,000

=

5.45 x 10- 3

(6-23)

514 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

+1.00 +0.80 +0.60

+2.00

'" :;:

.

+1.00

"'" +0.20 c:

~ ~ :::

{ -0.2:

~ 0.600

-0.40

-

0.400

E;

..

rS.,.

IS

-0.60 -0.80 -1.00

o Angle III

Figure 6-45. The elastic constants in a balanced orthotropic material. These constants and angles have the same meanings as in Figures 6-42, 6-43, and 6-44. EI "112

= - (- 0.0286) 5.45 =

The strains caused by "112

X 10- 3

= =

-4.66 X 10,000/5,000,000 'T 12

0.16 X 10- 3 -9.32 X 10- 3

(6-24) (6-25)

are

= 4,000/550,000 = 7.28 X 10- 3

(6-26)

EI

= -4.66

X

4,000/5,000,000

= -3.73

X

10- 3

(6-27)

E2

=

X

4,000/5,000,000

=

X

10- 3

(6-28)

-4.98

-3.98

The total strains, therefore, are "112

=

-2.04

X

EI

=

1.72

10- 3

E2

= - 3.82

X

X

10- 3

10- 3

Problems involving Figure 6-45 can be solved in an analogous manner. It must be kept in mind that equations 6-12,6-14,6-16,6-18, and 6-21 are valid and useful if the fibers and the resin behave together in accordance with the assumptions upon which their derivation is based. If only the values of EL , ET , GLT , and VLT are available, the intermediate values of Ej, G 12 , V12, and the values of ml and m2 can be estimated by means of these equations.

THE PROPERTIES OF PLASTICS 515

Composite Plates Fibrous reinforced plates are in practice often made up of several layers, each of which may be of different construction, such as mat, fabric, or roving. Furthermore, the various layers may be oriented at different angles with respect to each other in order to provide the best combination to resist some particular loading condition. Outside loads or stresses applied to a composite plate of this type result in internal stresses that are different in the individual layers. External direct stresses may result not only in internal direct stresses but in internal shear stresses as well, and external shear stresses may result in internal direct stresses as well as internal shear stresses. Figure 6-46 depicts a small composite plate made up of materials a and b having principal longitudinal and transverse directions La and Ta and Lb and Tb respectiVely. Several layers of each are present, with their total thicknesses as ta and tb, respectively, and an overall thickness of t. Outside stresses (TJ, (T2, and TI2 are applied in the 1 and 2 directions, as shown. The 1 direction makes angle a with La and reverse angle J3 with Lb. Angle a is considered to be positive and angle J3 negative. The internal stresses (Tla, (T2a, TI2a, and (Tlb, (T2b, TJ2b in the individual layers can be found by observing that the sums of the internal stresses in the 1 and 2 directions must be equal the external stresses in these directions and that the strains must be the same in all layers. These relationships may be written in the following forms:

(Tlata

(T2at a

T 12ata

+

(Tlbtb

+

(T2btb

+

T 12btb

r

=

(TIt; (Tlb

=

(T2t ; (T2b

=

(TIt -

(Tlata

(6-29)

tb (T2t

-

(T2at a

(6-30)

tb

TJ2t; TI2b

=

TJ2ata

TI2t -

(6-31)

tb

l

--T12

I

~~ la

I I

0"2+--

!

T12

al bl a2

goo

goo

I

I

Tb

T..

~ a3

~0"2

I

I

-+--

tl

2

t,. = tal + ta2 + t,.3

tb = tb) + tb2 t=ta+tb

Figure 6·46. A composite panel with layers a and b of different orthotropic materials oriented at arbitrary angles 0: and j3 with respect to stress 0"1> 0"2, and T12.

516 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

(6-32) (6-33) 'Y12a

= 'Y12

'Y12b

(6-34)

Strains and stresses are induced in each layer. Because the layers are firmly bonded together, the strains are the same in the a and b layers and are equal to the strains in the whole plate: -

(T2a V21a E2a

TJ2c mla-E La

Elb -

(T2b V21b E- 2b

EI2b m lb -E Lb

(Tla

{

10 la Ela

(Tlb Elb -

(6-35)

(6-36)

10 2

'YI2a

=

'YJ2a

=

(6-37)

{

Solving equations 6-26 through 6-37 leads to the following simultaneous equations: (6-38) (6-39) (6-40) in which AIJ

A21

A31

1

Elata

+--

AJ2

Elbtb

VI2a

VI2b

Elata

Elbtb

= AI3

A32

V21a

V21b

E2at a

E2bt b

1

A22

= A 23

E2at a

A33

+-E2bt b

AJ3

A 23

1

1

G I2a ta

GJ2bt b

- -mla -

---

ELata

ELbtb

m2a ELata

mlb

- -m2b ELbtb

=--+--

where A2 = aJ2, numerically. An application of the foregoing expressions may be illustrated by a cylindrical pressure vessel, as shown in Figure 6-47. The wall of this vessel, having an eternal radius of 127

THE PROPERTIES OF PLASTICS 517

mm (5 in.) and a wall thickness of 5.08 mm (0.20 in.), may be considered to be a thin plate. It is subjected to an internal pressure of 5.5 MPa (800 psi). The circumferential stress (1\ and longitudinal stress (12 in the wall are calculated

pro t

=

. 19 200 PSt '

pro 2t

=

. 9 600 PSt

(1\

=-

(12

=-

'

The stresses acting on a small part of the wall are therefore as shown in Figure 6-47. Three types of construction will be investigated, as shown in the three parts of Figure 6-47. All three employ the balanced fabric having the characteristics shown in Figure 645. In (1) the fabric is simply wrapped in layers a and b with the Land T directions laid in the circumferential and axial directions. In (2) the layers are laid at 45° to the axis of the cylinder, and in (3) they are laid at alternate 30° angles in left-hand and right-hand spirals as shown. In each instance ta = tb = 0.10 in.

'~ \~i':~~;=

0'2 + -

-

0'2

9600 psi

2 /0'1

p Tl

= 800 psi

= 5.00 in.

ro =4.80 in.

1

.-

9600 psi

t19,200 psi

~ Tb

--

90· Lb 9600 psi T"

~

2

t 19,200 psi 45"

9600 psi

.-

-

La Tb

9600 psi

go•

Lb T"

.-

9600 psi

2

~

19,200 .psi

19,200 psi

(1)

(2)

~

19,200 psi tatb • ~0.10\n.

(3)

"""1

~t4. 0.20;n.

Figure 6-47. A fiberglass-reinforced plastic thin-walled cylinder showing internal pressure alone.

518 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Referring to Figure 6-45, it can be seen that for Case 1

Vl3a

=

V21a

=

=

Vl2b

V21b

=

0.20

...

Equations 6-38 to 6-40 therefore become

Solution of these equations and reference to equations 6-29 to 6-31 shows that O"la

=

O"lb

=

0"1

=

19,200 psi

0"2a

=

0"2b

=

0"2

=

9,600 psi

'T12a

=

'T12b

=0

This proves what might have been expected intuitively, that because of symmetry with respect to the 1-2 directions chosen the internal direct stresses O"la, O"lb, 0"2a, and 0"2b are equal to the imposed stresses 0"1 and 0"2, and there is no internal shear stress. The same result is found for Case 2. In this balanced fabric ml = m2 = 0 at 45°, there is no shear distortion caused by direct stress, and the shear therefore is zero. In Case 3,

= 0.597

x 3 x 106

=

= 1.78 X 106 psi

G I2a

=

G I2b

VI2a

=

VI2b

=

V30°

=

V60°

=

mla

=

m30°

=

0.775,

mlb

=

-ml a

=

-0.775

m2a

=

m60°

=

0.775,

m2b

=

-m2a

=

-0.775

1.82 x 0.5 x 106 V21a

=

=

0.91

V21b

=

X

106

0.523

The values of mlb and m2b are negative, because the 30° angles of orientation of the longitudinal direction Lb of layers b are measured in the negative direction, whereas it is positive for the a layers.

THE PROPERTIES OF PLASTICS 519

Equations 6-38 become

The first two of these equations are exactly like the first two equations for Cases 1 and 2 and show that the internal direct stresses are equal to the imposed; that is, (11

=

19,200 psi

(12

=

9,600 psi

The third equation is not, however, equal to zero, and its solution, together with equation 19c, shows that 'rJ2a

=

6,750 psi

'rl2b

=

-6,750 psi

Appreciable shear stresses are set up within the body of the cylinder wall when the layers are oriented as in Case 3, even though no shear forces are applied to the cylinder itself. The shear stresses in layers b are oriented in the direction opposite to those in layers a. The difference in the shear stresses between the two layers must be taken up by shear in the adhesive bond between them, in the layer of resin that holds the fiber-reinforced layers together. The difference is 6,750 - (-6,750) = 13,500 psi. This shear stress in the bonding resin is therefore seen to be high. In Cases 1 and 2 the orientation of the fibers with respect to the 1 and 2 directions chosen resulted in zero shear stresses associated with those directions, whereas in Case 3 the shear stresses were not zero. In all three cases, a symmetry of fiber orientations with respect to the stress directions resulted in internal stresses that were equal to the external stresses. These are special cases. In the more general case, the internal direct stresses in the individual layers are not necessarily equal to the external direct stresses, nor are they the same in the various layers. Furthermore, even symmetrical Case 3 leads to there being internal shear stresses when external shear stresses are absent. In the more general case it is even more true that internal shear stresses may be appreciable, or be absent, depending upon the magnitude of the external stresses and the orientation of the 1 and 2 directions with respect to the external stresses. A more general case is shown in Figure 6-48 in which the same cylinder is chosen as in Figure 6-47 except that a torsional effect equal to a twisting couple of 25,000 in.-Ib. has been added. The construction of the wall has also been changed. Layers a of ~n­ balanced material having the properties of Figure 6-44 are a total of 0.13 in. thick and are oriented at 15° to the circumferential direction, as shown. Layers b, of balanced material having the properties of Figure 6-45, are a total of 0.07 in. thick and are oriented at 45°, as shown. Referring to Figure 6-45, the properties are found to be

520 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK b layers

a layers ta = 0.13 in. a = 15° Ela = 0.703

tb

=

3.515

X

106 psi

E14 = 0.109 X 5 X 106 = 0.545 ])114 = 0.193 mla = 2.63 m14 = 2.94 G I14 = 0.93 X 0.5 X 106 = 0.465 X 106 psi

X

106 psi

X

5

X

106

=

0.07 in.

a = 45°

0.526 X 3 X 1
Elb

= E2b =

X

106 psi

Solving for the various constants, and substituting in equations 6-38 through 6-40, we find that 11.2412
+

= + 190,180

23.2030
+ 26.4668Tl2a = + 16,190

-4.0461 glb - 4.5156
The solution of the foregoing simultaneous equations leads to the following results for stresses in the a layers:

0"2

= 9600 psi T12

p ,; 0"2 = 9600 psi

= 920 psi - - ) 0"1

0"1

= 19,200 psi

') \ T,.-m J ~ 0"1

= 19,200 psi

= 19,200 psi

t

_ 9 2 0 psi

-920 psi

i

-9600 psi ~

!

2

19,200 psi

Figure 6-48. A fiberglass-reinforced plastic thin-walled cylinder with internal pressure plus twisting moment.

THE PROPERTIES OF PLASTICS 521 (Jla

=

21,100 psi

(JZa

=

5,200 psi

TlZa

=

4,740 psi

When these results are employed with equations 6-29 to 6-31, it is found that the stresses in the b layers are: (JIb

=

(J2b

= 17,800 psi

=

TI2b

15,700 psi

-6,150 psi

Bending of Beams and Plates Plates and beams of fibrous glass-reinforced plastics may be either homogeneous and isotropic or composite and nonisotropic, depending upon their structure. Mat-reinforced plates may be considered to be essentially isotropic, and the usual engineering formulas may be applied. Composite structures require suitably modified formulas, but otherwise the procedures for computing bending stresses, stiffness, and bending shear stresses are essentially the same as for isotropic materials. The differences and similarities may be brought out by considering two beams of identical overall dimensions, one isotropic and the other composite. Two such cross-sections are shown in Figure 6-49. For each crosssection it is necessary to know the stiffness factor, EI, to compute deflection, the section modulus to compute bending stresses, and the statical moments of portions of the crosssection to compute shear stresses. For isotropic materials (a) the neutral axis of a rectangular cross-section is at mid-depth, and the familiar formulas are Moment of inertia I

bd 3

= 12' stiffness factor =

EI

I bd 2 Section modulus = - = - for outermost fiber y 6 Bending stress

=

Shear stress =

(J

__y

6M

I

bd 2

= M- = -

VQ hi

=

for outermost fiber

V. . 23 bd for maximum shear at neutral aXIs

(6-41) (6-42) (6-43) (6-44)

For composite materials the neutral axis is not necessarily at the mid-depth point of a rectangular section and it must first be found: Neutral axis X

= LEi AiXi/LEi Ai

(6-45)

in which Ei , Ai, and Xi are the modulus of elasticity, the cross-sectional area (bdi), and the distance from some reference line, such as the bottom of the cross-section, to the center of gravity of any particular layer.

522 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 11

~E

11 AIEl A2E2 .!£ -Aa E3 A4E, AsEs

A,E Neutral axis

0/2

(/I)

(b)

U)OO" /I

11

c

is 0

-II

~

...0

(1)

(2)

~

:!!

- r---.~ --(3)

~

0

~1 ~

t

~ 0

II

1

... d

(4)

~

0-r.:::r-~

11

l8 0 "'r.~ _..Q. - - f - - . -

!!It-....

......ri:

0

'"

__t

d

It)

"'0=

0

IF

(5)

-I

(c)

~

is is '"0 ...'"0 ~

1.

~t fn

It)

...

:3'

0

...

N

Neutral axis ,

is ...~ ~

t

~

lit

I

....i£

~ IE

::u..

~I~

10'" (d)

Figure 6-49. a) Cross-section of an isotropic beam; b) cross-section of a composite beam made of layers of different materials; and c) cross-section of a composite beam having properties indicated. E1 = 5.0 X lOb psi (11 = 40,000 psi E2 = 3.0 x lW psi 0"2 = 25,000 psi E3 = 1.0 X 106 psi (13 = 5,000 psi E4 = 5.0 X 106 psi (14 = 40,000 psi Es = 3.0 X 106 psi (15 = 25,000 psi

= EI = ~Ei Ii

Stiffness factor

(6-46)

in which Ei and Ii are, for any particular layer, the modulus of elasticity and the moment of inertia about the neutral axis. Bending stress

C1

= MEyytEI

(6-47)

in which y is the distance from the neutral axis to any point and Ey is the modulus of elasticity of the layer at that point. The maximum bending stress does not necessarily occur at the outermost (top or bottom) fiber, as it does in isotropic materials. Shear stress

T

= VQ'thEI

(6-48)

in which V is the total shear on the cross-section, T is the shear stress intensity along some horizontal plane, and Q' is the weighted statical moment EiAiY about the beam's

THE PROPERTIES OF PLASTICS 523

neutral axis, of the portion of the cross-section between the horizontal plane in question and the top or bottom outer edge of the cross-section. An example of the foregoing is illustrated in Figure 6-49(a) in which a composite beam is made up of five layers having three differential moduli of elasticity, and three different strengths, as shown. The neutral axis, found by applying equation 6-45, is 0.415 in. from the bottom of the cross-section. The distances from the neutral axis to the centers of the individual layers are computed, and the stiffness factor EI calculated by means of equation 6-46. This is found to be

EI = LEi Ii = 0.174

106 lb. in. 2

X

Bending stresses are next computed for the top and bottom edges of the cross-section and the outer edge of each layer; that is, the edge of each layer farther from the neutral axis. From these the bending moment the cross-section is capable of carrying can be computed. This may be done, for example, by applying a bending moment M of one in.lb. and computing the unit bending stresses. These unit bending stresses, divided into the strengths of the individual layers, give a series of calculated resisting moments the smallest of which is the maximum bending moment the beam is capable of carrying without exceeding the strength of any portion of the cross-section. For a unit bending moment M = 1 in.-Ib. ITy

EyY EI

=-

from equation 6-47.

Plane a-a bob

e-e dod e-e

f1

Ey

y 0.385 0.185 0.085 0.115 0.315 0.415

in. in. in. in. in. in.

'5 3 1

X


1~ X 1~

X

I x 106 5 x 106 3 X 106

11.1 3.19 0.49 0.66 9.07 7.16

psi psi psi psi psi psi


--40,000/11.1 25,000/3.19 5,000/0.49 5,00010.66

40,000/9.07

25,00017 .16

M

3,600 in.-Ih. 7,800 in.-Ih. 10,200 in.-Ih. 7,600 in.-Ih. 4,400 in.-Ih. 3,500 in.-Ih.

If, for example, the beam were a simple beam carrying a load Won a lO-in. span, as shown in Figure 6-48, the bending moment at the center of the span would be WLI4. Setting this equal to 3,500 in.-Ib. gives load Was 1,400 lb. Shear V is then WI2 or 700 lb. Using this value, the shear stress intensity at various horizontal planes in the beam may be computed by means of equation 6-48. For planes b-b, c-c, and d-d, for example: Plane

Layers

EA,

y'

Q'

T

bob

g

0.2 x 5 x 106 0.2 x 5 x 106 +0.1 x 3 x 106

0.285" 0.285 0.135

0.285 x 106

1,150 psi

0.326 x 106

1,315 psi

{~

0.2 x 5 x 106 +0.1 x 3 x 106

0.215 0.365

0.324 x 106

1,310 psi

e-e dod

1

524 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

These would be the critical planes, because they represent planes between layers of different materials, and consequently the resin alone would largely carry the stress. The shear stress at the neutral axis would be slightly higher and might or might not represent the critical plane, depending upon the structure of the material in layer three. REGRIND AND RECYCLING

In practically all processing plants it is necessary to reclaim reprocessable TP scrap, flash, rejected parts, and so on. The reclaiming ofTSs is usually not done, however, since they cannot be remelted. They can nevertheless be granulated and used as fillers for certain TS or TP plastics. If possible, the target is to eliminate having a scrap, because it has already cost money and time to go through the process; granulating just adds more money and time. Different types of granulators are available, from many different suppliers, and the selection of a granulator depends on such factors as the type of plastic used, the type of reinforcement, the product's thickness and shape, and so on. There are some series units in which subsystem granulators incrementally reduce the size from a large one to the required small size [10-12, 24]. When reinforced plastics are granulated, the length of their fibers is reduced. When they are reprocessed, with virgin materials or alone, their process ability and performance definitely change. So it is important to determine if the change will affect final part performance; if it will, a limit for the amount of regrind mix should be determined. An easy "cutting" unit is required, to granulate with minimal friction, as too much heat will destroy the plastic. The general-purpose types have definite limitations. Blending with virgin material definitely influences, and can significantly change, the melt-processing conditions and the performance of the end product. Recycled material is denser and usually has a variable-size regrind that could affect a product's properties, as shown in Figure 6-50 and 6-51.

TenSile strength IASTM 0638·721

Impact strength IASTM 0256·731

12.000

460 N" Ion

S/'I,\,i

40

450 440

~ 20

200

10.000

8000

6000

II

;'

f

HOPE

..e

~

POIYSlyrt' n e

"160 11

i< 120

HOPE

10

0

W

t>

4000

~

10

~ 2000

Per cent elongation IASTM 012381

60

SAN

8

HOPE

6

SAN Polystyrene

20 40 60 80 Scrap content, %

100

a

20

40

60

Scrap content. %

80

100

a

20

40

60

80

100

Scrap cont.nt, %

Figure 6-50. How regrind levels affect the mechanical properties of certain formulations of plastics "once through" the fabricating process and blended with virgin material. The regrind, or scrap, amount is a percentage by weight.

THE PROPERTIES OF PLASTICS 525

100

-

.c en

--

95

-----j----1 25

40

90

c

85

III

80

!!!

Tensile Strength

60

CI

C

--.... CI

;:::

c

IU IU

c

IU U

IU

1:1..

100

Impact Strength

95

25

90

40

85 80 ~____~____~__~60 1st 2nd 4th 3rd Number of times molded (Heat history effects performances) Figure 6-51. An example of the potential effects of regrinding on the performance of an injection-molded TP mixed and blended with virgin material.

GUIDE FOR PLASTICS IDENTIFICATION Throughout this book references are made to different classifications of materials, products, and test specifications and standards. The "References" section at the end of this book lists the organizations that provide this type of information.

Plastics Classification ASTM D 4000 A "classifying" materials standard for plastics that can serve many of the industry's needs has been issued for ASTM. This standard, designated D 4000, or ISO 1043, is entitled "Standard Guide for Identification of Plastic Materials." It provides for an easy means of identifying plastic materials used in fabricating parts. Ever since classification systems were adopted years ago for materials such as 1030 steel and elastomers, there has been activity toward issuing this guide. The approach used follows the ASTM unified classification systems for steel and elastomers. This guide provides properties tabulated for unfilled, filled, and reinforced plastic materials that are suitable for all methods of processing into parts. This standard is required to reduce the growing number of material specifications and the paperwork and manhours needed to ensure that parts of known quality are being produced from commercially available materials. The D 4000 standard will eliminate the many certifications required for the same material that a processor might have to obtain from several vendors for one or more customer's. Table 6-27 provides a basic outline that identifies the D 4000 line callouts (specifications). The classification system and its subsequent line callouts is intended to be a means of identifying plastic materials used to fabricate end items or parts. It is not intended for the selection of materials. Material selection should be made after careful consideration of the design and performance required of the part, the environment to which it will be

526 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 6-26. Characteristics of Glass Fiber Reinforced Polyesters (TS) Polyester Type

Characteristic

General purpose

Rigid moldings.

Flexible resins and semirigid resins

Tough, good impact resistance, high flexural strength, low flexural modulus.

Light-stable and weatherresistant Chemical-resistant

Resistant to weather and ultraviolet degradation.

Flame-resistant

High heat distortion Hot strength Low exotherm

Extended pot life Air dry Thixotropic

Highest chemical resistance of polyester group; excellent acid resistance, fair in alkalies. Self-extinguishing, rigid.

•

Service up to 500°F, rigid. Fast rate of cure (hot), moldings easily removed from die. Void-free thick laminates, low heat generated during cure. Void-free uniform, long flow time in mold before gel. Cures tack-free at room temperature. Resists flow or drainage when applied to vertical surfaces.

Typical Uses Trays, boats, tanks, boxes, luggage, seating. Vibration damping; machine covers and guards, safety helmets, electronic part encapsulation, gel coats, patching compounds, auto bodies, boats. Structural panels, skylighting, glazing. Corrosion-resistant applications, such as pipe, tanks, ducts, fume stacks. Building panels (interior), electrical components, fuel tanks. Aircraft parts. Containers, trays, housings. Encapsulating electronic components, electrical premix parts-switch-gear. Large complex moldings. Pools, boats, tanks. Boats, pools, tank linings.

exposed, the fabrication process to be employed, the inherent properties of the material that are not covered in this document, and the various economic factors. This classification system is based on the fact that plastic materials can be arranged into broad generic families and then, using their basic properties, arranged into groups, classes, and grades. A system is thus established that, together with the values describing additional requirements, permits as complete a description as desired of the selected material (see Tables 6-28 through 6-32). Plastic materials are to be classified first on the basis of their broad generic family, identified by the letter designations shown in Table 6-28. These letters represent the standard abbreviations for plastics in accordance with ASTM abbreviations for D 1600. For example, PA indicates polyamide (nylon). A generic family is based on the broad chemical makeup of its base polymer. By its designation, certain inherent properties are specified. The generic family is then classified into groups according to their general chemical composition. These groups are further subdivided into classes and grades, as shown in the basic property table that applies. The letter designation applicable is followed by a three-digit number indicating its group, class, and grade. The basic property tables have been developed to identify the commercially available reinforced plastics into groups, classes, and grades. These tables are found in the ASTM standards listed in D 4000.

THE PROPERTIES OF PLASTICS 527

Table 6-27. ASTM D 4000 Line Callout

o

2

3

4

5

Cell Requirements

Specific Group

Broad generic type

I

6

Reinforcement % Reinforcement

I Group Class Grade I

Table

I

Ixxxxxi Physical Properties

o=

One digit for expanded group, as needed. I = Two or more lette.. identify the generic family, based on Abbreviations D 1600. 2 = Three digits identify the specific chemical group, the modification or use class, and the grade by viscosity or level of modification. A basic property table will provide property values. 3 = One letter indicates reinforcement type. 4 = Two digits indicate percent of reinforcement. S = One letter refe.. to a cell table listing of physical specifications and test methods. 6 = Five digits refer to the specific physical parameten listed in the cell table.

The format of this system was prepared to permit the addition of property values for future plastics. To facilitate the incorporation of future materials, or where the present families require expanding a basic property table, a number preceding the symbol for the generic family is used to indicate that additional groups have been added to the table. This digit, coupled with the first digit after the generic family, indicates the group under which the material is to be found in the basic property table.

Rubber/Elastomer Classification ASTM D 2000 When D 4000 for plastics was being prepared, D 2000 was used as a guide, since it provided many years of international experience. ASTM D 2000 has proven a very useful standard, as previously noted.

Identifying Plastics To identify a specific plastic, the detailed techniques of characterization available from different industry laboratories are used, as well as the more conventional chemical analysis and synthesis methods routinely performed in various laboratories. To provide a quick way of identifying plastics, refer to Figure 6-52 the following chart. It is only a guide, not foolproof. A detailed chart would cover a wide range of plastics (see Figure 6-52). Although the chart may appear to be somewhat formidable at first glance, only three simple tests are necessary to identify all the plastics shown. No special equipment is needed-just water, matches, and a hot surface-and the only sensors required are one's eyes and nose. The first step is to try to melt the material to determine whether it is a thermoset or a thermoplastic. This is usually done with a soldering iron, but any implement with a temperature of approximately 260°C (5000f) could be used. If the material softens, it is a thermoplastic; if not, it is a thermoset.

528 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

If the material is found to be a thermoplastic, the next step is to find out whether its specific gravity is greater than or less than 1. This is done simply by dropping a sample in water. If the material floats, its specific gravity is less than 1; if it sinks, its specific gravity is greater than 1. The thermoplastics that have a specific gravity less than 1 are the polyeolefins-polypropylene and polyethylene. The final step for both thermosets and thermoplastics is a burn test, which should, of course, be performed in a well-ventilated area. The material should be held with pliers or clamps, and ignited with long wooden matches or a Bunsen burner. If there is only a small piece of material to test, it is best to break it into several parts, as it might take several tries to identify the odor and observe the other effects noted on the chart. The major difficulty in interpreting a bum test is that the bum rate and color of the flame of many plastics are affected by fillers, fire retardants, and other additives. However, in most cases the odor is not affected by these additives. It is recommended that you first perform the tests on a styrene drinking glass, a polyethylene milk bottle, or some other known plastic. This practice will prove invaluable when it is time to identify an unknown material. The identification tests reviewed in this section are only a quick way possibly to get information about the type of plastic. They should not replace the laboratory analysis and testing of the material for definitive identification (see Chapter 9).

COMPUTERIZED DATABASES There are more than fifteen thousand plastics compounds alone, so to review what is available as well as keeping up with the constantly proliferating new types for a specific set of design requirements can seem daunting. Nevertheless, with a logical approach to design and engineering this can be done practically. However, it would probably be impossible to keep up to date manually even for the veteran. Manual searching capable of doing the job at an affordable cost has become difficult to arrange for. On-line computerized databases can cut through this information overload by organizing a material's properties into a manageable format. Such programs not only significantly reduce time but also present a host of new options. Besides doing a relatively fast, efficient materials search on what is available today, some databases also offer integration with CAD/CAE/CAM systems to support processing, finite element analysis, and other programs. See Appendix D for information on the databases available. To make the databases more practical and useful, major international agreements are being arrived at to set uniform methods for sample preparation and test methods. Basically, numerous test standards exist that in many cases are either not in accord with each other or are so only regionally [528-36]. In order to meet the rising demand for information by the economy, industry, and research interests, about four thousand databases are available worldwide. Nearly all supply technical literature, economic information, patent references, and manufacturers' addresses. Materials databases with numerical values are a relatively small part of these programs. Because the majority of these databases are from individual manufacturers of plastics, there is only limited comprehensive, neutral information on most materials. The German federalist ministries of the Economy and of Research and Technology recognized this situation and during the 1980s launched programs to assist in the development of comprehensive factual databases. Within this framework, the Deutsches Kunst-

<.C

\1'1 N

~I

PE

pp

Melts & drips

Other characteristics

...

Slow

...

...

I I Sinks

I All others I

. ..

CTFE

... . .. "

. ..

Burnt Acidic hair

. ..

PVF

ASS Acrylics

Acetal

Slow

Oap

Yellow with sparks

Slow

Some black smoke

Fast

Burnt sugar

Rancid butter

Black Some smoke smoke with soot with soot

Slow

Yellow

Yellow with blue tip

Self-extinguishing

I Urea

Yellow with blue edges

Poly_r

Dense smoke with soot

Fast

Camphor

Pale yellow

Fast

Faint apple

Yellow

Sample Slight Black burns black smoke with soot completely smoke

Fast

ciJ Collula .. Poly· nitrate urothan

Froths

Slow

Blue with yellow tip Burnt wool or hair

Nylon

Polyester

Slow

Phenol

Orange or yellow

carbonate

Poly·

Black Black smoke smoke with soot with soot

Fast

Odor of sulphur

Orange

Poly· sulfone

'Dri';;;"

Difficult to ignite smoke

Slow

Phenol

White smoke

Slow

Hydrochloric acid

~ges

PVC Yellow Yellowish with green orange

PPO

~

PPO Polysulfone PVC

Nylon Polycarbonate

I

Black smoke

Contin:Jes

amine

Pungent

Yellow

Epoxy

to burn

None

Bright yellow

Self-extinguishing

Black smoke

with soot

cinnamon

Sour

I

Silicone

I I Continues to burn I

Yellow with blue edges

®

Swells and cracks

Mayor may not

Swells and be self-exiting

Formaldehyde

formaldehyde Yellow with greenish blue edge

Phenol

cracks

J

Burn a small corner of the sample

I

Thermo.ts

Fish like

Illuminating Burning gas or rubber marigold Fast

I Melamine Phenol formaldehydo formaldehyde Yellow with Yellow blue tip

Poly· styrene Yellow

smoke

Black

of phenol

Faint odor

Vellow

I

Collulose Collulo.. Collulo. acotate acotate butyrate propionate

Formalde· Vinegar hyde

Black No smoke smoke with soot

Slow

Acrid

Blue with Blue yellow edges

ABS

Ves

Other characteristics

Odor

Color of flame

~81

Observation

against the sample

or a hot rod (SOO°F)

Press a soldering iron

I PLASTICS MATERIALSJ I

Acetals Cellulose acetate Cellu lose acetate butyrate Cellulose propionate Cellu lose nitrate Polystyrene Polyurethane

~.-

I

I I Continues to burn I

No

PTFE

Drips

PTFE CTFE PVF FEP

No flames

I

I Burn a small corner of the sample

Burnt Acetic hair acid

...

FEP

Ves

I

Drop a small

::~rlein

Figure 6-52. Identification Procedures for Representative Plastics.

Fast

Blue with Blue with yellow yellow tip tip Acrid or diesel Paraffin fumes

Speed of burning slow < 3 inches fast> 3 per min.

Odor

Color of flame

Observations

~

Floats

I Thermoplastics

Softens

530 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 6-28. Standard Symbols for Generic Families, with Referenced Standards and Cell Tables

Standard Symbol ABS AMMA ASA CA CAB CAP CE CF CMC CN CP CPE CS CTA CTFE DAP EC EEA EMA EP EPD EPM ETFE EVA FEP FF IPS MF PA PAl PARA ... etc.

Plastic Family Acrylonitrilelbutadiene/styrene Acrylonitrile/methylmethacrylate Acrylonitrile/styrene/acrylate Cellulose acetate Cellulose acetate butyrate Cellulose acetate propionate Cellulose plastics, general Cresol formaldehyde Carboxymethyl cellulose Cellulose nitrate Cellulose propionate Chlorinated polyethylene Casein Cellulose triacetate Polymonochlorotrifluoroethylene Poly(diallyl phthalate) Ethyl cellulose Ethylene/ethyl acrylate Ethylene/methacrylic acid Epoxy, epoxide Ethylene/propylene/diene Ethylene/propylene polymer Ethylene-tetrafluoroethylene copolymer Ethylene/vinyl acetate Perfluoro (ethylene-propylene) copolymer Furan formaldehyde Impact styrene Melamine-formaldehyde Polyamide (nylon) Polyamide-imide Polyaryl amide

ASTM Standard

Suggested Reference Cell Tables for Materials Without an ASTM Standard Unfilled

Filled

D--E E D 706 D 707 E E H E E

D D H D

D 1562 F H E

H D

H E F F H

H D

F

D

H

F

H

H

H

H

G

G

(See PS) D4066

stoffinstitut (DKI; German Plastics Institute) established the materials database called Polymat. This program brings greater availability into a plastics market in which a general perspective is becoming increasingly difficult to obtain. This database contains information on plastics, supplying about thirty to fifty properties for each material. At present some six thousand plastics, from about seventy manufacturers, are stored. The concept of the Polymat database was based on the following criteria: 1) the database is neutral, independent of raw-material manufacturers; 2) anyone can use the database; 3) all the products on the European market should, if possible, be included; 4) since testing is carried out in accordance with a variety of different international standards, the relevant standard, as well as the testing conditions, should be registered; 5) during the

Table 6-29. Reinforcement-Filler Symbols and Tolerances Symbol

Material

C

Carbon and graphite fiber-reinforced Glass reinforced Lubricants (i.e., TFE, graphite, silicone, and molybdenum disulfide) Mineral reinforced Reinforced-combinations/mixtures or reinforcements or other fillers/reinforcements

G L

M

R

Tolerance

± 2 Percentage points ± 2 Percentage points By agreement between the supplier and the user

± 2 Percentage points ± 3 Percentage points (based on the total reinforcement)

Table 6-30. Suffix Symbols and Requirements Symbol A

B C

D

E

F

Characteristic Color (unless otherwise shown by suffix, color is understood to be natural) Second letter A = does not have to match a standard B = must match standard Three-digit number 001 = color and standard number on drawing 002 = color on drawing Not assigned Melting point-softening point Second letter A = ASTM D 789 (Fisher-Johns) B = ASTM D 1525 Rate A (Vicat) C = ASTM D 1525 Rate B (Vicat) D = ASTM D 3418 (Transition temperature DSC/DTA) E = ASTM D 2116 (Fisher-Johns-high temperature) Three-digit number = minimum value °C Deformation under load Second letter A = ASTM D 621, Method A B = ASTM D 621, Method B First digit 1 = total deformation 2 = recovery Second and third digit x factor of 0.1 (deformation) = % min. 1 (recovery) Electrical Second letter A = dielectric strength (short-time), per ASTM D 149 Three-digit number x factor of 0.1 = kV/mm, min. B = dielectric strength (step by step), per ASTM D 149 Three-digit number x factor of 0.1 = kV/mm, min. D = dielectric constant at 1 MHz, per ASTM D 150, max. Three-digit number x factor of 0.1 = value E = dissipation factor at 1 MHz, per ASTM D 150, max. Three-digit number x factor of 0.0001 = value F = arc resistance, ASTM D 495, min. Three-digit number = value [Other methods under review, per ASTM D 257 and D 1531) Flammability Second letter A = per ASTM D 635 (burning rate) 000 = to be specified by user B = per ASTM D 2863 (oxygen index) Three-digit number = value %, max .

. . . etc

531

""N

IN

Property

To be determined

("F)

Tensile strength, per ASTM D 638, MPa, min.! (psi) Flexural modulus, per ASTM D 790, MPa, min.! (kipS/in.2) Izod impact per ASTM D 256, JIm, min. 2 Deflection (temperature, per ASTM D 648, 1820 kPa), °C, min.

2J/m x 18.73 x J()-3 = ft.·lbf.lin.

'MPa x 145 = psi.

*Other cell tables are in 0 4000.

5

4

3

2

Designation Order Number

Unspecified

Unspecified

Unspecified

Unspecified

0 Unspecified

130 (266)

160 (320)

2 40 15 (2,100) (5,800) 3,500 600 (87) (510) 15 30 200 (392)

3 65 (9,400) 6,500 (940) 50 230 (446)

85 (12,000) 10,000 (1,400) 135

4

260 (500)

300 (572)

5 6 110 135 (16,000) (19,600) 13,000 16,000 (l,89O) (2,300) 270 425

Cell Limits

Table 6-31. Cell Table G Detail Requirements*

330 (626)

7 160 (23,200) 19,000 (2,800) 670

8 9 185 Specify (26,800) value 22,000 Specify (3,200) value 950 Specify value 360 Specific (680) value

THE PROPERTIES OF PLASTICS 533

Table 6-32. Data Developed Based on ASTM D 4000 Tensile Strength

Tensile Modulus

Material

MPa

ksi

GPa

106 psi

ABS ABS-PC DAP POM PMMA PAR LCP MF Nylon 6 Nylon 6/6 Nylon 12 PAE PBT PC PBT-PC PEEK PEl PESV PET PF PPO PPS PSU SMA UP

41 59 48 69.0 72.4 68 120 52 81.4 82.7 81.4 121 52 69.0 55 93.8 105 84.1 159 41 54 138 73.8 31 41

6.0 8.5 7.0 10.0 10.5 9.9 17.5 7.5 11.8 12.0 11.8 17.6 7.5 10.0 8.0 13.6 15.2 12.2 23.0 6.0 7.8 20.0 10.7 4.5 6.0

2.3 2.6 10.3 3.2 3.0 2.1 11.0 9.65 2.76 2.83 2.3 8.96 2.3 2.3 2.2 3.5 3.0 2.6 8.96 5.9 2.5 11.7 2.5 1.9 5.5

0.33 0.38 1.50 0.47 0.43 0.30 1.60 1.40 0.40 0.41 0.34 1.30 0.34 0.34 0.32 0.51 0.43 0.38 1.30 0.85 0.36 1.70 0.36 0.27 0.80

FlexuaJ Strength

Impact Strength

ft·

MPa

ksi

11m

Hardness, lbf.lin. Rockwell

72.4 89.6 117 98.6 110 82.7 124 93.1 113 110 113 138 82.7 96.5 86.2 110 152 129 245 62 88.3 179 106 55 82.7

10.5 13.0 17.0 14.3 16.0 12.0 18.0 13.5 16.4 16.0 16.4 20.0 12.0 14.0 12.5 16.0 22.0 18.7 35.5 9.0 12.8 26.0 15.4 8.0 12.0

347 560 37 133 21 288 101 16 59 53 64 64 53 694 800 59 53 75 101 21 267 69 64 133 32

6.5 10.5 0.7 2.5 0.4 5.4 1.9 0.3 1.1 1.0 1.2 1.2 1.0 13.0 15.0 1.1 1.0 1.4 1.9 0.4 5.0 1.3 1.2 2.5 0.6

RI03 R1l7 E80 RI20 M68 R122 R80 M120 R119 R121 R122 M85 R117 R1l8 R1l5 R120 MI09 M88 R120 MI05 R1l5 R123 M69 R95 M88

Flame Rating, UL94

HB HB HB HB HB HB VO VO V2 V2 V2 VO

HB V2

HB VO VO VO

HB HB VO VO

HB HB HB

search, all properties should be capable of being linked with one another as desired; and 6) the sources used for the database are the technical data sheets and additional information supplied by raw-material manufacturers, and various lectures, pUblications, and measured data from different institutes. In order for such an extensive project to remain manageable, certain requirements were necessary. Initially the data were confined to TPs, TSs, TPEs, and casting resins. To be included in this group were the TSEs, composites, foams, semifinished products, and others. Polymat'completed its initial work in 1989. New plastics products on the market and updated additional information on existing products will continue to be added. Data no longer available are still accessible to the user in a memory file. Each plastic in this database is first characterized by descriptive data such as its trade name, manufacturer, product group, form of supply, or additives. Then follows complex technical information on each material, with details of fields of application, recommended processing techniques, and special features. The central element of this material database is the numerical values it gives on a wide range of mechanical, thermal, electrical, optical, and other properties. All these items can be searched for individually or in the combination of properties that was the subject of the enquiry.

534 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

A number of material suppliers offer information on their products on floppy discs for use on personal computers. One especially important one, called Campus, is a database concept started by four German material manufacturers who use a uniform software. This database, developed jointly by BASF, Bayer, Hoechst, and Hulls, provides for other manufacturers to join later. It is given in the form of diskettes in German, English, French, Italian, or Spanish. Each diskette contains the uniform test and evaluation program and the range of the respective material producers. It runs on IBM-compatible personal computers under the MS-DOS operating system [528, 536]. In order to understand the possibilities of these two databases, a comparison can be made of a central database like Polymat and Campus as an example of one from manufacturers. The Polymat central database provides the following: 1) all the products of the various firms represented on the European market are included; 2) the search is independent of the manufacturers and can be performed for all the products of all the manufacturers; 3) available are not only the values contained in the list of basic values but other data specified in such standards as DIN, ASTM, and BS (although a search can nevertheless be confined to products whose data conform to the list of basic values); 4) the information is presented only once and is then maintained centrally; and 5) a selection can be made between a greater number of materials and manufacturers. The manufacturers' database, Campus, provides use that is free of charge and no charges for data transmission. The actual value of the table of basic values described in Campus lies in its effect on the standardization and streamlining of testing. In the long term, the nonparticipants in the material market will not be able to remain outside this development. This comparison shows how these two electronic information systems are not comparable, because they pursue completely different objectives. The material manufacturers' databases provide information on products whose manufacturer and product class are already known. Polymat also gives information if the manufacturer and product class are not known, but only if the requirements with regard to the material can be described. Polymat can also be used if a replacement material !s being sought for a product that can no longer be supplied. Furthermore, Polymat is also capable, for example, of answering the question of possible manufacturers of nylon 6, or of how many different nylon 6 grades an individual manufacturer can supply. If only the trade name is known, its manufacturer or distributor in Europe can be traced. This is especially important in the case of foreign products sold by a trading company under the same trade name. When Polymat was being established, it was necessary to decide on whether to charge the material manufacturer but not the user, or make no charge to the manufacturer but charge the user. DKI chose not to charge the user. Neither database generally offers the possibility of integrating into it the greater number of values and test data that may already be established by users or processors. These organizations have data for their own internal use, and their goal has been to integrate all these types of data sources. Such in-house databases are at present available under operating system BS 2000 and in conjunction with the database software known as Adabas. With regard to the common European market, the European Economic Community (EEC) has undertaken numerous activities concerned with materials and material information systems. In one demonstration program for material databases eleven such databases from various countries in the EEC are being cooperatively developed with joint standards for terminology, data presentation, database access, and the user interface of

THE PROPERTIES OF PLASTICS 535

search commands, aids, and menus. For the materials class of plastics, Polymat was selected to participate in this cooperative work. Interesting developments will no doubt occur from which the users of central material databases in the entire EEC area can benefit.

SELECTION WORKSHEETS Selecting an optimal material for a given part must obviously be based on analysis of the requirements to be met. A simplified approach involves comparing the specific service requirements to the potential properties of a plastic. What follows is a simplified but practical material-selection approach. This "longhand" system, which is a basis of the fastest computerized databases, follows these steps: 1) select the design criteria from a worksheet (see Table 6-33) and check off only the major criteria across the worksheet, keeping it simple but realistic; 2) refer to the selection chart (see Table 6-34) and transfer the bold-faced numerical rating in each selected criteria column to the worksheet; for example, if toughness is one criterion, list 6, 4, I, 2, 4, and 2 on the worksheet from top to bottom in the toughness column; 3) add up the numbers across the worksheet in Table 6-33 to the "point subtotal" column to find the resin group with the lowest-point subtotal, which will be the best for a given application on a performance basis; 4) add in the cost factor and total it, to find again the resin group with the lowest number, again the best for the application on a cost-performance basis. Finally, repeat the first four steps, but this time use the small numbers on the selector chart in Table 6-34 and only for the resin group that was found to be the best. The resin with the lowest final total will be the best for the application on a cost-performance basis. Such a selector worksheet can include specifically what the designer requires, with appropriate numerical ratings. Tables 6-35 and 6-36 provide examples of how to use these simplified worksheets in evaluating two different products.

SELECTING MATERIALS The material information and data presented in this chapter and other sections have provided a variety of useful selection guides. The following section elaborates on some of the data seen earlier. These tables and figures present general overall data, ratings, and guides concerning different properties. It should.be remembered that the values given here and elsewhere in this book are representative rather than precise. These values vary depending on the type of material, the manufacturing process, and the condition and method of testing. Thus, for example, the tensile strength of a PC given in one table could be quite different from that in another table. The procedure to follow is to properly identify a plastic, usually by its manufacturer's name, its trade name, the manufacturer's grade or identification listing, and by what its data sheet says about its properties, the goal being to have minimum and maximum values. The data presented in this book are not intended to be used as a substitute for more up-to-date and accurate information on the specific plastics. Such details can be obtained from in-house sources, testing laboratories, and various institutions or the material's supplier. The data here are provided instead as a comparative guide to help in understanding the performance of plastics and in making the decisions that must be made when developing a logical approach to design when compromises have to be made [1, 10-12, 40,62-64,99,249,262,278, 386,444,485,497,519-36].

0\

W

'" Criteria

--

_.-

_.-

_.

---

Strength Short-Teno Long-Teno and Heat Heat Stiffness Toughness Resistmce Resistance

--

-

L ...

Point Subtotal Cost

----

Point Total

Toughness: 1bc ability 10 withstand impacting at high strain rates. Short-tcnn heal resistance: The abilily to withstand CII:POSUrc 10 elevated IcrnpcralUreS for a limited period of lime wilhout distortion. Long·tcrm heal reslslance: The ability to relaID a high level of room-temperature mechanical properties after exposure to elevated temperature for a sustaincd period. Environmental resistance: The ablhty to withstand exposure to solvents and chemicals. Dimensional accuracy in molding: The ability to produce warpfrce, high-tolerancc molded parts. DimensIOnal stability: 'The ability to maintain the molded dimensions ariel exposure to a broad ransc of temperatures and environments. Wear and frictional propenies: The ability of the plastic to resist removal of material when run against a mating metal surface. The lower the frictional values. Ihc better the relatIve rating. Cosl: 'The relative cost per cubic inch.

relative ratings.

Strength and stiffness: The ability to resist instantaneous applications of load while exhibiting a low level of strain. Materials that demonstrate a proportionality between SlIcss and strain have been assigned better

-

Wear and Dimensional Accuracy in Dimensional Frictional Resistance Molding Stability Properties

Environmental

RabngS: I-most desirable; 6-1easl desirable. Large numbers indicate group classification. small numbers the specific: MSins within thai group.

PEP ETFE

Auorocarbons

High Temp. Resins PPS Polyamide-imide

Arylates Modified PPO Polycarbonate Polysulfone Polyethersulfone

Polyester Polyacetal

6 6/6 6/10, 6/12

Other Crystalline Resins Nylons

Olefins Polyethylene Polypropylene

Styrenics ABS SAN Polystyrene

Groups

Resin

G/R

~

Material Characteristics

Table 6-33. Selection Worksheet (Courtesy ICI-LNP)

'I

w

c.n

2 1

I 2

4 2 2

4 5

2 6

3

1

3 5

1 2

2 1

3 1 2 3

2 3 1 4 5

2 1

3

I 2

2 1 2 1

2

4 3 2 1

2 1 3 2 5

2 1

3

I 2

4

2

1

4

6

Toughness

2

3 1

2

4

6

Short-Term Heat Resistance

1 2

2 1

4 3 2 1

2 1 3 1 2

2 1

3

1 2

1

1

3

4

5

6

Long-Term Heat Resistance

1 2

1 2

3 4 2 1

5 4 3 2 1

2 1

3

1 2

2 1

5

4

3

6

Environmental Resistance

2 1

1 2

4 1 2 3

1 2 2 2 3

1 1

1 2

3

6

1 4

4

1 5

Dimensional Accuracy in Molding

Ratings: I-most desiJable; 6-least desiJable. Large numbers indicate group classification, small numbers the specific ",sins within Ibat group.

ETFE

FEP

Fluorocarbons

High Temp. Resins PPS Polyamide-imide

Arylates Modified PPO Polycarbonate Polysulfone Polyethersulfone

2 I

Other Crystalline Resins Nylons 6 616 6110,6112 Polyester Polyacetal

3

2

3

2 1

Strength and Stiffness

Olefins Polyethylene Polypropylene

Styrenics ABS SAN Polystyrene

Resin Groups

2 1

2 1

4 3 2 1

4 3 2 1 2

1 1

3

2 1

2 1 6

4

5 5

Dimensional Stability

Table 6-34. Glass-Reinforced TP Compound Selection Chart (Courtesy ICI-LNP)

1 2

2 1

4 3 1 2•

3 2 3 4 1

2 1

1 2

3

Cost

1

~6

1

:3 4 !4 4 !5

2

1 2

6 ;2 3 !1

Wear and Frictional Properties

\11

W

CD

Criteria

ETFE

Fluorocarbons FEP

High Temp. Resins PPS Polyamide-imide

Arylates ModifiedPPO Polycarbonate Polysulfone Polyethersulfone

6110,6112 Polyester Polyacetal

616

Other Crystalline Resins Nylons 6

Olefins Polyethylene Polypropylene

Polystyrene

SAN

ADS

Styrenics

GIR.. Resm Groups

~

Material Characteristics

5

2 I 3 4

6

2

3

1

5

2 3 I 4 5

2

4

2

1

4

6

X

X 3

Toughness

Strength and Stiffness

5

2 I 3 2

2

1

3

2

4

6

X

Short-Tenn Long-Tenn Heat Heat Resistance Resistance

5 4 3 2 I

1

2

5

4

3

6

X

Environmental Resistance

Dimensional Accuracy in Molding Dimensional Stability

Wear and Frictional Properties Point Total

2 23

Cost

~ I

4 I

3

:~ 11 17

14 13

11

9

6 17

5 14

13 4 17

16

10 12

I! 8

16 1 17

21

Point Subtotal

Table 6-35. Gasoline-powered Chain-saw Housing: Approach Resulting in Selection of Nylon 6/6 or 6

..0

Ion W

Design Criteria

Fluorocarbons FEP EfFE

High Temp. Resins PPS Polyamide-imide

Arylates Modified PPO Polycarbonate Polysulfone Polyethersulfone

Polyester Polyacetal

6/6 6/10,6/12

Other Crystalline Resins Nylons 6

Olefins Polyethylene Polypropylene

Styrenics ABS SAN Polystyrene

Groups

Resin

G/R

Material Characteristics

~

6

2

3

1

5

3

x

Strength and Stiffness Toughness

2

2

1

3

2

4

6

x

Short-Term Heat Resistance

1

2

1

1

3

4

5

6

x

Long-Term Heat Resistance

1 2

1

2

5

4

3

6

x

Environmental Resistance

Dimensional Accuracy in Molding Dimensional Stability

Wear and Frictional Properties Point Total

2 23

Cost

11 10 6 16

~6 ~5 ~

14 4 18

11 3 14

17 1 18

21

Point Subtotal

Table 6-36. Impeller for Chemical-handling Pump: Approach Resulting in Selection of PPS

540 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The information that follows is categorized by certain main subjects; however, some of the data in one specific subject will also come under other subjects as selection guides: Mechanical properties ofunrein/orced and reinforced plastics: See Figures 6-53-59 and Table 6-37-39. Temperature properties: See Figures 6-60-66 and Tables 6-40-48. Chemical properties: See Figure 6-67 and Tables 6-49-51. Other properties: See Figure 6-68 and Tables 6-52-56.

Other parts of this book contain selection guides that may also prove useful, such as some figures in Chapters 1 and 3 and a number of tables in Chapters 2 and 3.

SELECTING MATERIALS UNDER DYNAMIC LOADING Hysteresis Measurement Gives Dynamic Characteristics One objective of the dynamic testing of materials and components, in addition to determining the number of cycles to break, is to establish the causes and pattern of failure. Normally, in such tests the stress and strain (extension) are measured and an elastic modulus derived from these values. However, this can be only the initial step in the direction of developing a complete description of the failure pattern. This analysis is based on the work by F. Orth and G. W. Ehrenstein Kassel [300]. One method that has been developed for measuring hysteresis during a dynamic test allows for four different paraqteters to be determined simultaneously: stresses, extensions (strains), stiffnesses, and the mechanical energy involved. These factors are determined separately for the tension and compression phases, enabling a finely differentiated analysis of the behavior of the material. In addition, the damping factor can be calculated from measuring the stored energy and the energy loss, f<,lr both linear and nonlinear viscoelastic behavior. If the dynamic behavior of a material or a component must be guaranteed, the material and its processing conditions are normally subject to continuous checking to ensure that the properties do not vary. The better way, of monitoring the dynamic properties directly, is confronted with the current high cost of testing. The hysteresis measurement procedure offers the possibility of obtaining the maximum of information on the dynamic behavior of a material or a component in a very short time.

Characteristics: Definition and Calculation To determine the values of material characteristics, the stress and extension signals are plotted against each other. Damping (mechanical loss) in a material causes a phase shift between strain and stress. When the two signals are superimposed, a hysteresis diagram is obtained on which four different characteristic quantities can be defined: stresses, strains, elastic moduli, and mechanical losses. The maximum and minimum stresses and strains can be read straight off the diagram (see Figure 6-69). The upper path of the stress from Eu to Eo is indicated by the upper curve, (1ok(E), and the lower one correspondingly by the curve, (1uk(E). The mean curve (1mk results from these two as (6-49)

THE PROPERTIES OF PLASTICS 541

Special significance attaches to the intersection of the middle curve with the mean value of stress, (6-50) since it defines the origin of the relative coordinate system. The extension at this point is the mean extension Em, which is a measure of the creep of the material under load. During loading the test sample takes in energy; this is not all recovered on unloading. The difference, which is the energy loss Wv , corresponds to the area enclosed by the curve (see Fig. 6-70, left):

(6-51) The stored energy is obtained from the surface area between the middle or mean curve and the mean stress (see Fig. 6-70, right):

Ws

=f


dE

(6-52)

By means of the relative coordinate system, division of the total energy into upper and lower energy (index 0 and u) can be undertaken. The ratio of lost energy to stored energy is the relevant damping A.

A = Wv Ws

u

u

u

(6-53) (6-54)

=

A g

Wv Ws

g g

(6-55)

If the material is linear-viscoelastic, there is a pha~e difference, 8, between the extension and the stress, and the hysteresis diagram takes the form of an ellipse. Equations 6-51, 6-52, and 6-53 can in this case be solved analytically:

Wv

=

Ws =

1T • {1 •

{1 •

E • sin 8

E + cos 8

(6-56) (6-57) (6-58)

where {1 and E are the stress and strain amplitudes, respectively. In equation 6 the energies and the damping are angular functions. On the other hand the definition of energies according to 6-51 and 6-52 has the advantage that it is valid for both linear-viscoelastic materials and nonlinear-viscoelastic materials. Damping-the quotient of energy loss to energy stored-is dimensionless and independent of stress and strain.

to.)

"'"

I.n

0.98 1.04 1.12

20

30

20 30 20 40 10 20 20 35

1.64 0.98 1.04

40 10 20 30 10

1.22 1.35 1.38 1.55 1.22 1.58 1.22 1.35

1.12

1.10 1.22 1.28 1.54 1.63 1.25 1.37 1.23 1.37 1.30 1.26 1.43 1.52 1.04 1.18 1.21

10 20 30 10 30 15 30 13 30 30 10 30 30 10 30 20

90 83 97 124 33 97 100 110

(13,000) (12,000) (14,000) (18,000) (4,800) (14,000) (14,500) (16,000)

1.2 1.8 2.5 1.5 48.0 3.0 1.8 1.4

2.0

3.0

3.0 4.0 3.0 2.0 4.0

152 (22,000) 43 (6,200) 45 (6,500) 47 (6,800) 50-59 (7,2008,600) 57-68 (8,3009,900) 68-83 (9,90012,000)

36 59 100

131

3.0 2.0 1.4 2.4 2.0 4.0 3.0 4.0 3.0 4.0 9.0 2.0 4.0 4.0 3.0 5.0

(9,400) (11,000) (13,000) (10,400) (12,000) (15,000) (24,000) (14,000) (25,100) (19,600) (12,000) (17,500) (19,000) (5,200) (8,600) (14,500)

65 76 90 72 83 104 166 97 173 135 83 121

Tensile Elongation, %,0638

"To convert MPa to psi, mUltiply by 145; to convert GPa to psi, multiply by 145,000. "To convert Jim to ft·lbf/in., divide by 53.38.

SAN

polyurethane PVC

polystyrene high heat copolymer high heat terpolymer polysulfone

polypropylene, chemically coupled

poly(phenylene oxide), modified poly(phenylene sulfide) polypropylene

polyester, thermoplastic polyethylene

nylon-6, 12 polycarbonate

nylon-6,6

nylon-6

acetal

ABS

Plastic

Specific Gravity, 0792

Glass-fiber Content, wt%

Tensile Strength, MPa", (psi) 0 638

8.3 6.5 6.0 11.6 0.7 0.8 8.6 10.4

4.6

3.9

14.1 2.5 3.7 4.4 3.7

4.6 5.1 6.3 6.6 7.7 5.9 7.2 6.2 9.0 8.3 5.2 8.6 8.3 2.5 5.0 6.4

(1,200) (940) (870) (1,680) (100) (116) (1,250) (1,510)

(670)

(570)

(2,050) (360) (540) (640) (540)

(670) (740) (910) (960) (1,100) (860) (1,050) (900) (1,300) (1,200) (750) (1,250) (1,200) (360) (720) (930)

Tensile Modulus, GPa", (kips/in2) 0638

(14,800) (15,500) (16,800) (15,500) (16,500) (23,000) (29,000) (25,100) (34,100) (28,000) (16,000) (20,400) (28,000) (6,700) (12,900) (18,600)

131 123 138 173 43 145 131 155

(19,000) (17,800) (20,000) (25,100) (6,200) (21,000) (19,000) (22,500)

(37,000) (7,800) (8,300) (9,100) (10,40013,600) 81-106 (11,70015,400) 90-131 (13,00019,000)

255 54 57 63 72-94

102 107 116 107 114 159 200 173 235 193 110 141 193 46 89 128

Flexural Strength, MPa", (psi) 0 790

Table 6-37. Guide for Selecting Glass Fiber

(650) (710) (930) (880) (1,050) (780) (1,000) (650) (1,300) (1,100) (590) (1,000) (1,100) (360) (710) (750)

7.9 5.7 5.9 10.7 0.6 6.9 7.6 9.3

4.6

3.7

(1,100) (830) (860) (1,550) (90) (1,000) (1,100) (1,350)

(670)

(540)

13.0 (1,900) 2.4 (350) 3.6 (320) 4.3 (620) 3.5 (510)

4.5 4.9 6.4 6.1 7.2 5.4 6.9 4.5 9.0 7.6 4.1 6.9 7.9 2.5 4.9 5.2

0790

Flexural Modulus, GPa", (kips/in2)

(12,000) (14,000) (15,100) (10,000) (11,700) (14,000) (24,100) (13,500) (27,000) (20,000) (14,000) (17,000) (18,000) (5,100) (5,900) (17,500)

110 76 124 138 35 83 121 45

(16,000) (11,000) (18,000) (20,000) (5,100) (12,000) (17,500) (6,500)

145 (21,000) 41 (5,900) 45 (6,500) 47 (6,800) 43-44 (6,2006,400) 44-47 (6,4006,800) 45--48 (6,5007,000)

83 97 104 69 81 97 166 93 186 138 97 117 124 35 41 121

Compressive Strength, MPa", (psi) 0695

(1.3) (1.2-1.4)

(1.1)

(1.5) (0.81)

(2.00) (2.19) (2.00) (2.40) (1.8) (1.4) (\.7) (1.8)

(0.99)

(1.2) (1.1) (0.99) (0.99) (0.80) (1.5) (2.19)

59 80 64 80 747 80 64 53

(1.1) (1.5) (1.2) (1.5) (14.0) (1.5) (1.2) (0.99)

69-91 (1.3-\.7)

69-80 (1.3-1. 5)

80 43 59 69 64-75

64 59 53 53 43 80 117 53 107 117 107 128 96 75 91 96

Impact Strength, Izod Notched, J/m b 0 256 (ft-lbf/in.)

olio

VI

w

polyurethane PVC SAN

polystyrene high heat copolymer high heat terpolymer polysulfone

polypropylene, chemically coupled

poly(phenylene oxide), modified poly(phenylene sulfide) polypropylene

polyester, thermoplastic polyethylene

nylon-6,12 polycarbonate

nylon-6,6

nylon-6

Acetal

ABS

Plastic

20 20 35

10

20 30 20 40 116 149 179 185 54 82 102 104

142-146 (288-298) 146-149 (295-300)

20 30 (241) (300) (354) (365) (129) (ISO) (216) (219)

(511) (261) (270) (280) (275-286)

266 127 132 138 135-141

40 10 20 30 10

(20S) (210) (212) (255) (325) (385) (399) (475) (486) (390) (280) (289) (415) (230) (255) (289)

98 99 100 124 163 196 204 246 252 199 138 143 213 110 124 143

10 20 30 10 30 15 30 13 30 30 10 30 30 10 30 20

Glass-fiber Content, WI %

Heat-deflection Temp at 1.7 MPa", "C, ("I') D 648

(2.6) (2.4) (2.1) (2.5)

(1.1)

(1.4) (3.0) (2.1) (2.0)

(1.3)

(1.2) (1.8)

(1.3)

(1.5)

(1.5)

(2.3) (2.1) (1.7) (2.9) (2.4) (1.7)

4.0 3.6 2.7 2.3 6.1 4.1 3.8 2.9

(3.4) (2.3) (2.1) (1.6)

(1.5) (1.3)

(2.2) (2.0)

4.1 (2.3) 3.6 (2.0)

2.0 4.7 4.3 3.8 4.5

4.1 3.S 3.1 5.2 4.3 3.1 2.7 2.7 2.3 2.2 3.2 2.3 2.5 5.4 3.8 3.6

Coefficient of Linear Thennal Expansion, 10-" em/cdC (10-5 in.lin. "1'), D 696

(450) (180) (190) (210) (199)

(180) (180) (230) (261) (199) (230) (225) (261) (230) (261) (270) (250) (180) (199) (241)

104 116 163 171 43 66 82 88

(219) (241) (325) (340) (109) (151) (180) (190)

110 (230) 121 (250)

232 82 88 99 93

82 82 110 127 93 110 107 127 110 127 132 121 82 93 116

77 (171)

Maximum Temp, Continuous Use,oC

0.28 0.10 0.2 0.18 0.4 0.09 0.24 0.21

0.04 0.04

0.01 0.05 0.05 0.04 0.05

0.3 0.3 0.2 0.22 0.2 1.8 1.3 1.0 0.9 0.2 0.14 0.12 0.06 0.08 0.06 0.06

Water Absorption, 24h, %, D 570

16.7 16.7

101• 101•

101•

101•

15.7 15.7 19.7 IS.9 15.0 16.5 19.3 19.7

20.0 17.3 17.3 16.5 16.9

101• 10 1• 10 1• 101• 10 1•

101• 101• 101• 101• 1012 1013

17.7 18.3 18.9 20.0 18.9 16.5 16.1 20.9 19.7 19.7 17.3 18.9 23.6 26.8 24.0 16.5

D 149

V/fLm ,

Dielectric Strength, Dry,

10" 10 1• 101• 101• 10 1• 10 15 1015 101• 101• 1013 1015 10 15 10 1• 10 1• 10 1• 10 17

Volume Resistivity, O-cm, D 257

0.003 0.002 0.004 0.002 0.007 0.002 0.002 0.001

0.006 0.006

0.002 0.007 0.006 0.006 0.006

0.003 0.002 0.002 0.006 0.005 0.007 0.004 0.007 0.004 0.004 0.005 0.003 0.003 0.005 0.003 0.003

Mold Shrinkage, em/em, D 955

c.n

:t

·Ultimate strength.

Wood (dry) parallel-to-grain Douglas fir Redwood Southern pine Steel Mild, low-carbon Cable, high strength Concrete Stone Structural, lightweight Brick masonry Aluminum, structural Iron, cast Glass, plate

Material

180* 130* 300* 18,000 25,000*

3,500 3,500* 4,500* 30,000 85,000* 36,000*

200* 150* 300* 30,000 20,000* 10,000*

(24.1) (24.1) (31.0) (207) (586) (248)

20,000

(248)

36,000

36,000 (248) 275,000* (1,896) (1.38) (1.03) (2.07) (207) (138) (69.0)

600

500 450

(1.24) (0.80) (2.07) (124) (172)

(138)

(3.45) (3.10) (4.14)

Shear

3,500 (24.1) 4,500 (31.0) 5,000 (34.5)

Compression

(41.4) (44.8) (58.6)

6,000 6,500 8,500

Tension

Strength (psi) (MPa) (Yield values except where noted)

Table 6-38. Properties of Different Selected Materials

3,500 2,100 4,500 10,000 25,000 10,000

(24.1) (14.5) (31.0) (69.0) (172) (69.0)

29,000 (200) 25,000 (172)

1,700 (11.7) 1,300 (8.96) 1,700 (11.7)

Modulus of Elasticity (E) (ksi) (GPa)

5.5 5.5 3.4 12.8 6 4.5

6.5 6.5

2 2 3

Coefficient of Thermal Expansion COp-I) (I Q-6)

t.I1

t.I1

..

2,560 1,920 1,920 1,280 2,400 1,760 1,600 1,440 2,880 800 320

160 120 120 80 150 110 100 90 180 50 20 40 490 160 30 45 25 62

Aluminum Brick masonry construction Cement plaster Concrete masonry construction, hollow blocks Concrete, stone, reinforced Concrete, structural lightweight, reinforced Earth, sand, loose Earth, topsoil, packed Glass Gypsum board Insulation, rigid Plywood Steel Stone Wood, Douglas fir Wood, oak Wood, redwood Water, fresh

ambient conditions.

·Values in this table are intended to be representative rather than precise. Most material densities vII}' considerably. depending upon type and/or

7,850 2,560 480 720 400 990

640

kg/m3

Ib.lft. 3

Material

Table 6-39. Weights of Different Selected Materials·

~

THERMOSET POLYESTER ALKYD VINYL ESTER NYLON (INCLUDING AROMATICS) POLYIMIDES POLY AMIDE-IMIDES THERMOPLASTIC POLYESTERS (INCLUDING AROMATICS' POLYSULFONES POLYSTYRENE/COPOLYMERS (EXCEPT ABS) EPOXY POL YPHENYLENE SULFIDE PHENOLIC PPOBASED UREAS PVC • COPOLYMERS

Figure 6-53. Overview of mechanical properties.

THERMOSET POLYESTER ALKYD VINYL ESTER NYLON (INCLUDING AROMATICS' THERMOPLASTIC POLYESTERS (INCLUDING AROMATICS' POLYCARBONATES/ ALLOYS POL YSULFONES POLYSTYRENE/COPOLYMERS (EXCEPT ABS) EPOXY POL YPHENYLENE SULFIDE ABS POLYACETAL PHENOLIC PPOBASED POLYIMIOES MELAMINE

THERMOPLASTIC ELASTOMERS POLYMETHYLPENTENE POLYBUTYLENE FURAN SILICONE POLYETHYLENE. COPOLYMERS CELLULOSICS POLYURETHANE PVC • COPOLYMERS FLUOROPLASTIC / COPOLYMERS POLYPROPYLENE PHENOLIC ABS EPOXY POLYSTYRENE THERMOPLASTIC POLYESTERS POLYAMIDE·IMIDES POLYIMIDES ALKYD VINYL ESTER

POL YBUTYLENES THERMOPLASTIC ELASTOMERS FLUOROPLASTICS NYLONS CELLULOSICS POLYTHYLENE • COPOLYMERS POLYURETHANE PVC • COPOLYMERS EPOXY THERMOSETTING POLYESTERS ALKYDS VINYL ESTERS POLYPROPYLENE POL YCARBONATE MELAMINES PHENOLIC POLYIMIDE ALLYLS ACRYLIC ABS PPOBASED POL YARYL ETHOR GLASS-REINFORCED SILICONE POLYSTYRENE COPOLYMERS

"""

VI

••

amm

10

PSI x 103

30

TENSILE STRENGTH

20

Figure 6-54. Tensile strengths of different plastics.

o

-

•

-

__ . i"l"itttFn:n:ttt:n:Z1ZZZizzrzzZZZ/l1d

IlZZlZZZZI

[T"fIIII""'TTUWTWUWUI

-

STANDARD AND NON·GLASS REINFORCED GRADES

GLASS REINFORCED GRADES

-

IZlll1lZlllZmI

40

50

3-50

9.5

0.35-1.0 3.0 ..... 5 3.8.....4 3.5.....0 0.65-3.0

4~.5

5.0-11 0.2-11 1.9-9 9.0 0.6-5.5 7.5

6.0-11 5.0-8.7

ACRYLIC POLYURETHANE CELLULOSICS GLASS REINFORCED POLYETHYLENE POLYETHYLENE AND COPOLYMERS POLYARYL ETHER GLASS REINFORCED SILICONE SILICONE FURAN POLYBUTYLENE POLYMETHYLPENTENE THERMOPLASTIC RUBBER

ALLYLS

GLASS REINFORCED FLUOROPLASTICS FLUOROPLASTICS AND COPOLYMERS PVC AND COPOLYMERS GLASS REINFORCED PVC

MELAMINE-FORMALDEHYDE

UREA-FORMALDEHYDE

POLYPROPYLENE

5.5-13.6 5.0-10.5 5.0-13.0 12 2.0· 7.4 0.5-12

~:~~~7

PHENOLIC GLASS REINFORCED PPO BASED PPO BASED

POLYACETAL

GLASS REINFORCED POL YPHENYLENE SULFIDE POLYPHENYLENE SULFIDE GLASS REINFORCED ABS ABS

EPOXY

POLYSTYRENE AND COPOLYMERS (EXCEPT ABSI

GLASS REINFORCED POLYSULFONES POL YSULFONES

POL YCARBONATE AND ALLOYS

THE RMO PLASTIC POL YESTE RS

POL YIMIDES AND POLY (AM IDE·IMIDESI

NYLONS INCLUDING AROMATICS

THERMOSET POLYESTER, ALKYD, VINVL ESTER

5.0-18 3-12 14.5-17 6.5-12

2-13 19.5 13.4 8.5-19 2.5-8 6.9-18.5 8.5-10

5-20

1.5-12.0

9-20

17-23 10.2-12.2

2.6-26 10.8-25 2.5-12 9.6-25 7.5-9.5

1.3-28 7.4-14

0.5-13

I»

111

..

10

2.0

psi x 103

Figure 6·55. Compressive strengths of different plastics.

0

trllIIIIIIIIII

...... --

..",.

.-

STANDARD AND NON-GLMS REINFOIICED GRADES _ _ GLMS REINFOIICED GRADES

so

NDBREAK

I

1

70

2.0-6.8 8-6.8

7


17.6-17.11 16-16.4 10-1. 10-13 1.7-10 10

~:=~:

14-21 12.11 2.0.0 8.4--11

21.0

1 22.0 13.1

~~8:~2.& I 0.1-22

1.7--40 32.& 1.1-38 211-3& I 2.0-32 r 18.2-30 I 8.8-2.0 r 12.8-30 111-21.4 13-22 I 4-17 r

~I

2&--4&

:::1

18-70 I 10-38 r 111-&0 I 12-38 r

GLASS liFO POL VETHVLENE • COPOLYMERS POLVETHVLENE • COPOLVMERS POLVMETHVLPENTENE

POLVPRDPVLENE

GLASS RFD SILICONE FURAN FLUOIIOPLASTICS. COPOLVMERS GLASS RFD FLUOROPLASTICS

PPOBABED

POL VACETAL 110% DEFI

PVC • COPOLYMERS GLASS RFD PVC SULFONES IVIELDI GLASS liFO SULFONES IVIELDI GLASSIIFDPOLV· PHENVLENE SULFIDE GLASS liFO POLVCARBONATE POLVCARBONATE POLVURETHANE ACRVLlC

Aas

THERMOPLASTIC POLVESTER NVLONS INCLUDING AROMATICS POLVSTVIIENE • COPOLYMERS IEXCEPT ASSI

POLVIM IDE • POLVAMIDE·IMIDE GLMS RFD POLVIM IDE CELLULOSICS ALLVLS

EPOXV

THERMOSET POL VESTER ALKVD. VINVL ESTER MELAMINE· FORMALDEHVDE UREA·FORMALDEHVDE

PHENOLIC

THERMOPLASTIC RUBBER

\,Q

""

V1

-

_a

PSI x 103

Figure 6·56. Flexural strengths of different plastics.

o

--

STANDARD AND NON·GLASS REINFORCED GRADES

• ,umfiiur22""""

-

rJ1IIlI'

--

_

IIDZIIIZZI GLASS REINFORCED GRADES NO BREAK

f

ABS POLYSTYRENE. COPOL YMERS EXCEPT ABS MELAMINE·FORMALDEHYDE

27 12-14 } lo·!:ftl 14-23 } 11-16 18.5-20 } 12.8-17.5 7-19 0.7-19 10-18 4.2-18 13.5 1.8-18 10-14 7-11 } 5-9 11 11 2-7 10.7 5.5-9.3 0.8--9.0 4 .. 6.5

POLYARYL ETHER GLASS REINFORCED POLYETHYLENE POLYETHYLENE. COPOLYMERS GLASS REINFORCED FLUOROPLASTICS FLUOROPLASTICS FURAN POLYMETHYLPENTENE

POLYPROPYLENE

ACRYLIC POLYURETHANES UREA·FORMALDEHYDE PVC. COPOL YMERS GLASS REINFORCED PVC CELLULOSICS GLASS REINFORCED SILICONE

PPO BASED

POLYACETALS

SULFONES GLASS REINFORCED POL YPHEYLENE SULFIDE POL YPHENYLENE SULFIDE

23-30 } 15.4-18.6 28 20 16-28 } 13.. 14

THERMOPLASTIC POLYESTERS INCLUDING AROMATICS POLYCARBONATE GLASS REINFORCED POLYCARBONATE ALLYLS

NYLONS INCLUDING AROMATICS

PHENOLIC GLASS REINFORCED POLYIMIDE POLYIMIDE. POLY(AMIDE·IMIDEI

13-38 } 6.6-18 13.5 15-32 ll-fi } 6-1

lH:fB.6 49.6 7.1-30.7 18-41 I 5-281

THERMOSET POLYESTERS. ALKYDS. VINYL ESTERS EPOXY

10-80 } 1-24

THERMOPLASTIC RUBBERS 7-80} 6-23

20 FLEXURAL MODUWS (0Pa)

J

15r~~···~··:··:··:..:..~..__________~P:H~::O:U:C____________________ ••••••

10 r.......

... ...PET ...

PAil ........... ....

........ .

...... .-..

.•.•.•.....•.•••••.•...•• :~:::::::::::::::::::::::::::::::::

50

150

200

70 60 ";;;

0. 50

0

~~ ";;; I

C.c.

QJ-

40

Glass Reinforced

30

• Mg

45,. GRPEt. -:l5,. Gil PET

.3O'f. Gil ~h c 20 Unreinforced ( PA 66-33'f.-GR-P~ ~ • 3m. GItPBT ~ Vi ~ Acetal-----10 :-PBTI ....!.-,A8s y O~__~~~~~----~--~~~--~~--~~ 100.000 10.000 1.000 100

-0>

C, ·-

Flexural MOdulus -

1Q-l psi

Figure 6-58. Example of strength versus stiffness for unreinforced and reinforced TP composites and die-cast metal. 550

-

I.n I.n

2

I

•

I

I

• •

I 10

,. 1.

18

FT·LBS/INCH OF NOTCH

12

20

GLASS REINFORCED GRADES

I/llIJ1l1lI1l

Figure 6·59. Impact strengths of different plastics.

0

I-

I-

rIIIP

dB

I

--

..........-

-

-

/llllZZI

STANDARD AND NON-GLASS REINFDRCED GRADES

_

22

~

24

~

28

~

28

30

..l...-....J

I ACRYLIC

0.3-15 2-12 1-2.4 3-10 2.3 8 0.3·-8 0.25-8

004-4.5

1.3-1.6 1.1-1.6

0.3-1.2 0.25-0.4

0.6-1.6 0.23-1

104

0.7-2.3 0.8.. 1.8 2.2

UREA FORMALDEHYDE

1 POLYMETHYLPENTENE

THERMOPLASTIC POLYESTERS

POL YPHENYLENE SULFIOE GLASS-REINFORCED PPS SULFONES

POLYACETAL

PPOBASED GLASS REINFORCED PPO BASED POLYARYL ETHER GLASS REINFORCED SILICONE POLYSTYRENE AND COPOLYMERS (EXCEPT ABSI

1 ABS

1 ALL YLS

POL YAMIDE·IMIDE & POL YIMIDE

gj:~\

GLASS REINFORCED POLYIMIDE

0.26-2.6

PHENOLIC

MELAMINE·FORMALDEHYDE

IpOLYCARBONATE

POLYPROPYLENE

ALKYDS, VINYL ESTERS

0.3-18 0.2-8 17

t.z~6 g:~4~~

I

0.6-20 1-6

1THERMOSETTING POLYESTERS,

I

FLUOROPLASTICS GLASS REINFORCED FLUOROPLASTICS NYLON INCLUDING AROMATICS GLASS REINFORCED NYLON CELLULOSICS, POLYETHYLENE AND COPOLYMERS GLASS REINFORCED POLYETHYLENE POLYURETHANE, PVC AND COPOLYMERS GLASS REINFORCED PVC

1 EPOXY

2.0-30 0.3-5.0 0.5-30 0.2->70

1.1 OA-NB 1.0

2.S-NB 8.0 0.8··NB 2.2-304 0.6-NB

NO BREAK (NBI POLYBUTYLENE, THERMOPLASTIC RUBBER

400

225-350 310-320 250-320 250-320 180-310 250-300 300

Figure 6-60. Overview of temperature properties.

NVLONS THERMOPLASTIC POL VESTER POLVARVLETHER POLVMETHVLPENTENE POL VCARBONATE ALKVOS CHLORINATED POL VETHVLENE

-- -- -- -- -- --

300-400 150-400 200-350 400

340-400

~

300-450 212-450

ALLVLS PHENOLICS POLVETHERSULFONES POLVSULFONES MELAMINES THERMOSET POLVESTER UREAS ACRVLATE RU8BER FLUOROSILICONES

400-600 300-550 520-545 175-500 500

_ _ _ _ _ _ __

_OF 500-800

FLUOROPLASTICS POL V AMIDE·IM IDE EPOXV POL VPHENVLENE SULFIDE

...!!!-1fQ..N~

POLVIMIDES

>300o F 11490 C)

HIGH TEMPERA TURE

I

I

PPO BASED POLVPROPVLENE POLVURETHANE VINVLS POLVBUTVLENE ACETALS ABS80SAN POL VSTVRENE ABSIPOL VCARBONATE ALLOV ACRVLICS CELLULOSICS POLVETHVLENE & COPOLVMER

STANDARD TEMPERATURE 75 - 3000 F (25 -1490 F)

125-205 190-200 140-200 120-200 120-180

160-220

205-250 190-250 150-230 225 195-220

170-260

CONTINUOUS OPERATING TEMPERATURE

I

CHLORINATED POLVETHVLENE POL VURETHANE FLUOROSILICONE SILICONE FLUOROPLASTICS

« 2SOC)

LOW TEMPERATURE

< 750 F

I

-90 -180 -300

-65

-60

Time. h Figure 6-Cil. Heat-resistance properties of resins retaining 50 percent of properties obtainable at room temperature with resin exposure and testing at elevated temperature. Zone 1: Acrylics, cellulose esters, LOPE, PS, PVC, SAN, SBR, UF, etc. Zone 2: Acetals, ABS, chlorinated polyether, ethyl cellulose, EVA, ionomer, PA, PC, HOPE, PET, PP, PVC, PUR, etc. Zone 3: PCTFE, PVOF, etc. Zone 4: Alkyds, fluorinated ethylene-propylene, MF, polysulfone, etc. Zone 5: TS acrylic, OAP, epoxy, PF, TS polyester, PTFE, etc. Zone 6: Parylene, polybenzimidazole, silicone, etc. Zone 7: PAl, PI, etc. Zone 8: Plastics in R&O etc. Since plastics compounding is rather extensive, certain basic resins can be modified to meet different heat-resistance properties.

1_ '!..,..--

Figure 6-62. Classifying plastics by range of continuous or rather long-time heat exposure. 553

Figure 6-63. Basic guide to flash ignition and self-ignition temperatures, per ASTM D 1929.

9.0

SPECIMENS NO LONGER BREAK

1.0 -i-7.0

"

...~

... 6.0

iis

i

IMMCT STYRENE

5.0

POLYSTYRENE

Figure 6-64. Example of Izod unnotched-impact strength versus temperature. 554

555

Table 6-40. Thermal Properties of Different Plastics Thennal Conductivity

Material Acrylonitrile-butadiene-styrene (ABS) ABS-polycarbonate alloy (ABS-PC) Diallyl phthalate (DAP) Polyoxymethylene (POM) Polymethyl methacrylate (PMMA) Polyarylate (PAR) Liquid crystal polymer (LCP) Melamine-formaldehyde (MF) ~ylon 6 Nylon 616 Amorphous nylon 12 Polyarylether (PAE) Polybutylene terephthalate (PBT) PC PBT-PC PEEK Polyether-imide (PEl) Polyether sulfone (PESV) PET Phenol-formaldehyde (PF) Unsaturated polyester (UP) Modified polyphenylene oxide alloy (PPO)

°C

op

W/m·k

Btu . in.lhr.. ff . op

60 60 130 85 90

140 140 265 185 195

0.27 0.25 0.36 0.37 0.19 0.22

1.9 1.7 2.5 2.6 1.3 1.5

220 130 75 75 65 160 120 115 105 250 170 170 140 150 130 80

430 265 165 165 150 320 250 240 220 480 340 340 285 300 265 175

0.42 0.23 0.25 0.25

2.9 1.6

200 140 80

390 285 175

0.20 0.25 0.22

1.7 1.5

0.17 0.25 0.12

1.7 0.8

0.17 0.26

1.8

(mod)

Polyphenylene sulfide (PPS) Polysulfone (PSU) Styrene-maleic anhydride terpolymer (SMA)

Table 6-41. Tensile Strength versus Temperature

Engineering Polymer

UL Temperature Index

PPS PES Nylon 6/6 PET P-Sulfane PAl Phenolic-GP Phenolic-HR Phenolic-Glass

220 180 130 140 150 200 150 160 170

556

Tensile Strength at 23°C (73"F) x 1,000 psi (MPa) 23.2 (160) 22.8 (157) 31.0 (214) 19.0 (131) 17.4 (120) 27.5 (190) 8.0 (55) 8.5 (59) 10.0 (69)

Percent Retention of Tensile Strength at 100°C (212 oF)

1500C (302 "F)

200°C (392 oF)

48.3 85.5 51.3 38.1 86.1 72.8 83.7 71.9 75.1

34.9 57.7 39.7 21.0 13.3 59.1 51.2 57.6 59.7

4.7 13.7 7.2 29.9 50.1 52.0 69.9

Table 6-42. Flexural Strength vs. Temperature

UL Engineering Polymer

Temperature Index °C

Nylon 6/6 Nylon 6/6 PET PET Phenolic-GP Phenolic-HR Phenolic-Glass

115 120 150 155 150 160 170

Flexural Strength at 23°C (73°F) x 1,000 psi (MPa) 17.4 35.8 29.2 37.6 8.8 10.2 15.9

Percent Retention of Flexural Strength at

(120) (247) (201) (259) (61) (70.3) (110)

100°C (212 "F)

150°C (302 "F)

200"C (392 oF)

36 51 46 51 67 77 82

29 41 30 33 62 64 70

23 36 24 26 42 48 54

Table 6-43. Heat-Deflection Temperature vs. Temperature Test Result at Deflection Temperature

ASTM D 648 Deflection Temperature at 264 psi (1.82 MPa) °C ("F)

Material P-Sulfane PBT PET PEl PAS PPS Phenolic-HR Phenolic-Glass

175 200 210 210 215 260 210 270

Maximum Stress Load psi (MPa) 1,050 525 1,050 1,050 275 275 2,100 2,100

(347) (392) (410) (410) (419) (500) (410) (518)

(7.2) (3.6) (7.2) (7.2) (1.90) (1.90) (14.5) (14.5)

Degrees Deflection in 4 Hours 5 9 8 8 Failed 4 4

Test Results at Deflection Temperature Plus 10° Maximum Stress Load psi (MPa) 275 275 525 275

(1.90) (1.90) (3.62) (1.90)

275 (1.90) 2,100 (14.5) 2,100 (14.5)

Degree Deflected! When Occurred Failed!<1 Failed!<2 Failedl<4 Failed!<1

hour hours hours hour

Failed!<2 hours 7°/4 hours 2°/4 hours

Table 6-44. Tensile Strength of TP Composites at Elevated Temperatures Tensile Strength (103 psi) at Temperature Base Resin

200"F (93.3)

300"F (149)

350"F (176)

4OO"F (204)

4500F (232)

500"F (260)

GLASS-FIBER REINFORCED PES 30 23.0 PEl 30 25.8 PPS 23.0 40 PEEK 30 25.0 HTA 30 19.5 PFA 20 6.2 PEK 30 26.4

16.0 19.0 11.8 17.5 15.2 4.3 19.7

13.8 13.7 9.0 10.2 12.2 3.2 14.5

11.0 10.7 6.7 8.0 11.1 2.5 8.9

7.0 6.7 5.2 6.3 10.0 2.3 7.3

X X 2.5 5.4 6.7 1.8 6.7

X X X 4.5 X 1.1 5.2

CARBON-FIBER REINFORCED PES 30 26.0 PEl 30 34.0 PPS 30 26.0 PEEK 30 30.6 PEEK 40* 35.5 HTA 30 21.5 5.0 PFA 10

18.2 22.6 13.3 25.2 29.0 16.8 3.1

15.4 16.3 10.2 15.2 17.4 13.5 2.4

12.5 12.4 7.6 9.8 11.2 12.3 1.8

8.1 8.0 5.9 7.9 9.0 11.0 1.7

X X 2.8 6.8 7.7 7.5

X X X 5.4 6.0 X 1.0

x

Fiber (wt%)

73°F (23°C)

1.3

= No effective strength.

'Long·fiber (Verton) composite (ICI-LNPJ.

557

Table 6-45. Flexural Modulus of TP Composites at Elevated Temperatures Tensile Strength (106 psi) at Temperature 73°F (23°C)

200°F (93.3)

300°F (149)

350°F (176)

400°F (204)

450°F (232)

500°F (260)

GLASS-FIBER REINFORCED PES 30 1.260 PEl 30 1.380 PPS 40 1.900 PEEK 30 1.500 UTA 30 1.250 PFA 0.520 20 PEK 30 1.530

1.150 1.200 1.500 1.450 1.150 0.110 1.480

1.120 1.150 1.000 0.950 1.130 0.075 1.000

1.100 1.120 0.750 0.550 1.100 0.065 0.650

0.400 0.420 0.450 0.350 1.050 0.060 0.500

X X 0.250 0.350 0.450 0.055 0.475

X X X 0.350 X 0.030 0.450

CARBON-FIBER REINFORCED PES 30 2.050 PEl 30 2.500 PPS 30 3.100 PEEK 1.950 30 PEEK 40* 4.400 UTA 30 2.000 PFA 10 0.530

1.880 2.160 2.400 1.720 3.900 1.900 0.115

1.800 2.050 1.600 1.300 2.800 1.800 0.080

1.780 2.000 1.230 0.750 1.600 1.750 0.070

0.740 0.750 0.750 0.480 1.020 1.700 0.065

X X 0.400 0.480 1.020 0.850 0.055

X X X 0.480 1.020 X 0.035

Base Resin

Fiber (wt%)

x = No effective strength. 'Long-fiber (Yerton) composite (ICl-LNP).

Table 6-46. Shear Strength of TP Composites at Elevated Temperatures Shear Strength (l03 psi) at Temperature 73°F (23°C)

200°F (93.3)

300°F (149)

350°F (176)

400°F (204)

450°F (232)

500°F (260)

GLASS-FIBER REINFORCED PES 30 11.6 PEl 30 14.0 PPS 40 12.0 PEEK 14.6 30 UTA 10.4 30 PFA 20 3.8 PEK 16.0 30

9.9 13.4 10.3 11.8 9.0 3.2 12.9

8.5 11.0 10.2 8.8 7.9 2.8 9.5

7.6 9.0 9.3 6.8 7.2 2.0 8.7

4.8 6.2 5.0 6.0 5.8 1.7 8.1

X X 2.5 5.0 3.5 1.6 7.8

X X X 4.0 X

CARBON-FIBER REINFORCED PES 30 10.5 PEl 30 13.1 PPS 30 9.5 PEEK 30 12.4 PEEK 40* 15.5 UTA 30 9.5 PFA 10 2.8

8.6 12.5 8.2 10.1 12.6 7.2 2.4

7.3 10.0 8.0 7.5 9.4 7.0 2.1

6.6 8.4 6.5 5.8 7.3 6.5 1.5

4.1 5.8 4.0 5.1 6.4 5.0 1.3

X X 2.0 4.2 5.3 2.3 1.2

X X X 3.4 4.1 X 1.1

Base Resin

Fiber (wt%)

x = No effective strength. 'Long-fiber (Yerton) composite (ICI-LNP). "To convert psi to pascals (Pa), multiply by 6.895 x

558

UP.

1.5 6.7

Table 6-47. Tensile Stress Relaxation of TP Composites at Elevated Temperatures Decrease in Applied Stress (%) with Time* at Temperature Base Resin

Glass (wt%)

73"F (23°C)

200"F (93.3°C)

300"F (149)

35O"F (176)

4OO"F (204)

PES PEl PPS PEEK PEEK HTA PEK

30 30 40 30 40t 30 30

7/8/9 7/9111 3/5/9 13/14/16 19/21125 71718 12113/15

20/21125 13/16125 20/21122 17/21123 21123/27 14/16122 15/18/20

33/35/39 32134/38 26127/28 25/28/30 29/32137 23/27/35 18/21124

35/40/57 34/39/55 26128/32 28/32135 33/35/40 30/35/50 23/25/29

61n4/90 58/69/86 26133/34 30/33/40 35/37/42 39/47/59 26127/30

45O"F (232)

500"F (260)

XlXIX XlXIX XlXlX

XIXIX XlXIX XlXlX

32138/40 36138/43 45/63/55 27/28/31

32138/41 38/39/44 XlXlX 28/29/32

'Thm: values indicate pen:ent stIess relaxation for I h, S h, and IS h. Example: 7/8/9 indicates 7% at I h, 8% at S h, and 9% at IS h. tLong-carhon (Verton) composite (ICI-LNP). X indicates sample would not sostain the test load. Initial stIess for all tests was 2,SOO psi.

How They Rank in Stress Relaxation Base Resin

Glass (wt%)

PES PEl PPS PEEK PEEK HTA PEK

30 30 40 30 40* 30 30

Temperature ("F) 73 (23°C)

200 (93.3)

300 (149)

350 (176)

400 (204)

450 (232)

500 (260)

3 4

6 5 3 4 7 2

6 5 2 3 7 4

7 6 2 3 4 5

X X 2 3 4 5

X X X 2 3 4

X X X 2 3 X

6 7 2 5

The lower the number, the higher the retained stIess at the indicated temperature. 'Long-carbon (Verton) composite (ICI-LNP).

Table 6-48. Polymeric Materials for Extreme-Temperature Applications Polymer Polyphenyls Polyphenylene oxide Polyphenylene sulfide Polybenzyls; polyphenethyls Parylenes (poly-p-xylylene) Polyterephthalamides Polysulfanyldibenzamides Polyhydrazides Polyoxamides Phenolphthalein polymers Hydroquinone polyesters Polyhydroxybenzoic acids

Comments Decompose at 530°C (986°F); infusible, insoluble polymers. Decomposes close to 500°C (932"F); heat cures above 150°C (302"F) to . elastomer; usable heat range - 135 to 185°C ( - 211 to 36s"F) Melts at 270 to 315°C (578 to 599°F); cross-linked polymer stable to 450°C (842°F) in air; adhesive and laminating applications. Fusible, soluble, and stable at 400°C (752"F); low molecular weight. Melt above S20"C (968"F); insoluble; capable of forming films; poor thermal stability in air; stable to 400 to 525°C (752 to 977"F) in inert atmosphere. Melting points up to 455°C (851 "F); fibers have good tenacity, elongation, modulus. Melting points up to 330"C (626"F); soluble; good fiber properties. Dehydrate at 200"C (392"F) to over 4OO"C (752"F) to form polyoxadiazoles; good fiber properties. Some melting points above 4OO"C (752"F); give clear, flexible films. Melting points of 300"C (572"F) to over 400"C (752"F); formable into fiber and film. Soluble polymers with melting points of 335°C (635"F) to over 400°C (752"F). Films melt at 380 to 450"C (716 to 842"F); stable to oxidation but not to hydrolysis; tough, flexible films; good thermal stability. (cont'd)

559

Table 6-48. (Continued) Polymer Polyimides Polyarylsiloxanes Carboranes Polybenzimidazoles Polybenzothiazoles PolyquinoxaIines Polyphenylenetriazoles Polydithiazoles Polyoxadiazoles Polyamidines Pyrolyzed polyacrylonitrile Polyvinyl isocyanate ladder polymer Polyamide-imide Polysulfone Polybenzaylene benzimidazoles (pyrrones) Polybenzoxazoles Ionomer Diazadiphosphetidine Phosphorous amide epoxy Pbosphonitrilic Metal polyphosphinates Phenylsilesesquioxanes (phenyl-T ladder polymers)

560

Comments Commercial film, coating, and resin stable up to 600"C (l,1l20f'); continuous use up to 300"C (S72Of'). Good thermal stability 400 to SOO"C (752 to 9320f'); coatings, adhesives. Stable in air and nitrogen at 400 to 4SO"C (752 to 8420f'); elastomeric properties for silane derivatives up to S38"C (l,OOO"F); adhesives. Detelopmentallaminating resin, fiber, film; stable 24 hours at 300"C (S72Of') in air. Stable in air at 600°C (l,1l2°F); cured polymer soluble in concentrated sulfuric acid. Stable in air at SOO"C (9320f'); tough, somewhat flexible resins; make film, adhesive. Thermally stable to 400 to SOO"C (752 to 9320f'); make film, fiber, coatings. Decompose at S2S°C (9770f'); soluble in concentrated sulfuric acid. Decompose at 450 to 500°C (842 to 9320f'); can be made into fiber or film. Stable to oxidation up to SOO"C (9320f'); can make flexible elastomer. Stable above 900"C (l,62SOf'); fiber resists abrasion with low tenacity. Soluble polymer that decomposes at 38S"C (72SOf'); prepolymer melts above 405°C (76 I Of'}. Service temperatures up to 288°C (SSOOf'); amenable to fabrication. Thermoplastic; use temperature -102"C (-IS20f') to greater than 150°C (3020f'); acid and base resistant. Thermally stable to 600"C (l,1l20f'); insoluble in common solvents; good mechanical properties. Stable in air to 500°C (9320f'); insoluble in common solvents except sulfuric acid; nonflammable; chemical resistant; film. High melt and tensile strength; tough; resilient; oil and solvent resistant; adhesives, coatings. Thermoplastic up to 350°C (6620f'); tbennosetting at 357"C (7070f'); cured material has good thermal stability to SOO"C (9320f'); amenable to fabrication. Soluble B-staged material; amenable to fabrication; good thermal stability. Retention of properties in air up to 399"C (7S00f'). Polymers stable to better than 400"C (7S20f'). Soluble; bigh molecular weight; infusible; improved tensile strength; bigh thermal stability to 525°C (977"F) in air; film forming.

1.0

7.0

i

6.0

~

z

:::::

III

~.O

oJ

~

... ... ... u :! ~

4.0

IE

Z

~

3.0

2.0 HIGH DENSITY I'OLYEllM.EJ«

1.0

0.0 Z20

110

140

100

10

20

-20

-10

-100

TEMPERA~I"1

Figure 6·66. Example of Izod notched-impact strength versus temperature.

CHEMICAL RESISTANCE

GOOD

FLUOROPLASTICS POLYIMIOES POL YOLEFINS ACETALS POL YPHENYLENE SULFIOE ALLYLS EPOXIES IONOMERS POLYAMIDE-IMIOE

POLYESTERS ITP 81 TSI SILICONES PHENOLICS POLYSULFONES NYLONS VINYLS POLYURETHANES ACRYLICS ALKYDS STYRENE-ACRYLONITILE AMINOPLASTICS POLYARYL ETHER POL YARYL SULFONE

STYRENICS CELLULDSICS POL YCARBONATE

Figure 6-67. Overview of chemical properties. 561

~

""N

r

_

I

I

I

LIMITED

I

I

Figure 6·68. Overview of other properties,

EPOXY FURAN PHENOLIC POL YAMIOE·IMIDE POLYIMIDE POLYPHENYLENE SULFIDE SILICONES

II

THERMOSETTING POLYESTER POLOLEFINS POL YSTYRENES POLYURETHANES THERMOPLASTIC ELASTOMERS UREAS VINYLS VINYL ESTERS

_

I

COLORABILITY

UNLIMITED

I

ABS ACETAL ACRYLIC ALKYD ALLYL CELLULOSIC FLUOROPLASTICS MELAMINES NYLONS ..,.. ......... ....

I

I

LOW FRICTION < 0.8 ON STEEL

ABS POLYACETALS SOME NYLONS eOLYSULFONES THERMOPLASTIC POLYESTERS FLUOROPLASTICS GRAPHITE, MOLYBOENUM DISULFIDE AND FLUOROPLASTIC FILLED COMPOUNDS

I

I

I

FLUOROPLASTICS

I

J

HIGH FREQUENCY HIGH VOL TAGE POLYETHYLENE AND CCPOI. YMERS CHLORINATED PVC IHIGH VOl TAGE) POLYSTYRENE ANDCOPOLVMERS POlYPROPYLENE OLEFINIC THERMC»LASTIC RU88ERS AROMATtC POLYESTERS

MC.T PLASTIC MATERIALS

lOW VOLTAGE

I

SULFIDE

POL YPHENLENE

POLYETHYLENE & Cc::.Ot.YMERS POLY IMIDE MELAMINES

r ......... "

I INHERENTLY-V-O

I

J

70-_

.. -

200

230 05-200

135-_

50-310

NO TRACK 15-420

I

--.~

POLYESTER (TP & TS) POLYOLEFINS STYRENICS POL YURETHANES SILICONES UREAS

_....

WITH MODIFICATION OR COMPOUNDING

.1

I LOWEST OENSITIES

< 1.0 UNFILLEO

I

I

0.93 0.93-0.96

0.92~.95

0.910-0.940

0.896-0.899 0.908-0.917

0.89~.91

0.88~.90

0.83-0.84

NYLON POLYSULFC THIN SECTION POL YETHYLENE POLYPROPYLENE POLYVINYLS POLYBUTYLENES IONOMER POLYMETHYLPENTENE FLUROPLASTICS

THICK SECTION> 0.100 IN. ACRYLIC POLYCARBONATE POL YSTYRENE CELLULOSSICS SAN POL YESTE ~ - . __ .

.)

R SULFONE

TRANSPARENCY (COLORLESS)I

I

POLYMETHYLPENTENE POL YOLE F IN TPR POLYPROPYLENE POLY (ETHYLENE CO PROPYLENEI POLYBUTYLENE POLYETHYLENE (INCLUDING UHMWPE) POLY (EHTVlENE CO VINYL ACETATE) POLY (ETHYENE CO ETHYL ACRYLATE) IONOMER

l

ABS ACRYLIC ALKYD CELLULOSE ACETATE ALLYLS EPOXY

I

FLUOROPLASTICS IONOMER MELAMINES POL YPHENYLENE SULFIDES POLYSULFONES VINYLS POLYIMIDES POL YAMIDE·IMIOES

I

FLUOR~LASTICS

CELLULOSICS

~

I

NON-FLAMMABILITY

I

I

ARC RESISTANCE >200 SEes

ACRYLICS ALKYDS

I

ELECTRICAL PROPERTIES

'?C. LOW FREOUENCY

I

I

NYLONS POLYURETHANES POLYACETALS ACRYLIC/PVC ALLOY ABS/POLYURETHANE ALLOY POLYSULFONES PPO BASED THERMOPLASTIC POLYESTER POLYIMIDES SOME POLYSTYRENES FLUOROPLASTIC COPOl YMERS SOME FLUOROPLASTICS UHMWPE GRAPHITE FILLED COMPOUNDS

I

MISCELLANEOUS PROPERTIES

ABRASION RESISTANCE 10,000 MG/KC

I<

I

THE PROPERTIES OF PLASTICS 563

Figure 6·69. Definitions of stress and extension (strain) on a hysteresis loop. Load-dependent stiffnesses can be defined by way of the mean curve (see Fig. 6-71). The compression stiffness at minimum load emerges at the lower end of the diagram and, analogously, the mean- and tensile stiffnesses at points O'm and 0'0 for mean and maximum loads:

0-

(~:) la=au

Eu

=

Em

=

(::)Ia=a m

Eo

=

(~:) la=ao

= -:- tan au E

0-

= -:- tan am E

0-

= -:- tan a o E

(6-59)

(6-60)

(6-61)

The dynamic secant modulus E dyn is obtained from the maximum and minimum stresses and strains: (6-62)

Equipment and Procedure for Hysteresis Measurement The equipment used is shown in Figure 6-72. The material undergoing testing is stressed by a servohydraulic device, and the signals from the force-measuring cell, and the extensometer or a strain gauge, are simultaneously amplified and digitized. The computer reads in data from the transient recorder at predetermined intervals and calculates the values of the hysteresis parameters. They are then printed out and stored with the data concerning the hysteresis loop. The function generator controls the testing machine by reference to the set point. The set point of the function generator is set by the computer. The computer thus controls the test cycle and records the data. It is also possible to attach a temperature controller.

~

UI

Aromatic hydrocarbons

Amines

Aldehydes

A

A

:;:"

A

B

A C

A A

C C

C

A

C C

.S!

~

B

A

~

:I

."

A B

A

A B

A A A

A

A

A B

A A A

A B

I ;f

A B

A

A

}

B

A

A A

A A

A C

B

A

A C

B

B

.f

.?;>

A A A

A A A

B

A

C

A

A

C C C

A A A

A A A

C

A

A C

A

A A

A

A B

A

A A

.f

A

A

it

oS

!!

i

-!

..!! ~

f0

C

C

C

A

A C

C C C

A C

C

C

A C

;f

~

C

C

B

A

A C

A A C

B C

C

A

B

C

A

A

A

C

C A A

A A

A

A

A A

~ ;f

]»

C

A

A

A A A

A A

A

A

A A

::!

ell

------

C A A

A A

A

A A

A

B C

::E

'i! l i i I i ..!! ~

i!!

-8 ~ ell

.

Material

C

B

A B

!

:9

~

I., J 1 8

----

B

A B

J

§-

;f .g 11

~

~

p

------------

A

A B

A A

A

A A A

C

A A A

A

Bases, strong

C

B C

A

Bases, weak

B C

A C

A C

Acids, organic, weak Acids, organic, strong Automotive Automotive, fuel Automotive, lubricants Automotive, hydraulic Solvents Aliphatic hydrocarbons Aliphatic hydrocarbons, halogenated Alcohols

0

:t:

J J

Acids and bases Acids, weak Acids, strong

Chemical Class

u

t I

~

~

»

~ ~

p

A

A

A

A

A B

A

B

B

A

A B

A B

C C

B

B

A C

~

"

ell

11 e

A

B

A

B

A B

A A

C C

C

C

C C

§ :;;:

.~

------

A A

A B

B

A

A B

'" '" ;;;

·iii"

11

.!l

ell

Example

Toluene, xylene, naphtha

Heptane, hexane Ethylene chloride, chloroform Ethanol, cyclohexanol Acetaldehyde, formaldehyde Aniline, triethanolamine

Dilute mineral acids Concentrated mineral acids Dilute sodium hydroxide Concentrated sodium hydroxide Acetic acid, vinegar Trichloroacetic acid

Table 6-49. Chemical Resistance of Various Plastics, Steel, and Aluminum by Chemical Class

~

IJ1

C

A B C

Oxidizing agents, weak

Water, ambient Water, hot Water, steam A C C

C

C

B

B

C

C

B

B C

A

A

A

A

C

--

B

C B

C B A A

B

B A

------

B

A

A C C

A

A C C

-A B C

A A A B

A

A A A

A

B

---- A

C -- B

A

A

--

A

C

C

A A

A

C

A

-- B ---- A

B

A

C B

C

A C C

A

C

A

A

C

A

C C

C

A A

----

A

A C C

A

A

---A

A

----

--

A

B

C

-- B

C

A C

A

A

--

C

------

B

C C

C

A A A

B

C

B

A

A

A

B B

A

C C C

C

C

B

A

A

A

C B

A

B B

A

C

B

B

A

A

A B

A

Laundry and dishwashing detergents, soaps Zinc chloride, cupric sulfate 30% hydrogen peroxide, bromine (wet) Sodium hypochlorite solution

Phenol Ethel acetate, dioctyl phthalate Butyl ether, diethyl ether Methyl ethyl ketone, acetone

Chlorobenzene

Source: (Hoechst-Celanese)

·-2ooF

Room temperature except hot water, steam and " ... Generally, extended expusure (more than a week) data were used

C-generally not recommended.

B-some effect.

A-minimal effect.

chemicals listed do not necessarily correspund to the ones on which the individual ratings are based.

that for the chemicals in that class found in the literature reviewed the rating was generally an "A." There may he other chemicals in the same class for which the rating would he • "C." Finally, the typical

no effect to toIaI dissolution. Therefore, an "A" rating for a particular plastic exposed to a particular class of chemicals should not he inlelpreted as applying to all chemicals in that class. The rating simply means

given material to specific chemicals in anyone class can vIII}' significantly. Indeed, during the preparation of the table, the effect on one plastic of various chemicals in the same category ranged from essentially

This infonnation is presented for instructional purposes and is not intended for design. The data were extracted from numerous sources, making consistent rating assignments difficult. Further, the response of any

C

A

Miscellaneous Detergents

Oxidizing agents, strong

B

Ketones

B

B

Ethers

Inorganic salts

C B

----------

Aromatic hydrocarbons, halogenated Aromatic, hydroxy Esters

~ ~

\11

Acetals Acrylics Acrylonitrile-ButadieneStyrenes (ABS) Aramids (aromatic polyamide) Cellulose acetates (CA) Cellulose acetate butyrates (CAB) Cellulose acetate propionates (CAP) Diallyl phthalates (DAP, filled) Epoxies Ethylene copolymers (EVA) (ethylene-vinyl acetates)

Plastic Material

200 (93.3°C)

77 (25°C)

~

8-

:I

.g

2

5

3 5 5 2-4 2 5

2 4 4

1-2

5

2

2-4 5 5 1 2 2

77

2 5

3

3

3 3

2 3 3-5

200

.

4 3-4 5

1-2 5

4

4 4

4 5 5

200

2

3

3 3

1-2 5 3-5

77

f

~

I~

~

!.g I

I~ I

i J

~

1-4 5 4

)

)

til

~

1 !t

p

~ ~

p

;...

fI)

2-5 3 2-4 3

1-3

2

2

1-2 2

3

2

3 4

200

2 2

I

~ ~

~

i

B

2

3

3 3

4

1-5 2

77

2-3

1-2

4

3 3

3

5 4 1-4

77

~

;...

3

2 5

;...

!

~ J

3-4 5

2-3

5

5 5

4

5 4-5 5

200

::E

Q:I

e f'B i :g G)

*

5

5 5

5

2-5 5 2-4

200

Temperature ("F)

l

77

5

8 ! ~

i

1 *

~ J

Environment

4

2

3

3 3

2

5 5 1-5

77

~

I G)

8

4-5 5

4

5

5 5

5

5 5 5

200

~

fI)

2 2

3-4

5

5 5

5 3-5

77

fI)

..,~

1

'" rl

j fI)

Table 6-50. Effects of Elevated Temperature and Chemical Agents on Stability of Plastics*

0.01-0.10 0.05-0.13

0.2-0.7

4-5 3-4 5

1.3-2.8

2-7 0.9-2.0 5 5 5

0.6

0.22-0.25 0.2-0.4 0.1-0.4

% Change by Weight

·<1

2

2-3 5 5

200

~u

j

fI)

~

.....

cr>

~

• A rating of I indicates greatest stability.

Ethyleneltetraftuoroethylent copolymers (ETFE) ethylene 6'" Fluorinated propylenes (FEP) ~ Perftuoroalkoxies (PFA) ~ Polychlorotriftuoroethylenes (CTFE) Polytetraftuoroethylenes ®(TFE) Furans Ionomers Melamines (filled) Nitriles (high barrier alloys of ABS or SAN) Nylons Phenolics (filled) Polyallomers

2

2

1 4

1 4

I

I

1 4 1 2-4

4 1 4

I

2

1 4

4 1 1-4

3

2 3

1 2

2 5

2 4 3 2-4

2 1 2

1 4 1 2-5

4

2 3

2 1 2 1 3 5 1

2 4 3 2-4 5

1 2 2 2-5 5 1 3

1 4 3 5 5 4

5 1 2 3-5

5 5 4

5 5 3 5

2

I

1 1 1 1-5

2 3

I

4 2 5

(cont'd)

0.2-1.9 0.1-2.0 <0.01

0.01-0.20 0.01-1.4 0.01-1.30 0.2-0.5

0

0.01-0.10

<0.03

<0.01

<0.03

\11

g;

Polyamide-imides Polyarylsulfones (PAS) Polybutylenes (PB) Polycarbonates (PC) Polyesters (thermoplastic) Polyesters (thermoset-glass fiber filled) Polyethylenes (LDPEHDPE-Iow density to high density) Polyethylenes (UHMWPE-ultrahigh molecular weight) Polyimides Polyphenylene oxides (PPO) (modified) Polyphenylene sulfides (PPS)

Plastic Material

5 4

5

4 3

4 2

5 5 5 5 3-5

200 (93. 3°C)

4 3 5 2 1-3

(25°C)

77

C

'"

8o CLl> <~

~

u

.~

2

3

4

1 2

2

77

CLl

£

3

4

5

4

2

4

5

3 5 5 5 5 4

200

3

4

2 4 4 5 3 2

1 3 5 1 3-5 3 5

77

->

o

-

..c u

B I:l _ '" I:l'" .t::

"t:l

200

'"

]c: ~~ ~ £

u

.~

2

2

77

'"

3

1 2 2 5 3-4 3

200

4

3 2 1 5 2 3

77

5

4 2 3 5 5 5

200

3

1-2

1 3 2

2

77

en ~

I:l '" O"t:l .!:; .~

~

'"

I:l

en~

g

bI)

bI)

Temperature (OF)

~ ~

CLl"t:l

~ f;l

'"

'" 1;! ~.:=

Environment

Table 6-50. (Continued)

4 2

1-2

3 1 3 1 4-5 3

200

77

2

1-3

1 2 2

2 2

'"

e~

2

5 2

3-5

3 4 4 1 3-5 4

200

- 0>< en

I:l

bl)1:l

~

77

1 2

3

2

3 1 5 2 3-4

I

CLl

1 3

4

3

4 3 5 3-4 4-5

200

~~

'"

B B

'" '"~ ~

I:l

"t:l

.... B

I:l

..::0

'"

<0.05

0.3-0.4 0.06-0.07

<0.01

0.00-0.01

0.22-0.28 1.2-1.8 <0.01-0.3 0.15-0.35 0.06-0.09 0.01-2.50

% Change by Weight

N

"
=, '"~

'" ~

$

1/1

• A rating of I indicates greatest stability.

Polyphenylsulfones Polypropylenes (PP) Polystyrenes (PS) Polysulfones Polyurethanes (PUR) Polyvinyl chlorides (PVC) Polyvinyl chlorideschlorinated (CPVC) Polyvinylidene fluorides (PVDF) Silicones Styrene acrylonitriles (SAN) Ureas (filled) Vinyl esters (glass-fiber filled)

3 4 3 2-4

2 3 I

4 5 3 3

4 4 1-2

2

4 I

3 5 2

I

I

4 5

I

2

4 4 5 4 4 5 4

4 2 4 4 3 4 4

1-2

4 3

5 2-3 5 5 4 5 5

3 4

5 5

5 4-5 5 5 5 5 5

2

1 2-3

3 3

2 3 2

4 3 3

5 3

2

4

3

5 2

4 3

2

I

2-3 3-4 5 2

3-4 5 2

4

2-3 5

3-4 5 2

2-3

5

1

I

5

I

I

2 2

4 3

4 2 2

I

2-3 4

1

3 3

5 4

2

4 5 3

4-5 5

3-4

I

2 4

3

3 2 4 3 4 4 4

2 4-5

4 5

5

4 4 5 4 5 5 5

0.4-0.8 0.01-2.50

0.1-0.2 0.20-0.35

0.04

0.5 0.01-0.03 0.03-0.60 0.2-0.3 0.D2-1.50 0.04-1.00 0.04-0.45

VI

Cl

22.7**

26.7

23.2

21.0

17.7

6.4

20.5

22.7

26.7

23.2

PEII30

PPS/40

PEEKl30

HTAl30

PFAl20

PEKl30

PES/30

PEII30

PPS/40

F19.3 XI6.S F23.9 F22.4 X15.3

F20.0 F17.7 A25.7 A2S.3 F20.0 F19.7 E21.0 E21.0 Al7.0 A16.4 A6.0 AS.9 E20.5 E20.S

Initial Nitric Strength Acid, 10%

PES/30

Base Resin! Glass (%)

A20.4 FIS.4 F23.S F22.4 X16.2

A24.0 FIS.2 E26.7 E26.4 F20.2 F19.3 E21.0 E21.0 A17.0 AIS.9 E6.3 A6.1 E20.5 E20.S

Sulfuric Acid, 10%

A20.4 FI9.S E26.0 F23.9 A21.1

E22.7 E22.7 E26.7 E26.4 E23.0 E22.7 E21.0 E21.0 E17.7 A17.0 E6.4 E6.3 E20.5 E20.S

Water

FIS.O XI6.S XS.O XC E23.0

F20.2 FI7.S F20.3 XIS.S E23.2 E23.2 E21.0 E21.0 F15.0 F14.7 E6.3 E6.2 E20.5 E20.S

Motor Oil

E22.7 E22.7 E26.7 A24.2 E23.2 E23.2 E21.0 E21.0 El7.7 E17.7 E6.4 E6.4 E20.5 E20.S

E22.7 A21.3 E26.7 E26.0 A20.9

E22.7 E22.S E26.7 F24.0 FIO.7

3 Days at ISO°F

E22.7 E22.7 E26.7 E26.2 A22.0 A21.3 E21.0 E21.0 El7.7 E17.7 E6.3 E6.2 E20.5 E20.S

7 Days at 73°F

Ammonium Ethylene Hyd.lO% Glycol

A21.S A21.S F24.0 F21.0 F19.5

E22.0 A21.7 E26.0 A2S.S E23.2 E23.2 E21.0 E21.0 E17.7 E17.7 E6.4 E6.4 E20.5 E20.S

Transm. Oil

E22.5 E22.2 X19.0 XC E23.2

E22.7 E22.7 X19.0 XC E23.0 E23.2 E21.0 E20.S E17.7 E17.7 E6.4 E6.4 E20.3 E20.0

Gasoline

XII.O XC X19.0 XC E23.0

X12.0 XC F22.0 XIS.O E23.2 E23.2 E20.5 A19.2 F14.0 F13.2 E6.4 E6.3 E20.5 E20.S

Brake Fluid

X6.0 XC X15.0 XC E23.2

X12.0 XC X16.0 XC E23.2 E23.1 E20.5 A20.2 XII.O XC E6.3 E6.2 E20.5 E20.S

F17.3 XC X7.0 XC E19.7

F19.3 XIO.O X20.0 XC A21.1 F19.3 E20.5 E20.S A15.9 F14.6 E6.4 E6.4 E20.3 E20.1

X16.3 XI4.S X5.0 XC X16.0

F19.3 F19.3 X17.0 XC FIS.I XIS.S E21.0 E21.0 E17.7 A16.4 E6.4 E6.4 E20.5 E20.S

TricholoroSkydrol 500 Toluene ethylene

A20.9 Fl9.1 A24.S XI7.S F19.5

A21.6 F20.2 A25.0 F20.0 F20.4 F20.0 E20.5 A20.0 E17.7 E17.7 E6.4 E6.4 E20.5 E20.S

Freon TF

Table 6-51. Tensile Strength (10 3 psi) versus Liquid Chemical Environment of High-Temperature TP Composites (Courtesy ICI-LNP)*

.....

In

-

17.7

6.4

20.5

22.7

26.7

23.2

21.0

17.7

6.4

20.5

HTAl30

PFAl20

PEK/30

PES/30

PEI/30

PPS/40

PEEK/30

HTAl30

PFAl20

PEKl30

F17.0 XIS.O F23.2 F21.S XI2.S XI2.S E20.4 XU.O F13.S XU.S X4.4 X4.3 A19.0 XIS.O

X13.7 E20.S XI2.S A16.3 F13.S F5.0 FS.O E19.9 AIS.S

F17.6 E21.0 A19.0 A16.3 FI3.S E6.4 E6.3 E20.2 AIS.9

FlS.4 XIS.9 A25.4 X19.3 X15.1 X12.3 E21.0 XIS.I F14.2 X12.0 F5.1 FS.I A19.1 XIS.S

XI2.S E20.4 AIS.9 A16.1 F13.3 A5.9 AS.S E20.2 AIS.7

Xl1.6 X6.1 X19.9 XIS.O XlO.2 X9.7 A19.9 X14.6 X12.5 XU.O F4.S X4.7 AlS.5 XI4.S

""To convert psi 10 Pascals, multiply by 6,895.

Values in bold Iype an: for sample strained 0.25% in ftexure during tesl.

*Test samples were ASTM type V tensile specimens.

21.0

PEEK/30

X15.1 XIS.I F15.9 XI2.S X13.0 XU.S F5.1 X4.S FlS.0 XI4.S

XD XD

X15.0 X13.1

Al2.0 AlS.9 FIS.S F14.2 FI3;S F5.5 FS.3 A19.5 AIS.S

F19.7 E21.0 E21.0 E17.7 E17.7 A5.S AS.S E20.5 E20.S

F17.3 XIS.4 A25.S XIS.2 X17.2 X17.2 E21.0 FIS.9 F15.1 F14.3 F5.2 FS.I A19.5 AIS.O

E22.7 E22.0 E26.0 F23.S X17.0 X17.0 E21.0 E21.0 E17.7 E17.7 A5.S AS.S E20.5 E20.S

1 Day at 300°F

F20.4 E21.0 E21.0 E17.2 A16.1 F6.0 FS.6 E19.9 AIS.6

A21.7 AlI.O F22.0 X14.0 X16.7 X16.7 E21.0 AIS.9 E17.7 AI6.S F5.0 F4.S E20.5 A19.0

FI9.S E21.0 E21.0 E17.7 E17.4 F5.1 FS.O E20.5 E19.9

E22.4 E22.1 XlS.0 XC E23.0 F20.1 F17.5 F17.3 E17.7 E17.7 A6.0 FS.6 E20.5 FIS.O

Al2.2 A19.1 FIS.S E17.7 E17.7 A6.0 AS.9 E20.5 AIS.7

XlO.O XC XlS.0 XC E23.0 AlO.9 F17.2 F17.1 F13.6 F13.3 F5.1 FS.O E20.5 E20.S

Al1.0 FlS.0 F17.7 F13.S F13.S F5.1 FS.O E20.5 E20.S

XD XD XD XD X17.2 X16.7 FlS.5 X13.S F14.0 X13.2 F5.0 FS.O E20.5 FIS.S

F19.0 A19.3 FIS.S F15.0 F13.7 A5.S AS.6 E20.5 AIS.6

X16.1 XC XD XD X15.1 X13.7 FI5.S X13.3 X12.7 XC F5.2 FS.O E20.5 FI7.S

X14.4 A19.1 XIS.O F13.5 XU.S F5.5 FS.4 E20.5 AIS.9

FI9.S X13.0 A24.5 X13.0 X16.9 X16.0 A19.3 F16.6 E17.7 E17.7 F5.7 FS.S E20.5 E20.3

FIS.6 A19.6 AIS.9 E17.7 E17.0 E6.4 E6.4 E20.5 E20.S

E A

= Excellenl (0 10 3% loss of lensile strength) = Acceptable (3 10 10% loss) F = Fair (10 10 25% loss) X = Unacceptable (more than 25% loss) C = Sample crazed or cracked D = Sample dissolved or severely tackified

Key to ratings:

XC XC X14.0 XC E23.2 XI4.S A19.3 FIS.3 X6.0 XC F5.0 F4.S E20.5 E20.S

XI6.S E20.3 A19.2 X7.0 XC F5.0 A4.9 E20.5 E20.S

IJ1

;j

Blue-yellow

Blue-yellow Blue-yellow Orange yellow (green) Green Orange yellow

0.S5-O.9

0.91-0.93

0.93-0.96

1-1.25

Polypropylene

LOPE

HOPE

Epoxy

No Yes Softening Yes (trans)

Yellow (Green) Yellow-orange Blue mantle Yellow-orange

1.12-1.46

1.15-1.65

l.lS-1.19

Vinyl chloride

Acrylic

1.1-1.16

Polyester

Ethyl cellulose

1.09-1.14

Nylon Yes

Yes (trans) Yes

Blue mantle Yellow Blue mantle Yellow Blue white

1.07-1.0S

Polyvinyl butyral

Yes Yes

1-2.4 1.05-1.0S

Yes (trans) Yes (trans) No

(trans)

Yes

Color

Chlorinated PE Polystyrene

As is

Specific Gravity

Material

Flame Color (Copper Wire)

White to Green Some Black

Black

Black

Black

White

White

White

Smoke Density

Little

Dense

Dense

Candle wax

Very little

Sweet (resinous) Acrid chlorine Floral burnt fat

Sweet

Burnt hair

Sweet marigolds Rancid butter

Phenolic

Candle wax

Heavy

Solvents

Very little

Very little

Odor

Table 6-52. General Characteristics of TPs

Toluene

Toluene

Sec-amyl alcohol

Toluene Oiethyl benzene

Toluene (slowly slight) Dipropylene glycol Toluene

Comments

Clear bead

No Drip

Softens

Swells, froths Drips

Drips, swells

Soot, no drip

Some soot

Drips, swells

Drips, swells

Drips, swells

\11

...... w

Blue mantle Yellow

1.41-1.42

1.58-1.75 1.62-1.72

Acetal

Saran Vinylidene chloride Tetrafluoroethylene

2.1-2.3

Intense white

1.35-1.40

Cellulose nitrate

(Green) Yellow

Dark yellow Mauve blue

1.27-1.34

Orange-yellow

1.20

Polycarbonate

Cellulose acetate

Dark yellow

1.19

Vinyl acetate

No

Yes Yes

Yes

Yes Yes

No

Yes

Black

Black

Black

Burnt hair

Sweet

Formaldehyde

No odor

Phenolic sweet Acetic vinegar

Acetic

Furfuryl Alcohol and acetionitrile Dipropylene Glycol and Acetionitrile

Sec-hexyl alcohol CycJohexanol Acetonitrile Toluene

Chars

Heavy black

Drips

Burns, charred bead

Chars

Some swell

....

VI 'I

Specific gravity color Possibilities By-products of cure Molding pressures Molding temp., OF Shrinkage, % Tensile strength, 103 PSI Elongation, % Modulus of elasticity tension, 103 PSI Compressive strength, 10-3 PSI Aexural strength 103 PSI Impact strength (Izod) Heat resistance OF (continuous) ("C) Heat distortion, "F (OC) Water absorption, %, (24 hrs.) Dielectric strength V per mil Dielectric constant (60-106 CPS) Dissipation (power factor 60-106 CPS) Arc resistance, sec. Burning rate

Property

4.0-7.5 0.01-.15

Tracks Very low

13-21 0.2-1.0 250-500 (121-260) 115-550 (46-288) 0.08-0.15 400-500 3.3-5.0 0.002-.050 45-120 Slow to selfextinguishing

8.5-18.5

0.2-4 250-500 (121-260) 140-400 (60-204) 0.15-0.60

200-420

2.8-5.2

0.003-.028

Slow to selfextinguishing

125

12-15

15-21.5

13-36.5

360-400

0.25-0.40 160.250 (71-121) 165-260 (74-127) 0.03-0.04

11-17

7.5-10

1.5-2.0 4-5

2.6 4--6

5 3--6.5

Selfextinguishing

100-145

0.015-0.080

None to slow

250-360

0.0008-0.01

3.2-5.2

4.3-7.6

230

0.0018-0.013

3.2-131.5

250-725

0.1-0.5

0.3-0.5

200-500

270--680 (132-360) 0.11-0.60

>900 (482) 298 (148)

300-400

0.25-0.8 >600 (315)

7.1-49.5

4.7-40.0

<1-10 1.9-28.5

330-660 0.1-0.3 5-27

H 2O Low-high

1.42-1.90

Polyimides

0.25--6.0 >600 (315)

9-14

9-15

280-360 1-1.5 4-5

Good H 20, RCOOH Low-high

1.30-1.34

Silicones

0.24-0.40 210 (99)

11-14

40-45

270-360 1-1.5 7-9

270-320 1-1.2 6-9

<70-330 0.1-0.4 4-13

<70-320 2-8 6-10

1.40-1.48 Very good H 2O Medium-high

1.30-1.86 Limited H 2O Low-high

1.10-1.4

Melamines

Good None O-high

Phenolics

1.10-1.45

Epoxies

Very good None O-high

Polyesters

Table 6-53. General Characteristics of TSs

\II

'I

\II

Arc resistance, Scratch resistance, colorability

Good general properties, low cost

Major advantages

Ease of fabrication, clarity with flame retardancy, moderate dissipation (power) factor

Fair High cost

Excellent Colors limited

Machining qualities Major limitations

None

Attacked by some

Generally resistant

Attacked (ketones and chlorinated solvents) Good Cure shrinkage Dermatitis, difficult to mold release Low shrinkage, excellent adhesion

Good

pH Decomposes

Slight

Attacked

Effect of strong alkalis Effect of organic solvents

Attacked

None

None to slight

Effect of weak alkalis

None

None Decomposed by oxid. acids, none to slight with regular org. acids Slight to marked function of

None Attacked by some

None None to considerable

Effect of weak acids Effect of strong acids

Slight color change None Decomposes

Darkens

None

Yellows

Effect of sunlight

Heat resistance, low dissipation (power) factor

Fair to good Very high cost

Attacked by some

Slight

None to slight

None to slight Slight

None to slight

Heat resistance

High cost

Very resistant

Attacked

Slowly attacked

Resistant Slowly attackedresistant

Table 6-54. Example of Qualitative Plastics Environmental Ratings

Material Family ABS Acetal Acrylic Allyl ASA Cellulosic Epoxy Fluoroplastic Melamine-formaldehyde Nylon Phenol-formaldehyde Poly (amide-imide) Polyarylether Polybutadiene Polycarbonate Polyester (TP) Polyester-fiberglass (TS) Polyethylene Polyimide Polyphenylene oxide Polyphenylene sulfide Polypropylene Polystyrene Polysulfone Polyurethane (TS) (TP) SAN Silicone Styrene butadiene Urea formaldehyde Vinyl

Abrasion Resistance

Weather Ability (Natural)

F G P G F F-P G G G G G VG G G F G G G VG G G G P G VG F F G

F-P F G F P F-G F E F-G F-P G F F F-G F F G P F-P F-G G F-P F-P F-P E-G F VG G

G

F

G

G

"Those with "No" ""Iuire special paint. primer. or ptqJOintiDll surface preparation. Code: E = ExccUent; VO = Very Good; 0 = Good; F = Fair; P = Poor.

576

Paint Ability· No No No No No No Yes No Yes Yes Yes No No No No No Yes No No Yes No No No No No No No No Yes No

Transparent

Translucent

Yes

Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes

Table 6-55. Example Showing Permeability of Some Plastics

Type of Polymer ABS (acrylonitrile butadiene styrene) Acetal-homopolymer and copolymer Acrylic and modified acrylic Cellulosics acetate Butyrate Propionate Ethylene vinyl alcohol copolymer Ionomers Nitrile polymers Nylon Polybutylene Polycarbonate Polyester (PET) Polyethylene Low density Linear low density Medium density High density Polypropylene Polystyrene General purpose Impact SAN (styrene acrylonitrile) Polyvinyl chloride Plasticized Unplasticized Polyvinylidene chloride Styrene copolymer (SMA) Crystal Impact

Specific Gravity (ASTM D 792)

Water Vapor Barrier

Gas Barrier

Resistance to Grease and Oils

101-1.10

Fair

Good

Fair to good

1.41

Fair

Good

Good

1.1-1.2 1.26-1.31 1.15-1.22 1.16-1.23 1.14-1.21

Fair Fair Fair Fair Fair

Fair Fair Fair Very good

Good Good Good Good Very good

0.93-0.96 1.12-1.17 1.13-1.16 0.91-0.93 1.2 1.38-1.41

Good Good Varies Good Fair Good

Fair Very good Varies Fair Fair Good

Good Good Good Good Good Good

Good Good Good Good Very good

Fair Fair Fair Fair Fair

Good Good Good Good Good

1.04-1.08 1.03-1.10 1.07-1.08

Fair Fair Fair

Fair Fair Good

Fair to good Fair to good Fair to good

1.16-1.35 1.35-1.45 1.60-1.70

Varies Varies Very good

Good Good Very good

Good Good Good

1.08-1.10 1.05-1.08

Fair Fair

Good Good

Fair Fair

0.910-0.925 0.900-0.940 0.926-0.940 0.941-0.965 0.900-0.915

577

Table 6-56. Examples of Properties of TP and TS Elastomers· Non-Oil-Resistant Rubben

Generic Name

Weight of Base Elastomer

Lb.lin.' (g/cm') Specific Gravity

Durometer range

fit 1liil~

'1iloS!

i

Tensile strength, psi (reinforced) (MPa) Elongation % (reinforced) Creep, room temperature

Compression set Electrical resistivity Impermeability, gas

j ~

Resistance to

Impact Abrasion Tear Cut growth 25O"F (121"C)

Tensile strength;

I

~ :l

J j

Psi (M Pal, at

elongation, 'Ib, at

400"F (204"C) 250"F (l21"C) 400"F (204"C)

Creep at 212'F (l00"C) Heat aging at 212'F (IOO"C)

Flame resistance Stiffening, 'F ("C)

Low temperature

Styrene-Butadiene (GRS) (SBR)

Butyl

EP Rubber

0.033 (0.91) 0.93

0.034 (0.94) 0.94

0.033 (0.91) 0.92

0.031 (0.86) 0.86

30-100 4,000 (27.5)

40-100 G 2,000 (13.8)

30-100 F 2,000+ (13.8)

500 E G E G

450 E G E F

300-800 F F E

E E E E

E E F G

G G G E

G G P G

1,800 (12.4) 125 (0.862) 500

1,200 (8.27) 170 (1.17) 250

1,000 (6.90) 350 (2.41) 250

2,000 (13.8)

0

Resilience

1l

Natural Rubber

Brittle pt., 'F ("C)

SO

G G P

F E(-29to -46) P

0-50 (-18 to -45) -80 (-62)

-lOto -40 (-23 to -40)

-SO

-20 to -50 (-29 to -46) -90

(-62)

(-68)

F-G E P-G F-G P P P G P-F

F G P G E F-G F-G P P P F P

E E E P E E E P F-G P E F

E G E P G-E G-E G-E P F P P P-F

P P-F

P-F P

F G

F-G G-E

F-G F-G P-G

F-G G P-G

F-G G G

G G G

E

E

Elastic cord, tires

Tires

F-E Air and gas

P Weather-

-20 to -50 (-29 to -45)

-SO

.J

.ll

fluids

Properties

P

Gasoline, kerosene. etc. (aliphatic Hes) Benzol, toluol, etc. (aromatic HCs) Degreaser solvents (halogenated HCs) Alcohol Synthetic lubricants (diester) Hydraulic

Subjective

F G

Water Acid Alkali

Silicates Phosphates

Taste Odor

Nonstaining

Bonding to rigid materials Typical applications

SO

400 (2.76) 300-500 0-120

F (-23 to -40) E P

(-62)

Weather Oxidation Ozone Radiation

60 G G (-18 to -45) P

0

30-90 G 2,000-3,000 (13.8-20.7) 500 F F E G

containers

resistant parts

·Ratmgs lItt In thiS decreasing order: Outstandin,. ElI.l;Cllcnt. Good. Fair. Poor. Note. 0 = ouiscanchng. E "" cJ.cellem. G = good; F - fau-; P ., poor.

578

Oil-Resistant Rubbers Neoprene 0.044 (1.22) 1.23

Acrylonitrile (BUNA-N) 0.036 (1.00) 1.00

4()-95 E 3000+ (20.7) 65()-S5 F-G F-G F G

Polysulfido O.04S (1.33) 1.34

Heat-Resistant Rubbers Polyurethane 0.039 (1.0S) 1.05

Silicone 0.036 (1.00) 0.95

Auorosilicone 0.049-0.06 (1.36-1.66) 1.35-1.65

Auoroelastomers 0.05-0.07 (I.3S-1.94) 1.4-1.95

Acrylics 0.040 (1.11) 1.10

2()-9O G 1,()()()-3,500 (6.9()-24.1) 4Q()-6()() G G P G

2()-SO F 5Q()-1,500 (3.44-10.3) 2Q()-550 P P-F F E

55-100 G-E 4,()()()-S,000 (27.6-55.2) 25()-SOO G-E E G G

25-S0 P-E 6Q()-1,500 (4.14-10.3) 9()-SOO F-E G-E E P

4()-SO P-G SQ()-I,350 (5.52-9.31) l00-4S0 F-G G-E E P

65-90 F 1,5Q()-3,000 (10.3-20.7) 100-450 G G-E G E

4()-9O G 1,700 (11.7) 450 F F-G F G

F

E E E F-E

P-F P-G P-G P-G

P-F P-G P-G P-G

P-G G P-G P-G

P F-G F G

G G-E G G

G G

P P-F P-F P

1,500 (10.3) ISO (1.24) 350 ()-IOO

700 (4.S3) 130 (0.90) 120 20

700 (4.S3) Under 25 (0.17) 140 Under 25

1,800 (12.4) 200 (I.3S) 300 140

S50 (5.S6) 400 (2.76) 350 200

S50 (5.S6) 350 (2.41) 160 150

3Q()-SOO (2.07-5.52) 15()-300 (I.03-2.Q7) IQ()-350 5()-16O

1,300 (S.96) 225

F-G G G

E G P-F

P G P

E F-G P-F

E E F-G

E E F-G

G-E 0 E

F E P-F

+ 1()-20 ( -12--29) -45 (-43)

+ 3()--20 ( -1.1--29) -65 (-54)

-1()--45 ( -23--43) -60 (-51)

-1()--30 (-23--34) -60 ( -51)

-6()--120 (-51--84) -9()--14O ( -6S--96)

-70 (-57) -9()--120 ( -6S--84)

+ 2()-+ 30 ( -6.6---1.1) +1()-+20 ( -12--6.7)

+35-+10 (1.7--12) -10 (-23)

E G E F-G G G G G P

G F-G P F-G E G F-G E G

F P

E F-G

E E E F-G G F G E E F-G G G

E E E G G P-F P-F E P-F F-G G P

E E E F-E G P-G P-F P-F P P-F G P-F

E E E F G G G G F F-G G G

E 0 0 F-G G G-E P-G E E G E F-G

E E E P-G F F P E P P

G P

F P

P-F P-F

P

P G

G G

G P

G P

F-G F-G G-E

F-G G P-G

P-F P P-F

G G G

G G 0

G G 0

F-G G P-G

F-G F-G G

G-E Weather- and oilresistant parts

G-E Gasolineand oilresistant parts

p

E

P

F-G Solventresistant parts

F-G Abrasionresistant parts

F-E High- and lowtemperature, non-oilresistant parts

G High- and lowtemperature, oilresistant parts

P-G Hightemperature solventresistant parts

400 150

p

G

G Hightemperature, oilresistant parts

(cont'd)

579

Table 6-56. (Continued)

Generic Name Weight of Base Elastomer

.~

t: ~

(,I)

.~~

8

8.e-g :l 8. Ojfile

£ S[(l if

Durometer range Resilience Tensile strength, psi (reinforced) (MPa) Elongation % (reinforced) Creep, room temperature Compression set Electrical resistivity Impermeability, gas

OJ

..

'8

..c ~

Resistance to

::il

~

i3

e ~

.~

~

8.

£

·&!1

~ "e ,g= ;.

Jj

Subjective Properties

Tensile strength; Psi, (MPa) at elongation, %, at

250"F 4OO"F 250°F 400°F

(l21°q (204°q (l21°q (204°q

Stiffening, "F (OC) Brittle pI., OF eq

Specialty Rubbers

Chlorosulfonated Polyethylene (Hypalon)

Various Types

0.040 1.10 50-95 G 1500-2500 (10.3-17.2) 200-500 F F-G G E

Thermoplastic, Thermally conductive, Damping, High and low permeability For specific details consult manufacturer.

F E G -30 to -50(-34 to -46) -60 (-51)

Hydraulic fluids

G P-F

Taste Odor Nonstaining

conductive,

500 (3.45) 200 (1.38) 60 20

E E 0 F-G G E E F P-F P-F G P

Silicates Pbosphates

Electrically

F-G G F-G F-G

Weather Oxidation Ozone Radiation Water Acid Alkali Gasoline, kerosene, etc. (aliphatic HCs) Benzol, toluol, etc. (aromatic HCs) Degreaser solvents (halogenated HCs) Alcohol Synthetic lubricants (diester)

Bonding to rigid materials Typical applications

580

Impact Abrasion Tear Cut growth

Creep at 212°F (loooq Heat aging at 212"F (lOOoq Harne resistance Low temperature

""

Lb.lin.' (glcm') Specific Gravity

Weather-Resistant Rubbers

F-G G E F-G Weather-resistant coatings

Thennoplaslic Elastomers

Styrenebutadiene

Olefinic

Urethanes 1.15-1.25

7SA-6SA

92A-63D

Low 5,800-6,400 (40.0-44.1)

SOA-60D Medium

650-2,000 (4.48-13.8)

High-Damp 3,000-7,000 (2.07--48.3)

180-250

200-500

500-800

0.042~.04S

0.93-1.113

0.88-1.02

3S-9SA

6S-90A, SOD Low

400-1,350 P G

P-G High

F-G G

F-G G

Maximum use temp, "F ("C) 150P -40(-40)

EVA andEEA 0.03~.34

0.031~.037

High 500-3,000 (3.45-20.7)

Polyester 0.043 1.20

0.03~.038

250 (121) P -60 (-51) -76(-60)

F-G P

o E

200 (93.3) P -40(-40)

0.91~.9S

800-4,000 (5.52-27.6) 50-1,200

P

G G

F

G E

F-G F-G

250-300 (121-149) P

-70(-57)

140(60) P -20 (-29)

-94 (-70)

F-P F F

E E E

E-G G-F G-F

G G G

G-F G-F G-F

G F F Very poor Very poor Very poor Very poor

E G-F

P F

E G

G-F P P G-F

E G F F-P

G G 0 G G-F P G

G Gaskets, seals, footware

P Gaskets, seals, toys

G Household items, toys

G Gears, industrial parts

P P P F

F Household items, toy

581

582 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

c

{

Figure 6-70. Definitions of the mechanical energy components on a hysteresis loop; energy loss (left) and stored energy (right).

Figure 6-71. Definitions of stiffness on a hysteresis loop. Two test patterns have proved to be useful: the increasing-load test and the dynamic

fatigue test. In the increasing-load test, the load is increased by a set amount to a series of prescribed values, after which the specimen is unloaded to a load level where no failure occurs. If the parameters return to their initial values after unloading, no irreversible damage has occurred at the preceding stress level. This method is suitable for a rapid study of the stress dependence of dynamic characteristics, while in the dynamic fatigue test the time dependence of changes in characteristics is recorded. With this test also, it is often good policy to unload to a lower level to allow possible damage to be detected.

Hysteresis Measurements on Fabric Laminates A TS polyester resin/glass-fabric laminate containing about 60 wt.-% glass was studied. Figures 6-73 to 6-77 reproduce the original diagrams from an increasing-load test, using a 10 Hz sine-wave alternating load cycle, with (R = -1). The load was increased successively by 10 N/mm2 from 60 N/mm2 to break at 100 N/mm2 (see Fig. 6-73a). Each load level was maintained for 15,000 cycles and then reduced again to the starting level. The temperature on the surface of the test sample was measured during the test. The energy loss changes with the level of load (see Fig. 6-37b). It heats up the sample, and the surface temperature increases correspondingly (see Fig. 6-73c). Whenever the energy loss changes, the temperature falls or rises exponentially. The damping shows a similar pattern to the energy loss (see Fig. 6-73d). However, the different load levels

THE PROPERTIES OF PLASTICS 583

ranSlent recorder dIgItal

servo -hydraulIC

testing machme

IntermedIate store personal (omputer 1--+.1.----1--1

Figure 6-72. Experimental arrangements for hysteresis measurement.

can be compared better, since damping is independent of stress (see equation 6-58). Damping first exhibits a maximum, then falls off and later increases again. The origin of the first damping maximum is the creation of transverse cracks between bundles of long fibers. The development of transverse cracks later decreases, until a point is reached when no further cracks occur. This situation in cross-laminated composites is called the characteristic damage state. The edges of the transverse cracks rub against one another and so raise the energy loss. After a while the crack edges become smooth and the frictional losses become less. Thus, damping decreases after cracking stops, because no more rubbing surfaces are created and the frictional loss per unit of crack area decreases. The second rise in damping before breakage indicates the occurrence of a second stage of damage to the sample. Lengthwise cracks are formed at the crossover points of the weave. These grow along the length of the fiber bundles and lead to delamination. During the compression phase, this results in partial crumpling of individual layers, and breakage. During dynamic fatigue tests, damping shows the same kind of behavior as in increasing-load tests. Figure 6-74 shows the pattern of change of damping during a tension/compression test at 70 N/mm 2 (R = -1). A damping maximum and minimum can again be seen. Thus, in dynamic fatigue tests three characteristic features occur: The damping maximum that characterizes the completion of transverse cracking The damping minimum that signals the creation of longitudinal cracks The breaking point

~

j

...

Q)

(a)

U5

I00r====----;::====~____;;::::====="~ ml-,----~ L...J Ll o~r

__________________~-------

-,O ;-;----~

-IOO~

o

C) Q)

c

i )0

30

I

0.015

W

<:::J

I

o

o

I

10

20

30

so

0

6O.tO! 70

34,----c---~--------~----~--~--~

~}~-~ ~E¥ :

~

:::I

~ Q) a. E

Q)

~

o

C)

c

.0.

(d)

20

0.045

...

(c)

10

;t;;i;;;:):L)1

I/) I/)

.Q >.

(b)

_ _ _,.........,

E ~

10

20

40

30

60 .1 0) 70

:~n;; ~· i :{l I o

10

20

30

0

)0

60 .,0) 70

Number of cycles Figure 6-73. Patterns of change in characteristic quantities observed during an increasing-load test on TS polyester-glass fabric laminate containing 60 percent glass by weight.

0,040 r-----..,.....------.......,.....--------. 0,036 ~---i---~_+_-------30_-----J-

0,032 ~-----

R = -1

= 70 N/mm f = 10 Hz

C1max

C>

c .0.

E

~

0.028 L.-_ _--:-_ _ _~_ _ _ _ _ _

..l...__ __

10 3

10'

Number of cycles Figure 6-74. Change of damping at 70 N/rrun2 • 584

101.

THE PROPERTIES OF PLASTICS 585

160,-----------r-----.---N/mmZ 120r---+--1~~-~--~-~

..... 40r-----~----_+----<'~------r-----~

O~-~--~---~-~~-~

10 2

Number of cycles Figure 6-75. Break curve (O'B) and curves for damping extremes ('Ymin, 'YmoJ on stress versus number of cycles diagram.

If these three points for a number of different stresses are transformed into a diagram of stress versus number of cycles, two other curves appear, in addition to the breaking line (see Fig. 6-75). The damping maximum curve characterizes the completion of transverse crack formation, and the damping minimum curve signals the creation of longitudinal cracks.

Hysteresis Measurements on Shoe-Sole Material The demands made on the soles of jogging shoes are very high. For one, the material has to be soft so that the wearer's joints are as lightly loaded as possible. For another, the material must give back the highest possible proportion of the energy it stores when it is loaded, so as to minimize the expenditure of energy during running. So as to be able

to judge the suitability of different materials as soles for running shoes, some means of characterizing the above-mentioned properties must first of all be found. After this, the parameters must be measured by a simulated-service test. The values of materials' characteristics change with time because of aging. The constancy over time of a characteristic should be noted as a third criterion for the choice of a soling material, since shoes ought to remain usable for the longest possible time. The physical characteristic used for the first requirement (material softness) is the compression modulus. If the modulus is low, the load on the joints is less than if it is high. The damping coefficient quantifies the proportion of mechanical energy not given back by the sole. It is the ratio of the energy lost to that stored (see equation 6-55) and is thus a measure of the energy lost in the shoe during running. The efficiency, IJ., can be calculated from the damping. It indicates the ratio of the work recovered (stored energy) to the total energy put in: IJ.=

1

Wv

+ Ws

=--

A + 1

(6-63)

586 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

20r-----~--------------------._----_o------~

(J)

:::l

"S u

o E

c: o 'in (J)

Q) ....

a. E o

()

5~----~------~------------~---------------

o

10

15

20

25

,,10 3

30

Number of cycles Figure 6-76. Compression moduli of three PUR foam materials.

Simulation of a Loading Cycle

During running, a material undergoes cyclic loading. The foot presses down and the sole is compressed. The compressive load increases from zero to some maximum value. After this the foot rolls off the sole and unloads it until the foot descends again. The time taken to reach the maximum load of 1 N/mro2 is some 40 ms, the contact time about 450 ms, and the total cycle time between successive descents of the foot is about 700 ms. This curve pattern was simulated by the function generator. The setting of amplitude and prestress is carried out by the computer controlling the test cycle. Differences in Patterns of Damage

Three different systems of soft polyurethane (PUR) foam were tested, all of which are used for the soles of jogging shoes. Test samples 30 mm square and 18 mm thick were used. It turned out that all the samples were severally damaged during the first 1,000 test cycles and that after this damage proceeded approximately linearly with time (see Figs. 6-76 and 6-77). In Figure 6-78, two hysteresis curves for sample 3 are shown. They were measured shortly after the start of the test, at 15 cycles, and before the end of the test, at 28,730 cycles, respectively. It is clear that the energy loss (area enclosed by the curves) becomes smaller with an increasing number of cycles. A shift of the hysteresis curve, caused by quasi-permanent set, is observed. Also, the upper and lower gradients change with increasing time under load. This qualitative information can be quantified by means of the trend in values of material characteristics. The lower slope describes the force/extension behavior in the region of greatest compressive loading when the foot descends (see Fig. 6-76). The elastic modulus of three samples increases rapidly at first; but later differences become evident. Samples I and 3 exhibit a slower rate of increase, while the modulus of sample 2 at the end of the test is almost twice its value after the first few load cycles. The efficiency of all three materials becomes essentially constant after rapid initial increase in value (see Fig. 6-77).

00 0/0

90

?

60 Q)

.(3

:= w

,

fr

>-

0

c

/2

70

" '3

60L-----~------------------------------------

o

S

10

1S

20

Number of cycles Figure 6-77. Efficiencies of three PUR foam materials.

Or-------------------------~-------------

N/mml

-0,4 f------+H'------~_____7"~__7'_::......------------

-0,6 f----Htf------l.-f-+......,.--------....:....----------II)

~

-0,6~--~--------_++_~------------------------------

Ci5 -1L--A~

________- L______________________________

o

-20

20

0/0

..

Extension (Strain) Figure 6-78. Hysteresis curves for a PUR foam material (sample 3) after 15 and 28,730 load cycles.

587

588 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Material Choice Exemplified By Soling for a Jogging Shoe Different materials show different dynamics behavior under simulated running loads. The mechanical characteristics of materials change somewhat during fatigue testing because of internal damage. The biggest changes occur during the first 1,000 cycles (see Figs. 6-76 and 6-77). For making qualitative judgments about the suitability of the three materials as soling materials, it is useful to analyze the region from 1,000 cycles onward, after the runningin phase. The compression moduli of samples 1 and 3 increase some 10 percent to 10 N/mm2 from here to the end of the test, while the value for sample 2 rises to some 20 N/mm2, about twice the value of the other two samples. As time goes on this material will load the joints very much more heavily. The second criterion is the efficiency, the ratio of recoverable ("stored") energy to total energy. After the running-in phase, it stays almost constant at 78 percent for sample 3, 85 percent for sample 1, and 90 percent for sample 2. This means that with a sole made from material 3, 22 percent of the energy a sportsman uses in running is transformed into heat; with material 2, in contrast, it is only 10 percent. But material 2 is very hard for running shoes, and the efficiency of material 3 is worse than that of material 1. Thus it appears that material 1 is the most suitable for the soles of jogging shoes.

Hysteresis Measurements for Material and Part Tests Hysteresis measurements provide values for a number of characteristics of materials that describe their behavior under dynamic loading. The damping coefficient has proved to be particularly successful for describing the behavior both of materials and of components. By example of the glass-fabric laminate it has been shown how, with the help of the damping factor, the degree of damage can be described. The increasing-load test provides the possibility of rapidly recording the development of damage. An increasing-load test lasts about one hour; in this time it readily shows up changes in material characteristics, and the development of internal damage that occurs in vibration fatigue tests. With this method the dynamic behavior of different materials or materials modifications, and the effects of production changes on a material, can be discovered rapidly and cost effectively. Furthermore, the hysteresis-measurement procedure is very effective when values of dynamic characteristics and efficiencies are required as indices of quality: for example, in the choice of suitable foam materials for sport shoes, the pattern of modulus change under dynamic load, or the damping index, can be used.

Chapter 7

THE PROCESSING OF PLASTICS

The various processes discussed in this chapter are used to fabricate all types and shapes of plastic products, ranging from household convenience packages to electronic devices and many others-including the strongest products in the world, used in space vehicles, aircraft, building structures, and so on. Proper process selection depends upon the nature and requirements of the plastic, the properties desired in the final product, the cost of the process, its speed, and the product volume. (Remember that a plastic may also be called a polymer or a resin.) Some materials can be used with many kinds of processes, but others require a specific or specialized machine (see Fig. 7-1). Plastics consumption by process is given in Table 7 -1. Numerous fabrication process variables play an important role and can markedly influence a product's aesthetics, performance, and cost. This review provides information on the effects on performance and cost of changing individual variables during processing, including upstream and downstream auxiliary equipment. Many of these variables and their behaviors are the same in the different processes, as they all relate to temperature, time, and pressure. This chapter contains information applicable to all processes characterized by certain common variables or behaviors, such as plastic melt flow, heat controls, and so forth. It is essential to recognize that for any change in a processing operation there can be advantages and disadvantages. The old rule still holds: for every action there is a reaction. A gain in one area must not be allowed to cause a loss in another; changes must be made that will not be damaging in any respect. All processes fit into an overall scheme that requires interaction and proper control of its different operations. An example is shown in Figure 7-1, where a complete block diagram pertains to a process. This FALLO (Follow ALL Opportunities) approach can be used in any process by including those blocks that pertain to the fabricated product's requirements. Knowledge of all the processing methods, including their capabilities and limitations, is useful to a designer in deciding whether a given part can be fabricated and if so by which process. Certain processes require placing high operating pressures on plastics, such as those used in injection molding, where pressures may range from 13.8 to 206.9 Pa (2,000 to 30,000 psi). Because of these pressures and the fact that three-dimensional parts are molded, injection molding is the most complex process, but it is easily controlled. Lower pressures are used in extrusion and compression, ranging from 1.4 to 69 Pa (200 to 10,000 psi), and some processes, such as thermoforming and casting, operate at relatively little pressure. Basically, with the higher pressures it is possible to develop 589

Figure 7-1. Alpha I, the centerpiece mUltipurpose machine for GE Plastics' Polymer Processing Development Center in Pittsfield, Mass. is in the center of a facility with a dozen more different processing machines. This one-of-a-kind multipurpose machine measuring 75 ft. long x 50 ft. wide x 55 ft. tall can produce parts up to six feet square. It is the world's first machine to combine more than seven processes, including injection molding, flow molding, tamping, coinjection molding, gas injection molding, compression molding, flow forming, side-by-side, and proprietary processes. 590

THE PROCESSING OF PLASTICS 591

Table 7-1. Plastics Consumption by Processes Material Extrusion Injection Blowing Calendering Coating Compression Powder Others

Percent by Weight 36 32 10 6 5

3 2 6

Approximate values, U.S. and worldwide.

tighter dimensional tolerances with higher mechanical performance, but there is also a tendency to develop undesirable stresses (orientations) if the processes are not properly understood or controlled. A major exception is reinforced plastics processing at low or contact pressures. Regardless of the process used, its proper control will maximize performance and minimize undesirable process characteristics. Practically all processing machines can provide useful products with relative ease, and certain machines have the capability of manufacturing products to very tight dimensions and performances. The coordination of plastic and machine facilitates these processes. This interfacing of product and process requires continual updating because of continuing new developments in manufacturing operations. The information presented throughout this book should make past, present, and future developments understandable in a wide range of applications [1-14, 62-69, 537-673]. Most products are designed to fit processes of proven reliability and consistent production. Various options may exist for processing different shapes, sizes, and weights (see Table 7-2). Parameters that will help one to select the right options are 1) setting up specific performance requirements; 2) evaluating materials' requirements and their processing capabilities; 3) designing parts on the basis of material and processing characteristics, considering part complexity and size (see Fig. 7-2) as well as a product and process cost comparison (see Table 7-3); 4) designing and manufacturing tools (molds, dies, etc.) to permit ease of processing; 5) setting up the complete line, including auxiliary equipment (Fig. 7-1 and Chapter 8); 6) testing and providing quality control, from delivery of the plastics through production to the product (see Chapter 9); and 7) interfacing all these parameters by using logic and experience or obtaining a required update on technology. Polymers usually are obtained in the form of granules, powder, pellets, and liquids. Processing mostly involves their physical change (thermoplastics), though in some cases chemical reactions occur (thermosets) (Chapter 2). A variety of processes are used. One group consists of the extrusion processes (pipe, sheet, profiles, etc.). A second group takes extrusion and in certain cases injection molding through an additional processing stage (blow molding, blown film, quenched film, etc.). A third group consists of injection and compression molding (different shapes and sizes), and a fourth group includes various other processes such as thermoforming, calendering, and rotational molding. The common features of these groups are 1) mixing, melting, and plasticizing; 2) melt transporting and shaping; 3) drawing and blowing; and 4) finishing. Mixing, melting, and plasticizing produce a plasticized melt, usually made in a screw (extruder or injection).

~

\11

N

2,B

2, A

B. Short sections can be molded. C. Also caleoderiDg process.

I, C

Extrusion

2. SecondaJy process. A. Combine two or more pans with ultrasonics, adbesives, etc.

1. Prime process.

Bottles, necked containers, etc. Cups, trays, open containers, etc. Tanks, drums, large hollow shapes, etc. Caps, covers, closures, etc. Hoods, housings, auto parts, etc. Complex shapes, thickness changes, etc. Linear shapes, pipe, profiles, etc. Sheets, panels, laminates, etc.

Injection Molding

2

Blow Molding

2 2

2

Reaction Injection Molding

2

2, A

2, A

forming

Thenno-

2

Rotational Molding

2

2,B

Compression and Transfer Molding

Table 7-2. Examples of Competitive Processes versus Different Products

2

2

2

2

2

Matched Mold, Spray-up

THE PROCESSING OF PLASTICS 593

Table 7-3. Cost Comparison of Plastic Products and Different Processes (Cost Factor x Material Cost = Purchased Cost of Product) Cost Factor Process Blow molding Calendering Casting Centrifugal casting Coating Cold pressure molding Compression molding Encapsulation Extrusion forming Filament winding Injection molding Laminating Match-die molding Pultrusion Rotational molding Slush molding Thermoforming Transfer molding Wet lay-up

Overall

Average

111.6 to 4 11/2 to 5 11/2 to 3 11/2 to 4 11/2 to 5 11/2 to 5 1% to 10 2 to 8 111.6 to 5 5 to 10 11/8 to 3 2 to 5 2 to 5 2 to 4 11/4 to 5 11/2 to 4 2 to 10 I1h to 5 11/2 to 6

11/8 to 2 21/2 to 31/ 2 2 to 3 2 to 4 2 to 4 2 to 4 11/2 to 4 3 to 4 11/8 to 2 6 to 8 13/16 to 2 3 to 4 3 to 4 2 to 31/ 2 11/2 to 3 2 to 3 3 to 5 13/4 to 3 2 to 4

Melt transporting and shaping involve applying pressure to the hot melt in order to move it through a die or into a mold. The drawing and blowing technique stretches the melt to produce orientation of the different shapes (blow molding, forming, etc.). The final feature of processing, finishing, is the usual solidification of the melt. Many product designs are inherently limited by the economics of the process that must be used to make them. For example, TSs cannot be blow molded, and to date they have limited extrusion possibilities. Many hollow parts, particularly very large ones, may be produced more economically by the rotational process than by blow molding. The need for a low quantity of parts may eliminate certain molding processes and indicate the use of casting or others. The extrusion process has fewer problems with TPs than does injection /,

Blow Molding Injection Molding

I'

Compression

~

Thermoforming

I'

I~ "

Extrusion

/

Large

Small Part Size

Figure 7·2. A design guide for processing characteristics in regard to size and complexity.

594 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

molding but has greater problems in dimensional control and shape. However, as the plastic leaves the extrusion die, it can be relatively stress-free. It is drawn in size and passed through downstream equipment from the extruder to form its shape as it is cooled, usually by air and/or water. The dimensional control and the die shape required to achieve the desired part shape are usually solved by trial-and-error settings, with the more experience one has in the specific plastic and equipment being used meaning the less trial time needed. Other analyses can be made. Compression and injection molds, which are expensive and relatively limited in size, are employed when the production volume required is great enough to justify the molds' costs and the sizes are sufficient to fit available equipment's limitations. Extrusion produces relatively uniform profiles at unlimited lengths. Casting is not limited by pressure requirements and large sheets, for example, can be produced. Calendered sheets are limited in their width by the width of the material's rolls, but are unlimited in length. Vacuum forming is not greatly limited by pressure, although even a small vacuum distributed over a large area can build up an appreciable load. Blow molding is limited by equipment that is feasible for the mold sizes. Rotational molding can produce relatively large parts. The tendency of injection molding and extrusion to align long chain molecules in the direction of flow results in their having markedly greater strength in that direction than at right angles. With an extruded pressure pipe, for example, its major strength could be in the axial/machine direction when the major stresses in the pipe wall are circumferential. With the proper processing-condition controls, the directional properties required can be obtained. If in an injection mold the plastic flows in from several gates, the melts must unite or weld where they meet. This process may not be complete, however, especially with filled plastics, so the welds may be points of weakness. Careful gating with proper processing control can allow welds to occur where stresses will be minimal. The nature of the process may have profound influence on such properties as lzod impact strength. Figure 7-3 compares the impact strengths of three PP formulations that were processed either by injection or compression molding. In this example the 1M process resulted in a drastic reduction of impact strength over that offered by the CM. This situation can be reversed by varying the processing conditions. Certain products are most economically produced by fabricating them with conventional machining out of compression-molded blocks, laminates or extruded sheets, rods or tubes. It may be advantageous to design a product for the postmolding assembly of inserts, to gain the benefit of fully automatic molding and automatic insert installation, even though inserts could greatly increase the cost of TP parts. The choice of molder and fabricator places no limits on a design. There is a way to make a part if the projected values justify the price; any job can be done "at a price." The real limiting factors are tool-design considerations, material shrinkage, subsequent assembly or finishing operations, dimensional tolerances, allowances, undercuts, insert inclusions, parting lines, fragile sections, the production rate or cycle time (see Fig. 74), and the selling price. Applying the following principles, applicable to virtually all manufacturing processes, will aid the designer in specifying parts that can be produced at minimum cost: 1) maintaining simplicity; 2) using standard materials and components; 3) specifying liberal tolerances; 4) employing the most processable plastics; 5) collaborating with manufacturing people; 6) avoiding secondary operations; 7) designing what is appropriate to the expected level of production; 8) utilizing special process characteristics; and 9) avoiding processing restrictions.

THICKNESS, mm

9

0

0.5

1.0

2.0

1.5

2.5

3.0

3.5 12 II

8 M.I.

f!

'"

I

10~

I ....

9 I....

.... 7

l.?

.2-

l.?

6

8

.... 5 U

7

Z

UJ

c:: .... Vl

....

I l.?

Vl

....

U

6 :E

....

4

5 Il.?

W

W

3: 3

all materials injection

l.?

Z

4 3: l.?

mOldOO]

....J ....J

< 2

IJ..

0

c:: ....

< c..

< c.. ~

Z

UJ

...==- =cb:-....._..::-.:-iO 0

0.02

0.04

0.08

0.06

0.10

0.12

Z

3 :J ....J 2

< IJ..

0 0.14

THICKNESS, in.

Figure 7·3. An example of the effect of processing conditions on lzod impact strength for PP with different melt index flow behaviors.

Part thickness, in.

2.5

3.75

5.0

6.25

7.5

8.75

10

010

0.15

0.20

0.25

0.30

0.35

0.40

110 .c

II;

.,u

90

i:[

70

.,

50

Part thickness, mm

Figure 7·4. An example of cycle time during the injection molding of TPs as a function of part thickness. 595

596 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

TOLERANCES AND SHRINKAGES After preliminary study, the designer has to define the part geometry. This process usually passes through several stages, beginning with preliminary drawings and sketches that indicate the basic design and functions. More detailed sketches will show the appropriate wall thickness, ribs, radii, and other structures, including the tolerances that are required to be met. Chapter 3 included a review on tolerances with the factors that influence how to predict part shrinkages (see Table 7-4 through 7-6). To meet tolerances or shrinkages needs more than applying simple lItithmetic-the part designer and toolmaker, the plastic's behavior and particularly its fabrication method all have to participate together [10-12,

62-68, 551-95]. Using the calculated shrinkage theory can dictate how much oversize to cut the tool (the mold or die) if a part has a relatively simple shape. For other shapes some critical key dimensions of the part will, more often than not, not be as predicted from the shrink allowance, particularly if the item is long, complex, or tightly toleranced. The important factors that influence the shrinkage of a specific plastic in using a specific machine, such as injection molding, by causing it to vary and not follow the values like those in Table 7-4-6, are flow direction, wall thickness, flow distance, and the presence of reinforcing fibers. Determining shrinkage involves more than just applying the appropriate correction factor from a material's data sheet. Shrinkage is caused by a volumetric change in a plastic as it cools from a molten to a solid form. As discussed in Chapters 3 and 7, shrinkage is not a single event but occurs over a period of time. Most of it happens in the mold, but it can continue for up to twenty-four to forty-eight hours. This so-called postmold shrinkage is when a part might be constrained in a cooling fixture. Additional shrinkage can occur when frozen-in stress is relieved by annealing or exposure to high service temperatures. The main considerations in mold design affecting part shrinkage are to provide adequate cooling, proper gate size and location, and structural rigidity [10-12]. Of the three, cooling conditions is the most critical, especially for crystalline resins (see Chapter 2). The cooling system must be adequate for the heat load. Slow cooling increases shrinkage by giving resin molecules more time to reach a relaxed state. In crystalline types, having longer cooling time leads to a higher level of crystallinity, which in turn accentuates shrinkage. Proper cooling, along with having an overall melt-flow analysis of how the material will react in the mold, by the mold designer, will eliminate or at least be capable of controlling the potential problems of shrinkage and warpage. This analysis can also include the best gate locations. A number of the computer-aided flow-simulation programs now offer modules designed to forecast part shrinkage-and, to a limited degree, warpage-from the interplay of resin and mold temperatures, cavity pressures, stress, and other variables in mold-fill analysis. The predicted shrinkage values in various areas of the part should be used as the basis for sizing the mold cavity, either by manual input or feed-through to the mold-dimensioning program. All the programs can successfully predict a certain amount of shrinkage under specific conditions (see Appendix D, under "Shrinkage").

MODEL BUILDING Model building, prototype construction, soft tooling, and pattern making are the key steps in product design. These stages, which are usually expensive and time consuming, often

Table 7-4. Guidelines for Nominal TP Mold Shrinkage Rates Using ASTM 1/ 2 -inch-thick Test Specimens

1/4 -

and

Avg. Rate* per ASTM D 955 Material ABS Unreinforced 30% glass fiber Acetal, copolymer Unreinforced 30% glass fiber HOPE, homo Unreinforced 30% glass fiber Nylon 6 Unreinforced 30% glass fiber Nylon 6/6 Unreinforced 15% glass fiber + 25% mineral 15% glass fiber + 25% beads 30% glass fiber PBT polyester Unreinforced 30% glass fiber Polycarbonate Unreinforced 10% glass fiber 30% glass fiber Polyether sulfone Unreinforced 30% glass fiber Polyether-etherketone Unreinforced 30% glass fiber Polyetherimide Unreinforced 30% glass fiber Polyphenylene oxide/PS alloy Unreinforced 30% glass fiber Polyphenylene sulfide Unreinforced 40% glass fiber Polypropylene, homo Unreinforced 30% glass fiber Polystyrene Unreinforced 30% glass fiber

0.125 in. (3.18 rnm)

0.250 in. (6.35 rnm)

0.004 0.001

0.007 0.0015

0.017 0.003

0.021 NA

0.015 0.003

0.030 0.004

0.013 0.0035

0.016 0.0045

0.016 0.006 0.006 0.005

0.022 0.008 0.008 0.0055

0.012 0.003

0.018 0.0045

0.005 0.003 0.001

0.007 0.004 0.002

0.006 0.002

0.007 0.003

0.011 0.002

0.013 0.003

0.005 0.002

0.007 0.004

0.005 0.001

0.008 0.002

0.011 0.002

0.004 NA

0.015 0.0035

0.025 0.004

0.004 0.0005

0.006 0.001

'Rates in in.lin. (Courtesy ICI-LNP)

597

Table 7-5. Example of Wall-Thickness Ranges and Tolerances for RP/Composite Parts Thickness Range*

Molding Method

Min., mm (in.)

Max., mm (in.)

Maximum Practicable Buildup Within Individual

Part

Hand lay-up Spray-up

1.5 (0.060) 1.5 (0.060)

30 (1.2) 13 (0.5)

Vacuum-bag molding Cold-press molding Casting, electrical

1.5 (0.060)

6.3 (0.25)

1.5 (0.060) 3 (.125)

6.3 (0.25) 115 (4.5)

Casting, marble

10 (.375)

25 (1)

EMC molding

1.5 (0.060)

25 (1)

Matched-die molding: SMC Pressure-bag molding Centrifugal casting

1.5 (0.060)

25 (1)

3 (.125)

6.3 (.25)

2: I variation possible

± 0.25 (0.010)

2.5 (0.100)

4.5% of diameter

5% of diameter

Filament winding

1.5 (0.060)

25 (1)

Pipe, none; tanks, 3: I around ports

Pultrusion

1.5 (0.060)

40 (1.6)

None

Continuous laminating Injection molding

0.5 (0.020)

6.3 e14)

None

±0.4 mm for 150-mm diameter (0.015 in. for 6-in. diameter); ±0.8 mm for 750-mm diameter (0.030 in. for 30-in. diameter) Pipe, ± 5%; tanks, ± 1.5 mm (0.060 in.) 1.5 mm, ±0.025 mm e/16 in. ±O.OOI in.); 40 mm, ±0.5 mm (11/2 in. ± 0.020 in.) ± 10% by weight

0.9 (0.035)

13 (0.5)

1.3 (0.050)

13 (0.5) 6.3-13 (0.25-0.50)

Min. to max. possible 2: I variation possible 3: I possible as required

Rotational molding Cold stamping

1.5 (0.060)

No limit; use cores No limit; use many cores No limit; over three cores possible 3-13 mm e/s-1/2 in.) 3-115 mm e/s-41/2 in.) 10-13 mm; 19-25 mm e/s-1/2 in; 3/4-1 in.) Min. to max. possible Min. to max. possible

Nonnal Thickness Tolerance, mm (in.) ±0.5 (0.020) ±0.5 (0.020) ±0.25 (0.010) ±0.5 (0.020) ±0.4 (0.015) +0.8 (.031)

±0.13 (0.005) ± 0.13 (0.005)

± 0.13 (0.005) ±5% ±6.5% by weight; ± 6.0% for flat parts

*Thickness may be varied within parts, but prolonged cure times, slower production rates, and the possibility of warpage may result. If possible, the thickness should be held unifonn throughout a part.

598

THE PROCESSING OF PLASTICS 599

Table 7-6. Recommended Dimensional Tolerances for RP/Composite Parts Dimension, mm (in.)

Class A* (fine tolerance), mm (in.)

Class Bt (nonna! tolerance), mm (in.)

0-25 (0-1) 25-100 p-4) 100-200 (4-8) 200-400 (8-16) 400-800 (16-32) 800-1,600 (32-64) 1,600-3,200 (64-128)

±0.12 (±0.005) ±0.2 (±0.008) ± 0.25 (± 0.010) ±0.4 (±0.016) ±0.8 (±0.030) ± 1.3 (±0.050) ±2.5 (±O.loo)

±0.25 (±0.010) ±0.4 (±0.016) ±0.5 (±0.020) ±0.8 (±0.030) ± 1.3 (±0.050) ±2.5 (±O.loo) ±5.0 (±0.2oo)

Class C** (coarse tolerance), mm (in.)

±0.4 ±0.5 ± 0.8 ± 1.3 ±2.0 ±3.8 ±7.0

(±0.016) (±0.020) (± 0.030) (±0.050) (±0.080) (±0.150) (±0.280)

'Class A tolerances apply to parts compression molded with precision matched-metal molds. BMC, SMC, and preform an: included. tClass B tolerances apply to parts press molded with somewhat less precise metal molds. Cold-press molding, casting, centrifugal casting, rotational molding, and cold stamping can apply to this classification when molding is done with a high degree of care. BMC, SMC, and preform compression molding can apply to this classification if extra can: is not used. "Class C applies to hand lay-up, spray-up, vacuum bag, and other methods using molds made of RP/C material. It applies to parts that would be covered by Class B when they an: not molded with a high degree of care.

account for more than half the time taken for design. Almost every type of product design involves creating one or more prototypes prior to production_ Different modeling systems that have recently become available now permit making parts in a matter of a few hours or overnight where previously this had taken from days to months [11]. For industrial designers using CAD, even seeing a part on a high-resolution graphics screen is sometimes not enough. The physical design can bring to life a high-tech design, along with formerly unnoticed flaws. By quickly forming three-dimensional conceptual models from design ideas, designers can evaluate a design concept, demonstrate its feasibility, then sell the new idea. The process called stereolithography is basically a method for building complex parts automatically, by successively printing across sections of photopolymerized plastics on top of each other until all the thin printed layers can be joined together to form a whole part. The chemical key to the process, photopolymerization, is a well-established technology in which a photo initiator absorbs UV energy to form free radicals that then initiate the polymerization of the liquid monomers. The degree of polymerization is dependent upon the total amount of light energy absorbed. This system enables having precise control over the process and constructing parts with complex geometries [587-88].

MOLDS AND DIES Molds are used to produce components by one of the usual molding or forming procedures. Dies are used for extruded, stamped, and some heat-formed products. Production molds or dies are usually made from steel for high-pressure molding that requires heating or cooling channels, strength to resist the forming forces, and wear resistance to withstand the wear of plastics, particularly that which has glass and other abrasive fillers. Most blow molds are cast or machined from aluminum, beryllium copper, zinc, or Kirksite [10-12]. The design and construction of a mold plays a significant role in the dimensional integrity of its product (see Figs. 7-5 and 7-6). The cavity that forms the final product can be shaped using a variety of steel removal methods from jig grinding to wire-feed electrical discharge machining. Each removal method has a corresponding range of vari-

TME M OL.DMAKER$ DI VISION

THE SOCIETY OF THE PLASTICS INDUSTRY, INC.

3150 Des PlaInes Avenue (RIver Roadl. Des j)lalnflS III 600'8 hrephone 31212976150

QUOTE NO. _ _ __ OATE _ _ __ DELIVERY REO _ _ __

FROM

TO

Genllemen: Please submil your Quolal lon for a mold as per following specifications and drawin gs: COMPANYNAME __ __ _ Name 1. _ _ _ __ __ __ _ ~~=~---------~~~---~~---_ BIP No _ _ _ _ _ _ _ _ _ _ _ Rev No. _ _ _ No. Cav. _ __ 2. BI P No Rev. No. _ _~ No . Cav _ __ of 3. BI P No. Rev. No. _ _ _ No. Cav. _ __ Partls No. of Cavities :

Design Charges:

Prfce:

Delivery:

Other (specify) Transfer Type of Mold: 0 Injection r CompressIon Special Features Mold ConstNction Leader Pins & Bushings in K.O. Bar Standard Spring Loaded K.O. Bar 0 3 Plate C Inserts Molded In Place Stripper Spring Loaded Plate Hot Runner l Knockout Bar on Stationary Side Insulated Runner r Accelerated K.O. Other (Specify) _ __ Positive K.O. Return ,... Hyd. Operated K.O Bar Mold Base Steel ~ Parting Line Locks 0 111 Double EjectIon 0 112 Other (Specify) _ __ D .3

o o o o o

o o o

Cores Cavilies Hardened Pre·Hard Other (Specify) _ __

o o o

Finish Cavities Cores SPE!SPt Mach. Fin ish Chrome Plate Texture Other (SpeC Ify) _ __

_ __ Material Cores Ca.itles [J Tool Steel r Beryl. Copper o Steel Sinkings [ Other (Specify) _ __

o o o

Press Ctamp Tons _ _ _ __ MakelModel _ _ _ __

Coolin Ca.ltles Core f Inserts CJ ~ Reta iner Plales r Olher Plates 0 r Bubblers tJ Other (Specify) _ __

Cores Cavities T 01 Gate Side Action K.O. Pins Cores Cavities Blade K.O. Angle Pin C Edge o Sleeve n Center Sprue Hydraulic Cyl Stripper All Cyl 'J Sub·Gate D Air D Pin Point Posilive Lock Special Lifts , Other (SpeCIfy) _ __ Cam ::J Unscrewing (Auto) K.O Actlvaled Spring Ld. Removable Inserts (Hand) Olher (SpeCIfy) _ __ Other (Specify) _ __ Design by: 0 Moldmaker Customer Type 01 Design: 0 Detailed Design Layout Only Limit Switches: 0 Suppl ied by _ _ _ _ _ _ _ __ Mounted by Moldmaker Engraving: 0 Ves L' No Approxlmlte Mold Size: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

o o o o o o

Huters Supplied By: [J Moldmaker Customer Duplicating Clsts By: OMoldmaker "Customer Mold Function Try·Out By: rl Moldmaker - Customer Tooling Modells or Masterls By: r Moldmaker r Customer Customer Try·Out Material Supplied By: ~ Moldmaker Terms subject to Purchase Agreement. TillS quotation holds for 30 days . Spec ial Instructions: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ __ The prices Quoted are on the baSIS of piece part print . models or designs subm itled or supplIed. Should there be any change in the fina l design. prices are subject to change By _ _ __ _ _ _ _ _ _ _ _ _ _ __ _ _ ____ Title _ _ _ _ _ _ _ __ __ ___ OlllrlbuHon~ Use or "''11$ 3 pari form IS 'ecommenOed as toliowS 1) Wh ile oiInd yellOW ~lI!Inl Wl lh leoue:,.1 to QuOle Plnll, . m. i nl.,"e~ m ... ell1j'~ Itle 21 White onQtnal returned Wllh QuOHlllOn YlI!IlIow '1I!IlalneCi In Moldmaker's achve hIe

Figure 7·5. A guide for mold quotation. Courtesy Moldmakers Division, Society of the Plastics Industry.

600

a-

S

Dls!on PIOC"'"CI on QYlnhl

molO SP9C I'lc. 1I0ns

'.Que.1

Mole Din

eore pins l ie

Cort and ('''' ''Y Shde . no ,nSIUS

... chln lng

PoII,'.. n; Ailimbly

•• nchlng Dell llino Fill ing

o

o

Re l'IS!' to' pfOCuCI1on

OC •• ,.... ( "gln.e.,,"; ,.Ie.se

Ptotess lIanOl ros.

In Dl."1 I'l-O,, "I) C In'C»etltonfS) TOuch·uptS)

C de lllleO I"S~Cljon Correctlonl 10uc.n·uD

Vendor I'lout (add ,tlon.'1

CorrleltOns.

M OIO JunellOnl"g M. ,ol dimenS Ions

V.ndo, I'Y-oul (10,.1 ,

Figure 7-6. Time considerations must be allowed for in going from designing to producing an injection mold able to fabricate products that meet performance requirements. The time-line events in weeks provide a guide on events that could occur.

A.'1lew mold coneeOIl, Fln. hze ,,,oauci Clraw ,ng Pllet order Al le lse Of.wlng Pll ee orael

A.".e.... QUO"I

'Isue QuOl1

SCleen canOlo.1t vendors

St.

$e1 ..CI mOlding mac'.. n.

Decide numbe, Of elvllI"

DeClOI'

602 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

ation. Although the tolerances of these processes are geometry dependent, some processes are more accurate than others. Studies can be performed to determine an average range of variation for each method. Among other important tool design and construction considerations for determining tool performance for a given dimension are the mold's construction details, such as its main parting lines. All these factors add to the overall variations in related dimensions of the mold's products.

DRYING HYGROSCOPIC PLASTICS Of the various plastics available, such TPs as nylon, PC, PMMA, PUR, PET, and ABS are among those categorized as hygroscopic. These absorb moisture, which then has to be removed before the plastics can be converted into acceptable products. Low concentrations, as specified by the resin's supplier, can be achieved through having efficient drying systems and properly handling the dried resin prior to and during molding or extrusion. Drying hygroscopic resins should not be taken casually. The simple tray dryers or mechanical convection hot-air dryers that may be adequate for nonhygroscopic resins are simply not capable of removing water to the degree necessary for the proper processing of hygroscopic resins or their compounds, particularly during periods of high humidity. The effect of having excess moisture manifests itself in various ways, depending on the process being employed. The common result is a loss in both mechanical properties (see Fig. 7-7) and physical properties, with splays, nozzle drool between shot-size control, sinks, and other losses that may occur during processing. The effects during extrusion can also include gels, trails of gas bubbles in the extrudate, arrowheads, wave forms, surging, lack of size control, and poor appearance.

HEAT HISTORY, RESIDENCE TIME, AND RECYCLING The process of heating and cooling TPs can be repeated indefinitely by granulating scrap, defective parts, and so on. During the heating and cooling cycles of injection, extrusion, and so on, the material develops a "time to heat" history, or residence time. With only limited repeating of the recycling, the properties of certain plastics are not significantly affected by residence time. However, some TPs can significantly lose certain properties.

OO~----~~---+----~r---~~----~

.!

...5~ 8O..---+--~--+---..:t----f o.,. 70~----~-----+--~~+-----~--~~ 80~----;------+------~~--;-----~

~~----~----~~--~~----~----~ .01 "" H,O

Figure 7-7. An example of the effects of moisture on the mechanical properties of a hygroscopic PET upon injection molding.

THE PROCESSING OF PLASTICS 603

Some examples of change in properties for injection-molded parts were shown in Figure 6-26. If incorrect methods were used in granulating recycled material, more degradation will occur [11]. PROCESS CONTROL

Adequate process control and its associated instrumentation are essential for product quality control. The goal in some cases is precise adherence to a control point. In other cases, maintaining the temperature within a comparatively small range is all that is necessary. For effortless controller tuning and the lowest initial cost, the processor should select the simplest controller (of temperature, time, pressure, melt-flow, rate, etc.) that will produce the desired results. One example of controls used with injection molding is seen in Figures 7-8-10. Based on the process-control settings, different behaviors of the plastics will occur. Some examples of these behaviors are shown in Figures 7-11 for injection molding and 7-12 for extrusion. Regardless of the type of controls available, the processor setting up a machine uses a systematic approach that should be outlined in the machine or control manuals. Once the machine is operating, the processor methodically makes one change at a time, to determine the result. Two basic examples are presented in Figures 7-13 and 7-14 to show a logical approach to evaluating the changes made with any processing machine. As the injection-molding machine is very complex with all the controls required to set it up, these examples refer to the injection-molding process [11]. TROUBLESHOOTING

With all types of plastics processes, troubleshooting guides are set up to take fast, corrective action when products do not meet their performance requirements [10-12]. This problem-solving approach fits into the overall fabricating-design interface. One brief example of troubleshooting an RP/composite is in Table 7-7. A simplified approach to troubleshooting is to develop a checklist that incorporates the basic rules of problem solving such as 1) have a plan, and keep updating it, based on

Table 7-7. Troubleshooting RP Processses Problem Nonfills

Excessive thickness variation

Possible Cause Air entrapment Gel and/or resin time too short Improper clamping and/or lay-up

Blistering

Demolded too soon Improper catalytic action

Extended curing cycle

Improper catalytic action

Solution Additional air vents and/or vacuum required Adjust resin mix to lengthen time cycle Check weight and lay-up and/or check clamping mechanisms such as alignment of platens Extend molding cycle Check resin mix for accurate catalyst content and dispersion Check equipment, if used, for proper catalyst metering Remix resin and contents; agitate mix to provide even dispersion

i

"

, " =:.... <

~

Holtl ptessute Holtl ptessute Holtl ptessute Builtl up Time Builtl uP Rate sctewRPM SCtew Contlitio n sc w,ip • finte al fill Speetl ovetali Fill sp d Machine Ftictee io n

,

BOOSI ?tessute I~ Boosl floW ck?teSsute ---- / / t:r, Ba Back ptes sute Bul\tl\lp,\me Back ptessute BUlltl UP Rale

Maletial

?ull Back o·,slance

=-=-- -lit! S\\OI Si2e o·,slance

• Clamp Open _ Holtl "met lime Seiling

=

=_:========= =:==:=-=__=-==__=-== -=__=-==__=-==_ _ _O • C Cl"" p' O Cle,w "~e osin'"g ,im Cu team,im

rwerage "o \d Pressure

olding mach ine controls .

Figure 7 .8 . Injection m

ressure

f-.-......... Pea~ p\a stic P

~ ~

\ic pressure

pea~ "ydrau

cyc\eTime

#-. . . . . .

iscOSiW~

Mell comptessib iliW

0 / ~ MeiIV

screVl RPM preSSure

.-?~Y'"·"'_'"' ,~ Bac~ NOllie Mell,e Moltl ,empetampetalute lute

'Jirgln

FilletS Ma\e•tla\~ ~RelnlotcReeglmintel nIS

a=> <.n

2·Plale 3·P1311 Slacked CamAction. COrl Pullers Inserls Olhers

Tolerances

Part Shape

Figure 7-9. Control factors important to mold operation.

MOld Design Cost and Cycle Time

MOld Cooling Aulysl.s

Type Bushln.

Runner Diameter Cavlly Lo cations Others

Temperatur.~ Runner Length

Tool L11e

Humber 01 Paris Required

606 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK CAE - IJORKSTATlON

CPU

\linch • • 1,r Disk

IBMIAT

PRODUCTION PLANNING AND CONTROL (PPe>

CPU

Ho..ster

\r'lnch~s1fr

Top' Cortrldgl'

Disk

Tap .. Cor1rldg.

Serial Bus I I I 32 INJECTIONI MOLDINGI MACHINES:

PER LINE I I

L-O *

6Q[I\!!l.BLe8~MGELHifI!Hi\!..L

TMCOHC£PT E'port Systeo for Molding

"S "CO FA [SE PP

-

MaterIal Selection MoldIng L Cost Optl.lzatlon Floo An.lysls Co.puterlzed Shrinkage E~al~atlon Past-Processor

PPP - Production Planning

SOH

Set Data Manage.ent

PDM

Production Data

Pa(~age

Manag~ment

Figure 7-10. An overview of the Computer-Integrated Injection Molding (CIIM) machine from TMConcept, which includes shrinkage control (Metstal is a Swiss machine manufacturer).

the experience gained; 2) watch the processing conditions; 3) change only one condition or control at a time; 4) allow sufficient time for each change, keeping an accurate log of each; 5) check housekeeping, storage areas, granulators, etc.; and 6) narrow the range of areas in which the problem belongs-that is, machine, molds-dies, operating controls, material, part design, and managyment.

INSPECTION Inspection variations are often the most critical and most-overlooked aspect of the tolerance of a fabricated part. Designers and processors base their development decisions on inspection readings, but they rarely determine the tolerances associated with these readings. The inspection variations may themselves be greater than the tolerances for the characteristics being measured, but without having a study of the inspection method capability this can go unnoticed. Inspection tolerance can be divided into two major components: the accuracy variability of the instruction and the repeatability of the measuring method. The calibration and

THE PROCESSING OF PLASTICS 607

accuracy of the instrument are documented and certified by its manufacturer, and it is periodically checked. Understanding the overall inspection process is extremely useful in selecting the proper method for measuring a specific dimension. When all the inspection methods available provide an acceptable level of accuracy, the most economical method should be used. As the overall fabricating tolerance is analyzed into the sources of its variation components, the potential advantage of analytical programs comes into play with their ability to efficiently process all these factors. All the empirical tolerance ranges for each tooling

f

SHRINKACE

WITH FLOW

-

ACROSS FLOW

t

-

SHRINKAGE IN LINE OF FLOW

MOLD TEMPERATURE-+

t

SHRI N KA.GE

WITHFLDW

-

GATE

SHRINKAGE

\ . RESTRICTED GATE

~, PACKI~IG

t

i'OST MOLDING SHIliNKAGE OF TALLINE POLYMER

&"

MELT TEMPERATURE_

t

SHRINKAGE

R~

~ CAVITY

TI;;CKNESS_

t PACKING TIME

ANNEALED

AGEING TIME ~

---...

t

TIt.4E_

COLD MOLD

AREA

DIFFERENTIAL SHRINKAGE- -_ _ _ _ _-_

CAVITY TH ICKNESS---'"

t

-------

~

Figure 7·11. Examples of how injection-molding machine settings affect certain properties of plastics, including shrinkage.

608 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

method and inspection method are stored in data files for easy retrieval. For each critical dimension the program sums all the component tolerances and computes a ± overall tolerance for each critical dimension. The program then provides a tabulated estimate of the achievable processing tolerances and pinpoints the areas that contribute most to the overall required tolerance. This information is useful in identifying the needed tolerances, which in practice can be expected to exceed the initial design tolerances.

t

FILM OPTICAL PROPERTIES

--------

t

HAZE VALUE

DIE LAND LENGTH---.

DIE TEMPERATURE_

t

ALM IMRO.CT STRENGTH

HIGH

BLOW UP RATIO

+

FILM IMPACT STRENGTH

LO

FILM IMPACT STRENGTH

HIGH BLOW UP RATIO

~

/ 8l.C1tY UP RATIO ___

FREEZE LINE HEIGHT_

t

--------

t

TENSILE STRENGTH

LOW BLOW UP RATIO

~, DIRECTION

MACHINE DIRECTION

t

ALM TEAR STRENGTH

COOLING RATE

-....

~-(") IRECTION

~~~WJN(MD) BLOW UP RATIO

----p.

BLOW UP RATIO - - - - .

+ -

LOW BLOW UP

FILM TEAR STRENGTH

--JD MD

-

TO -MD HIGH BLOW UP

FREEZE LINE HEIGHT---.·

Figure 7-12. An example of how blown film-extrusion machine settings can affect the properties of plastics.

!

Flash area

~

a

'"'"

D-

...E a:

I Short shot area

----4.~

Mold temperature

Figur.e 7-13. A two-dimensional molding-area diagram (MAD) that plots injection pressure (that is, ram pressure) against mold temperature. Within this area all parts meet the performance requirements. However, rejects can develop at the edge of the diagram, because of machine and plastics variations .

.;;;

0.

~

::: ~

a.

Q)

E <0

a:

t

Figure 7-14. After a three-dimensional molding-volume diagram (MVD) is constructed, it can be analyzed to find the optimum combination of melt temperature, mold temperature, and injection or ram pressure. 609

610 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Establishing initial design tolerances is often done on an arbitrary, uninformed basis. If the initial estimated tolerance proves too great, a lower-shrink plastic could be used to reduce the shrinkage range. However, if the key dimension was across a main parting line, the tooling could be redesigned to eliminate the condition and consequently reduce variation from tool construction. Even with all these data processed by a computer, estimating tolerances is difficult if they are not properly interrelated with the highly dependent factors of the part and tooling designs. INJECTION MOLDING

The injection-molding (1M) process is greatly preferred by designers because the manufacture of parts in complex shapes and three dimensions can be more accurately controlled and predicted with 1M than with other processes. As its method of operation is much more complex than others, 1M requires a thorough understanding. Figures 7-15 and 716 show schematics of the load profile and the molding cycle that highlight the way in which the melt is plasticized (softened) and forced into the mold, the clamping system for opening and closing the mold under pressure; the type of mold used, and the machine controls [1, 2, 10, 12,576-600]. Plastic moves from the hopper onto the feeding portion of the reciprocating extruder screw. The flights of the rotating screw cause the material to move through a heated extruder barrel where it softens (is made fluid) so that it can be fed into the shot chamber (front of screw). This motion generates pressure (usually 50-300 psi [0.35-2.07 MPa)), which causes the screw to retract. When the preset limit switch (or a position transducer, on newer machines), is reached the shot size is met and the screw stops rotating. At a preset time the screw acts as a ram to push the melt into the mold. Injection takes place

Table 7-8. An Example of 1M Processing Temperatures used with Heat Resistance for Engineering TP*

Type

(0C)

Processing Temperature. deg. F eC)

Polyetheretherketone (PEEK) Polyphenylene sulfide (PPS) Polyaryleneketone

Semicrystalline

290 (143)

650 (343)

SemicrystaIline

185 (85)

630 (332)

SemicrystaIline

400 (204)

700--780 (371-416)

Polyarylene sulfide Polyetherimide (PEl)

Amorphous Amorphous

Polyarylether Polyethersulfone (PES) Polyamide-imide (PAl)

Amorphous Amorphous Amorphous

410 (210) Varies 450 (232) 545 (285) 476 (247) 510 (266) 470 (243)

625-650 (329-343) Varies 575-650 (302-343) 650-700 (343-371) 650 (343) 575 (302) 650 (343)

Polyimide

Pseudothermoplastic

Tg deg. F Polymer

·Typical commodity TPs use about 400" to 550"F (204° to 2SS°C},

480 {249) 482 (250) 536 (280) 536 (280)

680 660 660 660

(360) (349) (349) (349)

THE PROCESSING OF PLASTICS 611

at high pressure (up to 30,000 psi [207 MPa] melt pressure in the nozzle). Adequate clamping pressure must be used to eliminate mold opening (flushing). The melt pressure within the mold cavity ranges from 1 to 15 tons/sq. in. and is dependent on the plastic's rheology/flow behavior (see Chapter 2). Time, pressure, and temperature controls indicate whether the performance requirements of a molded part are being met. The time factors include the rate of injection, duration of ram pressure, time of cooling, time of plastication, and screw RPM (see Fig. 7-16). Pressure factors are injection high and low pressure, back pressure on the extruder screw, and pressure loss before the plastic enters the cavity, which can be caused by a variety of restrictions in the mold. The temperature factors are mold (cavity and core), barrel, and nozzle temperatures, as well as the melt temperature from back pressure, screw speed, frictional heat, and so on. The large number of variables summarized in Figure 7-8 will cause part changes if not properly controlled. The basic settings for these variables are provided by the plastic's producer: the injection barrel temperature, melt temperature (see Table 7-8), the cavity melt pressure, and so on. However, the final settings are determined by the processor on a specific machine and mold. Even though most of the literature on processing specifically identifies or refers to thermoplastics (TPs), as in this book, some thermosets (TSs) are used (TS polyesters, phenolics, epoxy, etc.). The TPs reach maximum heat prior to entering "cool" mold cavities, whereas the TSs reach their maximum temperature in "hot" molds.

Productivity Productivity is directly related to cycle time. There usually is considerable common knowledge about a part's geometry and process conditions that will provide a minimum cycle time. Practices such as using thinner wall sections, cold or hot runners for TPs or hot or cold ones for TSs (see Figs. 7-17 and 7-18), narrow sprues and runners, the optimal size and location of coolant (or heat) channels, and lower melt or mold heat, will when possible decrease the solidification time.

Nozzle

Screw

2.25 in. dia. (4 sq. In.)

Sprue

Mold Runner i - --
~!:=:7~~2~~~~~~~!~;i:;=::;"]

Cylinder, HydraulIC 7.161n dla. (40 sq. in.)

Molded part 4,000 psi

15,000 psi

Figure 7-15. An example of pressure loading on plastic melt during injection molding (see Appendix B for English to metric conversions).

t-______________~CO MP LETE

CYCLE~____________~

&0 SEC.

I NJE CTIO COOLI G 3D S EC-------1-'~ 2 S SEC

EJECTIO

1--- - -

uuu ...........

"''''''' SC R EW-R AM TR AVEl SEc..

a

N

PL AST IC COOLI G I 4 7 SEC DWELL

RAM

S SEC

rRH

TR AVEL

8 SEC

N

I

17SEC

5 SEC.,

GA E SEAL

SHRI

TI ME 17 SEC

-

MOLD

I

K AGE OCC URS ~ OLO

- ----I

3D SEC

5 S C

Figure 7-16. An example of injection molding cycle processing thermoplastics.

';'.--i.l~t:..d - CAVI1Y

AND MOLDEO PART

/"I""t<E--< --MOLD SEPARATES RUNNER PUNCH 011 fORCE PLATE

I"t"-E<E-~-


MOLD SEPARATES PUNCH 011 FORCE PLATE ~

~~~~~~~~~~~~~~~~1=~~HOT

ELECTRIC HEATED FOLD M"HIRUNNER NSULATED NOZZLE

~t--<-- MOI.O SEPARATES

Figure 7-17. Types of injection molds. 612

THE PROCESSING OF PLASTICS 613

Numerous factors affect the elapsed time required to eject a part, as different plastics can have dramatically different melt behaviors. Many of these influences are poorly, if at all,. understood. Some critical ones are the coefficient of expansion, melt rheology, thermal diffusivity, and the thermomechanical spectrum. Although the usual and important ways to optimize time are based on part design and process conditions, it can be shown that additional and significant decreases occur by using modified molding compounds via additives, alloying ratios, molecular weight distribution, and so on.

I'K...- - " - --ElECTRIC HEATER

-INSULATED RUNNER

CAVI TY AND MOLDED PART ...--~'"

-

MOLD SEPARATES I NSULATING SHELL

.........._ - - MOLTEN POLYMER

HOT MANIFOLD

I"'t"'~<--MOLO

SEPARATES

I NJECIION MOLD 1fGi~'1$~;;;~~-HEATING UNI T

Figure 7-18. Three further types of injection molds.

614 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Mathematical models of part geometry and melt flow within the mold cavity, which are available for mold design, are useful tools for optimizing cycle times. They allow a wide variety of plastics materials and process parameters to be evaluated in a convenient, cost-effective manner [10]. CAD and engineering algorithms offer a continually higher level of sophistication in determining the best heat distribution throughout the part. In actual practice, parts are not ejected according to a measurement of their internal or wall temperatures. Ejection times are set through secondary characteristics of part heat, such as the ability of the part to withstand the forces of ejection, the occurrence of sink marks or other thermal warpage, and the part's overall gloss or appearance. The prediction of sink marks appears amenable to CAD/CAE calculation. Part appearance still plays a larger role in the art of 1M.

Screw Design With practically all machines, only the cylinder temperature is directly controlled. The actual heat of the melt, within the screw and as it is ejected from the nozzle, can vary considerably, depending on the efficiency of the screw design and the method of operation. The factors affecting melt heat include the time the plastic remains in the cylinder; the internal surface heating area of the cylinder and the screw, per volume of material being heated; the thermal conductivity of the cylinder, screw, and plastic; the heat differential between the cylinder and the melt; and the amount of melt turbulence in the cylinder. In designing the screw, some balance must be maintained between the need to provide adequate time for heat exposure and to maximize output most economically. In general, heat-transfer problems have led screw designers to concentrate on making screws more efficient as heat-transfer devices. As a result, the internal design and performance of screws will vary considerably to accommodate the different plastics used. Most machines are single, constant-pitch, metering-type screws to handle the majority of plastics.

Air Entrapment Air can be entrapped in the melt during processing when plastic pellets or flakes are melted in a normal air environment, as in a plasticating extrusion process or injection barrel, compression mold, casting form, or spray system and the air cannot escape. Generally, the melt is subject to a compression load, or even a vacuum, which causes a release of air, but in some cases the air is trapped. If the air entrapment is acceptable, no further action is required. However, it is usually unacceptable, for reasons of both performance and aesthetics. Changing the initial melt temperature in either direction may solve the problem. With a barrel and screw, it is important to study the effects of temperature changes. Another approach is to increase the pressure in processes that use process controls. The particle size, melt shape, and the melt delivery system may have to be changed or better controlled. A vacuum hopper feed system may be useful. With screw plasticators, changes in screw design may be helpful. Usually, a vented barrel will solve the problem. The presence of bubbles could be due to air alone or moisture, plastic surface agents or volatiles, degradation, or the use of contaminated regrind. With molds such as those used for injection, compression, casting, or reaction injection, air or moisture in the mold cavity will be the culprit. So the first step to resolving a bubble or air problem is to be sure what problem exists. A logical troubleshooting approach can be used.

THE PROCESSING OF PLASTICS 615

Shot-to-Shot Variation During 1M, shot-to-shot variations can occur. The major causes of inconsistency are worn nonreturn valves, bad seating of a nonreturn valve, a broken valve ring, a worn barrel in the valve area, or a poor heat profile. To identify the cause, follow a logical procedure. Any problem caused by the valve will cause the screw to rotate in the reverse direction during injection. To locate the trouble, pull and inspect the valve and check the OD of the ring for wear. The inspector looks for a broken valve stud (caused by cold startup when the screw is full of plastic), bad seating of the ring or ball (the angles of the ring ID and the seat must be different, in order to ensure proper shutoff action at the ID of the ring), and a broken ring. Check the dimensions of the valve and compare them with those determined before using the machine.

Purging Purging has always been a necessary evil, consuming substantial amounts of materials, labor, and machine time, all nonproductive. In 1M it is sometimes necessary to run hundreds of pounds of resin to clean out the last traces of a dark color before changing to a lighter one. Sometimes there is no choice but to pull the screw for a thorough cleaning. Although there are few generally accepted rules on how to purge, the following tips should be considered: I) try to follow less viscous with more viscous resins; 2) try to follow a lighter color with a darker color of resin; 3) maintain the equipment; 4) keep the materials-handling equipment clean; and 5) use an intermediate resin to bridge the temperature gap (such as that encountered in going from acetal to nylon), and use a PS as a purge.

Molds A mold must be considered one of the most important pieces of production equipment in the plant. It is a controllable complex device (see Fig. 7-19) that must be an efficient heat exchanger. If not properly designed, handled, and maintained, it will not be an efficient operating device. Hot melt, under pressure, moves rapidly through the mold. Water or some other medium circulates in the mold to remove (for TPs) or add (for TSs) heat. Air is released from cavities to eliminate melt burning or voids in the part. All kinds of actions operate, including sliders and unscrewing devices [2, 10, 11]. Parts like knockout pins as well as air are ejected at the proper time. These basic operations in tum require all kinds of interactions, including such parameters as fill-time, hold pressure, and other variables, as shown in Figures 7-8 and 7-9. Each of the plastics used has special distinctive properties. Some are abrasive or corrosive; others require very tight heat control and pressure. Settings that work for one resin probably will not work for another, or a machine change to a duplicate will probably require different settings. Shrinkage requires special attention. Crystalline resins shrink more than amorphous ones (see Chapter 2). Differential shrinkage can cause warpage. With tight part tolerances it is necessary to leave more, rather than less, steel in the mold so that corrections requiring metal "cutting" can correlate processing with tolerances. CAD and CAE programs are available that can aid in mold design and in setting up the complete process [11]. These programs are concerned with melt flow to part solidification and the meeting of performance requirements. Many different factors are incorporated, including heat transfer, thermal conductivity, thermal expansion, the coefficients of friction, machine and mold operating setup, and so on.

616 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Ring - - - - - - - - - - - S p r u e Bushing

~----------Locating

'"'~""~,.,.,77~"7i:J-"-Cm""7/':,.,.,?7m"":01_Front Clamping Plate ~ClampSlot

-Front Cav. Retainer PI. of>-l'>J---Water Channels ---Cavity _ _~.....--Force (Male Cavity) Guide Pin Bushing Rear Cav. Retainer PI. Pin

~t~~~IH~~ t~~1~0~~I~~~--Push-back

Support Plate It--t++---+--Ejector Pin Sprue Lock Pin Support Pillar \ ~Ejector Retainer PI. L-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _~\ ~Ejector Plate LClampSlot Ejector Housing

Figure 7-19. A two-part standard mold.

Molding Variables versus Performance There are variables during molding that influence part performance. The information presented here shows how melt flow variables behave to influence products' properties. A flow analysis can be made to aid designers and moldmakers in obtaining a good mold. Of paramount importance is controlling the fill pattern of the molding so that parts can be produced reliably and economically. A good fill pattern for a molding is one that is usually unidirectional in nature, thus giving rise to a unidirectional and consistent molecular orientation in the molded product. This approach helps avoid warpage problems caused by a differential orientation, an effect best demonstrated by the warpage that occurs in thin center-gated disks. In this case all the radials are oriented parallel to the flow direction, with the circumferences transverse to the flow direction. The difference in the amounts of shrinkage manifests itself in terms of warpage of the disk. In order to achieve a controlled fill pattern, the mold designer must select the number and location of gates that will result in the desired pattern. Flow analysis can help by allowing the designer to try multiple options for gate locations and evaluate the impact on the molding process. This analysis often can be conducted with the product designer to achieve the best balance of gate locations for cosmetic impact and molding considerations. Figures 7-20 through 7-29 show various flow patterns, orientation patterns, and property performances (see also Chapter 3) [11]. In the practical world of mold design, there are many instances where design tradeoffs must be made in order to achieve a successful overall design. While naturally balanced runner systems are certainly desirable, they may lead to problems in mold cooling or increased cost due to excessive runner-to-part weights (see Fig. 7-17). Additionally, there are many cases such as parts requiring mUltiple gates or family molds in which balanced runners cannot be used. Flow analysis tools allow successful designs of runners to balance for pressure, temperature, or a combination of both.

SURFACE HIGHL.Y ORIENTEO

., t

f

, /

CORE ORIENTATION FROW BUL.K-SHEAR - -

\ '\

SUB-SURFACE ORIENTATION FROM HIGH SHEAR NEAR WALL

Figure 7-20. Cavity melt flow looking at a part's thickness.

-

SHEAR TIiINNING LAYER J.

-.

-

-

""

FAST~

PWG FLOW

f'--""""

~

---'----'

--

-

-- ~

ORIENTATION

--

--

-1

--I

&DW~

Figure 7-21. The effects of different fill rates.

Cavity

Parlin!! I,ne

(A)

(8)

Figure 7-22. Plastic melt does not flow uniformly through the diaphragm of plate mold (A) in the compensating phase, but spreads in a branching pattern (B). 617

Figure 7-23. Flow paths are determined by part shape and gate location. Flow fronts that meet head on will weld together, forming a weld line. Parallel fronts tend to blend, however, producing a less distinct weld line but a stronger bond.

Figure 7-24. Example of flow lines or weld lines in a telephone handset where the gate was located at the top-center part of the handle. 618

STRESS PARALLEl TO ORIENTATION

STRESS PERPENDICULAR TO ORIENTATION Figure 7-25. The effect of orientation on strength: the highest tensile strength is in the direction parallel to the orientation.

.------A + B

100

C

~

•

80~--~--~--~---~

a

2

3

4

INJECTION TIME, SECONDS

100 ~

en

VI

9

L

!

450

470

.q ~

90

(!)

80

490

STOCK TEMPERATURE~F

100 ~

:i 90 9

A

(!)

80

80

100

120

140

160

180

MOLD 'rEMPERATURE~F Figure 7-26. The effect of molding conditions on the gloss of an ABS plastic.

619

Direction of Orientation a

Direction of Orientation b Figure 7·27. An example of locating a gate to obtain the required performance of a retainer product that is subject to being flexed in service. a) Retainer edge gated; b) retainer center gated; and c) left and center retainers (between fingers) that were edge gated, with the failed retainer on the right which was center gated. 620

c Figure 7-27. (Continued)

~5

u

b4.5r--___ z

SI NGLE GATED NOTCHED SPECIMENS

g;4 ~

u 3 .5

z

0::

L&.J

3

a.. 2 .5 en o z 2

:::l

o

a.. 1 .5

b o u..

--

-----

_---.e

"""""",,- DOUBLE GATED WELD LINE "," SPECIMENS

1

460 440 STOCK TEMPERATURE - 0 F Figure 7-28. Izod impact strength of ABS plastic (1 x ! x 3 in. specimens). 380

400

420

621

622 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

-

I--~

DUCTILE

---.1

5.0

::z:

~

~

4.0

......

REMOLDED ALL SAMPLES AT 430°F

U)

CCI

.....

~

u..

::z:

3.0

= = w w

~

c::I Z

:::;)

~

w

C[

I: ~ U) ~

~

1-- BRITTLE ~I

~

2.0

::E w

C[

~

~

c::I

z

::IE

2i

.....

0

::IE

1.0

U)

CCI

C[

C

~

0 400

en

430

460 490 520 MEL T TEMPERATURE (OF)

550

580

Figure 7-29. Impact strength versus melt temperature in white ABS plastic.

Molding Techniques In addition to conventional injection molding, specialized 1M techniques are used to meet certain product requirements that result in cost reductions or having the necessary molding capabilities to produce given products. Some examples are provided. See also at the end of this chapter "Other Processes," on coining and using fusible cores, including the use of injection molding. In addition there are also those methods involving molding with rotation stretching and orienting techniques that differ from injection stretch blow molding [12], continuous molding like producing Velcro strips [12], injection-molding metals, liquid plastic injection molding, foam molding, structural sandwich molding, and parts consolidation (see Fig. 7-30) and others [12].

THE PROCESSING OF PLASTICS 623

Figure 7-30. Du Pont's Arylon polyarylate and Rynite TP-polyester resins eliminate parts and simplify assembly of Xetec Corp. 's three-dimensional circuitboard for lighting ballasts. A special two-shot molding process removes the need for metal fabrication and yields a low-cost, reliable component for use in rapid-start fluorescent lighting ballasts.

Coinjection Coinjection basically means that two or more different plastics are "laminated" together. These plastics could be the same except for color. When different plastics are used, they must be compatible in that they provide proper adhesion (if required), melt at approximately the same temperature, and so on (see Table 7-9). Two or more injection units are required, with each material having its own injection unit. The materials can be injected into specially designed molds-rotary, shuttle table, and the like [12]. The term co injection can denote different products, such as sandwich construction, double-shot injection, multiple-shot injection, structural foam construction, two-color molding, and inmolding. Whatever its designation, a sandwich configuration has been made in which two or more plastics are "laminated" together to take advantage of the different properties each plastic contributes to the structure. This form of injection has been in use since the early 1940s. Many different advantages exist. For example, 1) it combines the performance of materials; 2) it permits the use of a low-cost plastic such as a regrind; 3) it provides a decorative "thin" surface of an expensive plastic; 4) it includes reinforcements; 5) it permits the use of barrier plastics (see Chapter 4), and more. Coinjection molding is being redefined today in light of the approaches now available for molding such multicomponent parts as automotive taillights, containers, business machine housings, and so on.

624 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7·31. The three-channel coinjection system shown here simultaneously injects two different plastic melts. Courtesy of Battenfeld of America.

There are three techniques offered for multiple-component injection, called the onechannel, two-channel, and three-channel techniques (see Fig. 7·31 for the three-channel type) [12].

Gas Injection Parts can be molded by gas injection using injection-molding machines. This process is most effective and economical when used for large parts. It offers a way to mold parts with only 10 to 50 percent of the clamp tonnage that would be necessary in conventional 1M [12]. The technique-practiced in several variations, with some patented-involves the injection of an inert gas, usually nitrogen, into the melt as it enters the mold. This is not structural foam, as no foam core is produced; instead, the gas forms a series of interconnecting hollow channels in the thicker sections of the part. The gas pressure is maintained through the cooling cycle. In effect, the gas packs the plastic into the mold without a second-stage high-pressure packing in the cycle as used in 1M, which requires high tonnage to mold large parts. Molded-in stresses are minimal. The thick but hollow sections provide rigidity and do not create sink or warpage problems. The cycle time is reduced because the thick sections are hollow. As the gas is not mixed with the melt, there is no surface splay, which is typical of low-pressure structural foam molding [12]. Gas injection is now being used with commodity and engineering resins.

EXTRUSION The extruder, which offers the advantages of a completely versatile processing technique, is unsurpassed in economic importance by any other process. This continuously operating process, with its relatively low cost of operation, is predominant in the manufacture of shapes such as films, sheets, tapes, filaments, pipes, rods, in-line postforming, and others. The basic processing concept is similar to that of injection molding (1M) in that material passes from a hopper into a cylinder in which it is melted and dragged forward by the movement of a screw. The screw compresses, melts, and homogenizes the material. When the melt reaches the end of the cylinder, it usually is forced through a screen pack prior to entering a die that gives the desired shape with no break in continuity (see Fig. 7-32) [10, 12,601-25].

a~

\11

=

good adhesion. -

=

Source: Battenfeld

+

0

+

+

acetate

0

+

+

+

+ +

+

+

acetate

vinyl

Ethyl

+ +

6

+ +

616

Nylon Nylon

+

0

+

+

0

+ +

+

+

+ +

+

camonate HDPE LOPE

Poly-

+ + +

0

+

+

acrylate

meth-

Polymethyl-

+

methylene

Polyoxy-

+

+

0

+

PP

+ + +

+

+

PS

+ +

PPO

purpose

General-

+

+ +

0

0

0

PS

Impact

High-

Polytetra-

+

+

+

thalate

tereph-

methylene

+ +

+

+

+

PVC

Rigid

0

+ + +

+

0

PVC

Soft

Styrene

+ +

+

+

+

+ +

bile

Acryloni-

poor adhesion. 0 = no adhesion, blank indicates no recommendation (combination not yet tested). The addition of fillers or reinforcements leads to a deterioration of adhesion between

+

0

+

+

+

+

+

+ +

+ +

+

bile

ABS

raw materials for skin and core.

+

ABS Acrylic ester acrylonitrile Cellulose acetate Ethyl vinyl acetate Nylon 6 Nylon 6/6 Polycarbonate HDPE LDPE Polymethylmethacrylate Polyoxymethylene PP PPO General-purpose PS High-impact PS Polytetramethylene terephthalate Rigid PVC Soft PVC Styrene acrylonitrile

Materials

acryloni- Cellulose

ester

Acrylic

Table 7-9. Compatibility of Materials for Coinjection

626 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Hasin

Screw

Thermocouples

Compression Section

Breaker Plate

Figure 7·32. A cross-section of a single-screw extruder.

A major difference between extrusion and 1M is that the extruder processes plastics at a lower pressure and operates continuously. Its pressure usually ranges from 1.4 to 10.4 MPa (200 to 1,500 psi) and could go to 34.5 or 69 MPa (5,000 or possibly 10,000 psi). In 1M, pressures go from 14 to 210 MPa (2,000 to 30,000 psi). However, the most important difference is that the 1M melt is not continuous; it experiences repeatable abrupt changes when the melt is forced into a mold cavity. With these significant differences, it is actually easier to theorize about extrusion and to process plastics through extruders, as many more controls are required in 1M. Good-quality plastic extrusions require homogeneity in terms of the melt-heat profile and mix, accurate and sustained flow rates, a good die design, and accurately controlled downstream equipment for cooling and handling the product. Four principal factors determine a good die design: internal flow length, streamlining, the construction materials, and uniformity of heat control. Heat profiles are preset via tight controls that incorporate cooling systems in addition to heater bands. Barrels use forced air or water jackets. In some machines water bubbler channels are located within the screws. On leaving the extruder, the product is drawn by a pulling device, at which stage it is subject to cooling, usually by water or blown air. This is an important aspect of downstream control if tight dimensional requirements exist or conservation of plastics is desired. Usually lines do not have adequate control of the pulling device. The processor's target is to determine the tolerance required for the pull rate and to see that the device meets the requirements. Even if tight dimensional requirements do not exist, the probability is that better control of the pull speed will permit tighter tolerances and reduce the material output. As the molecules of the melt flow are aligned in the direction of the output from the die, the strength of the plastic is characteristically greater in that direction than at right angles. Depending on the product's use, this mayor may not be favorable . The degree of orientation can be controlled. The success of any continuous extrusion process depends not only upon uniform quality and conditioning of the raw materials but also upon the speed and continuity of the feed of additives or regrind along with the virgin resin (upstream). Actually, only thermoplastics go through extruders; markets have not developed to date for extruded thermosets. Variations in the bulk density of materials can exist in the hopper, requiring controllers such

THE PROCESSING OF PLASTICS 627

as weight feeders and perhaps requiring some type of packing feed, such as rams, screw packers, and so on. In extrusion, as in all other processes, an extensive theoretical analysis has been applied to facilitate understanding and maximize the manufacturing operation. However, the real world must be understood and appreciated as well. The operator has to work within the many limitations of the materials and equipment (the basic extruder and all auxiliary upstream and downstream equipment). The interplay and interchange of process controls can help to eliminate problems and aid operation with the variables that exist. The greatest degree of instability is due to improper screw design, or using the wrong screw. Proper instrumentation, particularly barrel heat, is important to a diagnosis of the problem. For uniform and stable extrusion it is important to check periodically the drive system, the take-up device, and other equipment, and compare it to its original performance. If variations are excessive, all kinds of problems will develop in the extruded product. An elaborate process-control system can help, but it is best to improve stability in all facets of the extrusion line. Some examples of instabilities and problem areas include 1) nonuniform plastics flow in the hopper; 2) troublesome bridging, with excessive barrel heat that melts the solidified plastic in the hopper and feed section and stops the plastic flow; 3) variations in barrel heat, screw heat, screw speed, the screw power drive, die heat, die head pressure, and the take-up device; 4) insufficient melting or mixing capacity; 5) insufficient pressure-generating capacity; 6) wear or damage of the screw or barrel; 7) melt fracture/sharkskin (see Chapter 2), and so on. Finally, one must check the proper alignment of the extruder and the downstream equipment. Proper alignment and isolation of the vibrators is a must for high-quality, high-speed output. Regardless of their particular designs, all extruders have the function of conveying plastic and converting it into a melt. For this purpose, both single- and multiple-screw extruders are suitable, but they all have individual characteristic features. Practical and theoretical data show that each type has its place. The single-screw machine dominates. However, other types are available, such as twin-screw extruders, which are often used to achieve improved dispersing and mixing, as in the compounding of additives [10, 12].

Screen Packs Melt from the screw is usually forced through a breaker plate with a screen pack. Extra heat develops when melt goes through the screens, so some heat-sensitive materials cannot use a screen pack. The function of a screen pack is initially to reduce the rotary motion of the melt and remove large unmelted particles and other contaminants. This situation can be related to improper screw design, a contaminated feedstock, poor control of the regrind, and so on. Sometimes screen packs are used to control the operating pressure of extruders. However, there are advantages in processing with matched and controlled back pressure, operating within the required melt pressure, as this can facilitate mixing, effectively balancing out the melt heat. In operation a screen pack is backed up by a breaker plate that has a number of passages, usually many round holes ranging from 1 to h in. in diameter. One side of the plate is recessed to accommodate round discs of wire screen cloth, which make up the screen pack (usually 40- to l00-mesh screen). Pressure controls should be used on both sides of the breaker plate to ensure that the pressure on the melt stays within the required limits. Based on the processing requirements, manual to highly sophisticated screen changers are used.

628 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Barrel and Screw Materials The majority of barrels and screws are made from special steels, which are nitrided to a minimum depth by special techniques. Low-alloy steels are sometimes used with wearresistant liners. Usually the wear on these bimetallic cylinders is almost three times that of the others. In the processing of abrasive materials, feed sections are sometimes finished in hard metal or other special materials and matched with the screws. If there is wear in the extruder, the greatest damage is always on the screw. Often only a new screw is used as a replacement, as it is assumed that the barrel is not damaged. However this assumption is usually a fallacy. If the screw is worn out, the barrel has been affected to some extent and may well need complete replacement. The rate of wear is increased considerably when the feed contains fillers such as titranium dioxide, glass fibers, and so on. There are many variables that cause damage to the barrel and the screw. If a problem is likely to occur frequently, protect the screw and consider using barrels with replaceable inner liners. The wear problem relates directly to the overall performance of the extruder to the extent that it is impossible to produce products that will remain within the tolerances required. (Note that what has been said about barrels and screws here also applies to other processes like 1M and blow molding that use screw plasticators.)

Energy Consumption Like the output capacity, the energy efficiency of an extruder is dependent on the torque available on the screw, screw RPM, heat control, and material being processed. Unfortunately, costly energy losses can occur, ranging from 3 to 20 percent, due to various factors, with the major loss occurring in the drive mechanism. Energy consumption is a major factor in production costs, as well as all equipment efficiency. Many extruders, as well as other equipment, are usually overpowered. This situation may be better than using underpowered equipment, but processes should not waste energy, resulting in higher product costs.

Gear Pumps Gear pumps, also called melt or metering pumps, have been standard equipment for decades in textile fiber production and postreactor polymer finishing. In the 1980s they established themselves in all kinds of extrusion lines. They consist of a pump, a drive for the pump, aJj.d pump controls located between the screen pack or screw and the die. Two counter-rotating gears will transport a melt from the pump inlet (extruder output) to the pump discharge outlet (die). Gear rotation creates a suction that draws the melt into a gap between one tooth and the next. This continuous action from tooth to tooth develops surface drag that resists flow, so some inlet pressure is required to fill the cavity. Melt pumps are most appropriate when the screw and die characteristics combine to give relatively poor pumping performance by the total system. This can happen when die pressures are low but more often occurs when they are extremely high, from 35 to 55 MPa (5,000-8,000 psi), or when the melt viscosity is extremely low. When pumps are used to increase the production rate by reducing the extruder head pressure without a corresponding increase in the screw speed, the extrudate solids' content is often increased. The result is an inferior product. This problem often necessitates additional

THE PROCESSING OF PLASTICS 629

filtration, which serves only to increase pressure and may counteract many of the benefits expected from the pump, as well as increasing the financial investment even further. Depending on the screw design, the extruder often creates pulses, causing the production rate to fluctuate. Some products usually cannot tolerate even minor fluctuations, and a pump often can assist in removing these minor product nonuniformities. In general, a pump can provide output uniformity of ±0.5 percent or better. Products include films down to 0.75 mil thickness, precision medical tubing, HIPS with 1,600 kg/h (3,500 lb.l hr.) output, fiber-optic sheathing, fibers, PET magnetic tape, PE cable jacketing (with the weight/ft. variation reduced from 14 to 2.7 percent), and so on. Pumps are very helpful to sheet extruders who also do in-house thermoforming, as they often run up to 50 percent regrind mixes. This normally variable-particle-size mix promotes surging and up to 2 percent gauge variation. Pumps practically eliminate the problem and make cross-web gauge adjustments much easier. Pumps are recommended in 1) most two-stage vented barrels where output has been a problem, such as ABS sheet; 2) extremely critical-tolerance extrusions, such as CATV cable, where slight cyclic variations can cause severe electrical problems; 3) coextrusion, where precise metering of layers is necessary and low pressure differentials in the pump provide fairly linear outputs; and 4) twin screw extruders, where pumps permit long wear life of bearings and other components, thus helping to reduce their high operating costs. Besides improving gauge uniformity, a pump can contribute to product quality by reducing the resin's heat history. This heat reduction can help blown film extruders, particularly those running high viscosity melts such as LLDPE and heat-sensitive melts such as PVC. Heat drops of at least 20 to 30°F will occur. In PS foam sheet extrusion, a cooling of 10 to 15°F occurs in the second extruder as well as a 60 percent reduction in gauge variation by the relieving of backpressure. All melts require a minimum heat and backpressure for effective processing. Although gear pumps can eliminate or significantly improve many processing problems, they cannot be considered a panacea. However, they are worth examining and could boost productivity and profits significantly. Their major gains tend to be principally in melt stability, temperature reduction in the melt, and increased throughput with tighter tolerances for dimensions and weights.

Dies The function of a die is to accept the available melt from an extruder and deliver it to takeoff equipment as a shaped profile (film, sheet, pipe, filament, etc.) with minimum deviation in cross-sectional dimensions and a uniform output by weight, at the fastest possible rate. A well-designed die should permit color and compatible resin changes quickly with little off-grade material. It will distribute the melt in the die flow channels so that it exists with a uniform density and velocity (see Fig. 7-33) [10]. The flow rate is influenced by all the variables that can exist in preparing the melt during extrusion-namely, die heat and pressure with time in the die. Unfortunately, in spite of all the sophisticated polymer flow analysis and the rather mechanical computeraided design capabilities, it is very difficult to design a die. An empirical approach must be used, as it is quite difficult to determine the optimum flow channel geometry from engineering calculations. It is important to employ rheological flow properties and other melt behavior (see Chapter 2) via the applicable CAD programs for the type of die required. The most important ingredient is experience, which, for the novice, is properly recorded in a computer program. Nevertheless, die design has remained more of an art

630 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

CHOKE BAR ADJUSTME DIE BLADE ADJUSTMENT

MANIFOLD CHAMBER , THAT CAN INC TEAR -DROP SHAPE .

ADJUSTABLE DIE BLADE

ELiVERY

DIE BODY

CAST-IN HEATERS

Figure 7-33. A simplified schematic of an extruder sheet die with a choke bar, adjustable die lip, and so on. Some dies, other than sheet dies, do not require these controls, but others may require more of them. There are many different dies designed to meet different processing requirements, such as spiral dies for blown film.

than any other aspect of process design. Design experience can work only if the operator of the processing line has developed the important ability to debug [10]. A well-built die with adjustments-temperature changes, restricter/choker bars, valves and other devices-may be used with a particular group of materials. Usually a die is designed for a specific resin meeting its particular rheological behaviors (see Chapter 2). To simplify the processing operation, the die design should consider certain factors, if possible. The goals are to have the extrudate (product) of a uniform wall thickness (otherwise the heat transfer problem is magnified); to minimize the use of hollow sections; to minimize narrow or small channels; and to use generous radii on all comers, such as a minimum of 0 .5 mm (0.02 in.). An "impossible" or difficult process can still be designed, but it requires extensive experience (both practical and theoretical), with trial-and-error runs, to make it practical. The design of the product should consider ease of processing (see Figs. 7-34 through 7-38 and Table 7-10). BASICS OF FLOW

The non-Newtonian behavior of a plastic (see Chapter 2) makes its flow through a die somewhat complicated. One characteristic of plastic is that when a melt is extruded from the die, there is some swelling (see Fig. 7-37). After exiting the die, it is usually stretched or drawn down to a size equal to or smaller than the die opening. The dimensions are then reduced proportionally so that in an ideal resin the drawn-down section is the same as the original section, but smaller proportionally in each dimension. Because of the meltelasticity effects of the material, it does not draw down in a simple proportional manner; thus, the drawdown process is a source of errors in the profile. Errors are significantly reduced in a circular extrudate, such as wire coating. These errors must be corrected by modifying the die and takeoff equipment (see Fig. 7-35). There are substantial influences on a material created by the flow orientation of the molecules, so there are different properties in the flow direction and perpendicular to the flow. These differences have a significant effect on the performance of the part (see Chapter 2).

THE PROCESSING OF PLASTICS 631

POOR

,/

I

V

Y

v=::/

~ BETTER

L

~

Igl~~

POOR

BETTER

POOR

BETTER

POOR

~ BETTER

., .11;.10 't1.'L

/

(r,"

...

lJ-': -- : BEST

POOR

POOR

BETTER

POOR

BETTER

BEST

IIEST

Figure 7-34. The influence of a part design on reducing extrusion process variables.

Another important characteristic of melts is that they are affected by the orifice shape (Fig. 7-36). The effect it produces is related to the melt condition and the die design (land length, etc.), but a slow cooling rate can have a significant influence, especially with thick parts. Cooling is more rapid at the comers; in fact, a hot center section could cause a part to "blow" outward or include visible or invisible vacuum bubbles. The popular coat-hanger die, used for flat sheet and similar products, illustrates an important principle in die design. The melt at the edges of the sheet must travel farther through the die than the melt that goes through the center of the sheet. Thus, a diagonal melt channel with a triangular dam in the center is used to restrict the direct flow to some degree. The principle of built-in restrictions is used to adjust the flow in many dies. With blow molding dies and profile dies, the openings require special attention to provide the proper product shape (see Figs. 7-37 and 7-38).

Orowdown from pull rolls

Orawdown from pull rolis

Die

Figure 7-35. The effect of land length on swell.

00 00 O,e Shape

Die Shope

L

~

Part Sh.pe

Part Shape

Product

. '~L:·:",-=---Ote~

@:>'

~~,/

Square 0.. Yields

o.slorted Sechon

Product

Figure 7-36. The effect of die orifice shape on the extrudate. 632

SLOW COOL

a

DIE EXIT

n

SLOW HEAT

tl

FAST COOL

~

{)

FAST HOT

t)

COLD STRETCH

t>

HOT STRETCH

Figure 7·37. Examples of how temperature, pressure, and takeoff speed (time) variations can potentially influence the shape of an extrudate.

Dimensions of die orifice ~--------1.378

Dimensions of final product ~-----------1.252in.---------~

Figure 7·38. Examples of changes in a PVC profile shape from the die orifice to the product where no dimensions remained the same. 633

0-

""...

WaU thickness Angles Profile dimensions, ±mm(in.) 0-3 (0-'/,) 3-13 ('/,-'1,) 13-25 ('/z-I) 25-38 (1-1'/,) 38-50 (1'/z-2) 50-75 (2-3) 75-100 (3-4) 100-125 (4-5) 125-180 (5-7) 180-250 (7-10)

Dimension ± 8% ±3°

0.25 mm (0.010 in.) 0.50 mm (0.020 in.) 0.63 mm (0.025 in.) 0.68 mm (0.027 in.) 0.90 mm (0.035 in.) 0.94 mm (0.037 in.) 1.3 mm (0.050 in.) 1.7 mm (0.065 in.) 2.4 mm (0.093 in.) 3.0 mm (0.125 in.)

0.18 mm (0.007 in.) 0.30 mm (0.012 in.) 0.43 mm (0.017 in.) 0.63 mm (0.025 in.) 0.75 mm (0.030 in.) 0.90 mm (0.035 in.) 1.3 mm (0.050 in.) 1.7 mm (0.065 in.) 2.4 mm (0.093 in.) 3.0 mm (0.125 in.)

0.18 mm (0.007 in.) 0.25 mm (0.010 in.) 0.38 mm (0.015 in.) 0.50 mm (0.020 in.) 0.63 mm (0.025 in.) 0.75 mm (0.030 in.) 1.1 mm (0.045 in.) 1.5 mm (0.060 in.) 1.9 mm (0.075 in.) 2.4 mm (0.093 in.)

ABS

± 8% ±2°

Polystyrene

± 8% ±2°

Rigid Vinyl (PVC)

0.25 mm (0.010 in.) 0.38 mm (0.015 in.) 0.50 mm (0.020 in.) 0.68 mm (0.027 in.) 0.90 mm (0.035 in.) 0.94 mm (0.037 in.) 1.3 mm (0.050 in.) 1.7 mm (0.065 in.) 2.4 mm (0.093 in.) 3.0 mm (0.125 in.)

± 8% ±3°

Polypropylene

0.25 mm (0.010 in.) 0.38 mm (0.015 in.) 0.50 mm (0.020 in.) 0.75 mm (0.030 in.) 0.90 mm (0.035 in.) 1.0 mm (0.040 in.) 1.7 mm (0.065 in.) 2.4 mm (0.093 in.) 3.0 mm (0.125 in.) 3.8 mm (0.150 in.)

±IO% ±5°

Flexible Vinyl (PVC)

Table 7-10. Some Dimensional Tolerances for Plastic Profile Extrusions

0.30 mm (0.012 in.) 0.63 mm (0.025 in.) 0.75 mm (0.030 in.) 0.90 mm (0.035 in.) 1.0 mm (0.040 in.) 1.1 mm (0.045 in.) 1.7 mm (0.065 in.) 2.4 mm (0.093 in.) 3.0 mm (0.125 in.) 3.8 mm (0.150 in.)

±IO% ±5°

Pol yethy lene

THE PROCESSING OF PLASTICS 635

Special Dies Some special dies, shown in Figures 7-39 and 7-40, produce interesting flow patterns and products, such as tubular to flat netting dies. For circular output a counterrotating mandrel and orifice have semicircular-shaped slits through which the melt flow emerges. If one part is held stationary, a rhomboid or elongated pattern is formed; if both parts rotate, a true rhombic mesh is formed. When the slits overlap, a crossing point is formed where the emerging threads are "welded." For flat netting, the slide is in opposite directions.

Coextrusion Coextrusion provides multiple molten layers-usually using one or more extruders with melts going through one die-that are bonded together. This technique permits using melt heat to bond the various plastics (see Tables 7-11 and 7-12), or using the center layer as an adhesive. Coextrusion is an economical competitor to conventional laminating processes by virtue of its reduced materials-handling costs, raw materials costs, and machine-time cost. Pinholing is also reduced with coextrusion, even when it uses one extruder and divides the melt into at least a two-layer structure. Other gains include elimination or reduction of delamination and air entrapment. In the past, a processor desiring to enter the field had little choice of equipment, but the increased interest in coextrusion has produced a proliferation of equipment. With rapidly changing market conditions and the endless introduction of useful materials, the design of machines has become much more involved. It is important that the processor have flexibility in making selections, but not at the expense of performance, dependability, or ease of operation. One should provide for the material or layer thickness necessary in product changeover without allowing high scrap rates. The goal should be to incorporate scrap regrind in the layered construction. It is important to be able to control the individual layer distribution across the width of the die. It is normal, as the viscosity ratio or thickness ratio of the polymers being combined increases, for the individual layer distribution(s) of the composite film to become displaced. Viscosity differences influence a reduction and saving of materials. A number of techniques are available for coextrusion, some of them patented and available only under license. Basically, three types exist: feedblock, multiple manifold, and a combination of these two (see Table 7-12).

Table 7-11. Examples of Compatibility Between Plastics for Coextrusion

LOPE HOPE PP Ionomer Nylon EVA

LOPE

HOPE

PP

Ionomer

3 3 2 3

3 3 2 3

2 2

1

1

2 1

3

3

3

3 3 2 3 3 3

Code: I. Layers easy to separate. 2. Layers can be separated with moderate effort. 3. Layers difficult to separate.

3

Nylon

EVA

I I

3 3 3 3

3 3

1

3

~

SPROC E OAIVE

~

~./

U ~

DRIVE CHAIN

r~

----------

Round pIaStlc neltN'IQ A

::::::;:=========

1- -.- - - - -

:t

C

-------~--''--

0- - - - - - E -----------Flal nettmg •. . •.•• W1Ih d"f rem po5lllDM 01 doe lops

o

DIE OPE ING

-:"'+-=;=*=~:'---~H--: WIRE FEED

S OT

WIRE

A-

sfoPoF~s

INTERMITTENT

~ ~

I · · · i~~.~

~

thickness using ." Old IN'IQ In etOa lINd lUbe die

v~ lUbe ~

00IIer..... pet1cnted tubing pattern: USIn!I bnIJ mandRIl die

Figure 7-39. Examples of special action dies that produce round and flat products. 636

Figure 7-40. Example of netting as it exits the die. This mesh, produced by extrusion from a counter rotating die design, originally patented in 1956, followed by a postextrusion stretching process, is available in almost every conceivable form.

Table 7-12. Comparison of Feedblock and Multimanifold Coextrusion Dies Characteristic Basic difference

Cost Operation Number of layers Complexity Control flow

Layer uniformity Thin skins Viscosity range Degradable core material Heat sensitivity Bonding

Multimanifold

Feedblock Melt streams brought together outside die body (between extruder and die) and flow through the die as a composite Lower Simplest Not restricted; seven- and eight-layer systems are commercial Simpler construction; no adjustments basically Contains adjustable matching inserts, no restrictor bar

Individual layer thickness correction of ± 10 percent Better on dies >40 in. Usually limited to 2/1 or 3/1 viscosity range of materials Usually better More Potentially better; layers are in contact longer in die

Each melt stream has a separate manifold; each polymer spreads independently of others; they meet at die pre-land to die exit Higher Generally restricted to three or four layers More complex Has restrictor bar or flow dividers in each polymer channel; but with blown film dies control is by individual extruder speed or gearboxes Restrictors and manifold can meet ±5 percent Better on dies <40 in. Range usually much greater than 3/1

Less

637

638 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Orientation Orientation consists of a controlled system of stretching plastic molecules to improve their strength, stiffness, optical, electrical, and other properties. This process, which has been used for almost a century, became prominent during the 1930s for stretching fibers up to ten times. Later it was adapted to stretching film and sheet and, more recently, blow-molded containers. Many other products take advantage of its benefits for producing tape, pipe, profile, and thermoformed parts, etc. (see Table 7-13). Practically all plastics can undergo orientation, although certain types find it particularly advantageous (PET, PP, PVC, PE, PS, PVDC, PVA, and PC). Of the 15 million tons of plastic film sales annually worldwide, about 16 percent are sales of oriented material (see Chapter 2). In extrusion the most important orienting processes are used for blown film, flat film and sheet, and blow molding. During blown-film processing the blow-up ratio determines the degree of circumferential orientation and the pull rate of the bubble determines longitudinal orientation (see Figs. 7-41 through 7-43) [10]. The optimum stretching heat for amorphous resins (PVC, etc.) is just above the glass transition temperature; for the crystalline types (PET, PE, and so on) it is just below the melting point. During the stretching process the structure changes because of crystallization, thus usually necessitating an increase in heat if further deformation is planned. Afterward, the orientation is "frozen in" by lowering the heat or, with crystalline types, set by increasing the crystalline portion. With orientation, the thickness is reduced and the surface enlarged. If film is longitudinally stretched in the elastic state, its thickness and width are reduced in the same ratio. If lateral contraction is prevented, stretching reduces the thickness only. In orienting film or sheet the processor uses a tentering frame (typically used for many decades in textile weaving), which is enclosed in a heat-controlled oven, with a very

Nip Rolls

ROll 01 FIlm

I

Extruder

Idler Roll

Figure 7-41. A schematic of a basic (vertical-up) blown film line. The hot-melt bubble exits the die with an air tube in its center that provides internal air pressure to the bubble that is pulled by the nip rolls via the bubble collapsing frame.

THE PROCESSING OF PLASTICS 639

Table 7-13. Example of Orientation Used to Fabricate Different Types of TP Film Tapes Ranges of Application Carpet basic weave

Tarpaulins Sacks

Ropes Twine Separating weave Filter weave Reinforcing weave Tapestry and home textiles Outdoor carpets

Decorative tapes Knitted tapes, sacks, and other packagings, seed and harvest protective nets Packaging tapes

Fleeces

Demands Made

Rate of Stretching

Thennoplastic

Low shrinkage High strength Temperature stability Specific splicing tendency Matt surface High strength

1:7 1:5

PP PETP

1:7

High strength High friction value Specific elongation Weather resistant High tensile strength Specific elongation Good tendency to splicing High tensile strength High knotting strength High strength Low shrinkage Abrasion resistant Low shrinkage Specific elongation Temperature resistance UV-resistance Low static charge Unifonn coloration Textile-type handle Low shrinkage Wear resistance Weather resistance Elastic recovery Unifonn coloration Defined splicing Effective surface Low specific gravity High knotting strength Low splicing tendency Suppleness UV-resistance High strength Low splicing tendency Fiber properties

1:7

PP PE PP PE

1:9 to 1:11 (15)

PP

1:9 to 1:11 1:7 1:7 1:5 1:7 1:5

PP PPIPE PP PP PETP PP PETP

1:7

PE

1:7

PP

1:5

PETP

1:6

PP with blowing agent PP PE

1:6.5

1:9 1:7 1:7

PP PETP PP and blends

accurate and gentle air flow used to hold the oven at the required orienting heat (Fig. 744). The frame has continuous speed control and diverging tracks with holding clamps. As the clamps move apart at prescribed diverging angles the hot plastic is stretched in the transverse direction, resulting in single orientation (0). To obtain bidirectional orientation (BO) an inline series of heat-controlled rolls are located between the extruder and tenter frame. The rotation of each succeeding roll is increased, based on the longitudinal stretched properties desired.

640 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Processing Lines Each line has interrelating operations, as well as specific line operations, to simplify processability. Usually the extruder is followed by some kind of cooling system to remove heat at a controlled rate, to cause plastic solidification. It can be as simple a system as air or water cooling, or a cooled roll contact can be used to accelerate the cooling process. Some type of takeoff at the end of the line usually requires an accurate speed control to ensure product precision and save on material costs by tightening thickness tolerances. The simplest device might be a pair of pinch rolls or a pair of opposed belts (a caterpillar takeoff) . A variable-speed drive is usually desired to give the required precision. Some examples of different extrusion lines are shown in Figures 7-41 and 7-45 through 7-51 [10].

1

M .D.- Machine Direction

T.O. ......-

M .0 .

T .D.- Transverse Direction

Layflat Width

_Main Nip Rolls

~----------------~-

Layflat Die Dia.

= Blow Ratio (BR)

Bubble of Film

Figure 7-42. The tenninology and layout for blown film.

THE PROCESSING OF PLASTICS 641

Typical Converter Film 4 :1 DDR/SUR @ 1.5 MIL

Balanced Shrink FIlm 1:1 DDR/BUR @ 1.5 MIL ILLUSTRATIO N

ILLUSTRATION

THEORY

THEORY

4 :1 BUR 1.5 M IL .....-4-:-:-:-1-::D:-:::D:-:::R~

@ 60 FPM

Line Speed 4 :1 BUR 6 MIL ......---::-::-::-@ 15 FPM 1:1 DDR une Speed

---::-2...." :1-:S,..."U,...,,R,......._1 . 5 MIL S:1 ODR @ 120 FPM Line Speed

32" Dia. Bubble 50" Layfla t Frost Line

~

Film

2 :1 BUR 1:1DDR

~..

I

Melt

~..

12 MIL

@ 15 FPM Line Speed

16 " Dla Bubble 25" Layflat BUR= Blow Up Ratio DDR= Draw Down Ratio FPM= Feet Per Minute

S" Dia . DIE .024" DIE GAP '" DIE LAND 15 FPM Melt Velocity

Figure 7-43. Examples of unoriented and oriented blown film.

IN-LINE POSTFORMING

Information on the technique of forming is covered later in this chapter under "Forming." In-line postforming, or postextrusion processing, refers to the special processing that may be done to the extrudate, usually just after it emerges from the die but before the material has a chance to cool. When the material is worked in such a state it is known as in-line processing, as opposed to cutting, forming, or other processing, which might be done on the cold extrusion. In-line processing is usually done close to the extruder and is done automatically, with little or no extra labor on the part of the machine operator. The extra processing, which may involve shaping, cutting, re-forming, or a surface modification of the extrudate, can considerably increase the value of the extrusion without materially increasing its cost, but it may also be done to enable the use of a lower-cost die, as for example flattening a tubular extrusion into an oval so that a much cheaper circular die can be used.

8

OVEN

(a)

Nonol lenled Oriented (machine direction) Oriented (transverse direction) DRAWDOWN

DIE SWELL L - - - - t

400"/.

PULLER

Tenllte I trength 5,000 psi 4,000 psi 25,000 psi

(b)

Figure 7-44. Use of a tender frame to biorient film or sheet. In a), the feeder-roll speed to puller-roll speed ratio can be set, such as 1 : 4, and simultaneously the ratio of width can be set, as 1 : 4. However, the machined direction ratio is usually accomplished prior to the plastic's entering the temperature-controlled oven that contains the tenter frame, by having it move around heat-controlled rolls where the rotational speed of the rolls increases from one roll to the next. Part b) is a schematic of the drawdown phenomenon with swell to produce orientation in the machined (longitudinal) direction.

EXTRUDER

TRIMMER (SLITTER)

\

,, / POWERED ------ ' ) CARRIER ROLLS

~

,

"

\-.

RUBBER (R C) NIP 0 P IN H , ROLL ./

\

\

---2 (o}mor~)

STAINLESS-STEEL NIP (OR PINCH) ROLL (driv~n )- - -_.

WATER - COOLED HIGHLY POLISHED CHILL ROLLS

Figure 7-45. Schematic of a chill-roll system for a flat-film extrusion line.

642

,,

\

, •

\

f

~

t

""'.$'..

"0'/0

""0; ?oS'/
"""

"I

.P0o.:

""

"0

(

~o<:­

.;:,c"

<;

("

0' '"

'"01-, 0""Q.

(

~~o, ..r.~

~

("

(

"0

/'Q.

'". . .

'"0

."?o..

'

,,(

,<:-0,

~

,<:-"

" ~.. "" "0' 0,,-_/0"

~-t-o,

"T

.$' """0".. .P,..{o

' ..

0'>.9.

,,~

q<1'

~'"

r:-'"

~Ci

.P~

.Po

"'. .~" .

(a)

"

"'0"

0,,-

,....1-,

{o

.P,..

0,

0"",

(,"'''

oS'

".P{. ~

Q.

(

..r."~

qO

r:-o, ,...~

~'"

(:0'"

~Ci

.,,p<:-

"0-;'0

(

.. '"

"'>

(

(

p,

r:f

(

<:: (

« (

granulate

...& ~o ,~" ~",<:-o ~" ...o~ ,,,,, ~Ci

-t-o, p, ~." ",-t' 0- e p, -t-o,A- p, ,,'" p, ,...&~:F',s ~-t' ",,\ ~
.q.""'0/1 "b?)

..

-I-,

"" ,.... .P"b

I",

.P"'o "'>

.

"'o"F"'{o°' +:9:>-",......~ "'>6:°/(,/1""..'}o.... "".. .P" '" 'o'~ O/l'~ "s 00.: "s ".. -..?

$(, 0/"",

"Q.

0',..'" "".. .$'''''' $"" ~.. oS''''' "q"?o0;.. .Q.$'1 ..,.. ..,.. ,...., ..,.. "(,(, ""6

0,,- ""

"

~ ~

.Ii"....;:,.s "I

"('(,oo.,:9{.

~ .Po

'""?o "'"0",

(

~rf'

",.;:,'"

(" (

~~

,,-0

'('

(:0'" .....~Ci

",,0<:-

~'"

Figure 7-46. Top: An example of the downstream equipment used with the extruder. Bottom: The processing steps in the chill-roll process used in an oriented film-tape line.

lopt'S _ "

processing

Slep~

-

~

t

ri.; -;

I \'Fg

d

To draIn

Pump

a

Au xlhary Cooling Tank

b

b

c

Windup untl

d

e

~

Pressure calibration of rigid pipe using plug insert with water spray cooling (a. pipe die, b.pressure calibration , c.water spray cooling , d. drag lugs on conveyor belt and e. caterpillar take· all puller)

Figure 7-47. Examples of calibration systems for pipe and tube extrusion lines.

Extruder

Bleed

DIfferentIal pressure SIzIng unll

Differential pressure sizing for flexible tubing

~lmlTI ~ ~ i -IIT - -

UU]II I til

a b c

Vacuum lank calibration of rigid pipe used with water baths (a. pipe die, b.vacuum with discs, c.heated zone water baths and d. caterpillar take·off puller)

Hopper Dryer

Shear

Figure 7·48. A sheet line using a three-cooling-roll stand.

Haul-Off Unn Cooling Tank

Figure 7-49. The important downstream equipment used in pipe and profile extrusion.

SAW, CUT, AND SHEAR

PUNCH AND DRill

EMBOSS

~w

01

~"C'BCI

L::IJ~

HOT STAMPI NG

. 1.

PRINTING AND DECORATING

APPLY TAPE

HEAT SEAL AND ULTRASONIC WELDING

FORMING

0

WR~Jrt::j S~L

WELD

Figure 7-50. The many off-line operations that are performed on extruded plastics. 645

646 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~

~ c""'''' '""

CROSSHEAD COATER

FEED ROLLS

FINISHED-PARTS COLLECTOR

Figure 7-51. A schematic of noncontinuous profiles that are being automatically coated with plastics.

BLOW MOLDING Blow molding (BM), the third most popular method of plastics processing, offers the advantage of manufacturing molded parts economically, in unlimited quantities, with little or virtually no finishing required. It is principally a mass-production method (see Figs. 7-52 and 7-53). The surfaces ofthe moldings are as smooth and bright, or as grained and engraved, as the surfaces of the mold cavity in which they were processed. Among the special techniques available are stretch blow molding and coextrusion. One can improve the cooling efficiency and reduce cycle time with gases (C02 , etc.). Other developments include shuttle postcooling, insertion of printed film in the mold to avoid the need for subsequent decorating, and so on. Blow-molded parts demonstrate that, from the technical and cost standpoints, BM offers a promising alternative to other processes, particularly injection molding and thermoforming. The technical evolution of BM, plus accompanying improvements and new developments in plastics, has led to new BM parts (see Fig. 7-54). With the coextrusion technology now established and the hardware in place, the variety of achievable properties can readily be extended by the correct combination of different materials (see Fig. 7-55). The potential for BM products includes much more than the simple bottles that have been made for many decades. Now the expertise and economics of the method are such that many ideas once deemed futuristic are much closer to realization. Blow molding offers a number of processing advantages, such as molding extremely irregular (reentrant) curves, low stresses, the possibility of variable wall thicknesses, the use of polymers with high chemical resistance, and favorable processing costs. Reentrant curves are the most prominent features-so much so that it is difficult to find examples that do not incorporate them (see Fig. 7-56) [11]. They combine aesthetics with strength and cost benefits [1, 10-12, 62-65, 626-34]. A significant difference exists between BM and 1M. BM usually requires only .17 to 1.03 MPa (25 to 150 psi) pressures, with possibilities for certain resins of up to 1.38 to 2.07 MPa (200 to 300 psi). For 1M, the pressure is usually 13.8 to 137.8 MPa (2,000 to 20,000 psi), and in some cases up to 207 MPa (30,000 psi). The lower pressures generally result in lower internal stresses in the solidified plastics and a more proportional stress distribution. The result is improved resistance to all types of strain (tensile, impact, bending, environmental, etc.). As the final mold equipment for BM consists of female molds only, it is possible simply by changing machine parts or melt conditions to vary the wall thickness and weight

THE PROCESSING OF PLASTICS 647

of the finished part. If the exact thickness required in the finished product cannot be accurately calculated in advance, this flexibility is a great advantage from the standpoint of both time and cost. With BM it is possible to produce walls that are almost paper thin. Such thicknesses cannot be achieved by conventional 1M but, with certain limitations, can be produced by thermoforming. Both BM and 1M can be successfully used for very thick walls. The final choice of process for a specific wall section is strongly influenced by such factors as tolerances, reentrant curved shapes, and costs. Blow molding can be used with plastics such as PE that have a much higher molecular weight (MW) than is permissible in 1M (see Chapter 2 on MW). For this reason items can be blown that utilize the higher permeability, oxidation resistance, UV resistance, and so on of the high-MW plastics. This feature is important in providing resistance to environmental stress cracking (see Chapter 4). This extra resistance is necessary for plastic bottles used in contact with the many industrial chemicals that promote stress cracking. With BM, the tight tolerances achievable with 1M are not obtainable. However, in order to produce reentrant-curved or irregularly shaped 1M products, different parts can be molded and in tum assembled (snap-fit, solvent-bonded, ultrasonically bonded, etc.). In the BM of a completely irregular/complex product, even though the 1M tolerances

Figure 7-52. This diesel truck surge fuel tank (left), next to a stainless-steel tank, was extrusion blow molded of glass-fiber-reinforced Du Pont Zytel nylon. The stainless-steel tank cost more and required welding and assembly. The weight went from 10 to 4lbs. with the RP nylon. A new corrugated wall design also minimizes deflection, resists harsh under-the-hood temperatures and solvents, and has other advantages.

Figure 7-53. This Navistar International truck fascia is of multiple-extrusion blow-molded parts made from Himont's HiFax polypropylene. The fascia parts, which range from 8 to 17 lbs. and come in lengths from 24 in. to 6 ft., are molded on 50-lb. accumulator head capacity machines. The ability to mold the PP in an integral color eliminates the need to paint.

Figure 7-54. This aquacycle, with its PP blow-molded wheels that incorporate paddles or fins on their sides, operates most efficiently. 648

Body loyer BondIng agrnl / / Borner layer / / Bonding ogenl . V Body loyrr mel regnnd

Figure 7-55. Coextrusion blow molding can provide multiple layers, from two to at least six, flash-free with easy, high-speed production.

Figure 7·56. A section of an HDPE blow-molded, double-waIled, integral handle for a container lid. 649

650 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 7-14. Manufacturing Cost Comparison of 16-oz. Blow Molded Bottles Standard Extrusion Blow-molding 2-Parison Head 4-Fold

Stretch Blow Molding PVC (2) Single-parison Heads 4-Fold

$270,000

$450,000

$850,000

$ 9.00

$ 14.85

$ 28.33

2.80 13.00 2.50 1.50 2.25

4.65 13.00 5.35 2.00 3.75

10.20 13.00 11.00 4.00 4.50

Total hourly mc costs 3.O--Bottle specs. hourly/annual prod. 3.1-16 oz. finish wt. (454 g) -regular 37 g (1.3 oz) -stretch PVC 20 g (0.7 oz) -stretch PET 20 g (0.7 oz) Cycle time/bottles per hour bottles per yr., millions 4.O--Annual costs 4.1-16 oz. (454 g) Resin: -37 g $.70/lb. ($70/0.45 kg or $1.54 kg) -20g $.661lb ($1.46 kg) -20g $.60/lb ($1.32 kg) Machine costs

$ 31.05/hr.

$ 43.60/hr.

$ 71.03/hr.

8.4 sec.!I,714 10,286

7.5 sec.! 1,920 11,520

186,300

261,600

426,180

total p.a. Royalty (PET) Du Pont-per year Cost per thousand

$771,500

$596,550

$1,060,540 30,000

$ 75.00

$ 51.78

$

1.000Machine cost incl. head, molds, ancillaries (lie. fee, stretch PVC and PET) 2.O--Hourly machine costs Depre'n, 5 yr. 30 K hr, $/hr. Financing cost, 5 yr. 12.5% Labor, I man Energy at $.06 per kWh Floor space Maintenance and consumable mtl.

Stretch Blow Molding PET

4,000 24,000

$585,200 $334,950 $634,360

45.44

Notes: 1. Figures are not to be considered as absolute costs, but rather reflect comparisons between various machine options.

2. All calculations are based upon 100 percent efficiency. 3. All bottle weights are finish weights (flash being considered 100 percent reusable).

cannot be equaled, the cost of the container is usually less (see Figs. 7-52-56). No secondary operations such as assembly are required. Other advantages are also achieved, such as significantly reducing (if not eliminating) leaks, reducing total production time, and so forth. Blow molding can be divided into three major processing categories: 1) extrusion BM (EBM), which principally uses an unsupported parison; 2) injection BM (IBM), which usually uses a preform supported by a metal core pin; and 3) stretch BM, for either EBM or IBM, to obtain bioriented products, providing significantly improved cost-to-perfor-

THE PROCESSING OF PLASTICS 651

mance advantages. Almost 75 percent of the BM processes are EBM, almost 25 percent are IBM, and the other 1 percent or so use other techniques such as dip BM [12]. About 75 percent of all IBM products are bioriented. These BM processes offer different advantages in producing different types of products, based on the materials to be used, the performance requirements, the production quantity, and costs. Blow molding requires an understanding of every element of the process, starting with the basic "extruder" used in conventional extrusion and 1M machines. (For information on the machines used to plasticate/melt materials for BM, see the previous sections on injection molding and extrusion.) With EBM, the advantages include high rates of production, low tooling costs, incorporation of blown handleware, a wide selection of machine builders, and so on. The disadvantages are a usually higher scrap rate, the use of recycled scrap, and limited wall thickness control or resin distribution. Trimming can be accomplished in the mold for certain types of molds, or secondary trimming operations may have to be included in the production lines, and so forth. With IBM the major advantages are that no flash or scrap occurs during processing, it gives the best of all thickness and material distribution control, critical neck finishes can be molded to a high accuracy, it provides the best surface finish, low-volume quantities are economically feasible, and so on. The disadvantages are its high tooling costs, the lack (to date) of blown handleware (there is only solid handleware), its being more or less limited to relatively smaller blown parts (whereas EBM can easily blow extremely large parts), and so forth. Similar comparisons exist with biaxially orienting EBM or IBM. With respect to coextrusion, the two methods also have similar advantages and disadvantages, but mainly major advantages. With IBM, for example, PET can be processed (mono- or multilayer) and stretched into the popular two- and three-liter carbonated beverage bottles. Table 7-14 provides a cost comparison of the different BM techniques for PVC and PET, the plastics predominantly processed in BM. EXTRUSION BLOW MOLDING

In EBM, a parison (tubular type of hot melt) is formed by the extruder melt output (see Fig. 7-57). Turning continuously, the screw feeds the melt through the die head, generally as an endless parison directly through a die. A die head can have one or more openings, so one or more pari sons can be extruded (see Fig. 7-59). The size of the part and the amount of material needed to produce a part (shot size) dictate whether or not an accumulator is required. The basic nonaccumulator machine offers a continuous flow of plastic melt. With an accumulator the flow of the parison through the die is cyclic (see Fig. 758). The connecting channels between the extruder and the accumulator, as well as the accumulator itself, are designed to prevent restrictions that might impede the flow or cause the melt to hang up. Flow paths should have low resistance to melt flow, to avoid placing an unnecessary load on the extruder. To ensure that the least heat history (residence time) is developed during processing, the design of the accumulator should provide that the first melt in be the first to leave when the ram empties the chamber; the aim should be to have the chamber totally emptied on each stroke. When the parison exits the die and reaches a preset length, a split cavity mold closes around it and pinches one end of it. Usually a blow pin is located opposite the pinched end of the "tube." Compressed air inflates the parison against the pinched end of the

652 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

A

1=

f~ll~f PRESS PLATEN

\

t_

1=:11 _t

II

I I lsi I I

I I I I

I I

[IJ~i ~

D

Par/son being extruded

u Compressed air inflates par/son

DQO 0

Blown container being ejected

Figure 7-57. The basic extrusion blow-molding process. A = a parison (plastic tubular melt leaving the extruder die) cutter; B = parison; C = the two blow-mold cavities, D = the blow pin.

D,e position Machine interface inputsl outputs

Parlson control

Profile control signal

Parison

Figure 7-58. A schematic of an accumulator head with a programmable process controller to control such melt characteristics as the rate of melt flow to form a parison, the profiling thickness of the parison as it extrudes from die, etc. The controller interrelates with the extruder and molded product performance.

"tube." Compressed air inflates the parison against the female cavity of the mold surfaces. Upon contact with the relatively cool mold surface, the blown parison cools and solidifies to the part shape. Next the mold opens, ejects the part, then repeats the cycle by again closing around the parison, shaping it, and so on. Various techniques are used to introduce air. It can enter through the extrusion die mandrel, through a blow pin over which the end of a parison has dropped (see Fig. 757), or through blowing needles that pierce the parison. The wall distribution and thickness of the blown part are usually controlled by parison programming, the blow ratio, and part configuration.

THE PROCESSING OF PLASTICS 653

Figure 7 -59. A multiple die head (three parisons) blow molding three containers simultaneously.

Melt Viscosity Melt properties are of critical importance to BM, more so than for conventional extrusion. To a large extent they determine the quality achieved. The melt viscosity decides whether sagging or lengthening of a parison during extrusion can be compensated, particularly in noncircular pari sons (see Fig. 7-60). Because engineering resins have so far been used mainly with 1M, most processors attempt to use easy-flowing, low-molecular-weight 1Mgrade resins. But in BM, particularly EBM, the objective is very different; the melt should be viscous and of high molecular weight (high melt strength) (see Chapter 2). This requirement generally also ensures another important feature-better impact strength. The melt viscosity should be nearly independent of the shear rate and the processing heat.

Parison Thickness Control Electronic parison programming is an effective way to control material usage and improve both quality and productivity. The most common method used is orifice modulation (see

654 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 0.850

1.750

0.950

1.800

1.600

1.700

0.800

0.900

WITHOUT DIE SHAPING

1.200

1.~

1.500-

1.150

1-1.600

1.500

1.200

WITH DIE SHAPING

Figure 7-60. A noncircular blow-mold die, shown with and without a wall-thickness die shape (the dimensions are in mm).

Fig. 7-59). The die is fitted with a hydrualic positioner that allows positioning of the inside die diameter during the parison drop. The OD to ID relationship of the tapered die orifice opening is varied in a programmed manner to increase or decrease the paris on wall thickness. In regard to parison control, a compromise is necessary between the desired net weight and the need to maintain a sufficient safety margin over a set of minimum specifications, which include minimum wall thickness, drop speed, drop strength, dimensional stability, and fluctuations in net weight. Most of these parameters can be affected directly by the molder's ability to control the parison wall thickness. The most common and practical way of doing this has been to adjust the gap between the die and the mandrel (see Table 7-15).

Pinch-off The pinch-off is a critical part of the EBM mold, where the parison is squeezed and welded together, requiring good thermal conductivity for rapid cooling and good toughness to ensure long production runs. The pinch-off must have structural soundness to withstand the resin pressure and repeated closing cycles of the mold. It usually must push a small amount of resin into the interior of the part to slightly thicken the weld area. It also provides a cut through the parison to make a clean break later when flash is removed.

THE PROCESSING OF PLASTICS 655

Table 7-15. Examples of Differently Performing Extrusion Blow Molding Dies Type Die

Feature

Simple die

Fixed die gap

Die profiling

Pennanentiy profiled; preferred in die land area

Die centering

Open-loop axial die gap control

Can be pennanentiy shifted laterally to correct parison drop path Can be axially shifted during extrusion

Servohydraulic closed-loop axial die gap control

As above, with greater speed, accuracy, and flexibility

Stroke-dependent die profiling

Pennanentiy ovalized die gap

Die/mandrel adjustable profiling

Settable adjustment of die gap profile

Servohydraulic closed-loop radial die gap control

Programmable ovalization and shifting of die gap

Advantage/Disadvantage Simple; inexpensive; no adjustment facility Fixed circumferential wall thickness change; timeconsuming; complex Compromise between required drop path and equal wall thickness Equal circumferential wall thickness change possible; no feedback Equal circumferential wall thickness change possible, with feedback Fixed, unequal circumferential wall thickness change possible; affects entire parison length Settable, unequal circumferential wall thickness change possible; rapid optimization Programmable circumferential wall thickness change possible, independent of parison length

Most molds use a double-angle pinch-off with 45° angles and a 0.25 mm (0.010 in.) land; see view (1) in Figure 7-61. When the blow part is large relative to the parison diameter, the plastic will thin down and even leave holes on the weld line, requiring pinch-off (2). Using shallow angles (15°) has a tendency to force the plastic to the inside of the blown part, thereby increasing the thickness at the weld line. A pinch-off with dam (3) also helps to solve problems. The flash pocket's depth is related to the pinchoff and is very important for proper molding and automatic trimming. A gross miscalculation of pocket depth (which must be learned through experience) can cause severe problems. For example, if the pocket depth is too shallow the flash will be squeezed with too much pressure, putting undue strain on the mold, mold pinch-off areas, and machine-clamp press section. The molds will be held open, leaving a relatively thick pinch-off that will be difficult to trim properly. If the pocket is too deep, the flash will not contact the mold surface for proper cooling. In fact, between molding and automatic trimming, heat from the uncooled flash will migrate into the cool pinch-off and cause it to heat up and create unwanted problems like sticking to the trimmer. Or, during trimming, it can stretch instead of breaking free.

Clamping The improvements made in clamping units provide a great variety of movement and action in the larger BM machines. Small machines still need certain improvements to ensure good flash removal with low deformation of the clamping units. The most important

656 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7·61. A few typical pinch-off double-angle designs.

development has been the use of proportional valves in hydraulic systems. With this technology a machine runs more smoothly and more exactly, to permit a wide variety of action in the mold, as well as accurate control of the closing speed (see Fig. 7-62). The delayed closure action in the final phase of mold closing determines pinch-off weld formation, and the reproducibility of this delayed closure phase ensures the uniformity of BM parts.

Shrinkage The shrinkage behavior of different resins and the part geometry must be considered. Generally, shrinkage is the difference between the dimension of the mold at room temperature of 22°C (72°F) and the dimensions of the cold blown part, usually checked twenty-four hours after production. The elapsed time is necessary to allow the part to shrink. Trial and error determines what time period is required to ensure complete shrinkage. The coefficients of expansion and different shrinkage behaviors depend on whether the plastic materials are crystalline or amorphous (see Chapter 2). Lengthwise shrinkage tends to be slightly greater than transverse shrinkage. Most horizontal shrinkage occurs in the wall thickness rather than a body dimension. With PE, higher shrinkage occurs with the higher-density polymers and thicker walls. Lengthwise shrinkage is due to the greater crystallinity of the more linear types of plastics. Transverse shrinkage is due to slower cooling rates, which result in more orderly crystalline growth.

THE PROCESSING OF PLASTICS 657

UNILOY-Side Shift

Blow pin centered on mold

..

UNILOY-Side Shift

Blow pin centered on container neck

Figure 7-62. Mold movement can locate the blow pin in any horizontal position required.

Injection Blow Molding Injection blow molding basically has three stages (see Fig. 7-63). In the first stage, hot melt is injected through an injection molding machine nozzle into a manifold and into one or more preformed cavities. An exact amount of resin is injected around core pins. Hot liquid from a heat-control unit is directed by hoses through mold-heating channels around the preform cavity; these channels have been predesigned to provide the correct heat control on the melt within the mold cavity. The melt heat is decreased to the required amount. The two-part mold opens, and the core pins carry the hot plastic to the second-stage blow-mold station. Upon the mold's closing, air is introduced via the core pin, and the plastic blows out and contacts the mold cavity surface. Controlled chilled water circulates through predesigned mold channels around the mold cavity (usually 4-21°C, 40-50°F), and solidifies the hot plastic. The two-part blow mold then opens and the core pin carries the complete blown container to the third stage, which ejects the part. Ejection can be done by using a stripper plate, air, a combination of stripper plate and air, robots, and so forth. IBM can have four or more stations or stages. A station can be located between the preform and blow-mold stages to provide extra heat conditioning time. A station between the blow mold and ejection stages can provide additional cooling and such secondary operations as hot stamping, labeling, and so on. A station between the ejection and

658 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

D Injecting preform

c=:::{ll]

)

Blow molding and ejection

Figure 7-63. The basic injection blow molding process. preform stages can be used to detect if ejection has not occurred, to add an insert to the core pin, and so forth. There are several different IBM methods available, with different means of transporting the core rods from one station to another. These methods include the shuttle, two-parison rotary, axial movement, and rotary with three or more stations used in conventional 1M clamping units. A variation of IBM is displacement BM or dip molding. Here a premeasured amount of hot melt is deposited into a cupel the shape of a preform. A core rod is inserted into the cupel, displacing the melt and packing it into the neck finished area. It then moves to a blow station, where it basically follows the procedure described above. The advantages of IBM include the lower-cost machines and the fact that a nearly stress-free bottle is blown.

Stretching and Orienting High-speed EBM and IBM are taken an extra step in stretching or orienting. Figure 7-64 shows stretched IBM; with EBM the stretching action is similar to this, as it occurs during compressed air inflation (see Fig. 7-57). In EBM the parison, which is mechanically held at both ends, is stretched rather than just blown. Stretching can include the use of an expanding rod within the IBM preform or an external gripper with EBM. By biaxially stretching the extrudate before it is chilled, significant improvements can be obtained in the finished containers. This technique allows the use of lower-materialgrade resins or thinner wall thicknesses, with no decrease in strength; both approaches reduce material costs. Stretched BM gives many resins (mono- or bioriented) improved physical and barrier properties. The process allows wall thicknesses to be more accurately controlled and allows weights to be reduced.

THE PROCESSING OF PLASTICS 659

ill

Inject preform

Reheat preform

Stretch blow molding and ejection Figure 7·64. A schematic of stretch injection blow molding.

Design Guidelines Basically, BM products should be designed with generous radii at their comers and edges. Fillets and rounds should be employed wherever possible in comers, ribs, and edges (see Fig. 7-65). Such parts will possess more uniform wall thickness and, as a result of more uniform and faster cooling, internal stresses will be reduced. With all this action, various techniques can also be used for strengthening parts: Lower blow ratios, which make for thicker wall sections. Flutes with concave ribbing, resulting in increased strength. Horizontal ribs, adding strength without increasing wall thickness. They are superior to vertical ribbing for drop impact and container-collapse resistance. Because certain plastics, such as PET, must be cooled below their glass transition temperature (see Chapter 2) before they can be removed from the mold, undercuts of any significant depth will not strip from a mold. Moldings of PET can usually be

660 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7-65. Automotive panels like these extrusion blow-molded ones have generous radii at their corners and edges. stripped from undercuts as deep as 0.010 inch per inch (0.003 mm/mm) of cavity width, but probably not much greater. The maximum allowable undercut is influenced by the design of the undercut and the usual factors such as the thickness of the sidewall. Diameter change applies to round containers and increases rigidity and strength by increasing wall thickness at the reduced diameter or blowing radius. Increased performance or stiffness is achieved using elementary strength of material theory with shapes such as corrugation, 1- and T-shapes, and ribs. Tapering imparts an angular support at its junction with a walled section differing in shape or direction. Wall thickness also increases with the decreasing diameter and blow ratio. Tolerances

Lengthwise shrinkage in BM products tends to be slightly greater than transverse shrinkage. Most of the horizontal shrinkage occurs in wall thickness rather than in body dimension. In the case of polyethylene, higher shrinkage occurs with the higher-density polymers and thicker-walled products. The former is due to greater crystallinity of the

THE PROCESSING OF PLASTICS 661

more linear polymer and the latter to the slower cooling rates that result in more orderly crystalline growth. Mold shrinkage is dependent on many factors, such as resin density, stock temperature, mold temperature, part thickness, and blowing air pressure. Typical polyethylene blow molding shrinkage is as follows: Low-Density Polyethylene: Thickness up to 0.075 in. Thickness over 0.075 in. High-Density Polyethylene: Thickness up to 0.075 in. Thickness over 0.075 in.

0.010-0.015 in.lin. 0.015-0.030 in.lin. 0.020-0.035 in.lin. 0.035-0.055 in.lin.

Once the operating conditions are established, tolerances of ± 5 percent may be expected. The cavity defines the shape of the article. As with other molding processes, cavity dimensions are enlarged slightly to compensate for resin shrinkage. For polyolefins, particularly in the neck finish, slightly higher shrinkage rates are used for extrusion blow molding than for injection blow molding. The amount for the body is 2 percent and is as high as 3.5 percent for the neck finish. Rigid resins, however, remain about the same. The most common special feature of the mold is the "quick change" volume control insert. Rigid volume control is a must for certain products such as dairy containers. Unfortunately, an HOPE container slowly shrinks and changes in size for many hours after molding. Because of production "volume control" requirements some dairies fill containers molded half an hour before and then switch to fill containers molded several days before. Volume control inserts that displace the difference in size are added to the mold (usually as a disk in the side wall) to ensure that volume and fill levels are the same in both containers at the time of filling. The device works because HOPE shrinkage is reduced to virtually zero for the life of the container when filled with milk or juice and stored at cold temperatures. Several sizes are available to suit particular molding needs. Sound product design is the first step in mold design: the mold maker is the last step. The mold maker is a highly skilled craftsman whose judgment and ability are used to duplicate the cavities, hand finish and fit the parts, engrave the lettering, and manufacture the molds. The mold maker's commitment, attention to detail, and pride are as important a contribution as the design itself. Shrinkage may be reduced by raising the blowing pressure and lowering the temperature of the mold. Rapid cooling is desired, but cooling too rapidly can cause surface imperfections, distortion, and frozen-in stresses. Raising the stock temperature, while it may not appreciably affect outside dimensions, causes more of the shrinkage to occur in wall thickness, because the higher melt temperatures lessen strain recovery and reduce blowing stresses. Shrinkage must be taken into consideration when designing a mold and, if necessary, must be confirmed by data and measurements obtained from preliminary trials. Close tolerances on the capacity of the blown item also influence mold construction. Where multiple molds are required, capacity from one mold to another must be closely held. In general, some of the casting processes have been more successful in reducing capacity variation than in duplicating or hobbing. For final capacity adjustment, an inserted bottom plate or other adjustable insert is advisable. By this method, slight changes can be made on each mold insert without noticeable variation in finished parts.

662 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Plastic will sometimes thin out at an insert or parting line. With some materials, like low-density polyethylene, this is of little importance, but with high-density polyethylene and similar materials it can be a serious problem. When inserts are used, they should be fitted tightly. The parting-line miter should be as close as possible, and sufficient clamping pressure should be provided to prevent partial opening of the dies. Solid molds have a slight technical advantage over molds with inserts in these instances. An interchangeable fit with other parts requires serious consideration in mold construction. This type of construction includes plug fits, screw- and snap-type caps, snap beads, and so on. It is strongly advised to use inserts for the interchangeable portion. This would allow for changes of inserts to accommodate different shrinkage conditions, different materials, and the like. Inserts may be welded on or embedded during molding. Weldedon parts must be melted onto the blow-molded article. Inserts that are to be embedded must be anchored with the aid of undercuts, knurling, or tags. They may consist of metal or other materials. For internal embedding they are placed on mandrels, for external embedding, in retaining holes in the mold cavity, which should be in line with the mold's closing movement.

Handles Many designers are familiar with the fact that the extrusion blown process can include a "blown" handle; a typical product is the HDPE milk bottle. What many do not recognize is that the injection blown process can also readily include a handle that is not blown, as shown in Figure 7-66. This integral carrying handle was issued in a French patent granted to an Italian company. This patent incorporates a traditional jug handle above the blown portion of an injection blow-molded container, as for example a pitcher. The handle is molded as part of the preform and is undisturbed when the container is blown. A direct extrapolation to stretch blow-molding technology would incorporate the jug handle immediately below the neck finish of an injection-molded preform. It would be necessary to mold such a preform in a split mold that suits itself to production on certain rotary-type injection stretch blow equipment. Another approach to a carrying handle provides for gripping features being molded to a ring or strap placed just below the neck finish. Various carrying handles are shown in Figure 7-67. Such a preform does not require a split injection mold and adds only a nominal amount to the clamp force and platen area of an injection-molded preform. Attention would have to be given to the arrangement of the cavities in the injection mold, with the possibility that the number of cavities could be restricted in some situations. Such a handle would not affect bottle shape or height, which provides design freedom for either aesthetic or functional purposes. Such a handle could be incorporated, for instance, on a soft drink bottle without compromise to the strictly engineered shape. An approach (one more of many that a designer could conceive) to producing an integral pouring handle or strap is shown in Figure 7-67. There is an appendage in this design to the preform below the neck finish, which can be attached at its lower end to the blown portion of the container. The appendage may be a portion of an injectionmolded preform or it may be attached to the preform prior to stretch blow molding. Attachment of the bottom end of the handle can be made in many different ways, including welding, gluing, or use of a restraining means such as a wraparound label. The lower end of the handle may be used as a part of the blow mold surface and, if suitably

THE PROCESSING OF PLASTICS 663

2

3

Figure 7-66. An integral carrying handle for an injection blow-molded product. 1) The precision "neck" mold, including a solid handle; 2) the preformed core and blow pin; 3) the basic water-cooled bottle female mold; and 4) the injection nozzle of the injection-molding machine.

configured, may produce a mechanical interlocking to the blown portion of the bottle. The handle may be semielliptical in cross-section to conform comfortably to the hand.

Collapsibility Containers An interesting and practical design involves a patented bellows-collapsible bottle produced in conventional BM equipment by the Collapsible Bottle of America Co. These "foldable" in contrast to "passive" bottles provide advantages and conveniences such as 1) reducing storage, transportation, and disposal space; 2) prolonging product freshness by reducing oxidation and loss of carbon dioxide; and 3) providing continuous surface access to foods such as mayonnaise and jams. Best of all, they provide the consumer with a different, futuristic, fascinating package. The marked advantages of the collapsible bottle over conventional packages justify the product's trade name: "The Smart Bottle" (see Figs. 5-12 and 5-13).

Complex Irregular Shapes As mentioned, BM products comprise more than the major market of relatively conventional bottles. Complex, irregular shapes can be blown to meet different practical and engineering structures, from small to large parts. For example, double-walled components,

664 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~

Figure 7-67. Ring, strap, and integral pouring handles.

large insert moldings, and hollow-ribbed structures in products range from chairs to truck parts (see Figs. 7-52-56, 7-65, 7-68, and Table 7-16). All types of action, such as side motions, tapered slides, and unscrewing mechanisms are used in blow molds (see Fig. 7-69). These actions have been used extensively for a long time in injection molding [12]. Designs in them have many possibilities to develop useful, complex parts [11]. Their geometry involves reverse folds or inwardly protruding channels and narrow projections. Blow molding offers designers innovations in high-performance parts. Effective design requires new or at least different thinking on the part of the designer. To make the most of this opportunity requires thinking in terms of both the overall finished parts and the interdependence of the fabrication process and design. The closed hollow shape produced can take many forms and can be used in many different ways, from a hollow container to a satellite dish or dish panels for the larger dishes like the radomes in Chapter 3. One hollow extruded parison form can produce several parts at once, similar to the way a multicavity injection mold works [12]. Or multiple, single-walled parts can be made as symmetric parts of a single molding. Parts can be blow molded in one piece, whereas in injection molding two or more parts must be welded together. Certain BM parts can also be lower in cost or provide better performance. It is possible to produce a high-strength one-piece plastic with chemical and temperature resistance that allows for containing high internal pressures at high temperatures. Some examples of part made like this include under-the-hood items in autos and paint oven environments. Having lower molded-in stress means that large blow-molded panels are less likely to warp. Hollow structures give better strength-to-weight ratios, while molded-in ribbing or channels provide additional strength, with extremely complex shapes possible. BM also allows for bold styling and aesthetic highlighting, which are not always economically

THE PROCESSING OF PLASTICS 665

Observe proper blow ratio for side dud

Trim

Section through 0 hollow wall blow molded pori

Slots are a secondary aelion

I

Compressed flange for mig

----===--__ Single piece

Figure 7-68. An example of blow-molded spoiler air duct.

Figure 7,(,9. A mold used in extrusion blow molding a complex shape that includes a threaded forming core. Shown are the three-part mold in the open and closed positions, with the blow pin located in the top two sections of the mold.

possible with other processes. Smooth-lined styling lends itself well to BM, and surface textures ranging from matt to a class A automotive finish are easily obtained. Moldedin color can also result in significant savings in total finished part cost. Other examples of high-performance, high-strength BM plastic parts are readily available to designers. FORMING

Formed or shaped plastics provide a great variety of marketable products, in a wide size range (see Fig. 7-70). Different techniques are used, with thermoforming being the most productive and most diversified. Other techniques are basically similar to thermoforming but usually use less heat than it requires and are more limited as to the type of plastic

666 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 7-16. Hollow and Structural Blow-molded Shapes Industry Automotive

Furniture

Appliance Business machine Construction Leisure

Industrial

Application

Required Properties

Spoilers Seat backs Bumpers Underhood tubing Work stations Hospital furniture Office furniture Outdoor furniture Air-handling equipment Air-conditioning housings Housings Ductwork Exterior panels Flotation devices Marine buoys Sailboards Toys Canoes/kayaks Tool boxes, ice chests Trash containers, drums Hot-water tanks

Low temperature, impact, cost Heat distortion, strength/weight Low temperature, impact dimensional stability Chemical resistance, heat Flame retardance, appearance Flame retardance, c1eanability Flame retardance, cost Weatherability, cost Flame retardance, hollow Heat distortion, cost Flame retardance, cost Cost Weatherability, cost Low temperature, impact strength cost, weatherability Low temperature, impact strength cost, weatherability Low temperature, impact strength cost, weatherability Low temperature, impact strength cost, weatherability Low temperature, impact strength cost, weatherability Low temperature, impact strength, cost Low temperature, impact strength, cost Low temperature, impact strength, cost

used; these processes include cold forming, stamping or compression forming, flow molding, rubber pad molding, diaphragm forming, coining, forging, and so on. Formed parts are used in many different applications and production lines (form, fill and seal, etc.). Food, electronic devices, medical products, and other parts use continuous thermoforming operations at the end of high-speed production lines to reduce the handling of products, provide hermetically sealed contents, reduce costs, and so forth [10-12, 6264, 651-62].

Thermoforming Thermoforming usually consists of heating extruded thermoplastic sheet, film, and profile to its softening heat, then forcing the hot, flexible material against the contours of a mold by pneumatic means (differentials in air pressure are created by pulling a vacuum between the plastic and the mold, or the pressure of compressed air is used to force the material against the mold), mechanical means (plug, matched mold, etc.), or combinations of pneumatic and mechanical means. The process involves 1) heating the sheet (film, etc.) in a separate oven and then transferring the hot sheet to a forming press, 2) using automatic machinery to combine heating and forming in a single unit, or 3) a continuous operation feeding off a roll of plastic or directly from the exit of an extruder die (postforming). Practically all the materials used are extruded TPs; very small amounts may be calendered or cast. To date few thermosets are used, as the markets have not developed for them. These TSs can be either unreinforced or reinforced. Practically any TP can be used, but certain types are easier to use, permitting deep draws without tearing or excessive thinning in certain areas such as corners. The ease of forming depends on the materials' characteristics; it is influenced by minimum and maximum thickness, pinholes, ability of the material to retain heat profile gradients across the surface and the thickness, the controllability of applied

THE PROCESSING OF PLASTICS 667

Figure 7·70. Here ABS 76 x 230 in. sheets are conveyed to an IR heating oven in back of a console where the sheets are individually heated and formed into I5-ft. outboard-powered runabouts. The entire automated process of conveying the sheet, heating, forming, and cooling takes only ten minutes. stress, the rate and depth of draw, the mold geometry, the stabilizing of uniaxial or biaxial deformation, and, most important, minimizing the thickness variation of the sheet. PS and other plastics with a high melt viscosity are particularly desirable. HIPS is an example of a plastic with the special characteristic of being able to be stretched to over 100 times its original length. Table 7-17 lists some of these plastics and how they can be formed. Coextruded sheets are used to meet different design requirements (see Fig. 7-71 through 7-73). In many applications of conventional thermoforming, low-cost tooling is used compared to that of other processes, particularly in cases of limited production or the forming of large parts. Thermoforming of thin parts has an advantage over most other processes, where very thin walls cannot be produced. An example of its use is in skin and blister packaging. To improve the strength or structural performance of formed plastics, the processor can utilize design features such as corrugations, box shapes, and so on. These features are easily incorporated with thermoforming. One example of minimizing the thickness of a product and improving its strength is a formed drinking cup with a rolled edge (which, incidentally, also eliminates cutting lips). With most forming (not including bending) there can be up to 50 percent scrap trim or web. This material could be wasted, but it is usually recycled and blended with virgin resin. Individual sheet stock formed into round shapes could have 50 percent or more scrap. With square forms, there could be up to 25 percent scrap. The y;arious thermoforming techniques are generally described in terms of the means

668 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 7-17. Thermoformed Plastics and Their Products Polystyrene Polystyrene foam Acrylic Rigid vinyls Acrylonitrile butadiene styrene (ABS) Cellulose acetate Cellulose propionate Cellulose acetate butyrate High-density polyethylene Nylon Polycarbonate Polypropylene

Various packaging applications, including transparent meat trays, trays for cookie and candy boxes, blister packages Meat trays, egg cartons, take-out food containers Signs and other outdoor applications like motorcycle windshields, snowmobile hoods, and recreational-vehicle bubble tops Lighting panels, signs, relief maps, bus-interior panels, dishes and trays for chemicals, blister packages, automobile dashboards Recreational-vehicle components, luggage, refrigerator liners, businessmachine housings Blister packages, rigid containers, machine guards Machine covers, safety goggles, signs, shipping trays, displays Skylights, outdoor signs, pleasure-boat tops, toys Camper tops, canoes, sleds Reusable trays, outdoor signs, surgical equipment, meat trays Outdoor lighting, face shields, machine guards, aircraft panels and ducts, signs Truck-fender liners, drinking cups, juice and dairy-product containers and lids, test-tube racks

used to form the sheet, such as bending, vacuum forming, pressure forming, plug-assist forming, matched-mold forming, and so on. The different methods enable the processor to form different-shaped products to meet various performance requirements. An evaluation of these methods shows that simple to complex shapes can be formed, with the shape as well as the surface condition able to be accurately controlled outside, inside, or on both sides. The basic concept of thermoforming is shown in Figure 7-74 (see Reference 10 for the different thermoforming techniques).

Other Forming Methods Certain thermoplastics, when formed, requiring handling that is normally not available with conventional thermoforming machines, so other processes of formation have evolved. Most of these methods tend to reduce the amount of heat required or event eliminate it entirely. One popular technique is high-pressure forming, which is like conventional compression molding. The techniques that are used modify conventional metalworking tools. They can be classified as 1) cold forming (performed at room temperature with unheated tools), 2) solid-phase forming (plastic is heated to below its melting point and then formed), and 3) compression molding of reinforced composite sheets (heat is used). Other methods so used are classified as forging (which includes closed-die forming, opendie forming, cold pressing, etc.), stamping, rubber pad or diaphragm forming, fluid forming, coining, spinning, explosive forming, and so on. Cold forming and solid-phase forming include the use of ABS/PC, PC, conventional PP, and HMWHDPE. By using solid-phase forming, processors can make more efficient use of ultrahigh molecular weight, high density plastics that are difficult or impossible to process by other methods. Forming by these techniques can usually use existing metalworking equipment, with minor modifications. Tooling is inexpensive, and production rates can be high. Flash, trim, or weld lines can be eliminated by using some of these processes. Thermoplastic composites can be stamped to produce high-performance parts. Fiber reinforcements can be used, including glass, graphite, aramid, and so on, in different patterns (short fibers, woven, etc.). Products are molded in quick, high-productivity

UNIFORM THICKNESS

BONO

Figure 7-71. The bonding of two thennofonned parts from coextruded sheets here resulted in a fuel tank. The tank's inside has a nylon surface to resist gasoline; the outside uses PP to provide a combination of low cost and the required support strength. WHITE

CLEAR

\

\

,

f

.,.../

I

\

COLOR

Figure 7-72. A coextruded sheet for the production of a three-color thennofonned container with an integrally hinged lid . CLAMPSTRIP (RECLAIMABLE)

I

CLAMP STRIP (RECLAI MABLE)

\

COLOR OR CLEAR

\I

\

\

COLOR OR CLEAR

Figure 7-73. The addition of a single-plastic clamping strip at each side of a coextruded sheet permits scrap reclaim of the thennofonned trim waste.

I

HEATER

)\«(\\\\(

(1\

VACUUM ON

Figure 7-74. The basic concept in vacuum thennoforming. 669

670 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

processes, using less energy than that needed to manufacture comparable aluminum and steel parts. Tooling costs decrease because of part consolidation. Stamping involves two quite different forming processes: solid-state forming and flow molding, or fast-compression molding. Each has its advantages and disadvantages.. Postforming

A popular forming technique that has provided both performance and cost advantages, principally for long production runs, is applied as the plastic sheet, film, or profile exits an extruder. Upon leaving the die, and retaining heat, the plastic is continuously postformed. With this type of in-line system the hot plastic is reduced only to the desired heat of forming. All it may require is a fixed distance from the die opening. Cooling can be accelerated with blown air, a water spray, a water bath, or some combination. Examples of some postforming techniques are shown in Figures 7-75 through 7-79. This equipment, like others, requires precision tooling with perfect registration.

REINFORCED PLASTICS/COMPOSITES When methods of processing plastics are discussed, the category of reinforced plastics or composites represents many different fabrication processes, some of which have been covered already. Some of these processes are listed in Figure 7-80. A broad range of properties exist for RPs (see, for instance, Table 7-18). As reviewed in the RP section of Chapter 6, both TSs and TPs are used in RPs, thus providing it a broad range of properties [1-21, 27-33, 62-70, 635-50, 663]. Choosing the optimum process encompasses a broad spectrum of possibilities. In some situations only one process can be used, but generally there are options. Influencing EMBOSSING ROLL EMBOSSED EXTRUSION

EMBOSSING ROLL EMBOSSED EXTRUSION

\

~

EXTRUDER Ole

BACKUP ROLL..-/" EMBOSSING ROLL

POSITIVE ROLL

~

~

\

\

DEEPLY EMBOSSED EXTRUSION

DEEPLY EMBOSSED EXTRUSION RUBBER BACKUP ROLL

"NEGATIVE ROLL

Figure 7-75. In-line postforming with an extruder: embossing one or both sides with shallow or deep patterns.

-

C1\

"

Reinforcement wt'll> MPa

ksi

GPa

10" psi

Modulus

Tensile

MPa

Flexural Strength ksi MPa

Strength ksi

Compressive

J/m

Spray(a)

1.~2.3

1.~2.3

0.27-0.33 0.27-0.33

0.22-0.27 1.5-1.85

0.22-0.33

0.22-0.33

480-720 540-720 120-300

60-360

48-300

"F

kV/em

kV/in.

Dielectric Strength

350-400

175-205

1.5-2.3

200-300

175-230 350-450

95-150

80-120 200-300

80-130 200-325

80-120 200-300

80-160 200-400

350-400 120-160 300-400

350-400 120-240 300-600

205-260 400-500

175-205

175-205

175-205 350-400 80-160 200-400 205-260 400-500 120-180 300-450

'C

Heat Distortion at 1.8 MPa

1.5-2.3

1.3-1.8 0.1~.26

120-240

1.2-1.6 1.3-1.1 0.1~.25

0.17-0.23

W/m·K

8m' in.1h . ft2 . "F

Thermal Conductivity

48-144 96-264

ft·lbr/ft

Impact Strength

30-50 glass5.5-12 0.8-1.8 110-190 16-28 100-170 15-25 210-640 ~18 polyester 60-120 8-20 11-17 1.6-2.5 120-210 18-30 100-210 15-30 430-1,150 Compression(a) 15-30 glass-SMC 55-140 Compression(a) 25-50 glass 10-40 100-210 15-30 530-1,050 25-30 6.2-14 0.~2.0 70-280 mat-polyester 170-210 30-80 glassFilament 550-1,700 80-250 28-62 4.0-9.0 690-1,850 100-270 310-480 45-70 2,150-3,200 epoxy winding(a) Pultrusion(b) 40-80 glass mat-polyester 410-1,050 60-150 28-41 4.0-6.0 690-1,050 lOO-ISO 210-480 30-70 2,400-3,200 Paltrusion(b) 30-50 glass 530-1,350 25-30 210-340 30-50 80-210 12-30 6.~17 1.0-2.5 170-210 mat-polyester Pultrusion(c) 30-55 glass mat and roving-vinyl 270-1,600 15-40 140-340 20-50 70-280 10-40 6.~21 1.0-3.0 100-280 ester resin Pultrusion(e) 30-55 glass mat and roving10-30 100-280 15-40 210-1,350 7-35 5.5-17 0.8-2.5 70-210 polyester resin 50-240

Process

Tensile Strength

Table 7-18. An Overview of RP Properties and Processes

EXTRUDATE

\

VACUUM ......, PORTS ./

/~UUM

PLENUM

VACUUM DRUM

DRUM SECTION

Figure 7·76. An in-line vacuum forming embossing roll with water-cooled temperature control. PRESSURE·FORMING DIES

UPPER CONVEYOR

VACUUM·FORMING DIES

Figure 7·77. An in-line vacuum/pressure fonner for plastic sheet with matched, water-cooled, forming molds on a continuous conveyor system. This system can be used with different profiles, such as small and large tubes producing corrugated tube or pipe. Moving molds would have corrugated tubular cavities with vacuum/pressure/water-cooling lines.

- - - ROTATING MANDREL

Figure 7·78. An in-line coil former that can produce telephone cords, springs, and so on using extruded round, square, hexagonal, and other shapes. 672

THE PROCESSING OF PLASTICS 673 ROTATING RING

,

TWISTED EXTRUSION

Figure 7-79. In-line fixed/rotating rings used to twist extrudate.

process selection are quantity, size, thickness, tolerances, the type of material, and performance requirements (see Fig. 7-81). Regarding tolerances, as mentioned in other chapters, resins with fillers or reinforcements are generally far more stable in meeting tight tolerances. (In fact, the TSs, whether unreinforced or reinforced, are more dimensionally stable than other resins.) RP parts are fabricated by processes using pressures that range from contact (or no pressure) through moderate 350 to 700 kPa (50-100 psi) on up to thousands of psi. Temperatures can range from room temperature to the usual 121 to 315°C (250 to 600°F) and on up, particularly for certain high-performance TPs. The time cycles can range from seconds to minutes, hours, or even days. Processing may involve equipment that is simple to operate or requires extensive specialized equipment. Among the most common processes are contact molding methods (hand lay-up, spray-up, use of a vacuum bag, pressure bag, autoclave, etc.), matched mold methods (compression molding, transfer molding, resin transfer molding, injection molding, compression-injection molding, stamping, etc.), and other methods (filament winding, cold press molding, pultrusion, continuous laminating, centrifugal casting, encapSUlation, rotational molding, reaction injection molding, etc.). Tables 7-19 through 7-21 summarize some of these processes.

TYPE OF REINFORCEMENT

METHOD OF PROCESSING

Yarns/fabrics 0(00;=-------::::1,0 Hand lay-up Roving (Cllrr-~~_-"'~""o() Spray-up Chopped strands

_~~......."V

Chopped strand mats, HSB Chopped strand mats, LSB (includes continuousstrand mat)

~~~~~~

Veil and surfacing mats Sliver/woven roving Mat and woven roving combinations Milled fiber

~

Vacuum-bag molding Tooling Cold-press molding Casting Architectural paneling Centrifugal casting Pressure bag molding

~I.~....-"-' Premix - BMC

Preform - SMC Pultrusion

Glass beads and spheres

Filament winding

Glass flake

Injection molding Rotational molding Cold forming

Figure 7-80. Various reinforcements and processes used for manufacturing reinforced plastics or composites.

674 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Filament or

tape winding

150,000

·i ..: 1iG c

.

~

100,000

1i c ~

50,000

o~--~~--~----~----~

100

75

50

25

0

Percent reinforcement, by weight Figure 7-81. An example showing the strength of RPs versus the amount of glass reinforcement and the types of reinforcements.

Hand Lay-up

This method is the oldest, and in many ways the simplest and most versatile, process, but it is slow and very labor intensive. It consists essentially of the hand tailoring and placing of layers of (usually glass fiber) mat, fabric, or both on a one-piece mold and simultaneously saturating the layers with a liquid TS resin (usually polyester). The assemblage is then cured, with or without heat, and commonly without pressure (see Fig. 7-82). Alternatively, preimpregnated B-staged, partially cured dry material such as SMC may be used, but in this case heat is applied, with the probability of applying low pressure. Fabrication begins with a pattern from which a mold is made. The mold may be of any low-cost material, including wood, hard plaster or hydrostone, concrete, a metal such as aluminum or steel, and glass-fiber-reinforced polyester or epoxy. If only a few parts are to be made, a single mold will suffice; otherwise, multiple molds may be required. If the volume is large enough and speed is important, heating elements such as lines for steam or other fluids, or electrical heat units, may be incorporated. Automated equipment may also be installed (see Fig. 7-83). The mold may be male (plug) or female (cavity), depending upon which side of the formed part is to have the accurate configuration (the other side will be rough). Spray-up

An air spray gun includes a roller cutter that chops glass-fiber rovings to a controlled length before they are blown in a random pattern onto a surface of the mold simultaneously with a spray of catalyzed resin. The chopped fibers are coated with resin as they exit the gun's nozzle. The resulting, rather fluffy, mass is consolidated with serrated rollers to

Q\

""

'I

·Courtesy Owens-Coming Corp. tFRP = Fiberglass reinforced plastics.

Generally expected mold life (parts)

3,000

Continuous strand mat, preform, woven roving

Reinforcement

Part trim equipment

High shear type

Pressure feed pumping equipment req'd; mold halves clamped (methods range from clamp frame to pressure pod)

FRP, t spray metal, cast aluminum; gasket seal, air vents, self-sealing injection port

Resin compounding equipment

Cure system

Pressure

Mold construction

Resin Transfer Molding

None

FRP

Hand Lay-up

Continuous roving

1,000

Yes

roving, cloth

mat, woven

Chopped strand

Not needed

Room temperature

Spray-up

Open Molding

3,000

Continuous strand mat, preform, woven roving

Lows pressure press, capable of 50 psi (hydraulic or pneumatic mechanical); resin dispensing equipment not req' d but recommended

FRP, spray metal, cast aluminum, pinch (land)

Cold Press Molding

Continuous roving (specific orientations for higher strength)

(I 35-177°C)

350"F

Heated; normal range of 275-

Hydraulic; as high lIS 2,000 psi (13.8 MPa)

High grade steel; shear edge

Sheet Molding Compound

150,000+

150,000+

With optimum shear edges, minor trimming only

Continuous strand mat, preform, woven roving

High shear type

325°F (l07-163°C)

Heated; normal range of 225-

Hydraulic press, normal range of 100-500 psi (0.69--3.05 MPa)

Metal, shear edge

Mat-Preform

Compression Molding

Table 7-19. Process Comparison of Various Reinforced Plastics Manufacturing Techniques*

Figure 7-82. An example of a hand lay-up RP process where a glass-fiber-reinforced TS polyester resin boat structure has been cured and is being removed from its mold.

Figure 7-83. An automated-integrated RP lay-up process that uses TS-preimpregnated reinforced sheets that are in the B-stage of curing.

676

THE PROCESSING OF PLASTICS 677

Table 7-20. Comparison of Resin Transfer Molding, SMC Compression, and Injection Molding Process RTM Process operation: Production requirement, annual units per press Capital investment Labor cost Skill requirements Finishing Product: Complexity Size Tolerance Surface appearance Voids/wrinkles Reproducibility Cores/inserts Material usage: Raw material, cost Handling/applying Waste Scrap Reinforcement flexibility Mold: Initial cost Cycle life Preparation Maintenance

SMC Compression

Injection

5,000-10,000 Moderate High Considerable Trim flash, etc.

50,000 High Moderate Very low Very little

50,000 High Moderate Lowest Very little

Very complex Very large parts Good Gel coated Occasional Skill dependent Possible

Moderate Big flat parts Very good Very good Rarely Very good Very difficult

Greatest Moderate Very good Very good Least Excellent Possible

Lowest Skill dependent Up to 3 percent Skill dependent Yes

Highest Easy Very low Cuts reusable No

High Automatic Sprues, runners Low No

Moderate 3,000--4,000 parts In factory In factory

Very high Very high Years Years Special mold-making shops Special machine shops

squeeze out air and reduce or eliminate voids. As in hand lay-up, the first layer of a gel coat may be applied over the mold, followed by successive passes of the sprayed-on composite before a final gel coat is applied. If required, inserts and so on can be included during the spraying operation.

Vacuum Bag A molded part made by hand lay-up or spray-up is allowed to cure without the application of external pressure. For many applications this approach is sufficient, but maximum consolidation usually is not achieved with its use. There is some porosity, fibers may not fit closely into internal comers with sharp radii but tend to spring back, and resin-rich or resin-starved areas may occur because of drainage, even with thixotropic agents. With moderate pressure these defects can be overcome, with an improvement in mechanical properties and better quality control of parts. One way to apply such moderate pressure is to enclose the "wet-liquid resin" composite and mold in a flexible membrane or bag, and draw a vacuum inside the enclosure. Atmospheric pressure on the outside then presses the bag or membrane uniformly against the wet composite. Pressures commonly range from 69 to 283 kPa (10-14 psi).

678 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 7-21. Guide to Compatibility of Materials and Processes Thermoplastics

Thermosets

'"

'" :s!

'0


u

u

ell

I:Q

.,

~ .,

~

....

.,

~

Process

....

~

~

i i i'"

><

"3 ., ~

~ ..s

~

~

I

IQ IQ

~

t.l

0(

iQ :::

!Z .aZ

::: '"

~

~ ~

::: '"

1:!

j

'"

.§.

'" ;;, c.. 8c..

~ :::

"0

~

>.

~

0

~

>.

:::

ell

I:Q

0(

::: e'" .,

~

..

~

i i i'"

Injection molding Hand lay-up Spray-up Compression molding Preform molding Filament winding Pultrusion Resin transfer molding Reinforced reaction injection molding

Pressure Bag If more pressure is required than what is available with the bag vacuum system, a second envelope can be placed around the whole assemblage and air pressure admitted between the inner bag and outer envelope. This method is also called the vacuum-pressure bag process.

Autoclave Still higher pressures can be obtained by placing the vacuum assemblage in an autoclave. Air or stream pressures of 690 to 1,380 kPa (100-200 psi) are commonly achieved. If still higher pressures are required and the danger of extremely high air pressures is to be avoided, a hydroclave may be used, employing water pressures as high as 6,900 kPa (1,000 psi). The bag must be well sealed to prevent infiltration of high-pressure air, steam, or water into the molded part. In these processes an initial vacuum mayor may not be employed.

Pressure Bag Molding In this process seamless containers, tanks, pipes, and other products can be made. A preform is used that is made by a spray-up, a mat, or a combination of materials (using a perforated screen to produce the preform). Only enough resin is included in the preform to hold the fibers in place. Usually it amounts to about 0.5 percent by weight and is compatible with the matrix resin. An inflatable elastic pressure bag is positioned within the preform and the assembly is put into a closed mold, which could be a drum, for instance. Resin is injected into the preform and the pressure bag is inflated to about 345 kPa (50 psi). Heat is applied and the part is cured within the mold. When cure is complete, the bag is deflated and pulled through an opening at the end of the mold, and the part removed.

THE PROCESSING OF PLASTICS 679

Compression and Transfer Molding Compression and transfer molding (CM and TM) are the two main methods used to produce molded parts from generally thermoset (TS) resins. Compression molding was the major method of processing plastics during the first half of this century because of the development of a phenolic resin (TS) in 1909 and its extensive use at that time. By the 1940s this situation began to change with the development and use of thermoplastics (TPs) in extrusion and injection molding (1M) processes. CM originally processed about 70 percent (by weight) of all plastics, but by the 1950s its share of total production was below 25 percent, and now that figure is about 3 percent. This change does not mean that CM is not a viable process; it just does not provide the much lower cost-to-performance benefit of TPs, particularly at high production rates. In the early 1900s resins were almost entirely TS (95 percent by weight), but that proportion had fallen to about 40 percent by the mid-1940s and now is about 3 percent. During this century, TSs have experienced an extremely low total growth rate, whereas TPs have expanded at an unbelievably high rate. Regardless of the present situation, CM and TM are still important, particularly in the production of certain low-cost parts as well as heat-resistant and dimensionally precise parts. CM and TM are classified as highpressure processes, requiring 13.8--69 MPa (2,000 to 10,000 psi) molding pressures (see Figs. 7-84 through 7-86). Some TSs, however, require only lower pressures of down to 345 kPa (50 psi) or even just contact (zero pressure). Compression molding is the most common method of molding TSs. In this process, material is compressed into the desired shape using a press containing a two-part closed mold and is cured with heat and pressure. This process is not generally used with TPs.

Foam Reservoir Molding This low pressure process, also known as elastic reservoir molding, consists of making basically a sandwich of resin-impregnated open-celled flexible polyurethane foam between the face layers of fibrous reinforcements. When this composite is placed in a mold and squeezed, the foam is compressed, forcing the resin outward and into the reinforcement. The elastic foam exerts sufficient pressure to force the resin-impregnated reinforcement into contact with the mold surface.

Resin-Transfer Molding Resin-transfer molding (RTM) is a closed-mold, low-pressure process in which a preplaced dry reinforcement preform is impregnated with a liquid resin (usually polyesters, although epoxies and phenolics may be used) in an injection or transfer process, through an opening in the center of a mold. The preform is placed in the mold and the mold is closed. A two-component resin system (including catalyst, hardener, etc.) is then mixed in a static mixer and metered into the mold through a runner system. The air inside the closed mold cavity is displaced by the advancing resin front and escapes through vents located at the high points or the last areas of the mold to fill, as in injection molding. When the mold has filled, the vents and resin inlets are closed. The resin within the mold then cures and the part can be removed. The advantages of RTM are that the molded part has two finished surfaces, and the overall process may emit a lower level of styrene vapor if the polyester resin used contains

MOlD (HEATED)

MOlDCLOSEO

(a)

PRESS OPEN

MOlD (HEATED)

PRESS CLOSED

(b)

Figure 7-84. Schematics of a) compression molding and b) transfer molding.

rototable insert part -

rotatoble stud inser t

-

c::.:::J

_

EJims,. . ,.Q

L

mold

~

Centerline Parallel to Mold Movement

Cen te rline Perpendicular to Mold Movement

(a)

(b)

L:f*LJ]

L-Q-iJ

C

~ -

~

c:=tl (c)

~

re

rotating

¢D

~ (d)

Figure 7-85. Examples of compression-molding threads in various positions. 680

par t ing line will appear on threads

core

Figure 7·86. This 4,OOO-ton compression press is here fabricating truck and coach hoods, as well as fenders of glass-fiber-TS polyester. This view shows a hood that has been molded, with its separately molded grill attached.

681

682 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

styrene. The mold, unlike a compression or TP stamping mold, is completely closed to define stops prior to final part formation/curing. This procedure provides a more reproducible part thickness and tends to minimize trimming and deflashing of the final part.

Reaction Injection Molding RIM is very similar to RTM. In the reinforced RIM (RRIM) process a dry reinforcement preform is placed in a closed mold. Next a reactive resin system is mixed under high pressure in a specially designed mix head. Upon mixing, the reacting liquid flows at low pressure through a runner system to fill the mold cavity, impregnating the reinforcement in the process. Once the mold cavity is filled, the resin quickly completes its reaction. The complete cycle time required to produce a molded part can be as little as one minute (see Tables 7-22 and 7-23). The advantages of RRIM are similar to those listed for RTM. However, RRIM uses preforms that are less complex in construction and lower in reinforcement content than those used in RTM. The RRIM resin systems currently available will build up viscosity rapidly, resulting in a higher average viscosity during mold filling. This action follows the initial filling with a low-viscosity resin.

Table 7-22. Property Comparison and Design Guidelines of Resin-Transfer Molding versus Spray-up, Hand Lay-up, Mat/Preform, and SMC Molding Design Parameter Minimum inside radius, in. (mm) Molded-in holes In-mold trimming Core pull and slides Undercuts Minimum recommended draft (deg.) Minimum practical thickness, in. (mm) Maximum practical thickness, in. (mm) Normal thickness variation, in. (mm) Maximum thickness buildup, heavy buildup (ratio) Corrugated sections Metal inserts Bosses Ribs Hat section Raised numbers Finished surfaces

Resin-Transfer Molding

Spray-up

Hand Lay-up

Mat/Preform

'/4 (6.35)

'/4 (6.35)

'/4 (6.35)

'/4 (6.35)

No No Difficult Difficult 2 to 3

Large No Difficult Difficult 0

Large No Difficult Difficult 0

0.080 (2.0) 0.500 (12.7) ±0.01O ( ±0.25)

0.060 No limit

0.060 (1.5) No limit

±0.020 ( ±0.50)

±0.020 ( ±0.50)

2 : I Yes Yes Difficult Difficult Yes Yes 2

Any Yes Yes Yes No Yes Yes I

Any Yes Yes Yes No Yes Yes I

(1.5)

Sheet Molding Compound

'/'6 (1.59) Yes Yes Yes Yes No Yes No Yes '/4 to 6-in. depth 1 to 3; above 6-in. depth 3, or as required 0.030 0.050 (0.76) (1.3) 0.500 1 (12.7) (25.4) ±0.008 ±0.OO5 ( ±0.2) ( ±O.l)

2: I Yes Yes Difficult Yes Difficult Yes 2

Any Yes Yes Yes Yes No Yes 2

THE PROCESSING OF PLASTICS 683

Table 7-23. Cost Comparison of Resin-Transfer Molding versus SMC Compression and Injection Molding Process RTM Process operation: Production requirement, annual units per press Capital investment Labor cost Skill requirements Finishing Product: Complexity Size Tolerance Surface appearance Voids/wrinkles Reproducibility Cores/inserts Material usage: Raw material, cost Handling/applying Waste Scrap Reinforcement flexibility Mold: Initial cost Cycle life Preparation Maintenance

SMC Compression

Injection

5,000-10,000

50,000

50,000

Moderate High Considerable Trim flash, etc.

High Moderate Very low Very little

High Moderate Lowest Very little

Very complex Very large parts Good Gel coated Occasional Skill-dependent Possible

Moderate Big flat parts Very good Very good Rarely Very good Very difficult

Greatest Moderate Very good Very good Least Excellent Possible

Lowest Skill dependent Up to 3 percent Skill dependent Yes

Highest Easy Very low Cuts reusable No

High Automatic Sprues, runners Low No

Moderate 3,000-4,000 parts In factory In factory

Very high Very high Years Years Special mold-making shops Special machine shops

Stamping In the stamping process, a reinforced thermoplastic sheet material is precut to the required sizes. The precut sheet is preheated in an oven, the heat depending on the TP used (such as PP or nylon, where the heat can range upward from 520°F or 600°F). Dielectric heat is used to ensure that the heat is quick and, most important, to provide uniform heating through the thickness and across the sheet. After heating, the sheet is quickly formed into the desired shape in cooler matched-metal dies, using conventional stamping presses or SMC-type compression presses. Stamping is a highly productive process capable of forming complex shapes with the retention of the fiber orientation in particular locations as required. The process can be adapted to a wide variety of configurations, from small components to large box-shaped housings and from flat panels to thick, heavily ribbed plfrts (see Figs. 7-87 and 7-88).

Pultrusion In contrast to extrusion, in pultrusion a combination of liquid resin and continuous fibers is pulled continuously through a heated die of the shape required for continuous profiles. Shapes include structural i-beams, L-channels, tubes, angles, rods, sheets, and so on.

684 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7-87. These roofing panels made from Azloy (GE Plastics' Azdel, a thennoplastic RP) were featured on the GE Living Environmental Concept House seen in Chapter 1. These panels were designed to provide the highest flame and fire protection listing given for roofing systems as well as to allow easy installation and excellent weatherability.

The resins most commonly used are polyesters with fillers. Other resins, such as epoxies and urethanes, are used where their properties are needed. Longitudinal fibers are generally continuous rovings. Glass-fiber material (mat or woven) is added for cross-ply properties.

Filament Winding In filament winding (FW), continuous filaments are wound onto a mandrel after passing through a resin bath, unless preimpregnated (prepreg) filaments or tapes are used, which eliminates the resin bath. The shape of the mandrel is the internal shape of the finished part. The configuration of the winding depends upon the relative speed of rotation of the mandrel and the rate of travel of the reinforcement-dispensing mechanism. The three most common types are helical winding, in which the filaments are at a significant angle with the axis of the mandrel (see Fig. 7-89); circumferential winding, in which the filaments are wound like thread on a spool; and polar winding, in which the filaments are nearly parallel to the axis of the mandrel, passing over its ends on each pass (see Fig. 7-90) [13, 40]. Different configurations can be employed on successive passes and the orientation of the filaments be tailored to the stresses set up in the part. For example, with pipe, continuous helical winding can be employed on a segmental mandrel, an extruded mandrel,

THE PROCESSING OF PLASTICS 685 1:00 - r - - - ; , - - - - - - - - - - - - r 300

12~

AZDEL stamr1ng

250

100

200

7:0

150

100

50 Part<; pt'l hour _ _--.I rlt 80° Job ~ 'ff Ie II'll CY L---,---r---..-.---r----.-,----,-----'

20 40 60

Parts

per shift

per year f hOLJsa n OS

80 100 120 1 40 160

Pa:-t r:yc:lp

t1l11f'

seconds

Figure 7-88. An example of TO stamping on GE's Azdel versus other processes in regard to production cycle time as opposed to production rates based on single-cavity molds.

or on release film placed on a stationary mandrel. Other filament winders include braiding machines, loop wrappers, small to large storage tank machines, rectangular box-frame machines, and many different special fiber-placement machines with several degrees of freedom for intricate shapes. As filaments are continuous and tightly packed, they permit a high filament-to-resin ratio. This capability often results in products having the highest strength-to-weight ratio obtainable in any structures (see Fig. 7-91). Even though most FW uses glass filaments, all types of filaments can be used. Precautions must be observed if superior properties are to be achieved. Glass fibers are strong, but as glass they are subject to a severe loss in strength with surface abrasion. They must be carefully handled and processed to avoid such deterioration. In a lay-up for FW, as well as others, plastic abrasion-resistant fibers or (usually) film can be included. This construction permits parts to operate in severe load environments, such as under vibration, twisting, and so on, and eliminates or at least significantly reduces glass-toglass abrasion where a high fiber-to-resin ratio exists. Other types of fibers should be studied to determine whether fiber damage can occur when the part is in service. Certain fibers, with or without resins, might be brittle, and other problems could develop. The designer of the part should have knowledge of the potential problems. If problems do develop, steps can be taken during processing to overcome them. If unwanted porosity occurs, for instance, liners (gel coatings, elastomeric materials, etc.) can be included during FW.

Injection Molding As explained in the previous major section on injection molding, most 1M parts are made from TP, and some of TP uses milled glass fibers to improve part performance. Other fibers have seen limited use to date. TS compounds usually include reinforcements.

686 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Cycle 5 To Close

,.. Trovel

- Center LIne

Ax,~

Dwell Area

Band Width Impregnation Both

Carnage

Figure 7-89. A schematic showing helical filament winding.

Void Content

Voids are generally the result of the entrapment of air during construction of a lay-up, particularly with the use of hand lay-up and low-pressure processing methods. It is possible to have void contents of 1 to 3 percent. Depending on the application, voids can cause a reduction in part performance, particularly in certain environments and after lengthy outdoor exposures. If voids are undesirable, procedures can be used to reduce or eliminate them, such as applying a vacuum during the process. Another preventive method is to squeeze out air during lay-up by a roller or a spatula. The following method can be used to estimate void content: Percent voids

=

100 - 100a

(~ + ~

+

~)

where

= b = c = a

specific gravity of product

= 2.55 gravity of cured resin = range

specific gravity of fiberglass specific

1.18 to 1.24

d = resin content, by weight

e

=

glass content, by weight

f

=

filler content, by weight

g

=

specific gravity of filler

This method is not exact, because the assumption is made that the resin system has the same density with reinforcement as it does in an unreinforced casting. The net result is a possible overstatement of the void content. In addition to air entrapment, entrapment of volatiles can occur with certain resins that release them during processing.

THE PROCESSING OF PLASTICS 687

OTHER PROCESSES Reaction Injection Molding The RIM process involves the high-pressure impingement mixing of two or more reactive liquid components and injection of the mixture into a closed mold at low pressures. Large and thick parts can be molded using fast cycles with relatively low-cost materials. Its low energy requirements with relatively low investment costs make RIM attractive [121,664-73]. Different materials can be used such as nylon, polyester (TS), and epoxy, but TS polyurethane (PUR) is predominantly used. Almost no other plastic has the range of properties of PUR-a modulus of elasticity in bending of 200 to 1,400 MPa (29,000203,000 psi) and heat resistance from 90 to over 200°C (122-392°P), the higher values are for chopped glass-tiber-reinforced RIM, or RRIM. Table 7-24 compares PUR-RIM and injection moldings of unrein forced and glass-tiberreinforced thermoplastics in the production of parts with large surface areas. RIM is also comparable to resin transfer molding (RTM) in regard to the process of TS resins and the molding of large surface areas in that both processes offer the ability to tailor the reinforcement to the application. RRIM generally delivers faster cycles than other processes but needs much more expensive high-pressure dispensing equipment to ha.ldle the fast-reacting resin systems.

Liquid Injection Molding LIM has been in use longer than RIM, but the two processes are practically similar. The advantages it offers in the automated low-pressure processing of (usually) thermoset

Table 7-24. Comparison of RIM and Injection Molding of Unreinforced and Reinforced Plastics in the Production of Parts with Large Surface Areas

Plastic temperature, °C (OF) Plastic viscosity, Pa . S Injection pressure, bar (psi) Injection time, s Mold cavity pressure, bar (psi) Gates Clamping force, t Mold temperature, °C (OF) Time in mold, s Annealing Wall/thickness ratio Part thickness, typical maximum, cm (in.) Shrinkage, % Unreinforced Reinforce-glass parallel to fiber vertical to fiber Inserts Sink marks around metal inserts Mold prototype, months Mold alterations

PUR-RIM

Injection Molding

40-60 (50-140) 0.5-1.5 100-200 (1450-2900) 0.5-1.5 10-30 (l4~30) 1 80-400 50-70 (122-158) 20-30 30 min. @ 120°C (248°F) 110.8 10 (3.9)

200-300 (392-572) 100-1,000 700-800 (10,100-11,600) 5-8 300-700 (4400-10,200) 2-10 2500-10,000 50-80 (122-176) 30-80 Rarely 110.3 1 (0.4)

1.30-1.60

0.75-2.00

0.25 1.20 Easy Practically none 3-5 (epoxy) Cost-effective

0.20 0.40 Costly Distinct 9-12 (steel) Costly

TYPE Of WIIiOING

CONSIOCRATIONS

Hoop or CIf(umlertntl.1

ttl " 1n1.f'i1: .. If Cot1'l~~le CO. ""t:: "11 m.lncSftl f '" pus ot ClI1h'tf' R 's.alol UUj" e c'"' be mldt at ,any tlmoe ill 1 I "Khnl £N1,,.,n

H. III With W,de Ribbon

Sunplf f'qu',,"" nl iIII tlh pfO .lill~ tOf 'iI'Io ,de Slr"drDn af XCu' It 1.11.0, 01 UffUl1'I lD i11 ~1 !i~\ P ~'ul """,h,ne Ina

Nn,

m

,"poOh 01 Mi•• 'filLIII~ tot Il, ...

m~nd l l!J.

Heh. W,lh Narrow RIbbon I nd Mod lum 0' Hllh A "~I.

Muthpft! P'''~ " Cllfllrt nfCtW' l 10 coYfl Nndlei P'Ol'.I'!1mtj c.n\h,p bttlll "'ft UltI,"C on 1!1d tNtWlll:' rO

,tt.'

lI'%flW H.h, wllft Low Wlndln, Anile

~

ftt..lfl A .. rn

1.. 1

to' C.Uf!

~ muU

limed "f((IWly .,n nyrK!,ti lC:uhon

0,..,,11 .)1 c.:kh cOO of (,IH" f \1,.0 .. 1: m.aw M n u\J1y to,a"I(II,. ~t'OA l1WJ PI! rtt silpp.1 r

'.tt"

Flbtrs posthond ,'ou

end 01

to ~It~ InJ k»op on If .. t'UJ CI' U UI . e F.tfu, lfnd to tlOUP Hom 'ItlbOct ,nlO,ope dtotfln l U'H.~ ,t.. ffUI t.'Jl'lIlltl ttnd

e d ttmfl ,. Ion

fl".,

tnd 0' Urn.~t s.lIo t . s.Qftd liP 01 m,a"d,f4 . 1 e.ch end 01 Ulr

0«11 110 I I ! Kft

fI,

1 'lro'-! 'i I f,hlw dnu ]b.lt 10

d rtl PClS'hO'l" n II~ tit II'It'<. In 'Jr '

pra

I, ,,, . llP)Qu

,,,Of ltuOUil'l

wtn..a11 J~ d ... toH\

JJ'IOfI.n

s.11C " n Ii M'Ces\.ll ~ If crou rf~" orl UUt 11 not u'\td ClotS Iftj on U" •• lt'

11 rtqUlltC I.... ~Ply

P'o "l"Iuned '01 lin

LanllludlOll

~

o.t (..tw

t I nd

rnou!'\t !'lUI

_

'ola, W,ap

f'S

UlI bf uWO to

'QtI

Of

IS

I'RJnd,et ,pefdi

wll~ 1 lI ~ e up

u P ~"

ItQUI,ed

Mand,eI must It '"'lll n mobOlllns dunnJ p,lU

{

f~1I!

".a1I"n .1" be lost PoIiU " rilp I"'tlC:h.I!'It elll be u~ 'Of n,mo .... ubtK;" ,.. 11n. tIIolrtl dfYtC'f

0'

10'" "",m'"I""

1111 bll~ ",I (.Jill. e ft , f' SolI lndle' S4'1ffd I,Ip t _h. n-... U tf tun I" I'nfCI tUCll j \111·111 UWllf4!! ..,. 01 Hp .Ib bon 11'1

Inl lll< btlo... boul

Zero

f.

lt d

Hllit

R I,UI(,II'II,,"'P tHt.-.Hn

c.

.POA, ,t'ie.'

""",t, Ihol

odlnl mac;r.l(le ,tQllll r d,

fNncllfl

dCM 10 :LuMlOtl Ihill. Ch~'actenst,n flu 'tIl1th ","c. flbboft ' f

IU lns '"

PIKne hllt,.1

Ita to 01 tlmil 1'1'1011,)1\ Ie ~rlUt. '~ I IJ~ mull be idlll\1 Ie In '.IUY S lI.n cr~l1 t ' Rdlll.ClnSh, ... ot U".. It 10 min

___

Ihtn 101,11 , ,UKt"

l. . .. !e

"Iff

d. ftll.

dose to SUPDOH '!I,1t ("r'" I'\Irtd.tC mouon Of f btll 'lIPdl , ,",p Fibers

~u

be tl IJ

-------------

WftrOUO l il er fOl rlbbon:.o

h nltqlut!d

10 Ple"hf rltlbon Inl,.lllt.

Pot, •• "

'hoff., ..

0 ~"jnt " lltt , .. ,n

"I hM '

t ·

,m de 'HI 101 til t.·'IpaH . 1nd 1"1 Hlllc.al fIlI(hlne 'I I ollh fuollammtd (IOU 'Hd 'I 1I 111 lIiI .nd pol... . 11m: mcnc

sIco.l,.

F ig . 11 ·51

Mec:hlnery reqUired tor diff,rlnt tvpe! of 'ilament Winding (Reprinted by permtuion hom Pt.UICI Dat", & Proc."mg Janulry

1965. CopV,.ghl 1965, L• • PUblishIng CorI>o","on, L.I)e'lvYltI.,

tiL)

Figure 7-90. Examples of different filament-winding patterns that can be used in designing to meet different perfonnance requirements.

688

Lone

r... lllul (0,'

f('t.I~ ,W,\ j.J I, .. ~ lo r I f!.~ .. 1 'lIIIlIlIIh"r -.:It,.~ 11,,1 ... ., ....:C' 11lOh(0I1 IS 1101

,,"dow1

r.o Ijrtlm("lh.o', qul.fiI OIht'.I11 JI

Simple Sph."c I I

'or b"IC

~M"''''fUIfI ... "e~'ltI\lt

l,nC' 'rquUC'lIlf.llt 'U

I ... ,=tr1U'I,R

51'" w,~. moi on 01 Uti Ilf" II fM,UUt"d f, r UUlJr.e 11011" no cron fHd AI oIn rn of lII.utd. (Ion rm" ntc:fUll)' be

(Jus. (lUI' ""lit; bKomn fltUSI'Wt PglJr .'ip nl,ch,nt '"",. be IUrd I' II" t CI"

S,n,,'al 10. .~'h.t.

~

I IS ,r'I(h"oIllon 1'5101'11;(' fr'lOU h.

Iplt 1iph'flcal .,ndll't

l.ttI UllI.lif Ot CIOU

h!

I, ..

True Spr'tf'flciJl

P.alh DI tlbt'r ... PICI,,,,"'" td II) , ~ ~ Uftl'OtFll .. rr Itll( ~u ,net '5'It'n(.t1'1. 10 ~II It t, 01"1 ttlht f @'

M,sc..'laneaus

'Of wccn:s.lul Ml"'f'nl "'''~If'l

II '"1.1\1 ttl!!'

!o hInd .... nd liolll\ no 1t1t .it,S D"~f'll 01 'Ibfn. Oft ",.,l\drl'l ~rtMI'. fO~~blf

SIIf'(IIII'Tt "4nf: bt" tPPloxh Other.11@' (,0IT'01'1 I'rOlumm"1 !;If ,II tI'IotlOllI ot "fllCII mxhJnt!! IrqlJlltdi.

MJ(hlnl! to IfPloducl! tT'IOltOl'lt af Ptill'td

'IUndln ProCt,mnw-d ..~ts 'lUI)' bt .e(lul,HI.

mohons In Ifu"t

Figure 7-90. (Continued)

689

690 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7-91. A precision filament-wound solid-fuel rocket motor case.

resins-fast cycles, low labor cost, low capital investment, energy saving, and space saving-may make LIM competitive to potting, encapsulating, compression transfer, and injection molding, particularly when insert molding is required.

Foaming Many different foamed plastic products are produced, and practically all the processes reviewed in this book can be used to make them, particularly extrusion, calendering, casting, RIM, and injection molding. Almost all plastics can be used to make these cellular-core structures, which range from flexible to very rigid objects. Basically, the resin is mixed with a blowing agent, which can be a solid, liquid, or gaseous substance that imparts the cellular character to the product. Blowing agents are classified as either physical or chemical [10].

Expandable Polystyrene EPS molding illustrates the use of blowing agents. Resin beads containing a blowing agent are supplied to the molder in solid form. Each about 0.1 to 0.3 mm in diameter, these beads or spheres contain a small amount of a hydrocarbon liquid, usually pentane, that is used as the blowing agent. The process involves two major steps. The first consists of a preexpansion of the virgin beads by heat (steam, hot air, radiant heat, or hot water). Steam is the most-used medium, as it is the most practical and most economical.

THE PROCESSING OF PLASTICS 691

The next step conveys these beads, usually through a transport tube by air, to the mold cavity(s). The final expansion occurs in the mold, usually with steam heat, either by having live steam go through perforations in the mold itself or by means of steam probes that are withdrawn as the beads are expanding. During expansion the beads melt together, adhering to each other and forming a relatively smooth skin, filling the cavity or cavities. With small parts, multiple cavities can be used. After the heat cycles the cooling cycle starts. Because the EPS is an excellent thermal insulator, it takes a relatively long time to remove its heat prior to demolding, or the part will distort. Cooling is usually done by directing a water spray on the mold. To facilitate removal, particularly for complex shapes, mold-release agents are used. An outstanding property of EPS is its extremely low density (when compared to other processes), which-by alteration of the preforming treatment-can be varied according to the end use. Other types of plastics are employed to produce expandable plastic foam (EPF), including PE, PP, PMMA, and ethylene-styrene copolymers. They can use the same equipment, with only slight modifications. These plastics have different properties from those of PS and open up new markets; they provide improved sound insulation, resistance to additional heat deformation, better recovery of shapes in moldings, and so on.

Rotational Molding Rotational molding is a simple, basic, four-step process that uses a thin-walled mold with good heat-transfer characteristics. This closed mold requires an entrance for insertion of plastic and, most important, the capability to be "opened" so that cured parts can be removed. These requirements are no problem. Liquid or dry-powder plastic equal to the weight of the final part is put into the mold, which rotates simultaneously about two axes located perpendicular to each other. With slow rotation about each axis, the material inside the mold tumbles to the bottom, creating a continuous path that covers all mold surfaces equally. The next step involves heating the mold while it is rotating. Molds can be heated by a heated oven, a direct flame, a heat-transfer liquid (either in a jacket around the mold or sprayed over the mold), or electric-resistance heaters placed around the mold. With uniform heat transfer through the mold, the resin melts to build up a layer of molten plastic on the mold's inside surface. After the required heat-time cycle is completed, the mold is ready for cooling, which is accomplished with the mold rotating continually. Cooling is usually done by air from a high-velocity fan or by a fine water spray over the mold. After cooling, the final step is to remove the solid hallow part and reload the mold with plastic [10]. This process is capable of molding small to large hollow items with uniform wall thicknesses, using certain plastics. Its production rates, compared to those of other processes, can be low. However, the total cost of equipment and the production time for moderate-sized and, especially, large parts are also low. Large parts range up to 85,200 I (22,000 gal.) in size, with a wall thickness of 3.8 cm (1.5 in.). One tank used 2.4 t (5,300 lb.) of XHDPE; the first charge was about 1.5 t (3,300 lb.), followed by .45 t (1,000 lb.) and finally another .45 t (1,000 lb.). Molds can be of any shape and can include corrugated or rib constructions to increase their stability and stiffness (if not impossible, large, flat walls can be difficult). The thickness of their walls is limited to allow heat penetration. Figure 7-92 is an example of a mold shape with its rotating mechanism.

692 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 7-92. A mold and its rotating mechanism, as used in rotational molding.

Powder Coating Powder coating is a solventless coating-a coating that is not dependent upon a sacrificial medium such as a solvent, but is based on the performance constituents of solid TP or TS resins. It can be a homogeneous blend of the plastic with fillers and additives in the form of a dry, fine-particle-size compound similar to flour. The three basic methods are the fluidized bed, electrostatic spray, and electrostatic fluidized bed processes. The advantages of the process include its minimizing air pollution as well as water contamination, and increased part performance when coated, resulting in cost savings. This is basically a chemical coating, so it has many of the same problems as solution painting. If not properly formulated, the coating may sag at high thicknesses, show poor performance when not completely cured, reveal imperfections such as craters and pinholes, and have poor hiding, with low film thickness.

Vinyl Dispersion Vinyl dispersions are fluid suspensions of special fine-particle-size polyvinyl chloride resins in plasticizing liquids. When the PVC is heated to about 148 to 180°C (300-355°F), fusion or mutual solubilization of the resin and the plasticizer takes place. The dispersion then turns into a homogeneous hot melt. When the melt is cooled below 50 to 60°C (122140°F), it becomes a tough vinyl product. With vinyl dispersions the processor can use convenient liquid-handling techniques such as spraying, pouring, spread coating, and dipping. This system permits products to be made that would otherwise require costly and heavy melt-processing equipment. The term plastisol is used to describe a vinyl dispersion that contains no volatile thinners or diluents. Plastisols often contain stabilizers, fillers, and pigments, along with the essential dispersion resin and the liquid plasticizer. All ingredients exhibit very low volatility under processing and use conditions. Plastisols can be made into thick fused sections with no concern for solvent or water blistering, as with solution or latex systems, so they are described as being 100 percent solids. It is convenient in some instances to extend the liquid phase of a dispersion with organic volatiles, which are removed during fusion. The term organosol applies to these dispersions.

THE PROCESSING OF PLASTICS 693

Casting

Sorne TPs and TSs begin as liquids that can be cast and polyrnerized into solids. In the process various ornamental or utilitarian objects can be ernbedded in the plastic. By definition, casting applies to the formation of an object by pouring a fluid rnonornerpolyrner solution into an open rnold where it cornpletes its polyrnerization. Casting can also lead to the formation of filrn or sheet, rnade by pouring the liquid resin onto a rnoving belt or by precipitation in a chernical bath. Casting differs frorn rnany of the other techniques described in this book in that it generally does not involve pressure or vacuurn casting, although certain rnaterials and cornplex parts rnay require one or the other. Calendering

The calendering process, used in the production of plastic filrns and sheets, converts plastic into a rnelt and then passes the pastelike rnass through nips of a series of heated and rotating speed-controlled rolls into webs of specific thickness and width. The web rnay be polished or ernbossed, either rigid or flexible [10]. Coining

This process, which has been used at least since the early 1940s, cornbines the best of injection rnolding and cornpression rnolding. The process is also called injection cornpression or injection starnping. Basically, it involves using an 1M rnachine to rnelt a plastic (unreinforced or reinforced TP or TS) and direct a fixed amount of the rnelt into a cornpression rnold. As shown in Figure 7-93, a cornpression rnold has a rnale plug that fits into a fernale rnold. When rnelt enters the rnold, it is not cornpletely closed. Thus, the rnelt literally flows unrestricted in the cavity and is stress-free. After injection is cornpleted, the rnold is closed, with the pressure on the rnelt rnade uniform. The result is practically a stressfree solid part held to tight tolerances. This process can result in a faster rnolding cycle than that of conventional cornpression or even injection rnolding. Its disadvantage is the extra cost for the closing rnechanisrn and its control, which becornes insignificant when all the performance requirements that are gained are evaluated. Fusible-Core Technique

The basic fusible-core technique, a take-off of cored rnetal casting, rnakes it possible to produce sirnple to very cornplex hollow structural products. It involves using a fusible core inside the plastic part or structure. This core permits forming the desired plastic shape. The core rnaterial is a type that will not collapse or change shape during a pressureternperature-tirne processing cycle. Shape is not usually the problern, since the core rnaterial is restricted. The core rnaterial to be used depends on the actual processing requirernents, particularly ternperature. It can range frorn a wax to different ratios of zinc-alurninurn eutectic rnixtures (alloys) to special fusible eutectic alloys. The core rnaterial has to rnelt below the rnelt ternperature of the plastic. These shaped cores are usually inserted in a rnold cavity where it is retained by the rnold (such as is used with a rnold core puller) or by "spiders" (as used in certain rnetal core supports for extrusion dies). After processing, the core rnaterial

694 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK InJ~ctlon CompreSSIon CO,"lnq

Prccornl"es"on

Figure 7-93. A schematic of the coining process, which combines injection molding and compression molding.

is removed by heating it to its melt temperature. Release is via an existing opening or a hole drilled through the plastic to the core. This technique is used in such different processes as injection molding, compression molding, and various RP methods. For example, during 1943-1944 the fusible-core method (at that time called the lost wax method) was used with hand lay-up of glassfiber-reinforced TS polyester to fabricate box beams. These hexagonal beams were molded with the RP material (in a resin impregnated or prepreg form) wrapped around an extruded rectangular solid wax. In tum these wrapped beams were placed next to each other. They could be shaped to meet required designs, such as an aircraft wing airfoil shape. This lay-up would be sandwiched between formed plates, usually of aluminum, and the complete fusible core-RP construction cured using external heat or by compounding a curing agent in the resin. After curing, the individual beams would be subjected to sufficient heat to melt and remove the wax from the RP. These RP boxes would be used as the core for a sandwich construction utilizing RP for the sandwich skins (see Chapter 5 on sandwich structures). Some of the original development in fabricating the original allplastic aircraft design concept used these fusible-core sandwich structures. The fusible-core technique provides the designer with a different processing method for each different shape. Some examples of recent products produced with it include the tennis racket and automobile engine intake manifold. These designs permitted injection molding in bent or curved shapes.

THE PROCESSING OF PLASTICS 695

SELECTING PROCESSING The processing infonnation presented in this and other chapters has provided a variety of useful selection guides, particularly Tables 7-4-6 and 7-19-23 and Figures 7-2, 7-3, and 7-88. The process selection guides in Tables 7-25 to 7-32 and Figures 7-94 to 7-97 provide additional different aspects to consider. For any given part, the most important processing requirements should be detennined based on the plastic to be processed, the quantity, and the dimensions of size and the tolerances. Process selection is a critical step in product design. Failure to select a viable process during the initial design stages can dramatically increase development costs and timing. It is important to recognize that the process can have a significant effect on the perfonnance of the finished part. The following examples of the considerations in choosing a process are based on what has been reviewed throughout this book. 1. The nature of the process may have a profound influence on a product's mechanical strength. 2. Excessive heat during processing can consume sacrificial heat stabilizers for certain plastics, rendering stabilization levels insufficient to ensure long life at elevated temperatures. Thennal degradation usually results in embrittlement (tests can be conducted to detennine the remaining levels, as in Chapter 9). 3. The slow cooling of crystalline polymers, such as HDPE and PP, can allow large crystal fonnations to develop. Such crystals embrittle the resin and make it prone to stress cracking. 4. The rapid cooling of certain plastic parts can result in "frozen in" stresses and strains (particularly with injection molding). The stresses may decay with time, in a viscoelastic manner. However, they will act like any other sustained stress to aggravate cracking or crazing in the presence of aggressive media and hostile environments like UV radiation. 5. Annealing at temperatures below the Tg (glass transition temperature, see Chapter 2) where material becomes leathery is not necessarily beneficial. For example, annealing a PC greatly accelerates both its crazing and rupture under sustained loading. In general, the annealing of plastics results in lowering its properties; however, its dimensional stability may be improved. Heating a material to above its Tg , however, results in the relief of internal stresses. 6. Knit or weld lines fonn where the melt flow during processing meets after flowing through separate gates in an injection mold or after being parted by either "spiders" in an extruder die or bosses in an injection mold. Because the material is not well mixed in the zone of the knit or weld line, the seam thus fonned can be weak or brittle under long-tenn or impact loads. This problem can easily arise with fiberreinforced plastics, where 40 to 60 percent of their strength can be lost, since fibers fail to knit together at their seams. 7. In RPs, insufficient compaction and consolidation of a composite before resin cure will result in air pockets, incomplete wet-out and encapsulation of the fibers, and/or insufficient fiber or unifonn fiber content. These deficiencies lead to loss of strength and stiffness and susceptibility to deterioration by water and aggressive agents.

a. ~ a.

0.001

1-2 0.01

rough).

0.01-0.125 (0.25-3.18) Yes Yes' Yes Yes Yes Yes Yes Yes

0.01 (0.25) Yes Yes Yes Yes Yes Yes Yes Yes

=

0-1

0.01-0.125 (0.25-3.18)

Material None

configurations

Simple

Casting

0

Material >0.25 (6.4) 0.125 (3.18)

bodies

Hollow

tSpeciai mold required. 2Not recommended. 30nly flexible material. 400ly direction of extrusion. ~Possible with special techniques. 6Fusing premix/yes. 'Rated I to 5 (1 = very smooth. 5

Minimum draft (deg.) Minimum thickness, in. (mm) Threads Undercuts Inserts Built-in cores Molded-in holes Bosses Fins or ribs Molded in designs and nos. Surface finish' Overall dimensional tolerance (in.! in., plus or minus)

(mm)

radius, io.

Limiting size factor Maximum thickness, in. (mm) Minimum inside

Major shape characteristics

Part Design

Blow Molding

1-2 0.001

0.01-0.125 (0.25-3.18) Yes NR' Yes No Yes Yes Yes Yes

>1

plane Equipment 0.5 (12.7) 0.125 (3.18)

in one

Moldable

Compression

1-2 0.005

0.001 (0.02) No Yes Yes Yes Yes 4 Yes Yes No

NR'

(150) 0.01-0.125 (0.25-3.18)

Material

profile

cross section

Constant

Extrusion

0.005

0.015 (0.38) No NR' Yes Yes Yes No No' No

2-3

with surfaces of revolution Equipment 3 (76) 0.125 (3.18)

Structure

Filament Winding

Process

0.001

0.005 (0.1) Yes Yes' Yes Yes Yes Yes Yes Yes


Equipment 6 (150) 0.01-0.125 (0.25-3.18)

limitations

Pew

Injection

4-5 0.005

0.03 (0.8) No NR' Yes Yes Yes Yes No· Yes

Equipment 2 (51) 0.06 (1.5)

plane

in one

Moldable

Matched Die Molding

2-3 0.01

0.02 (0.5) Yes Yes 3 Yes Yes Yes Yes Yes Yes

Material 0.5 (12.7) 0.01-0.125 (0.25-3.18)

bodies

Hollow

Rotational

Table 7-25. Basic Processing Methods as a Function of Part Design

1-3 0.01

0.002 (0.05) No Yes l NR' Yes No Yes Yes Yes

(76) 0.125 (3.18)

Material

Moldable in one plane

fonning

Thermo-

1-2 0.001

0.01-0.125 (0.25-3.18) Yes NR' Yes Yes Yes Yes Yes Yes

Equipment 6 (150) 0.01-0.125 (0.25-3.18)

Simple configurations

Compression

Transfer

4-5 0.02

0.06 ( 1.5) No Yes Yes Yes Yes Yes Yes Yes

0

plane Mold Size 0.5 (12.7) 0.25 (6.4)

in one

Moldable

Wet lay-up (Contact Molding)

\Q

~

'I

UNDER 250°F THERMOPLASTICS

LARGE AREA

THERMOFORM FOAM HEAT SEAL WELD ROTOFORM BLOW MOLD ADHESIVE BOND STRUCTURAL FOAM RIM

I

I I

I EXTRUDE

I

LONG LENGTHS

Figure 7-94. Basic guide to process selection.

LOW-PRESSURE LAMINATION FILAMENT WINDING COMPRESSION HIGH-PRESSURE LAMINATION POST FORM ADHESIVE BOND MACHINE PULTRUSION

OVER 250°F THERMOSETS

I

I

I

COMPRESSION TRANSFER INJECTION LAMINATION PULTRUSION

I

HIGH-VOLUME

I

CASTING MACHINING LOW PRESSURE LAY-UP POST FORM SPRAY-UP RESIN TRANSFE R

LOW-VOLUME

I

I MACHINE THERMOFORM COMPRESSION CASTING ROTOFORM FOAM ADHESIVE BOND

I

LOW-VOLUME

INJECTION BLOW MOLD THERMOFORM EXTRUSION ROTOFORM RIM

HIGH-VOLUME

I I

I

I LESS THAN 250°F THERMOPLASTICS

OVER 250°F THERMOSETS

I

I

I SMALL PART LESS THAN 1 III ft LESS THAN Sib

I

I

LARGE PART OVER 1 111ft OVERS . .

I

PART TO BE FORMED

~ ~

=

L';"

VACUUM

I

I SPRAY UP

BAG

I

I

L Sl7 LE I

I

INJECTION

CONTINUOUS LAMINATION

I

MATCHED TOOLING

AUTOCLAVE

BAG

PRESSURE

MAT PREFORMS

I

MATCHED TOOLING

I

INJECTION

PREMIX

PREFORM

I

MAT

PREFORM

I COM;LEXl I~

TOOL

MATCHED

COST PREDOMINANT

I

VOLUME

I

I

CONTINUOUS LAMINATION

I

PREFORMS

I

INJECTION

COMPRESSION

~-l

INJECTION

MATCHED TOOL

I

FABRIC PREPREGS

SMCMAT PREFORM

c$S-~$

(UNDER 3)1 5 FT)

SMALL PART

FABRIC· PREPREGS

ALL ~~ATCHED TOOL COMPRESSION MOLDING

COMPRESSION

I

I

FASRIC PREP REGS

MAT

I

HIGH

RESIN TRANSFER

PReFORM

Si.1cr.1AT-

SHAPE

-BMC. PREMIX PREFORMS ANO PREPREGS

J

I

BMC&PREMIX·

STRENGTH PREDOMINANT

I

Figure 7-95. Overall guide to reinforced TS plastics' process selection.

I

MATCHE D TOOL

SPRAV uP

FILAMENT WINDING

I

WINDIHG

AUTOCLAVE

CONTACT

CONTACT

I ~';"';LEX 1 I

I

'-4ATCHEO TOOL

AUTOCLAVE

PRESSURE BAG

I

VOLUME

RESIN TRANSFER

I

BAG

VACUUM

PUL TRUSION

I

CONTACT

I

I ~--$

COST PREDOMINANT

SPRAY UP

PRESSURE BAG

1

SHAPE

PUl TRUSION

I

PRESSURE

RESIN TRANSFER

BAG

AUTOCLAVE

CONTACT

I

FILAMENT WINDING

VACUUM BAG

FIL4MENT

s17~

PUL TRUSION

I

SPRAY UP

$$-C$=J $

I

STRENGTH PREDOMINANT

{OVER 3)1 5 FTt

.---

LARGE PART

REINfORCED PLASTICS PART TO BE PROCES5EO tTHERMOSeTl

THE PROCESSING OF PLASTICS 699

These examples show the kinds of alterations that the processing of plastics can have on the performance of the product. As discussed throughout this book, the many different plastics tend all to behave in different patterns, so where a particular problem could develop with one material, it might have little or no effect on another, even if the base resins are the same but contain different additives or reinforcements. Regardless, the problems that might arise should be eliminated at the outset. In some cases the designer will not have the ability to choose freely from all the design, material, and process alternatives. For example, a design is often heavily constrained by the need to fit an existing assembly, and the material and process may be determined largely by the need to use existing facilities. However, to optimize results the designer should establish the extent of any design freedom early in the design process and explore the design, material, and process alternatives within these bounds. Before final selection of the process, the entire process of production should be considered, including such secondary operations as painting and decorating. Chapter 8 provides more information on both the upstream and downstream auxiliary equipment that may have to be considered in order to be efficient in regard to product performance and cost.

Shape Both shape and design details are heavily process related. The ability to mold ribs, for example, may depend on material flow during a process or on the flowability of a resin reinforced with glass. The ability to produce hollow shapes depends on the ability to use removable cores, including air, fusible or soluble solids, and even sand. Hollow shapes can also be produced using cores that remain in the part, such as foam inserts in RTM or metal inserts in 1M. The geometric symmetry of a part can also influence process selection. For example, an axis of symmetry in a long, narrow part may suggest selecting an extrusion or pultrusion process. Similarly, the need for hollow sections in the part could suggest blow molding or rotational molding. In order to handle materials that melt, flow, and solidify quickly, it is necessary to use a mechanical process such as injection molding, which as a process could still be limited by the time available with the particular machine in question to fill the mold cavity before the melt solidifies; thus, high pressures are used to increase the speed of mold filling [11]. Each process has certain characteristics that can be summarized by determining whether 1) its ribs and bosses are feasible, depending on whether one or both sides of the part reproduce the tool (mold) surface; 2) the sequence of material injection or some other process and tool closure allows of having deep vertical sections in the surface wall; 3) the material's viscosity is high enough to allow the use of slides and cores in the tool without their being gummed up with material flowing into the slide mechanism; 4) hollow sections or containers are feasible; and, finally, 5) whether hollow or foam-filled box sections can be produced to increase section stiffness.

Size Part size is limited by a process's pressure and the available equipment, whereas the ability to achieve specific shape and design detail is dependent on the way the process operates. Generally, the lower the processing pressure, the larger the part that can be produced. Other restrictions are the size of the equipment that is available, the length of flow, and the material's reaction time. With most labor-intensive methods, such as hand

700 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Unidirectional Fiber Orientation

REINFORCEMENT TYPES: F.lamenls. Rovings PROCESSES: Pullrus.on. F.lament W.nd.ng. RTM. RIM

Bidirectional Fiber Orientation

REINFORCEMENT TYPES: F.laments. Rovings. Woven FabriCS. Braid.ng PROCESSES: F.lament Winding. Hand Layup. Compress.on Mold.ng. Inleet.or. Molding. Vacuum Bag. Stamp.ng. Co.ning. Pultrusion

Multidirectional Fiber Orientation

REINFORCEMENT TYPES Chopped Strands. M.lled Fibers. Mats PROCESSES: Hand Layup. Compress.on. InJection. Spray· UP. Vacuum Bag. Autoclave. RTM, RIM, Rotahonal. Stamping. Autocalve. Co.n.ng

Figure 7-96. Guide to the directional properties of composites versus processes.

lay-up, slow-reacting TSs can be used and there is virtually no limit on size. With some processes, size is limited only by the size of the equipment that is either available or can be produced. A general guide to practical processing thickness limitations is (in in.): injection molding, 0.02 to 0.5; extrusion, 0.001 to 1.0; blow molding, 0.003 to 0.2; thermoforming, 0.002 to 1.0; compression molding, 0.05 to 4.0; and foam injection molding of 0.1 to 5.0. The functions and property characteristics of a part will be largely determined by the performance requirements and material selected for fabrication. The basic requirement of the process is its capability of handling a suitable material. For example, if a major function requirement is for resistance to creep under high loads, it is probable that a longfiber RP will be necessary, which would immediately eliminate such processes as blow molding and conventional injection molding.

Surface Finish Another consideration is the ability of a material to provide a surface that is compatible with the requirements of the application: a smooth finish, molded-in color, textured surface, or other surface. The compatibility of the major processes with in-mold coating and other insert-surfacing materials, and their compatibility with surface decoration secondary processes, could also be important. It should be recognized that surface finish can be more than just a cosmetic standardit also affects part quality, mold cost, and delivery time. The surface can be used not only to enhance clarity for the sake of appearance but to hide surface defects such as sink marks. The Society of Plastics Engineers/Society of Plastics Industries standards range from a No.1 mirror finish to a No.6 grit blast finish. A mold finish comparison kit consisting of six hardened tool steel pieces and associated molded pieces is available through SPE/SPI [12].

...;::

I

PLASTER

elECTRO FORM

I AL I

LOVOL

Figure 7-97. Guide to tooling selection .

FILLED EPOXY

SPRAYED METAL

I

I AL

HI

ELECTJOFORM

PLASTIC

AL

I

STEEL tMACH HOBBEDI

I

HI VOL

I

FILLED EPOXY

I

EPOXY· FIBERGLAS

I

I AL I PLASTER

LOVOl

MEHANITE

ELECTROFORM • BACK UP

I AL I BoCu

HI VOl

CASTAL

I I I

STEEL

FILLEDIRIM EPOXY

I

I

STEEL

I

STEEL ) CAST AL ( MACHINE H088ED ELECTRO MACH. AL FORM. BACK·UP KIRKSITE

I

HI VOL

LOW VOL

HI VOL

I WOOD I PLASTER

LO

ELECTROFORM

I

I CASTAL

HI VOl

I

I

I

I

PLASTIC

I

SILICONE

I

ELASTOMER

I

SPRAYED METAL

I

elECTRO FORM

I

STEEL

WOOD

I AL

PLASTIC

I

ELASTOMER

I

PLASTIC ITPOR TSI

DIP METAL

DIPPED METAL

I I

SPRAY METAL

ELECTROFORMED

L..--r------!ELASTOMER

SHEET METAL

I

LOVOL

~-ROTOFORM

702 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Various types of surface finishes are available for plastics and composite parts, such as smooth, textured, molded-in color, and in-mold coating. A textured part surface can be obtained through either a textured mold cavity or a postmold paint process. The former method is the most commonly used. A wide variety of texture designs is available. The surface smoothness, and to some degree the texture, of a plastic or composite is as dependent on the materials used in it as on the surface of the cavity. For example, certain textured surfaces can be obtained only if the proper steel or metal is used in the mold cavity. Also, with any mold the proper cavity steel based on the plastic processed will significantly reduce its wear and tear and extend its useful life [11, 12].

Cost The production flexibility of the fabrication process is often the single most important economic factor in a plastic part. The component's size, shape, complexity, and required production rate can be primary determinants. As an example, small numbers of large objects tend economically to favor casting, as well as the RP's hand lay-up or spray-up process, with a minimal tooling cost and maximum freedom for design changes. Many products favor injection and compression molding or long runs in extruders, with their automation capabilities to minimize labor costs. Often the shape dictates the process, such as centrifugal casting or filament winding being used for cylindrical parts, rotational molding for complex hollow shapes or extremely large parts, and pultrusion for constant cross-sections requiring extremely high strength and stiffness. These general examples should be considered broadly, since individual processes can all be designed for a specific product capability to meet performance requirements at the lowest cost. A further important point is that major costs can be incurred in the operations required after fabricating: trimming, finishing, joining, attaching hardware, and so on. Observing the following design practices will help reduce costs and improve processing and performance, whatever fabrication method is selected: 1) strive for the simplest shape and form; 2) use the shape of the product to provide stiffness, reducing its required number of stiffening ribs; 3) combine the parts into single moldings or extrusions as much as possible, to minimize assembly time and eliminate designing fasteners and so on; 4) use a uniform wall thickness wherever possible, and make changes in thickness gradually, to reduce stress concentrations; 5) use shape to satisfy functional needs like slots for hoisting, hand grips, and pouring; 6) provide the maximum radii that are consistent with the functional requirements; and 7) keep tolerances as liberal as possible, but once in production aim for tighter tolerances, to save plastic material and probably reduce production cost. Minimizing cost is generally an overriding goal in any application, whether a process is being selected for a new product application or opportunities are being evaluated for replacing existing materials. The major elements of cost include capital equipment, tooling, labor, and inefficiencies such as scrap, repairs, waste, and machine downtime [12]. Each element must be evaluated before determining the most cost-effective process from among the available alternatives. To a great extent, the selection of a process for a new plastic or composite product will be dictated, or at least restricted by, the limitations imposed by design and cost factors. These factors frequently coincide to indicate the preferred processes. Some plastics will have a limited scope of processes available, but others will be more flexible. And parts for limited production or small volumes may require processes with low capital and tooling costs to make them economical.

Table 7-26. Specific Processing Methods as a Function of Part Design Good

Process

Finish, Varying Vertical Spherical Box Slides/ Both CrossRibs Bones Walls Shape Sections Cores Weldable Sides section

Thermoplastics Injection Injection compression Hollow injection Foam injection Sandwich molding Compression Stamping Extrusion Blow molding Twin-sheet forming Twin-sheet stamping Thermoforming Filament winding Rotational casting

Y Y Y Y Y Y N

Y Y Y Y Y Y Y

N

Y

Y

Y N N Y N N

Y Y Y Y Y Y

N N Y N N N

Y Y Y Y Y Y N Y N N N N Y N

N N N N

Y Y N N

N N N N

Y Y Y Y

Y Y Y Y

N N N N

N N Y Y

N N N N

N N N N

Y Y N N

Y Y Y Y

N N

Y

N

Y

N

N

Y

N

Y Y Y

Y N

N N N N N

N Y

N

N Y

Y

N

Y

Y Y Y N Y

N

Y

N

Y

N

N

Y

Y

N Y Y

Y Y Y Y

N Y N Y N

Y Y Y

N N N N N

N N N

Y Y Y

N N

Y

Y Y Y Y Y

Y Y N Y Y Y

N N N N N N N N Y Y N N Y Y

N N Y Y N N N Y Y Y Y N Y N

Y Y Y Y

Y Y Y Y

N N N N

Y N N N

Y N Y Y

Y Y Y Y

Y Y

Y

Y Y Y

Y Y Y

N Y

N Y

Y

Y Y Y Y Y

Y Y Y Y Y Y

Y N Y Y Y Y N

Y Y Y Y Y Y N Y N N N N Y N

N N N N N N N

N/A

Y Y N N

N

Y Y Y Y Y Y Y

Thermosetting Compression Powder Sheet molding compound Cold-press molding Hot-press molding High-strength sheet molding compound Prepreg Vacuum bag Hand lay-up Injection Powder Bulk molding compound

ZMC Stamping Reaction injection molding Resin transfer molding, or resinject High-speed resin transfer molding, or fast resinject Foam polyurethane Reinforced foam Filament winding Pultrusion

N N

N/A

Y Y

Y

Note: Y, yes; N, no; N/A, not applicable.

703

.....

~

e

X

X

X X X X X X X X X

X X X X X X X X X

X X

X X X

X X

X

X

X

X

X

X

X X X

X

X

X X X X X X

X

X

X X

X X

X

X X X

X

X

X X

X X

X

X

X X

X X

X

X X X

X

X X X

X

X X

Injec- Compres- TransCold Casting Molding Coating tion sion fer

·Compounding pennits using other processes.

Cellulosic Epoxy F1uoroplastic Melamine-formaldehyde Nylon Phenol-formaldehyde Poly (amide-imide) Polyarylether Polybutadiene Polycarbonate Polyester (TP) Polyester-fiberglass (TS) Polyethylene Polyimide Polyphenylene oxide Polyphenylene sulfide Polypropylene Polystyrene Polysulfone Polyurethane (TS) (TP) SAN Silicone Styrene butadiene Urea formaldehyde Vinyl

ASA

ABS Acetal Acrylic Allyl

Material Family

X

X X

X X

X

X

X

X

X X

X X

Structural Foam

X

X X X X(TP) X X X

X X(TP) X

X X X X X X

X

X

X X

X X X

Extrusion

X X X

X X

X

X

X

X

X X

X X

X

X X X X X

X

X

X

X X

X X X

X

X

X

X

X

X

X

X X

X

X X X

X

X

X X X

RP Dip LamiSheet Molding Fila- and nating Forming FRP ment Slush Blow

Table 7-27. Molding Process Guide to Plastic Materials*

X

X X

X X

X

X

X X X

X

X

X

X

Rotational

Table 7-28. General Information Relating Processes and Materials to Properties of Plastics Thermosets

Properties

Polyesters Properties shown also apply to some polyesters formulated for thermoplastic processing by injection molding

Simplest, most versatile, economical and most widely used family of resins, having good electrical properties, good chemical resistance, especially to acids

Epoxies

Excellent mechanical properties, dimensional stability, chemical resistance (especially alkalis), low water absorption, selfextinguishing (when halogenated), low shrinkage, good abrasion resistance, very good adhesion properties Good acid resistance, good electrical properties (except arc resistance), high heat resistance Highest heat resistance, low water absorption, excellent dielectric properties, high arc resistance Good heat resistance, high impact strength Good electrical insulation, low water absorption

Phenolics Silicones

Melamines Diallyl phthalate

Processes Compression molding Filament winding Hand lay-up Mat molding Pressure bag molding Continuous pultrusion Injection molding Spray-up Centrifugal casting Cold molding Comoform l Encapsulation Compression molding Filament winding Hand lay-up Continuous pultrusion Encapsulation Centrifugal casting Compression molding Continuous laminating Compression molding Injection molding Encapsulation Compression molding Compression molding

Thermoplastics Polystyrene

Nylon Polycarbonate Styrene-acrylo-nitrile Acrylics

Vinyls

Acetals

Polyethylene

Low cost, moderate heat distortion, good dimensional stability, good stiffness, impact strength High heat distortion, low water absorption, low elongation, good impact strength, good tensile and flexural strength Self-extinguishing, high dielectric strength, high mechanical properties Good solvent resistance, good long-term strength, good appearance Good gloss, weather resistance, optical clarity, and color; excellent electrical properties

Excellent weatherability, superior electrical properties, excellent moisture and chemical resistance , self-extinguishing Very high tensile strength and stiffness, exceptional dimensional stability, high chemical and abrasion resistance, no known room temperature solvent Good toughness, light weight, low cost, good flexibility, good chemical resistance; can be "welded"

Injection molding Continuous laminating Injection molding Blow molding, Rotational molding Injection molding Injection molding Injection molding Vacuum forming Compression molding Continuous laminating Injection molding Continuous laminating Rotational molding Injection molding

Injection molding Rotational molding Blow molding (conr'd)

705

Table 7-28. (Continued) Properties

Processes

Very high heat and chemical resistance, nonbuming, lowest coefficient of friction, high dimensional stability Very tough engineering plastic, superior dimensional stability, low moisture absorption, excellent chemical resistance Excellent resistance to stress or flex cracking, very light weight, hard, scratch-resistant surface, can be electroplated; good chemical and heat resistance; exceptional impact strength; good optical qualities Good transparency, high mechanical properties, heat resistance, electrical properties at high temperatures; can be electroplated

Injection molding Encapsulation Continuous pultrusion Injection molding

Thermosets Fluorocarbons

Polyphenylene oxide modified Polypropylene

Polysulfone

Injection molding Continuous laminating Rotational molding

Injection molding

'Comoform is an extension of the cold molding process which utilizes a thermoformed plastic skin to impart excellent surface to a cold molded laminate.

Table 7-29. Economic Comparison of Three Different Processes Production Considerations Typical minimum number of parts a vendor is likely to quote on for a single setup

Relative tooling cost, single cavity

Average cycle times for consistent part reproduction Is a multiple-cavity tooling approach possible to reduce piece costs?

Are secondary operations required except to remove sprue?

Structural Foam 250 (using multiple nozzle equip. with tools from other sources designed for the same polymer and ganged on the platen) Lowest. Machined aluminum may be viable, depending on quantity required 2 to 3 minutes el4 in. nominal wall thickness) Yes

No

Sheet Molding Compound

Injection Molding 1,000 to 1,500

500

20 percent more. Hardened-steel tooling

20 percent to 25 percent more. Compressionmolding steel tools 11/2 to 3 minutes

40 to 50 seconds

Yes. Depends on size and configuration, although rapid cycle time may eliminate the need. No

Not necessarily. Secondary operations may be too costly and material flow too difficult Yes, e.g., removing material where a "window" is required (often done within the molding cycle) (cont'd)

706

THE PROCESSING OF PLASTICS 707

Table 7-29. (Continued) Production Considerations

Structural Foam

Sheet Molding Compound

Injection Molding

Range of materials that can be molded

Similar to thermoplastic injection molding

Finishing costs for good cosmetic appearance

40 to 60 cents per sq. ft. of surface (depending on surface-swirl conditions)

Limited; higher cost

Unlimited; cost depends on performance requirements None, if integrally colored; 10-20 cents per sq. ft. if painted

None, if secondary operations such as trimming are not required. Otherwise 20 to 30 cents per sq. ft. of surface

In summary, when considering alternative processes for producing plastic and composite products, the major concerns usually involve 1) limitations that may be imposed by the material, because not all materials can be processed by all methods; 2) limitations imposed by the design, such as the size, single-piece versus mUltiple-piece construction, a closed or open shape, and the level of dimensional and tolerance accuracy required; 3) the number of products required; and 4) the available capital equipment. Certain equipment may already be available and in use, although it may not necessarily be the one needed for the lowest production cost.

Table 7-30. Guide to Compatibility of Processes and Materials Thermoplastics

Thermosets

tl~

;f Injection molding Hand lay-up Spray-up Compression molding Preform molding Filament winding Pultrusion Resin transfer molding Reinforced reaction injection molding

~ i:Q~

~

C'-l

B tl., '"

i

i

~

i

1

"3 8 If -<

'"] ~

~

~

~

~ .; If

J

i

!I If

~0 j

8

~

If

~

C'-l i:Q

-<

~ If

!

~

If

I-<

~

'"

.;

If... B ~

;f

tl

If

Table 7-31. Design Recommendations for Choosing an RP Process Contact Molding, Spray-up Minimum inside radius, in. Molded-in holes Trimmed-in mold Built-in cores Undercuts Minimum practical thickness, in. (mm) Maximum practical thickness, in. (mm) Normal thickness variation, in. ' (mm) Maximum buildup of thickness Corrugated sections Metal inserts Surfacing mat Limiting size factor Metal edge stiffeners Bosses Fins Molded-in labels Raised numbers Gel coat surface Shape limitations Translucency Finished surfaces Strength orientation Typical glass percent by weight 'Note: N.A.: NOI applicable. N.R.: Not recommended.

708

Matched Die Premix! Molding Molding with Preform Compound or Mat

Pressure Bag

Filament Winding

Continuous Pultrusion

%

'Is

N.A.*

1/32

I/S

Large No Yes Yes

Large No Yes Yes

N.R.* Yes Yes No

N.A. Yes N.A. No

Yes Yes Yes Yes

Yes Yes Yes No

0.060

0.060 (1.5)

0.010 (0.25)

0.037 (0.94)

0.060

(1.5)

(1.5)

0.030 (0.76)

0.50 (13)

(25.4)

3 (76.2)

I (25.4)

I (25.4)

0.25 (6.4)

±0.020 (±0.5l)

±0.020 ( ±0.5l)

±0.01O (±0.25)

±0.OO5 (±O.l)

±0.OO2 (±0.05)

±0.OO8 (±0.2)

As desired

N.A.

As desired Yes

2 to I maximum Yes

Yes No Press capacity

Yes Yes Press dimensions

Yes

Yes

1/4

As desired As desired Yes

Yes

Yes Yes Mold size

Yes Yes Bag size

Yes

N.R.

Yes Yes Yes Yes Yes

N.R. Yes Yes Yes Yes

No No Yes No Yes

No N.R. Yes No No

Yes Yes No Yes No

Yes N.R. Yes Yes Yes

None

Flexibility of the bag Yes One Orientation of ply

Surface of revolution Yes One Depends on wind

Constant cross-section Yes Two Directional

Moldable

Moldable

No Two Random

Yes Two Random

45--60

50-75

30-60

25

30

Yes One Random

30-45

Circumferential In longitudinal only direction Yes No Yes Yes Lathe bed Pull capacity length and swing Yes No

~

50+*

I. Smaller variety of finishes available, such as chrome or baked enamel 2. No R.F.1. and grounding capabilities 3. Harder to retrofit to frame or skins 4. Thicker wall 5. Higher tool costs than with brakeforming

for complex configurations

8. Reduced tooling costs

shipping

5. Greater design freedom 6. Better sound damping 7. Reduced damage from

canning

3. Dent resistance 4. Elimination of oil

-Even with limited quantities. tDcpcnding on unit volume and part size.

~'"

c:

]]g

·ti "'''''

.g .S

e~

]

.1

J

2. Fewer parts required for assembly

time

product integrity; less final product-inspection

tolerances; increased

1. Fabrication economy: less assembly time; tighter dimensional

Foam vs. Sheet Metal

15 to 30

I. No heat sink capabilities 2. No R.F.I. and grounding capabilities 3. Fewer available finishes for cosmetic appearance 4. Higher finishing costs 5. Thicker walls 6. Possible internal voids

8. Better impact resistance

weight

maintenance 3. No trim dies required 4. Lighter weight 5. Higher impact resistance 6. Better sound damping 7. Better strength-to-

2. Longer tool life, lower

costs

I. Much lower tooling

Foam vs. Die Casting

Up to 30

I. Increased finishing costs (surface swirl) 2. Heat distortion 3. Thicker wall 4. Lower physical properties 5. Possible internal voids

5. Greater inherent structural capabilities 6. Lower shipping costs 7. Large parts more economical 8. Lower tooling costs 9. Better sound damping

resistance

I. Uniform physical properties throughout the part 2. Warping and sink marks reduced or eliminated 3. No resin-rich areas to cause configuration problems 4. Higher impact

Foam vs. Sheet Molding Compound

50+

part size vs. quantity 3. Thicker walls 4. Higher tooling costs

distortion 2. Poorer economies of

I. More prone to heat

7. Better sound damping

properties

5. More design freedom 6. More uniform physical

stability

4. Better dimensional

I. More consistent part reproducti<>n 2. Lower labor 3. Simplified assembly

Foam vs. Hand Lay-up Fiberglass

15 to 20t

3. Longer cycle time 4. Thicker walls 5. Poorer high-volume economics 6. Less equipment available for various shot sizes

I. Poorer surface finish 2. Application of cosmetic detail for appearance parts

capability 5. Better sound damping 6. Lower internal stresses 7. Sink marks reduced or eliminated 8. Inherent structural strength

functional engineering 2. Better low- to mediumvolume economics 3. Lower tooling costs 4. Better large-part

(Many process similarities exist) I. Flexibility for

Molding

Foam vs. Injection

Table 7-32. Comparison of Structural Foam with Five Other Processes

Chapter 8

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS

Many different types of auxiliary equipment (AE) and secondary operations (SO) can be used to maximize overall processing productivity and efficiency and reduce the fabricated product cost (see Figs. 8-1 and 8-2). Their proper selection, use, and maintenance are as important as the selection of the processing machines such as the injection molder or extruder and they can cost more than the base machine. The processor must determine what is needed, from upstream to downstream, based on what the equipment has to accomplish, what controls are required, the ease of operation and maintenance, safety devices, energy requirements, and compatibility with existing equipment wanted, and so on. This chapter provides examples of this selection procedure and its importance in evaluating all the equipment required in a processing line. Details on the equipment available can be obtained from plastics industry trade publications, usually compiled in annual (appendix) issues. These and other pertinent publications are included in the Reference section at the end of this book. [2, 5-12, 674-94]. All this equipment has to be properly interfaced to operate efficiently. In the past, most of the equipment did not properly, or at least easily, interface mechanically and/or electronically. The term protocol, as it relates to an interface, means a set of rules governing the communication and transfer of data between machines and equipment. The Society of the Plastics Industry's Machinery Division formed a special committee in 1987 to develop its Communication Protocol Standard Development Kit. (Other countries have protocols as well.) This kit includes a complete reference manual and program simulation software. The SPI made the kit available late in 1989 under a paid-license agreement to members for $2,500 and nonmembers for $5,000. The SPI kit covers the primary processing machine communication protocol with chillers, blenders, dryers, water systems, discrete mold-temperature controllers, and the like. The protocol provides for centralized setup and monitoring of auxiliaries by the primary machine. The test simulation software that is a key part of the kit ensures uniform interpretation of technical specifications. It is a combination of a detailed technical specification of hardware requirements such as the type of cable, the connector and electrical interface, and software requirements. The kit protocol shows how electronic information moves through the system. The SPI standard references the standards of the Electronics Industry Association, American National Standards Institute, and Institute of Electrical and Electronics Engineers. 711

Imprln,.,

Figure 8-1. An injection-molding production line that starts upstream with materials being delivered to the 1M machine and progresses through the downstream equipment where the finished product (hangers) are leaving the plant. This flow diagram defines the manufacturing sequence and capacity requirements.

Figure 8-2. Examples of auxiliary equipment in a production line that starts at the right end at top and moves to the left end at bottom. 712

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 713

MATERIAL HANDLING In most processes, for either small or large production runs, the cost of the plastics used compared to the total cost of production in the plant may be at least 60 percent. The proportion might be only 30 percent, but it is more likely to exceed 60 percent, so it is important to handle material with care and to eliminate unnecessary production problems and waste. Where small-quantity users or expensive engineering resins are concerned, containers such as bags and gaylords (storage container holding 1,000 lb. of material) are acceptable, but for large commercial and custom processors these delivery methods are bulky and costly. Resin storage in this form is also expensive. Any large-scale resin-handling system has three basic subsystems, for unloading, storage, and transfer. For a complete system to work at peak efficiency, processors need to write specifications that fully account for the unique requirements of each subsystem. The least efficient component, no matter how inconsequential it may seem, will limit the overall efficiency of the entire system.

Energy Conservation

Energy conservation is only one of many factors that should be considered in the selection of an automated materials conveying system, as well as in all equipment used in the processing line. Fortunately, any steps taken to save energy will also save money, in most cases. The traditional arguments favoring the silo are its savings on resin costs, labor savings through the elimination of handling bags and cartons, the savings of a costly warehouse inside floor space, and energy savings. For example, if a plant used a large quantity of resins and did not use silos, during the winter months bags or gaylords would be delivered repeatedly through open delivery doors, and warm air would be lost. With automatic delivery from silos, all resin-handling lines are kept as short as possible. There is no reason for these lines to conform to the right angles of walls; they should follow a straight line from the resin's source to where it has to be delivered. There are graphs from systems suppliers that show the relationship between the lengths of conveyor lines and the power requirements [12]. The graphs also show the horsepower required, based on different factors such as the length and diameter of the delivery pipe, the position of the pipe, the type of resin being conveying, the size of the hopper at the machine, and the Ib.lhr. that can be delivered.

PARTS HANDLING The logic and approach used in materials handling also apply to the use of handling equipment to move processed parts. Parts-handling equipment (PHE) does not resemble the humanoids of science fiction. Robots are blind, deaf, dumb, and limited to a few preprogrammed motions, but in many production jobs that is all that is needed. They are solutions looking for a problem. Most plants can use some degree of PHE, which can substantially increase productivity. The use of PHE can range from simple operations to rather complex ones with sophisticated computer controls. Although the concept of automatic operation is appealing, its ultimate justification, as for material handling and process controls, must be made on the basis of economics. At times it may provide the solution to handling a part that otherwise would be damaged.

714 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

FINISHING AND DECORATING

The finishing of plastics includes different methods of adding either decorative or functional surface effects to a plastic product (see Figs. 8-3 and 8-4 and Tables 8-1 and 82). Plastics are unique in that color and decorative effects can be added to them prior to and during processing. Two-or-more-color plastic parts are easily processed. Decorative surface textures can be incorporated on the surface of practically any product during or after processing. Decorative foils of plastic or aluminum or other materials placed on the surface during injection molding, extrusion, blow molding, and so on can also improve the performance of the plastic. JOINING AND ASSEMBLING

Different methods are used for joining and assembling plastic products, as summarized in Figures 8-5 and Tables 8-3 to 8-6. Information on assembly using methods such as inserts, press fits, and snap fits, was provided in the "Structural Design Analysis" section of Chapter 5; see also Chapter 11.

Adhesives and Solvents There are many different adhesive systems used to join or assemble products. There are solvent systems for most TPs, but not for thermosets. Monomeric or polymerizable cements can be used for most TPs and TSs. There are certain plastics with outstanding chemical resistance, such as the polyolefins, that preclude the use of many cements, but they generally require some form of surface treatment prior to adhesion, such as flame treatment. Solvent bonds work because they react chemically with the plastic. However, they literally destroy it, so it is important to limit such factors as the length of time and depth of the plastic soak. The solvent could cause either immediate or delayed damage. If a part contains excessive internal strains, the solvent could release the strains and cause cracking, surface defects, and so on. (To evaluate a plastic's reaction with a solvent, PART TO BE OECORATEO

I

I

SURFACE PREPARA nON

I FLASH REMOVAL

I TUMBLING WHEE LABRATOR HAND FILING MACHINING

I

I I SURFACE COATING

BONO PREPARATION

I

SANDBLAST PLASMA ETCH SOLVENT ETCH CHEMICAL ETCH

PAINTING OYE APPLICATION METALLIZATION by VACUUM and ELECTROPLATING SILK SCREEN TRANSFER

HOT STAMPING and EMBOSSING DECALS IN-MOLD DECORATION

Figure 8·3. A guide to decorating selection.

INTEGRAL COATING

COLOR CONCENTRATES IN RESIN LIQUID ADDED AT PRESS OR EXTRUDER COLOR MIX IN TUMBLE BARREL

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 715 Multiple-level paint and foil 1st color all over ( 1 st color all over

~nd

Foir--...

Foi",

~ c:ol~r

(

__

~nd

~col~r

Masking step .030" x .030"

Masking step .030" x .030"

Foil surface around raised clear area

Multiple decoration on flush surfaces

Clear rec!. window raised .060" above surrounding surface

Groove for foil stop-off: .060"w x .040"d. For paint mask: .030" x .030". Groove color is from base mat'!.

Radiused edge and shallow drop (18 ) to foil surf. prevents foil wrinkles at corners

~ Bright metal frame or edge 2-pc: acrylic ASS

Separate metallized or plated 1st surf. clear~ ASS frame assembled by ~~ heat staking. (Leave clearance for gate removal.) 2nd surf paint or f o i l -I

Molded-In letters for metallizing

Two-piece assembly avoids higher cost of metallizing mask and gives choice of other material for frame.

For small letters: .025" x .025" with .005" radius on internal edges For large letters: .050"w x .030"d with .003-.004" crown on bottom to increase brightness. Use 3-5 draft on groove walls.

l-pc acrylic 1st surf. Clear~st surf. metallize

2nd surf. paint

o~aSking

groove .050"w x .040"d

Figure 8-4. An example of the dimensions recommended for masking and stop-off grooves.

I PART TO BE ASSEMBLED I

I

I

I

ITHERMOPLASTICS I

I THERMOSETS I I

I HIGH VOLUME I

I

I

MOLOED-IN INSERTS MECHANICAL FASTENERS ELECTROMAGNETIC. INDUCTION HEATING OF ADHESIVE ADHESIVE BONDING (DRIFILM • WET)

I

I LIIWVOLUME

I

I

I

ADHESIVE BONDING MECH FASTENERS

I HIGH VOLUME I I

I

SOLVENT BONDING HOT TOOL WELDING ULTRASONIC WELDING ADHESIVE BONDING ELECTROMAGNETIC • INDUCTION BONDING SPRING. VIBRATION WELDING DIELECTRIC HEAT SEALING

I

I

I LllWVOLUME

1

I

GAS WELDING ADHESIVE ELECTRD-MAGNETIC • INDUCTION BONDING ULTRASONIC HOT TOOL WELDING SPIN WELDING

Figure 8·5. A guide to part assembly selection.

after a part is processed it is immersed in a solvent to determine whether strain patterns exist. The reaction with the solvent can be correlated with the processing versus part performance. ) The solvent action described does not mean that adhesives are harmful; they have been used successfully for over a century. But no matter what action is taken in joining (Table 8-3) or assembling (Table 8-4), the processor should determine whether there are limitations to its use.

~

....

-

Roller coating

Wiping

Electrostatic spray

Painting: Conventional spray

The Process

Generally for one-color, overall coating.

One color per pass; multicolors achieved in multistation units.

All plastics can be decorated. Some work, not much, being done on powder coating of plastics.

Can be used for most materials. Products range from medical containers to furniture.

Can be used for most materials.

Spray gun, high-voltage power supply; pumps; dryers. Pretreating station for parts (coated or preheated to make conductive).

Standard spray-paint setup with a wipe station following. For low production, wipe can be manual. Very highspeed, automated equiment available.

Roller applicator, either manual or automatic. Special paint feed system required for automatic work. Dryers.

Generally one-color painting, though multicolor possible with side-by-side rollers.

Solids, multicolor, overall or partial decoration, special effects sush as woodgraining possible.

Can be used on all materials (some require surface treatment) .

Spray guns, spray booths, mask washers often required; conveying and drying apparatus needed for high production.

Effect

Paint's sprayed by air or airless gun( s) for functional or decorative coatings. Especially good for large areas, uneven surfaces, or relief designs. Masking used to achieve special effects. Charged particles are sprayed on electronically conductive parts; process gives high paint utilization; more expensive than conventional spray. Paint is applied conventionally, then paint is wiped off. Paint is either totally removed, remaining only in recessed areas, or is partially removed for special effects such as woodgraining. Raised surfaces can be painted without masking. Special effects like stripes.

Applications

Equipment

How It Works

Table 8-1. Printing and Decorating Systems (Courtesy of Plastics Technology)

'I

'I

...

Depositing, in a vacuum, a thin layer of vaporized metal (generally aluminum) on a surface prepared by a base coat.

Metallizing: Vacuum

Electroplating

Heat Transfers

Hot Stamping

Ink is applied to part through a finely woven screen. Screen is masked in areas that won't be painted. Economical means for decorating flat or curved surfaces, especially in relatively short runs. Involves transferring coating from a flexible foil to the part by pressure and heat. Impression is made by metal or silicone die. Process is dry. Similar to hot stamp but preprinted coating (with a release paper backing) is applied to part by heat and pressure. Gives a functional metallic finish (matte or shiny) via electrodeposition process.

Screen Printing

Most thermoplastics can be printed; some thermosets. Handles flat, concave, or convex surfaces, including round or tubular shapes. Can handle most thermoplastics. A big application area is bottles. Fiat, concave or cylindrical surfaces. Can handle special plating grades of ABS, PP, polysulfone, filled Noryl, filled polyesters, some nylons.

Rotary or reciprocating hotstamp press. Dies. High-speed equipment handles up to 6,000 partsIhr.

Ranges from relatively simple to highly automated with multiple stations for, say, front and back decoration. Preplate etch and rinse tanks; Koroseal-lined tanks for plating steps; preplating and plating chemicals; automated systems available. Metallizer, base, and topcoating equipment (spray, dip or flow), metallizing racks.

Most plastics, especially PS, acrylic, phenolics, PC, unplasticized PVC. Decorative finishes (e.g., on toys), or functional (e.g., as a conductive coating).

Most materials. Widely used for bottles; also finds big applications in areas like TV and computer dials.

Screens, fixture, squeegee, conveyorized press setup (for any kind of volume). Dryers. Manual screen printing possible, for very lowvolume items.

(cont'd)

Metallic finish, generally silver but can be others (e.g., gold, copper).

Very durable metallic finishes.

Multicolor or single color; metallics (not as good as hot stamp).

Metallics, wood grains or multicolor, depending on foil. Foil can be specially formulated (e.g., chemical resistance) .

Single or multiple colors (one station per color).

-

==

.....

Film or foil inserted in mold is transferred to molten plastics as it enters the mold. Decoration becomes integral part of product. Printing of a surface directly from a rubber or other synthetic plate Roll-transfer method of decrating. In most cases less expensive than other multicolor printing methods. Uses embossing rollers to print in depressed areas of a product. From simple paper labels to multicolor decals and new preprinted plastic sleeve labels.

In-the-Mold Decorating

Valley Printing

Labeling

Offset Printing

Flexography

Special process using a soft transfer pad to pick up image from etched plate and tamping it onto a part.

Deposition of a metallic finish by chemical reaction of water-based solutions.

Spray

Tamp Printing

Unifonn metallic coatings by using electrodes.

How It Works

Cathode sputtering

The Process

Ranges from low-cost hand presses to very expensive automated units. Drying, destaticizers, feeding devices. Embosser with inking attachment or special package system. Equipment runs the gamut from hand dispensers to relatively high-speed machines.

Manual, semi- or automatic press, dryers.

Activator, water-clean and applicator guns; spray booths, top- and basecoating equipment if required. Metal plate, squeegee to remove excess ink, conical-shaped transfer pad, indexing device to move parts into printing area, dryers, depending on type of operation. Automatic or manual feed system for the transfers. Static charge may be required to hold foil in mold.

Discharge systems-to provide close control of metal buildUp.

Equipment

Table 8-1. (Continued)

Used largely with PVC, PE for such areas as floor tiles, upholstery. Can be used on all plastics. Used mostly for containers and for price marking.

Single- or multicolor decoration.

Most plastics, especially polyolefins and melamines. For parts where decoration must withstand extemely high wear. Most plastics. Used on such areas as coding pipe and extruded profiles. Most plastics. Used in applications like coding pipe.

All sorts of colors and types.

Generally two-color maximum.

Multicolor print or decoration.

Single- or multicolor.

Single- or multicolor--one printing station per color.

Metallic (silver and bronze).

Metallic finish. Silver and copper generally used. Also gold, platinum, palladium.

Effect

All plastics. Specially recommended for oddshaped or delicate parts (e.g., drinking cups, dolls' eyes).

High-temperature materials. Unifonn, precise coatings for applications like microminiature circuits. Most plastics. For decorative items.

Applications

'"

.........

Two-shot molding

Inserted nameplates

In-mold label

Engraved mold

Investment: moderate to high

Labor cost: high Investment: moderate Unit cost: high Labor cost: high

Investment: none to moderate Unit cost: high

Labor cost: low Investment: moderate Unit cost: high Labor cost: high

Unit cost: low

Economics

Limited

Partially limited

Unlimited

Limited

Aesthetics

Somewhat restricted

Restricted

Somewhat restricted

Unrestricted

Product Design Chemistry

Good durability

Not critical

Good durability

Not critical

Good durability

Critical

Good durability

Not critical

Done in the Mold

Two molding operations

Longer molding cycles

Longer molding cycles

No extra operations

Manufacturing

Comments

(cont'd)

Good where maximum abrasion resistance necessary.

Allows three-dimensional as well as special effects.

Good for thermoplastics and thermosets. Automatic loading equipment becoming available.

Best for simple lettering and texture.

Table 8-2. Guide to Plastic-Decorating Methods (Courtesy of Plastics Technology)

0

hi

......

Heat transfer

Hand painting

Flexographic

Electrostatic

Applique

Labor cost: high Investment: low Unit cost: low to moderate Labor cost: low to moderate Investment: low to moderate

Investment: moderate to high Unit cost: high

Labor cost: high Investment: moderate to high Unit cost: low to moderate Labor cost: low Investment: moderate to high Unit cost: low Labor cost: low

Unit cost: high

Economics

Unlimited

Somewhat limited

Somewhat limited

Limited

Somewhat limited

Aesthetics

Somewhat restricted

Unrestricted

Moderate durability

Restricted

Somewhat restricted

Unrestricted

Product Design

Good durability

Good durability Critical

Critical

Moderate to good durability Critical

Critical

Good durability

Not critical

Chemistry

Done after Molding

Table 8-2. (Continued)

Requires little 1I00r space

Hand operation

Automates well

Hand operation

Manufacturing

Multicolor graphics.

Dry process, tool contacts product.

Wet process, tool contacts product.

Wet process, tool contacts product. Sometimes requires top coat.

Dry process, no tool contact with product.

Allows unusual effects.

Comments

...

'I N

Offset

Nameplates

Metallizing

Labeling

Hot stamping

Labor cost: moderate Investment: high

Investment: low to moderate Unit cost: low

Unit cost: moderate to high Labor cost: moderate to high Investment: high Unit cost: high Labor cost: moderate to high

Labor cost: low to moderate Investment: low to moderate Unit cost: low to moderate Labor cost: low to moderate Investment: low to high

Unit cost: low

Unlimited

Unlimited

Limited

Unlimited

Limited

Restricted

Somewhat restricted

Somewhat restricted

Somewhat restricted

Somewhat restricted

Moderate to good durability

Critical

Good durability

Good durability Less critical

Moderate to good durability Critical

Less critical

Good durability

Critical

Automates well

Adaptable to many situations

Requires special technological know-how

Adaptable to many situations

Requires little floor space

Multicolor graphics.

(cont'd)

Wet process, tool contacts product.

Multicolor graphics.

Produces bright metallics. Dry process, tool contacts product.

Wet and dry process, no tool contact with product.

Multicolor graphics.

Dry process, no tool contact with product at times.

Produces bright metallics.

Dry process, tool contacts product.

~

hi hi

Woodgraining

Spray

Silk screening

Offset intaglio

Labor cost: high Investment: moderate to high

Unit cost: moderate Labor cost: moderate Investment: moderate Unit cost: moderate Labor cost: moderate Investment: moderate to high Unit cost: high

Labor cost: moderate Investment: moderate

Unit cost: low

Economics

Specialized

Limited

Somewhat limited

Limited

Aesthetics

Specialized

Unrestricted

Somewhat restricted

Unrestricted

Product Design

Good durability

Critical

Good durability

Critical

Good durability

Moderate to good durability Critical

Critical

Chemistry

Done after Molding

Table 8-2. (Continued)

Mostly hand operated

Requires much floor space

Flexible operation

Requires little floor space

Manufacturing

Wet process, tool contacts products.

Wet process, no tool contact with product.

Wet process, tool contacts product.

New process.

Wet process, tool contacts product.

Comments

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 723

Table 8-3. Reference Chart to Help in Selecting the Proper Method of Fastening Thermoplastic Materials

Thennoplastics

Spin and Mechanical Vibration Thermal Ultrasonic Induction Fasteners Adhesives Welding Welding Welding Welding

ABS

G

G

G

G

G

G

Acetal

E

P

G

G

G

G

Acrylic

G

G

F-G

G

G

G

Nylon Polycarbonate Polyester TP Polyethylene

G G G P

P G F NR

G G G G

G G G G

G G G G-P

G G G G

Polypropylene

P

P

E

G

G-P

G

Polystyrene

F

G

E

G

E-P

G

G NR G F

G G G G

G NR E F

E NR G G

E NR G F

G G G G

Polysulfone Polyurethane TP PPO modified PVC rigid

Remarks Body type adhesive Recommended Surface treatment for adhesives Body type adhesive Recommended

Surface treatment for adhesives Surface treatment for adhesives Impact grades difficult to bond

E = Excellent, G = Good, F = Fair, P = Poor, NR = Not recommended.

Ultrasonic Welding Ultrasonic welding is an economical method for joining small- to medium-sized plastic parts of the same or similar plastics. Certain polymers may not weld if they contain specific fillers, such as particularly glass fibers if they are of a high concentration. This technique is rapid and can be fully automated. Welding occurs when high-frequency (2040 kHz) vibrational energy is directed to the interface between the two parts, creating localized molecular expansion causing the plastic to melt. Pressure is maintained between

Table 8-4. Reference Chart to Help in Selecting the Proper Method of Fastening Thermoset Plastic Materials

Thennosets

Mechanical Fasteners

Spin and Vibration Thermal Ultrasonic Adhesives Welding Welding Welding

Induction Welding

Epoxies Melamine

G G G F

G G E G

NR NR NR NR

NR NR NR NR

NR NR NR NR

NR NR NR NR

Phenolics Polyester Polyurethane Silicones Ureas

G G G F F

E E E G G

NR NR NR NR NR

NR NR NR NR NR

NR NR NR NR NR

NR NR NR NR NR

Alkyds

DAP

E=Exce11ent. G=Good, F=Fair, P=Poor, NR=Not recommended.

Remarks

Material notch sensitive

Material notch sensitive

""

~

'I

50-70 30-70 50-70 35-50 60-80 60-80 20-50 20-60 60-70 60-70

2,000-8,000 8,000-11,000 7,000-12,000 8,000-9,500 800-6,000 3,000-6,000 3,500-8,000 8,000-11,000 5,000-9,000 3,000-5,000

50-70 30-50 50-70 40-50 70-90 70-90 30-60 20-50 50-70 50-70

50-70 20-50 50-70 40-50 60-80 60-80 20-50 20-50 60-70 60-70

20-30 65-80 65-80

20-30 60-75 60-75

8,000-10,000 2,400-8,500 3,000-7,000

50-70 65-80 65-80

50-70

50-70

50-70

2,400-9,000

Hot-Plate Welding

10-15

Friction Welding

10-15

Hot-Air Welding

7,000-13,000 7,000-13,000 6,000-9,000 6,000-13,000

"To convert psi to Pascals, multiply by 6,895.

Thennosetting plastics Epoxy Melamine Phenolic Polyester Thennoplastics Acrylonitrile butadiene styrene Acetal Cellulose acetate Cellulose acetate butyrate Ethyl cellulose Methyl methacrylate Nylon Polycarbonate Polyethylene Polypropylene Polystyrene Polystyrene acrylonitrile Polyvinyl chloride Saran

Original Tensile Strength (psi)*

30-50 60-70 60-70

50-80

Dielectric Welding

25-50 25--60 50-70 50-70

40-60

80-90 40-60

90-100 90-100

30-60

Solvent Welding

50-80 40-60 20-40 5-15 10-30 20-40 20-50 20-50 50-70 50-70

50--60 50-80

40-60

50-80 50-80 50-80 50-80

Adhesive Bonding

Table 8-5. Percent Tensile Strength Retention with Different Welding Techniques

60-90

60-100 60-100 60-100 60-100

Polymerization Welding

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 725

Table 8-6. Materials for Plastics Hardware Application Hinges Knuckle and pin

Material

Snap-in

Snap-on

Clasp

Drive-pin

ABS Acetal Acrylic Cellulosic Fluorocarbon Polycarbonate Polyethylene Polyamide Polypropylene Polystyrene Polyurethane Vinyl

j j

j j

j j j

j j j

j

j

j

j

j

j

j

j

Ball-grip

Integral

j j j j j

j j j j j

j

j

j j

j j j

j

j

j

j

the two parts after vibration stops, and the melted polymer immediately solidifies. The entire welding process nonnally takes place in less than two seconds. It has high strength, which sometimes approaches the strength of the base material, if the joint design is correct and the equipment is properly set. If it is not properly bonded, poor bonds can be created or nonair-tight contact occurs. Table 8-7 shows the types of welds that can be made.

Vibration Welding In vibration welding, two plastic parts are rubbed together in either linear or angular displacement, producing frictional heat that results in a melt at the interface of the two parts. Different bonding joints can be used to eliminate having flash that is visible at the joints; basically, recesses within the bond exist. The vibration is in the fonn of highamplitude, low-frequency, reciprocating motion. With circular parts a rotary motion is used. When the vibration stops, the melt cools and the parts become pennanentIy welded in the alignment that is held. Typical frequencies are 120 and 240 Hz, with amplitudes range from 0.10 to 0.20 in. of linear displacement. Vibration welding, like ultrasonic welding, produces high-strength joints for materials that can be melted. However, it is much better suited to large parts and irregular joint interfaces. Moisture in materials does not usually have an adverse effect on the weld as it does with ultrasonics.

Spin Welding Spin welding is just a special fonn of vibration welding. Because it is such a popular technique, it is considered a special assembly method.

Radio Frequency Welding With this type of process, welding occurs from the heat created by the application of a strong radio frequency (RF) field to the selected joint region on those plastics that are

Table 8-7. Plastics' Characteristics with Regard to Different Types of Ultrasonic Welding Applications Percent of Weld Strength*

Welding Spot Weld

Staking and Inserting

Swaging

Near Fieldt

Far Fieldt

95-100+ 95-100+ 90--100 1

E E E

E E E

G F F

E E G

G E P

95-100 95-100+

E E

E E

G F

E E

G-P E

95-100+

E

E

F

E

E

95-100+ 95-100+ 2

E E

E E

G G

E E

G G

95-100+

E

E

G

G

F

65-703 95-100+ 4 95-100

G G E

E E E

P P G

G E E

G G G

95-100+

E

E

G

G

F

95-100+ 90--100+ 95-100+

E E E

E E E

G G F-P

E G G

G F E-G

90-100+ 2 90-100+

E G

E G

F-P F

G G

F F

90-100 95-100+ 95-100+ 2 80-90 95-100+

G G E F E

E E E G G

G G G-F P F-P

G E E G G

G-F G E F G-F

95-100+ 2

E

E

F

G

G-F

90-100 90--100 90--100 90--100 85-100

G G E E E

G-F G-F E E E

G G G G F

P P G-P G-P G

P P F-P F-P F-P

40-100

G

G-F

G

F-P

F-P

Material

General-purpose plastics ABS Polystyrene unfilled Structural foam (styrene) Rubber modified Glass filled (up to 30%) SAN

Engineering plastics ABS ABS/polycarbonate alloy (Cycoloy 8(0) ABS/PVC alloy (Cycovin) Acetal Acrylics Acrylic multipolymer (XT-polymer) Acrylic/PVC alloy (Kydex) ASA Methylpentene Modified phenylene oxide (Noryl) Nylon Polyesters (thermoplastic) Phenoxy Polyarylsulfone Polycarbonate Polyimide Polyphenylene oxide Polysulfone

High-volume, low-cost applications Butyrates Cellulosics Polyethylene Polypropylene Structural foam (polyolefin) Vinyls Code: E

= Excellent,

G

= Good,

F

= Fair, P = Poor.

·Weld strengths are based on destructive testing. 100+% results indicate that parent material of plastic part gave way while weld remained intact.

tNear field welding refers to joint IHigh-density foams weld best. 2Moisture will inhibit welds.

1/4

in. or less from area of hom contact: far field welding to joint more than

'Requires high energy and long ultrasonic exposure because of low coefficient of friction. 'Cast grades are more difficult to weld due to high molecular weight.

726

1/4

in. from contact area.

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 727

not transparent to RF. The RF is usually applied by a specially formed metal die in the shape of the joint desired, which also applies the clamping pressure needed to complete the weld after the plastic melts. This is a fast process that is sensitive to heat buildup. This type of welding, usually referred to as heat sealing, is widely used with flexible TP films and sheets such as plasticized PVC and PUR. It can also be used to join film to plastic molded parts.

Heat Welding Many TPs can be heat welded. However, as with other welding techniques, certain fillers or too much of a particular filler, could prevent good bonding. There are nevertheless, certain fillers that can improve bonding action. Heat welds can also be used with friction or spin welding of TP joints.

Electromagnetic and Induction Welding This type of welding uses a radio frequency magnetic field to excite fine, magnetically sensitive particles that are either metallic or ceramic. The particles can be embedded in a preform, filament ribbon, adhesive, coextruded film, molding compound, and other materials. The most common is to include an extra part such as a preform containing the magnetically active particles. The preform is placed at the joint's interface and exposed to an electromagnetic field. Then electromagnetically induced heat is conducted from the particles through the preform and to the part joint as the parts are pressed together.

Summary These and various other joining and assembly methods are used. As noted, certain plastics can be solvent bonded or heat bonded, with or without fillers and reinforcements. One problem that can develop after production starts is to change a plastic's composition but continue using the same basic polymer. This action may be taken to improve performance or cost. The assumption could be made that the joining action would be the same. It might indeed, but without checking this could be disastrous. For example, with a higher percentage of calcium carbonate or glass fibers (additives that are basically not affected by solvents or heat), there would be no joining capability. Thus, it is important to check out the performance of any new material.

MACHINING Each type of plastic has its own unique properties and machining characteristics, which are far different from those of the metallic or nonmetallic materials familiar to many processors. TPs are relatively resilient compared to metals and require special cutting procedures. Even within a family of plastics (PE, PC, PPS, etc.), the cutting characteristics will change, depending on the fillers and reinforcements. Elastic recovery occurs in plastics both during and after machining, so provision must be made in tools' geometry for sufficient clearance to allow for it. This is so because of the expansion of compressed material due to elastic recovery (see Chapter 2). This causes increased friction between the recovered cut surface and the cutting surface of the tool. In addition to generating heat, this abrasion affects tool wear. Elastic recovery also

728 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

explains why, without proper precautions, drilled or tapped holes in plastics are often tapered or become smaller than the diameter of the drills used to make them. As heat conductivity in plastics is very slow, essentially all the cutting heat generated will be absorbed by the cutting tool. The small amount of heat conducted into the plastic cannot be transferred to the core of the shape, which causes the heat of the surface area to rise significantly. This heat must be kept to a minimum or be removed by a coolant to ensure a proper cut. For many commodity TP resins the softening, deformation, and degradation heats are relatively low. Gumming, discoloration, poor tolerance control, and poor finish are apt to occur if frictional heat is generated and allowed to build up. Engineered TP resins such as nylon and TFE-fluoroplastic have relatively high melting or softening points. Thus, they have less tendency to become gummed, melted, or crazed in machining than do plastics with lower melting points. Heat buildup is more critical in the plastics with lower melting points. Thermoset resins generally have the fewest problems of any plastics during machining.

Cutting Guidelines The properties of plastics must be considered in specifying the best speeds, feeds, depths of cuts, tool materials, tool geometries, and cutting fluids. Machining data are available from machinery handbooks as well as the plastic material and cutting machinery suppliers. Note that some plastics may be cut at higher speeds with no appreciable loss of tool life, but higher speeds usually result in thermal problems, especially with the commodity resins. The guidelines for tool geometry start by reducing frictional drag and heat. It is desirable to have honed or polished surfaces on the tool where it comes in contact with the work. The geometries of the tools should be such that they generate continuous-type chips. In general, large rake angles will serve this purpose because of the force directions resulting from these angles. Care must be exercised to keep rake angles from being so large that the brittle fracture of workpieces results and chips become discontinuous. Drill geometry should be made to differ from that used for metals by employing wide, polished flutes combined with low helix angles, to help eliminate the packing of chips, which causes overheating. Also, the normally 118-degree point angle is generally modified to 70 to 120 degrees. Round saws should be hollow ground, with burrs from sharpening removed by stoning, and handsaws and jigsaws should have enough set to give adequate clearance to the back of the blade. This set should be greater than is usual for cutting steel. It is always better to relieve the feed pressure near the end of a cut to avoid chipping. The proper rate of feed is important and, because most sawing operations are hand fed, experience is required to determine the best rate. Attempts to· force the feed will result in heating of the blade, gumming of the plastic, loading of the saw teeth, and an excessively rough cut. Chrome plating the blade reduces friction and tends to give better cuts. Above all, the saw-whether band or circular-must be kept sharp. Circular saws are usually from to i in. thick. The width of bandsaws is usually f« to ! in. Both TP and TS resins can be sawed by using cutoff machines with abrasive wheels. This equipment is used to cut rods, pipes, L-beams, and so on. With appropriate wheels, properly used, clean cuts can be made. If necessary, water is used to prevent overheating. Practically all cutting and machining operations can be performed on plastics if it is kept in mind that their properties vary from soft and flexible to hard and brittle, that

n

AUXILIARY EQUIPMENT AND SECONDARY OPERATIONS 729

some are weak and others strong, some soften upon heating and others do not, and that they may contain a wide variety of additives that will affect their machining characteristics. The cutting and rake angles, relative rates of cutting speed and feed, types of cutting edges from plain metal to diamond saws, various coolants, and other factors affecting machining must be adjusted to the particular plastic. When it is properly carried out with appropriate tools, machining can be readily accomplished.

Chapter 9

TESTING AND QUALITY CONTROL

The demand for quality products, and the competition this creates, requires designers to build in quality early in the design cycle. The available conventional, and particularly computer-aided, testing helps designers meet this important challenge. Throughout this book, many different tests have been presented (see the Index, under "Test Methods," for these tests). This action was necessary to have the properties they represent properly defined and applied in their respective areas of design interest. This chapter is an overview to help the designer, processor, material supplier, customer, and others understand how to evaluate tests. The individuals required to conduct tests should review the applicable and required specifications or standards to ensure that procedures and test equipment are being properly used. It is of special importance to make sure that the most up-to-date procedures are being observed [1-14, 35, 355, 432, 695-750]. Designers and processors should keep quality under control and demand having consistent materials that can be used with a minimum of uncertainty. Plant quality control (QC) is as important to the end result as selecting the best processing conditions with the correct grade of plastic, in terms of both properties and appearance. After the correct plastic has been chosen, probably its bleeding, reprocessing, and storage stages of operation need to be frequently or even continuously updated. The processor should set up specific measurements of quality to prevent substandard products from reaching the end user. How deeply one gets involved depends on the performance requirements. If all that is required is to weigh the part, that is all that one does. Testing and QC are the most discussed but often the least understood facets of business and manufacturing. Many companies spend a high percentage of each sales dollar on QC. Usually it involves the inspection of components and parts as they complete different phases of processing. Parts that are within specifications proceed, whereas those that are out of spec are either repaired or scrapped. The workers who made the out-of-spec parts are notified that they produced defective parts and that they should correct their mistakes. The most important testing is that done on the finished part. In tum, tests done on materials and during processing must all be related to the final part performance. Unfortunately, there is no single set of rules designating which tests are to be conducted in order to manufacture a part repeatedly with zero defects. The tests depend on the required performance. For example, if a part is to operate where any type of failure could be catastrophic to life, then extensive-and usually expensive-testing is necessary. The approach just outlined is after-the-fact QC in that all defects caught in this manner will already be present in the part being processed. This type of QC will usually catch 731

732 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

defects, and it is necessary, but it does little to correct basic problems in production. One of the problems with add-on QC of this type is that it constitutes one of the least cost-effective ways of obtaining a high-quality part. Quality must be built into a product from the beginning; it cannot be inspected into the process. The closest add-on, afterthe-fact quality control can come to improving the quality built into a part is to point out processing, material, and designed-in defects to the departments or persons responsible for them. The object instead should be to control quality before a part becomes defective. There are many different approaches to setting up QC. For example, mechanical properties can be considered the most important of all, and there are many factors that determine the mechanical behavior of plastics. As seen throughout this book, the factors that influence meeting the design requirements of properties include the resin composition (fillers, molecular weight distribution, morphology, etc.), the processing method and machine controls, the capability of auxiliary equipment, and part performance requirements. Considering what the critical areas of a process are, one can understand that sometimes a test or measurement of resin viscosity is all that is needed.

BASIC VERSUS COMPLEX TESTS Choosing and testing a plastic when only a few existed that could be used for specific products would prove relatively simple if the selection were limited, but the variety of plastics has proliferated. Today's plastics are also more complex, complicating not only the choice but the necessary tests. Fillers and additives can drastically change the resin's basic characteristics, blurring the line between commodity and engineering resins. Entirely new resins have been introduced with exoteric molecular structures. Therefore, resin suppliers now have many more sophisticated tests to determine which resin best suits a product or fabricating process. Material suppliers and developers routinely measure such complex properties as molecular weight and its distribution, stereochemistry, crystallinity and crystalline lattice geometry, and detailed fracture characteristics (see Chapter 2). They use complex, specialized tests such as gel permeation chromatography [12], wide- and narrow-angle xray diffraction, scanning electron microscopy, and high-temperature pressurized solvent reaction tests to develop new polymers and plastics applications. For the product designer, however, a few basic tests will help determine which plastic is best for a given product.

SPECIFICATIONS AND STANDARDS The industry specifications and standards are regularly updated to aid processors in controlling quality and to meet safety requirements, and thus they will prove useful to anyone who must choose tests and QC procedures. For example, the ASTM, UL, and DIN (see below) tests are among the most important ones. Organizations involved in preparing or coordinating specifications, regulations, and standards include the following: ASTM. American Society for Testing and Materials. UL. Underwriters Laboratories. DIN. Deutsches Instut, Normung, Germany. ACS. American Chemical Society. AMS. Aerospace Material Specification of the Society for Automotive Engineers (SAE). ANSI. American National Standards Institute. ASCE. American Society of Chemical Engineers.

TESTING AND QUALITY CONTROL 733

ASM. American Society of Metals. ASME. American Society of Mechanical Engineers. AWS. American Welding Society. BMI. Battele Memorial Institute. BSI. British Standards Institute. CPSC. Consumer Product Safety Commission. CSA. Canadian Standards Association. DOD. Department of Defense. DODISS. Department of Defense Index & Specifications & Standards. DOT. Department of Transportation. EIA. Electronic Industry Association. EPA. Environmental Protection Agency. FMRC. Factory Mutual Research Corporation. FDA. Food and Drug Administration. FMVSS. Federal Motor Vehicle Safety Standards. FTC. Federal Trade Commission. IAPMO. International Association of Plumbing & Mechanical Officials. IEC. International Electrotechnical Commission. IEEE. Institute of Electrical and Electronic Engineers. IFf. Industrial Fasteners Institute. IPC. Institute of Printed Circuits. ISA. Instrument Society of America. ISO. International Organization for Standardization. JIS. Japanese Industrial Standards. MIL-HDBK. Military Handbook. NADC. Naval Air Development. NACE. National Association of Corrosion Engineers. NAHB. National Association of Home Builders. NEMA. National Electrical Manufacturers' Association. NFPA. National Fire Protection Association. NIST. National Institute of Standards & Technology (previously the National Bureau of Standards). NIOSH. National Institute for Occupational Safety & Health. NIST. National Institute of Standard Testing. NPFC. Naval Publications & Forms Center. NSF. National Sanitation Foundation. OFR: Office of the Federal Register. OSHA. Occupational Safety & Health Administration. PLASTEC. Plastics Technical Evaluation Center of DOD. PPI. Plastics Pipe Institute of the Society of the Plastics Industry. QPL. Qualified Products List. SAE. Society of Automotive Engineers. SPE. Society of Plastics Engineers. SPI. Society of the Plastics Industry. STP. Special Technical Publications of the ASTM. TAPPI. Technical Association of the Pulp and Paper Industry.

These test procedures and standards are subject to change, so it is essential to keep up to date. It may be possible to obtain the latest issue on a specific test (such as a simple

734 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

tensile test or a molecular weight test) by contacting the organization that previously issued it. For example, the ASTM issues new annual standards that include all changes. The Annual Books of ASTM Standards contain more than seven thousand standards published in sixty-six volumes [695] that include different materials and products. There are basically four volumes on plastics: 08.01-Plastics I; 08.02-Plastics II; 08.03Plastics III, and 08.04-Plastic Pipe and Building Products. Other volumes include information on plastics and composites. The complete ASTM index lists under different categories the standards for the different products, types of tests (by environment, chemical resistance, etc.), a statistical analyses of different test data, and so on. The ASTM issues other useful information for the designer, included in its Special Technical Publications (STPs). The STPs are not concerned with ASTM standards. Some examples of STPs are STP 701, "Wear Tests for Plastics: Selection and Use," R. Bayer, ed., 1980, 106 pages; STP 736, "Physical Testing of Plastics," R. Evans, ed., 1981, 142 pages; STP 816, "Behavior of Polymeric Materials in Fire," E. L. Schaffer, ed., 1983, 121 pages; STP 846, "Quality Assurance of Polymer Materials and Products," Green, Miller and Turner, ed., 1985, 142· pages; and STP 936, "Instrumental Impact Testing of Plastics and Composite Materials," S. L. Kessler, ed. 376 pages. ORIENTATION AND WELD LINES

The preceding chapters, particularly Chapters 2, 3-5, and 7, have reviewed molecular orientation and weld lines, also called knit lines. This melt flow can be deliberate or undesirable. During processing, such as by injection molding and extrusion, orientation or weld lines can occur. Products can be drawn in one direction (uniaxially) or in opposed directions (biaxially). Weld lines can form during molding when hot melts meet in a cavity because of flow patterns caused by the cavity configuration or when there are two or more gates. With extrusion dies, such as those with "spiders" that hold a center metal core, as in certain pipe dies, the hot melt that is separated momentarily produces a weld line in the direction of the extrudate and machine direction. The result could be a poor bond at weld lines, dimensional changes, aesthetic damages, a reduction of mechanical properties, and other such conditions. To illustrate the influence of processing on mechanical properties, the test specimens in Figure 9-1 can be analyzed and related to what can happen in a fabricated product. It shows three sets of injection-molded specimens where the same plastic is processed in all specimens. There are three sets of similar specimens: a tensile one on top, a notched Izod impact one on the right side, and a flexural one on the left. The top set has a single gate for each specimen, the center set has double gates that are opposite each other for each specimen, and the bottom set has fan gates on the side of each specimen. The highest mechanical properties come with the top set of specimens, because of its melt orientation being in the most beneficial direction. The bottom set of specimens, with its flow direction being limited insofar as the test method is concerned, results in lower test data performance. With the double-gated specimens (the center set) weld lines develop in the critical testing area that usually results in this set's having the lowest performance of any of the specimens in this figure. Fabricating techniques can be used to reduce the potential problems in a product. However, the approach used in designing the product and its mold or die is most important to target, to eliminate unwanted orientation or weld lines. If potential problems exist, the design can incorporate the necessary changes, or make them later. This approach is no different from that of designing with other materials like steel, aluminum, or glass.

TESTING AND QUALITY CONTROL 735

Figures 9-1. Injection-molded test specimens that can be related to orientation and weld lines.

TYPES OF TESTS The details of procedures and the like for all types of tests-physical, mechanical, chemical, optical, insulation, electrical, and so on-are reviewed in the different specifications such as the ASTM standards. These procedures explain the reasons for a test, how to conduct it, how to determine the results, and sometimes provide information on long-term test results such as fatigue, creep, and so on. With all the available tests, confusion could exist in deciding which ones to conduct if experience in them does not exist. To eliminate confusion it is important to determine what could occur on a product that would be damaging. Here basic logic has to be used-examine the tests that incorporate and evaluate the potential damage. Examples of th~se tests are presented in Tables 9-1 to 9-7.

THERMOANAL YTICAL TESTS Thermoanalytical (TA) methods characterize a system in terms of the temperature dependency of its thermodynamic properties and the physiochemical reaction kinetics of TPs and TSs. The techniques reviewed here include only differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), thermomechanical analysis (TMA), and dy-

Table 9-1. ASTM Test Methods by Subject ASTM No.

Subject

Mechanical Testing D638 D 695 D 2344 D 3039 D 3518 D 732 D 785 D790 D 953 D 2344 D 3410

Tensile Properties of Plastics Compressive Properties of Rigid Plastics Apparent Horizontal Shear Strength of Reinforced Plastics by Short Beam Method Tensile Properties of Oriented Fiber Composites In-plane Shear Stress-Strain Response of Unidirectional Reinforced Plastics In-plane Shear Rockwell Hardness Flexural Properties of Plastics and Electrical Insulating Materials Bearing Strength Short Beam Shear Test for Compressive Properties of Oriented Fiber Composites

Fatigue D 3479 D 671

Tension-Tension Fatigue of Oriented Fiber Resin Matrix Composites Flexural Fatigue of Plastics by Constant-Amplitude-of-Force

Impact D 256 D 1822 D 3029

Impact Resistance of Plastics and Electrical Insulating Materials Tensile-Impact Energy to Break Plastics and Electrical Insulating Materials Impact Resistance of Rigid Plastic Sheeting or Parts by Means of Tup (Falling Weight)

Creep

D 2990 D 2991

Tensile, Compressive, and Flexural Creep and Creep-Rupture of Plastics Stress Relaxation of Plastics

Physical Properties D D D D D

570 792 1505 2734 3355

Water Absorption Specific Gravity and Density of Plastics by Displacement Density of Plastics by the Density-Gradient Technique Void Content of Reinforced Plastics Fiber Content of Undirectional Fiber/Polymer Composites

Thermal Properties D D D D

648 746 3417 3418

Deflection Temperature of Plastics under Flexural Load (HOT) Brittleness Temperature Heats of Fusion and Crystallization Transition Temperatures

Thermal Expansion D 696 E 228

Coefficient of Linear Thermal Expansion of Plastics Linear Thermal Expansion of Rigid Solids with a Vitreous Silica Dilatometer

Thermal Conductivity

Cll7

Steady-State Thermal Transmission Properties by Means of the Guarded Hot Plate

Electrical Properties D 149 D 257 D 495 D 150

736

Dielectric Breakdown Voltage and Dielectric Strength of Electrical Insulating Materials at Commercial Power Frequencies Electrical Resistance Arc Resistance A-C Loss Characteristics and Permittivity (Dielectric Constant) of Solid Electrical Insulating Materials (cont'd)

TESTING AND QUALITY CONTROL 737

Table 9-1. (Continued) ASTM No.

Subject

Wear Resistance D 673 D 1242

Mar Resistance of Plastics Resistance of Plastic Materials to Abrasion

Chemical Resistance C 581 D 543

Chemical Resistance of Thennosetting Resins Used in Glass Fiber Reinforced Structures Resistance of Plastics to Chemical Reagents

Flammability Tests D635 D 2843 D 2863 E662

Rate of Burning Smoke Density Oxygen Index Smoke Emission

Weatherability Tests D 1499

D 2565

D 4141 E 838 G 23 G 26 G 53

Operating Light- and Water-Exposure Apparatus (Carbon-Arc Type) for Exposure of Plastics Operating Xenon-Arc Type (Water-Cooled) Light- and Water-Exposure Apparatus for Exposure of Plastics Conducting Accelerated Outdoor Exposure Testing of Coatings Performing Accelerated Outdoor Weathering Using Concentrated Natural Sunlight Operating Light-Exposure Apparatus (Carbon-Arc Type) With and Without Water for Exposure of Nonmetallic Materials Operating Light-Exposure Apparatus (Xenon-Arc Type) With and Without Water for Exposure of Nonmetallic Materials Operating Light- and Water-Exposure Apparatus (Fluorescent UV-Condensation Type) for Exposure of Nonmetallic Materials

namic mechanical analysis (DMA). Others are also available that are useful in the processing plant (see Figs. 9-2 and 9-3).

Differential Scanning Calorimetry Differential scanning calorimetry (DSC) directly measures the heat flow to a sample as a function of temperature. A sample of the material weighing 5 to 10 g (18-36 oz.) is placed on a sample pan and heated in a time- and temperature-controlled manner. The temperature usually is increased linearly at a predetermined rate. The DSC method is used to determine specific heats (see Fig. 9-4), glass transition temperatures (see Fig 95), melting points (see Fig. 9-6), and melting profiles, percent crystallinity, degree of cure, purity, thermal properties of heat-seal packaging and hot-melt adhesives, effectiveness of plasticizers, effects of additives and fillers (see Fig. 9-7), and thermal history. DSC is also used to determine the percentage of crystallization (see Fig. 9-6). A significant consideration in using polyolefins is their susceptibility to crystallization. The molder needs to know how rapidly material crystallizes as it is cooled. A comparison of materials from different lots will indicate whether they will crystallize in the same manner under the same molding conditions. (Polyolefins are provided in both nucleated and nonnucleated grades. A nucleating agent is added to a material to increase the material's rate of crystallization, a factor bearing on the performance of parts molded from that material.)

Table 9-2. ASTM Test Methods in Alphabetical Order Property Apparent density Free flowing Nonpouring Bulk factor Specific gravity Density Mold shrinkage Flow temperature Rossi-Peakes Melt-flow rate, thermoplastics Molding index Dielectric constant; dissipation factor at 60 Hz, I kHz, I MHz Volume resistivity I min at 500 v Arc resistance high voltage, low current Dielectric strength Short time Step by step Tensile strength Elongation Modulus of elasticity, tensile Flexural strength Tangent modulus of elasticity, flexural Compressive strength Modulus of elasticity, compressive Impact resistance, Izod Notch sensitivity Hardness, durometer Hardness, Rockwell Haze Luminous transmittance Index of refraction Water absorption 24-hr. immersion Long-term immersion Brittleness temperature Coefficient of linear thermal expansion Deflection temperature Vicat softening point Flammability Oxygen index Deformation under load Rigid plastics Nonrigid plastics Dynamic mechanical properties Logarithmic decrement Elastic shear modulus Creep Creep rupture

738

ASTM Test Method D 1895 Method A Method B D 1895 D 792 Method A or B D 1505 D 955 D 569 Procedure A D 1238 D 731 D 150 D 257 D 495 Stainless-steel electrodes D 149 Sec. 6.1.1 Sec. 6.1.3 D 638 D 638 D 638 D790 D790 D 695 D 695 D 256 Method A D 256 Method D D 2240 D785 Procedure A D 1003 D 1003 D 542 D 570 Sec. 6.1 Sec. 6.4 D 746 D696 D 648 D 1525 D635 D 2863 D621 Method A Method B D2236

D2990 D2990

Table 9-3. ASTM Test Methods by Material Sleeving, Tubes, Sheets, and Rods 0229 0348 0349 0350 0709 0876 01202 01675 01710 o 2671 03394

Rigid sheet and plate materials Laminated tubes Laminated round rods Flexible treated sleeving Laminated thennosetting materials Nonrigid vinyl chloride polymer tubing Cellulose acetate sheet and film TFE-ftuorocarbon tubing TFE-ftuorocarbon rod Heat-shrinkable tubing Insulating board

Molding and Embedding Compounds 0700 0704

Phenolic molding compounds Melamine-fonnaldehyde molding compounds 0705 Urea-fonnaldehyde molding compounds 0729 Vinylidene chloride molding compounds D 1430 Polychlorotriftuoroethylene plastic (PCTFE)

01636 01674

Allyl molding compounds Polymerizable embedding compounds

Table 9-4. Military Specifications for Materials Specification Number

Material Description Thennoplastic Polysulfone Polyamide-imide Polyetheretherketone Polyether-imide Polyether sulfone

MIL-P-46120B MIL-P-46179A MIL-P-46183 MIL-P-46184 MIL-P-46185

Thennoset Resin, polyester, low-pressure laminating Resin, phenolic, laminating Resin-epoxy, low-pressure laminating Resin solution, silicone, low-pressure laminating Resin, polyimide, heat resistant, laminating

MIL-R-7575C MIL-R-9299C MIL-R-9300B MIL-R-25506C MIL-R-83330

Table 9-5. An Example of UL Standards for Materials Number

Title

UL94 UL 746A UL 746B UL 746C UL 7460 UL 746E

Tests for Flammability of Plastic Materials for Parts in Devices and Applications Polymeric Materials-Short Tenn Property Evaluations Polymeric Materials-Long Tenn Property Evaluations Polymeric Materials-Use in Electrical Equipment Evaluations Polymeric Materials-Fabricated Parts Polymeric Materials-Industrial Laminates, Filament Wound Tubing, Vulcanized Fibre, and Materials Used in Printed Wiring Board

739

740 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 9-6. Examples of Aerospace Material Specifications for Materials Number AMS 3628C AMS 36468 AMS 36560 AMS 3684A AMS 3709A AMS 3756

Tide Plastic Extrusion and Moldings, PolycaIbonate, Gene{lli Purpose Polychiorotriftuoroethylene (PCTFE) Sheet, Molded, Unplasticized Polytetraftuoroethylene Extrusions, Nonnal Strength, as Sintered, Radiographically Inspected Resin, Polyimide, Seal~-High Temperature Resistant, 315·C, or 600"F, Unfilled Syntactic Foam Tiles Polytetraftuoroethylene Moldings, Glass Fiber Filled 75 PTFE Resin, 25 Glass, as Sintered

DSC is a very useful technique for monitoring the level of antioxidant in, for example, polyolefins such as polypropylene. One of the materials most susceptible to oxidation, polypropylene experiences some brittleness and cracking, with the amount depending partly on the end use of the molded part. Antioxidants are added to extend the service life and protect the material during the molding operation, but they are sacrifically oxidized to protect the polymer during molding. Once the antioxidants have been depleted, the material is again vulnerable to oxidation. The end user of the part needs the antioxidant protection, however, and will not be well served if the antioxidant is used up during the molding operation. Therefore, the molder needs to ensure that sufficient antioxidant is in the raw material before processing and that enough antioxidant remains in the material after molding to meet the customer's needs.

Thermogravimetric Analysis This method measures the weight of a substance heated at a controlled rate as a function of time or temperature. To perform the test, a sample is hung from a balance and heated in the small furnace on the TGA unit according to a predetennined temperature program. As all materials ultimately decompose on heating, and the decomposition temperature is a characteristic property of each material, TGA is an excellent technique for the characterization and quality control of materials (see Figs. 9-8 and 9-9).

Table 9-7. ASTM D 4000-type Material Specifications Plastic Material

ASTM Standard

Phenolic Polyamide (nylon) Polycarbonate Polyoxymethylene (acetal) Polyphenylene sulfide Polypropylene Polystyrene Styrene-acrylonitrile Thennoplastic elastomer, ether-ester Thennoplastic polyester (general) Styrene-maleic anhydride Thennoplastic elastomer-styrenic Acrylonitrile-butadiene-styrene

04617 04066 03935 04181 04067 04101 04549 04203 04550 04507 04634 04774 04673

Report management terminal

Sample log-in terminal

LI MSnOOO

550

Program development term inal

Printer

Program development term inal

Ana lyuf;lIl disciplines dala stations

Gas chromatography

550

UV N IS sPeCtroseopy 3600

1:15

Liquid chromatography

Fluorescence sPeCtroscopy

3600

1:15

I nfrared SPeCtroscopy

3600

3600

Elemental analysis

Thermal analysIs 3600

3600

Manual data entry

Figure 9-2. Examples of plastics identification in a computer-aided chemistry laboratory. Courtesy Perkin-Elmer Corp.

DSC

:

TG

-01/

MEASURES

MEASURES WEIGHT CHANGES ATTENDING'

MEASURES DIMENSIONAL CHANGES ATTENDING.

• SPECIFIC HEAT • HEATS OF TRANSITION

• REACTION • DETERIORATION

• THERMAL EXPANSION • CONTRACTION

• ONSET TEMPERATURES • RATES OF REACTION

• DEHYDRATION • EVAPORATION

• SOFTENING (PENETRATION) • STRETCHING (EXTENSION)

• PURITY

Figure 9-3. Examples of thermal analysis test methods. Courtesy Perkin-Elmer Corp. 741

Range 5 mcal/sec Heating rate: 20°C/min Weight: 11.52 mg

..J----

PMMA \

~

i

(

i

150

:I:

II

,...-.-.~\

___ ."..T. = 110°C g

\

Endothermic

\

1\\

t

Basel i ne

I t...lr..-

~

150

T(OC)

Figure 9-4. The results of using DSC to determine the heat capacity of PMMA near the glass transition temperature (Tg ). Range: 5 mcal/sec Heating rate: 20°C/min Polycarbonate

t

Endothermic

60

Figure 9-5. DSC here identifies the glass transition temperature (Tg) for the amorphous plastics PC, PMMA, and PS, indicating a minimum temperature for processing these plastics. Range: 10 mcal/sec Heating rate: lOoC/min Weight: 7.1 mg

u

S!

t

=iij

E

Endothermic

!

.. 54.5 cal/gm ~ % Crystallinity = 68.4 cal/gm X 100% ~

~

=79.7%

... ~ '"

Area = ~Hf = 54.5 cal/gm

20

40

60

80

100

120

140

160

180

T (oC)

Figure 9-6. Here DSC determines the melting point and percent crystallinity of HDPE. 742

TESTING AND QUALITY CONTROL 743

Endothermic

u w en

LOPE

::J

«

I

t

/i~

Exothermic

~-----~ ~\--------

u :E

Melting of lOPE

LOPE +/ blowing agent

~----~~----~------~------~--~ 20 60 100 140 lao 220 T(Oeg C) Differential scanning calorimetry is used for determining the effects of additives and fillers from a process and quality-control point of view. The above graph characterizes LOPE foam.

Figure 9·7. DSC has related here the effects of additives and fillers that can be used in quality control, in this case for an LDPE foam. TGA fiberglass reinforced nylon

l00r-~~==~======~~==

2% Moisture

..

75 80% Nylon

~ 50

~

25

Heating rate: BO°C/min. Atmosphere: Air

O~

o

lB% Fiberglass

______~______~____~·~______~ 250 500 750 1000 T(OC)

Figure 9·8. The TGA process determines the amount of glass-fiber reinforcement in nylon.

The properties measured by TGA include thermal decomposition temperatures, relative thermal stability, chemical composition, and the effectiveness of flame retardants. TGA also is commonly used to determine the filler content of many thermoplastics. A typical application of TGA is its use in compositional analysis. For example, a particular polyethylene part contained carbon black and a mineral filler. The electrical properties were important in the use of this product and could be affected by the carbonblack content. TGA was used to determine the carbon-black and mineral-filler content for various lots, which were considered either acceptable or unacceptable. The samples

744 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Heating rate: 40· C.lmin. A tmosphere: air

1001----..........

75

50

i'*

25

0.5% residue from flame retardant

~ OL--_.....L-_----1.:~~...._ _ , /....._ _L

o

200

400

600

800

1000

T rC.) Figure 9-9. Characterizing the flame retardant in polypropylene by using TGA. were heated in nitrogen to volatilize the PE, leaving carbon black and a mineral-filler residue. The carbon content was then determined by switching to an air environment to bum off the carbon black. The weight loss was a direct measure of the carbon-black content.

Thermomechanical Analysis This system measures dimensional changes as a function of temperature. The dimensional behavior of a material can be determined precisely and rapidly with small samples in any form-powder, pellet, film, fiber, or as a molded part. The parameters measured by thermomechanical analysis are the coefficient of linear thermal expansion, the glasstransition temperature (see Figs. 9-10 and 9-11), softening characteristics, and the degree of cure. Other applications of TMA include the taking of compliance and modulus measurements and the determination of deflection temperature under load. Tensile-elongation properties and the melt index can be determined by using small samples such as those cut directly from a part. Part uniformity can be determ~ned by using samples taken from several areas of the molded part. Samples also can be taken from an area where failure has occurred or continues to occur. This permits comparisons of material properties in a failed area with properties measured either at an unfailed section or from a sample of new material. Samples may also be taken from within a material blend, to ensure that a uniform blend is being supplied. The results of such testing can be used either for evaluation of part failure or in the acceptance testing of incoming materials or parts. In basic mechanical testing, the mechanical characteristics that can be tested include expansion, penetration, extension, flexure, and compressive compliance. Photoelasticstress analysis allows the stress distribution to be visually displayed, and strain gauging allows the stress distribution to be approximated. Residual stress, also known as moldedin stress, can be measured by a variety of techniques.

Heating rate: 10 C./min. 0

Range: 0.1 mm.

t

Mode: expansion Load: zero

.~ c ! an

~188°C. o T

rc.)

I

I

I

I

I

50

100

150

200

250

Figure '·10. TMA is here used to determine the coefficient of expansion and the glasstransition tempemture (Tg) of an epoxy-graphite fiber composite.

Temperature, 0 F

500

395

375

355

415

435

r

/

E

E ~

Tgj

300

C> C

."

.&. U C

) "re.

o

~ 200 CD

E

is

100

180

175

OJ' CTE = SO.6 ( 185

-

195

224

V.-4"""

2OO"C

(Start of both runs)

"I

o

-

225 D C /

400

CD

475

455

10- I /K

205

215

221

CTE.5783 x 10-'/,

:.s--

'I1'

CTE='"

225

i,.-"

235

245

Temperature, DC

Figure

'·11. The TMA of oriented plastic's coefficient of thermal expansion. 745

746 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Dynamic Mechanical Analysis The dynamic mechanical analysis (DMA) procedure measures the viscoelastic properties (modulus and damping) of a material as functions of time and temperature. The material is deformed under a periodic resonant stress at a low rate of strain. Microprocessor datareduction techniques then provide graphical and tabular outputs of these properties as functions of time or temperature. The values determined, and the modulus and damping data, aid in establishing realistic structural design criteria. The speed of analysis provides high throughput and low labor cost. Precise temperature control can be used to simulate processing conditions. The breadth of material types range from rubbery to very high stiffness. The data obtained correlate both the structure-property and property-processing characteristics.

NONDESTRUCTIVE TESTING In the familiar form of testing known as destructive testing, the original configuration of a specimen is changed, distorted, or even destroyed for the sake of obtaining such information as the amount of force that the specimen can withstand before it exceeds its elastic limit and permanently distorts (usually called the yield strength) or the amount of force needed to break it (the tensile strength). The data collected in this instance are quantitative and could be used to design structural parts that would withstand a certain oscillating load or heavy traffic usage. However, one could not use the tested specimen in the part. One would have to use another specimen and hope that it would behave exactly the same as the one that was tested [10-12]. Nondestructive testing (NDT), on the other hand, examines a specimen without impairing its ultimate usefulness. It does not distort the test specimen's configuration but provides a different type of data. NDT allows suppositions about the shape, severity, extent, configuration, distribution, and location of such internal and subsurface defects as voids and pores, shrinkage, cracks, and the like. Most materials contain some flaws, which mayor may not be cause for concern. Flaws that grow under operating stresses can lead to structural or component failure, whereas other flaws may present no safety or operating hazards. Nondestructive evaluation provides a means for detecting, locating, and characterizing flaws in all types of materials, while the component or structure is in service if necessary, and often before the flaw is large enough to be detected by more conventional means. The following is a brief guide to nondestructive evaluation methods [12].

Radiography Radiography is the most frequently used nondestructive test method. X rays and gamma rays passing through a structure are absorbed distinctively by flaws or inconsistencies in the material so that cracks, voids, porosity, dimensional changes, and inclusions can be viewed on the resulting radiograph.

Infrared Spectroscopy Infrared (IR) spectroscopy records spectral absorptions in the infrared region using pyrolysis, transmission, and surface-reflectance techniques. Exposing the sample to light in the infrared range and recording the absorption pattern yield a "fingerprint" of the material.

TESTING AND QUALITY CONTROL 747

Infrared spectroscopy is used for identification of plastics and elastomers, polymer blends, additives, surface coatings, thickness (see Fig. 9-12), and chemical alteration of surfaces [10-12]. This is one of the most common analytical techniques used with plastics. The easy operation and availability of this type of equipment have contributed to its popularity. Although the infrared spectrum characterizes the entire molecule, certain groups of atoms give rise to absorption bands at or near the same frequency, regardless of the rest of the molecule's structure. The persistence of these characteristic absorption bands permits identification of specific atomic groupings within the molecular structure of a sample.

X-ray Spectroscopy This method identifies crystalline compounds by the characteristic X-ray spectrums produced when a sample is irradiated with a beam of sufficiently short-wave-Iength X radiation. Diffraction techniques produce a fingerprint of the atomic and molecular structure of a compound and are used for identification. Fluorescence techniques are used for quantitative elemental analysis.

Nuclear Magnetic Resonance Spectroscopy Proton magnetic resonance (NMR) spectroscopy characterizes compounds by the number, nature, and environment of the hydrogen atoms present in the molecule. Identification is possible because of the characteristic absorptions of radio-frequency radiation in a magIR·Source Chopper Molor

I

I

I

I I

I

Beam Selector

...I, II ~

_

I

Dividing Mirrors

Figure 9-12. This infrared system determines wall thickness by measuring how much IR radiation passes through the wall of the container.

748 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

netic field as a result of the magnetic properties of nuclei. NMR techniques are used to solve problems of crystallinity, polymer configuration, and chain structure.

Ultrasonics In ultrasonic testing the sound waves from a high-frequency ultrasonic transducer are beamed into a material. Discontinuities in the material interrupt the sound beam and reflect energy back to the transducer, providing data that can be used to detect and characterize flaws. When an electromagnetic field is introduced into an electrical conductor, eddy currents flow in the material. Any variations in material conductivity due to cracks, voids, or thickness changes can alter the path of the eddy current. Probes are used to detect the current movement and thus describe the flaws. When flaws or cracks grow, minute amounts of elastic energy are released that propagate in the material as an acoustic wave. Sensors placed on the surface of the material can detect these waves, providing information about the location and rate of flaw growth . • These principles form the basis for the acoustic emission test method. Although they have been commercially available for the past twenty-five years or so, ultrasonic detectors never really caught on as a diagnostic or maintenance tool. The biggest problem with ultrasonic detectors was their inability to produce measurements as accurately or consistently as could many competing devices used for nondestructive testing. The advent of microprocessing is dramatically improving the ability of ultrasonics to detect the wall thickness of metals and plastics, to determine particle dispersion in suspensions, and to detect potential leakage and faulty parts.

Liquid Penetrants The liquid penetrant method is used to identify surface flaws and cracks. Special lowviscosity fluids containing a dye will penetrate into a flaw or crack when placed on the surface of a part. When the surface is washed, the residual penetrants contained in the part reveal the presence of flaws.

Acoustics In acoustic holography, computer reconstruction provides a means for storing and integrating several holographic images. A reconstructed stored image is a three-dimensional picture that can be electronically rotated and viewed in any image plane. The image provides full characterization and detail of buried flaws.

Photoelastic Stress Analysis Photoelastic stress analysis helps a processor determine why a part broke and how to prevent similar failures in the future. Parts ranging in size from structural composites to tiny thermoplastic heart valves can all be tested easily. The test method is also a valuable tool for predicting where prototype parts may fail, such as in poorly designed features

TESTING AND QUALITY CONTROL 749

like comers, ribs, or holes. Likewise susceptible are parts with improper processing conditions, including poor mold design or an inconsistent mold temperature. Photoelastic analysis, one or several related testing techniques, is easy to use and usually a more economical and positive method than computer analysis. From the information it provides, the test can lead to better-designed, lower-cost products. Traditionally used to test the integrity of metal parts, photoelastic analysis is now being used to physically test thermoplastics as well as thermosets. For transparent plastics, the analysis can be made directly on the plastic. For nontransparent plastics, a transparent coating is used. Actual parts and representative models can be tested by a simple procedure. The former may be stressed under actual use conditions, whereas models are tested under simulated conditions. Although theoretical analytical methods such as finite element analysis (see Chapter 5) offer a chance to solve complex stress problems, there are many causes of strain in parts that cannot be reliably tested by these expensive computer-oriented techniques. For instance, strains arising from the assembly of components and those caused during processing are extremely difficult problems to analyze without physically testing the part. Photoelastic analysis is more than just another pretty experimental stress test (see Figs. 9-13 and 9-14). When examined under a polariscope, the colorful interference pattern can be used to survey stress distribution and the degree of strain. This analysis ultimately helps pinpoint which manufacturing function-design, processing conditions, or assembly techniques-led to part failure or might do so in the future. Interference patterns for coatings and models are analyzed in the same way. The photoelastic color sequence shows stress distribution in the part. In order of increasing stress the sequence is black, gray, yellow, red, blue-green, yellow, red, and green. Black and gray areas show low strains, whereas a continued repetition of red and green color bands indicates extremely high concentrations of stress. An area of uniform color is under a uniform stress. The degree of strain is indicated by a fringe order, which is simply a collection of black bands appearing in close proximity to each other between colors in the stress pattern. As the stress configuration increases, so does the number of black bands in a fringe order.

Vision Systems Inspections There are many opportunities for automatic vision systems in controlling quality and productivity of molded containers, such as inspection, gauging flaw detection, verification, counting, character reading, identification, sorting, robot guidance, location analysis, and adaptive control. The inspection covers the feed rate of materials into equipment, parison shape/drop distance, preform shape/neck geometry/molded-in specks or flaws, container shape/neck geometry/size, and so on. As an example, equipment is available to detect minute flaws at line speeds of up to 51,000 preforms/hr. (see Fig. 9-15) for an on-line Qualiplus system diagram to check PET injection blow mold (Chapter 7) bottle preforms' neck geometry and specks or flaws at rates up to 850/min. The system inspects each product with video cameras containing 180,000 photoelectric cells, using stroboscopic lighting to eliminate motion blur. The image is then analyzed by a gray-scale image processor that recognizes 256 different shades of gray and compares it to a standard image preset by the user. The system reportedly can detect flaws as small as 0.04 mm (oh of a sq. in.).

750 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 9-13. The photoelastic stress pattern for these two parts molded during the same production run shows that the processing conditions changed, resulting in failure of the top hinge (right), because of stress concentration.

~~~Aln

-!JUlY

Figure 9-14. Photoelasticimetry as an aid to the design of a load involved comparing the stress pattern with the preengineered load requirements. This example shows a spring element.

Microtoming and Optical Analysis Quality control of plastic molded parts can use optical techniques. In this procedure thin slices of the material are cut from the part and microscopically examined under polarized light transmitted through the sample. Study of the microstructure by this technique enables rapid examination of quality-affecting properties. This kind of approach can provide the molder with information for failure analysis, part and mold design, and processing optimization [12]. Thin sectioning and microscopy are old techniques, having been applied to biological samples for many years. Furthermore, metallurgists have used similar techniques in the microstructural analysis of metals to determine their physical and mechanical properties and to aid in failure analysis. Microtoming enables slices of plastic to be cut from opaque parts. These slices are so thin (under 30 !Lm) that light may be transmitted through them. The sample can then be analyzed under a microscope. Another useful technique is to use the microtome to slice down through a specimen until the specific level to be examined is reached. This method reveals a series of sequential levels, each smooth enough for viewing without need for polishing. The usual method is to cut, mount, and polish. When a series of cuts is needed,

TESTING AND QUALITY CONTROL 751

Preform

".-

Stroboscope

Flaw detector

Figure 9-15. This on-line inspection system from Qualipus (Stamford, Conn.) uses highresolution video cameras to detect flaws at speeds of 51,000 preforms/hr. These PET injectionblow-molded bottle preforms are checked for both neck geometry and molded-in specks or flaws at rates up to 850 preforms/min.

it becomes necessary to regrind and repolish. The microtome technique eliminates these tedious steps. An attractive aspect of the microtome analysis procedure is the speed with which results can be obtained. Generally, the sample can be rough cut from the product with a hacksaw and secured in a microtome vise, although in some cases it is necessary first to embed the sample in a block of epoxy. The slicing is a simple procedure. usually slices 8 to 15 fLm thick will be produced. These are mounted on a microscope slide using mounting cement and a cover plate. Polymers are often categorized as either amorphous or crystalline (see Chapter 2). Some can exist in either or both forms, and so it is common to discuss the degree of crystallinity when referring to the microstructure of a part. Often the effects of molding are clearly exhibited by observing the transition from the amorphous skin of a part to the crystalline core. Much of the analysis of plastics' microstructures is fairly straightforward. It is easy to tell whether you are dealing with a crystalline or an amorphous material by observing the sample using polarized light. Amorphous areas appear black, but crystalline areas can be clearly examined. The explanation for this effect is that in the case of crystalline polymers the molecules crystallize and fold together in a uniformly ordered manner,

752 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

whereas the amorphous polymers do not produce crystallites and occur randomly positioned. Thus, under polarized lighting crystalline materials exhibit multicolored patterns, whereas amorphous materials appear black. In this way the crystalline microstructure can be examined. Features of the crystalline polymers are readily discernible, whereas those of the amorphous polymers are not. It is interesting to notice differences in the crystalline structure found with different materials. A comparison was made between a nylon 6/6 micrograph and that produced from one of acetal homopolymer. The acetal has a characteristic structure very different from the square crystallites seen in the structure of the nylon. This difference is related to the propensity of nylon to supercool to a greater extent than most crystalline materials, whereas acetals crystallize much more rapidly. Optical techniques can be used for both quality control and failure analysis. Stress concentration can for a variety of reasons be a principal failure mode. One of these reasons relates to the use of contaminated or mixed materials, which may be caused by the presence of foreign materials or improper machine cleaning. Incorrect regrinding procedures, improper dry coloring methods, and the use of the wrong pigment are additional causes of this condition. Stress concentrations that resu.t from material contamination can be detected by observing the break area by reflected light. Particle size and dispersion can be found by examination under transmitted polarized light. Using polarized light it is possible with crystalline materials to identify residual stresses caused by incorrect melt flow and sharp comers emanating from poor part design. Impact, bending, and other physical stresses imparted to the part during service can also be identified. Generally, it is necessary to know whether or not you are dealing with a stressed-inservice part. Then it is possible to determine whether residual stresses resulted from service or occurred in molding. Stresses imposed in the molding process usually appear as regular patterns in the flow line direction, whereas those that result from imposed stresses created in service tend to exhibit semicircular arc-shaped configurations. Another source of stress involves the use of the microtome itself, since with some materials induced stresses are not difficult to create. These are usually found along the edges of the sample, where frequently the microstructure becomes smeared. Fortunately, stress caused by the microtome is not difficult to detect when viewing the specimen. Optical examination of the microstructure will determine whether or not correct mold temperatures were used in the production of parts from crystalline or partially crystalline polymers. With these thermoplastics the degree of crystallinity achieved depends on the temperature of the mold and the melt and the time that the pressure on the melt is maintained. In the case of acetal, the use of a cold mold results in fast dissipation of heat from the melt into the mold wall. Consequently, the threshold limit for the formation of crystallization nuclei is quickly reached and a skin is formed on the parts that has an amorphous appearance but is actually crystalline, although to a much lower extent than the spherulitic region that is formed below the skin. The thickness to the "amorphous" zone is dependent on the mold and melt temperatures and other factors. From the micrographs it is possible to judge the extent of what might be loosely described as the amorphous skin, the transcrystalline zone, and the spherulitic core. From their relative proportions it is not only possible to estimate the processing conditions that were employed to produce the parts, but also to predict part performance. Particles may remain unmelted within the molten mass of plastic. These particles inhibit the formation of crystallization nuclei. Since the thermal conductivity of plastics is poor, the length of

TESTING AND QUALITY CONTROL 753

time for the material to cool controls to a large extent the length of the molding cycle. Reducing melt temperature to shorten cycles and increase production reduces the quality of the product. Mold temperature, melt temperature, and screw operation all interact to influence part quality. Computer Image Processor An important aspect of the machine vision system, image processing, is performed by a computerized unit called the vision engine. Many of these units have been designed for specific types of analysis, as for example gauging or pattern recognition. Many applications are highly data intensive and, with certain types of image-capturing devices, could require a high order of computing power. Many of the applicationsspecialized processors use special techniques to simplify the analysis problem and reduce the data-processing load. For any given application, therefore, it is important to match the characteristics of the vision engine to the specific needs of the job. Machine vision systems can be classified as configurable, task-specific, or custom (dedicated). Configurable systems are basically nonspecialized systems that can be adapted for a specific application. They can be converted to other uses if the original application terminates. Task-specific equipment performs a single function, such as measuring dimensions. While it can accommodate to a variety of objects, measurement is all it can do. Some task-specific systems, however, use configurable vision engines. In these installations it is the peripheral equipment-the camera-mounting arrangement, lighting, part fixturing, and material-handling devices-that makes the system task specific. As with the generically configurable systems, the vision engines in the task-specific types can be used in other applications consistent with their performance envelope. Customized, dedicated systems are analogous to fixed automation: the system becomes obsolete when the application disappears. Only the individual components may be reused.

COMPUTER TESTING Although both designing and testing have sophisticated software to assist them, these areas have remained largely isolated from one another. However, increases in hardware power and availability of special software have now linked the two disciplines. Programs are available allowing design to take advantage of test data so that testing can benefit from design data. Software to link designing and testing comes from several sources. Some vendors of CAD software offer test data analysis modules so that information can be easily exchanged and compared. And suppliers of finite-element-analysis and modal-analysis software are creating ways to use the other's data in their programs. Modal-testing software typically will allow designers to test prototype changes in a computer, once the original prototyping is done. A computer solution could take as little as thirty seconds, whereas modifying an actual physical prototype might take as long as a week.

QUALITY AUDITING Some organizations have a documented quality assurance program that includes an audit program. A quality assurance program usually contains three tiers of documentation: the

754 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

quality assurance manual, system-level procedures, and instructions. The purpose of an audit program is to evaluate the existence and adequacy of the QA program and ensure that the manufacturer's operations are in compliance with it. Putting a program in writing does not ensure that it will be followed, nor does it, in and of itself, provide the feedback necessary to correct and update programs and processes. The audit fills both these gaps. By monitoring product, process, and system, and by rating performance against a predetermined scale the auditor determines the need for corrective measures.

RELIABILITY AND QUALITY CONTROL Reliability is the probability that a product will perform satisfactorily for a specified time under the stated operating conditions. This implies probability, duration, and a specification of what is considered satisfactory performance, which necessarily incorporates the use environment. By comparison, quality control is the determination, by measurements, that production materials and processes are within the specified tolerances. Reliability is a design function, quality control a manufacturing function. Both are essential to satisfactory product performance. Basic to any design is an accurate understanding of what is desired. Because cost increases with reliability, good design and engineering mandate that only the necessary level of reliability be specified. This requires rigorous analysis of user needs so that a quantitative performance specification can be developed. Performance specifications state what is needed to satisfy the system requirements. They include the environment of use, performance requirements, and the stipulated reliability.

FAILURE ANALYSIS Product design or the establishing of QC control bases involves a postmortem examination of the failure itself, utilizing every means at the designer's disposal. In such situations the designer must be systematic in exploring and evaluating all the possibilities. Failure analysis procedures for unreinforced metal and plastic parts have been developed since at least the early 1940s. Their predictable behavior in failure modes thus makes the establishment of cause fairly straightforward. It is generally accepted that the sources of fracture can be grouped into three basic categories: 1) design deficiencies or material misapplications includes poor assessment of service conditions, selection of inadequate material, poor design details, oversimplification of load and load paths, or inadequate attention to environmental stresses; 2) manufacturing or process discrepancies occur in spite of the fact that fabricating processes should be controlled by precise specifications, but out-of-compliance conditions can occur. Typically, problems could occur because of incomplete cure, voids, use of incorrect materials, contaminants, or cure at improper temperature; and 3) unexpected service conditions that refer to load; environmental and damage conditions beyond those reasonably anticipated in the design. RP/composites generally do not fail in the same way. The methodology for analyzing failure in composite parts used in structural applications can be rather complex. For one thing, fracture may occur from a multitude of diverse causes, and more than one cause may contribute to the failure. However, procedures have continually been developed and updated for failure analysis since the 1940s [1, 5-14, 27-34, 62-68].

TESTING AND QUALITY CONTROL 755

Table 9-8. Example of ASTM Test Methods Applicable to RP/Composites ASTM 0494, "Acetone Extraction of Phenolic Molded or Laminated Products" ASTM 01867, "Copper-Clad Thermosetting Laminates for Printed Wiring" ASTM 01823, "Etching and Cleaning Copper-Clad Electrical Insulating Materials and Thermosetting Laminates for Electrical Testing" ASTM 02861, "Flexible Composites of Copper Foil with Dielectric Film or Treated Fabrics" ASTM D709, "Laminated Thermosetting Materials" ASTM D1532, "Polyester Glass-Mat Sheet Laminate" ASTM C582, "Reinforced Plastic Laminates for Self-Supporting Structures for Use in a Chemical Environment" ASTM D3039, "Tensile Properties of Oriented Fiber Composites" ASTM D2408, "Woven Glass Fabric, Cleaned and After-Finished with Amino-Silane Type Finishes, for Plastic Laminates" ASTM D241O, "Woven Glass Fabric, Cleaned and After-Finished with Chrome Complexes, for Plastic Laminates" ASTM D2660, "Woven Glass Fabric, Cleaned and After-Finished with Acrylic"

SELECTING TESTS

As explained, there is no single set of rules designating what tests are to be conducted. Tests depend on required performance (as explained in the beginning of this chapter). Guides to selection of test methods have been reviewed (see Tables 9-1 to 9-7). Table 9-8 reviews tests for RP/composites. QUALITY AND CONTROL

Quality in products begins with good design, which in tum allows for simplifying selection of tests. Unfortunately, so often product design projects start with little appreciation for a good problem statement, an identification of requirements and objectives, and a reasonable schedule that includes all company functions involved. Most of all, what is usually lacking is a complete understanding of the end-user of the product, system, and/or environmental design being considered. The FALLO approach, as summarized in Figure 1-3, includes this type of quality and control.

Chapter 10

COMPUTER-AIDED DESIGN

With the help of computer technology, the designer has the capability of managing the enormous number of design, material, and process applications available when designing with plastics and reinforced plastics (composites). In the past, the designer more often than not had to limit the material, process, and design choices to those that were familiar. However, the computer now provides a means of developing a number of design concepts, quickly evaluating them against established functional and performance criteria in a wide range of materials and processes, and making the necessary design modifications. This approach provides finished drawings and specifications that can be furnished with increased accuracy and confidence, usually in less time than before. In addition, the computer provides many other advantages. If one dimension is changed that affects many others, all the dimensions will be changed automatically by the computer at the same time the initial change is made. It is of course possible that entering the initial three-dimensional geometry for various design concepts into a computer can be more time consuming than developing concept sketches with conventional long-hand methods. However, subsequent revisions and analyses of that geometry can be performed many times faster on computer than manually. It is the concept of database sharing that has made Computer-Aided Design (CAD) a tremendous time- and work-saving tool for the designer. In the past, the evaluation of design concepts to determine their viability was a rather long, tedious task. Now, by utilizing a CAD system the three-dimensional geometry initially entered can be redefined, manipulated, and edited quickly. Other actions can also be taken easily and quickly, such as applying finite element analysis and testing and making economical analyses for various plastic materials and processes [1-13, 56-66, 374-84, 414, 419-31, 751-806]. For example, three-dimensional wire-frame models can be constructed to help visualize the product and serve as a framework for other modeling programs. (More .details on wire frame and other modeling techniques are presented later in this chapter.) These wireframe models, which are constructed in three dimensions, can normally be viewed in orthographic, isometric, or perspective views, often at the same time. Changing one view alters all the other ones. Wire frames are easily modified, and many design variations can be constructed in a relatively short time. With a wire-frame model a designer can blend one shape into another while holding specific parameters. Thus, a wall section or volume of a product can be held constant while the shape changes incrementally over a given number of repetitions. This CAD approach often yields a potential configuration that would not have been possible to arrive 757

758 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

at with time-consuming manual methods. This capability of repositioning products or the parts of an assembly is a great aid in refining and checking certain operations and functional uses of the product. Parts can be lined up on a common center and clearances be edited and verified from the available databases (see Appendix D). Many hours of cumbersome materials selection and cumbersome calculations can thus be eliminated. Moving parts can be incrementally or dynamically rotated about an axis and interferences be checked at each position. Color technology is a further benefit to the checking process, particularly in mechanical load profiles, the plastic melt temperature during processing, mechanical or geometric profiles, and others. For example, interferences can be more readily found when each part or position is in a different color. Another important capability of the CAD system is in the manufacturing of molds or dies to produce the designed product. A database can be copied for numerically controlled tool paths already scaled for shrink factors, to enable the mold or die designer to prepare a preliminary or production mold or die design.

MOLD DESIGN Because of the complexity of examining all the possible variations of technique, style, and approach in plastic-product design, let us focus on an illustration of how to apply Computer-Aided Design/Computer-Aided Manufacturing (CAD/CAM) to the product mold design process. This illustration is pertinent to producing a final designed product that is to be injection molded. The approach here can be used as a guide in fabricating designed products by other processes that use some type of tool such as extrusion dies, blow molds, compression molds, and others. This illustration assumes that the product design has already been established and optimized for structural integrity, wall thickness, and cosmetic appeal. The task at hand is then to produce an optimum mold for a given level of production volume and to produce parts at an optimum quality level. CAD/CAM can be applied by the mold design and production teams in order to deliver successfully the molded product at the lowest possible cost and with the shortest possible lead time. In order to provide a better understanding of how to utilize fully the unique features of CAD/CAM systems in the mold design process, this section discusses the differences between a manual method of designing molds and the corresponding method using CAD/CAM technology. The flow charts showing this comparison are provided in Figures 10-1 and 10-2. Close examination of these charts indicates that a considerable difference exists in both the design methodology and the engineering emphasis for molds designed with and without the benefits of CAD/CAM technology. In general, molds that are manually designed depend heavily on the experience of the mold designer and have a strong emphasis on paper as the communications and driver medium for subsequent manufacturing operations. These molds also rely on a number of iterations after the mold is built to achieve the product function and fine tune the performance of the mold. In the CAD/CAM approach, the product model database serves as the communications and manufacturing driver medium, analysis programs augment the mold designer's experience, iteration is conducted more in the design phase prior to cutting cavity steel, and there are generally fewer molding trials.

CAD/CAM for Mold Design Without going further into the details of how CAD/CAM software is being applied to mold-design applications, it is appropriate to discuss what benefits can be derived from

COMPUTER-AIDED DESIGN 759

--

Figure 10-1. A flowchart of conventional moldmaking activities.

successful application of the technology. Those firms and individuals who have invested time and money in learning and implementing CAD/CAM systems have identified the following primary benefits of their use: Productivity improvement Quality enhancement Turnaround time improvements More effective utilization of scarce resources Each of these primary benefits is now further described in detail [10-12].

Productivity Many types of productivity benefits have been documented and verified that result from CAD/CAM being effectively applied to plastic design tasks. The first benefit is an actual

760 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

A£VI8E

Figure 10-2. A flowchart of moldmaking activities using CAD/CAM.

increase in the productivity of the mold design itself. CAD/CAM software provides the basic tools to provide productivity increases from a low ratio of 2 to 1 up to firms with specific applications demonstrating lO-to-1 increases or more. Much of the achievable benefit depends on the degree of commonality or standardization present in the type of molds or parts that a particular firm designs. In this respect the CAD/CAM technique known as group technology can be of great benefit in the capture of similar part designs. This technique shows great promise as a tool to enhance the overall productivity of the design process. Other tools such as finite element modeling can help to reduce the amount of prototyping and testing required to design plastic products successfully. Other software capabilities such as the creation of shaded images allow the aesthetic appeal of a product

COMPUTER-AIDED DESIGN 761

to be evaluated without requiring the construction of physical models or prototype molded parts. A second productivity benefit is obtained in mold manufacture. CAD/CAM technology facilitates the use of numerically controlled (NC) machining technology in the fabrication of the molds themselves. As such, much more of the mold can be cut in single setups on a single machining center. In tum, this reduces the number and complexity of manual setup operations so that more time is spent "making chips." Additionally, the ability to build three-dimensional product models in the database and to automatically generate tool paths from these models reduces the amount of effort spent in defining section views, calculating pickup points, and the like. The availability of graphic tool-path generation and verification eliminates the manual data-entry steps and proofing cuts previously required on the shop floor. The availability of communications software for the direct numerical control (DNC) of machine tools eliminates the time-consuming and error-prone process of punching paper tape to drive the machine tools in the shop. An additional productivity benefit is a reduction in the amount of time required for mold start-up, a substantial cost element of running a molding plant. Quite often, management either overlooks or chooses to accept this cost element as a fact of life. In fact, the debug and fine-tuning time for a mold can be greatly reduced by effective utilization of CAD/CAM. It has been demonstrated that analysis programs can be of great assistance in enhancing the quality of first-time mold designs. Additionally, they provide a great means to define the molding "process window," or extremes of acceptable process conditions, before the mold is put on the production floor. These factors result in a better use of molding press time and less disruption of production activities in the molding plant. Molders have reported differences of as much as 10 to 1 in start-up time, which they have attributed to the better accuracy and quality of molds designed and constructed using CAD/CAM. The last great area of productivity improvement is the one with the greatest financial payback-the part production environment. Integration of analysis tools into the design process provides much of this production benefit. One major plastics product manufacturer reported a 20 percent improvement in the cooling time requirements for 80 percent of the molds that were analyzed. High-production molds have in fact yielded cost savings in excess of $100,000 per year by the application of cooling analysis techniques. Flowanalysis techniques provide the benefits of being able to safely design runner systems of smaller diameters. They also allow the design of unbalanced runners while avoiding the problems of overpacking, and at the same time provide molds with higher production yields. All these results yield substantial reductions in material usage and manufacturing costs. Economic analysis tools allow true optimization of the molding operation within a real-world operating environment. Trade-offs can thus be evaluated between real-world variables such as the number of cavities to build versus press size and capability, product tolerance requirements, and product quality requirements. The result is a minimization of the total cost per piece at a given production volume level.

Quality The quality benefits of CAD/CAM are perhaps the most underrated of all the benefits. Drawings produced by CAD/CAM systems from three-dimensional models have been shown to be of a consistently higher quality than those produced manUally. Dimensions are totally defined by the geometry in the database, and as such are never incorrect. Tolerance stack-ups and other tolerance-related issues can also be calculated by the

762 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

CAD/CAM system, resulting in far fewer tolerancing errors. Complex geometries such as sculptured surfaces and blending radii are totally described in the database and thus not subject to ambiguities in drawing interpretation. Another of the means of enhancing quality is in the mold designs and molds themselves, by reducing the number of errors caused by redefinition of the product geometry. This geometry is transferred first to the shrink-corrected geometry, then to the cavity and core details, and finally to the machined components. Each of these steps is driven directly from the original product model, with little possibility for error other than that which is operator induced. Even the probability of operator error is reduced, as many of the tedious tasks of redefinition have been eliminated, with the result that the operator does a more consistent job. The accuracy of the mold cavity and core with respect to the product model is also a great aid in yielding quality enhancements. By utilizing NC in moldmaking and eliminating dependence on second- and third-generation tool-making aids such as die models, the conformance of the mold core and cavity to the product requirements is virtually guaranteed. The limits of accuracy obtainable are often those of the NC machine tool plus the error created by hand finishing operations. The accuracy of parts is another quality aspect to be considered. Many of the analysis packages promote a better understanding of molding process parameters and the interrelationships between process variables. This contributes to a better ability to control previously mysterious phenomena such as warpage by better process control and cooling system design. The final quality benefit is that of reproducibility. This includes reproducibility from cavity to cavity and mold to mold. Additionally, if cavities are severely damaged and require newly built cavities from new shop setups, the use of NC will result in cavities closer to the originals than those constructed manually. The ultimate result of the previously mentioned benefits is more consistent product quality.

Turnaround Time Companies that are designing and using plastic products in today's fiercely competitive business environment are sorely aware of the time it takes to design, manufacture, and debug the tooling for injection-molded products. The ability of CAD/CAM to speed up the plastic part and mold design process, the mold manufacturing process, and the startup and debugging process makes it a great aid in solving these age-old lead-time problems. Teamed with other prototype moldmaking techniques, CAD/CAM is helping to revolutionize the plastics product development cycle. Plastic products are now designed, analyzed, and evaluated for both technical and economic feasibility entirely without paper or physical models. Prototype molds, when required, are now turned out in one fourth to one half the time of their production counterparts, resulting in greater degrees of product design quality from the testing of molded rather than machined prototypes. Additionally, lead times for production molds are being pared down by many of the same techniques, resulting in more effective utilization of assets such as cash and inventory.

Resource Utilization The final benefit of CAD/CAM for moldmaking· is that it allows the effe~tive utilization of scarce resources, especially skilled labor. It is well known that while the usage of plastics materials is increasing at a steady rate, the population of skilled moldmakers is

COMPUTER-AIDED DESIGN 763

on the decline. This creates a problem for those firms intending to continue manufacturing injection molds, either as a primary business or as a part of a larger business. Since sociological changes are reducing the skilled moldmaker labor pool available for employment, a new means to continue mold production without dependence on that skilled labor base must be developed. CAD/CAM provides the opportunity to reduce the elements of moldmaking that require high skill levels, thereby enabling skilled moldmakers to be used on tasks that cannot be accomplished by any other means. It is possible, and indeed probable, that the use of CAD/CAM for the repetitive and routine tasks of moldmaking will enhance the quality of work life for skilled moldmakers by providing more challenge and job satisfaction. CAD/CAM MODELING

Having seen the major benefits of CAD/CAM for plastic design applications, it is appropriate to discuss the technology by which these benefits may be gained. In the normal engineering environment, products and mold designs are usually presented as a series of orthographic projections in the form of engineering drawings. These projections allow for representing a three-dimensional world as if it had been photographed and reduced to a planar image. To define on a drawing that we are working with orthographic views of the product lets the mind synthesize a three-dimensional image of the product. In many cases, however, orthographic images fall short of the mark in their ability to describe a complex geometry. This necessitates the creation of auxiliary and section views of the product in order to communicate the geometric description accurately. The methodology behind producing product designs and mold designs via CAD/CAM technology is analogous to that in the normal method, except that the description of the product or mold is contained in a product model database. The product model differs from a drawing in that the model is generally a three-dimensional representation of the real-world object. This representation can completely describe that object without auxiliary views or supplementary information. In this sense the product model fulfills the requirements of a classic definition of a model. A model has been defined as a representation to show the structure of something; an image to be reproduced in more durable material; a pattern or mode of structure or formation. By these definitions one can anticipate that while moldmaking has always relied on the production of wood patterns, die models, or

other tool-making aids to be able to accurately reproduce the geometry defined via paper media, CAD/CAM technology is eliminating many of those requirements. It is therefore appropriate to include in this discussion on CAD/CAM how the concept of a product model applies to the area of mold design. It is best to begin with a basic discussion of the types of modeling tools available today and how these can be used to construct the databases used in product and mold designs. In order to construct an effective database it is important to note that it must integrate and consolidate all the information requirements used by engineering and manufacturing operations subsequent to the initial product or mold design. This is a sizable task, which is why database integrity and completeness should be a major concern to CAD/CAM system purchasers and users. Consider a few of the many application areas using such a database: Engineering functions Design Drafting

764 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Analysis Technical publications Manufacturing functions Machine control applications Robotic applications Tool and mold design applications Quality management applications Communication with other business functions Materials management (bills of material) Cost estimating A graphic representation of these relationships of sharing a common database is depicted in Figures 10-3 and 10-4. As can be imagined, in order to drive the number of applications areas previously discussed, an extremely robust database must be constructed and maintained. The comprehensiveness of the database is one of the primary differences between CAD/CAM systems today. Each application that reuses this database recognizes the benefits of improved productivity and qUality. This is attributable to the fact that the already existing information in the product model does not need to be redefined, which would present the chance of introducing errors. Many will probably be asking at this point, "Aren't there a number of 2-D CAD/CAM systems on the market today, and if so why aren't they being discussed?" The answer is that certainly systems are being sold that are called CAD/CAM systems and basically satisfy the need for automating the drafting function. However, they also fall short of being able to satisfy the information requirements of the previously listed functions. This applies particularly to the area of mold design, where the problems of clearly representing complex surface geometries and other spatial relationships lead to the conclusion that 2D systems are not adequate to reap fully all the benefits that CAD/CAM has to offer. The three-dimensional product model provides the means by which many of these benefits can be obtained. The 3-D model clearly depicts all the spatial relationships between the items of interest in the product model. Terms such as blend to suit, which were prevalent on the drawings of the past, are now replaced by exact and reproducible mathematical descriptions of the surface curvatures desired. This has led to a new era in

TH CAD/CA

DATABASE

ENGI EERI G APPLICATIONS

rFigure 10-3. The engineering functions that use a product-model database.

COMPUTER-AIDED DESIGN 765

Figure 10.4. The manufacturing functions using the same product-model database as the engineering functions.

quality and dependability of product representation made possible now by these 3-0 modeling approaches.

Modeling Methods Applied to Part and Mold Design These are currently three predominant methods of building 3-0 models of products and storing them in a database, each with its own costs and benefits. The following overview of each of these modeling techniques summarizes the benefits and costs associated with each technique. It is interesting to note that because major changes in the power of computing hardware and software are occurring rapidly, these costs and benefits are appropriate only to today's "snapshot" of the state of the art in CAD/CAM. As this evolutionary process continues, it wiU bring other benefits not yet imaginable, along with a hardware base whose cost-to-performance ratio makes realization of these benefits easier to achieve. The three major methods of modeling in common use today are wire frame modeling, surface modeling, and solids modeling. Each of these modeling methods and its potential

application to mold design will be discussed in detail [11,12]. Wire-Frame Modeling

Wire-frame modeling is the simplest of the CAD/CAM modeling methods. With this technique the product model is constructed as a collection of geometric entities. Typical ones used in wire-frame construction are points, lines, arcs, b-splines, strings, and the like. The wire-frame method of modeling is similar to orthographic projection drawing in that each of the lines and other entities represents edges of the physical surfaces of the product. Unlike orthographic projection, however, the three-dimensional database allows the graphic display devices (terminals) of the CAD/CAM system, such as the one depicted in Figure 10-5 to automatically display isometric views of the product from any perspective the user desires, thus communicating the three-dimensional nature of the product. In these views of the object, lines and such are connected at their ends, thus portraying the appearance of a frame built of wire elements, hence the term wire-frame modeling.

766 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 10-5. A typical CAD/CAM system's components, including the workshop (foreground), central processing (background. right), and plotter (background. left).

The simplicity of this modeling method also implies simplicity in the database, thus resulting in superior system performance. The computing power required and storage requirements are minimal. Manipulation of the geometry to display different views is relatively fast and easy, owing to the small number of data elements that require mathematical transposition. The major disadvantage of this approach is that all the entities to describe the geometry are simultaneously displayed. For products with complicated geometries, this may result in many lines being visible on the screen at the same time, often resulting in a complex, confusing screen image. Hidden lines cannot be removed unless a surface or plane bounded by a the wire-frame geometry is described to the computer. This then allows it to determine which entities are behind that surface and therefore invisible. Additional CAD/CAM features and techniques are described later in this chapter to assist in organizing wife-frame entity information more effectively. The wire-frame modeling approach is best applied to rectilinear objects without complex surface geometries. This applies to a reasonable portion of plastic products, especially those that are more functional than aesthetic in nature. Such products might include gears, cams, brackets, and other nonappearance-type items. In mold design this applies to objects such as mold bases, cavity blocks, and leader pins. The surfaces of these objects are flat or circular and generally need no subsequent manufacturing operations. Hence, addition of surface information to the database description of these items only adds to storage space requirements, without producing any tangible benefit. Wire-frame modeling is also useful for describing mold features that require relatively simple manufacturing operations, such as drilling and pocket and profile machining. Machining of this type is referred to as "two and one-half axis" machining. Machine cuts are made at positions or contours relative to the X and Y axes and at a fixed position relative to the Z axis. Since a contour relative to the Z axis is not being machined, it is counted as only a half axis. Wire-frame geometry suffices to describe clearly the cutter

COMPUTER-AIDED DESIGN 767

Figure 10-6. A mold base modeled in wire-frame geometry.

path boundary as a series of two-dimensional curves. The cutter's depth can be established from the distance between two curves known to be parallel (e.g., the top and bottom of a pocket). An example of wire-frame geometry is shown in Figure 10-6, where a standard mold base is described by a collection of wire-frame entities. Since these entities are not ambiguous, positions such as the ends of the lines or the centers of the circles may later be used in the construction of additional geometry or for machining operations. Figures 10-7 through 10-12 provide examples of more-complex wire-frame geometries [10-12]. Surface Modeling

The next modeling method is surface modeling, a technique that can describe the total outer boundary of a part. Surface modeling differs from wire-frame modeling in that not only does it allow one to describe the edges of the geometry to be represented, but also all of the faces. A surface model is used to represent all the outside faces or boundaries of the product. By using such a model ambiguities in the wire-frame approach are immediately eliminated. With a surface model every point on a product's surface is definable, either by explicit coordinates of a key point or by interpolation using an explicit set of parametric equations to specify the points between the key points. Most of today's surface-modeling methods use a number of key points at prescribed intervals to bound or define the surface. Through these points a curve is fitted, in the same way that a french curve is used on a drawing board to fit a curve through a number of random points. As the CAD/CAM system must use a precise set of mathematical equations to define such a curve, this method makes every point along the length of the curve definable. Imagining these curves as a type of spatial grid, by understanding the mathematical base upon which the set of curves was built, makes it possible to identify uniquely any point on the surface by interpolation using the same set of mathematics used in the construction of the curves

Figure 10-7. An example of designing a bottle with three-dimensional OAD analysis. Infonnation on the bottle's design is stored in the CAD system's computer and with the applicable software can immediately provide all the types of infonnation needed such as its volume, weight, and head space.

Figure 10-8. For evaluating the depth of a fonn, 3-D wire frames tend to communicate more realistically than artists' renderings or traditional 2-D drawings. The 3-D wire-fonn system can be converted into two- or four-dimensional view by just touching appropriate computer keys.

768

------- -----

---

- ----.::.:!:.HE AX 15"

"Tt£ CURVE"

Figure 10·9. An example of using two lines ("the axis" and "the curve") to create a 3-D wire frame.

1IIIIIIIIIIIIrlllllllllllllllili Figure 10·10. In this example a wire fonn is converted to a four-perspective orthographic/isometric drawing.

Figure 10-11. An example of a complex shoe shape depicted in a 3-D wire-frame system. 769

770 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 10-12. An example of developing a wire-frame system to depict the image in the mold cavity. It is generated by CAD/CAM using data on the dimensions of a bottle and includes other information stored in the data bank (or incorporated at the time the image is displayed): the shrinkage characteristics of the plastic based on how it is affected by the blow-molding process, the temperature/pressure/time cycle of the melt, the die or mold materials of construction, and so on. themselves. Surface modeling is therefore particularly appropriate when complex 3-D geometries need to be described without any ambiguities. Many plastic products fall into a category of this type. Enclosures, covers, and consumer items often require not only a basic function to be provided but considerable aesthetic appeal to enhance market attractiveness. CAD/CAM and surface modeling provide the means not only to define a surface but also to evaluate the aesthetic appeal of the product and directly control the finished product by an unambiguous description of all surface curvatures and radii. The practical implications of surface modeling in mold manufacturing are manifold. First, since every point on the surface may be explicitly defined, it is possible to create a tbree- to five-axis tool path for a milling machine, to follow the surface contour. This is of great practical significance in the area of mold design. Because it can generate tbreeaxis contour tool paths, surface modeling is particularly appropriate for describing the geometries of cavities and cores within the mold. Other benefits include the ability to calculate angles of incidence and angles of refraction for light rays, which allow for computing shaded images. These in tum greatly enhance the ability of humans to comprehend the surface geometry. An example of shaded imagery is illustrated in Figures 10-13 and 10-14, which show a surface-modeled plastic molding (a cover for an automobile ignition computer) as a surface model. Figure 10-13 shows the primary construction of the surfaces, which are displayed for speed and efficiency reasons as sort of a wireframe representation. Stretching between the wires, however, like canvas on an airplane wing, is a mathematically defined surface geometry. The existence of this geometry allows us to display the same product in a shaded image, as in Figure 10-14. The shaded image is preferable to the wire frame in terms of ease of comprehension of the geometry of the product as well as its aesthetic appeal. A moment's reflection about the concept of a surface should make it clear that it has

COMPUTER-AIDED DESIGN 771

Figure 10-13. A plastic molding, an automobile ignition computer housing, and modeled-in surfaces. no inside or outside. It is merely a spatial curve of zero thickness. This is another primary reason why surface is so important in mold design. Assuming that the product designer does an effective job of defining a product model using the surface-modeling technique, the CAD/CAM system can mathematically adjust this surface to account for the effects of a plastic material's shrinkage. The resultant surface model (after accounting for shrinkage) is both the surface of the part and the surface of the mold. Since this surface is totally described at every point, we can select a distance between successive cuts from a ball type and mill, then calculate a tool path that always keeps the cutter tangent to the surface. The final benefit of the surface model is that it allows mass-related properties to be computed for the product model (e.g. volume, surface area, and moments of inertia), and also allows section views to be automatically generated. This is due to the fact that the intersection of a cutting plane with the surface can be calculated with certainty, resulting in another curve that becomes the section view. Because of these benefits, surface modeling is quickly becoming the workhorse of the CAD/CAM field for describing geometries typical of those of molded products. There is a cost that accompanies the benefits of surface modeling, however. The system now requires far more computing resources than those required by wire-frame modeling techniques, in the form both of computing power and database storage space. These needs are created by the necessity for more complex computations for such operations as automatically intersecting, trimming, and filleting entire surfaces. The state of the art allows users to do this type of construction interactively, using the wire-frame types of displays shown in Figure 10-13. As computing power and speed increase, however, it will become possible to do surface manipulation with the shaded image types of displays shown in Figure 10-14, which will greatly enhance the user friendliness of CAD/CAM systems.

Figure 10-14. A shaded-image representation of the same ignition housing shown in Figure 10-13.

772 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Solids Modeling The concept of solids modeling has become the ultimate in the modeling of real objects today. The solids model takes the concept of the surface model one step further in that it assures that the product being modeled is valid and realizable. A solids model of the product may be created in a variety of methods, such as the Boolean addition and subtraction of primitive shapes. These geometric primitives would include such objects as cubes, spheres, and cylinders. Other solids-modeling techniques include the sweeping of 2-D profiles through space, and even the possibility of "sewing" together the edges of surface models. Through a solids model the mass and boundaries of a product are represented in totally defined terms. An example of solids modeling is shown in Figure 10-15, in which not only the geometry and surface boundaries are known, but the mass properties of the object are instantly available upon the creation of the model. In surface modeling the system cannot distinguish between the inside and the outside of the product being modeled. In solids modeling, the system recognizes the difference, and during construction it can avoid possible inconsistencies using surface mathematics. One example of such inconsistencies is the renowned Mobius strip, in which a surface turns over on itself. An additional benefit of solids modeling is that the mass properties of the product are generally available immediately from the database. The key concept of the solids model states that every point in space can be determined to be either inside or outside the object being modeled. Whereas surface models define every point on a surface itself, solids models can determine every point within a solid object. It is this characteristic that brings about a great interest in solids modeling. This technique enables one to determine solutions to some complex problems of practical interest, such as automatic interference checking. With such a set of important benefits, solids modeling may appear quite attractive. However, these benefits are not without their

VOLUME MASS

• 0.,714""'. • 0.7I4I7M~

SURFACE MIA: 0.....' ••02 CEITItOIDI X • ·O.-..a7E-DZ Y • ·0_

'IRST YOMEITS.

n • ·O.54l7G8IE·OI

Z • ·O.IS8I1107E.OI

" . -0.78 . . . . ·01

rz. ·0.1241750£-02

MOMENTS OfF I.IITIA:

IIIX·

O~·02

In • O.3232I6IE·OZ III • o.887SZIlE-OI

PRODUCTS OfF INERTIA

Figure 10-15. A solid model with the bidden lines removed and with its associated mass properties.

COMPUTER-AIDED DESIGN 773

price. Interactive solids modeling demands a great deal of computing power to accommodate the practical designs of today. In many instances it is perceived to be too slow for widespread application. In many cases, solids models are based on the addition and subtraction of simple primitive geometric shapes that are easily defined. Although this speeds up the design process, it is inadequate to describe solid objects of a class defined by sculptured surfaces. Given these circumstances, the additional benefits gained by using solids modeling in plastic part and mold design applications are often outweighed by the costs of the additional time and computing power required to accomplish the task. Remember that hardware and software technologies are changing so rapidly today that solids modeling may not require such a drastic trade-off in the very near future.

ADDITIONAL CAD/CAM FEATURES USED IN PLASTIC PART AND MOLD DESIGN Group Technology Group technology is a technique by which parts may be characterized and classified into parts having similar geometric features. Once a part is classified, a code number is assigned to it. The numbers within the code-numbering system each have a significance related to a description of the product. Once a database of designs is encoded, when the requirement for a new design is introduced the existing database may be searched for exact or close matches of a part description, which is provided by the significant code number. This tool prevents accidental duplication of existing parts and encourages using existing parts that are close to the design needs. Not only are design costs reduced by this method, since existing designs need only be "edited" to produce new designs, but substantial manufacturing costs may be saved as well. Parts can sometimes be "made from" existing parts by the addition of a machining or assembly operation, at a lower cost than manufacturing the new part from scratch. Further, it is often feasible to install interchangeable inserts in the existing part mold and thus allow a single mold to produce two or more similar components. Manufacturing process information may be similarly retrieved by a group technology code number, so that group technology may sometimes be thought of as an artificial experience tool.

Finite Element Modeling Finite element modeling is a technique whereby a material continuum is divided into a number of patches, or "finite elements," and the appropriate engineering theory is applied to solve a variety of problems. The initial (and probably still dominant) use of finite element modeling was for the solution of structural engineering problems. The technique is currently being applied by a number of companies and research institutions in the design of plastic products. CAD/CAM systems provide the means to create a "mesh" of finite elements directly from a product model database, by automatic and semiautomatic means. The model described by the mesh is then analyzed and the results displayed, if desired, by graphic means. A mesh and the corresponding analysis results are shown in Figures 10-16 and 10-17. For structural designing with plastic materials several unique requirements exist. Because plastic materials may have nonlinear and anisotropic material prop-

774 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 10-16. A typical finite element mesh displays the distortion caused by mechanical loads and predicted by analysis.

Figure 10-17. The maximum principal stress values in each element, as predicted by analysis and displayed graphically.

erties, finite element programs for the analysis of polymer structures should possess the ability to characterize material nonlinearities and anisotropy. Additionally, materials that are reinforced by glass or other types of fibers will be influenced by the degree and direction of fiber orientation in the molded product. Several research institutions are currently working on computer programs to predict this orientation in molded parts. Finite element modeling is receiving increasing interest in plastics not only for structural design purposes but in the area of predicting plastic flow and heat transfer as well. Figures 10-18 and 10-19 show the finite element mesh and resulting pressure distribution for a bottom view of the computer housing previously described in the discussion on surface modeling. In this case the isobar pattern helps identify the total pressure required to fill the part, as well as the uniformity of pressure distribution and the location of the weld lines formed by flow around the holes in the part. The combination of ease of modeling provided by CAD/CAM systems and easier-to-use finite element systems will help to cause this computer-aided engineering tool to proliferate in years to come.

COMPUTER-AIDED DESIGN 775

Figure 10-18. A finite-element mesh applied to the ignition computer housing shown in Figure 10-13 and 10-14.

Digitizing Thus far we have discussed the methods used to define a product's or mold's geometry from scratch or from a designer's concept. In many cases other work may have been done that can expedite the creation of a database for the product. Traditional methods for product design and development have often involved the use of an appearance model or pattern where consumer testing or industrial design personnel may approve the appearance or functionality of the product prior to engineering detail design work. The ability to reach out and touch a physical model may be indispensable for certain design environments. If such a model exists, there is a strong possibility that it may be used as a medium to expediently create a product model database and thus capture many of the benefits of CAD/CAM technology.

ISOBARS

AlCJlUI

Figure 10-19. The results of flow-analysis predicting the isobars, or lines of constant pressure, at the instant of complete mold filling.

776 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Digitizing describes the technique by which a physical model is re-created in digital form by in some way scanning the physical model and building a database of points through which lines and curves are later constructed. This technique may be used in either a 2-D or a 3-D digitizing mode. In the 2-D mode a set of 2-D models, such as drawing Mylars, are placed on the table of a digitizing board or tablet. A number of points are created by using a tool positioned over the model at various locations. Then a pulse is sent to the computer, which records lhe X and Y coordinates of the device at that instant. This 2-D digitizing is useful when the input model or master exists in the form of a 2D pattern. If a substantial number of 2-D curves are available at varying Z depths, a 3-D model can be created by curve fitting the points along the Z direction (axis) simultaneously with the X and Y directions. The most common way to create a digitized model for 3-D objects is from a 3-D model. In this method a measurement device such as a coordinate measuring machine is programmed to traverse to known X and Y locations, and the Z height of the model is measured and recorded at those locations. This series of points is then fitted with smooth curves such as b-splines, which are then used to develop surfaces like bsurfaces. Although this method of geometry creation seems relatively straightforward, it is not without problems or pitfalls. The first consideration is whether the accuracy of the model and the digitizing method being used suffice for the application. If it is not sufficient, a more precise method of creating the geometry must be pursued. Additionally, digitizing works best for surfaces or geometries of relatively slowly changing curvatures. This means that geometries such as sharp comers or crease lines may cause problems. If a geometry such as a sharp crease in the product model exists, care must be taken to place a series of points on that crease and to stop the crease at the crease line. This is due to the mathematics of fitting curves through a series of points and the fact that the mathematical methods used do not handle drastic curvature changes between points very well. The only alternative is to increase the point density of the surface being modeled, but this enlarges the size of the database, causing a slowing of the CAD/CAM system's response. The digitizing method may be successfully applied, however, if the user is careful in the method of creating geometry and has a reasonable understanding of how the CAD/CAM system being used fits curves through points. Other Features

Other design features may now be summarized, such as layering [12]. Layering is a method by which the display of selected geometric entities may be turned on or off at the discretion of the user. The drawingboard analogy to layering is the overlaying of multiple tracings or Mylar sheets during the design process, to build up assembly layouts from a series of detail drawings. Most CAD/CAM systems support a feature of creation of groups of geometric entities that may be selected or manipulated together. The concept of a pattern is similar to that of a group, with one important exception; the entities used to construct a pattern may not be selected or changed once the pattern is created. With large-scale geometry manipUlation the system allows large blocks of geometric entities to be copied and moved to new positions or to be rotated or mirrored about an axis. The concept of a local coordinates or construction plane is one that makes modeling in 3-D convenient. This feature allows the user to set an absolute coordinate system and build a model relative to that system. Because the computer can solve mathematical

COMPUTER-AIDED DESIGN 777

------ ,.-------

Figure 10-20. The model and drawing concept, the model here being a 3-D representation. This drawing documents the model through a series of two-dimensional views.

equations quickly, a new coordinate system can be established at each location and orientation wanted within the model. A number of CAD/CAM systems today support the concept of model and drawing modes of geometry creation. Doing so allows those companies that require the output of paper drawings to create them (see Fig. 10-20) while still maintaining the same high integrity of the designed product model database. The verification of geometric relationships system permits any relationship between two geometric entities in the database to be quickly established by the computer. This means that rather than having to scale a drawing or insert a dimension in the database, a function exists that allows the designer to understand the distances and angles between the geometry existing in the database [12]. When automatic dimensioning and automatic tolerance analysis database systems are used, this allows the designer to select tolerances on the drawing and let the system calculate a statistically derived sum of those tolerances. The minimum and maximum dimensions will be obtained as a result of the tolerances applied. Other features used in plastic product and tooling design are available.

PROCESS-ANALYSIS TOOLS Recent years have seen a tremendous growth in the nUQ1ber of software products available to serve the needs of manufacturing functions. The products that have evolved over this period are tools that serve to replace the rules of thumb of the past with analytical analyses based upon sound theoretical principles. These products combine the benefits of relative ease of use with the speed of the computer, resulting in tools that can be cost-effectively applied to a large number of problems. Over the last decades, specific software products have arisen to serve the needs of the injection-molding industry. These tools are most effectively applied prior to construction of the mold, but they can be applied after the fact to solve process-related problems. Three types of analysis tools have emerged recently and are providing major benefits to molders. The types of analyses commercially available fall into three categories: 1. Flow analyses 2. Cooling analyses 3. Economic and plant-operating analyses

778 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

In general, these analysis tools fall under the domain of computer-aided engineering (CAE). The key word there is aided. Analysis tools in no way replace skill or education in the basics of plastic material properties, mold design, or processing. What analyses do is supplement the knowledge of trained individuals, making them more productive and more accurate in their predictions. The basic methodology behind CAE technique is that a design or process is proposed as the first step. The designer then constructs a model or representation of the specific design, using a prescribed method. The computer is then used to rapidly evaluate the results of both the input conditions and the model that the designer has described. The output conditions are listed by the computer, and the designer evaluates the consistency of results with his experience, then determines how the design must be modified to achieve acceptable results. The process is repeated until a successful design is achieved. In this manner the computer aids the designer by calculating results much more rapidly and with greater precision than is humanly possible. The skill and experience of the designer are still reflected in the final results. The CAE technique can be effectively applied, as an example, in the injection-molding field to 1. Maximize the probability of first -time plastic part or mold functionality. 2. Solve process problems such as warping, dimensional inconsistency, and long cycle times. 3. Reduce molding costs such as mold start-up costs, part-molding costs, material costs, mold rework costs, and scrap and regrind costs. Two major types of analyses are discussed below-flow analysis and cooling analysis. It is important to review the literature, as more and more products are being introduced to serve the needs of the molding industry each year. The use of these CAE tools will assure that mold design and part processing can be accomplished with the greatest possible speed and effectiveness [11,12].

Melt-Flow Analysis Over the last decade the field of flow analysis has gained increasing importance in injection molding. Flow analysis has provided rational solutions to many of the hard-to-understand effects that cause problems in the injection-molding process. These effects have included warping, molded-in stress, excessive fill pressures, part flashing, and others. The interrelationships between part design and molding process parameters that cause problems of this nature were not well understood in the industry. Practical experience was often insufficient to identify potential problems, and too limited to have encountered the full range of molding problems that can be addressed by such techniques as flow analysis. Hence, much prototyping and mold fine tuning was necessary before successful moldedproduct results would be achieved. Computerized flow analysis has emerged as a powerful tool to aid in the implementation of applying injection molding as the production process of choice to a widening spectrum of products. The ability of modem digital computers to perform complex calculations in short periods of time has been the breakthrough that made flow analysis a tool applicable to increasing numbers of new parts (see Figs. 10-21 and 10-22). Furthermore, technology advances in computer hardware have allowed these flow models to increase in their sophistication and accuracy while at the same time bringing the cost of the analyses into

Figure 10-21. How to describe a part to a computer is summarized here. The user's major task is to accurately describe the geometry of the part to be molded so that the programs can analyze the melt flow through the mold cavity. Simple parts present no problem, and sometimes the flow length and part weight are all that are needed to balance the runner and gate system. For complex parts, such as this surrounding for a headlight, a lay-flat graphic approach is used. The part is first "flattened" to create a 2-D graphic representation. Then a series of circular-flow fronts is drawn, sectioning the mold. These reflect the radial flow of the melt from the gate and could be thought of as fronts for successively larger short shots.

Figure 10-22. Computer graphics can display a flow pattern as isobars (lines of equal pressure) like these against an isometric representation of the part being formed within the mold. 779

780 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

a range at which they can be applied to a large number of new designs. The flow-analysis tools can be successfully applied and utilized by three different groups in the product development process: the product designer, the mold designer, and, finally, the injection molder. Product Designers

Product designers can apply flow analysis to many questions like the following: Will the part fill at all? This age-old question concerns many designers, especially those who design large injection-molded components such as covers, enclosures, furniture, and the like. The relationships among material structural properties, cosmetic properties, and processing properties are generally hazy in the designer's mind, and flow analysis provides a way to evaluate different materials in the design stage and to evaluate the processing-related characteristics in a scientific manner. What is the minimum practical wall thickness for the part? This question is a primary consideration for the cost of the molded product. The ability to use thin walls on the product results in obvious savings in material, which many times comprises more than 40 percent of the finished-product cost. A less obvious advantage, however, is the overall benefit in cycle time for the product that comes from using thinner walls. The cooling time of an injection-molded part is known to be a function of the square of the wall thickness of the part, so reductions in wall thickness have a substantial impact on the cycle time of the molded product. This is of great benefit in increasing the productivity of the molding plant and thus is ultimately reflected in product costs. Can gates be located acceptably? The ability of plastic materials to be formed into attractively styled shapes has long been recognized. This has led to an increasing use of plastic materials for applications requiring high degrees of aesthetic appeal. The proper use of flow-analysis tools can help assure product designers that sufficient latitude exists in the design to allow gates to be positioned to protect the aesthetic properties of the design, while at the same time allowing production of the item at reasonable cost.

Mold-Cooling Analysis Prior to the introduction of computer modeling tools, the cooling of injection-mold cavities was a complex and often misunderstood phenomenon. The placement of cooling lines in the mold was often the last portion of the design to be considered, resulting in many molds that ran at low levels of productivity and producing molded parts of questionable qUality. Modem techniques that allow the evaluation of a proposed cooling system design prior to construction of the mold are allowing mold designers to evaluate thoroughly the quality of their designs and revise them to achieve more efficient and balanced cooling ofthe mold (see Fig. 10-23). These techniques are generating large savings in cycle time, parts cost, and tryout and tuning time for new molds. The importance of proper cooling system design can be summarized in two major areas of interest to the plastics industry: quality and productivity. Quality has become a major area of emphasis in the industry over the last few years. The quality of the product that can be produced from an injection mold can be directly attributed to three factors: the accuracy of fabrication of the mold cavity and core, the repeatability of the molding machine used, and the correct design of the mold to produce the part. Part of the correct design of the mold will be a cooling system that will extract heat from the melt in the mold cavity at the maximum rate possible, uniformly throughout the mold.

COMPUTER-AIDED DESIGN 781 MOLD MATERIAL DATA BANK Select cavity and core matenal codes for thermal properlles data

~-

-

MOLDCOOL FILE Enter data on all aspects of the part, mOld, and processing condition

~

~

COOLING LINE DATA FILE Describes and dimensions all coolanthandhng components in each mold plate, core and cavity (channels, bends, baffles, bubblers, etc,)

+

f-.-.

CIRCUIT PROGRAM Organizes the cooling channels into circuits to insure equal pressure drop and temperature rise to promote even-balanced cooling

+

OPTIMIZATION PROGRAM (2) Specifies coolant inlet temperature ~ pressure, and flow rate to obtain specified mold surface temperature (or part temperature). Calculates cooling time for cycle determination

L

PLASTIC MATERIALS DATA BANK Select matenal code for thermal properties data

DESIGN PROGRAM (Cooling System Capacity) Computes the total heat load from parameters that pertain to mold cooling outlined in the MOLDCOOL FILE Gives a listing of common channel diameters, effecllve length, depth and Pitch to match heat load with coohng capacity

~ PROPOSED COOLING LAYOUT Done by designer manually from output supplied by DESIGN PROGRAM

AUXILIARY PROGRAMS: CURE: Calculates theoretical coohng time for part to target design efforts and check analysis results EQUI~: Matches optimum condillons with proper auxiliary equipment specifications to ensure the best use of capital and energy HELP: Summons assistance from Application Engineenng-answers are never more than 24 hours away

1 MOLDCOOL-FAST fast-running program for quick results BaSIC,

2 MOLDCOOL-OPTIMUM Same output as MOLDCOOL FAST, but uses more sophisticated, two dimensional finite-
Figure 10-23. A mold-cooling analysis diagram used in a CAD system--the Moldcool program by Moldflow Australia Pty" Ltd,

Several molding quality problems can be avoided if uniform cooling is designed into the mold, Warpage of the molded part is one problem whose roots may lie in the proper design of the cooling system for the mold. Another cooling problem that can be related to cooling system design is the homogeneity of properties in parts molded from crystalline resins. The morphology of the crystal structure developed in the molded part is strongly dependent on the material's temperature history and cooling times. Minor changes in these variables can lead to major changes in crystal formation and thus in the mechanical properties of the molded product. Thus, molds with uneven cooling profiles may produce parts with unacceptable mechanical properties, possibly resulting in premature mechanical failure of the molded product. The second major benefit of effective cooling system design is the impact on productivity. Molds that can be designed for optimal cooling will produce parts in the shortest possible cycle time. This results in a direct cost saving of several types. It has been shown many times that large energy savings on injection-molding equipment can be gained by decreasing cycle time. Additionally, effective cooling system design can result in the use of higher coolant temperatures, which imply less chiller capacity requirements or less plant water usage. There are also capital equipment savings accompanying these benefits.

782 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Capital outlays for new equipment can be minimized once shorter cycle times reflect themselves in higher plant capacity without an increase in the number of molding machines. Likewise, maximizing the cooling-water temperature needed to cool the mold results in a reduction in chiller requirements and a corresponding savings in equipment costs. In examining the problem of the cooling of plastic parts being formed by injection molding, it is possible to separate the problem into three distinct elements: 1. Cooling of the melt 2. Conduction from the melt to the waterline 3. Convection cooling by the waterline These elements must be considered in combination in order to understand the cooling performance of the mold. Improper mold design may result in a high thermal resistance that will lead to the quality and productivity problems previously mentioned. The preceding overview provides strong arguments for careful consideration of the cooling system design for injection molds. In short, the successful application of cooling analysis tools can result in higher quality in the molded product as well as lower operating and production costs for the molder [11, 12].

DESIGN DATABASES The Database Concept One of the key factors required to implement a CAD/CAM program that will allow increased productivity is the effective use of a database, especially design databases (see Appendix D). A design database typically contains graphics, plastic materials information, parts or components that are frequently used in mold designs, as well as standard design methods to be applied to new designs. There are at least two methods to create a design database, both of which are described below. Before these methods are described, it is necessary first to clarify the exact meaning of database. A database can be defined as one of the largest elements in a hierarchy of information structure. In order to illustrate this concept, consider the following types of information elements, ranked in order of their increasing information content: Data item. A single unit or specific piece of information. If we were to consider using a line as an illustration, a data item about a line might be the X coordinate of one of the ends of the line itself. Data record. A collection of data items in a logical sequence. Again using the illustration of the line, the data record of the line might contain the X, Y, and Z coordinates of both ends of the line. Data file. A collection of data records, generally organized around some common attribute. Once again, a number of lines can be organized into a data file that describes the geometry of a specific standard part. Database. A collection of files, again usually organized around some common attribute or organized for a specific purpose. Continuing with this example, a number of data files that describe standard components can be further organized into a database that describes all the standard mold bases available from a particular manufacturer.

COMPUTER-AIDED DESIGN 783

To further clarify the concept of a data hierarchy, whose highest tier is the database, an analogy of a paper file system is presented in Figure 10-24. In this analogy the data item is a specific line contained on a page of paper. The data record is the page of paper containing many data items. The next level of the hierarchy is the data file, which corresponds to the many sheets of paper contained in a file folder. The last tier of this structure is the database, which corresponds to the entire filing cabinet.

Graphics Databases The concept of a graphics database is essential to understanding the nature of a CAD/CAM system. Such a system takes all the design information that would normally reside on a piece of paper and builds that information into files to create the database. Each graphic entity, such as lines, arcs, circles, points and surfaces, is represented by a number of data items. These items are further collected in a data record and a number of these records are then collected into a file, which generally describes a molded part or the mold itself. A number of these files may then be grouped together to form the database. In the same concept of a hierarchy of information we may consider one of the key tools for boosting the productivity of design, the library of parts. This tool is a database of geometry, which is recalled and used as quickly as the computer can associate it with the particular design in question. A library of parts is particularly useful when items such as injection molds are to be created from standard components.

Defining the Library Database The simplest method of creating a database is by combining a collection of part files. To do so a designer will generate all the parts required on a CAD/CAM system, then store them in a logical catalog structure so that they may be easily retrieved at a later date.

PAPER SYSTEM

DOCUMENT

--

---------r-fI LI NG CABINET I FILE

RECORD FILE

DATA BASE Figure 10-24. An analogy of the database concept.

784 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Most CAD/CAM software packages support a command that allows a part currently existing in the database to be inserted into the part being designed. Additionally, many software packages support a method of preparing parts in a figure of associated graphics, rather than a collection of discrete entities, which speeds up the operation of inserting previously defined geometry into a new design. The major benefit of this overall technique is that once the part is created it never needs to be redefined and can be used over and over again. Unfortunately, there are also several disadvantages to this technique. The first constraint is that the parts must not change, or the geometry will require editing after insertion. Depending on the particular type of system, this may present problems, as some systems do not support the editing of graphics figures described in the last paragraph. Second, if there is a large number of unique combinations of geometry (i.e., for mold bases), an extremely large database will be required. This implies that a large amount of work will be required to create each component, in addition to large amounts of storage being needed to keep the database on line. Since on-line storage is relatively expensive, another computer program, called a database management system, is usually employed. Such a system (see Appendix D) will provide the convenience and economics of storage off line but allow rapid retrieval of the components desired. One means to eliminate the problems caused by databases of parts files is to use what is known as graphics language programming. The concept of a graphics language program is that a computer program builds the required geometry only when it is needed; through interactive programming it allows the operator to intervene and change the geometry as required. The language in which this program is written is called a graphics language, since the language facilitates the creation of graphic entities automatically. An example of the use of such a program might be given by the need for graphics for a standard ejector pin. This pin could be available in a number of different lengths but must be cut to a specific length to be properly installed in the mold. A graphics language program can then be written to prompt the user for the required diameter, length, and location of an ejector pin to be installed. The system will then read a data file of the corresponding head diameter and thickness and create the geometry and display the graphics to describe the part completely. The advantage of the graphics programming technique is that it drastically reduces the amount of on-line storage space required to support large libraries of common geometry. Additionally, by allowing operator intervention we may now build standard parts with unique characteristics, such as ejector pins "cut to length." Variable parameters such as A and B plate thicknesses and the desired locating rings or sprue bushings can be interactively entered at the program run time, creating any of many unique mold bases. Design standards can also be incorporated into the graphics programs, thus decreasing the probability of designer errors and promoting standardization in both the design and manufacturing operations. An example of design standards might be the uniform recessing of screw heads in a design, which would allow cutting tools to be set once and would help identify potential interferences caused by the shop's inadvertently counterboring too deeply. The major disadvantage of the graphics programming technique is that the speed of inserting new geometry in the design is generally slower than an "insert part type" command. Depending on the complexity of the geometry and the software in question, this mayor may not be a disadvantage. In general, the user needs to assess this performance issue for his particular application with the specific software package available on his system. It suffices to say, however, that graphics programming for families of parts

COMPUTER-AIDED DESIGN 785

applications like component libraries has shown itself to be an extremely beneficial feature of CAD/CAM systems. Users considering the potential purchase or effective usage of these systems should give graphics programming languages and their ease of use careful consideration in their implementation activities.

COMPUTER-INTEGRATED MANUFACTURING Until now we have been discussing CAD/CAM and the various state-of-the-art techniques for its application in the field of mold design. Before closing, it might be appropriate to discuss briefly how CAD/CAM is changing in relation to computer-integrated manufacturing (CIM). When addressing the field of computer-aided manufacturing today most speak solely of numerical control as the means of utilization of the product model database. Obviously, the field of manufacturing is composed of many processes that do not rely on numerical control as their means of control. Computer-integrated manufacturing is the natural evolution of CAM into serving an ever-widening scope of manufacturing processes. The unique element of CIM, however, is that it tends to integrate or tie together all the manufacturing processes into one coherent unit, all of which share a common database. CIM involves the appropriate combination of hardware and software that allows the manufacturing processes and functions to draw information from, and contribute information to, the database. This information is shared by all the respective functions, including the business ones, as a primary businessinformation system. As illustrated in Figure 10-25, the CAD/CAM system can be envisioned as a central hub of primary information such as the product model and its attributes, and secondarily derived information such as toolpaths, bills of materials, and so on. Just as in the manual method of doing business the engineering information stored in a database begins the product delivery cycle, this basic information about the geometry and attributes of the product is transferred and reused by many subsequent business functions, such as materials planning and manufacturing engineering. We are moving toward the realization of CIM when we will be able to pass information efficiently to and from each of the manufacturing functions to the database. MYTHS AND FACTS

Recognizing that, as with all products and actions in this world nothing is prefect, there are always new developments that make life easier by taking an asymptotic approach to perfection. First of all it is best to do away with some of the myths about computers. The computer will not design a part, not make critical design decisions, and not produce the end product at the touch of a button. What the computer will do is make the designer's life easier and more productive. The designer now no doubt has a 32-bit computer right at his or her fingertips and is no longer limited to paper, pencils, calculators, and the like. Designers can now design, if willing, by starting to think in three dimensions. They can view a product from any angle and display many views at once. For some it may have been painful to draw an attractive isometric view, but the computer handles this with ease. Designers can now zoom to any degree of depth in a graphic and work on a very small portion of a part until the desired result is achieved. Much more, like panning, can be performed. And document tracking is now handled by proven database management software packages. Some of the most important capabilities that CAD/CAM has to offer the designer

786 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 10-26. Summary of the design analysis capability available through computer-aided software that is used to reduce the new designed product lead time.

are a vast amount of software and databases available on the same computer that is running graphics processors. This consolidation is a great advantage to the designer who realizes it. All the geometric data that in the past had to be determined and punched in are now right there in the designer's 3-D database so eloquently called the model. With the correct processors, any type of numerical-solution program can be utilized, and with the correct postprocessors the results displayed will immediately come back on the graphics screen. Powerful finite element analysis programs like MSC Nastran, Ansys, Superb, and Moldflow can now be fully analyzed, with more new types becoming available almost daily. Programming-experienced designers can, using a systems graphics language, develop their own applications software to do any specialized task for the design at hand. Again, the geometric database is there for the taking. For the aggressive designer, CAD/CAM means a whole new realm of better and more accurate ways of completing a successful design project (see Fig. 10-26).

CAPABILITY AND TRAINING The decisive step in switching to CAD/CAM is the choice of a suitable CAD/CAM system. In spite of recognizable endeavors at standardization in the area of interfaces, there remains a diversity of systems, which is more likely to deter than to motivate. In addition, there is a large number of unfamiliar terms that are often used as casually as if they were an integral part of our language. Many computer system and software suppliers provide advice on specific applications.

COMPUTER-AIDED DESIGN 787

Task Planning Design concept Model-build

17 Time savings from CAE due to: Bener conceptual work Geometry avail. from earlier steps Basic productivity gain Fast response to changes Fewer revisions Reduced parallel path activity

Test OesignPhase 1 Build toolingPhase 1 OesignPhase 2 Build toolingPhase 2

Start production

Figure 10-25. Infonnation that can be generated, stored, and maintained on a CAD/CAM system. This is especially important if the areas of design, work planning, and production are still relatively unstructured. Working with CAD/CAM compels one to follow an orderly approach with systematic action. For computer-aided management, the drawings to data files must be clearly numbered and designated. In many organizations, the support group needed for this purpose must still be created before the new system can gain access. An adequate number of personnel with practical experience should be available for a well-planned entry into computerized technology. The method of working with CAD/CAM means that the areas of design, work planning, production, and quality assurance must move closely together to have the proper interface. It is recommended that the specialists selected for training in CAD/CAM possess wide knowledge in these fields. It is not surprising that the computer and its peripherals head almost every designer's or engineer's list of "tool's I cannot do without." However, it is possible that fewer than 20 percent of design engineers are in fact truly computerized [751]. The uncertainty caused by inadequate training and the expectations placed on CAD/CAM systems, which as a result are not fulfilled, bears no relation to the savings made in training costs. Only if sufficient time and facilities are provided to training purposes can the hoped-for results be attained. Working with CAD/CAM requires experience, but it creates flexibility. The goal of simplification and flexibility in design necessitates adopting standard procedures and automatic routines, where possible, by the computer systems. Anyone who works quickly

788 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

and efficiently at the design stage can also commence earlier with such different programs as testing (see Chapter 9) of the product and will thus have more time available for modifications and corrections. In summary, the design and processing of free-form surfaces are scarcely possible without the use of the computer. Three-dimensional designing has been greatly simplified by the use of high-performance software in conjunction with highly developed monitors. Complex 3-D shapes can now be viewed from any direction, turned, mirrored, and cut. In addition, there are numerous design aids such as for the joining of surfaces by tangential junctions, surface generation according to mathematical functions, joining with a constant or variable radius, generating wall thicknesses by offset, and. many other useful and practical design aids.

Chapter 11

DESIGN FEATURES THAT INFLUENCE PERFORMANCE

BASIC DETRACTORS AND CONSTRAINTS In Chapters 3 through 5, 7, and 8 in particular some of the design problems of plastics were discussed. This chapter provides more detail on this important subject. Even though some of the analyses here will pertain to a specific process, many will relate to other processes, so it is best to review them all. Also, most of the characteristics of plastics that detract from their effectiveness refer to injection molding and provide guidelines for all the processes. Although it is not actually the product designer's job to do so, the designer should have some idea of where problems can develop, based on how a mold or die is designed and manufactured [1, 2, 5-14, 21-32,40-42,45-46,50,55-68, 166,

208,225,253,409,478,498,564-70,593,637,674,807-15]. The successful design and fabrication of good plastic products requires a combination of sound judgment and experience. Designing good products requires a knowledge of plastics that includes their advantages and disadvantages and some familiarity with processing methods. Until the designer becomes familiar with processing, a fabricator must be taken into the designer's confidence early in development and consulted frequently during those early days. The fabricator and mold or die designer should advise the product designer on materials behavior and how to simplify processing. The designer should not become restricted by understanding only one process, particularly just a certain narrow aspect of it (see Chapter 7) [10-14, 32, 33, 40-42]. One of the earliest steps in product design is to establish the configuration of the parts that will form the basis on which strength calculations will be made and a suitable material selected to meet the anticipated requirements. During the sketching and drawing phase of working with shapes and cross-sections there are certain design features with plastics that have to be kept in mind to avoid degradation of the properties. Such features may be called property detractors or constraints. Most of them are responsible for the unwanted internal stresses that can reduce the available stress level for load-bearing purposes. Other features, which are covered in this chapter, may be classified as precautionary measures that may influence the favorable performance of a part if they are properly incorporated [4-6]. Although there is no limit theoretically to the shapes that can be created, practical considerations must be met. These relate not only to part design but also to mold or die design, since these must be considered one entity in the total creation of a usable, 789

790 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

economically feasible part. In the sections that follow, various phases considered important in the creation of such parts are examined for their contribution to and effect on design and function. Prior to designing a part, the designer should understand such basic factors as those summarized in Table 11-1 and Figure II-I. Success with plastics, or any other material, for that matter, is directly related to observing design details. For example, something as simple as a stiffening rib is different for an injection molded or structural foam part, even though both parts may be molded from the same plastic (see Fig. 11-1). However, a stiffening rib that is to be molded in a low-mold-shrink, amorphous TP will differ from a high-mold-shrinkage, crystalline TP rib, even though both plastics are just injection molded (see Chapter 2). Ribs molded in RP/composite plastics have their own distinct requirements (see Chapter 5). Hollow stiffening ribs of the type produced by thermoforming, blow, or rotational molding have the same function, but they are designed to be totally different [807]. The important factors to consider in designing can be categorized as follows: part thickness, tolerances, ribs, bosses and studs, radii and fillets, draft or taper, holes, threads, color, surface finish and gloss level, decorating operations, the parting line, gate locations, molded or extruded part shrinkage, assembly techniques, mold or die design, production volume, the tooling and other equipment amortization period, as well as the plastic and process selected. The order that these factors follow can vary, depending on the product to be designed and the designer's familiarity with particular materials and processes. Preparing a complete list of design constraints is a crucial first step in plastic part design; failure to take this step can lead to costly errors. For example, a designer might have an expensive injection mold prepared, designed for a specific material's shrink value, only to discover belatedly that the initial material chosen did not meet some overlooked design constraint. Flammability, glass-fiber fillers to provide a higher modulus, and other requirements are best considered before a tool is made. Otherwise, the designer may have the difficult if not impossible task of finding a plastic that does meet all the design constraints, including the important appropriate shrink value for the existing mold. Such desperation in the last stages of a design project can and should be avoided. As emphasized from one end of this book to the other, it is vital to set up complete checklist of product requirements, to preclude the possibility that a critical requirement may be overlooked initially. Recognize that the "impossible" as well as the "approaches" (~an

SHe

moIdetI

TW:T :olLJF I"lee

100

molded

ermofermed.

llIew & ro 0 lana! mo(ded

ASS

1.01.'_

.750 or 5W

Figure 11-1. An example of how different plastics and molding processes can affect the design details of a stiffening rib.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 791

to be avoided can in most cases still produce excellent products; however it is easier to take the direction with the least number of problems.

Tolerances and Shrinkages Two different forms of shrinkage must be considered when designing to meet tolerances: the initial shrinkage that occurs while a part is cooling, called the mold or die shrinkage, and that which occurs after as many as twenty-four hours, called the after-shrinkage or after-swell. Some plastics are more stable than others after aging, regardless of their initial shrinkage. In many cases low shrinkage may indicate greater stability. As noted earlier, some plastics have zero shrinkage, with others having little or a high degree of shrinkage; see Chapters 3 to 7. Large, unpredictable shrinkages can make close-tolerance designing almost impossible, so these must be indicated on the drawings. If it has been determined in advance that a part must be postcured, stress relieved, or baked, allowance must be made for probable additional shrinkage. These requirements must be specified on the drawings. Especially for long runs, mold or die design is an important factor. The metals that will be used, particularly in mold cavities, and the forces required will largely be determined by the complexity of the product design. This complexity will, of course, dictate in tum the intricacy of the tool design that will eventually be used. In general, pack hardening, oil hardening, and prehardened steels are used, with materials such as beryllium copper and electroformed cavities finding use in applications for specialized purposes [10-12]. Chrome plating is frequently used to protect parts from the corrosive effect of volatiles present in some materials. Plating not only produces high luster and prevents tool staining but also eliminates the sticking of parts on removal from a mold or die. However, plating will only duplicate an existing surface, so the tool itself must be highly polished before being plated. The intricacy required in tool design, and the commensurate costs, may be leading factors when choosing a plastic. If the material must flow around pins, spiders, and projections, it must have suitable flow properties and weld properly, leaving only minimal flow marks (see Chapter 7). Thus, the availability of the proper material becomes an important consideration.

Residual Stresses Such processing-induced residual stresses that influence properties as mechanical, physical, environmental, and aesthetic factors (which also exist in other materials like metals and ceramics) can have favorable or unfavorable effects, depending on the application of the load with respect to the direction of the stresses or orientation. For example, at room temperature an unoriented PS is a brittle, glassy, amorphous polymer, whereas a uniaxial oriented PS is highly anisotropic. High tensile strength, elongation, and resistance to environmental stress crazing and cracking are achieved in the direction of orientation. However, an oriented PS is weaker and more susceptible to stress crazing in its transverse direction than is an unoriented PS. A biaxially oriented PS is strong and tough in all directions [12, 19,42,201,225]. Residual stresses and molecular orientation play an important role in the toughness enhancement of cold-worked plastics, because toughness is primarily based on the mechanics of craze formation and shear band (crazes and flaws) formation. The shear bands determine the fracture mode and toughness of a polymer when subjected to impact loads. The amount of energy dissipated will depend on whether the material surrounding the

'I I.C

N

Q

Maximum practical thickness, in . (mm)

-r

to

Minimum practical thickness, in. (mm)

t \jJ I" (25.4)

_050" (1.3)

Yes

Ir./

II

Undercuts

Minimum recommended

Yes

~I/

(

Core pull & slides

Yes

I,.

No

No

Yes

Yes

00

I" (25.4)

(1.5)

.060"

(6.35)

.250"

.030" (0.76)

to 6" (6.35-152 mm) depth: 1 to 3 6" (152 mm) dcpth and ovc.r: JO or as required

Yes

Ycs

Yes

Yes

11\1"

(3 . 18)

1/16"

(1.59)

Yes

draft, in.r

~form

Molding

1/16"

Bulk Molding Compound

(1.59)

~

~J

c;)

Trimmed in mold

Molded-in holes

Minimum inside radius, in _ (mm)

Molding Compound

Sheet

Compression Molding

.500" (12.7)

0.35" (0.89)

Yes

Yes

No

Yes

( 1.59)

1'16"

Injection Molding (Thermoplastics)

Table 11-1. Design Guides for Processes versus Product Requirements*

I,.

I,.

.500" (12.7)

.OBO" (2.0)

20 30

No

No

Yes

No

(6.35)

No limit

( 1.5)

.060"

0"

Yes

No

No

Large

(6.35)

Molding

~ss

Spray-up and Hand Lay-up

Cold

794 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

flaws deforms plastically. For toughness enhancement the residual stresses play an important role in the suppression of craze formation, by avoiding the stress state that promotes brittle fracture. The term residual stresses identifies the system of stresses that are in effect locked into a part, even without external forces acting on it. For instance, minute stresses may be induced in a material by nonuniform heating and cooling. The production of residual stresses is usually the result of nonhomogeneous plastic deformation occurring during thermal and mechanical actions, arising from changes in either volume or shape. Thermal treatments like quenching (rapid cooling) and annealing (slow cooling) introduce changes in physical and mechanical properties. For example, with sheet plastic the stresses created by quenching are the result of uneven cooling, when the surfaces cool faster than the core. This produces nonuniform volume changes and properties throughout the thickness. The compressive stresses on the surfaces of the quenched plastic produce tensile stresses in the core, which maintain the equilibrium of the forces.

Cold Working Generally, a variety of mechanical deformation processes cause the nonuniform deformation that results in the formation of residual stresses. This nonhomogeneous deformation in a material is produced by the material's parameters-largely its process parameters, such as the tool geometry and frictional characteristics. For example, the rolling of a strip can be accomplished by using squeeze rolls. In the rolling process, parameters with a small roll diameter and little reduction produce deformation penetration that is shallow and close to the surface, whereas the interior of the strip remains almost undeformed. After the removal of the deformation forces and a complete elastic recovery, this condition produces compressive residual stresses at the surface and tensile residual stresses in the core. The logic of this situation is that the surface material is forced to elongate more than the relatively rigid core permits. When there is a large diameter and much reduction, the deformation penetration occurs deeper in the core and there is a tendency for the plastic to lag at the surface, the result of friction at the material-to-tool interface. Thus, the cold working processes like rolling, drawing, extrusion, and forging produce residual stresses along with their molecular orientation [509]. Generally, the more discussed and technically reviewed residual stresses are in injectionmolded parts. Their presence can often be detected qualitatively by immersing TP parts in appropriate stress-cracking solutions for a short time, then observing the crazing caused by surface tensile residual stresses (see Chapter 4). Such methods are ineffective for a part with compressive or insufficient tensile stresses on its surface. To determine their magnitude includes the layer removal technique of microtoming (see Chapter 9).

Stress Concentrations Sharp comers should always be avoided in designing, particularly when working with TP injection-molded parts. Although sharp-cornered designs are common with certain sheet metal and machined parts, good design practice in any material dictates the use of generous radii, to reduce stress concentrations. Reinforced plastics and composites and metal parts will often tolerate sharp comers, because the stresses at their comers are low compared to the strength of the material or because localized yielding redistributes the load. However, neither of these factors should be relied upon in TP molded parts. Sharp

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 795

comers, particularly the inside comers, can cause severe molded-in stresses as a material shrinks onto the comer, as well as poor flow patterns, reduced mechanical properties, and increased tool wear. The elementary formulas used in design are based on structural members having a more or less constant cross-section, or at least only a gradual change of contour, but these conditions are seldom found in practice. The presence of shoulders, bosses, grooves, holes, threads, and comers results in modifying the simple stress distribution and in localized, higher stresses. This localization, known as the stress concentration factor, is defined as K = maximum stress divided by nominal stress. Localized high stresses must in most cases be determined experimentally rather than theoretically. The photoelastic technique is one of the more effective methods used to do this. To interpret a photoelastic diagram qualitatively it is sufficient to know that the number of fringes (the density of lines) is proportional to the absolute stress level. Basically, in the vicinity of a sharp comer all fringes converge toward the apex. Having a high density of lines at this point indicates the presence of high stress level. At a rounded comer there will be considerably less concentration. Besides the molding problems, sharp comers often cause premature failure because ofthe stress concentration. To avoid these problems, inside comer radii should be equal to one-half the nominal wall thickness, with a 0.020 in. radius considered as a minimum for parts subjected to stress and a 0.005 in. minimum for the stress-free regions of a part. Having inside radii less than 0.005 in. is not recommended for most materials. Outside comers should have a radius equal to the inside comer plus the wall's thickness. Figure 11-2 illustrates quantitatively the effect of fillet radius (r) on stress concentration factor (K). Assume that a force or load is being exerted on the cantilevered section shown. As the radius is increased, with all other dimensions remaining constant, r/t increases

LOAD ~

a:

.-

"c~r !~

3.0

e ()

~ 2.5

<: 0 j:::

<:(

g: ~ 2.0

()

r=

~

() CJ) CJ)

UJ

g:

V2 t GOOD DESIGN STANDARD

1.5

CJ)

1.0 0.5

1.0 RATIO, ~

1.5

t Figure n-2. The effect of the fillet radius on the stress-concentration factor. To find the stress from having a small radius, multiply the calculated bending stress by K.

796 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

proportionally and the stress-concentration factor decreases, as shown by the curve. The K factor can be reduced by 50 percent, from 3.0 to 1.5, by increasing the rlt sixfold, from 0.1 to 0.5. This curve shows how readily the K factor can be reduced by using a large fillet radius. A fillet with an optimal design is obtained with an rlt of 0.5. A further

increase in the radius reduces the stress concentration by a marginal amount. Stress concentration factors on the same order of magnitude have also been determined for grooves, notches, holes, screw threads, bosses, and ribs [810]. INJECTION MOLDING Design Concepts

In designing a totally new product or redesigning an existing one to improve the product, bring about cost savings, or some combination of these or other reasons, consideration should be given to the key advantages of injection molding (1M). These advantages include the ability to produce finished, multifunctional, or complex molded parts accurately and repeatedly in a single, highly automated operation (see Chapter 7). While keeping this in mind during the initial planning stage, one should also be aware of the general design considerations presented in this section [1, 2,5-12,40-46,50-70, 86, 231,409,564,569,593, 807]. Many parts of an injection mold will influence the final product's performance, dimensions, and other characteristics. These mold parts include the cavity shape, gating, parting line, vents, undercuts, ribs, hinges, and so on (see Table 11-2). The mold designer must take all these factors into account. At times, to provide the best design the product designer, processor, and mold designer may want to jointly review where compromises can be made to simplify meeting product requirements. With all this interaction, it should be clear why it takes a certain amount of time to ready a mold for production. Thus, in the design of any 1M part there are certain desirable goals that the designer should use. In meeting them, problems can unfortunately develop. For example, the most

Table 11-2. Functions of an Injection Mold Mold Component Mold base Guide pins Sprue bushing (sprue) Runners Gates Cavity (female) and force (male) Water channels Side (actuated by cams, gears, or hydraulic cylinders) Vents Ejector mechanism (pins, blades, stripper plate) Ejector return pins

Function Perfonned Hold cavity (cavities) in fixed, correct position relative to machine nozzle Maintain proper alignment of the two halves of a mold Provide means of entry into mold interior Convey molten plastic from sprue to cavities Control flow into cavities Control size, shape, and surface texture of molded article Control temperature of mold surfaces, to chill plastic to rigid state Fonn side holes, slots, undercuts, threaded sections

Allow escape of trapped air and gas Eject rigid molded article from cavity or force

Return ejector pins to retraced position as mold closes for next cycle

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 797

common mold design errors of a sort that can be eliminated usually occur in the following areas: Thick/thin sections, transitions, warp, and stress Multiple gates and weld lines Wrong gate locations Inadequate provision for cavity air venting Parts too thin to mold properly, as diaphragms Parts too thick to mold properly Plastic flow path too long and tortuous Runners too small Gates too small Poor temperature control Runner too long Part symmetry vs. gate symmetry Orientation of polymer melt in flow direction Hiding gate stubs Stress relief for interference fits Living hinges Slender handles and bails Thread inserts Creep or fatigue over long-time stress (extremely important) As seen in previous chapters, different plastics have different melt and flow characteristics. What is used in a mold design for a specific material may thus require a completely different type of mold for another material. These two materials might, for instance, have the same polymer but use different proportions of additives and reinforcements. This situation is no different than that of other materials like steel, wood, ceramics, and aluminum. It is important to recognize that the drawing of a plastic product will not specifically spell out the way many of its details will be carried out in the mold design. Some features adversely affect the strength and quality of the molded product. In most cases, these problem details can be modified by the designer to minimize their adverse effects on the properties of the part [12, 46]. What follows is a general summary of how to reduce problems to tolerable limits. First, inside sharp comers should normally be shown as two intersecting straight lines, without specific indication as to the functional requirement or degree of sharpness. Inside square comers are stress-concentration areas, quite similar to a notch in a test bar. The Izod impact strength of notched and unnotched test bars shows the relative impact strength of each material at the two conditions. Thus, for example, polycarbonate has an impact strength of the notched Hn. test bar of 12 to 16 ft.-Ib.lin. of notch, whereas the same bar unnotched does not fail the test. Polypropylene has an impact strength 30 times greater in the unnotched than the notched bar. Nylon shows a drastic increase in impact strength as the radius increases from sharpness to A R. A similar trend exists for most other materials. These examples point out that brittleness increases with the decreasing of a radius in a comer. Visually, a radius of 0.020 in. on a plastic part may be considered sharp, with an influence on strength that is much more favorable than a radius of 0.004 in. To the moldmaker, a sharp comer is

798 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

usually easier to produce, but in the plastic part it is a source of brittleness and, in most cases, is highly undesirable. Inside sharp comers on plastic-part drawings are a frequent occurrence. It is the mold designer's responsibility to call attention to such strength degradation and invite appropriate corrective measures. Second, varying wall thicknesses from thick and thin sections in a part will lead to problems in molding. Having a uniform wall throughout a part gives it good strength and appearance. Thick and thin sections will have molded-in stresses, different rates of shrinkage (causing warpage), and possibly void formation in the thick portion. Since the parts in a mold solidify from their outer surfaces toward the center, sinks will tend to form on the surface of a thick portion. When thick (h in. and over) and thin (1 in. or less) portions are unavoidable, the transition should be gradual and coring should be utilized whenever possible. Third, sinks are not only the result of the causes listed above but also occur whenever supporting or reinforcing ribs, flanges, or similar features are used in an attempt to provide functional service without changing the basic wall thickness of a product. If the appearance of a sink on the surface is objectionable, the ribs and transition radius should be proportioned so that their contribution to the sink is minimal. Figure 11-3 is a guide to the dimensioning of ribs. Fourth, molded-in metal parts should be avoided whenever alternate methods will accomplish the desired objectives. If it is essential to incorporate such inserts, they should be shaped so that they will present no sharp inside comers to the plastic. The effect of the sharp edges of a metal insert would be the same as explained in the first point above, namely, brittleness and stress concentration. The cross-section that surrounds a metal insert should be heavy enough that it will not crack upon cooling. A method of minimizing cracking around the insert is to heat the metal insert prior to mold insertion to a temperature of 250 to 300°F so that it will tend to thermoform the plastic into its finished shape. The thickness of the plastic enclosure will vary from material to material. A reasonable guide is to have the thickness 1.75 to 2 times the size of the insert diameter. Fifth, most plastic parts are used in conjunction with other materials. If the use temperature is other than room temperature, certain compensatory steps must be taken to avoid problems arising from the difference in the thermal coefficients of expansion of the different materials. Many plastic materials expand about ten times as much as steel. Thus, careful analysis of the conditions under which the metallic materials are coemployed for functional uses is called for. In the automotive industry many long plastic parts are used in conjunction with metal frames. If proper compensation is not made for the

I

.--,---------~--~--------~ T

,---

-'R :.025-.045 APER-~tlMIN.

Figure 11-3. Rib and wall dimensioning.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 799

difference in thermal coefficients of the materials, buckling or looseness may result, causing noise or a poor appearance. Finally, plastic threads have a very limited strength and may be further degraded if the thread form is not properly shaped. The V-shaped portion at the outside of a female thread will present a sharp inside comer that will act as a stress concentrator and thereby weaken the threaded cross-section. A rounded form that can be readily incorporated in a molding insert will appreciably improve the strength over a V-shaped form. When selftapping, thread-cutting, or thread-form screws are used, their holding power can be increased if either the screws or plastic parts are heated to a temperature of 180 to 200°F at joining time. This will provide thermoforming action to some degree and keep the stress level caused by the joining action at a low point. These possible sources of problems in a molded part should be marked on the part drawing and explained to the product designer for corrective action or creating an awareness of possible product defects from design limitations. This is a necessary step in the chain of events in which the aim is to produce a tool that will provide parts for a good working product. Even if the mold's design, workmanship, and operation are carried out to the highest degree of quality, they cannot overcome a built-in weakness of product design. Sharp Corners

When a part drawing does not show a radius, the tendency is for the toolmaker (while making a mold) to leave the intersecting machined or ground surfaces as they are generated by the machine tool. The result is a sharp comer on the molded part. Such sharp comers on the insides of parts are the most frequent property detractors. The material data sheet should show the difference in impact strength between notched and unnotched test bars. In some materials this ratio is 1 to 30, but in others there is also a decided reduction in the strength of the notched bars. Some show no strength reduction, however. In a shaped product an inside sharp comer is an indication that a certain specified tough material acts in a brittle manner. Sharp comers become stress concentrators. The stress-concentration factor increases as the ratio of the radius R to the part thickness T decreases. An RIT of 0.6 is favorable, and an increase in this value will be of some limited benefit. If certain other details in this problem are properly counteracted, that will help in reducing stress concentration. In Figure 11-4, we can see that a concentric radius, in addition to eliminating the outside sharp comer, can play an important part in holding down the value of the stress concentration. Poor

Good

R,

Figure 11·4. An inside comer.

(min)

= 0.020 in.

800 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The ASTM Izod impact strength of nylon with various notch radii is shown in Figure 11-5. Thus, we see that with a radius of 0.005 in. the impact strength is about 1.3 ft.lb'/in., with an R of 0.020 in. it is 4.5 ft.-Ib'/in., and with an R of 0.040 in. it is 12 ft.lb./in. In most cases a radius of 0.020 in. can be considered a sharp comer as far as end use is concerned, a size that is a decided improvement over a 0 to 5 mil radius; therefore, it should be considered a minimum requirement and be so specified. If this radius of 0.020 in. causes interference, a comer such as shown in Figure 11-6 should be considered. The recommended radius not only reduces the brittleness effect but also provides a streamlined flow path for the plastic in the mold. The radiused comer of the metal in the mold reduces the possibility of its breakdown and thus eliminates a potential repair need. Too large a radius is also undesirable because it wastes material, may cause sink marks, and may even contribute to stresses, from having excessive variations in thickness.

0.040 -: c:

'"

...........

0.020

::I

/

" ro

a:

...u0

.......

./

~

Z

0.005

0.001

o

2

4

6

8

10

12

Izod Impact Strength (ft-Ib/in,l

Figure 11-5. A radius of notched Izod impact strength for nylon.

Radius just large enough to provide clearance over sharp corner.

Figure 11-6. A clearance radius for a sharp corner.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 801

1-1/2

I

1-1/2

t

t

f-L-L-L-+-t~

Figure 11·7. A solid-steel gear.

t

= Thickness through pitch line

I

I

Root diameter

Figure 11·8. A plastic design of the steel gear shown in Figure 11-7.

Uniform Wall Thickness

Wall requirements are usually governed by the load, the support needs for other components, attachment bosses, and other protruding sections. Designing a part to meet all these requirements while still producing a reasonably uniform wall will greatly benefit its durability. A uniform wall thickness will minimize stresses, differences in shrinkage, possible void formation, and sinks on the surface; it also usually contributes to material saving and economy in production. Most of the features for which heavy sections are intended can be modified by means of ribbing, coring, and shaping of the cross-section to provide equivalent strength, rigidity, and performance. Figure 11-7 shows a small gear manufactured from metal bar stock. The same gear converted to a molded plastic would be designed as shown in Figure 118. This plastic gear design, compared to copying the metal gear, saves material, eliminates stresses from having thick and thin sections, provides uniform shrinkage in teeth and the remainder of the gear, avoids the danger of warpage, with its thin web and tooth base prevents bubble formation and potential weak spots, and, having no sink in the middle of the thickness, provides a fun load-carrying capacity for the teeth.

802 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

A,

Poor \

'K: ____________, A, = A, + , LeRl / \

I

Ribs

J/'-_J-.- ---/

t

"Poor" No

"Good" Yes

Figure 11-9. Poor and better design features.

Figure 11-9 illustrates both well and poorly designed cross-sections. If a case exists where some thickness variation is unavoidable, the transition should be gradual, to prevent sharp changes in temperature during solidification.

Wall Thickness Tolerance When relatively deep parts are being designed, a tolerance for the wall thickness on the order of ±O.OO5 in. is usually given. What this tolerance should mean is that a product will be acceptable when made with this tolerance, but that the wall thickness must be uniform throughout the circumference. If we analyze the molding condition of such a part and assume that one side is made to minimum specifications and the opposite to maximum specifications, we find the following taking place: the resistance to plastic flow decreases with the third power of the thickness, which means that the thick side will be filled first, while the thin side will fill from all sides, instead of the gate side alone. This type of filling creates a pocket on the thin side and compresses the air and gases to such a point that the rising temperature caused by compression causes the material to be charred while the pocket is filling up.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 803

The charred material will create porosity, a weak area, and an electrically defective surface. Furthermore, the filling of the thick side ahead of the thin side creates a pressure imbalance generated by the 5 to 10 tons/sq. in. injection pressure that can cause the core to deflect toward the thin side, further aggravating the difference in wall thickness. This pressure imbalance will contribute to mold damage and make part production difficult if not impossible. We may conclude that the wall uniformity throughout the circumference must be within narrow limits, such as ± 0.002 in., whereas the thickness in general may vary from the specified value by ±0.005 in.

Flow Patterns Ultimately, part quality can be considered a direct outcome of a plastic melt's flow behavior in its mold cavity or cavities. Excessive restrictions and obstructions to the flow of material spell trouble in injection molding. Some examples of reduced-flow problems are illustrated in Figures 11-10 through 11-18.

POOR DESIGN

Good

Figure 11-10. A nominal thickness maintained throughout a part can simplify its melt flow. Good design has a minimum exterior radius of Ii times the wall thickness and a minimum of ! its thickness, which maintains uniform section thickness.

' " Poor Flow

Sharp angle Increases stress. InhIbIts ma/enaillow

Poor DeSIgn

I

Matenal must be forced /0 smaller area

~GOodFIOW

Generous radius relIeves stress. promotes

Good

~ma/eriaillow

DeSIgn L-____________________

~

~I

Ma/eflalllows smoothly to larger area

Figure 11-11. Approaches to consider when making changes in wall thicknesses.

~HJPl Heavy port cross section i Undesirable)

Ports cored, cross section (Preferred)

Figure H-ll. Parts having heavy cross-sections are subject to longer cycles and cures, laminations or skins, blisters, warpage, and increased fabricating costs.

Figure H-13. When molding an RP, thin sections sometimes lose strength, because fibers do not flow into a narrow space unless a suitable moldable material is used.

Figure H-14. Avoid having uneven sections, which cause distortion, warpage, cracks, sinks, and strains because of differences in shrinkage from one section to another. Where this situation exists or at least usually exists the problem is normally eliminated by changing the process controls which in tum usually requires a longer cycle time.

!~~e~

Uniform walls

Q I .

uestlonable

W:'

Heavy Thin

, Incorrect

~Thin ~OIlS Heavy

Correct

not critical Center gate preferred

Figure H-IS. The design of wall sections contributes to flow patterns and can be controlled to obtain the best end-product results. 804

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 805

I

-to.

100 in.

~"k

JJ!:i'~ J.IOO in.

Sink mark

Figure 11-16. The thicknesses of adjacent walls and ribs should be about 60 percent of the thickness of the main bodies to reduce the possibility of sink marks and promote better flow.

Figure 11-17. Sharp comers, straight sides, and improper venting impede flow, resulting in strains and possible cracks. Optimum flow requires using maximum radii at comers, reasonable inside and outside tapers, as well as proper vents at mold-parting lines, pockets, and blind holes.

Sinks

Figure 11-18. A crude but effective method for locating sink marks and designing them "out" of plastics parts is in effect to "roll a ball" down the wall's thickness; if it permits a sink, the part will be marked.

Parting Lines Parting lines (PLs) on the surface of a molded product, which are produced by the parting line of the mold, can often be concealed on a thin, inconspicuous edge of the part. Doing so preserves the good appearance of the molding and in most cases eliminates the need for any finishing. Figures 11-19 through 11-26 show parting-line locations on various part configurations.

NOT THIS, - PARIiNC liNt

_o~"'IN' ON'

-

8

--u:u-

PARTINC LINE

THIS

NOT TH IS

Figure 11-19. Examples of parting line locations. The extreme right diagram shows that having sharp comers could cause poor dimensional control or possibly nonflat surfaces with inadequate mold cooling.

~

P,-L

~L ~L '1. ~~ Flat porting preferred

Step porting lines

~L

p.

7L

Figure 11-20. Consider the parting lines when designing parts that may require odd or different radii, contours, and stepped parting lines.

~

)I

Purling line m.smolch Bobbing cross sections

Figure 11-21. Excessive wear on mold parting lines can create a mismatch on a molded part that will appear greatly exaggerated (a 0.0003 in. mismatch can appear to be 0.020 in.) or can create burrs from 0.001 to as much as 0.010 in. 806

~r ~;

[[[ [[

Figure 11-22. An example of a series of possibilities for designed mismatches.

Imperfect airgnment

Designed mismatch

Designed bead mismatch

Figure 11-23. A mismatch in alignment of two molded parts (such as a box and cover) usually is traceable to part warpage or difference in shrinkage; the misalignment at the parting line appears to be improved when a bead or a designed mismatch is utilized.

Solid bond Knurl

Figure 11-24. Allowing parting-line flash in gear teeth and on knurled parts requires expensive deflashing operations. To prevent flash, add a shroud to the gear or a solid band or bead to the knurl at the mold-parting line.

'7L Figure 11-25. Four possibilities for locating parting lines on assembly parts, boxes, and covers while yet maintaining good aesthetics. 807

808 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

-PL

y~j-PL.U:hM'a,e f=-T~ ~

"tSh,i"'a,e

DtShri kag

.

Maid cavity and pin Mold force Figure 11-26. The dimensions and tolerances given for parting lines should include on the final dimensions allowances for potential flashing conditions. All molds for TPs or TSs can open and flash to some extent, depending on the mold design and construction, the melt pressure, the mold-clamping force, and the type of material. For example, if the flash is 0.003 in. at the mold-parting lines, part dimensions A, B, C, D, and E will increase 0.003 in. (left diagram). The drawing at the right shows how to estimate mold components.

Gate Sizes and Locations Because of high melt pressure, the area near a gate is highly stressed, both by the frictional heat generated at the gate and the high velocities of the flowing material. Using a small gate is desirable for separating the part from the feed line, but not for a part with low stresses. Gates are usually two thirds of a part's thickness. If they are that large or larger it will reduce frictional heat, permit lower velocities, and allow the application of higher pressures for increasing the density of the material in the cavity. The product designer should caution the tool designer to keep the gate area away from load-bearing surfaces and to make the gate size such that it will improve the quality of the product. Some examples regarding the design of gates are given in Figures 11-27 through 11-32 [12].

Three -plate mold (ce nter gate I

Edge gate

Center gate

Figure 11-27. Injection molds are especially suitable for producing parts to meet different mold styles, as shown in these three examples.

Figure 11-28. Examples of some melt flows using too small a gate. The melt should fill the cavity uniformly.

tb2Z1F Edge gate

V

Ring gate

V' 6

Disk gate

Tunnel gllte

n

Center gate

m Leg gate

Figure 11-29. Among the various methods available for gating are edge, ring, center, tunnel, disc, and leg gates. Consider the material and design factors before selecting the type of gate.

Figure 11-30. Gate breaking and clipping are low-cost removal methods when the part design permits locating a recessed gate in a hidden area, as in this handle. This type of problem is eliminated by using hot runner molds for TPs and cold runner molds for TSs.

Hot runner gates

Top view

Figure 11-31. In molding a large, flat TP surface using hot or insulated runners, having one or more small gates can help reduce warpages. One way to camouflage the resulting gate scars is to use the bull's-eye design shown (with hot runners no gate problem develops).

Gale ground

Broken

Gale cui or broken

Sheared gale

Figure 11-32. Some gates must simply be cut, sheared with fixtures, ground, or finished, but those broken off from the parts to save finishing costs could have their recesses or protrusions remain. 809

810 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Tapers of Draft Angles

It is desirable for any vertical wall of a molded product to have an amount of draft that will permit its easy removal from a mold. Figures 11-33 and 11-34 show two basic conditions in which draft is a consideration. The first example is the most desirable application of the draft angle. The amount of draft may vary from 1 degree up to several degrees, depending on what the circumstances permit. A fair average may be from l degree to 1 degree. When a small angle such as 1 degree is used, the outside surfacethe mold surface producing it-will require a high directional finish, to facilitate removal from the mold. In the other example, as shown in Figure 11-34, there is a separating inside wall that should generally be perpendicular to the base. The draft in this case should be on the low side (1 degree) so that additional material usage is small, the possibility of having voids close to the base is avoided, and increased cycle time in manufacturing is minimized. Here again the vertical molding surfaces will demand a much higher surface finish, with polishing lines in the direction of part withdrawal. On shallow walls the draft angle can be considerably larger, since the influence of the enumerated drawbacks will be minor. The designer should be cognizant of the need for drafts on vertical walls. If problems are encountered during the removal of parts, stresses can result, the shape of the product can be distorted and surface imperfections be introduced. Some examples of different approaches to the design of drafts are shown in Figures

Angle not limited -

Figure 11-33. External wall taper.

Angle less than 10 Figure 11-34. Internal wall taper.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 811

11-35 through 11-38. One of the difficulties when applying draft to a part is the creation of heavy walls. The potential problem of removal can be remedied by the using parallel drafts on walls, as shown in Figure 11-38, where the walls are kept uniform. A guide to determining the amount of dimensional change required due to the draft angle is given in Figure 11-39 . .090·

.070·

.• " .080·

100· · "-

Co"tcl

I nCOII'tt:1

Figure 11-35. An example of side-wall taper for long-draw products .

.070"

080·

.080·

Slmplr

Dlfftcult

Figure 11-36. An example of designs for long-draw products. The left view has little clearance, but the right side is simplified by the clearance allowed by the walls' taper.

DRAFT

Figure 11·37. An example of using mUltiple drafts to permit ease of part removal.

O/MfNSION OIFFHI£NCE

Figure 11-38. Using parallel drafts with heavy walls.

"J

.'

lAI·

lAo

liz •

I"

2"

3"

4·

0 .0022

0.0044

0.0081

0.0175

0.0349

0 .0524

0.0699

.~

I"

~: ~

..

"~

~.

~

40

~ ,

2"

~

,

0.0044

0.0088

0.0174

0.0350

0.0698

0 .10A8

0.1398

0.0066

0.0132

0.02fi1

0.0525

0.10A1

0 .1512

0 .2091

....

,

.~

3"

f<:'!

l- 80

I:

;:' 4"

~

-

....

•

·i,.....:1

0.0088

0.0176

0.03A8

0.0700

0.1390

0.2096

0 .2796

4·

en

a:

0.0110

5"

0.0220

0.0A35

0.0875

0 .1745

0 .2620

0.3495

~ 120

..,

1&.1 I-

2

...J ...J

0·.0132

6"

0.0264

0.0522

0.1050

0.2094

0 .3144

0.41M

- 160

.... ~

7"

0.015A

0 .0308

0.0609

0.1225

0.24043

0.3668

0.4893

0 .0176

0 .0352

0.0696

0 .1400

0.2792

0.4192

0.5592 ~

0 .0198

0 .0396

0.0183

0 .1575

0.3141

0 .4716

0.6291

,

~.

8"

r.-:

200

~ 9"

'1 ~::.. .'

~

10·

~ ~;-:

1·

240 0.0220

0.0440

0.0870

0.1750

0 .3490

0 .5240

0.6990

Figure 11-39. The relation of the degree of taper per side to the dimension in in.lin. 812

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 813

Weld Lines

With molded parts that include openings (holes), problems develop. In the process of filling a cavity the flowing plastic is obstructed by the core, splits its stream, and surrounds the core. The split stream then reunites and continues flowing until the cavity is filled. The rejoining of the split streams forms a weld line that lacks the strength properties that exist in an area without a weld line because the flowing material tends to wipe air, moisture, and lubricant into the area where the joining of the stream takes place and introduces foreign substances into the welding surface. Furthermore, since the plastic material has lost some of its heat, the temperature for self-welding is not conducive to the most favorable results. A surface that is to be subjected to loadbearing should not contain weld lines. If this is not possible, the allowable working stress should be reduced by at least 15 percent. Some examples of different aspects pertaining to weld lines are shown in Figure 11-40.

(

~'>

'-----~.

...........

~ -­ ~Flow oround holes

Flow oround rl bs

L "--0------·I~~eld I

I

"

.. ~..::--\ ' ... -.....

"

,

...

~

I I

______'

I

I

Flo w at gate

Not this

This

G (

I

) ..

~ Weldlme

Opposite flow fronts produce a weld line that could also contain entrapped air.

-~::::-t:~Y:;.:.u::L lnSlde or

outs Ide

center

gOle

Direct flow in one direction to avoid undesirable welds.

Figure 11·40. Some examples of flow patterns to consider during the design stage to eliminate or at least minimize weld lines to obtain maximum strength.

814 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Vents, Trapped Air, and Ejectors Vents can be used to release air entrapped in cavities as well as moisture and gases formed from melts that were not previously removed during processing (see Chapters 7 and 8). Vents and ejectors may be considered when designing such parts, as shown in Figures 11-41 through 11-46.

yents required

,Plastic

F~P/ZZ/

vent

.~

Kick-out

~-out and/ or vent pins

Kick-out and/ or vent ~ stnpper sleeve

/

Kick-out

Vents

Figure 11-41. Parting lines can provide venting, but in deep pockets kick-out pins or vent pins incorporated into a mold can vent or remove air or gas.

Knockoutand~: __ vent blade

,-Plastic Edge! gate

section

Figure 11-42. Center-gate this type of part, if possible. If aesthetics prevent this approach and edge gating is requested, the material will flow around the edge and to the thin top section last. Thus, air-and any gas that may be present because of melt-will be trapped in the top, which may not be acceptable.

Trapped gas causes burn spot in center

Heavy wall

I

,/

,/

,/

,/

J---~--~

....i I - - -

Edge gate

Figure 11-43. Here the melt flow passes through heavy sections first, pushing air and gases into the topmost thin sections, where they become trapped. This design should either be center gated at its top, use thinner walls, or be designed, if possible.

UnevenWolI Section

Incorrect

Correct

POINTOF II!OLD WEAR

fS

GAS POCKETS

II I

(J ,,/ BLISTERS

I

j

\..._---BETTER DESIGN

Figure 11·44. These cross-sectional views show the results of uniform versus nonuniform wall thicknesses in eliminating air and gas entrapment.

Figure 11·45. To aid in eliminating air and gas entrapment, aim to reduce heavy or thick sections like those on the left of each pair.

Figure 11·46. Gates and ejector pins create marks and blemishes. Parts drawings should indicate surfaces that can be marred without creating problems, to aid in properly locating gates and ejector pins. 815

816 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Undercuts

Undercuts, whether external or internal, should be avoided if possible. In cases where it is essential to incorporate them in part design, a great many can be realized by appropriate mold design in which either sliding components on tapered surfaces or split cavity cam actions will produce the needed undercut. This obviously means increased tool cost, in the neighborhood of some 15 to 30 percent. Some conditions will, however, permit incorporating undercuts with conventional stripping of the part from the mold. Certain precautions are necessary in order to attain satisfactory results. First, the protruding depth of the undercut should be two thirds of the wall thickness or less. Second, the edge of the mold against which the part is ejected should be radiused to prevent shearing action. Finally, the part being removed should be hot enough to permit easy stretching and return to its original shape after removal from the mold (see Figures 11-47 through II-54). How easily the task can be accomplished depends on the material's elasticity and springback. Many threaded plastic caps are stripped from the cores instead of being unscrewed. Coarse threads with the crest of the core thread rounded and a material with good elongation and ability to spring back make it feasible to apply conventional part stripping. The undercut problem can be solved by the cooperation of the designer, moldmaker, and processor, since each product configuration presents different possibilities.

PL-

ndercut

Port cross section

Cam cross section

Molded part

Figure 11-47. A split mold with mold sliding-cam actions for outside undercuts will show seams in the molded part that may be objectionable. They can be partially hidden by locating them in the least-conspicuous areas when designing the molds.

Bf[Y rffP : :: ~der cu t formed

A

.

PL-

"Id pin Force and

-PL

cavity eams cutoff ............. Port cross sections

Figure 11-48. Inside undercuts can be made with removable wedges or ejector pins and ejector wedges, but seams will still show. The left view shows undercuts molded with removable wedges, the right view the mold knockout pin positioned through the permissible opening in the part and cavity cutoff approaches.

Figure 11-49. The mold for external undercuts to retain part in the required female cavity prior to ejection. When they are required to be retained on a male cavity undercuts are on the male plug.

Figure 11-50. An example of an undercut made possible by using an ejector pin.

~~ - -

Under cuI

Molded port Figure 11-51. The goal should be to avoid undercuts wherever possible, especially in small holes and on projections. With rigid plastics, undercuts usually require complex, more expensive molds and particularly molding operations. However, with flexible types of plastics stripping the mold will permit ease of removal from the mold. This action can be used with rigid plastics when it is practical to remove a part while the plastic is still soft, just prior to when the part should be removed. When certain parts that we removed at an earlier stage, cooling fixtures may be applicable. 817

I ""- ", ", ", I"""'""

", "Porting Iin~ n",essift1/~d by

spIll co,tily .... .....

",>- - Exler-nol

....

undercuts

Figure 11·52. External undercuts with this split-mold design eliminate the undercut problem but develop a parting line.

CORE & CAVITY "KISS-OFF" CREA TlNG OPENING IN SIDE WALL

J

USUALL Y REQUIRES 5° DRAFT MINIMUM

DIRECTION OF MOLD TRAVEL INJECTION MOLDED PLASTIC PART SIDE WALL OPENING UNDERCUT

HINGED DOOR WITH MOLDED-IN "THRU-HOLE" UNDERCUT CORE AND CA VITY CORE EXTENDS KISS-OFF THROUGH OPENING IN PART ---+-2IiO~FW;::.e:;:4- LOCKING TAB

ALTERNATE CORING FROM BOTH SIDES CREA TED THRU HOLE FOR HINGE PIN

... DIRECTION OF MOLD TRAVEL LOCKING TAB UNDERCUT

Figure 11·53. Creating undercuts with simple tooling. 818

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 819 I

+--1

Figure 11-54. Avoiding an undercut.

Blind Holes

In regard to molding products that include holes, it is important to ensure that sufficient material surrounds the holes and melt flows properly. Some examples of approaches and problems that can develop are to be seen in Figures 11-55 through 11-62 [2]. A core pin forming blind holes is subjected to the bending forces that exist in the cavity, due to the high melt pressures. Calculations can be made for each case by establishing the core pin diameter, its length, and the anticipated pressure conditions in the cavity. From technical handbooks we know that a pin supported on one end only will deflect forty-eight times as much as one supported on both ends. This suggests that the depth of hole in relation to diameter should be small, in order to maintain a straight hole. Sometimes a deep, small-diameter hole is needed, as in pen and pencil bodies. In this case the plastic flow is arranged to hit the free end of the core from four to six evenly spaced gates that will cause a centering action, and the plastic will continue flowing over the diameter in an umbrellalike pattern to balance the pressure forces on the core. When this type of flow pattern is impractical, an alternative may be a through hole or tube formation combined with a postmolding sealing or closing operation by spinning or ultrasonic welding. At the other extreme, consider a 1 in. diameter core exposed to a pressure of 4,000 psi with an allowance for deflection of 0.0001 in. and see how deep a blind hole can be molded under these conditions.

Mold port way and complete with drill Mold

Figure 11-55. Holes impractical to mold must be drilled, but they must not be so close to edges or comers that cracks result. A small-diameter hole is difficult to drill along its intended direction to any great depth, so the most practical approach in many products is to mold it part way, then drill it the remainder of the distance.

820 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

According to engineering handbooks the deflection, a, may be used to calculate this depth, I, as follows:

wP

a=-=

8EI

8

x

1,000r x 0.049

30,000,000

x

0.0039

Trd 4

I = 64 = 0.049d4 ~

where W

0.0625

= 0.0039

= total load =

I

= 0.0625 x

psi x d x I (projected area of pin)

= 4,000 x t x I = 10001 = length of pin; d = diameter of pin

a = deflection = 1110,000

_1_ _

_

10,000 - a - 8

x

1,OOOr

30,000,000

x

0.049

x

_ d fl

0.0039 -

. n e echo

r = 0.0045864 I

= 0.26" or slightly over diameter size

If a hole deeper than 0.26 in. is needed, we can calculate the amount of deflection that will be present and whether the calculated deflection will produce an opening of the necessary tolerance, as well as the kind of stress that will be generated in the pin, along with its corresponding life expectancy. Let us now assume that the desired depth of hole is i in. The deflection is then calculated as follows:

wP 1,000r a =-= 8EI 8 x 30,000,000 X 0.049 14

=

a

=- = 0 .00043" deflection 45,864

X

0.0039

0.0198 0.0198

.

The maximum stress, S, is found by S

= WI =

Z

=

0.098d 2

S

=

11 ,480 psi

2Z

1,000f 2 X 0.006125

=

=

1,000 x 0.1406 2 x 0.006125

0.006125

These results indicate that a hole with 0.0004 in. variation may be satisfactory, and if the pin is made of a springlike material, properly heat treated, it should last a long time.

-/" 4

Figure 11·56. A basic guide for good blind-hole design.

Port cross section

Mo td cross sect ion

;jcf,,;;~ Port cross section

Mold pin could bp.nd

Figure 11·57. When holes are too near an edge or comer, material may not "weld" properly around mold pins. Also, the flow of the melt can bend mold pins for blind holes when their length exceeds a diameter of 2! times and when holes are to be long with small diameters, even if these are anchored at both ends.

Suitable hole location Figure 11·58. When delicate parts are molded, mold designs should provide for proper hole locations and for replaceable blades and members in wall sections.

Square holes where possible ClCJCJOD ClCJCJCJD

Mold

Molded part Figure 11·59. Whenever possible with close-tolerance parts, consider the potential of using a laminated type of mold. 821

NOT THIS

THIS

Figure 11-60. Methods for molding holes or openings in side walls without undercutting mold movement. Courtesy E. I. Du Pont de Nemours & Co., Inc.

NOT THIS

NOT THIS

THIS

THIS

Figure 11-61. Whenever possible, chamfering should be used on open holes, because it reduces or eliminates the potential for rough molded comers, cracks, and the like.

near edge PL-

PL-"---Figure 11-62. Avoid part cross-sections that are so thin that they will be prone to cracking, sharp edges that will chip or break, and holes near enough the edges to cause them to chip. Sections should also not be so thin that melt will flow and weld in their thin edges. Certain plastics are prone to this type of action. 822

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 823

Bosses Bosses and other projections from the nominal wall are commonly found in injectionmolded plastic parts. These often serve as mounting or fastening points. Figure 11-63 shows some typical boss designs, along with common problems. As with rib design, avoiding overly thick wall sections is important, to minimize the chance of appearance or molding problems. When bosses are designed to accommodate self-tapping screws, the inside diameter and wall thickness must be controlled to avoid excessive buildup of hoop stresses in the boss. Ribs are frequently used in conjunction with bosses when lateral forces are expected. Special care must be used with tapered pipe threads, since they can create a wedging action on the boss. If there is a choice, the male rather than the female pipe thread should be the one molded into the plastic [2].

POOR

-

BETTER

BETTER

CORE FROM BELOW(PARALLEL DRAFT)

POOR

BETTER

BETTER

'ht

CONNECTING BOSSES TO OUTSIDE WALLS WITH RIBS

Figure 11-63. Design guides for molding bosses.

USE GUSSETS RA THER THAN VERY THICK BOSSES WHEN RESISTANCE TO LOADING IS REQUIRED

824 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Coring

The tenn coring in injection molding refers to the addition of steel to the mold for the purpose of eliminating plastic material in that area. Usually, coring is necessary to create a pocket or opening in the part, or simply for the purpose of reducing an overly heavy wall section (see Fig. 11-64). For simplicity and economy in injection molds, cores should be parallel to the line of draw of the mold. Cores placed in any other direction usually create the need for some type of side action (either a cam or hydraulic cylinder) or manually loaded and unloaded loose cores [2]. Blind holes in molded plastic parts are created by a core supported by only one side of the mold. The length of the core and depth of the hole are limited by the ability of the core to withstand the bending forces produced by the flowing plastic without excessive deflection. For this reason, the depth of a blind hole should not exceed three times its diameter or minimum cross-sectional dimension. For small blind holes with a minimum dimension below tin., the Ud ratio should be kept to two. With through holes the cores can be longer, since they are supported by the opposite side of the mold cavity. Sometimes the cores can be split between the two sides and interlocked when the mold is closed, allowing for the creation of long through holes. With through holes, the overall length of a given-size core can generally be twice as long as that of a blind hole. Sometimes, even longer cores are necessary. The tool can be designed to balance the hydraulic pressure on the core pin, thus limiting the deflection. Internal Plastic Threads

The strength of plastic threads is limited, and when molded in a part involving either an unscrewing device or a rounded shape of thread-similar to bottle-cap threads-they can be stripped from the core. Screw threads, when needed, should be of the coarse type and have the outside of the thread rounded so as not to present a sharp V to the plastic, which can produce a notch effect.

POOR DESIGN

SUGGESTED AL TERNA TlVES

SINK MARKS

~ WARPAGE

POOR RECTANGULAR PART WITH ROUND HOLES

Figure 11-64. Examples of coring in injection molds.

MA TCH OUTSIDE CONFIGURA TlON TO INSIDE CORES

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 825

If a self-threading screw can be substituted, it will not only appreciably decrease mold maintenance and mold cost but most likely, with proper type selection, also give better holding power. A screw that has a thin thread with relatively deep flights can give high holding power. If the screw or plastic is preheated to about 121°C (250"F), a condition of thermoforming in combination with material displacement will exist, thereby improving the holding power. When male plastic threads are being considered, the coarser threads are again preferred, and the root of the thread should be rounded to prevent the notch effect (see Figs. 11-65 and 11-66).

Molded-in Inserts If metal inserts are to be molded into a plastic product, their shape would present no sharp edges to the plastic, since the effect of the edges would be similar to that of a notch. A knurled insert should have the sharp point smoothed, again to avoid the notch effect. The practice of molding inserts in place is usually employed to provide good holding power for plastic products, but there are drawbacks to this method: it is dangerous to have an operator place an arm between the mold halves while the electrical power to the machine is turned on. It normally takes a pin to support the insert, and since this pin

Initially

Countersink

_--r~

Mold or machine <""<"'<,........'<""<""''"''''''~~ countersink

Pa rt ---t."'<"o:l~

Molded threads Tapped Figure 11-65. Holes having molded-in threads or ones that are to receive tapped threads should have molded-in countersinks to avoid chipping and burrs. If a threaded pin is being used to mold threads in a through hole, the lead thread on the mold pin should have a pilot thread for support.

Threads

TX I

I

t

ReqUires special mold design

Preferred design (X=X)

Figure 11·66. Automatic unscrewing is expensive when molding, unless the plastic part is designed to prevent turning during the unscrewing cycle. Where thread strength is a requirement, fine threads should not be molded, particularly in RPs and .composites where the threads could be mainly resin and thus very brittle. However, with proper molding procedures, based on type of compound or composite used and the design of the mold being gained through experience, no problem should develop.

826 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

is small in relation to the cored hole for the insert, it is easily bent or sheared under the influence of injection pressure. Should the insert fallout of position, there is danger of mold damage. Also, the hand placement of inserts contributes to cycle variation and with it potentially product quality degradation. Some of these problems can be overcome by higher mold expenditures, as for example shuttling cavities (see Figs. 11-67 through 11-80 and Table 11-3) [2]. On the other hand, the desired results in fastening can be attained by other means, as for example by coring holes in the part that will permit ultrasonic welding of inserts in place, coring a hole in the part that will be of a size when the part is removed from the mold that will permit a slight press fit plus a gain in the holding power from postmolding shrinkage, or coring a hole in the part that will permit dropping the insert and providing a retaining shoulder by spinning or ultrasonic forming. All these. assembly methods require the same time to perform as placing inserts in the mold, but they also lower machine time. There are probably several other means of accomplishing the desired result that depend on the circumstances at hand. In any event, molded-in inserts, in the long run, usually prove costlier and so should be avoided.

Table 11-3. Suggested Minimum Wall Thicknesses for Inserts of Various Diameters, in. (mm) Diameter of Inserts, in . Plastic Material ABS Acetal Acrylics Cellulosics Ethylene vinyl acetate F.E.P. (fluorocarbon) Nylon Noryl (modified PPO) Polyallomers Polycarbonate Polyethylene (H.D.) Polypropylene Polystyrene Polysulfone Surlyn (ionomer) Phenolic G.P. Phenolic (medium impact) Phenolic (high impact) Urea Melamine Epoxy Alkyd Diallyl phthalate Polyester (premix) Polyester T.P.

.125 (3.17)

.250 (6.35)

.062 (1.57) .093 (2.36) .078 (1.98)

.093 (2.36) .156 (3.96) .140 (3.56)

.062 .093 .125 .020 .125

.125 .156 .187 .030 .187 .187 .125 .125

.140 .187 .218 .040 .187 .250 .140 .187

(3.17) (1.57) (2.36) (3.17) (1.02) (0.64) (3.17) (1.57) (3.17) (1.57) (3.17) (3.17)

(1.57) (2.36) (3.17) (0.51) (3.17) .12~ (3.17) .093 (2.36) .062 (1.57)

(6.35) (3.17) (3.17) (6.35) (2.16) (1.52) (6.35) (3.17) (6.35) (3.17) (6.35) (6.35)

.500 (12.7)

.375 (9.52) .500 (12.7) .187 (4.75) .250 (6.35) .187 (4.75) .250 (6.35) .375 (9.52) .500 (12.7) N.R. N.R. N.R. N.R. .375 (9.52) .500 (12.7) .187 (4.75) .250 (6.35) .375 (9.52) .500 (12.7) .187 (4.75) .250 (6.35) .375 (9.52) .500 (12.7) .375 (9.52) .500 (12.7) Not Recommended Not Recommended .125 (3.17) .187 (4.75) .187 (4.75) .218 (5.53) .156 (3.96) .203 (5.16)

.125 .062 .093 .125 .040 .025 .125 .062 .125 .062 .125 .125

.250 .125 .125 .250 .085 .060 .250 .125 .250 .125 .250 .250

.375 (9.52)

(3.17) (3.96) (4.75) (0.76) (4.75) (4.75) (3.17) (3.17)

(3.56) (4.75) (5.53) (1.02) (4.75) (6.35) (3.56) (4.75)

.187 .218 .312 .050 .312 .312 .187 .250

(4.75) (5.53) (7.92) (1.27) (7.92) (7.92) (4.75) (6.35)

.750 (19.0)

1.00 (25.4)

.750 (19.0) .375 (9.52) .375 (9.52) .750 (19.0) N.R. N.R. .750 (19.0) .375 (9.52) .750 (19.0) .375 (9.52) .750 (19.0) .750 (19.0)

1.00 (25.4) .500 (12.7) .500 (12.7) 1.00 (25.4) N.R. N.R. 1.00 (25.4) .500 (12.7) 1.00 (25.4) .500 (12.7) 1.00 (25.4) 1.00 (25.4)

.250 (6.35) .312 (7.92) .281 (7.14)

.312 (7.92) .343 (8.71) .312 (7.92)

.250 .312 .343 .060 .343 .343 .250 .375

.281 .343 .375 .070 .375 .375 .281 .375

(6.35) (7.92) (8.71) (1.52) (8.71) (8.71) (6.35) (9.52)

(7.13) (8.71) (9.52) (1.78) (9.52) (9.52) (7.14) (9.52)

PUSH-IN TYPE INSERTS ADVANTAGES - SPEED AND LOW EQUIPMENT COSI DISADVANTAGES - HIGH INDUCED STRESS AND ONLY FAIR HOLDING POWER.

SELF-THREADING INSERTS ADVANTAGE INSTALLED WITH MINIMAL EQUIPMENI DISADVANTAGES - SLOW AND THEY CAN CREATE HIGH STRESSES.

• I EXPANSION TYPE INSERTS ADVANTAGE NO INSTAllATION EQUIPMENI DISADVANTAGES - LOWER PERFORMANCE AND MODERATELY HIGH INDUCED STRESS.

i

Figure 11-67. Examples of common threads metal inserts.

Before

After

Figure 11-68. The technique of using hot-roll plastic over inserts to obtain strengthened anchorages. 827

ft

-k-Insert hole

Insert

T.l.R.0.003m Mold pin

Figure 11·69. Either inside or outside diameters can be used to hold inserts in place during molding. Tight tolerance (± 0.001 in. with a maximum TIR of 0.003 in.) are required.

Inserts

Insert assembly

Inserf assembly

Figure 11·70. Assembling inserts after molding prevents problems such as potential melt flow over metal surfaces during molding and the scratching of plated inserts during any required deflashing. Pressed-in inserts require holes sized for proper fits.

!O'OOlin.

Insert Plastic Supported ot 0.0.

~

,-+0.001 in.

Insert Plostic

Supported at 0.0.

I I

Supported at 1.0.

Figure 11·71. Generally, through inserts must be molded ±O.OOI in. of length to ensure their making contact with both mold surfaces.

Assembly

~ ,

.

Shoul~dr Knurl Groove

~ ~~ !-Incorrect-!

+-1---

Figure 11·72. Knurls and grooves can serve to anchor inserts in molded products. They should be flush with the top surface or in contact with an assembling member to prevent their jacking out.

828

...,....-- Recessed shoulder

Blind hole insert Figure 11-73. During molding, any loose steel inserts falling into the molds will damage their cavities, but brass or soft metals like aluminum will usually crush, with minimal or no damage to the mold. The flow of plastic into the interior of an insert is impeded by using a blind-hole insert and a shoulder (about if x if in.) around the insert opening.

Molded in place

Hole molded or drilled

Figure 11-74. Some of the methods of assembling inserts include mechanical pushing and sonic pushing, or allowing hot plastic to shrink around the insert.

Plastic material

Normal shrinkage

Some materials may craze with this type insert Low shrinkage

Figure 11-75. Plastic generally shrinks away from metal when it is molded inside a metal insert. Both the insert's structure and the type of plastic used will determine the amount of shrinkage.

Insert~

)

~PIQSlie ~::.~ tparts

t:,

Inserts Plastic

Figure 11-76. Metal stamping and inserts of various shapes are usable in many ways. To prevent cracking and crazing during aging under use surround all inserts with reasonably thick plastic walls. Thin walls can crack and too-thin walls can also show sink marks. 829

Irregularly shaped inserts ~------~

Plastic

r---------,

Plastic

Figure 11·77. Irregularly shaped inserts protruding from the plastic complicate mold construction and add costly deftashing operations to the product.

I r -- ---:;-- i. 11... _____ .. '- _____II

~I~r~-~~~-~-~-~-~-~-~-~-I=I~"""Insert I L... _ _ _ _ _ _ _ _ _ _ _ _ _ .)1

--------------,

Low shrinkage possible crocks ond warpage Medium shrinkage Full shrinkage desirable conditions

Figure 11·78. When using metal embedments, the dimensions and tolerances of the plastic parts will be difficult to estimate, because metal tends to prevent plastic from shrinkage. The insert structure influences the final dimensions by controlling the movement of the plastic during aging, setting up stresses that can cause cracks and warpage.

wJl:' DUring molding

After shrinklnQ

Figure 11·79. Molded materials with high shrinkage values during cooling will shrink below the metal inserts, allowing the insert ends to extend beyond the part.

ULTRASONIC TYPE INSERTS

STUD TYPE ULTRASONIC INSERT

Figure 11·80. These examples of ultrasonic inserts, designed for excellent performance, result in fast action with very little induced stress. 830

1/32"(0.8mm) Edge

Easily ,Broken

Incorrect

1/32" (0.8mm)

1/32" (0.8 mm)

Figure 11·81. Thread design guides. Courtesy E. I. Du Pont de Nemours & Co., Inc.

iT Type 23 (ANSI T)

Type 25 (ANStBT)

Type with cutting edge on point (Hi·Lo)

Figure 11·82. Examples of self-threading screws.

Screws For mechanical assemblies using screws (see Figs. 11-81 through 11-83), they can be detached indefinitely, with the exception of self-tapping screws, which can be loosened and retightened only a limited number of times. The best guideline for the designer is to prefer any assembly design that converts eventual tensile loads to compression loads. For those plastics that are subject to crazing or stress cracking, compression loads tend to reduce this problem. When feasible, use metal-to-metal force-locking connections, particularly with many of the TPs, to release plastics significantly from stresses. The forces that can be applied with a single small screw can be surprisingly high. As in metals, consider the use of torque-limiting wrenches in designs where the degree of loading is critical. External and internal threads can be molded economically in plastic parts (see Fig. 1181). Screw threads produced by the mold itself using rotating cores, split inserts, or collapsible cores [12] will eliminate the normally expensive postmolding threading operations. Coarse threads can be molded easier than fine ones, so threads less than 32pitch should be avoided. American Standard screw threads should be designed and molded carefully. If the threads end up forming notches, a reduction in impact strength and ultimate elongation under tensile stress can be significant, depending on the type of plastic used. With certain applications and materials, trapezoidal and knuckle threads are better. 831

832 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Generally, the length of thread used should be more than 1.5 times the diameter, the section thickness around the hole more than 0.6 times the diameter. Avoid having feather edges, and limit tightening with the bolt shoulder. Bottle caps made from different plastics are extensively used. Some closures are of the simple cork type, but most are of the screw type. Strong, accurate threads can be molded, which represent undercuts. Simple designs should be used when permitted, such as wide-pitch threads. The thread should be designed to start about if in. from the end of the face perpendicular to the axis of the thread. It is usually practical to mold up to 32 threads per in.; more than this number can give certain molders trouble. Self-threading screws are an economical means of securing separable plastic joints. They can be either thread cutting or thread forming (see Fig. 11-82). To select the correct self-threading screw, the designer should know which plastic will be used and what its mechanical properties are, particularly its modulus of elasticity. These self-threading screws are driven into the molded part, eliminating the need for a molded-in thread or a secondary tapping operation. They differ in their thread spacing and body design. Thread-

Table 11-4. Simplified Example of the Effect of Rib and Cross-section Changes Cross Section Area, square inches Geometry

Maximum Deflection, inches

(mm2 )

Maximun Stress, psi (mPa)

0.0600 (38.7)

6800 (46.9)

0.694 07.6)

0.0615 (39.7)

2258 05.6)

0.026 (0.66)

0.1793 015.7)

2258 05.6)

0.026 (0.66)

(mm)

09.1 mm)

f--

0 75----11

~ ORIGINAL SECT/ON

.L

T

0080

(2.0)

~T

0040 '

---+-I

04,00

'..- ,I

1

ORIGINAL SECTION WITH RIB

~0'75-il

~ THICK SECTION

0.239

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 833

forming screws, which provide the highest stripping torques, have less tendency to damage threads in repeated assembly operations than do other types. The thread-forming screw displaces material as it is being installed in the receiving hole. This type of screw induces high stress levels in the plastic part, so it is not recommended for use with certain plastics such as those with a low modulus, unless careful procedures are used in forming the threads. Screws or threaded bolts with nuts require through-going holes but provide an easy assembly system. As shown in Figure 11-83, washers are recommended to distribute the load upon a larger area, wherever feasible or required. If a screw is tightened too far, excessive bending or tensile stresses will easily be created, possibly causing cracking based on stress-to-failure data curves. A change in design or the use of a spacer can convert tensile into compressive stresses. Different screw- and bolt-heads can be used, but flat-underside types of heads are best.

Ribs If there is sufficient space, the use of ribs is a practical, economic means of increasing the structural integrity of plastic parts without creating thick walls (see Chapter 3). Ribs are provided for spacing purposes, to support components, and for other uses. Table 114 shows a summary of the results of using a rib design. Although the use of ribs gives the designer great latitude in efficiently tailoring the structural response of a plastic part, ribbing can result in warping and appearance problems. In general, experienced design engineers do not use ribs if there is doubt as to whether they are structurally necessary. Adding ribs after the tool is built is usually simple and relatively inexpensive since it involves removing steel. There are certain basic rib-design guidelines that should be followed (see Fig. 11-84). The most general is to make the rib thickness at its base equal to one-half the adjacent wall's thickness. With ribs opposite appearance areas, the width should be kept as thin as possible. In areas where structure is more important than appearance, or with very low shrinkage materials, ribs are often 75 or even 100 percent of the outside wall's thickness. As can be seen in Figure 11-85, a goal in rib design is to prevent the formation of a heavy mass of material that can result in a sink, void, distortion, long cycle time, or any combination of these problems.

Figure 11-83. Examples of screw and bolt assemblies with the better designs on the right side.

834 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

DbJbJbJ

gggg

gggg

188BB Figure 11-84. An example of a guideline in ribbing-a bidirectional ribbing (right) can reduce or eliminate sagging or bending; a circular or diagonal ribbing (left) reduces or eliminates twisting. All ribs should have a minimum of lO of draft per side and a minimum radius of 0.005 inches at the base. Generally, the draft and thickness requirements will limit the height of the rib. Therefore, multiple, evenly spaced ribs are preferred to single, large ribs. Wherever possible, ribs should be smoothly connected to other structural features such as bosses, side walls, and component-mounting pads. Ribs need not be constant in height or width and are often matched to the stress distribution in the part. The first step in designing a rib is to determine the dimensional limitations, followed by establishing what shape the rib will have to be to realize a part with good strength and satisfactory appearance that can be produced economically. Figure 11-86 shows proportional dimensions of rib versus thickness. This arrangement will minimize voids (sinks), stresses, and shrinkage variations, and lends itself to trouble-free molding. If performance calculations indicate wall thicknesses well above those recommended for a particular material, one of the solutions to the problem is to achieve equivalent cross-sectional properties by ribbing. Heavy walls will cause a reduction in properties caused by poor heat conductivity during molding, thus creating temperature gradients throughout the cross-section, with resultant stresses. Cycle times are usually exceptionally long, another cause of stress. Also, close-tolerance dimensions are difficult to maintain, material is wasted, quality is degraded, and cost is increased. Solid plastic wall thicknesses for most materials should be below 0.2 in., preferably around 0.125 in., in the interest of avoiding these pitfalls. In most cases ribbing will provide a satisfactory solution; in others reinforced material may have to be considered. An example of how ribbing can provide the necessary equivalent moment of inertia and section modulus follows. A flat plastic bar I! in. wide, i in. thick, and 10 in. long, supported at both ends and loaded at the center, was calculated to provide a specified deflection and stress level under a given load. The favorable material thickness of this plastic is 0.150 in., and its rib proportions would be as in Figure 11-86. Using judgment as a guide, it would appear that the Ii in. width would require about two ribs. So, as a starting point, let us calculate the equivalent cross-sectional data as if we were dealing with two T-sections. According to standard engineering handbooks (under "Stress and Deflection in Beams," "Moments of Inertia," etc.), resistance to stress is expressed by the moment of inertia, and the resistance to deflection by the section modulus. By finding a cross-section with the two factors equivalent, we can assure equal or better performance in the ribbed design compared to the thick wall without ribs.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 835

The stress = WlI4Z and the deflection = W[2/48EI, where W = load, I = length of beam, Z = section modulus, E = modulus of elasticity, and I = moment of inertia. For the flat bar I = bd3/12, b = Ii, d = i; or I = (1.5 X 0.375 3)/12 = 0.0066 for the rectangular bar. The section modulus is

Z

I

0.0066

Y

0.1875

= - = - - = 0.0352

The moment of inertia of aT-section is

where b = 0.75, s = 0.15, h + h(T + t)/2. Therefore,

y

= 0.6, t = 0.8, T = 0.1, d = 0.75, A =

= d - [3bs 2 + 3ht(d + s) +

I(T - t)(h

area

= bs

+ 3s)]/6A

where y is the distance from the neutral point to the extreme fiber. Substituting the values into the formulas we have

A

= (0.75

y

= 0.75

x 0.15) + 0.6(0.1 2+ 0.08) - {(3

x 0.75

X

0.15 2)

= 0.1665

+ [3 x 0.6 x 0.08(0.75 + 0.15)]

+ 0.6(0.10 - 0.08)(0.6 + 0.45)}/(6 x 0.1665) _ 0 _ 0.0506 - .75

+ 0.1296 + 0.0126 0.999

= 0.75 - 0.193 = 0.557 I

= n[(4 x 0.75 x 0.153 ) + 0.63{(3 x 0.08) + 0.1}] - 0.1665(0.75 - 0.557 - 0.15)2

Z

= n[O.0102 + 0.073]

- 0.1665 x 0.00185

= 0.00693

= 0.00662

- 0.00031

= 0.00662 = 0.0119 0.557

Two of the T-sections would provide a higher moment of inertia and decreased section modulus than a sandwich structure. When placed on the end, the two ribs would make a channel that would give a moment of inertia of 0.018 and a section modulus of 0.035,

836 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~ ~VOIO

~ SIN

RIS TOO THICK

OT~Y'IB

USE A 0 GER DEEPER RIB

BEST

BtiTTtiR

----f

11 MIHIMVM

~ ~ VSIi MVLTIP!.£ RIBS

CORti fHIC FIlS FROM BACK

BETTER DES I GN

Figure 11-85. Examples of rib design characteristics.

2/3 t Figure 11-86. Rib proportion versus thickness.

values that are more than adequate for the purpose (see Fig. 11-87). It should be noted that the two-rib construction forming a channel would require only 70 percent of the material used in a solid bar (see Fig. 11-88). Other means of stiffening surfaces can also be used, if appearance permits, where areas can be domed and corrugated. The basic goal in any action that leads to greater rigidity is to specify a practical wall thickness that will optimize strength and processing and thus

DESIGN fEATURES THAT INFLUENCE PERFORMANCE 837 Force

0.375 in.

Rectangular bar, 0375 in. X 1.5 in .

•- - - - - 1 0 i n . - - - - . , J

1 Figure 11·87. A single-rib shape with good moldability that gives a section modulus and moment of inertia equivalent to a rectangular bar.

result in high-quality products. In addition to ribs, other features protruding from a wall, such as bosses or tubular shapes, should be treated similarly as far as transition radius, taper, and minimal material usage are concerned. The same principles are involved. Identical ill effects can be expected unless the recommended practices are incorporated.

Geometric Structural Reinforcements Besides ribs, there are other methods of improving section properties (see Chapter 5). Many of these can often be designed into functional or appearance features for the product. Basically, to increase load-carrying ability or stiffness it is necessary to increase either the properties of the plastic material or the section properties of the structure. Improving the material can be beneficial in certain cases by changing the grade or type, compounding or alloying, and incorporating fillers or fiber reinforcements (see Chapters 2-7). In regard to the structure, increasing the section properties (namely the moment of inertia and the section modulus) is usually required. As previously described, thickening wall sections will often prove a practical means of increasing section properties, but there can certainly be economic limitations in terms of material usage and molding cycles. Some typical

0.75 in. y

0.14 in.

t.----l.22 i n . - - -

1----1.5 in.-----.J

Figure 11·88. A two-rib shape with good moldability that gives a section modulus and moment of inertia equivalent to a rectangular bar.

838 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

examples of these structures and potential problems to consider are shown in Figures 1189 through 11-91 [2]. Some geometric shapes that provide the designer with alternate methods of increasing part supports include gussets, corrugating, and doming (see Chapter 5). Gussets are supporting structures for either the edge of a part or bosses. The design guidelines for gusset thicknesses, spacing, and taper or draft are the same as for ribs. The dimensions given in Figure 11-91 for height, length, and spacing are the lowest acceptable dimensions for good gusset design. Corrugating and doming provide the designer increased part performance without having to add ribs. Of these two methods, corrugating is the more effective. The alternative to corrugating is doming (see Fig. 11-89). When a part is domed, its wall is molded in a convex shape. In some cases this method may be preferred to corrugating for aesthetic purposes. From a structural standpoint, through, doming does not offer the reinforcement rigidity of corrugating. Snap Joints A snap joint is economical in two respects: it allows the structural member to be molded simultaneously with the molded part, and it allows rationalizing the assembly, compared with such other joining processes as screws. Table 11-5 provides a comparison of its advantages and disadvantages. The design guidelines that should be taken into consideration for obtaining the desired functions are described in Chapter 5 [409]. Some examples of the various types and their design considerations are shown in Figures 11-92 through 11-94.

HAT SECTION CROWNING CORRUGA T/ON

METAL REINFORCEMENT

BI-DIRECTIONAL CORRUGATION

Figure 11-89. Examples of geometric structural reinforcement techniques.

DOMING

Oil canning effect; left view is with even wall and other view shows redesigned double-tapered wall. GATE

- [--+--"

_JI

--+--I----1111

11.0--

...I~----Dia - - -....-11

Long, flat part (top) could tend to warp; add ribs to correct problem.

Warpage allowance and direction should be shown on the design drawing so that cooling fixtures, type plastic and molding techniques may be determined to achieve optimum performance in the molded part. Warpage allowance~,~

~

~:~

"Indicate warpage direction"

Figure 11-90. Examples of potential problems with different shapes.

A "c ,75 to 1T B ,,' Not less than 2T C " DIctated by Structural ReqUIrements T Wall ThIckness

Figure 11-91. Guidelines for designing a gusset.

839

840 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

The geometry for snap joints should be chosen in such a manner that excessive increases in stress do not occur (see Chapter 5). The arrangement of the undercut should be chosen in such a manner that deformations of the molded part from shrinkage, distortion, unilateral heating, and loading do not disturb its functioning. The following guidelines are recommended regarding the position of the snap joint to the in gate and the choice of the wall thicknesses in the area of flow to the place of joining: I) there should be no binding seams at critical points; 2) avoid binding seams created by stagnation of the melt during filling; 3) the molecules and the filler should be oriented in the direction of stress; and 4) any uneven distribution of the filler should not occur at high-stress points. (b) Snap-on fit

(a) Snap-in fit

(c) Separable snap joints for box cover

(d) Cap with two cantilever and two rigid lugs

(e) Discontinuous annular snap joint

(I) Detachable and non-detachable snaps

1. h = 0.00 75d if (~ ~ 1.2) 2. h = d

(0.0024~ + 0.005) if (~:::1.2)

I

/

-1- ",

/,. I

I

{

\

'-

..............

, 1\\ \ '

I

....... ~

---

/

/

h detachable

Figure 11·92. Examples of different snap-fit designs.

non-detachable

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 841

Table 11-5. The Advantages and Disadvantages of Snap Joints Advantages

Disadvantages

Can be easily integrated into the structural member Compact, space-saving form Takes over other functions like bearing, spring cushioning, fixing Higher forces can also be transmitted with proper designing Small number of individual parts Assembly of a construction system with little expenditure of production facilities and time

The fixing of the joined parts is weaker than in welding, bonding, and screw joining The conduct of force at the joining place is lesser than in areal joining (bonding, welding) Effects of processing on the properties of the snap joints (orientation of the molecules and of the filler, distribution of the filler, binding seams, shrinkage, surface, roughness and structure) Narrow tolerances are required in complicated applications (in plastics, this is associated in some cases with considerable expenditure) Inftuence of environmental effects (for example, distortion due to temperature differences) on the functioning Difficulties with a continuous loading of the snap joint

The stressing and joining forces should not act in the same direction if the joint has to absorb larger forces. Ann brackets should possibly be used to prevent the hook from escaping after joining. When there is a fracture of the snap hook as a result of overloading during the joining operation the cross-section should not be increased, but the hook should be designed to be more flexible. On account of the frictional forces and stresses that appear at the point of joining, the angle of joining should be chosen to be not larger than 60 degrees [811].

Integral Hinges Hinge dimensions for lids, boxes, caps, and many other products have by now been well established (see Chapter 5). Figure 11-95 shows the successful dimensions of a living hinge. They depend on the design shape but also on ensuring that during the molding operation its melt flow is through the hinge (perpendicular to the hinge's bending action) so that its molecules stretch to give a strong, pliable hinging section. Generally, in order to ensure proper hinge action it is flexed immediately after the part is molded while it is still hot, using mechanical action in the mold or flexing it manually. It is important to locate gates in the proper position in relationship to the thickness and flow pattern of the melt so that the melt flows properly through the hinge. An example of a poor flow condition is to have gates on opposite sides of a hinge, so that a weld line forms within the hinge, causing it to fail upon its first being bent. Some examples of a few of the thousands of successful living binge applications are shown in Figure 11-96. Others include dual-flap cap closures, lifting cap tabs, diaphragm valve flex hinges, and so on.

Mold Actions Molds can be designed to produce products that permit molding from very simple shapes to extremely complex ones. The complex shapes can include practically all those actions reviewed in this section that have been classified as being poor or difficult to mold. In

Sealing rib

Tamper seal Figure 11-93. Section of a linerless cap with a tamperproof seal.

Siandard threads dr Ive Snap Cap miGplace

,

J.

r

/ 5nap lin

/ lhreads disengage when Sn ap Cap seals. 1 en reengage 10 dllve cap do Io n Ilghler

& ,

I'

/

A

'\

A

"

Y

Figure 11-94. This nonbackoff cap, or NBO, from Sunbeam goes on like a screw cap and seals like a snap cap to provide a liquid-tight closure. These caps are injection molded from Marlex polypropylene supplied by the Phillips Chemical Co. Although conventional screw caps are easy to use, they can work loose or back off during shipping and handling. Although snap caps sometimes are difficult to operate, they provide a positive seal-the cap is either open or closed. Stiffen n9 shoulder

0.060 Web lond

0.005 to 0.015 os necessary for stiffness of hinge OC110n or for mold·fill requIrements

0.030

Thickness approxlrnotely equal ta sidewalls

+

All dimension, are in inches

Figure 11-95. An example of basic PP integral hinge design where the plastic melt flow in a one-way direction is from the thick section through the tin section of the hinge. This melt flow through the tin section causes a degree of molecular orientation that is directly related to its performance as a hinge . (All dimensions are in inches .)

842

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 843

a simple mold design, which is basically defined as when the mold separates into two or more sections, the part can be removed. In order for this to be done, certain geometric considerations must be met. First, the part should have no undercut sections that will lock if it is pulled from the mold. If shape is essential to function, a much more complicated mold is required where a portion of the mold is retracted to permit the undercut to be removed. This complicates the molding procedure and the mold, which may result in higher costs as well as a poorer quality part. The part will usually have some surfaces that are nearly parallel and perpendicular to the opening surface of the mold (the parting line), and pulling the part against these parallel surfaces could result in sticking and drag that would make removal difficult and damage the product. The product designer should restrict the number of undercuts to a minimum and consider carefully whether any undercuts in the design will present major problems in mold design. Moldings made from flexible plastic with small undercuts often allow forced mold release; that is, during the mold opening the molding distorts sufficiently, because of its flexibility, to jump free of the undercut. This method is not recommended without experience. In such cases a certain degree of deformation may have to be accepted. Generously rounded comers are a must if this method of mold release is to be used. For rigid plastics and large undercuts, use must be made of movable or rotating side cores, which then obviously influence mold constructions. Screw threads are an example of an undercut frequently met. To eliminate undercuts, consider tapering a wall so that a sliding shutoff can be used (see Fig. 11-97). Molded parts with undercuts (Le., articles that cannot be released in the direction of the mold opening) require molds with more than one parting line. For such articles various (1) Self-hinged spray closure

(2) Box lid hinge

(3) Single flap cal hinge closure with snap spud

ThicllRes$

6({C

Finger recess

land length

land length To Thickness RatiD At least 3 TD 1

Lower hinge

(4) Action of package design combining a hinge with snap fits. HlIIClCAVITT

I

,..,....,.,.,~ -;="

Figure 11-96. Some examples of living hinges.

844 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK Sliding Shu 011 Opening Par ling L,ne

Figure 11-97. Eliminate an undercut, requiring side cores, by using a sliding shutoff system.

Figure 11-98. A schematic of a mold with mechanically actuated side cores. methods have been developed that may be operated manually, mechanically, hydraulically, pneumatically, or electromechanically: Molds Molds Molds Molds

with with with with

side cores (Figs. 11-98 through 11-101) wedges (Figs. 11-100 and 11-101) rotating cores (Figs. 11-102 and 11-103) loose cores or inserts (Fig. 11-48)

The choice of method, or of a combination of these methods, is not only governed by the shape of the article and the properties of the polymer (its flexibility, rigidity, shrinkage, etc.) but also by the standards of quality to be met by the article. For articles with an external screw thread, for instance, either the first method or the third can be used. However, if the first is used, the mold parting line shows, which may be undesirable for aesthetic or design reasons. The method used should depend entirely on the circumstances of the article.

EXTRUSION Basically, the size of the die orifice controls the thickness, width, and shape of any extruded part dimension. In general, it is developed oversize to allow for the drawing and shrinkage that occur during conveyor cooling operations. The rate of takeoff also has significant influences on dimensions and shapes. This action, called drawdown, can also influence keeping the extrudate straight and properly shaped, as well as permitting size adjustments. The drawdown ratio (see Chapter 7) is the ratio of orifice die size at the exit to the final profile size.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 845 End cor

Figure 11·99. A schematic of a mold with side core action, a battery case molded of PP.

Figure 11·100. These cavity blocks for PVC pipe elbows use CSM414 prehardened stainless steel. Courtesy Crucible Specialty Metals Division of Colt Industries, Syracuse, NY.

The range of extrudable profiles is practically unlimited, but to realize a full, practical design with economic potential, particular attention must be given to factors like wall thickness, hollows and cores, legs and projections, comers and radii, and so on. The most important consideration in profile design is the balancing of various wall thicknesses. A profile with a uniform wall thickness throughout its cross·section is the easiest to produce. Having uneven walls will cause material flow variations between the large and small portions of the profile. Also, thinner sections cool faster, causing bowing or warpage toward the heavy side. To compensate, it is necessary to provide external cooling for the bulkier sections and, usually, some special orifice die design in which the land lengths

846 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

(distances along metal surfaces) are changed significantly in respect to their cross-sectional openings. This usually requires additional costs, equipment, and reduced extrusion speed, resulting in higher production costs [1, 5, 6-10, 25, 36-42, 46, 50-53, 62-65, 458,

459,601-16, 807-12]. Tolerances The penalty for having an unbalanced wall is the reduction of tolerance control. Tolerance limits are usually doubled. Also, with certain plastics it is more difficult to process them,

Figure 11·101. A mold with wedge side core action.

Figure 11·102. A mold with a rotating core that operates during mold opening and closing. The drive gear rotates via the worm shaft, which in turn transmits the rotation to the geared core. The core then unscrews the threaded, molded part.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 847

Figure 11-103. Cores can be positioned in rows, as this cutaway view of a closed-mold frame indicates. Each core resides within a gear that, when engaged by one of the parallel racks, causes the core to rotate and unscrew the molded caps. Courtesy Newark Die.

Figure 11-104. Sink marks can be eliminated by creating a design, a rib, or serrations.

such as those with low melt strength. Although the balanced wall is the ideal, having it is not always possible. Simply recognize that the unbalanced wall can be extruded. As discussed in the previous section on injection molding, a sink mark almost always occurs in extrusion on a flat surface that is opposite to and adjoining a leg or rib, because of unbalanced heat removal or similar factors. As with injection molding, sink marks can be concealed by adding a design feature, such as a series of serrations on the area where they occur (see Fig. 11-104). Figures 11-105 through 11-109 provide design features in extrusion that influence product perfonnance. The guiding principle should be to keep it simple whenever possible [41].

BLOW MOLDING Blow molding, discussed in detail in Chapter 7, provides designers with the capability to make products ranging from the simple to complex three-dimensional shapes. The BM process is especially amenable to the designer's goal of consolidating as much function as possible into a single part. Some of the features that can be incorporated include threads, inserts, fasteners, hinges, and others somewhat similar to those covered under injection molding. Hinges include the different mechanical types as well as integral hinges [5,7,8-11, 14,26,40-43,50-53, 111,213,626-34,784,807,809,813,814].

@ @ b.

c.

@

d.

e.

Figure 11·105. Examples of die designs to produce different profiles. a) The method of balancing flow to produce this shape requires having a short land where the thin leg is extruded. This design provides the same rate of flow for the thin section as for the heavy one. b) This die for making square extrusions uses convex sides on the die opening so that straight sides are formed upon melt exiting; the comers have a slight radius to help obtain smooth comers. The rear and sectional views show how part of the die has been machined away to provide short lands at the comers to balance the melt flow. c) In this die for a P shape, the hole in the P is formed by a pin mounted on the die bridge. The rate of flow in thick and thin sections is balanced by the shoulder dam behind the small-diameter section of the pin. The pin can be positioned along its axis to adjust the rate of flow to meet the melt characteristics. d) In this die to extrude a rather complicated, nonuniform shape, a dam or baffle plate restricts the flow at the heavy section of the extrudate to obtain uniform flow for all sections. The melt flows between the die plate and the dam to fill the heavy section. The clearance between the dam and the die plate can be adjusted as required for different plastics with different melt behaviors. e) In this die for extruding a quarter-round profile the die opening has convex sides to give straight sides on the right-angled portion, and the comers have a slight radius to aid in obtaining smooth comers on the extrusion.

848

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 849

Figure 11-106. Typical examples of balanced-wall extruded profiles.

~ c

o

Figure 11-107. Examples of successful profile shapes with unbalanced walls: a) a rigid PVC

shoe; b) an ABS house-trailer trim section; c) a flexible PVC armrest for a bus; and d) a rigid PVC insulator for an electrical bus bar. To produce a hinge during extrusion blow molding (see Figs. 11-110 and 111) the hinge is formed perpendicular to the parison flow from the die. With injection blow molding, the hinge is perpendicular to the melt flow in the preform mold. Here are some guidelines to mold living hinges with polypropylene (other plastics are similar but may require their own specific dimensions dependent upon their particular viscoelastic characteristics) : 1. The land of the hinge should be at least 1.5 mm (0.06 in.) wide for a proper flow pattern and at least wide enough so that when the part is bent in service it will not develop strains; too short a land length will cause the hinge to have limited flex life. 2. The minimum plastic thickness or pinch-off gap at the center of the hinge should be 0.25 to 0.38 mm thick (0.010 to 0.015 in.) and 0.5 mm wide (0.020 in.)

a.

REPLACEMENT SEAL

Jt\ .

,'.)

...

~'

.'. /

b.

c.

RIGIO VINYL

REPLACEME T SEAL

RIGIO VI NYL

Figure 11·108. Design tips for coextrusion: a) A dual extrusion for a modular cabinet wall panel; b) if the flexible sealing portion wears out from abrasion, a replacement flexible insert can be slid into the slot in a rigid portion; c) a cross-section of a dual extrusion (a ball-return trough for a billiard table); d) a bowling-ball return trough made from a 6-in.-diameter extruded tube with one or more layers. The tube is slit while still workable and guided over a forming die; e) typical dual extrusions of rigid and flexible PVCs; f) typical extrusions of rigid and flexible PVCs showing different applications; g) a cross-section of a window frame with a metal embedment; h) nonbondable plastic can be joined by keying or fitting; i) noncircular hollows are easier to form if each part of the surrounding wall is made from the same family of plastic. A) the rigid PVC base will remain flat and not bulge; B) the air pressure inside the hollow will cause the flexible base section to bulge; j) different applications for metal-embedment extrusions. 850

L

FLEXIBLE VINYL RIGID VINYL

~~~~~~~~

0.

..... 0: .•

EDGE PROTECTOR

/

.'. SEAL AND .:::. CAP

PROTECTIVE BUMPER

f.

e. FLEXIBLE VINYL ~

Ift~':'=2:;;=?;~;;:i='~1

Y.,,,, :R"" "'"

g.

WINDOW FRAME

NONBONDABLE DUAL EXTRUSIONS

ENE

h.

POLYETHYLENE

METAL- EMBEPMENT EXTRUSIONS

.-1111111(

STORM-WINDOW FRAME

EASY

A TUBULAR EXTRUSIONS

DIFFICULT I.

j. 851

852 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

~ A

a.

THAN

SHARP AS

b.

c.

B

A NO GREATER

F SHARP AS PRACT I CAL

B

TYPICAL

AVOID A HOLLOW WITHIN A HOLLOW

PREFERRED

OUTSIDE RADIUS HAS SAME CENTER AS INSIDE RADIUS

FOR 0.020' TO 0.040' WALL

Figure 11·109. Design tips for extruded profiles: a) hollow sections; b) recommended minimum radii for outside corners; c) recommended minimum radii for inside corners. 3. When the plastic melt flows across the small hinge gap, frictional heat will be gen-

erated. There should be sufficient cooling of the mold around the hinge area. 4. With an injection blow, the hinge gap is a difficult area to flow across. Therefore, the gate should be placed so that the molten plastic can flow perpendicular to the hinge, to ensure a good fill. If the melt flows along the length of the hinge, there is bound to be a short shot or cold weld at the hinge. 5. Shoulders and lips should be included in the two mating parts to help alignment. 6. The finished piece should be flexed immediately upon its ejection from the mold, while the heat from the mold is still in it. A flex angle between 90 and 1800 is recommended. The flexing action can stretch the hinge area by 200 percent or more; thus, the initial 0.25 to 0.38 mm thickness (0.10 to 0.015 in.) will be thinned down to less than 0.13 mm (0.005 in.). This elongation aligns the plastic's molecules and increases its tensile strength from 34 X 106 to 552 X 106 Pa (5,000 to 80,000 psi). An extrusion-blown hinge can be compared to a stamping. Hot- and cold-stamped hinges are made by compressing a sheet of material down to the desired thickness (about 0.25 mm, or 0.010 in.). Stamped hinges are less durable in flexing but strong in tearing. Even though blow molding provides the capability of providing multiple parts combined into one, by including hinges the designer has an added feature. The ability to produce many articulated parts in one shot opens new design possibilities. Because this hinge is practically free, the cost of these products is just the cost of the material and the molding.

HUlCle

Panson

Figure 11-110. An example of an extrusion blow-molded container with a living hinge in the as-molded position.

a.

b. Figure 11-111. A molded living hinge in the as-molded position (a) and a flexed position (b) with the dimensions in inches. 853

854 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

THERMOFORMING

Designers should follow the inherent nature of thermoforming (see Chapter 5), which basically uses flat panels instead of the solid, enclosed, boxlike, cylindrical, rodlike, or structural shapes of other processes [5, 7, 8, 10, 14,40-43,50-53, 213, 651-62]. They should be aware of and observe the material's depth-of-draw limitations, which can vary depending on the type of TP, the thickness tolerance of the sheet, and the degree of pinhole freedom it enjoys. Generally, for straight vacuum forming into a female mold, the depth-to-width ratio should not exceed 0.5 : 1. For drape forming over a male mold, this ratio should not exceed 1 : 1. For parts to be used with the plug-assist, slip-ring, or one of the reverse-draw methods, the ratio can exceed 1 : 1 and perhaps even reach 2 : 1 under normal circumstances [10, 40, 41]. However, shallow drafts are in general more readily formed than deep ones and result in more uniform wall thicknesses. Undercuts and reentrant shapes are possible in many designs. They require movable or collapsible mold members, but with small undercuts they can often be sprung from a female mold while the formed part is still warm. This type of action works best when the plastic has some flexibility, as do the TPEs, or is very thin. Guidelines for the maximum amounts of undercutting that can be stripped from a mold are as follows: 0.04 in. for acrylics, PCs, and other rigid plastics; 0.060 in. for PEs, ABSs, and PAs; and 0.100 in. for flexible plastics such as the PVCs. When female tooling is split to permit the removal of parts with undercuts, a parting line of the split halves becomes visible on the formed part. If this is objectionable, the designer can sometimes incorporate the parting line in the decoration of the part or at some natural line on the part. Sharp comers should never be specified, since they hamper the flow of material into the mold's comers. This results in excessive thinning of the materials and causes concentrations of stress. A minimum radius of two times the stock's thickness is recommended. It is also more desirable from several standpoints to have large, flowing curves in a thermoformed part than to have squared comers or rectangular shapes. The best parts have smooth, natural curves and drawn sections that are spherical or nearly so in shape. Their walls will be more uniform, they will be more rigid, their surfaces will be less apt to show tool marks, and their tooling and molds will be lower in cost. Notches or square holes should be avoided when punching formed parts. Round holes are preferred to oval ones for minimizing stress buildup. Some draft is required in side walls to facilitate the easy removal of the part from the mold. Female molds require less draft since parts tend to pull away from mold walls as they shrink during cooling. With female tooling, for most plastics the draft on each side wall should be at least! degree. For male tooling, it should be 1 degree (see Fig. 11-

112). Metal inserts are usually not feasible, because thin walls are not sufficiently strong to hold inserts, particularly if thermal expansion and contraction take place. Figure 11-113 shows a method of holding metal fittings. It may be desirable to increase the stiffness of thermoformed parts. Many such parts are panel shaped and made of thin walls, so they may lack rigidity. Corrugations, which if used are preferable in two directions, or an embossed pattern can add to their rigidity (see Chapter 5). With short-run production it may be more economical just to use thicker sheet plastic to gain stiffness. If the function of the part permits, use curved, dished, or domed surfaces to gain stiffness (see Fig. 1189). When thermoformed parts are stacked, without controlled spacing, they will jam to-

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 855 ~ Draft angle 1/4 min. for female tooling; 11_10 min. for mole tooling 0

iOOdraft

1~r>;::.:.!'"

R = 2 T or greater

Not this

This

Figure 11-112. An example of the draft required in side walls to facilitate the easy removal of a thermoformed part from the mold.

JU Not this

Metal insert Thermoformed plastic sheet

This

Figure 11-113. A recommended method for holding metal fittings in thermoformed parts.

Figure 11-114. An example of how to stack thermoformed parts using a boss design to avoid the jamming of parts.

gether, which could cause sufficient stress to cause parts to split. To avoid jamming and control the space between parts, a stacking boss or shoulder system can be used (see Fig. 11-114). Within this stacking area the plastic must be sufficiently rigid to prevent the deflection of bosses that would cause jamming. The height of the bosses is generally greater than their vertical cross-sections at the point of least taper; otherwise the tapered walls will interfere before the stacking sections can engage. There also other designs that can be used to eliminate jamming.

Tolerances Thermoformed parts lack the dimensional accuracy of injection- and compression-molded parts. With its low pressure, thermoforming reduces the degree to which the sheet being formed is forced to conform to the mold. Sheet variations, mainly in their thickness and degree of existing pinholes, affect the final accuracy of the part. This is particularly true

856 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

because tooling is generally one sided. The objective should be to use sheet with tight thickness controls that is pinhole free, rather than just to determine its weight (some fabricators "buy" by the lower-cost method, where weight is the controlling factor). Parts are affected dimensionally by the difference between their forming temperature and their product-use temperature. Thus, a plastic's coefficient of thermal expansion and contraction has a significant effect (see Chapter 2). The tooling generally used is inexpensive. High-precision tooling is usually not produced. The pressure, time, and temperature variations that can exist will affect the final part dimensions. Of these factors, evenness in heating the sheet before forming is usually the most important control. An allowance must also be made for postforming shrinkage (as previously reviewed; see especially Chapter 7). Molds should be designed oversize so that when shrinkage is complete the part dimensions will be correct to within the design tolerances. It should be noted that with really precise tooling accuracy is much more prevalent, especially with matched male and female molds and careful control of temperature, time, and pressure. The dimensional tolerances with the more conventional single-mold system are generally ±0.6 percent (±0.35 percent for close tolerances) with female molds, ±0.5 percent (±0.3 percent close) with male molds under 3 ft., ±0.8 percent (±0.4 percent close) with male molds over 3 ft., and ± 30 percent (± 10 percent close) for wall thicknesses.

REINFORCED PLASTICS AND COMPOSITES The approach of a designer planning an RP/composite part should be to determine from its size, shape, performance requirements, and quantity to be produced which method should be utilized (see Chapter 7). This basic approach is important, because design principles will vary considerably with the fabricating method. For example, there are limitations on the depth of draw obtainable with sheet-molding compound (SMC) that do not apply to preforms. Whenever possible, the product should be designed with a specific process in mind. It is also important to show the preliminary design to prospective fabricators. As noted on various occasions throughout this book, it is always sound advice to consult the prospective fabricator when using any extrusion or other process, but this is especially true in the case ofRP/C parts, since a wide variety of manufacturing processes exists [1-21, 26-33, 40-43, 46, 50-53, 81, 149-55,483-508,590,636-50,814,815]. When designing, a number of variables must be kept in mind. For example, the strength values for high-impact compounds containing long glass fibers generally remain high when they are compression molded. However, these values may drop as much as 50 percent upon being injection or transfer molded, because in these processes the fiber lengths are reduced, causing only the shortest fibers to flow with the resin into thin, intricate sections. Thus, these sections become weaker per unit of thickness than do the thicker sections. With certain parts the geometric design may offset any advantage to be gained by compression molding. There are products with thin, complex shapes that could use lessexpensive compounds and produce stronger parts by injection molding. Such an 1M part would have improved density and uniform fiber distribution to offset its other problems. Figures 11-115 through 11-128 [2] and Tables 11-6 and 11-7 provide guidelines for RP/C molding by the compression, transfer, matched-die, and stamping techniques.

Semi-posItive or londed type

Figure 11-115. Examples of the three basic types of closed compression molds principally used with TSs to provide for fabricating parts ranging from thin to thick or very heavy.

Pot tronsfet

Figure 11-116. Three basic types of transfer molds for TSs to fabricate parts with complicated shapes, inserts, and small-diameter holes.

§~§~;§;*l

Nominal TPanel thickness

t

Die draw angle

!

Stop block

L",,,m"m

Mold

core

3' draft angle

Figure 11-117. A guide to trimming or shearing the edges of parts that are compression molded.

Mold cavity

t

Die draw angle Shear~--------------------~

edge

~

t ~

Die draw angle

Figure 11-118. Matched mold shear edges. The better design is to use the lower angle (left). 857

L-_---'

Core

L

Knlfe.edge of 1001 sleel

Figure 11-119. An example of a knife shear edge.

PREFORM MAT

A

-

I

•

J

"'-1---'3 1-

1

, T

WAU THICKNESS

1

- 1---1 !.OO5-J

0.050 -1.00..3

WALL THICKNESS

WA LL THICKNESS

I

I

1----1 :.oosJ

to.OO3-0.005IM/IN. !0.010 IN./IN. . NOT ACROSS PARTING LINES TOLERANCES

C\\,?""u,

J

l

T

0.070"1.0·

0.030-0. 25 ..J

B 1....- - - - . . /

SHEET MOLD I NG COMPOUND .L

PREMIX

NOT ACROSS PARTING LI NES TOLERANCES

\\"., ..;

1

( -1.

~

!:.005::f

O.OOBIN/IN. NOT ACROSS PARTING LINES TOLERANCES

J\\

-nl'.?·UI

- ~N. RAD. b"'N. RA1 ~N CORNER RADIUS

]

6 b

LI BERAL RADII

NOT ADVISABLE

CORNER RADIUS

RAD.

CORNER RADIUS

6

LI BERAL RADII

6

NOT ADVISABLE

NOT ADVISABLE

6

U BERAL RAD II

6

Figure 11-120. Comparative infonnation on the matched·die molding of preform mat, premix compound, and sheet-molding compound. a) waIl thickness; b) tolerances; c) comer radius; d) radii. 858

A:1r" £ ..!!.

15

10'

FLANGE

.,

t

ff'~FSET O~LANGE Figure 11·121. Flanges used in premix molding. Left: An offset flange is used to increase stiffness. Right: The steps that can be taken to reduce warpage in long flanges. The C and D views show the preferred construction and design.

®b 6 . ..

llim

PREFORw MAT

---.!. :-MIN \

1°

I-- 8 ---l

~~~ {1

2·!

DRAFT

---I

t--A~

1--8

---l

I - - 8--1

Q

1]

A ' PART DIM . S = MOLD DIM. 0 .003 IN/IN. STEEL MOLD 0.002 I N./ I N. ALUMINUM MOLD

SHRINKAGE

!J

A:PART DIM . B-MOLD DIM. 0.003 I N./I N. STEEL MOLD 0.002 IN.!I N. ALUMINUM MOLD

SHRINKAGE

~

A(l

I( NESS

10 3°

6.

~~~ ~~ ~.Aj

A •PART DIM S = MOLD DIM. 0.001 I Nil N. STEEL MOLD 0.002 I N./I N. ALUMINUM MOLD

TA -THIC

,.

0 · 1"

MIN . DRAFT

DEPTH OF DRAW 1;3-

A

I-~

IJ

__

MIN. DRAFT

DRAFT

t--A--i

@I}B

.., .

..:~

DEPTH OF DRAW 33-18 18·

DRAFT

MQLQltHi

~ CON':UNO

,

FOR PARTS OVER 18 INCHES DEEP USE 2 _30

® Q

SHEET

__

A,l

T TA=THICKNESS T

OF WALL OF WALL 8 -1.5-A (PLENUM 8=2-A PREFORM) S - 2"A DIRECTED FISER PREFORM VARIATION IN WALL VARIATION IN WALL THICKNESS THICKNESS

I

SHRINKAGE ,[8

T

}

Al

--

1

A -THICKNESS OF WALL 8 ' 3. A

T

VARIATION IN WAll THICKNESS

Figure 11·122. Comparative information on the matched-die molding of preform mat, premix compound, and sheet-molding compound. 859

BETTER

BEST

AcA

Figure 11-123. A flange mounting construction. The flange mounting hole to be mounted on a flexible gasket should be in compression whenever possible.

HOLE MAX. BLIND MAX . THRU DIA. HOLE DEPTH HOLE DEPTH INCH INCH INCH _c. 25_ 0 .750 .f 5 50

I

DEPT H '--_ _- " " ' - T

{~~ .175 . OC -( 50 1.00

........_.-I--,.1O.l:I.

FHgg= .875 3 .DOC

~5OC

6.000

2.

-

2.

4

5. 6.0

-

DEPTH

Figure 11-124. Recommended hole sizes for through and blind holes in premix compression molding.

HOLE MIN DISTANCE MIN. DISTANCE FROM EACH FROM EDGE DIA. OTHER (I NCH) (I NCH\ (INCH) 0.187 0.187 0.125 0.187 0.250 0.250 0.250 0.312 0.312 0.281 0.375 0.375 0.375 1..5OC 10..500 0..562 0.750 0.500 0.750 0.500 1.000

Figure 11-125. Recommended distances between holes and distances of holes from side walls in premix molding. 860

puNCHEP HOLES

puNCHEP HOLES

PUNCH AND BLANK PIE

o

--I

©

®

I-DIA NO LESS THAN T

i t::1:$I::r-=~=~ T T

t

HOLE PUNCHED BEYOND ALLOWED TH I CKNESS

-i0l-OIAGONAL NO LESS ::::===t THAN T

51

o NO

®

@ CLOSER

I~q THAN~ A SHAVING PIE

LAMINATE

,xT @ !$¥I

tL:t

"*"

@it

(CAN BE SANDED OFF. HOLE PUNCHED PART WAY THROUGH

- "'-----

~ 0 AT LEAST

1·V~}g-, x /

45

CIRCULAR SHAVING KNIFE

.1..

PLUS 10""

®

IxT..l

I"""I3T""""1b"""""j'T"""@r= T

AVOID ERRATIC SHAPED HOLES

T

Figure 11-126. Some guidelines on die punching: a) the design of a punching and blanking die; b) a shaving die. Punched holes in laminated stock have limited sizes, compared to the thickness of the laminate shown in c), d), and e); f) a hole punched beyond the allowable thickness; g) a hole punched partway through; h) erratically shaped holes like these should be avoided. PREfORM MAT

4fNO

~OLDED HOLES

SHEET MOLDIN G

~

~

~

~S

~YES MOLDED HOLES

MOLDED HOLES

~.o ~ ~£S

~

UNDERCUTS

00

NO

@METALINSERTS

@~~ BOSSES

®~

~n."

RIBS

NO

PERCENT FIBER GLASS BY WT.

®

25 -50

YES

UNDERCUTS

~

UNDERCUTS

YES

OOYES

METAL INSERTS

METAL INSERTS

~ES

•

BOSSES

/A.;e

BOSSES

YES

I ~YGIdff!YES RIBS

RIBS

PERCENT FIB ER I PERCENT FI BER GLASS BY WT. GLASS BY WT.

15-35

15 - 30

Figure 11-127. Comparative information on the matched-die compression molding of preform mat, premix compound, and sheet molding compound. 861

BOLTS RIVETS SCREWS STAPlES GOOD GOOD NOT GOOD GOOO

BETTEA

NOT GOOD

,PrJ· NOT~ .~ ~ NOT GOOD

..

lETTER

f

PlAIN LAP JOINT OFFSET LAP JOINT ANGLE LAP JOINT FLANGE ANGLE LAP JOINT

CLdJ

I

BUTT JOINT WITH BAC K UP PLATE END CAP JOINT

Figure 11-128. Reinforced plastic parts can be fastened together by mechanical fasteners and adhesive bonding.

Table 11-6. Thermosetting Resins Used in Laminated Products Property Specific gravity Cost or price Heat resistance Physical properties Electrical properties Water resistance Machining qualities Molding pressures Molding qualities Advantages

862

Phenolics

1.3 Low

Melamine

1.48

Medium Excellent Excellent Good Good Excellent Excellent Fair Good Fair to good Fair to good High Low to high Excellent Good Good all-around Good all-around properties properties

Polyester

1.3 Low Good Good Good Good Fair to good

Low Excellent Many types and properties

Epoxy

Silicones

1.25

1.3

Medium-high Fair Good Excellent Excellent Good Low to medium Fair Shrinkage nil

High Excellent Fair Excellent Good Good Low to high Good Heat resistance

i

w

Under .062" 0.6) .062" to .093" 0.6-2.4) .093" to .12S" (2.4-3.17)

Material Thickness in inches (mm)

3" to 4" (76-102) ±.OOS (±0.13) ± .007 (±O.IS) ±.OOS (±0.20)

2" to 3" (SI-76)

±.004 (±0.1O) ±.006 (±O.IS) ±.007 (±O.IS)

Under 2" (SI mm)

±.003 (±O.OS) ±.OOS (±0.13) ±.006 (±O.IS)

Distance Between Holes

±.006 (±O.IS) ± .00S (± 0.20) ±.009 (±0.23)

4" to S" 002-127)

Maximum Punching Tolerances on Sheet Stock

± .001S (± 0.03S) ± .003 (± 0.076) ± .00S (± 0.137)

Size of Slot or Diameter of Holes

Table 11-7. Standard Tolerances for Punched Holes and Slots, inches (mm)

± .00S (± 0.20) ± .010 (±0.25) ± .01S (± 0.3S)

Overall Dimensions

864 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Tolerances Tolerances can vary depending on the location of the parting line and the depth of flash. As a rule, a mold should be laid out to give the best overall tolerances. Therefore, the parting line should not be shown on the part drawing. When this is possible, it will give the mold designer freedom of choice for the best producibility. If a dimension is extremely critical or other requirements could limit the location of the parting line, these factors should be noted on the drawing as a guide to the mold designer (of course the ideal product designer would also be capable of designing the mold). Alternatively, if a part is designed with a clearly defined parting line, dimensions should be shown that will not be affected by the flash, particularly in working with TS materials. This calculation could be appreciable, depending on the mold type, amount of wear, and so on. Figure 11-129 shows two methods of dimensioning, one that is affected by the flash thickness in four dimensions, the other in only two. It is also important to note that flash can affect dimensions in one plane but not in another, as shown in Figure 11-130. This figure also illustrates how the dimensions controlled by the cavity can remain unaffected by thick flash, whereas the dimensions controlled by a plunger can be seriously affected. Tables 11-8 and 11-9 provide guidelines on wall thickness, punched holes, and slots for specific RP/Cs. Pusslbie Flash

T Recommended Dimensioning Figure 11-129. An approach to use in dimensioning a part to meet optimum tolerances.

Figure 11-130. An example of how dimensioning can be affected by flash.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 865

Table 11-8. Tolerance Guide for Wall Thicknesses in High-Pressure Molding Processes Material

Minimum in. (mm)

Average in. (mm)

Maximum in. (mm)

Thermoplastics: Acetal ABS Acrylic Cellulosics FEP fluoroplastic Polyamide Polycarbonate Polyethylene (L.D.) Polyethylene (H.D.) Polypropylene Polysulfone Modified PPO Polystyrene SAN PVC-rigid

.015 .030 .025 .025 .010 .015 .040 .020 .035 .025 .040 .030 .030 .030 .040

(0.38) (0.76) (0.64) (0.64) (0.25) (0.38) (1.02) (0.51) (0.89) (0.64) (1.02) (0.76) (0.76) (0.76) (1.02)

.062 .090 .093 .075 .035 .062 .093 .062 .062 .080 .100 .080 .062 .062 .093

(1.57) (2.3) (2.4) (1.9) (0.89) (1.57) (2.4) (1.57) (1.57) (2.0) (2.54) (2.0) (1.57) (1.57) (2.4)

.125 .125 .250 .187 .500 .125 .375 .250 .250 .300 .375 .375 .250 .250 .375

(3.17) (3.17) (6.35) (4.75) (12.7) (3.17) (9.52) (6.35) (6.35) (7.62) (9.52) (9.52) (6.35) (6.35) (9.52)

Thermosets: Polyester premix, chopped glass Alkyd-glass filled Alkyd-mineral filled Diallyl phthalate Epoxy-glass filled Melamine-<:ellulose filled Urea-cellulose filled Phenolic-general purpose Phenolic-flock filled Phenolic-glass filled Phenolic-fabric filled Phenolic-mineral filled Silicone glass

.040 .040 .040 .040 .030 .035 .035 .050 .050 .030 .062 .125 .050

(1.02) (1.02) (1.02) (1.02) (0.76) (0.89) (0.89) (1.27) (1.27) (0.76) (1.57) (3.18) (1.27)

.070 .125 .187 .187 .125 .100 .100 .125 .125 .093 .187 .187 .125

(1.78) (3.17) (4.75) (4.75) (3.17) (2.54) (2.54) (3.17) (3.17) (2.4) (4.75) (4.75) (3.17)

1.00 .500 .375 .375 1.00 .187 .187 1.00 1.00 .750 .375 1.00 .250

(25.4) (12.7) (9.52) (9.52) (25.4) (4.75) (4.75) (25.4) (25.4) (19.0) (9.52) (25.4) (6.35)

ROTATIONAL MOLDING

Rotational molding, which uses single or multiple arms to hold the molds, is appropriate to different sizes and shapes of parts such as tanks or containers ranging from small squeeze bulbs of vinyl plastisol to large storage tanks (see Chapter 7). This technique can produce uniform wall thicknesses even when the part has a deep draw off the parting line or small radii. The liquid or powdered plastic used in this method flows freely into comers or other deep draws upon the mold's being rotated and is then fused by heat passing through the mold's wall [5, 10, 40-43, 50-53, 111]. This process is particularly suited economically to producing small production runs and large-sized parts, because molds are not subjected to pressure during molding and relatively inexpensive thin sheet metal molds can thus be used, if the part's shape allows. Lightweight cast aluminum and electroformed or vaporformed nickel molds, which are light in weight and low in cost, can also be used. Large rotational molding machines can be built economically, because they can use inexpensive, gas-fired, hot-air ovens and a relatively lightweight mold-rotating mechanism.

!

CE LE AA 0-3 0-5 0-7 0-9 0-10 N-l OPO-l

XXX

XXP

X

Orade

Paper Paper Paper Cotton Cotton Asb. Fab. Cont. OJ. Cont. OJ. Cont. OJ. Cont. OJ. Cont. OJ. Nylon OJ. Mat

Base Material

Phenolic Phenolic Phenolic Phenolic Phenolic Phenolic Phenolic Melamine Silicone Melamine Epoxy Phenolic Polyester

Resin 1.36 1.32 1.32 1.33 1.33 1.70 1.65 1.90 1.69 1.90 1.75 1.15 1.7

Specific Gravity .010-2 (0.25-51) .015-.250 (0.38-6.35) .015-2.0 (0.38-51) .031-2.0 (0.79-51) .015-2.0 (0.38-51) .062-2.0 (1.3-51) .010-2.0 (0.25-51) .010-3.5 (0.25-89) .010-2.0 (0.25-51) .010-2.0 (0.25-51) .010-1.0 (0.25-25) .010-1.0 (0.25-25) .062-2.0 (1.6-51)

Thickness, in. (mm) 110 100 110 105 105 103 100 120 100 120 110 105 100

Hardness

.60

1.0

6 1.8 1.4 2.2 1.95 3.00 2.7 2.7 .55 .80 .25

Water Absorption 225 250 250 250 250 275 290 .300 400 325 250 250 250

(107) (121) (121) (121) (121) (135) (143) (149) (204) (163) (121) (121) (121)

eC)

Continuous No Load Temp. F

20,000 (138) 11,000 (75.8) 15,000 (103) 9,000 (62.0) 12,000 (82.7) 12,000 (82.7) 23,000 (158) 37,000 (255) 23,000 (158) 37,000 (255) 40,000 (276) 8,500 (58.6) 12,000 (82.7)

LW

16,000 (110) 8,500 (58.6) 12,000 (82.7) 7,000 (48.2) 8,500 (58.6) 10,000 (69.0) 20,000 (138) 30,000 (207) 18,000 (124) 30,000 (207) 35,000 (241) 8,000 (55) 10,000 (69.0)

CW

Tensile Strength P.S.I. (MPa)

Table 11.9. Standard Tolerance Guide for Punched Holes and Slots in High-Pressure Laminated Grades of Sheet Stock, Rods, and Tubes, as Classified by the National Electrical Manufacturers' Association (NEMA)

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 867

When it is necessary to equal the production rates of other processes, the mold cost with rotational molding may exceed that of other processes such as flow molding. The resins used in rotational molding are generally more expensive than the pelleted resins used in many other processes, because they must be finely and evenly powdered, such as to a 35 mesh. However, this process generates low levels of regrind or scrap, even when it is operating poorly. Parts can have no flash at all if good molds are used. The molding of two different types of plastics in a single part may be accomplished to combine their specific properties and a better or lower-cost part. This process, called corotation, is similar to coextrusion or coinjection (see Chapter 7) in terms of the performance of the designed parts. An expensive plastic may be backed with a less costly material, and a skin surface layer can be backed with a foamed plastic molded in one operation. The dissimilar molding powders, which may have different softening temperatures, can be molded simultaneously or separately, depending on the processing conditions and the end product's requirements. Any greater than normal part thickness must usually be designed to form multilayered parts, especially if a foam component is to be included. Some combinations of materials are not feasible with this method. For instance, after molding the first layer against the mold wall, the second material cannot have a higher melt temperature, which, of course, would melt the first layer, probably causing them to mix. With rotational molding one inherent overall disadvantage exists-the complete cycle for a single mold is significantly longer than it is for many other processes. However, in many cases it is possible to run multiple molds on each arm or arms, to offset the effect of having slower cycles. The wall thickness in a mold can be changed just by increasing the amount of plastic put into the split mold, because the wall is basically produced by a coating or plating process that operates on the inside surface of the mold. However, changes in heating time would be necessary to fuse the resin properly. Thus, adjustments to parts' wall thicknesses can be made without a mold if a thicker wall becomes necessary for increased rigidity, impact strength, or load-carrying capacity. A maximum thickness does exist, based on the type of material used and the material chosen to construct the mold and the heat source. These factors all influence the rate of heat transfer through the plastic (see Chapter 2). Because in this process the resin is deposited on the mold without pressure, the finished part is generally stress-free.

Because parts in this process can have deep sections and relatively sharp comers, which could not be done in blow molding or thermoforming, this process is used to mold complex parts that may require three or four split molds. Also, as in other processes, limitations exist for both simple and complex shapes. For example, the parts' surface finish is dictated by the inside surface of the mold. This makes it easy to obtain smooth as well as textured surfaces on the same part. Raised or depressed letters, fluting, and other decorative inscriptions may also be molded. The inside surfaces of parts are influenced by the type of resin used and may be made smooth by selecting an easily flowing melt with a high melt index. Because such resins are sometimes chemically or mechanically inferior, the better resins may be made smoother by resorting to higher molding temperatures and longer cycle times, short of damaging the resin. In-mold decorating methods, such as decals that are deposited on the mold surface, are used that can become part of the finished part's surface and can be designed to provide increased structural performance, as also in injection molding, blow molding, and other processes. The preferred contour for any parting line is the straightest path possible. By this means, mold-construction costs can be reduced and demolding will be the easiest means

868 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

possible. When two parts like a container and its lid are to be molded together, as in blow molding, they may be separated after the molding by employing a removable cutter or annular wedge, at the parting line or by molding the part oversize to provide a resting flange, then cutting it to separate the parts. Ample draft is suggested on side walls, to facilitate part removal. A recommended minimum for most plastics is 1 degree. The lower-shrinkage plastics like PC and PMMA will require Ii to 2 degrees. Undercuts are possible, but they should be kept to a minimum. Making provisions for undercuts usually requires higher mold costs, because of having to use some type of action such as core pulls or splitting a mold to allow separation parallel to the undercut groove. Undercuts may also require extra time for unloading molds. Inside or outside comers should use large radiuses, not sharp ones. By doing so cracking, molded-in stresses, and undesirable part thickening will be prevented. A useful guide to the smallest allowable inside radius is h in. with! in. for optimal filling conditions. The goal should be to have a radius equal to the wall thickness for easier melt flow. Although rotational molding produces uniform wall thicknesses, comers in it can have greater variation than would the rest of the part. A sharp inside comer tends to heat at a slower rate, causing the plastic to flow away from it, thus making it thinner. Conversely, sharp outside comers heat at a faster rate and tend to hold the resin longer, thus building up more thickness. It is usually difficult to produce internal or external bosses and T sections, because they are not conductive to producing uniform walls. It is possible to produce interior extensions, by placing a metallic screen in contact with the inner mold wall. This screen heats up, attracts resin, and becomes covered, remaining in place after molding. By using this method a hollow part can be molded to have two or more separate chambers, with the screen being extended entirely across the inside of the mold. A hole can be formed by molding a dome and cutting it after molding. One technique that can be used for this is to mount securely on the inside mold wall a fluorocarbon (TFE) sheet "plug" to prevent resin from adhering to the mold at that location. Another method involves inserting machined brass plugs, pins, or tubing through the mold wall. During molding, the heat passes from the mold to the insert, causing resin to form around it. Care must be taken to select inserts that will heat easily and a resin that will not crack, because of the stresses created as the plastic shrinks around the insert upon cooling. Moldable holes and inserts can complicate molding and may require extra postmolding operations. Thus, the most economical designs are those that minimize the number of holes. With many resins, both external and internal threads can be molded, but sharp V threads should be avoided, because they cause the resin to bridge, resulting in incomplete thread fill. Rounded or modified buttress threads will allow improved thread fill. The stiffening of solid ribs or projections is possible and easily moldable if the requirement of maintaining uniform wall thickness is followed. A narrowed rib will not fill and will leave inside stringers. And a rib that is too keep presents even greater difficulties, preventing the melt from reaching the bottom before fusing. A small, shallow, narrow rib will fill completely but have limited strengthening effect. The correct rib design requires a wide gap to form the rib, with a generous draft so that the melt is allowed to fill uniformly, without bridging. As a guide, deep ribs that are four times the wall's thickness generally require at least five times the wall's thickness between the parallel sides of the rib, to prevent the plastic from bridging as it flows into the rib before fusing to the mold's wall.

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 869

Tolerances As noted, dimensional control in rotomolding is not as close as in other processes. Many factors influence the degree of this control: the type of resin, particularly the resin shrink rate, the part shape, the cooling rate-the slower the better-the uniformity of part wall thicknesses, the mold design, the control of the heat reaching the resin, and whether fillers are used (fillers reduce shrinkage and promote better dimensional control). A guide to dimensional tolerances for parts is to use a length and width ± if in. for the first in. and ±0.4 in. thereafter, with a wall thickness ±n in. and a wall thickness buildup at comers that is usually greater by 25 percent. The production rate is reduced and made more difficult when different rotational cycles must be employed to provide varying wall thicknesses in the same part, but other methods for changing wall thicknesses locally are also troublesome. They typically involve changing the amount of heat that is transferred by the mold to the plastic material in the designated area. Changing the mold wall's thickness or the color of the mold's outer surface are two methods of doing this. Having higher heat in one area will cause the wall thickness to be greater in that area. The wall-thickness variations that result from such steps are always gradual, with abrupt wall thickness changes being virtually impossible. The maximum recommended thickness buildup, when necessary, is 50 percent of the normal wall thickness for the part. The following recommendations apply to designed wall thicknesses of rotational molded parts: a preferred or desirable nominal wall thickness of 1/10 to i in., a minimum wall thickness of 0.050 to 0.060 in., a maximum wall thickness of i in., and an optimum thickness of glass-fiber-reinforced plastic parts of i in. All parts undergo considerable shrinkage in cooling after leaving the mold. The designer has to allow for this shrinkage, and its effects. As in other processes, shrinkage is in most cases caused by the thermal contraction of the plastic upon cooling. With TSs, molecular readjustments resulting from polymerization are also important (see Chapter 2). Typical shrinkages for standard rotomold resins under controlled conditions range from 1.25 to 2.5 percent. Such shrinkage results not only in a linear reduction of part dimensions but in distortion when wall thicknesses differ or connecting part elements pull other elements together as they shrink.

ASSEMBLY METHODS As noted throughout this book, designing a one-piece item is the ideal situation, because it precludes assembly. However, mechanical limitations and other considerations often make it necessary to join plastic parts, either to each other or to metal parts, to complete an assembly. In such instances the joining process can take an efficient approach to fabrication if a few precautions are observed and established procedures are followed. Various methods of joining can be used easily and successfully. As reviewed in Chapters 5 and 8, assembly methods include mechanical fastening, solvent and adhesive bonding, ultrasonic welding, and electromagnetic welding. Choosing the best method requires a basic knowledge of good joint design and a thorough understanding of the purpose of the joint, the geometry and nature of the components, the type of load involved, and the properties required in the final assembled part. Because of the potential complexities, careful product planning is essential to eliminate needless joints. Generally, the best joining system, in practically all cases, is never as good as a well-

870 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

designed part fabricated in one piece [5, 16, 40-43, 46, 50-53, 72, 105, 213, 683-94, 811]. One major problem of using assembly methods is potential overstressing. Typical of this structural situation is the bolt system with injection-molded parts, which can use standard fasteners like bolts, screws, nuts, locknuts, lockwashers, and other especially designed units. These threaded fasteners are usually made from steel, brass, or other high-strength metals, making it likely that the plastic parts being assembled may be overstressed before their fasteners break or strip. Proper design should thus prevent excessive stresses from occurring when using bolted assemblies [2]. The most obvious and logical method to prevent having high-stress assemblies is to control carefully the tightening of all fasteners, using properly adjusted torque-limiting drivers. This technique can work well when operations are confined to a factory assembly line, but if there is to be field or customer assembly this can present a difficult problem of control. Even when the torque can be controlled, a poorly designed bolted assembly will sometimes have an impractically low assembly torque. In this case, some design modifications will be required. Figure 11-131 shows some common examples of high assembly-stress problems that can easily occur in bolted assemblies, together with some practical solutions. MECHANICAL LOADING

The viscosity behavior of plastics makes them sensitive to strain rates as well as temperatures (see Chapters 2, 3, and 4). It therefore becomes important to define the rate, magnitude, duration, and type of mechanical stress and strain loading (i.e., tension, compression, flexure, and shear) along with temperatures during loading. The rate and duration of loading also determine whether creep or impact will be a factor in a given part's mechanical response [1, 2, 5-14, 29, 33,40-43,55-68,152,202,225,235,250, 270-74, 808]. Plastics are often used to replace stamped steel because their viscoelastic behavior at lower strain rates enables them to spring back rather than to deform under suddenly applied loads. This characteristic does add a safety factor to a design, but it also means that a part can undergo large, nonlinear deflections. Such deflections, defined as being greater than the wall thickness of the part, are difficult to analyze accurately with standard engineering equations. Conventional analysis would teIid to overstate the stresses, because it cannot take into account the ability of flexible plastics to absorb loading by redistributing stresses. Other conditions such as the path of the applied stress-strain loading and changes in the loading sequence introduce their own uncertainties into the analysis. Thus, in designing with plastics one should recognize that the analysis is based on a number of assumptions, which it may be advisable to test the validity of on prototype parts before signing off on the design. Besides dealing with mechanical loading, a well-thought-out design should anticipate abuse of the product, meaning the conditions under which the product is not intended to operate but it may still encounter. For example, the plastic housing of an appliance might end up being used as a stepstool. Designing for too wide a variety of abuse conditions can make a product overly expensive. One solution is to include in the design a preferential point of failure that will protect the product and be easy to repair or replace. For example, using an expendable shear pin would limit the amount of torque that could be transmitted to an otherwise vulnerable part of a product.

POOR DESIGNS

PREFERRED DESIGNS

POTENTIAL HIGH BENDING STRESS AS BOLT IS TIGHTENED

PLANNED GAP BETWEEN ADDED BOSSES PREVENTS EXCESSIVE BENDING OF HOUSING AS BOSSES TOUCH AND GO INTO COMPRESSION

TRUSS OR ROUND HEAD SCREW

t

FLAT-HEAD SCREW

~

..--PLASTICPART

7~-MrnLSUB-FRAME

FT=J ALTERNATIVE RECESSED HEAD DESIGN AVOIDS POTENTIALLY DANGEROUS WEDGING ACTION

POTENTIAL HIGH STRESS DUE TO WEDGING ACTION OF SCREW HEAD

SHOULDER SCREW

STANDARD SCREW

+ \...____~~~----~~PLASTIC PART

WHERE TORQUE CANNOT BE CONTROLLED, AS IS THE CASE WITH FIELD ASSEMBLY, SHOULDER SCREWS WILL LIMIT COMPRESSION ON PLASTIC PART, OTHER SOLUTIONS ARE:

FLANGE-HEAD SCREWS

LARGE WASHERS

SHOULDER WASHERS

Figure 11-131. Examples of bolted assemblies, stress problems, and suggested solutions. 871

872 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Mathematically speaking, developing stress fonnulas for analysis is done through a set of differential equations that must be solved, subject to a number of constraints. The equations for the pertinent stresses and deflections of a wide variety of structural shapes are provided in widely available textbooks, under the heading "Strength of Materials." Additional infonnation can be usually obtained from the major worldwide resin suppliers, along with their suggestions on how the handbooks' equations can most reliably be used with the materials of that supplier. (Chapters 3 and 5 discuss in detail these types of problems regarding the equations that are nonnally available.) As seen in Chapters 3 and 5, there are limitations in using handbook equations. The restrictions and idealizations that have come to be applied to the derivation of classical stress results should be kept in mind when they are used for design or analysis. Parts molded of plastic may deviate from all these values, rendering the handbooks' stress equations invalid to a greater or lesser degree. It has become common practice simply to ignore deviations from the ideal and use the standard equations without modification in the hope that the results will still prove reasonable approximations. Because this procedure clearly has inherent risks, its results should be checked against experimental observations or other validating data if at all possible. One other limitation to elementary equations is that the material being examined is typically assumed to be isotropic, with equal stiffness in all directions. However, molded plastic parts are likely to have at least some orientation, as a result of flow velocity gradients that exist during the melt-processing operation (see Chapter 7). Reinforced plastics and composites do exhibit strong anisotropy, because there is much higher stiffness along the fiber direction than in the transverse direction. A part can encounter two types of stresses-dynamic and residual. Dynamic stresses are simply those that result from the functioning of the part. Because this stress occurs only when the unit is used, its stress level is zero when it is not in use. The designer must as a result understand the part's function to calculate its dynamic stresses. Residual, or molded in, stresses are those that are initiated during fabrication, particularly with injection molding. These stresses can be caused by orientation of the plastic molecules or to an alignment of the fibrous reinforcements. Many factors influence the molecular or fibrous orientation, such as the design of the part, the design of the mold, and the processing conditions (see Chapters 3, 5, 7, and 9). The most common areas of high stress are the points of attachment, sharp comers, gate areas, sections behind bosses, and any openings in the part that may contribute to weld-line formation. It is always good design practice to minimize the amount of molded-in stress. However, most designs retain some level of this kind of stress. If this must be limited to ensure proper part functioning, a quality control test with limits should be specified, such as a part's heat-distortion temperature or impact, flexural, or tensile ratings.

Stress-Strain Polynomials Using a stress-strain polynomial approach can result in more accurate design calculations. In the fall of 1965, Du Pont designer James H. Crate suggested that calculations based on published stress-strain curves for specific plastics would be much more accurate and useful in their high ranges than were calculations based on a material's modulus of elasticity. This advice, from a period when very few were properly designing with plastics, has remained true. Now, designers using polynomials based on those curves may attain an even higher degree of accuracy [808]. In stress analysis, designers frequently use equations originally derived for materials

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 873

A

e

(e

~

0.001 ,n/,n

= 0001 mm/mm-approx)

o Figure 11-132. A typical tensile stress-strain curve for a metal. ,----------------------------.100

12,000

:§:

80

60 ~

e.

8,000

~

iil Q)

i7i

40 c7.i

50% Relative Humldlly

4,000

Average Air Exposure

20

2.5% MOisture Content

L-----------------__________~O

o

50

100

150

200

250

300

350

Strain (% Elongallon)

Figure 11·133. Tensile stress-strain data for Du Pont's Zytel 101 nylon at 23°C (73°F).

with linear stress-strain relations. In Figure 11-132, the nearly vertical segment of the line extending to point A describes the behavior of a metal. Metals typically have an initial linear tensile stress-strain relationship that "hardens" as the strain increases, the way it does here beyond point A. The tensile stress-strain curve for Du Pont's ZytellOl nylon, shown in Figure 11-133, is somewhat similar [808]. Because this material is not as stiff, however, the proportionally higher rate of elongation (strain) is readily apparent in the initial portion of the curve. Such engineering-manual curves, which detail a material's stress-strain relationship at different conditions, are often referred to in calculations. Unfortunately, their coarse scale and small size can easily defeat a designer's principal purpose in using them. The accuracy of calculations based on these curves can be significantly improved when a curve is described, with the aid of the proper computer software, as a polynomial equation. Using a computer permits, without referring to graphs, the calculation of the strength or deflection of a plastic just by using polynomial equations. A polynomial is the sum of two or more algebraic expressions or the sum of a finite number of terms that are each composed of a positive power of a variable that is multiplied by a constant. The experimental data from the tensile stress-strain tests that generated the curve for the Zytel 101 nylon resin in Figure 11-134 are accurately modeled by the computer in the following polynomial:

-------=:;;;_-.1

874 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK 16.000

1

12.000

80 0.2ln/min . 50% RH 73'F (5.1 mm/mln . 50% RH 23'C)

~

'i !!1

100

~

60 ~ gj

8.000

Cii

S

40

= - 9.57

x 10 - 1 X e 5 + 1 85 102 X e3 - 1.89 X 103

+ 1.68 X + 8.73 X 103 )( e + 8.64

X

e"

X

e2

OL-_ _ _ _ _ _ _ _ _ _ _ _ _

o

2

3

4

5

6

!!1

Cii

20 ~

0

7

Strain (% Elongation)

Figure 11·134. The stress-strain polynomial for Du Pont's Zytel 101 nylon.

S

= 1.05

X

- 5.69

10

X

X

Hf

eS X

-

e.

2

1.07 X Hf X e4 + 2.99 X Hf X e3 + 4.59 X 103 X e + 6.87 X 10- 1

where S = stress, in psi, and e = unit strain, expressed as the percentage of elongation per mm1mm or in.lin. as multiplied by 100 to obtain a percentage figure. As summarized by J. H. Crate, this polynomial describes the curve with a correlation of 0.9999 if the constant terms are taken to a tenth decimal place. As taken to the second decimal place, as shown, the error is small-only about one-half of 1 percent of the calculated stress. In the sample calculation that follows, a polynomial is used instead of the usual tensile modulus.

Conditions. A product containing a test bar of Zytel 80G 33L (glass-fiber-reinforced nylon), which must elongate 4.38 percent during assembly. The bars are preconditioned to equilibrium in 50 percent relative humidity air. The strain rate used can be no faster than the 0.2 in.lmin. rate from available data. Problem. What is the stress at this elongation? The tensile stress-strain curve for the nylon just described according to the conditions given is shown in Figure 11-135. The polynomial determined by the computer for this curve is

S

=

-9.57 X 10- 1 X eS + 1.85 X e4 + 1.68 X IOZ X e3 -1.89 X 103 X e2 +8.73 X lQ3 X e +8.64

Solution. Simply enter 4.38, the percentage of elongation in the problem, for e in the polynomial input. This gives a stress of 105.1 MPa (15,247 psi). General comments. Had this problem been calculated using published data assuming a linear relationship of tensile stress to strain (see Chapter 3), the answer would have been more than twice this amount. Then the designer would certainly have questioned its practicality, because the stress would have been higher than the tensile strength of the resin. By not using the polynomial but published data the following approach should be adapted. Under the same conditions, the flexural modulus of the Zytel material, which differs from its tensile modulus, becomes 5,068 MPa (735,000 psi). The flexural modulus E then is equal to the stress S divided by the strain e, or E = S/e. The stress is therefore

DESIGN FEATURES THAT INFLUENCE PERFORMANCE 875 12.000 i---------::::::;;iiiiiiii------=j80 0.2 In/min .. DAM. 73°F (S 1mm/mln .. DAM. 23°C) 60 8.000

~ ~ en

~ 40

4.000

S = lOS x 10 x e5 - 1 07 X 102 X e4 + 2 99 X 102 X e3 - S 69 X 102 X e2 + 4.S9 X 103 X e+ 687 X 10-'

~

'"'" ~ en

20

OL------------_~O

o

2

3

4

S

Strain (% Elongation)

Figure 11·135. The stress-strain polynomial for Du Pont's Zytel 80G 33L glass-fiber-reinforced nylon.

calculated as 32,193 psi (that is, 735,000 x 0.0438), resulting in twice as much as the plastic's actual tensile strength of 110 MPa (16,000 psi) under the same conditions. Another way to use polynomials, which requires trial-and-error computer entry, is to find the strain when the stress is either known or has been specified. As an example, in a final evaluation of the problem above, the designer may worry that the assembly conditions may not be ideal f6r field retrofits. The decision is then made to allow for a stress safety factor of two. Thus, the design stress becomes 15,247/2 = 7,624 psi. A sequence of tries then follows where 1.00 percent, then S

e

= =

1.10 percent, then S

= 7,018 psi = 7,551 psi

e

=

1.20 percent, then S

=

e

8,056 psi

Proceeding in this manner, a 1.114 percent strain is found to give the desired stress of 7,624 psi. Had the designer used the linear relationship of e = SIE, the result would have been 7,624 735,000

00104·· 1 04 =. 10./10. or. percent

Using this approach instead of the polynomial, the result would be off by 0.074 percent of the strain (1.114 - 1.040). Even though the stress is not particularly high, this is a significant error of 7 percent. Unfortunately, this high a degree of accuracy is not attainable in all stress-strain problems. Because polynomials are based on curves generated by using only one stress direction, they do not apply in problems involving two or more directions at the same time.

Chapter 12

CONCLUSIONS

Plastics provide the designer with many different materials and processes useful toward meeting all the varying types of product requirements. They are also capable of producing from simple to complex shapes and are economically beneficial. They can be made to have a long life, they resist corrosive environments, and are recyclable, degradable, and can meet practically any performance requirements. They also permit the fabrication of products whose manufacturing would be difficult if not impossible in other materials. However, designers must routinely keep up to date on developments with the more useful plastics and acquire additional information on how to process them. The emphasis throughout this book has been that it is not difficult to design with plastics and reinforced plastics/composites and to produce many different sizes and shapes of thermoplastic and thermoset commodities and engineering resins, whether unreinforced or reinforced. The bases of material and process selection should be product performance requirements, shape, dimensional tolerances, processing characteristics, production volume, and cost [1-8, 14,40-43,46,62-68,73, 106-12, 119, 136, 156, 163,695, 807, 809-940]. Some plastics can be worked by many different processes, but others require a specific proc,ess. Process selection can take place before material selection, when a range of materials may be available, or made first to meet performance requirements and only then have the app1ic"able process or processes chosen. Usually, in the latter situation only one special process can be used to provide the best performance-to-cost advantages. A particular design group may have its own processing capabilities. Unfortunately, some operations use just whatever equipment is available. This situation could either be very unprofitable, limit profitability, or restrict product performance. It is important to recognize that the fabrication process can markedly influence all aspects of product performance, including cost. Compared to other material-based industries, plastics have enjoyed an impressive growth rate over the century since their inception, but particularly since about 1940. The productdesign community was quick to recognize the design freedom and great versatility that plastics' materials and processing techniques afforded. Recognizing a growing marketing opportunity worldwide, international plastics material suppliers started an endless cycle of developing new and improved materials to meet continually new design needs. Processing machinery builders worldwide then responded with improved equipment and totally new processes, as conventional tool shops everywhere expanded their capabilities to include mold and die manufacturing for the plastics industry (see Fig. 12-1). 877

878 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

STRUCTURAL PLASTICS

Figure 12-1. The interrelationship among the methods of plastic applications and processing for the family of plastic materials. From "Studies in the Development of Plastics Industries," United Nations.

The design, development, fabrication, and, importantly, marketing of a new product will normally call upon many different kinds of design skills. A plastic product could include at least the product design, industrial design, piece part design, mold or die design, packaging design, and others. It is possible for one designer to have all these most desirable specialties kept up to date for a specific product category. Usually, however, the designer responsible for the product will have total responsibility for producing it with a profit and thus have to delegate work to all these specialists. This is no problem if the designer is careful to set up complete product requirements and follow through logically with the opinions of others as they properly come together. In a market economy, which is to say the real world, that is ruled by competition, plastics will be employed only in applications where they can be expected to bring an overall economic advantage compared with other competing materials. In this connection it is well to note that the biggest competitor to a given plastic may be another plastic. On the basis of an overall benefits assessment taking in the full service of a plastic product, it has been shown in millions of cases worldwide that the use of plastics not only makes economic sense but also makes a contribution toward conserving resources.

CONCLUSIONS 879

Despite all this optimism, it must not be denied that the plastics industry will continually be confronted with substantial problems. For one thing, new legislation will exert a greater influence on the marketing of all types of products and the development of new applications. Questions of product licensing, product safety, and, in particular, the methods for disposal of a product at the end of its service life will play an increasingly key role in this context. It will not be easy to find a simple, rapid solution to all such problems without involving the participation and efficiencies of the plastics industry. The present problems, as well as ones that can be expected in the future, will not reduce the high innovation potential of plastics, and the growth market can be expected to expand continually. The reason for this optimism is the past history and ability of plastics to produce a new generation of materials to meet new requirements. When the market is ready, plastics could become the world's most fire-resistant material (which is not typical today), the strongest material in the world, and other factors highlighted throughout this book. PRODUCT DIVERSIFICATION

As covered in this book and as is obvious with the millions of different plastic products produced and in use worldwide, plastics are used in all markets to meet all types of conditions (see Figs. 12-2 and 12-3). Newly designed products are literally endless, to meet every requirement (see Fig. 12-4). For example, consider Techturf, a rugged, natural turf system for golf courses that is based on a PP-mesh reinforcement from BriAg Industries Ltd., a York-based subsidiary of leI Plastics. It consists of a sand and fertilizer growing medium held together with 2-

Figure 12·2. High-perfonnance reinforced plastic/composite recreational boat exterior that meets all kinds of conditions for strength, rigidity, toughness, and resistance to saltwater, sun, sand, and so on.

880 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 12-3. A newly designed child-resistant closure for plastic pails used to package swimming pool chemicals. This product, called Life Latch, is from Morris Enterprises, of Tennessee. It employs a locking screw-top cover that is easy for an adult to open but difficult for a child. Pressing in the lid's latches with one's thumbs permits the lid to be turned and lifted off. Quantum's Petrothene LS556 HDPE injection-molding material is used for the lid and pail, with their PP8420-HK impact-grade PP for the molded latches to provide excellent chemical resistance, strength, stiffness, and ease of molding.

and 3-in. squares of mesh made of ICI's PP. The mesh forms an interlocking matrix through which grass roots can grow. This new turf system is reported to provide excellent drainage and wear resistance. Long established as the materials of choice for auto interiors but now also gaining favor for use in exterior body panels, plastics are continuing to make impressive inroads in under-the-hood performance parts, including the potential replacement of just about every component in the power train. Intake manifolds, camshafts, and engine blocks are just a few such parts under development for tomorrow's cars. The mostly plastic auto engine that has been a reality since the debut of the Polimotor in 1980, with its subsequent success as a power plant in racing cars has now been taken seriously in Detroit. The latest version of the Polimotor, for which the Rogers Molding Materials Division of the Rogers Corporation holds exclusive rights, is a 2.3-liter, dual-overhead cam, sixteen-valve, four-cylinder engine. It is an open-deck type where through-bolts hold the cam carrier, head, combustion chambers, and block together. The 60 percent by weight phenolic-glass-fiber-reinforced engine weighs 175 lb., as against the 300 lb. of its metal

CONCLUSIONS 881

Figure 12-4. This Total Environmental Control System in GE Plastics' Living Environments Concept House seen in Chapter 1 combines the functions of five home comfort appliances, using TPs to provide aesthetics, heat and corrosion resistance, light weight, and ease of serviceability and installation, and is designed for manufacturability.

counterpart. It improves mileage, runs more quietly, has less vibration, and is virtually free of corrosion. The phenolic RP compound used for this engine provides the required excellent dimensional stability, creep resistance, toughness, and strength retention at high temperatures. During operation this material is exposed to temperatures of 120 to 200°C (250 to 400°F), as well as bolt loads that can exceed 69 MPa (10,000 psi). The U.S. Postal Service has also turned to plastic, to develop a new stamp that could be sold by twenty-four-hour automatic teller machines anywhere. One of the two nonlick postage stamps developed uses a new proprietary Fasson-brand polymer-based pressuresensitive adhesive. It was developed by the U.S. Materials Group of Avery in Painesville, Ohio. These stamps are printed on a sheet the size of a dollar bill. With the cooperation of banks, consumers would buy stamps and the cost would be deducted from their accounts. Plastic is the only material that could meet the thickness tolerances of automatic bank teller machines. With this stamp, including its adhesive and backing sheet, plastic

882 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

can produce a sheet that is durable, will not double vend from the machine, and will not jam. The stamp was test marketed in 1990 through the First National Bank of Seattle. When the first plastic shotgun shells appeared on the market in the late 1950s, sporting goods stores could not stock their shelves fast enough. Replacing paper with PE in the shell's body spelled an end to the century-old problems of moisture absorption and misfires caused by spoiled powder. It also reduced shell jamming and enhanced the smooth operation of automatics that were then being introduced [843]. By the end of the 1960s Du Pont's supertough-polymer technology had emerged. This material, by the late 1980s, was embodied in designs for center-fire cartridges of .38-caliber pistols, using casings made of Du Pont's Zytel ST 901L supertough amorphous nylon. This casing was both strong enough and tough enough to withstand the explosive forces that are generated when a bullet is fired. It offers marksmen advantages ranging from consistently high accuracy to decreased chamber wear, low-friction extraction, elimination of corrosion, and less recoil but more economy. In conventional cartridges the mechanical crimp that holds a bullet tightly in place creates a drag when it is fired. In a plastic casing integrally molded "interference fits" lock the same standard bullet into place. But in this case, when the cartridge is fired the burning gases expand the casing, instantly freeing the bullet as they propel it out of the barrel. Advances in medical technology often require advances in materials technology, as when Friddle's Orthopedic Appliances in Honea Path, S. C., needed a plastic replacement for the metal halo that formed part of its cervical brace for patients with neck injuries. The adoption of magnetic resonance imaging (MRI) technology necessitated the replacement of this metal halo with a material that could offer high strength and stiffness for this critical medical application. It is made of a long-carbon-fiber-reinforced nylon 6/6 composite produced by LNP of ICI Advanced Materials. The composite is easy to fabricate, since it is injection molded. Previously, the halo had been fabricated of stainless steel, with the superstructure of aluminum, so it had to be removed before an MRI scan could be made, after which the skull pins had to be carefully reinserted. This repositioning is eliminated with the plastic halo, which also costs 33 percent less and weighs 30 percent less. Among the many plastics used in medical applications are the following: Acrylics. In bone replacement, corneas, adhesives, hemostatic agents, dentures, contact lens, artificial eyeballs. Cellulose acetate. In nerve regeneration, packaging material. Formaldehyde-treated polyvinyl alcohol sponge. In support and growth stimulator for blood vessels outside and into heart muscle, hemostatic agent in repair of liver and kidney wounds, abdominal aortic grafts, vascular shunts, synthetic skin. Fluorocarbons. In artificial cornea, blood vessels, heart-valve coatings, reconstructive surgery, bone substitution. Polyamide. In vascular implants, syringes, clamps, blood transfusion sets. Polycarbonates. In syringes, parts of heart-lung machine, baby bottles, containers. Polyester fiber. In aortic and peripheral artery transplants. Polyethylene. In tubing, syringes, oxygen tents, repair of incisional hernias, stomach wall support, repair of tissue damage, heart valves, contraceptive implants. Polypropylene. In syringes, sutures, containers. Polystyrene. In syringes.

CONCLUSIONS 883

Polyurethanes. In plastic surgery, vascular adhesive, bone adhesive. Polyvinyl chloride. In surgical tubing, blood collection and administration sets, repair of congenital and traumatic facial defects, surgical drapes, balloon-type splints, adhesive bandages. Polyvinyl pyrrolidone. In artificial membranes for filtration of body fluids. Silicones. In heart valves, tubing, catheters, defoamers in blood oxygenators, urethral valve, plastic surgery, tendon replacement, lubricants, tissue substitutes. It may be common soon throughout the United States to have laws requiring that new toilets use less water. Molded-in ribbing, a TP polyester resin, and a mechanical seaming technique play key roles in the innovative design of a pressure tank for Water Control International's water-saving system for toilet flushing. Called Flushmate, this pressurized system uses only 1.5 gals. of water per flush, as opposed to 5 to 8 gals. with conventional gravity flushing. The heart of this tank is its pressurization system, in three parts: a ribbed cylindrical component injection molded of Hoechst Celanese's Celanex 4330 TP polyester, and two metal end bells attached to the cylinder's integral flanges by a mechanical seaming technique similar to that used to seal metal cans. The cylinder measures 7.5 in. in diameter, is over a foot long, and weighs about 4 lb. An equivalent vessel made of a suitable metal would be two to three times heavier and cost twice as much. Although its normal system pressure does not exceed 240 kPa (35 psi), the tank is designed and tested to withstand a maximum pressure of ten times that. In designing this tank to withstand such pressures the integrally molded openings posed a design problem. With its two distinct openings at the top and one at the bottom, this vessel seemed to have hoop stresses too great for plastic. The solution was a ribbed design in a Hoechst Celanese resin that has a particular elasticity that works well. A total of twenty components are needed to make up the system's pressure regulator, valves, and supply units. All are made of Hoechst Celanese's Celcon acetal copolymer. Water treatment systems have now come to include plastics, such as Du Pont's Permasep permeators for reverse osmosis (RO) water desalination, introduced in 1969 (see Figs. 12-5 and 12-6). Since that time such systems have been used in thousands of installations around the world for desalination of brackish water and seawater and to treat waste effluents. These permeators come in four product types, according to the type of water to be treated (see Table 12-1). MATERIALS DIVERSIFICATION

As with end products themselves there is an endless need worldwide to develop new and improved plastic materials to meet new design performance and process ability requirements [3]. These new materials include plastics, composites, superconductors, ceramics, metals, aluminum, concrete, and others. Plastics have been revolutionizing our times. The greater understanding of plastics' performance and design capabilities assures their future growth in providing widespread benefits. One example of such a beneficial product is Monsanto Chemical's recently introduced low-color styrene acrylonitrile (SAN), which offers economical processing advantages and enhanced optical clarity, as compared to the acrylic, PC, and styrene/acrylic (NAS) types. From Dow Chemical comes the first in a family of new TSs to compete with the polyimides in electronics. Derived from bisbenzocyclobutene (BCB), it is an extension of Dow's Quatrex resin line with potential use in multichip modules and the like. BCB

884 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

has a dielectric constant of 2.66 and a dissipation factor of 0.00 1 at 1 kHz, with extremely low moisture pickup, rather easy processing, and a glass-transition temperature above 350°C (662°F). GE Plastics has developed a reactive polymerization technology for select TP engineering resins that radically improves the ability of this material to penetrate and bond to fiber reinforcements ranging from continuous strands to woven-glass mats. Such wetting and bonding are properties for which TS polyester resins have been used since 1942. The original method of bonding fiber to resin involved surface treatments of the glass. Later treatments involved modifying resins to improve an RP's performance. Usually, any fiber or resin treatment furnishes improvements only with certain resins [14, 17,2734, 150-53]. The GE technology yields ultrastrong composites while retaining its processing, economic, and end-use performance benefits. Based on a widely known chemistry called cyclics, this technology yields an initial molecular structure that produces extremely low polymeric viscosity, which improves its wetability. For example, these cyclics are used with many TS epoxy resins. One goal in synthesizing high-performance TPs for graphite-reinforced RPs and adhesives is to develop composites with enhanced properties to include improved toughness, thermal stability, and melt processability. Cyclotriphosphazene-based monomers and polymer precursors have now led to the development of high-temperature materials for aerospace applications [847]. NASA's new polymers have melting points from 111 to 138°C (232280°F). They are useful for obtaining heat- and fire-resistant composites or molding compounds with high structural strength.

Figure 12-5. Du Pont's hollow, fine-fiber Permasep permeator for reverse-osmosis water desalination.

CONCLUSIONS 885 An .. Telescoping Devlte

BMe Seal C~rn~r

Figure 12·6. Du Pont's spiral wound "Permasep" cartridge.

Du Pont's Fiber G is a tough, lightweight material developed for composite applications ranging from space satellites and auto bodies to electronics and undersea cables. Fiber G is stiffer than steel, yet weighs only one-fifth as much and has a modulus of elasticity of 206,700 to 964,600 MPa (30 million to 140 million psi). There are five grades of this fiber. With this combination of performance of properties to weight, the designer of a satellite can eliminate many pounds in a structure being launched into orbit. It now costs ten thousand dollars to launch each pound of a satellite. Thisweight saving applies equally well to such other products as autos and aircraft. The high-performance thermal properties of this fiber are of interest in the electronic parts industry. Interestingly, in an oxygenfree environment this fiber gets stronger as the temperature rises, all the way up to I, 700°C (4,OOO°F). Fiber G is made from petroleum processed into a liquid crystalline powder or mesaphase pitch, then transformed into filaments so delicate they would tum to dust at the slightest touch. Further manufacturing, including heating to more than 2,760°C (5,OOO°F), strengthens the fibers by realigning their microstructure. Finally, this pitch-based carbon fiber (PBCF) is wound and bound together with an epoxy finish to become Fiber G. It is often the case that an old idea bears reexamination. In 1889, T. V. Hughes and C. R. Chambers patented a process for pyrolyzing marsh gas in iron crucibles to produce electric lamp filaments, one of many ideas whose time had not yet come. Today, General Motors Research (GMR) physicists are using hydrocarbon vapors to catalytically grow carbon fibers. The strong, stiff, discontinuous carbon fibers produced by this method have potential usefulness in many composite applications. Conventional carbon fibers are currently based either on the polymer PAN (polyacrilonitrile) or petroleum pitch (see Chapter 7). Although they have many properties that would make them useful in many matrices like plastics, metal, and cement, their expense has limited their use. In contrast, the GMR method is an inexpensive way to produce carbon fibers (called Pyrograf, because the method uses pyrolysis to form graphitic carbon). The testing of the vapor phase process has been so promising that GM's Inland Division has built a pilot plant in Dayton to produce Pyrograf.

886 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 12-1. Du Pont's Permasep Permeators Based on the Type of Water to be Treated Product Type Seawater B-lO

Brackish Water B-9

B-15

Col

Description A hollow-fiber aramid membrane permeator designed for seawater and high brackish applications where long membrane life is needed. Replacement bundles available. A hollow-fiber aramid membrane permeator designed primarily for brackish-water applications with a moderate level of suspended solids in the feedwater stream where long membrane life is needed. Replacement bundles available. A spiral-wound, aramid membrane cartridge designed primarily for brackish-water applications with a higher level of suspended solids in the feedwater stream where long membrane life is needed. A spiral-wound, cellulose acetate membrane cartridge designed primarily for brackish-water applications with a higher level of suspended solids in the feedwater stream, where shorter membrane life is acceptable.

For years the use of aluminum plugs has required thermoformers to maintain temperature-control equipment, often resulting in greater fabricating costs. Recently, a growing number of designers have come to decide that plugs made with syntactic foam (see Chapter 7) are as durable as aluminum plugs but can also reduce production cost. For example, the Syntac 350 syntactic foam manufactured by Emerson & Cuming, Inc., of Canton, Mass., has low thermal conductivity and can withstand sustained temperatures up to 177°C (350Dp). Consequently, these plugs no longer alter the temperature of a plastic sheet as it passes through thermoforming machinery. The plugs now provide consistent performance, ensure the production of uniformly shaped packages, and require no support equipment. These plugs require no warm-up period at the start of each day, thus enhancing overall productivity. Because of their light weight, they increase the longevity of their expensive thermoforming machines. Blocks of this type of foam as purchased by thermoformers can be easily cut. The plugs can be easily cut by the fabricator into a wide variety of shapes and sizes. In contrast, aluminum plugs usually first require a pattern to be made, which is then sent to a foundry to produce a sand or plaster shaped cavity, after which the aluminum is poured to form a rough casting, then machined and polished. A new process to chemically separate commingled postconsumer plastics into a fairly pure regrind has been successfully tested at Rensselaer Polytechnic Institute in Troy, New York. This process is intended to be commercially available within a few years. These researchers have taken shredded, postconsumer, contaminated plastics and fed them into a column that selectively pulls out the polymers through a solvent process. In the first cut PVCs and PSs are taken out, and on the second cut LDPE is extracted, all at room temperature. On the third cut the column is heated to 71 °C (160°F), pulling out the PPs and HOPEs. Another extraction requires a temperature of 88°C (190Dp), netting the PETs. The drawback of this chemical-separation system thus far has been that the resulting polymer does not meet the specifications for virgin resins in some cases (see Chapter 8). The initial tests showed good separation efficiencies with PP and HDPE, but to date there remain some problems with the PVC/PS split. The property drawbacks are expected to be overcome by using other solvents before the extraction. The current cost is 13 cents per lb., including amortization of the capital start-up costs.

CONCLUSIONS 887

Even though new developments in materials will continue to occur, the plastics already in use are finding new applications, based on their individual characteristics. There are TPs with a memory, and others that can be bent, pulled, or squeezed into various useful shapes. But eventually, especially if one adds heat, they return to their original form. This behavior, known as plastic memory, can be annoying. But when properly applied, plastic memory offers some interesting design possibilities for parts (see Chapter 2). As discussed throughout this book, particularly in Chapters 3, 4, and 5, certain technical or engineering factors follow prescribed patterns. However, there are cases where they change because of the type of design applied. One example is Poisson's ratio of lateral strain to axial strain, which always falls within the positive range of 0 to 0.5 for any material. There is, however, a basic design configuration using a differently fabricated shaped cell structure that changes a material's strain behavior [870]. When a sample of polyurethane foam is stretched, its cross-section either moves or grows flatter, and when compressed it becomes thinner. Thus, a PUR has a negative Poisson ratio. This "reentrant" foam, as it is called by its inventor Roderic Lakes, a professor of biomedical engineering at the University of Iowa, is the only material thus far to exhibit negative action. The key to this behavior is in its microarchitecture. Whereas conventional foams have a convex cell structure, the ribs of each of the new foam's cells permanently protrude inward. When a tensile force is applied, the ribs push outward, causing this foam to expand laterally. With a compression force the ribs collapse into themselves, causing the material to contract laterally. Each reentrant foam starts out as a conventional plastic foam with a positive Poisson ratio and a convex, open-celled structure. The conventional foam is then compressed triaxially (that is, in three orthogonal directions) and placed in a mold. The mold is heated to a temperature slightly above its softening temperature, then cooled to room temperature. Any foam subjected to this conditioning that then possesses a permanent volumetric compression factor between 1.4 and 4.0 exhibits a negative Poisson ratio. When compared to conventional foams, reentrant foams display superior resiliency and toughness. This structure has applications where the redistribution of stresses is desired: in air filters, flexible fasteners, gaskets, sound-absorbing layers, fillers for highway joints, ankle wraps, wheelchair cushions, and many more. An air filter made of reentrant foam would, for example, address the problem of a pressure rise behind a clogged conventional filter. The pore space in the reentrant foam would open rather than close as the pressure increased, to prevent clogging. And if it were used in a flexible fastener, it would expand when the pull-out tension was applied.

EQUIPMENT IMPROVEMENTS New equipment nearly always offers potential or actual significant improvements in processing capabilities. Designers should always plan for the equipment they use to aid in meeting the goal of zero defects, as well as reduce production costs. There are many "old" machines in operation, especially in the United States, so in certain operations there could be ample room to improve and simplify plastics processing. In the United States there exist an estimated 80,000 injection-molding machines, 12,000 extruders, and 6,000 blow-molding machines. For each of these types of machines about 30 percent are under five years old, at least 35 percent are five to ten years old, and the rest are more than ten years old.

888 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Annual U.S. equipment purchases, including imports, have usually been about 5,000 injection machines, 1,500 extruders, and 1,000 blow-molding machines. Only about 10 percent of the U.S. machines use robots, compared to about 90 percent that use robots in Japan and other rather high percentages in different industrialized countries. The largest automated injection molding machine, from Battenfeld of America, that is capable of molding parts measures up to 6 x 5 x 4 ft. It is in a plant in Oak Brook, Ill., owned by Waste Management, Inc., which describes itself as the world's largest provider of waste- and environment-related services, with over $50 million in annual sales. This mold and 1M machine was designed to produce large waste containers from HDPE, replacing their original metal counterparts used by other companies like Du Pont. The HDPE containers are lighter in weight, have no need to be painted or repaired, are not subject to environmental wear and tear, and have a life span at least double that of metals, which last only two to three years. HDPE can be reground and reprocessed into containers. The 1.5 m3 (2 yd. 3) containers weigh 48 to 55 kg (l05 to 120 lb.), depending on the design. The clamping system uses two interconnected 4,5OO-ton plasticators (that is, injection units) positioned side by side just four inches apart. Two mechanically separated clamping units can be joined and be simultaneously under comprehensive closed-loop control to make ultralarge parts. Alternatively, the clamps can be used separately. To cope with the production of quite large parts, this unit incorporates multiaxis robots to demold and transport the molded part. The whole twin unit, which features a combined injection rate of 18 Ib.lsec., is managed by a Unilog 9000B control system that made its debut at the 1989 Dusseldorf Plastics Fair. The mold alone weighs 127 t (140 tons) for the 1.5 m3 (2 yd. 3 ) container, measures 305 x 153 x 216 cm (120 x 110 x 85 in.), and has an opening force of 545 t (600 tons). The top and bottom tie bars are 91 ft. apart and the side-to-side platen space is 18 ft. at a maximum mold thickness of 8i ft. In 1990 Davis-Standard shipped a tandem extrusion complex system to Taiwan that consisted of .25 cm (10 in.) and 31 cm (12 in.) diameter extruders for the high-speed production of biaxal1y oriented PP film. This tandem arrangement delivers 3,992 kglh (8,800 Ib.lhr.) of high-quality extrudate to the die. Upon its startup it was the largestknown such installation in the world. Tandem as a technique has become quite popular because it offers the advantage of high product throughput (see Chapters 2 and 7). The configuration also furnishes backpressures, by utilizing custom feed-screw designs for each extruder to allow certain speeds to be used for plastic melting, mixing, and conveying. These high-speed short-screw (that is, length/diameter) machines also reduce plastic residence time and can conserve floor space. Plastic's capability of saving on use and cycle time, its better surfaces and less stress, occur most notably with hollow, thick-sectioned parts made by the gas-injection process. This method uses a shot of material that is insufficient to fill the mold cavity, with the remainder of the injection-molding process being the usual 1M technique. Nitrogen gas is introduced through the mold's gating, pushing the melt to the hot cavity walls and forming hollow sections in the part. The locations, sizes, and shapes of the hollow sections are determined by the mold design, part design, amount of material and gas used, and control of the gas-injection process. The weight and cycle time of this process can be 50 percent less than for solid parts. The now-expanding approach of using telephone connections to unite the processor with the equipment supplier saves production downtime and cuts costs. Processors can now have their routine maintenance problems solved by the equipment suppliers' technicians in a matter of minutes, without a technician's having to travel to the plant for a

CONCLUSIONS 889

hands-on repair [12]. The result is drastic reduction in downtime from that which is typical for common processing-line problems. Importantly, such remote diagnostic systems represent relatively small capital investments for processors; a typical modem, for instance, costs only between $300 and $600. The use of a modem for diagnostic purposes comes as a natural progression of two trends in process-control technology: increased connectivity among computers, and improved diagnostic software. Coupled with the continuing shortage of good technical people, the incentives for processors and equipment suppliers to pursue remote diagnostics are strong. A survey of European injection-molding plants conducted by Phillips GmbH of Germany, one of the world's largest suppliers of industrial control systems, showed that 60 percent of all machine downtime resulted from operator error, 30 percent could be attributed to mechanical failure, 9 percent was due to faulty electrical systems, and just 1 percent resulted from faulty process controls. Several plants reported that having a modem did reduce significant amounts of downtime in all these categories.

THE SOLID-WASTE PROBLEM AND PRODUCT-DESIGN SOLUTIONS The waste-management problems of the United States and the rest of the world continually threaten to reach crisis proportions. The U. S. takes top billing, with each person producing at least 3! lb./day of waste, and to date it has literally done little to take corrective action. Such other countries as Germany and Japan generate about 2 Ib./day/person but have taken some positive corrective action. Industrialized countries have generated a lot of garbage for a long time but now are rapidly running out of environmentally acceptable landfills. Unfortunately, this problem expands with the world population. At present, more than 2 billion lb. of solid waste are pouring into waste streams annually worldwide. There is no single, simple answer to this problem. Different, limited approaches have been used successfully, and much more action has begun occurring here and internationally to integrate environmentally secure landfills, recycling, advanced waste-to-energy incineration, degradability, product design, waste-source reduction, industry support, public education and support, regulation support, and various economic considerations. We now should stop merely living with past problems and start solving them. Waste is a widespread, but solvable, problem; there is an abundance of possible cures and fixes, some good, others not so good. There are nevertheless logical approaches and facilities to check their reliability, rather than just criticizing them [164-200, 855-63]. This overview includes information and positive actions being now taken to provide solutions that will affect all materials. Because plastics usually receive the biggest emphasis, they are the main focus here. Plastics as well as other materials must all definitely be seen as problems (see Table 12-2). Practically all plastics can be made recyclable, incinerable, or degradable, but the conflict of product-performance requirements as against economics in most past applications has prevented these factors from being viable. Actions have thus been taken by the plastics industry here and abroad to make positive steps toward helping to reduce plastics waste by recycling, incineration, and so on. Unfortunately, generalizations that "plastics are bad" and "burning plastics always generate toxic products" are too often heard from customers and media representatives. More unfortunately, plastics packaging is a highly visible element in the waste stream. And the negative public perceptions about plastics sometimes lead to negative opinions about the companies that use them. These companies must then respond to consumer opinions to maintain their reputations.

890 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Table 12-2. Estimated Contributors to Solid Waste1 Percent

Percent by Weight

Waste Material

by Volume

37 18 10 9 8 7 11

Paper2 Yard Metal Glass Food Plastic

40 18 2 3 8 9-123

Others

13

'Total annual solid U.S. waste is estimated to be more than 300 billion lb. (136 billion kg). 'Includes, by volume, 12% in packaging, 12% in newspaper, 4% in cardboard, 2% in magazine, and 10% in others. 'By some reports, up to 16%. Note: Seven percent of 300 billion lb. is 21 billion lb. of plastics. Annual u.s. plastics consumption is about 67 billion lb. (30 billion kg), with domestic products at about 61 billion lb. (28 billion kg) (including about 33 percent, or 20 billion lb. [9 billion kg), in packaging) and imported products at about 6 billion lb. (2.7 billion kg) (contained in electronics, autos, appliances, packaging, medical products, and so on). Waste, like computer programming and nuclear physics, tends to be a subject shrouded in mystery and reportedly undentood by only the few. The annual U.S. Control and Service Environment Business is estimated at $80 billion or larger than the total computer business of all hardware and software, plus the telecommunications and airline businesses.

The plastics industry has fallen victim to an unrelenting international smear campaign, particularly in the United States, Discriminatory measures have been taken in a number of countries against plastic packaging, although scientific investigations have proven that certain products in fact have nothing to do with the rise in the amount of domestic refuse. On the other hand, the demand for plastics products among consumers, who readily appreciate the advantages of this material in day-to-day living, has risen so much that there is now a distinct possibility of disposal bottlenecks arising. The throwaway aspect, particularly with regard to fast-food packaging, of today's society has resulted in what has been billed in the press and by legislative bodies as the nation's solid-waste crisis. Ironically, in many modem composite landfills a high-density polyethylene (HDPE) liner is used to reinforce the conventional clay layers, as a way of minimizing leaching. The matter has become a crisis simply because many cities are facing the dilemma of how to dispose of their municipal waste. Many municipalities have filled existing landfills and establishing new ones is becoming more and more difficult. The plastics industry's response has been to commit itself even more firmly to recycling and to reaffirm its earlier position that waste-to-energy incineration is critical.

Statistics: Fact and Fiction There is no shortage of statistics about the growing municipal-solid-waste (MSW) disposal problems (seen in Table 12-2) as well as the amount and types of waste materials. Because plastics are lightweight, they translate from a rather low percentage weightwise to a large percentage by volume. Every leading study has shown that plastics make up a smaller share of solid waste than do paper and other materials. Still, in spite of the facts the public and legislatures continue to identify plastics as the major MSW offender. The result has been a growing proliferation of laws banning or limiting the use of plastic products. There exist generally held perceptions that the disposal of plastic wastes in landfills or incinerators is harmful to the environment and human health and that the environment and human health would somehow benefit if plastics were eliminated from the MSW stream. Many researchers have determined that these phobias are unwarranted, because

CONCLUSIONS 891

plastics cause little pollution of either land or air and are among the most readily combustible components in an incinerator. Wherever a specific plastic poses a potential pollution or other type of hazard, there are proper procedures available for properly and safely disposing of them. Another reason why plastics' role in the MSW problem is highly overrated is that they are durable materials that resist the effects of exposure to the elements. The fact that most plastics are not biodegradable may not sit well with many people, but it still does not change the percentages in Table 12-2 or in the MSW stream. It has been reported that the United States recycles only about 10 percent of all its waste, incinerates about 13 percent, and assigns the remainder to landfills. Japan recycles 50 percent, incinerates 34 percent, and landfills 16 percent. Western Europe recycles some 30 percent and has large-scale waste-to-energy incineration. These other countries have had to take earlier action, since they literally have no landfill areas in the way the United States does.

Landfill Fortunately, in the United States, as compared to other industrialized countries, municipal landfills can playa bigger role. They will continue to be needed at least for nonrecyclables and the ash from incinerators. They will no doubt be required to operate so that they are environmentally sound and meet all applicable regulations. Unfortunately, many former or existing sites are irresponsibly run and national and worldwide disgraces. Only some 15 percent of these old landfills are lined to restrict contamination by leaching of the surrounding area. The Environmental Protection Administration estimates that in 1987 about 18,500 U.S. landfills were operating, 1989 had about 8,000, and about 4,800 are projected by 1994. So while landfill sites are being reduced, the waste problem increases. It has been determined that practically nothing degrades fully in most landfills, which hold glass, food, paper, newspaper, aluminum, steel, plastics, and so on, in contrast with the popular belief that only plastics do not degrade. The reason for this is simply that the moisture and oxygen needed to encourage degradation are not present. The U.S. packaging industry estimates the following just for plastics:

Table 12-3. Estimated Disposition of

u.s. Plastics Used for Packaging

Percentages by Weight Year

Landfills

Recycling

Incineration

1987 1992 1997 2002

96.1 66.5 46 37

1 28 44 43

3 5 8 18

Year 1987 1992 1997 2002

Weight (per billion pounds) Estimated Total Solid Plastic Waste in U.S. Due to Packaging (6.8 (9.1 (11.3 (15.4

billion billion billion billion

kg) kg) kg) kg)

Biodegradation

o .5 2 2

892 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Recycling

Recycling of waste has become an important approach and a profitable business for many. At present very little in the way of plastics is recycled, which represents less than 1 percent of it by weight. In comparison, about 32 percent of aluminum, 25 percent of paper, and a small amount of steel are recycled profitably. It so happens that, technically, plastics can be one of the easiest materials to recycle, but economically this tends not to make sense. The reason is the present high cost of collecting, sorting, and processing plastics and other materials. Making any recycling system work requires public, industry, and local, state, and federal support. Fortunately, recycling efforts are on the increase, with more than 1,000 curbside recycling collection programs under way and expanding nationwide. For example, since 1987 Rhode Island has been successfully recycling MSW of all sorts in specific localities. The result has been the separating and selling of recycled paper, aluminum, glass, and plastics. Different procedures with their own auxiliary equipment are used to separate these materials, such as aluminum cans being separated from glass and plastic by eddy current equipment and conveyed to flatteners. Plastics separation involves granulating milk bottles (HDPE) and soda bottles (PET), which are also separable by color. At present 15 percent or some 60,000 tons per year of Rhode Island's residential waste is disposed this way. New York City, with its 15,000 tons of trash daily, has now turned to recycling to alleviate its landfill problems. By 1995 at least 25 percent of its trash is expected to be recycled. Business opportunities are gradually developing to produce many different products from recycled plastics such as detergent bottles, office equipment, highway barriers, wastebaskets, pallet strapping, tool boxes, fast food trays, wetland walkways, signs, hampers, boat docks, park benches, carpeting, irrigation pipes, and many others (see Figs. 12-7 and 12-8). Until recently, most recycling programs concentrated on paper, glass, and metal. There was little activity with plastics except in nine bottle-deposit states or some localities where commercial collectors were active. However, with the waste problem finally exposed, legal action aiming at curbing the use of plastics, limited action being taken with public support, and landfill sites disappearing and more waste developing, a new undeniable urgency exists to solve the MSW problem. Some of the companies already involved, who are are spending over $40 million per year in producing plastic recycled products or materials, include Rubbermaid Inc., Turtle Plastics Co. (Cleveland), Rubber Research Elastomerics Inc. (Minneapolis), DecTech Labs (Amherst, NH), Processed Plastics Inc. (Ionia, MI), Trio Products Inc. (Elyria, OH), IPPI (Greensboro, GA), MRC Polymers Inc. (Chicago), Mammers Plastic Recycling Corp. (Iowa Falls, IA), plus Mobay, General Electric, Dow Chemical, Hoechst, Bayer AG, Ellman, and others.

Incineration

Incineration solves a lot of solid-waste disposal problems-done properly, it can reduce the volume of solid waste by at least 90 percent. And waste-to-energy plants provide a reduction in disposal costs. With high-energy plastics of some sort in the waste, these significantly increase the ease of incinerating the other materials and provide a higher waste-to-energy economic value for the plant's operating costs. Plastics have the highest

CONCLUSIONS 893

Figure 12-7. Goodyear had this two-piece suit and matching tie made from recycled two-liter polyethylene terephthlate (PET) beverage bottles in 1978 and in 1990 donated it to the new Ripley's Believe It or Not Museum in Wisconsin Dells, Wisc. The recycling process that Goodyear developed shreds bottles into small flakes that can then be processed into a reuseable TP polyester resin. Goodyear had the suit made to demonstrate the versatility of its recycled PET.

stored energy value of any material, and also the lowest energy cost to produce and process into products. Most plastics bum cleanly, producing emissions of carbon dioxide, nitrogen oxides, and water vapor, but some produce unwanted by-products such as polyvinyl chloride (PVC). However, PVC and such other by-products can be safely burned at high temperatures of 980 to 1,650°C (1,800-3,OOO°F), using controlled oxygen input, sufficient cycle (residence) time, typically i to 2 minutes, and appropriate auxiliary equipment like scrubbers and solid salts. However, most U.S. incinerators operate below 87°C (1,600°F) and use only limited auxiliary equipment. For example, incinerated PVC generates undesirable chlorine (and bleached paper generates much more chlorine). Exhaust scrubber systems must be used to remove this chlorine. It has been estimated that some 135 waste incinerators operate in the United States and process about 13 percent of all trash. Another 93 plants now being built should pennit

894 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 12-8. Wildlife conservation and plastics recycling efforts were united by Custom-Pak, Inc., of Clinton, Iowa, in its manufacture of plastic wood-duck nest boxes. The boxes are injection molded from blends of HDPE reclaimed from milk bottles with virgin Petrothene HDPE from Quantum Chemical Corp. The double-walled outer-shelled plastic box resembles a tree trunk with a nesting cavity. It has an opening just large enough to permit entry by wood ducks, but not by predators such as raccoons.

the handling of a total of about 25 percent. With more incinerators being under considerations, perhaps half of our MSW could be incinerated by 2010 in energy-recovery plants. Thirty-eight states now have on-stream or planned waste-to-incineration plants. In Japan, 68 percent of their trash is already incinerated, in 1,915 smaller-sized plants. Modem incinerators can use the latest pollution-abatement technology equipment and other environmental safeguards to eliminate the exhaust contamination that might arise from certain plastics and other materials.

Degradable Plastics After a long period of continuing successes at improving plastics' durability, there is now more emphasis on using degradable plastics. As is generally recognized, plastics do not readily degrade. Some of the methods used to do so include UV-sun exposure (with appropriate additives), bacteria or enzymes (additives like starch to aid the microorganisms), water solubility, and others.

CONCLUSIONS 895

Degradable plastics have caused more public controversy than other approaches to providing a meaningful solution to MSW problems. Using degradables is definitely not an overall solution, but it does have some potentially useful applications such as trash bags and particularly mulch films. Degradability unfortunately conflicts with recycling, and there is not enough known yet about the products of degradation. Degradability is neither magic nor a quick fix. For example, polyethylene (PE) film bags that contain cornstarch as an additive to further biodegradability cause the starch to become carbon dioxide and water-but the PE remains, simply in a different form. The potential exists with any system that degradation may occur before the plastic product fulfills its intended life or purpose. If a degradable is accidentally mixed with nondegradable plastics for producing products, severe problems could erupt. Also, properly built landfills are basically not conducive to promoting the degradation of any material.

Product Design In product design there has always been the desire to use less of any material, because the result is usually a lower-cost product. On the other side of the issue is the use of more material to provide for a higher design safety factor beyond what is required. Thus, unfortunately, there are designs using more material than needed, particularly when using plastics. It is inexperience in designing with plastics that causes this problem. Many designers lack the know ledge of at least relating a material's performance to the processing variables that directly influence safety factors and the amount of a plastic to be used. With the flexibility that exists in designing with plastics, there are different approaches that may be used to reduce part weight, such as applying different shapes like internal ribbing, corrugations, sandwich structures, and orienting or prestretching. All this activity is aimed at producing products that use less in the way of materials and in tum let less material enter the solid-waste stream. Some designers, including myself, have habitually listed in product design specifications that specific environmental requirements should be met. A designer sometimes has an opportunity to use a material that provides no problem in the solid-waste stream or to use a design that lets lower-cost recycled plastics be used. In fact, blends of virgin (not previously processed) plastics with recycled plastics could permit the meeting of required product performance requirements. This approach has been used for the past century, but now there will be more use of it as more and more recycled plastics become available. However, the designer must take into account the usual lower performances that will occur with recycled plastics, which is not actually a problem. The recycled plastic will also have a degree of different contaminants that would eliminate its use in certain devices or products, such as in medicine, electronics, and packaging. However, within these rather small market applications there are designs with three-layer coextruded, coinjected, or laminated structures having the contaminated plastic as the center layer, isolated by "clean" plastics around it. This situation is not a problem in most products. Because the contaminants could be only a microscopic amount of steel particles, there could be increases in certain of the material's properties. Another method of reducing the quantity of plastics that has been used in certain products is to use engineered plastics with higher performance than the lower-cost commodity plastics. When applicable, this approach permits using less material to compensate for its higher cost. With a thinner-walled construction there could also be additional cost savings, since less processing heat, pressure, and time cycle is required.

896 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

To reduce significantly the use of plastics as well as other materials, designers could apply completely different nontraditional approaches to producing products, some of which have already been used and others for which an endless set of ideas seems to exist. Some examples include packaging liquids and creams in flexible plastic pouches or bags rather than the solid or rigid heavier plastic, glass, aluminum, and other containers. Food can also be packaged in tubes of aluminum or plastic with high-quality graphics.

Industry Support Worldwide efforts are being made by the plastic industry to accelerate solid-waste programs. The efforts include better methods of sorting waste such as separating vehicle parts, developing better methods to separate thermoplastics and thermosets, burning sewage sludge more efficiently with plastic wastes instead of fossil fuels, eliminating the use of the chemical chlorofluorocarbons (CFCs) that are used in certain plastic foamed products (alternates are being developed), removing certain additives and fillers in plastics that contaminate incinerators or landfills (such as cadmium and other heavy metals), and various others. Many plastics producers and consumer industry leaders have become involved in major recycling ventures: Mobil, Du Pont, Amoco, Dow Chemical, Proctor & Gamble, Heinz, Pillsbury, Beatrice, Campbell, Continental, and others. Certain companies have long been recycling, such as Wellman, Inc., of Shrewsbury, N.J., which is the largest U.S. recycler of postconsumer plastics and scrap plastic fibers. There has always been some degree of development, principally by equipment manufacturers, to improve the quality of recycled plastics and make them more competitive with virgin materials. To expand R&D, the Center for Plastic Recycling Research, a foundation established in 1984 by the Society of the Plastics Industry (SPI), is making important headway in solving some of the problems that tend to limit the use of reclaimable plastics. In December 1985 this group dedicated an experimental pilot plant at Rutgers University specifically to advance research into plastics recovery and reuse. The Plastic Bottle Institute of the SPI is another example of industries' participating in the recycling programs. PBI provides the industry with a list of recyclers identified by state, the type of resin (plastic) they handle, and their business category. These latter include the type of process, the minimum pounds required per pickup, the end use of the recycled plastics, and so on. For example, a few of the products that can be made from recycled plastic PET bottles (such as carbonated injection-stretched, blow-molded bottles) include strapping, twine, boards, and trash cans. With PET, bottle recycling purification levels are listed based on the suggested guidelines from recyclers and end users. These value levels, by the degree of processing, are, in descending order, pelletized, clean flake or granulated, uncleaned flake or granulated, baled bottles, and loose or bagged bottles. The PBI's Reference Guide includes information on recycling PET bottles into aromatic polyols, unsaturated polyester resins, recycling bottles for fuel value, and the like.

Public Education and Support Resolving the solid-waste crisis requires effective public education so that proper corrective actions can be taken. The intention is not to have isolated or incorrect slants on problems that just expand existing misinformation. Plastics are not perfect, nor is anything else, but they all play important parts in our society.

CONCLUSIONS 897

All materials, including plastics, definitely create major waste problems. Corrective action must therefore be taken in a positive direction. Effective education involves having people participate in programs such as not littering, separating trash, and purchasing or storing products wisely so that excess trash is not deliberately created. It is essential to provide the public with the real facts concerning solid-waste problems, to identify their causes and the actions to be taken for proper control and management of these problems. The public should recognize that human survival actually requires the use of plastics, because every industry uses it to benefit people, including its use in packaging. Furthermore, the manufacture of plastics requires only about 1.3 percent of the oil and gas consumed in the United States and less energy is required to produce plastics than such other materials as glass, steel, or aluminum. Plastics have also earned the name "white coal," because of their high heat value, which is about twice that of coal. Plastics would not be a simple substitute for other materials, if they were not to be used. Plastics have instead become the material of choice in many million-dollar markets and have made huge inroads in virtually all markets like packaging, construction, autos, boats, aircraft, electronics, medicine, and agriculture. Plastics are undeniably part of the MSW problem. Even though the industry is taking action to reduce the waste crisis, there will always be more to be done. However, because this problem should have been stabilized many years ago but was not, the corrective actions of all those involved cannot cause the problem to disappear overnight.

Economic Considerations

To obtain a clean, controllable solid-waste environment requires adopting rigid and appropriate policy regulations to be met by industry and the public. A coherent policy should provide an equal competitive opportunity to allow the cost of products to increase in order to pay for a cleaner environment. The unfortunate note of reality that usually has to be included is that nothing can be recycled against the grain of economics in our competitive environment. The challenge has been to learn how to collect, separate, reprocess, and market what has to be a lowcost material while at the same time making a profit. Definite progress is emerging in this regard. What has been occurring in the United States even more so internationally is extensive R&D in producing new equipment for handling and processing all types of waste, from pickup to the final stage of landfill, recycling, or incineration. Regulatory Support

Corrective action definitely requires having the proper and minimal government, state, and local regulatory support, with the understanding that it will not occur overnight. This support should include such incentives as tax-free financing for setting up plants like waste-to-energy incinerators. New York City now requires that its plastic bins used for collecting trash include 50 percent by weight of recycled plastics. In summary, the major actions that will make significant inroads on the solid-waste crisis involve 1) upgrading incineration; 2) upgrading landfills; 3) removing additives in plastics that cause ecological problems; 4) recycling; 5) expanding industry, public, and governmental support; and 6) designing products to use less material. In the meantime,

898 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

the use of plastics will continue to increase at their usual above-average rate as compared to other materials, as it has for over a century, since the first plastics were produced, in 1868. For the record, it has been reliably reported upon examining various environmental issues that the world is not only livable but healthy [158]. According to Melvin A. Bernarde and others, "Mobilization of the public from the fixed helplessness cultivated by media hysteria and development of an accountability on the part of speakers and writers who shape public opinion" can be the real threat and damage to our health, including our mental attitude. It is essential to recognize that there is always hope and a way to resolve manmade problems. TECHNICAL COST MODELING

The adoption of any technology for producing manufactured products is characterized by a wide range of processing , materials, and economic consequences. Although considerable talent can be brought to bear on the processing and engineering aspects, economic questions remain. Cost problems are particularly acute when the technology that will be employed is not fully understood, as much of cost analysis is based on historical data, past experience, and individual accounting practices. Historically, technologies have been introduced on the shop floor incrementally, with their economic consequences measured directly. Although incorporating technical changes in the plant to test their viability may have been appropriate in the past, it is now economically infeasible to explore today's wide range of alternatives in this fashion. Technical cost modeling (TCM) has thus been developed as a method for analyzing the economics of alternative manufacturing processes without the prohibitive economic burden of trial-and-error innovation and process optimization [10, 12]. TCM is an extension of conventional process modeling that particularly emphasizes capturing the cost implications of process variables and economic parameters. By coordinating cost estimates with processing knowledge, critical assumptions (processing rates, energy used, materials consumed, etc.) can be made to interact in a consistent, logical, and accurate framework of economic analysis, producing cost estimates under a wide range of conditions. For example, TCM can be used to determine the plastic process that is best for production without extensive expenditures of capital and time. Not only can TCM be used to establish direct comparisons between processes, but it can also determine the ultimate performance of a particular process, as well as identifying the limiting process steps and parameters. TCM uses an approach to cost estimating in which each of the elements that contribute to the total cost is estimated individually. These individual estimates are derived from basic principles and the manufacturing process. This reduces the complex problem of cost analysis to a series of simpler estimating problems and brings processing expertise rather than intuition to bear on solving these problems. In dividing cost into its contributing elements the first distinction to be made is that some cost elements depend upon the number of products produced annually, whereas others do not. For example, the cost contribution of the plastic is the same regardless of the number of items produced, unless the material price is discounted because of high volume. On the other hand, the per-piece cost of tooling will vary with changes in production volume. These two types of cost elements, which are called the variable and fixed costs, respectively, create a natural division of the elements of manufacturing part cost.

CONCLUSIONS 899

The variable cost elements are those elements of piece cost whose values are dependent on the number of pieces produced. For most plastics fabrication processes the principal variable cost elements are the material, direct labor, and energy costs. Fixed costs are those elements of piece cost that are a function of the annual production volume. Fixed costs are called fixed because they typically represent one-time capital investments (for buildings, silos, processing machines, etc.) or annual expenses unaffected by the number of parts produced (the building rent, engineering support, administrative personnel, etc.). Typically, these costs are distributed over the total number of parts produced in a given period. For plastics processes the principal elements are main machine cost, auxiliary equipment cost, tooling cost, building cost, overhead labor cost, maintenance cost, and the cost of capital. To demonstrate the use of such a comparative cost analysis, the production of a panel was analyzed according to different processes (see Fig. 12-9) [11]. In these case studies the following conditions existed: 1) the panels measured 61 x 91 cm (24 x 36 in.) with the wall thickness dictated by the process and part requirements so that the weights of the panels differed; 2) production was at a level of 40,OOO/yr.; 3) the plastics for all panels were of the same type, except that different grades had to be used, based on the process requirements, so that costs changed; 4) each panel received one coat of paint, except that the structural foam also had a primer coating; and 5) costs were allocated as needed to those processes that required trimming and other secondary operations. TCM can keep cost data current, based on cost changes from day to day, region to region, and so on. Of course, the means of keeping these data updated require that those costs be obtained on a regular basis and incorporated into the TCM.

$ 35 $ 30 $ 25 u; $ 20

e

~ $15

ii:

$10 $5 $0 c::J

Material

Inj

Foam (SJ

Finishing

EZ:::I

Overhead

T- Form ~

Equipment

Blow Ii:S:J

Labor

Processes: lnj = Injection molding Foam = Structural foam molding T-Form = Thermoforming Blow = Blow molding

Figure 12-9. Cost comparisons of panel production using a technical cost modeling program to show blow molding with the lowest piece cost.

900 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

SUCCESS BY DESIGN When desiring the success of a product many important aspects have to be considered, importantly, design. The key to analyzing design by designers, engineers, project managers, marketers, corporate managers, and others is whether it incorporates ergonomics and empathy. In other words, is it a product that truly understands the user or operator of the product? With designing there have always been the obvious factors of engineering, styling, and meeting performance requirements at the lowest cost. To some there appears to be a new era involving ergonomics, but this is not true. What is new is that there are easier methods always appearing on the horizon to simplify and meet all specific requirements of a design. Some designers operate only by creating a stylish outer appearance and letting engineers work within that envelope. Perhaps that is all the involvement needed to be successful, but a more in-depth approach would be better. Beginning with a thorough understanding of the user's needs and keeping an eye on ease of manufacture and repair, designers can work from the inside out. The envelope that emerges will then prove a logical and aesthetic answer to the design challenge. With the new materials and processes that have become available for over a century in plastics, the design challenge has become easier. Today's plastics and processes allow designers to incorporate and interrelate all the aspects that provide success. In products such as electronics, medical devices, transportation controls for aircraft, cars, and boats, and many others where user-friendly designing is required, it has to be obvious to all that plastics play the most important role.

DESIGN CONSIDERATIONS One factor that has continually done a great deal to harm the reputation of plastics is that in many cases so-called engineers have, after deciding tentatively to try to introduce plastics into a new application or to reduce the cost of a plastic product, lavishly copied the metal or plastic part proposed to be replaced. Too much attention cannot be spent upon recognizing the general principle that if plastics are to be used with maximum advantage and minimal risk, or used to make products to meet certain performance requirements, it is essential to cast aside all preconceived ideas of design in metals and to treat plastics on their own merits. A hard-and-fast rule, then, that should be kept by all intending to use plastics is "design for plastics." Designers for all kinds of products are regularly and pleasantly surprised to discover that plastics' materials and processes provide cost benefits. Eventually, even those who have been reluctant to do so come to acknowledge and accept the advantages of plastics over other materials for many applications. Chapter I highlighted how each material (plastics, steel, wood, etc.) has its own place in design. As a result of the ever-increasing use of plastics, designers for all kinds of products are being more and more required to learn how to use plastics. In many cases, designers who are experienced but have no formal training in plastics product design find themselves forced into the unfamiliar role of plastics product designers. In all too many cases the cost, quality, and reliability of the finished product suffer accordingly. Many of the plastic-product failures that consumers complain about could and should have been detected and corrected during initial design and development. Some degree of compromise is almost always necessary in designing parts in plastics and other materials. Arriving at the best compromise usually requires satisfying the

CONCLUSIONS 901

mechanical, electrical, or thermal requirements of the part, utilizing the most economical resin or composite that will perform satisfactorily and be attractive, and choosing a manufacturing process compatible with the part's design and the material chosen. Setting realistic requirements for these categories is of the utmost importance. Probably no plastic will provide a full 100 percent of the requirements for the desired performance, appearance, processability, and price. Then there are the "emotional" buyers for whom facts about usefulness are meaningless but for whom aesthetics is the real, and in fact the only, requirement. As described throughout this book, selecting the best-qualified material is not based simply on comparing numbers on published data sheets, because these values can be grossly misleading (see Chapters 3 through 6). For example, choosing the most economical plastic for a given part by comparing the cost per pound of various plastics is usually a mistake. Some plastics weight twice as much for a specific volume as others and thus would require twice as much of them to fill a part and possibly also cost twice as much to transport (see Chapter 2). A more meaningful comparison would be cost per specific volume such as per cm3 or in. 3 • But since there are expensive plastics far stronger than the lower-cost ones, cost-strength values should also be analyzed. In many applications paying more per pound or per cubic inch proves more economical if less material can be used, which usually results in faster processing. Making molds and dies, as well as assembling parts, can be quite costly, so the design should include as much function as possible in each product. A single part should when possible take the place of many individual parts, to eliminate assembly operations, reduce weight, and, frequently, improve overall structural integrity. Figure 12-10 gives a simplified explanation of this approach as compared to a sheet-metal enclosure with a welldesigned injection-molded plastic replacement. The designed part can also include easy access to other products, as illustrated in Figure 12-11 [2]. The minimum volume of plastic that will satisfy the structural, functional, appearance, and moldability requirements of an application is usually the best choice. This is in sharp contrast to machining operations, where one starts with a solid block of material and machines away only until what one needs to make the part function remains. Figure 1212 illustrates this contrast between a machined valve body and a properly designed, injection molded, equally functional part [2]. SHEET METAL

t

INJECTION MOLDED THERMOPLASTIC

@}'~.~

0}@/'

,.....,.L---

Figure 12·10. An example of an injection-molded one-piece housing (left) that replaced a fabricated sheet-metal housing (right) where many separate parts were required and had to be assembled.

902 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

Figure 12-11. The interior of GE Plastics' Living Environments Concept House features aGE Gecet TP resin substrate as part of a distribution floor developed by Infill Systems B. V. of the Netherlands, through which wiring and piping can be routed.

When designing heat-generating products such as appliances, consider the following guidelines: 1. Heating elements radiate energy, and all parts surrounding one will absorb some radiant energy. Generally, lighter-colored parts absorb less radiant energy than dark ones, because of reflection. 2. The radiant energy level is inversely proportional to the square of the distance from its source. A part 0.75 in. from a heat source is thus exposed to only one-ninth the radiant energy found on a part 0.25 in. away. 3. Using a reflector barrier or metallic adhesive-backed foil is an effective approach to avoiding radiant heat buildup in areas adjacent to the heat source. One example is the nearly all-plastic slide projector that has a hot electric lightbulb source; a metal reflector employing forced air directs the heat away from the plastics. 4. Double plastic walls provide an air passage outlet to help dissipate heat buildup away from its source. This construction allows the air between the walls to move or circulate. Convection dissipates excessive heat buildup by ventilation. 5. The most common way to protect plastic parts from high temperatures when close contact is required with a hot surface is by insulation. By placing a usually low-cost, low-thermal-conducting material between the part and the heat source, heat transfer can be sharply reduced or entirely eliminated.

CONCLUSIONS 903

Figure 12·12. An example of how to minimize material usage in plastic.

The temperature differential between the parts is inversely proportional to the crosssectional area perpendicular to the heat flow (A), the insulating material's thermal conductivity (K), and is directly proportional to the length of the flow path (L). Thus, thermal conductance is calculated as C = KAIL. Another term, thermal resistance (R), is defined as the reciprocal of thermal conductance, or R = lIC. A typical unit of thermal resistance is °C/watt. This can be considered as the temperature rise between two points per unit of heat transfered between them (see Chapter 2). Values for other materials are water at 0.36, mica at 0.14, and-a very effective medium-air at 0.016, with a vacuum being a nearly perfect medium. Design ideas in the automotive industry using plastics continue to be extensive. New ideas and applications for meeting performance-to-cost requirements are seemingly endless. There are all sorts of novel approaches to design ideas with plastics, such as those shown in Figure 12-13 [498]. The highest skill a designer can possess consists of making full use of the properties of materials to create truly distinctive products. In the process the designer needs to know and explore the limits of design. For example, limits are imposed by such factors as the manufacturing process's limits on the material that will determine the shapes to which the material can economically be converted, the physical properties of the materials that will limit their applications and useful environments, and the designer's imagination in combining form and function. In theory, the imagination is limitless, but in practice the first two limitations affect a designer's ability to exercise it (a real world exists where nothing is perfect; just direct one's effort toward perfection using an asymptotic approach). See Figures 1-1 to 1-3 and 1-17, which provide outlines on how to set up a new product idea or a design approach. In tum, a design must fit into a productivity pattern as shown in Table 12-4 and the product life cycles outlined in Figures 12-14 and 12-15 [1, 5, 710, 14, 15, 62-68, 105-25, 309, 450, 463, 636, 656, 722, 825, 830-940]. Most successful designers have the ability to develop products that will be instantly acceptable to a buyer. Their designs have.a recognizable, functional improvement along with some visual appeal to set their products apart from conventional ones. Too many new product designs or redesigns are nothing more than slight improvements that anybody could make with a minimum of thought. Many companies inch their way to progress with just such a slight change every now and then. This is the easy way to give the appearance of improving a product line with the least disruption of the manufacturing process and requires little adjustment by the sales staff, with the exception of printing new brochures. In order for a new design, redesign, design update, or model change to catch the interest of the design engineer who must use it, it should offer at least one clear-cut advantage over the model it supersedes. The process of inching ahead can afford the

~

....

----------..

..----------

----~-----

Figure 12-13. A design exercise to produce parts consolidation. High-volume car lines use glass-fiber-reinforced plastic and composite fender extensions, headlamp housings, rear finish panels, front grille panels, and others. There is a strong trend currently toward adding together adjacent components to make one unitary RP component. Cars thus use RP hoods, scoops, rear deck-lid spoilers, and other parts as distinguishing design features. Front-end parts consolidation. An upper-grille panel and a valance panel could be joined to make a one-piece front end. Further front-end parts consolidation. A two-piece hood in an SMC, or a wet system outer panel and a BMC or SMC inner support panel may give way to a one-piece, selfreinforced hood in SMC. It is possible to consider joining fenders and a one-piece hood into a single, rigid component. More front-end parts consolidation. The potential exists for a single joint-free front end assembly all molded as a single component. Rear-end parts consolidation. Deck lids take the same parts consolidation path as hoods. Here the rear finishing panels, valance panels, fender extensions, and taillamp housings (both hidden and with an exterior finish) are all united in one piece. Further rear-end parts consolidation. This is a single, jointfree molded component serving as a one-piece deck lid and rear-end panel with the addition of a quarter-panel section. The consolidation of large parts. The future may include one-piece molded chassis and even incorporate other parts. 904

--

/

Figure 12-13. (Continued) 905

A

Figure 12-13. (Continued) 906

jure 12·13. (Continued)

907

908 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

ISales volume

$

!

nme-+ +

Figure 12-14. An example of a product's life cycle.

Sales

Capacity Cost 01 sales Motivation costs Overhead

......

Order process Capacity Labor costs Eqpt.imatl. costs Overhead

Production -"'"

Capacity Labor costs Mati. costs Maintenance costs Motivation costs Qual. assur. costs Downtime history Downtime costs Eqpl. costs OSHA costs Overhead

--'"

.....

Shipping/ receiving

Capacity Labor costs Eqpt. costs Motivation costs . OSHA costs Inventory costs Inspection costs Overhead

Purchasing Capacity Labor costs Ordering costs Motivation costs Overhead

Figure 12-15. Another example of a product life-cycle chart.

writer of promotional copy a number of things to write about, but there should be one or preferably several technically sound, verifiable advantages to it. These should give the design engineer some reason to spend time learning about the product, trying it out, and perhaps eventually buying it (see Fig. 12-16). A new design or redesign should also incorporate some visual indication that there has been change. A new shape or form is great if you can get one, but a new color or label will do. When a buyer purchases something new, the product itself should look like it

CONCLUSIONS 909

New ....ice Is campared to ..tIeteddeviCfo

Does ... device have same

Indication statements?

_

No

Descriptive infarmatlon

,Red (In deciding. may canslder impad on safety and

New device has same Intended use and may be ·substantlally - - - - -..... equivalent"

as-...

deviceraquestecl

Does MW device have same technalagical characteristics _

No

No

c;:~:..": aHed safety

~--:~---.~

rNo

t

Ant the descriptive characteristics precise IIIOUF to ensure equlYlllance?

!

Yas

No Ant perfannllJD data avadable to oqulwalaace?

Perfarmance data required

L.

New ....ice has MW Intended use

_

Yes

I=-

Do the MW characteristics raise MW types of safety or

Do accepted sdentlftc methods exist lOr assessing eReds 01 the ... characteristics?

No

Yes

Ant performance data available --;;::--"1 to assess eRects of ... No I challlcterlstia?

+Yes

v,IS

Do perlonnance data _ _trate equivalence?

Do perfannance data de-mrate

.t

"Not substantially equivalent" determination

eHedlvenass)?

Yes

aIaout MW or _eted

~

Do the dillerances alter the intended therapeuticldlagnostldetc. Yes

equivalence?

Performance data required

----J

No

"Substantially equivalent" determination

Figure 12·16. An example showing the redesign of a medical device where a change in an FDA-approved original product is to be updated with a modification or change. This schematic shows the FDA's method for determining substantial equivalence. All these steps are needed to satisfy the FDA requirements for what is known as GMP (Good Manufacturing Practices). is new and different. Radical changes can be upsetting, but a big change is better than no change at all when dealing with the needs of the design community. Here is a checklist for designers looking for good things to say about their design:

1. Tell how it improves the intended function. What is its principal advantage, speed of operation, reliability, size, cost, material of construction, or other characteristics? Focus on the main user benefits. 2. Describe its efficiency of operation. If you can, document any operational improvement in terms of dollars, power, time, Or whatever. It gets the attention of other engineers. 3. Emphasize quality. Many design engineers will be interested in a quality product with longer life, a better appearance, quieter operation, improved durability, and greater convenience, as a component for use in their own quality products. This is particularly true now in office equipment and home appliances, where their environmental effect is a factor in sales. 4. Size, weight, portability, ease of handling, neatness of layout, fit, and compatibility are often crucial factors in design product selection. Visual appearance is always important to call attention to these advantages.

Table 12-4. Example of a Design Program Approach Detailed Requirements

Design Category

Subcategories

Establish functional and performance requirements

Estimated allowable size and shape

Product basic functions Aesthetics and marketing Shipping Available space Weight Standardization Strength and stiffness criteria Flexibility Process limitations

Establish structural requirements

Loads: Gravity Dead--Own weight superimposed Live--()ccupancy Snow Misc. Pressure Fluid

Earth Wind Dynamic Impact Seismic Handling and shipping Cyclic Temperature: . {Interior Service range- Exterior Gradient across component thickness No. of cycles--high to low No. of cycles--freeze-thaw Solar gain, surface air flow Liquid, moisture, and/or vapor tightness

Strength-weight ratios--relative significance Establish nonstructural requirements

Service environment: Corrosion {InteriOr resistance Exterior

Weathering Moisture wet--dry cycles { Freeze-thaw cycles UVexposure Rain abrasion Aging Moisture Temperature Fire safety Incombustibility Flame spread rate Toxic gases Fuel content Light Transparency transmission Translucency Opaque

Chemical { Soil Moisture Organic

{ Control of sunlight and solar heat Color (cant'd)

910

Table 12-4. (Continued) Design Category

Subcategories

Detailed Requirements Surface texture

{Aesthetics

Surface coatings

{Abrasion resistance

Thermal insulation Moisture and vapor penetration Electrical insulation

Preliminary design of component

{Barrier Gradient

{Condensation

{Dielectric properties

Establish cost targets

Examine economics for successful competition with similar products in conventional materials Consider total effect of new design on end product costs: materials, tooling, finishing, assembly, warehousing and inventory, quality control, packaging and shipping, and installation Consider effect on operating costs. Light weight is important in some applications

Establish production and marketing requirements

Number of identical pieces Minimum and maximum probable production rates Available plant Market locations Shipping costs Method of marketing Installation criteria, if applicable Cost restrictions imposed by competing products or technology. Prices can shift with short- and long-term changes in market conditions

Select size and general configuration

Consider end use and limitations of suitable plastics, efficient manufacturing processes, requirements for sufficient strength and stiffness with efficient use of materials, and cost

Select feasible plastics material or materials

Satisfy structural requirements with favorable cost ratios Satisfy nonstructural criteria with acceptable compromises and trade-offs where necessary Is efficient fabrication process available?

Select feasible manufacturing process or processes

Provide required size and configuration Tooling and plant capital costs to be appropriate for number of pieces and rate of production Compatible with available plant and marketing plan Provides required structural properties and quality control

Determine structural response based on approximate analysis

Develop suitably simplified concept of structural behavior to permit approximate determination of structural response-reactions, stress resultants, stability and stiffness requirements. Make appropriate assumptions within confines of laws of statics (cont d)

911

Table 12-4. (Continued) Design Category

Revise preliminary design of component

Develop final design of component

Subcategories

• Detailed Requirements

Establish design criteria for trial materials selected

Detennine suitable allowable design strengths and stiffness, taking into account type and duration of load, service environments, process effects, quality expectations, etc.

Proportion component for specific configuration and thickness

Detennine trial shape of plates, shells, and ribs, depth of ribs and sandwiches, and wall thicknesses to meet strength, deflection, and stability criteria Review economics and suitability

Develop significant details

Detennine concept and principal details for shop and field connections, penetrations, and other subparts (if required) Detennine materials for connections, coatings, subparts, etc.

Evaluate preliminary design

Review economics and suitability of materials and process based on preliminary proportions. Consider overall compatibility and practicality of all materials and parts in component as a system Does it meet functional and perfonnance requirements? Is it compatible with other components that may interact with it, relative to effects of expansion and contraction, structural support or movement, fire safety, etc.?

Review perfonnance and fun~tional requirements

Detennine if all original perfonnance requirements are feasible within economic objectives, or whether compromises and trade-offs should be considered

Optimize design to reduce cost or satisfy functional and perfonnance requirements

General configuration Configuration proportions such as rib depths, shell radii, fillet radii, etc. Material thickness Material alternatives-consider additives to tailor properties Process alternatives

Perfonn structural analysis of acceptable accuracy

Detennine structural response--stresses, support reactions, deflections, and stability-based on a structural analysis of acceptable accuracy. Determine acceptable accuracy based on economic value of component, consequences of failure, state-of-the-art capability in stress and stability analysis, margin of safety, knowledge about loads and materials properties, conservatism of loads, provisions for further evaluation by prototype testing

Establish final design criteria

Allowable stresses, strains, deflections Margins of safety against local and overall instability, vibrations, etc. Take into account type and duration of load, service environments, process effects, equality expectations

Evaluate proportions and design details; revise if necessary

Shape of plates, shells, ribs Depth of ribs and sandwiches Thickness of shells, flanges, and stiffeners Connections: Shop Field Edge conditions Penetrations Subparts, Inserts

Prepare engineering drawings

Drawings are sometimes prepared in two stages: Design drawings Detail or fabrication drawings (cont'd)

912

Table 12-4. (Continued) Design Category

Evaluate design by prototype and materials tests

Develop production and distribution system

Subcategories

Detailed Requirements

Prepare specifications for technical requirements of product and materials

Materials requirements including composition, quality standards, and minimum structural properties Fabrication requirements and standards, including dimensional tolerances, allowable defects, and minimum structural properties Requirements for prototype and quality control tests and procedures Shipping and handling Requirements for field assembly, installation, or erection

Prepare manuals or instructions for maintenance and repair

Periodic maintenance, recoating Service conditions: temperature limits, chemical exposure limits Repair procedures

Develop practical full-scale prototype for structural tests

Develop practical test program to demonstrate components ability to meet structural and performance criteria. Extent of such test program, if any, depends on economic value of "Component, number of units to be produced, consequences of failure, accuracy of structural analysis and design, margins of safety used in design, knowledge about service loads and environments, and difficulty of duplicating service loads and conditions in test

Test materials for structural properties and effect of service environment

Determine that materials produced in actual fabrication process will have the minimum structural properties and resistance to service environment assumed in the design. Extent of testing, if any, depends on available information about specific materials and processes to be used

Revise design, if required

Correct design and detail problems, if any, revealed in tests Modify materials, or process, if production materials' properties not adequate Protect or modify materials if service environment causes excessive degradation of properties

Pattern design and drawings Mold design and drawings

Take into account shape limitations and design rules that facilitate molding

Production process design and layouts

Take into account materials and configuration characteristics that simplify processes Automated processes are needed for high-volume production

Develop any special equipment Distribution and marketing plan

Production for inventory, or by special order Replacement part inventory Locate production facilities to optimize distribution

Procedures for packaging, storing, handling, and shipping

Identify special requirements for protection in handling and shipment

Installation requirements

Specify special requirements for assembly or installation

913

914 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

5. A designer's headache is the product that cannot be maintained, repaired, replaced, or renewed once it is incorporated in a design assembly. Most of us have been amazed to find such items and respond favorably to component designs that can be fixed. 6. Safety considerations are a positive factor when incorporated in a design component rather than being left as a final assembly project or add-on. Safety should be a prime consideration whenever there are exposed moving parts or pressure, hazardous substances, or unknown conditions that could result in injury during assembly or subsequent product use. Change-more specifically the rate of change-is the key to nearly all design activity. Change creates the need for new and better products and provides designers with the technical tools for making better products. There is little excuse for a lack of technical improvement in engineered products. There is also little excuse for not making products attractive, easy to use, and easy to maintain to benefit the buyer.

CHALLENGE REQUIRES CREATIVITY In order to find unique, creative solutions to difficult challenges that were not resolved by past tried and true techniques, one must get away from the conventional state of mind that is often unimaginative, frustrating, repetitive, and negative. The nature of some problems tends to invite unimaginative suggestions and attempts only to use past approaches. Problem solving in designing and producing products, as in business and personal problems, generally requires taking a systematic approach. If practical, make rather small changes and allot time to monitor the reaction of result. With whatever time is available, patience and persistence are required. However, when a problem is particularly difficult or only limited time exists, considering a new and imaginative approach with techniques that previously generated creative ideas. First generate as many ideas as possible that may be even remotely related to the problem. During the idea generating phase it is of critical importance to be totally positive: no ideas are bad. Evaluation comes later, so do not attempt to provide creativity and evaluation at the same time; it could be damaging to your creativity. Look for quantity of ideas, not quality, at this point. Now all ideas are good; the best will become obvious later. If possible, relate the problem to another situation and look for a similar solution. This approach can stimulate creative thinking toward other ideas. Try humor; do not be afraid to joke about a problem. The next step is to evaluate all the ideas. Consider categorizing the list, then add new thoughts, select the best, and try them. After all this action, if nothing satisfactory occurs, rather than give up look for that really creative solution-it is out there. You may be too close to the problem. Get away from the trees and look at the forest. Climb up one of the trees and look at things from a different perspective. Use your creative talents but, again, be positive. You have now creatively worked through the frustration and negativism that problems seem to generate. Your increasingly creative input will generate future opportunities. Now let us take the thoughts above and improve on them. In doing so let us avoid saying in effect "My mind is made up--do not give me the facts." Rather, let us use the approach that there is always room for improvement.

CONCLUSIONS 915

Designers spend little time in a vacuum. Most of their working hours are spent adapting, modifying, or otherwise improving on work previously performed by others. However, good ideas are as much in demand in these areas as any other.

THE FUTURE It can be said that the challenge of design is to make existing products obsolete or at least offer significant improvements. Despite this level of activity there are always new fields of industry to explore (see Fig. 12-17). Plastics will continue to change the shape of business rapidly. Today's plastics tend to do more but cost less, which is why in many cases they came into the picture in the first place. Tomorrow's requirements will be still more demanding, but with sound design plastics will satisfy those demands, resulting not only in new processes and materials but improvements in existing processing and materials. Research will no doubt become even more adept at manipulating molecules to the extent that the range of materials offered to industry will continue to present new opportunities and allow existing businesses to enjoy profitable growth (see Chapters 2, 6, and 7). A reading of the literature and patents being issued indicates that there is a great deal of commercially oriented research being aimed at further improvement and modification to the plastics family. Unfortunately, sometimes a new design concept is not accepted or may simply be ahead of its time. In 1483 Leonardo da Vinci designed what he called a spiral screw

Figure 12-17. An example of exploring new fields is that of fabricating film directly on farmland. Here two extruders are producing film that can in tum incorporate seeds, fertilizers, and so on so that they are applied uniformly or at a controlled rate. This type of system as well as those for extruding other products (pipe, rails, etc.) have all been in development for decades, but to date no known commercial applications exist.

916 DESIGNING WITH PLASTICS AND COMPOSITES: A HANDBOOK

flying machine. In 1942, Igor Sikorsky developed the R4B helicopter. Taking 459 years to bring a product to market seems a failure either in materials or perhaps the interoffice mail.

u

..0 :::> u

c

Q ~

Year

Figure 12·18. World consumption of plastics by volume. Steel

Plastics

I,OOO,Ooo~--+----+----t----+---t::*,,""--::~'---:1I----::lr-1

Aluminum

100,000 Rubber

::

Copper

0 I-

8o.

-

Zinc

10,000

1940

1950

1970

1960 Year

Figure 12·19. World consumption of plastics by weight.

1980

1990

2000

CONCLUSIONS 917

Plastics are now among the nation's and world's most widely used materials, having surpassed steel on a volume basis in 1983 (see Fig. 12-18). By the end of this century, plastics will have surpassed steel even on a weight basis (see Fig. 12-19). Plastics' materials and products cover the entire spectrum of the world's economy, so that its fortunes are not tied to any particular business segment. Designers are in a good position to benefit in a wide variety of markets: packaging, building and construction, electronics and electrical, furniture, apparel, appliances, agriculture, housewares, luggage, transportation, medicine and health care, recreation, and so on. The effective exploitation of product design opportunities is the key to success. In tum, success hinges on other factors, such as the proper selection of materials and using the best available processing equipment. Because new materials and equipment continue to be more productive and produce better quality products, one should stay abreast of new material and equipment developments and evaluate them logically. With designing, there is an extremely vast area for improving profitability by ensuring that the best available material and equipment are used to meet specific design performance requirements. To design, as to live, is to change, and to aim for perfection is to have changed often.

Appendix A

GENERAL INFORMATION SOURCES

SELECTED TRADE MAGAZINES AND PUBLICATIONS Advanced Materials & Processes. ASM International, Metals Park, OH 44073. Canadian Plastics. 1450 Don Mills Rd., Don Mills, Ont., Canada M3B 2X7. Chemical & Engineering News. 733 Third Ave., New York, NY 10017. Chemical Engineering. 1221 Avenue of the Americas, New York, NY 10020. Chemical Marketing Reporter. 100 Church St., New York, NY 10007. Chemical Week. 1221 Avenue of the Americas, New York, NY 10020. Design News. 275 Washington St., Newton, MA 02158. European Plastics News. 33-39 Bowling Green Lane, London ECIR ONE, England. European Polymer Journal. Fairwell Pk., Elmsford, NY 10523. Journal of Commerce. 110 Wall St., New York, NY 10005. KunststofJe. C. Hanser Verlag, Munich, Postfach 8604 20,8000 Munchen 86, Germany. Machine Design. 1100 Superior Ave., Cleveland, OH 44114. Materials Engineering. 600 Summer St., Stamford, CT 06904. Medical Device & Diagnostic Industry. 2416 Wilshire Blvd., Santa Monica, CA 90403. Modern Packaging. 205 E. 42 St., New York, NY 10017. Modern Plastics. 1221 Avenue of the Americas, New York, NY 10020. Paper. Film & Foil Convertor. 29 Wacker Dr .• Chicago. IL 60606. Plastics Compounding. 1129 E. 17 St., Denver, CO 80218. Plastics Design & Processing. 700 Peterson Rd., Libertyville, IL 60048. Plastics Design Forum. 1129 E. 17 St., Denver, CO 80218. Plastics Engineering . .14 Fairfield Dr., Brookfield Center, CT 06805. Plastics Focus. Box 814, Amherst, MA 01004. Plastics Machinery & Equipment. 1129 E. 17 St., Denver, CO 80218. Plastics Technology. 633 Third Ave., New York, NY 10017. Plastics World. 275 Washington St., Newton, MA 02158. Rubber & Plastics News. 1 Cascade Plaza, Suite 1302, Akron, OH 44308. SAMPE Journal. 843 W. Glentana, Covino, CA 91722. World Plastics & Rubber Technology. 4-7 Nottingham Ct., Short's Gardens, London WC2H 9AY England.

SELECTED TRADE ASSOCIATIONS AND PROFESSIONAL GROUPS Adhesives Manufacturers Assoc. 111 E. Wacker Dr., Chicago, IL 60601. American Chemical Society. 1155 16 St., NW, Washington, DC 20036. 919

920 APPENDIX A

American National Standards Institute. 1430 Broadway, New York, NY 10018. American Society for Testing and Materials. 1916 Race St., Philadelphia, PA 19103. American Society of Civil Engineers. 345 E. 47 St., New York, NY 10017. American Society of Mechanical Engineers. 345 E. 47 St., New York, NY 10017. American Society of Metals. Metals Park, OH 44073. Canadian Plastic Institute. 1262 Don Mills Rd., Suite 48, Don Mills, Ont. Canada M3B 2W7. Drug, Chemical & Allied Trade Assoc. 40-42 Bell Blvd., Suite 204, Bayside, NY 11361. Factory Mutual Research Corp. 1151 Boston-Providence Tpke., Norwood, MA 02062. Industrial Designers Society of America. 1142 E. Walker Rd., Great Falls, VA 22066. Malaysian Rubber Producers' Research Assoc. Malaysian Rubber Bureau, Washington, DC. Manufacturing Chemists Assoc. 1852 Connecticut Ave., NW, Washington, DC, 20009. National Assoc. of Corrosion Engineers. Box 1499, Houston, TX 77001. National Assoc. of Plastics Fabricators. 1100 Standard Building, Cleveland, OH 44113. Polyurethane Manufacturers Assoc. 4219 S. Wolcott Ave., Chicago, IL 60609. Rubber Manufacturers Assoc. 1901 Pennsylvania Ave., NW, Washington, DC 20006. Society of Automotive Engineers. 400 Commonwealth Dr., Warrendale, PA 15096. Society of Plastics Engineers. 14 Fairfield Dr., Brookfield, CT 06805. Society of Plastics Industry. 1025 Connecticut Ave., NW, Washington, DC 20036.

SELECTED GOVERNMENT PUBLICATIONS AND GROUPS National Institute of Standards & Technology (NIST). Polymer Building, Washington, DC 20234. Naval Publications & Forms Center. 5801 Tabor Rd., Philadelphia, PA 19120. Plastics Technical Evaluation Center (PLASTEC). Picatinny Arsenal, Dover, NJ 07801. U.S. Government Publications. Available through the Dept. of Commerce, 5285 Port Royal Rd., Springfield, VA 22161 (see Note below). Annual Survey of Manufacturers. Bureau of the Census. Census of Manufacturers. Bureau of the Census. Directory of Plastics-Knowledgeable Government Personnel. NIST. Foreign Trade Reports. Bureau of the Census. Standard Industrial Classification Manual. Office of Management & Budget. Statistical Abstracts of the U.S. Dept. of Commerce. Synthetic Organic Chemicals. International Trade Commission. US Industrial Outlook. Dept. of Commerce. Wholesale Price Index. Bureau of Labor Statistics.

Note: Some of the above publications are not available through the Springfield office, but they can guide you as titles and their availability change. Many more publications exist as listed in their different indexes.

Appendix B

CONVERSIONS

A. METRIC CONVERSIONS· To Convert (U.S. System)

To (Metric System)

Multiply by

To Convert (Metric System)

To (U.S. System)

Multiply by

Density Ib.lin. J Ib.lft.J Ib.lft. J Ib.lin. J

kg/mJ glcmJ kglmJ g/cmJ

27.680 0.0160 16.0185 27.68

kglmJ glcmJ kglmJ glcmJ

Ib.lin. J Ib.lft.J Ib.lft.J Ib.lin. J

0.000036 62.43 0.0624 0.036.13

Temperature in.l(in .. "F) "F

m1(m·°C) °C

1.8

m1(m· OC)

("F - 32)1

°C

in.l(in .. "F) "F

0.556 1.8°C + 32

"F

K

("F

K

"F

1.8K - 459.67

kPa MPa GPa bar

psi psi psi psi

0.145 145 145038 14.51

ft .. Ibf. in. ·Ibf. ft .. Ibf.!in. ft .. Ibf.!in. ft .. Ibf.!in.2 kW

0.7376 8.850 0.0187 1.87 0.4755 0.7355

(1.8)

+ 459.67)1 (1.8) Pressure

psi psi psi psi

kPa MPa GPa bar

6.8948 0.00689 0.00000689 0.0689

Energy and Power ft .. Ibf. in. ·Ibf. ft .. Ibf.!in. ft .. Ibf.!in. ft .. Ibf.!in. 2 kW U. S. horsepower

Jim J/cm kJ/m2

metric horsepower kW

1.3558 0.113 53.4 0.534 2.103 1.3596 0.7457

Jim J/cm kJ/m2

metric horsepower kW U.S.

1.3419 horsepower

Btu

1055.1

Btu

0.00095 (cont'd)

921

A. METRIC CONVERSIONS (Continued) To Convert (U.S. System)

To (Metric System)

Multiply by

To Convert

To (U.S.

(Metric

System)

Multiply by

System)

3.412 6.933

Btu Btu . in.! (h .. ft.' . oF)

W·h W/(m· K)

0.2931 0.1442

W·h W/(m' K)

Btu "lib. Btu "/(lb .. oF)

kJ/kg J/(kg . 0c)

2.326 4187

kJ/kg J/(kg . 0c)

V/mil

MV/m

0.0394

MV/m

V/mil

25.4

Ib.!min. Ib.lh.

glS

7.560 0.4536

gls

Ib.!min.

kglh

Ib.lh.

0.1323 2.2046

Btu Btu· in.! (h ..

ft." ~ Btu "!lb. Btu "/(lb.

0.4299 0.000239

oF)

Output

kglh

Velocity in.!min.

cmls

ft.!s.

mls

0.0423 0.3048

cmls

in.!min.

mls

ft.!s.

Pa's

poise Ib/ft . s

23.6220 3.2808

Viscosity poise Ib/ft . s ft'ls

Pa . s Pa's m'ls

0.1 1.4882 0.0929

millimeter millimeter centimeter centimeter meter meter

0.0254 25.4 2.54 30.48 0.3048 0.9144

Pa's m'ls

ft'ls

10 0.6720 10.764

millimeter millimeter centimeter centimeter meter meter

mil inch inch foot foot yard

39.37 0.0394 0.3937 0.0328 3.2808 1.0936

0.0016 0.155 0.0011 10.7639 1.1960

Length mil inch inch foot foot yard

Area inch'

miIlimeter'

inch'

centimeter'

foot'

centimeter'

foot'

meter'

yard'

meter'

645.16 6.4516 929.03 0.0929 0.8361

millimeter'

inch'

centimeter'

inch'

centimeter'

foot'

meter'

foot'

meter'

yard'

Volume, Capacity inch'

centimeter3

fluid ounce quart (liquid)

centimeter3

gallon (U.S.)

decimeter3 (liter) decimeter3 (liter)

16.3871 29.5735 3 0.9464 3.7854

centimeter3

inch 3

centimeter3

fluid ounce

decimeter3 (liter)

quart (liquid)

decimeter3 (liter)

gallon (U.S.)

0.061 0.0338 1.0567 0.2642

(cont'd)

922

A. METRIC CONVERSIONS (Continued) To Convert (U.S. System)

To (Metric System)

Multiply by

To Convert (Metric

To (U.S. System)

Multiply by

System) gallon (U.S.)

meter3

0.0038

meter'

gallon (U.S.)

264.17

foot'

decimeter' meter' meter3 m3/kg m3/kg

28.3169 0.0283 0.7646

decimeter' meter'

foot" foot 3 yard3 in. 3/lb.

0.0353 35.3147

foot' yard 3 in. 3/lb. ft.3/lb.

meter' m3/kg m4/kg

0.000036 0.0624

ft.3/lb.

1.3079 27.680 16.018

ounce

0.03527

Mass ounce (avdp.)

28.3495

gram

gram

(avdp.) pound

gram

453.5924

gram

pound

0.0022

pound pound U.S. ton (short)

kilogram metric ton metric ton

0.4536 0.00045 0.9072

kilogram metric ton metric ton

pound pound U.S. ton (short)

2.2046 2204.6 l.l023

N

lbf. dyne ton-f (2000

0.225 lOx 10' II X 10-5

Force lbf. dyne ton - f(2ooo Ibf)

4.448 I x 10-'

N N N

8.9 x 103

N N

lbf)

Metric Prefixes* Standard Metric Symbols

.... . ampere ...... . bar .... candelar C .... . celsiust g ....... . gram h ....... . hour A bar cd

Hz ....... hertz J ........ joule K ...... kelvin

kg .... kilogram L ......... liter m . ..... meter N ..... newton Pa ...... pascal S ..... siemens s ...... second t .... metric ton V . ....... volt W ....... watt

Numerical Value

Term

Symbol

lO lOz lO3 lO6 lO9 lO12 lO-1 lO-2 lO-3

deka hecto kilo mega giga tera deci centi milli micro nano pico

da h k M G T d c m

1~

lO-9 lO-12

J.I.

n p

*These prefixes may be used with all metric units. tForrnerly c.lled Centrigr.de.

923

924 APPENDIX B

B. SI UNIT PREFIXES Multiplication Factor

Prefix

I 000 000 000 000 000 000 = 10" I 000 000 000 000 000 = 10" I 000 000 000 000 = 10 12 I 000 000 000 = 10" I 000 000 = 10" 1000 = 103 100 = J02 10=10 0.1 0.01 0.001 0.000 001 0.000 000 001

= = = = =

Symbol

exa

E

asinT~

peta

P

asin~

T

as in terrace jig' a
tera giga mega kilo hecto deb

G

M k h da

10-1 10-2 10-3 10-" 10-"

deci centi milli micro nano

0.000 000 000 001 = 10-12 0.000 000 000 000 00 I = 10-"

pico femto

f

atto

a

0.000 000 000 000 000 00 I = 10-"

Pronunciation (U.S.)"

d m

.... n

P

deck' a


as in decimal as in sentiment as in military as in microphone nan' oh <.!!!!!! as in Nancy) peek' oh fern' toe @!!! as in feminine) as in .anatomy

Tenn (U.S.) one one one one one one one ten

quintillion quadrillion trillion billion million thousand hundred

one one one one one

tenth hundredth thousandth millionth billionth

one trillionth one quadrillionth one quintillionth

Abstracted from material by American National Metric Council, 1977. "The first syllable of every prefix is accented to ensure that the prefix will retain its identity.

C. THE PROPERTIES OF WATER One gallon of water weighs 8.3356 lbs. at 17°e (62°F). (Air free, weighed in vacuum.) One gallon of water contains 231 cubic inches or 0.13368 cubic feet. One cubic foot of water equals 7.4805 gallons. A cylinder 7" in diameter and 6" high contains one gallon. One cubic inch of water weighs .576 oz. The maximum density of water is at 39.1 OF. The freezing point at sea level is ooe (32°F). One cubic foot of water at 39.l oF equals .4335 lbs. per sq. inch. The boiling point of water at normal pressure is 1000e (212°F).

Appendix C

TRADE NAMES

There are thousands of trade names used in the plastics industry to include and identify its many different plastics materials. These listings are available in the publications listed above in Appendix A, as for instance Modern Plastics, Plastics Technology, and Plastics World. What follows gives examples of some of the more commonly encountered trade names. ACETRON. Filled and reinforced acetals. Polymer Corp. ACLON. CTFE. Allied Signal, Inc. ACRYLACON. Glass-concentrated SAN. Wilson-Fiberfil Internat., Akzo. ADIPRENE. Polyurethane resin. Du Pont. A-FAx. Amorphous PP. Himont U.S.A., Inc. ALATHON. Polyethylene resin. Du Pont. ALFfALAT. Alkyd resin. Hoechst Celanese Corp. AMOPLEX. Glass-compatible PP. Amoco Chemicals Co. ARAKOTE. Polyester resin. Ciba-Geigy Corp. ARALDITE. Epoxy resin. Ciba-Geigy Corp. ARCEL. Expandable PE copolymers. Arco Chemical CO. ARDEL. Polyarylate. Amoco Performance Products, Inc. ARETON. Polyaryletherketone. Du Pont. ARPAK. Expanded PE beads. Arco Chemical Co. ARYLON. Polyarylate resin. Du Pont. ATLAC. Polyester resin. ICI Americas, Inc. ATTANE. Ultra LDPE resins. Dow Chemical Co. AzoEL. Stampable TP-RP sheets. GE Plastics. BAYBLEND. PC/ABS blends. Mobay Corp. BA YDUR. Rigid structural urethane foam. Mobay Corp. BUDD. Urea molding compound. BIP Chemicals, Inc. BUDENE. Polybutadiene resin. Goodyear Tire & Rubber Co. BUTACITE. Polyvinyl butyral sheeting. Du Pont. BUTV AR. PVB resin. Monsanto Co. CADON. Styrene anhydride terpolymer. Monsanto Co. CALIBRE. PC resin. Dow Chemical Co. 925

926 APPENDIX C

CELDZOLE. Polybenzimidazole resin. Hoechst Celanese Corp. CELCON. Acetal copolymer resins. Hoechst Celanese Corp. CELLIDOR. CAB. Mobay Corp. CELSTRAN. Long-fiber-reinforced pellets. Hoechst Celanese Corp. CLEARTUF. PET resin. Goodyear Tire & Rubber Co. CLYSTAR. Shrink film. Du Pont. COMBOMAT. Woven roving mat. PPG Industries, Inc. CYCOLAC. ABS resin. GE Plastics. CYCOLOY. PC/ABS alloys. GE Plastics. CYCOM. Advanced composites. American Cyanamid Co. CYRLITE. XT Polymer. Cyro Industries. DACRON. PET Fiber. Du Pont. DELRIN. Acetal resins. Du Pont. DERAKANE. Vinyl ester resins. Dow Chemical Co. DURADENE. SB copolymers. Firestone. DUREL. Polyarylate resin. Hoechst Celanese Corp. DUROFLEX. PP resin. Shell Chemical Co. DYLARK. PS resin. Arco Chemical Co. ECDEL. Copolyester (TPE). Eastman Chemical Co. EKTAR. PP, TP polyester resins. Eastman Chemical Co. ELASTOFLEX. Flexible PUR systems. BASF Corp. ELASTOPOR. LD rigid PUR systems. BASF Corp. ELYACITE. Acrylic resin. Du Pont. ELYAX. EVA resins. Du Pont. ESTANE. PUR resins and compounds. B. F. Goodrich Co. FIBERLOC.

Glass-fiber-reinforced vinyl compounds. B.

F.

Goodrich Co.

FLUON. TFE resins. ICI Americas, Inc. FLUOREL. Fluorocarbon elastomers. 3M. FLUOROCOMP. Reinforced fluorocarbons (glass, etc.). ICIILNP. FLUOROMELT. Melt-processable fluoropolymers. ICIILNP. FORMYAR. Polyvinyl formal resins. Monsanto Co. FORTIFIED POLYMERS. Glass-reinforced TP compounds. ICIILNP. FORTILENE. PP resin. Soltex Polymer Corp. FORTRON. PPS resins. Hoechst Celanese Corp. GECET. PPO foam beads. GE Plastics. GELOY. Weatherable ASA terpolymer resins. GE Plastics. GEMAX. TP alloy. GE Plastics. GENOTHERM. PVC film and sheet. Hoechst Celanese Corp. GEOLAST. TP elastomers. Monsanto Co. HOSTATEC. Polyaryletherketone. Hoechst Celanese Corp. HYPOLON. Chlorosulfonated PE (TPE). Du Pont. HYTREL. Polyester TSE resins. Du Pont. IMPET. TP polyester. Hoechst Celanese Corp.

APPENDIX C 927 KADEL. Polyketone resins. Amoco Performance Products, Inc. KALREZ. Perftuoroelastomer (TSE). Du Pont. KAPrON. Polyimide film. Du Pont. KELVAR. Aramid fiber. Du Pont. KODAPAK. PET resin. Eastman Chemical Products, Inc. KODAR. Copolyester TP resins. Eastman Chemical Products, Inc. KRATON. Styrene block copolymer elastomers. Shell Chemical Co. KYNAR FLEX. PVDF resin. Pennwalt Corp. LEXAN. PC resin. GE Plastics. LoMOD. TP copolyer elastomer. GE Plastics. LOTRENE. LLDPE resins. CdF Chimie North America, Inc. LUBRICOMB. Lubricated TP composites. ICIILNP. LUCITE. Acrylic resin. Du Pont. LUSTRAN. SAN and ABS resins. Monsanto Co. MAGNACOMP. Magnetized injection-molding compounds. ICIILNP. MAGNUM. ABS resin. Dow Chemical Co. MAKROLON. PC resins. Mobay Corp. MARLEX. PE and PP resins. Phillips 66 Co. MELINEX. Polyester film. ICI Chemicals. MINDEL. Engineering resins, polysulfones, etc. Amoco Performance Products, Inc. MOVIOL. Polyvinyl alcohol resin. Hoechst Celanese Corp. MYLAR. TP polyester film. Du Pont. NEOPRENE. Chloroprene elastomers. Du Pont. NORPROP. Biaxially oriented PP film. Quantum Chemical Corp. NORYL. Phenylene oxide-based resins. GE Plastics. NY-KoN. Molybdenum disulfide-lubricated nylon. ICIILNP. OPPANOL. Polyisobutylene resin. BASF Corp. PELASPAN-PAC. Expanded PS resin. Dow Chemical Co. PELLETHANE. PUR elastomer. Dow Chemical Co. PLEXIGLAS. Acrylic sheet and molding compounds. Rohm & Haas Co. PLIOVIC. PVC resins. Goodyear Tire & Rubber Co. POCAN. PBT resins. Mobay Corp. PORON. Microporous plastics. Rogers Corp. PREvEX. Polyphenylene resin. GE Plastics. RADEL. Polyarylsulfone resin. Amoco Performance Products, Inc. RESIMENE. Melamine resins. Monsanto Co. RT/DuROID. Reinforced TFE plastics. Rogers Corp. RYNITE. PET resin. Du Pont. RYTON. PPS resin. Phillips 66 Co. SANOPRENE. TP elastomer. Monsanto Co. SARAN. PVDC resin. Dow Chemical Co. SCLAIR. Linear PE resin. Du Pont.

928 APPENDIX C

SCOTCHPLY. Reinforced plastics. 3M. SELAR. TP barrier resins. Du Pont. SUPEC. PPS resin. GE Plastics. SURLYN. lonomer resin. Du Pont. TEDUR. PPS resin. Mobay Corp. TEFLON. FEP, TFE, and PFA fluorocarbon resin. Du Pont. TEMPRITE. Chlorinated PVC. BF Goodrich. TERLURAN. ABS copolymers. BASF Corp. TEXIN. TP PUR resin. Mobay Corp. THORNEL. Carbon tiber. Amoco Corp. TORNAC. Hydrogenated nitrile elastomers. Polysar, Inc. TPX. Polymethylpentene resin. Mitsui & Co. U.S.A. TYRIL. SAN resin. Dow Chemical Co. TYRIN. Chlorinated PE resin. Dow Chemical Co. UDEL. Poly sulfone resin. Amoco Performance Products, Inc. ULTEM. Polyetberimide resins. GE Plastics. ULTRADUR. PBT resins. BASF Corp. ULTRAMID. Nylon resin. BASF Corp. ULTRAPEK. Polyaryletherketone. BASF. UPILEX. Polyimide resin. ICI Corp. VALOX. TP polyester resin. GE Plastics. VECTRA. Liquid crystal polymer. Hoechst Celanese Corp. VIBRATHANE. PUR elastomer resin. UniRoyal, Inc. VICTREX. PEEK, PSU, PES resins. ICIILNP. VINEX. Polyvinyl alcohol resins. Air Products & Chemicals, Inc. VITON. Fluorocarbon elastomers. Du Pont. XCAN. TS polyester resins. Amoco Performance Products, Inc. XENOY. TP alloy. GE Plastics. XYDAR. Liquid crystal polymer. Amoco Performance Products, Inc. ZYLAR. Clear impact terpolymer resins. Polysar, Inc. ZYTEL. Nylon resin. Du Pont. ZYTEL ST. Supertouch nylon resin. Du Pont.

Appendix 0

COMPUTERIZED SOFTWARE AND DATABASES

Software and databases expand design capabilities, simplify design analyses, speed up evaluations like stress analyses, provide more-complete analyses on materials, predict shrinkage and warpage and are otherwise valuable. For example, materials databases have been developed to optimize material selection for designers confronted with the large number of plastics available today. As covered in the "Selection" section of Chapter 6, optimizing material selection by reading data sheets has become impractical for many designers. Today, instead, databases are accessible over phone lines, on tape, floppy disc, and on databases to serve the printed page. They come from manufacturers and consortiums promoting the use of their own products, publishers compiling universal information, services that focus on specific performance areas, and from databases that exist solely to organize other properties on databases. All these types of useful information are now readily available. Examples are listed below of software, including databases that are likely to prove useful to the product designer, with abstracts of their capabilities. They are not complete and can become obsolete in a short period of time. It is therefore important to maintain contact with these sources and the many others that exist, to keep up-to-date in this fastmoving area of technology [519-36, 751-806].

DESIGN PRODUCTS CALS (Computer Aided Acquisition & Logistics Support. U.S. Standards Development Committee, comprised of industry and U.S. Dept. of Defense sources. A fast way to revise and update drawings in line with the way CAD drawings are handed today. CAM Station (Calma, Div. of Prime Computer, Inc., San Diego, CA). A systematic, economical way to manufacture complex three-dimensional parts without compromising function, quality, or performance. Features a full-function engineering workstation, powerful 2-D and 3-D modeling, advanced surface creation, industry-proven 3-D milling routines, full 2!-axis mill-drill functions, a postprocessor generator, tool-design capabilities in 2-D and 3-D, and a turnkey system. CARDD (Computer-Assisted Research, Design & Documentation System. Gates Energy Products, Inc., EI Paso, TX). In-house data-management software allowing designers to piece together quickly existing designs to form custom jobs. Provides measurably 929

930 APPENDIX D

improved turnaround times for new-product designs (prior fast design cycles required twenty-eight to thirty days; now available in as little as three min.). Most manufacturing firms rely on CAD systems to improve the speed and quality of drafting, but fully automating the design process requires having more than just computerized blueprints. By successfully integrating design automation with innovative design storage and retrieval systems, designers can help their companies become major players in today's quickly changing markets. Designview (Premise, Inc., Cambridge, MA). Unifies geometry and mathematics with dimensional-driven variation geometry to help visualize and analyze the behavior of mechanical systems and quickly evaluate alternatives. Pro/Engineer (Advanced Project Engineering, Haworth, Inc., Holland, MI). Attains accurate design data on a 3-D solid model. With part modifications the complete model will be accurately modified. Smart Model (lead, Inc., Cambridge, MA). Takes traditional CAD programs based on interactive geometric modeling systems a step further through the inputting of design rules, including information about tooling tolerances, lot quantity, processing parameters, and production deadlines. If any input parameters are subsequently changed, the system can automatically generate a new design, reevaluating all rules and part dependencies specified in the system. GRAPHICS

AutoCAD (Autodesk, Inc., Sausalito, CA). CADD-23 (Brodhead Garrett Co., Cleveland, OH). CADplan (Personal CAD Systems, Inc., Los Gatos, CA). MEGA CADD (Mega CADD, Inc., Seattle, WA). Personal Designer (ComputervisionlPrime Computer, Inc., Bedford, MA). VersaCAD (T&W Systems, Huntington Beach, CA). MECHANICAL ENGINEERING

ANALYTIX (Saltire Software, Beaverton, OR). This constructive variational geometry design system for mechanical engineering analysis provides kinematics, statics, dynamics, and tolerance analysis, integrated without requiring the user to enter formulas. Its algorithms provide fast response time even as drawings become more complex. Its geometry is entered by dimensional sketches, and feedback is given on whether a drawing is under, over, or consistently dimensioned. This software solves kinematics and statics problems analytically rather than numerically. The fully dimensioned drawings it provides can be made into a mechanism by simply setting one or more dimensions in motion while the others remain constant. Velocities and accelerations can be given to any dimensions on the drawing. Tolerances can be assigned to any dimension of the figure, and any distance or angle measured from the figure is given in both its true value and its tolerance range. COSMOS/M (Structural Research & Analysis Corp., Santa Monica, CA) Provides finite element analysis with capabilities in interfaces that are for structural (static, dynamic, etc.), nonlinear (plasticity, friction, damping, etc.), materials (isotropic, orthotropic, etc.), elements (3-D beams, sandwich shells, springs, etc.), heat transfer (steady state, convection, etc.), design optimization (miniweight, stress constraints, etc.), and CAD interfaces and others. DADS (Dynamic Analysis Design Systems, CADSI, Oakdale, IA). Performs nonlinear

APPENDIX D 931

large-displacement transient analysis and simulation. Allows for the modeling of realworld behavior of complex systems. DRAFT-PAK (Baystate Technologies, Worcester, MA). Allows CADD users to design instantly with high-level features, fasteners, mechanical elements, detailing symbols and more, by using powerful parametric programs. EUCLID-IS (Matra Datavision, Tewkesbury, MA). A CAD/CAM/CAE design system to help construct fast representations of complex assemblies. Its extensive array of tools and flexible users interface boost productivity. ME Workbench (Iconnex, Pittsburgh, PA). Analyzes designs in tolerances, linkages, and kinematics, component development or other such applications and provides practical, integrated design solutions to engineering problems. Its parametric modeling capabilities refine design concepts quickly and easily by analyzing stresses, forces, accelerations, and velocities. Transfers designs to CAD systems. NASTRAN (NASA, Washington, DC). Structural analysis via FEA. SAFE (Gulf Computer, INc., and Gulf General Atomic, Inc., TX). Structural analysis via FEA. SAP (University of California, Berkeley, CA). Structural analysis via FEA. STRUDL (MIT, Cambridge, MA). Structural design language with FEA and integrated civil-engineering systems.

MATERIALS AMDBS (Advanced Materials Data Base System; PDA Engineering, Inc., Costa Mesa,

CA). Focuses on plastics/composites, including glass, graphite, and aramid fibers with both TSs and TPs. Includes data from MIL-HDBK-5E and MIL-HDBK-17A, as well as AFWAL properties for advanced composites. Neat resins and single-strand fiber data also included. CARD (Computer-Aided Barrier Design; EVAL Co. of America, Lisle, IL). Reduces much of the development time and cost in designing multilayered packaging that must have barrier properties. Designer can use it to select barrier structures required to pass requirements like shelf life. Mathematical models predict the performance of constituent materials through the full range of performance requirements and processing, sealing, storage, and retail-sale conditions. CAPS (Computer-Aided Polymer Selection, Polydata Ltd., Dublin, Ireland). Provides selection from a complete overview of over 5,000 TPs from more than 65 important producers; 100 single values, stored for each grade, are continually revised, and users can be updated by diskette or modem. Based on a cursor-driven menu structure and user friendly. Low computer memory is required. All data accessible within seconds. User can enter own data, change or delete values, etc. CENBASE (InfoDexiCENCAD, Garden Grove, CA). Selected material based on technical requirements and cost data from over 12,000 grades of TPs, TSs, TPEs, RPs, and metals from more than 170 material suppliers worldwide. Each grade has 36 physical properties, including processability. EMA (Engineered Materials Abstract, ASM International, Metals Park, OH). Extensive coverage of the international literature on all aspects of engineered materials, including plastics and composites. Access to more than 200 other databases. EPOS (Engineering Properties on Screen, ICIILNP, Malvern, PA). Selection of over 600 LNP composites, each with at least 50 properties, including cost by weight and volume, friction and wear, chemical resistance, etc.

932 APPENDIX D

IPS (International Plastics Selector, D.A.T.A. Business Publications, San Diego, CA). Plastics material selection database reviews of more than 10,000 grades with up to 50 properties for each material. Includes physical, mechanical, electrical, flammability, and other characteristics. MAT.DB (Materials Property Data Base, ASM International, Metals Park, OH). Commercially updated available materials properties database developed for users interested in building their own database as well as for those who simply want access to precompiled databases. Provides flexible format for structuring diverse data types into one unified systems. Format accommodates plastics, metals, composites, wood. MEC (Materials Engineering Center, Dow Chemical Co., Midland, MI). Properties and parameters needed to specify plastic materials for engineering design, with both the product and its tooling defined. Effects of time, temperature, other environmental conditions included. MPD (Materials Property Data, National Materials Property Network, Inc., Columbus, OH). Not-for-profit corporation that addresses the need for easy, on-line access to highquality, well-documented numerical material property data. Enables the designer to canvass multiple databases for the materials information needed, making it possible to define properties' requirements interactively on a terminal and get back a list of databases with detailed documentation on the potential materials. Can also download the data into inhouse computer files. PLASCAMS (RAPRAT Technology, Ltd., Shawbury, Shropshire, England). Provides two search routes, enabling user to search more than 75 material properties. First, an elimination procedure identifies materials that satisfy certain essential criteria. Then the second search optimizes a procedure that can rank or order a short list of materials with the essential properties. Each material in the database gets ranked from 0 to 9 for each quality and it also has specific property data and lists of commercial suppliers plus typical applications. PLASPEC (Plaspec [PT), New York, NY). Complete, accurate up-to-date initial material selection available from all supplies through one source. Over 600 searchable characteristics listed, including special features of plastic materials, property data, cost. PLASTEC (Plastics Technology Evaluation Center, Plastec, U.S. Army, Picatinny Arsenal, Dover, NJ). Has been U.S. Dept. of Defense sponsored, since 1960, to provide the defense community with technical information services applicable to plastics, adhesives, and composites. Includes numerical data of military environments on deterioration of plastics and the like. UWS (Underwriters Laboratories Data Services, Underwriters Laboratories, Melville, NY). UL's worldwide independent third-party product safety certification. Query database can answer such questions as what grades of nylon are rated 140°C (2850f) with a 94Vo flame rating, which TPEs have a 94\'-0 rating, which PPs with a temperature index of 121°C (250°F) or greater are available from Germany, what wire styles are available for use internally in appliances rated at 104°C (220°F) with 300V and are oil resistant, and so on.

MATERIALS AND DESIGNS CAMPUS (Computer-Aided Material Preselection by Uniform Standards, KU-A-KFC, Geb. B207, Leverkusen, West Germany). Different European plastic material suppliers

APPENDIX D 933

started by BASF, Bayer, Huls, and Hoechst provide key data properties tested under the same conditions, including diagrams relating to tensile stress, creep, torsion, etc. Also rheological and thermodynamic data for all mold-design calculations. Free discs available to customers (see Chapter 6 under Computerized Databases for more information). BAYDISK (Bayer Diskette Information System for Plastics, Bayer-Mobay Corp., Pittsburgh, PA). Materials information, application advice, and calculation programs for plastics applications. Includes 1) RALPH (Recommended Admissible Load for Plastic of High-Quality), which establishes maximum loads for parts subject to mechanical stressing, and makes allowances for the type of loading (static tensile, static long-term, dynamic, impact), environmental conditions, internal and external structural characteristics (weld lines, notches, glass-fiber orientation, etc.). Stress-strain curves can be displayed for both short- and long-term loading, for any temperature and time, with recommended admissible loading limits; 2) FLAEMO, a program to establish moments of inertia, centers of gravity, outer fiber spacing, and surface areas for any desired cross-section; 3) FINEL, a program to calculate stresses, strains, and deformation for beams in any design with variable cross-sections and different boundary conditions. Also makes allowances for any nonlinearity with large-scale deflection; 4) BAYMAT, which contains thermal and rheological data for the design of molded parts and molds from the flow-engineering angle. Other programs to be included will be TFEW, to calculate temperature distribution as for cooling time in injection molding, and SNAP, for snap fits. WIS (BASF Corp., Parsippany, NJ). A materials database on floppy discs to assist the designer to select proper plastic materials or determine part dimensions. Includes at least four floppy discs: AGING TEP (Thermal Endurance Profile) contains thermal aging curves as well as important individual parameters required to compare materials. The calculations for the reduction factor for part dimensions and viscosity data area provided as either graphic representations or approximate coefficients. BEAMS, a simple PEA program, provides tension and deformation analyses for parts subjected to bending stresses but for which their cross-sections need not be constant and whose center line can be curved, such as snap connections. The calculations may take into account not only large deformations but also nonlinear stress-strain behavior. Provides direct access to GRAPH 1 (stress-strain functions), GRAPH 2 (static long-term stressing), and RHEODAT (rheological parameters). SCREWS calculates screw connections with self-tapping screws, including the geometry of the screw-plug cylinder, screw-removal force, and screw-in and overdrive moments for customary screw designs. AEDL (Applications Engineering Design Laboratory, B. F. Goodrich Co., Avon Lake, OH). A CAD/CAE database for products and molds using PVC to replace and compete with other plastics. It runs silicon graphics solid models, C-flow and C-cool software from Advanced CAE Technology, and PEA programs running IDEAS, from SRDC. EDD (Engineering Design Database, GE Plastics, Pittsfield, MA). Properties and information not found on data sheets, in design books, or in ordinary databases. Includes tensile static stress-strain, creep and fatigue, rheology and specific heat. All data are generated at several temperatures, stresses, strain rates, and times. Analyses include ribstiffening plate analysis, cooling calculations, assembly, structural foams, glass RPS. POLYFACTS (Du Pont, Wilmington, DE). Du Pont's engineering plastics and generic information on other plastics. Graphic curves for stress-strain, property versus temperature, weathering. Application guides provide lists of applications similar to users' input, with key properties. Troubleshooting guide and general processing information on materials is available.

934 APPENDIX D

SHRINKAGE C PACK (Advanced CAE Technology, Inc., Ithaca, NY). Calculates volumetric shrinkage of analyzed elements in terms of density differences between in-mold and stabilized conditions. Linear shrinkage prediction not yet available. IDEAS (SRDC, Milford, OH). This shrink and warp module provides volumetric and linear shrinkage data with some warpage forecasting. SIMVFLOW3D (Unisys, Inc., Boulder, CO). Calculates shrinkage as a function of density differences between packing pressure and stabilized conditions. SWIS (Moldflow, Kalamazoo, MI). Predicts linear shrinkage and, with unreinforced materials, part warpage. Planned for future commercial release as part of a refined software system. TMConceptlCSE (Computerized Shrink Evaluation, Plastics & Computer, Inc., Montclair, NJ). Develops the actual mold dimensions needed to meet specific product tolerances, taking into account part design, gate location and geometry, mold filling, process conditions, and postmold stabilization.

PROCESSING AND PART DESIGN PITA (Polymer Inflation Thinning Analysis, GE Plastics, Pittsfield, MA). Blow-molding PEA design model that accounts for both performance part design and the processing characteristics that influence part design. Modeled as a nonlinear elastic with the material properties dependent on temperature and strain rate. Accurately predicts wall thicknesses for complicated parts from small to large shapes. STAT (Sheet Thinning Analysis for Thermoforming, GE Plastics, Pittsfield, MA). This thermoforming PEA design model for programming and product development basically takes the same approach as PITA (just reviewed). See also the earlier section on "Materials and Designs" for more information on processing and part designs.

INJECTION MOLDS MOWFLOW (Moldflow Ltd., Trumball, CT). A series of software modules to analyze melt flow, cooling, shrinkage, warpage, and stress in parts. Covers the filling, holding, and cooling stages so that conditions within the mold are monitored until the parts eject. Uses a common geometric database (3-D-PEA). MOWTEMP (Moldflow Ltd., Trumbull, CT). Thermal analysis to optimize the cooling circuit. Calculates the water flow and heat-transfer capability of each section of cooling channels, the temperature distribution profiles at the metal-to-plastic interface, the pressure drop in the cooling circuit, and the coolant temperature rise. Shrinkage analysis modules are made both parallel and perpendicular to the flow for a series of individual layers through the part thickness at each finite element of the model. Opposite sides of a part having different cooling conditions will shrink to different degrees, including bending stresses or warpage that will show up in unbalanced stress analysis. Provides methods to balance this situation via better gate locations, etc. SIMUFLOW (Unisys Corp., Boulder, CO). Provides further analysis into its Mold Maker family, beyond simple 3-D design and drafting, by adding tools for PEA, mold filling, and mold cooling. Numerical control of part capabilities with Unisys's Opti Mold graphic mold design, including a database of standard mold bases and components.

APPENDIX D 935

TMConcept (Plastics & Computer, Inc., Montclair, NJ). An PEA flow program for mold design that evolves into an integrated system with capabilities for material selection, determination of molding conditions, flow analysis, cooling analysis, shrinkage analysis, part tolerances, cost optimization. Incorporates the injection rate, pressure, or clamp force capacity of 1M machines. Shrinkage programs available for different levels of precision molding. Aimed at developing process-fault analysis at the shop-floor level. All mold and product characteristics and their interrelations with the process are "memorized" in the molding machine's process-control system. New software for fault analysis will make full use of this large amount of information to help the operator prevent or correct problems caused by variations in materials and machine performance.

MANAGEMENT PMS (Plastics Management System, Data Technical Research, Jacksonville, FL). Fully integrated manufacturing and financial management system designed specifically for plastics processors working with injection molding, extrusion, blow molding, thermoforming, and so on with single- or multiple-plant operations. Product quotations, order entry, release ordering, inventory control, forward and backward production scheduling, material requirements, capacity planning, machine utilization, production experiences, product costing, tooling costs, other programs. Includes fully integrated financial accounting applications.

GENERAL INFORMATION CPR (Center for Research, National Institute of Standards and Technology, Gaithersburg, MD). Software to predict fire's behavior in buildings, transportation vehicles, etc. COSMIC (Computer Software Management and Information Center, University of Georgia, Athens). This organization collects, evaluates, and disseminates computer software developed by NASA and NASA contractors. Datapro (McGraw-Hill, Inc., New York, NY). Software master index that includes book publications in two volumes-vol. 1 on CAD/CAM/CAE systems and vol. 2 on engineering and scientific information, including subdivision on graphics, plastics, mathematics, and more. Books (R. R. Bowker Publications, New York, NY). The Software EncyclopediaGuide to Microcomputer Software includes sections on engineering and science. Software Catalog (Elsevier Science Publications, New York, NY). Six volumes pertaining to microcomputers, mini-computers, science and engineering, business software, health professions, systems software, others.

REFERENCES

1. Dorgham, M. A., and Rosato, D. V. Designing with Plastics Composites. Interscience Enterprises, Ltd. World Trade Center Bldg., 110 Avenue Louis Casai, Case Postale 306, CH-1215 Geneva-Aeroport, Switzerland. 1986. 2. Hoechst Celanese Corp. Designing with Plastics-The Fundamentals. Plastics Engineering Dept. Manual, 1989. 3. Canby, T. Y., and O'Rea, C. "Reshaping Our Lives-Advanced Materials." National Geographic, vol. 176, no. 6, pp. 746-81, Dec. 1989. 4. "Plastics in the 21st Century," Plastics World, Sept. 1987. 5. Rosato, D. V. Designing with Plastics. Rhode Island School of Design, Lectures, 19871990. 6. _ _ . "Product Design: Basic Processing Guide for Plastics." SPE ANTEC, pp. 1646-50, May 1989. 7. _ _ . "Advanced Designing with Plastics." Univ. of Lowell, Lowell, MA, Seminars, 1986. 8. _ _ . "Advanced Engineering Design-Short Course." ASME Engineering Conf., 1983. 9. _ _ . "Designing with Plastics." Medical Device & Diagnostic Industry, pp. 26-29, July 1983. 10. _ _ , and Rosato, D. V. Plastics Processing Data Handbook. Van Nostrand Reinhold, New York, 1990. 11. _ _ . Blow Molding Handbook, Munich: Hanser, 1989. 12. _ _ . Injection Molding Handbook. New York: Van Nostrand Reinhold, 1986. 13. _ _ . Filament Winding, New York: Wiley, 1964. 14. _ _ , and Schwartz, R. T. Environmental Effects on Polymeric Materials. Vol. 1: Environment, 2: Materials. New York: Wiley, 1968. 15. Rosato, D. V.; Fallo, W. K.; and Rosato, D. V. Markets for Plastics. New York: Van Nostrand Reinhold, 1969. 16. Rosato, D. V. Weighing Out the Aircraft Market. Plastics World (Aug., Sept. 1967). 17. Rosato, D. V., Capt. All Plastic Military Airplane Successfully Flight Tested. Wright-Patterson A.F. Base, Dayton, Ohio, 1944. 18. Rosato, D. V. "Materials Selection." Encyclopedia of Composites. S. Lee, Ed. New York, VCH,1990. 19. _ _ . "Materials Selection." In Mark-Bika1es-Overberger-Menges, Encyclopedia of Polymer Science & Engineering, vol. 9. New York: Wiley, 1987. 20. _ _ . "Thermosets." In Mark et al., Encyclopedia of Polymer Science & Engineering. vol. 14. New York: Wiley, 1988. 21. Rosato, D. V. "Polymers, Processes & Properties of Medical Plastics-Including Applications." In Szycher-Robinson, Synthetic Biomedical Polymers, Technamics, 1980. 22. _ _ . "Electrical Wire and Cable Plastic Coating-What's Ahead?" Wire & Wire Products, pp. 46-61, Mar. 1970. 23. _ _ . "What to Consider in Picking an Injection Molding Machine." Plastics Today, Jan. 1989. 937

938 REFERENCES

24. _ _ . "Injection Molding Thennosets." SPI Reinforced Plastics/Composites Annual Meeting, Seminar, Jan. 1985. 25. _ _ . "Extrusion: Technology, Markets, Economics." Univ. of Lowell, Lowell, MA, Seminar, 1987. 26. _ _ . "Choosing the Blow Molding Machines to Meet Product Demand." Plastics Today, Jan. 1989. 27. _ _ . "Reinforced Plastics/Composites." Univ. of Lowell, Lowell, MA, Seminar, 1986. 28. Du Bois, J. H., and Rosato, D. V. "From Laminates to Composites." Plastics World, Apr. 1968. 29. Rosato, D. V. "Non-Woven Fibers in Reinforced Plastics." Industrial Engineering Chemistry, vol. 54, No.8, pp. 30-37, Sept. 1962. 30. Milewski, J. M., and Rosato, D. V. "The History of Reinforced Plastics." Proceedings, ACS annual Conference, Mar. 19, 1980. 31. Rosato, D. V. Asbestos. New York: Van Nostrand Reinhold, 1959. 32. Lubin, G., and Rosato, D. V. "Application of Reinforced Plastics." 4th International RP Conference, British Plastics Federation. London, Nov. 25-27, 1964. 33. Schwartz, R. T., and Rosato, D. V. "Structural Sandwich Construction." In Dietz, A.G.H. Composite Engineering Materials, pp. 165-81. Cambridge, MA: MIT Press, 1969. 34. Rosato, D. V. "Testing and Statistical Quality Control." Univ. of Lowell, Lowell, MA, Seminar, 1986. 35. _ _ , and Lawrence, J. R. Plastics Industry Safety Handbook. Cahners, 1973. 36. Rosato, D. V. "25 Years of Polyethylene." Plastics World, Jan. 1967. 37. Rosato, D. V., Capt. "Theoretical Potential for Polyethylene." U. S. -British Correspondence, U.S.A.F. Materials Lab., Wright-Patterson Army Base, Dayton, Ohio, 1944. 38. Rosato, D. V. "Insulation News." Wire & Wire Products. pp. 75-79, Nov. 1970. 39. _ _ . "Wire & Wire Products-Insulation News." Wire & Wire Products, Dec. 1970. 40. Schwartz, S. S., and Goodman, S. Plastics Materials & Processes. New York: Van Nostrand Reinhold, 1982. 41. Beck, R. D. Plastics Product Design. 2nd ed. New York: Van Nostrand Reinhold, 1980. 42. Miller, E. Plastics Products Design Handbook. Part A: Components; Part B: Processes & Design. New York: Marcel Dekker, 1981, 1983. 43. Rosato, D. V. "Why Not Use Metal Wire in Filament Winding?" The Iron Age, Mar. 26, 1964. 44. _ _ . "Nose Cone of First U.S. Moon Vanguard Rocket Is Made in Manheim-U.S." New Era [Lancaster, PAl. Nov. 30, 1957. 45. Dym, J. B. Product Design with Plastics. Industrial Press, 1983. 46. Molding Tomorrow's Products Better-Today. Boonton Molding Co., Inc. 1 ed. (1933)-13 ed. (1973). 47. Pornnimit, B., and Kassel, G.W.E. "Self-Reinforcing Polyethylene Produced by Means of Extrusion." KunststofJe, pp. 43-46, July 1989. 48. Kaufman, M. The First Century of Plastics. Plastics Institute, 6 Manderville Place, London WI, 1963. 49. Rosato, D. V. "Reinforced Plastics for Rockets & Missiles." Aircraft & Missile Mfgr., Jan. 1958. 50. Frados, J. SPI Plastics Engineering Handbook. New York: Van Nostrand Reinhold, 1976. 51. Diamond, W. J. Practical Experimental Designs. Munich: Hanser, 1988. 52. Du Bois, J. H., and John, F. W. Plastics, 6 ed. New York: Van Nostrand Reinhold, 1981. 53. Levy, S., and Du Bois, J. H. Plastics Product Design Engineering Handbook. New York: Van Nostrand Reinhold, 1977. 54. Rosato, D. V. "Outer Space Parabolic Reflector Energy Converters." SAMPE Symposium, Phila., PA, June 2-5,1963. 55. French, T. E., et al. Engineering Drawing and Graphic Technology. New York: McGrawHill, 1986.

REFERENCES 939

56. Earle, J. H. Engineering Design Graphics, 5 ed. Reading, MA: Addison-Wesley, 1987. 57. De Long, G., et al. Engineering Graphics; Communication, Analysis, Creative Design. New York: McGraw-Hill, 1986. 58. Barton, L. 0 . Mechanism Analysis: Simplified Graphics & Analytical Techniques. New York: Marcel Dekker, 1984. 59. O'Neal, R., and Pennington, M. "Initial Visual Design Projections for an Interactive Work Environment." Rhode Island School of Design Newsletter, fall 1988. 60. Eide, A. R. Engineering Graphics Fundamentals. New York: McGraw-Hill, 1985. 61. Spencer, H. C. Engineering Graphics Problems. New York: Macmillan, 1981. 62. Engineered Materials Handbook. Vol. 1: Composites. ASM International, 1987. 63. _ _ . Vol. 2: Engineering Plastics. ASM International, 1988. 64. Composites. Vol. 1: Engineering Materials Handbook. Denver, CO: Edgell Communications, 1988. 65. Structural Plastics Design Manual. No. 023-000-00495-0. Supt. of Documents, GPO, 1982. 66. Heger, F. J. Structural Plastics Design Manual. ASCE, 1981. 67. Plastics for Aerospace Vehicles. MIL-HDBK-17A and 17B. Supt. of Documents, GPO: 1981. 68. Structural Sandwich Composites. MIL-HDBK-23A. Supt. of Documents, GPO: 1974. 69. Glanville, A. B. Plastics Engineers Data Book. UK: Machine Publ., 1971. 70. Ehrenstein, G. W., and Erhard, G. Designing with Plastics. Munich: Hanser, 1987. 71. _ _ . Hanser, 1984. 72. Bauhof, M. "Maintaining Bolt Preload in Plastic Assemblies." Du Pont Engineering Design, pp. 14-15, summer 1987. 73. Brandes, R. "How Critical Is the Design Function?" Modern Plastics, pp. 97-105, Apr. 1989. 74. Middendorf, W. H. Design of Devices and Systems. New York: Marcel Dekker, 1986. 75. Patton, W. J. Plastics Technology: Theory, Design and Manufacture. Englewood Cliffs, NJ: Prentice Hall, 1976. 76. McCormac, J. C. Design of Reinforced Concrete. New York: Harper & Row, 1978. 77. Van Vlack, L. H. Elements of Materials Science and Engineering. Reading, MA: AddisonWesley, 1985. 78. _ _ . Materials for Engineering Concepts and Applications. Reading, MA: Addison-Wesley, 1982. 79. Young, J. E., and Shane, R. S. Materials and Processes: Optimize Design. New York: Marcel Dekker, 1985. 80. Brown, R. L. Design and Manufacturing of Plastics Parts. New York: Wiley, 1980. 81. Bauer, E. Engineering Design of Plastics. New York: Van Nostrand Reinhold, 1964. 82. Saechting, H. International Plastics Handbook. Munich: Hanser, 1983. 83. Chanda, M., and Roy, S. K. Plastics Technology Handbook. New York: Marcel Dekker, 1987. 84. Pritchard, G. Developments in Reinforced Plastics. New York: Elsevier, 1985. 85. Crawford, R. J. Plastics Engineering. 2 ed. New York: Pergamon, 1987. 86. Lubin, G. Handbook of Composites. New York: Van Nostrand Reinhold, 1982. 87. "Composite Materials." Quarterly Review, Belgium Economics & Commercial Information, Belgium Foreign Trade Office, No.1, 1989. 88. Gerstle, F. P. "Composites." In Encyclopedia of Composites. S. Lee, Ed. New York: VCH, 1990. 89. Johnson, A. F. Engineering Design Properties ofGRP: A Joint BFP & NPL Project. British Plastics Federation. Her Majesty's Stationery Office, 1979. 90. "Plastics and Composites." Machine Design, Apr. 1989. 91. Quinn, K. R., and O'Brien, G. S. "High Temperature Advanced Composites." Proceedings, SPI Composites Inst. Conference, Feb. 1-5, 1988. 92. Nagarsheth, P. S. "An Integrated System Approach for Injection Molded Plastic Parts from Design to Fabrication." SPE ANTEC, pp. 220-23, May 1989.

940 REFERENCES

93. Shaeffer, R. E. Elementary Structure for Architects and Builders. New York: Prentice Hall, 1988. 94. Weston, J. W., and Peters, D. M. Engineer's Guide to Composite Materials. ASM International, 1987. 95. Cushion, R. F. Engineers-Malpractice-U.S.A. New York: Wiley, 1987. 96. Troitzsch, J. International Plastics Flammability Handbook. Munich: Hanser, 1983. 97. Schaffer, E. L. Behavior of Polymeric Materials in Fire. ASTM, STP 816, 1983. 98. Mihalich, J. "Silicone Polymer Boosts Flame, Flow, Mechanical Properties in Olefin Polymer Systems." Plastics Today, Aug. 19, 1989. 99. Landrock, A. H. Handbook of Plastics Flammability and Combustion Technology. Park Ridge, NJ: Noyes Pub!., 1983. 100. Marmathy, S. Design of Buildings for Fire Safety. ASTM, STP 685, 1969. 101. Gill, J. C. "Brominated Polystyrene Flame Retardants: A Step Forward." Plastics Compounding, pp. 77-81, Sept.-Oct. 1989. 102. Aseeva, R. M., and Zaikor, G. E. Combustion of Polymer Materials. Munich: Hanser, 1986. 103. Castino, L., and Harmathy, J. Fire Risk Assessment. ASTM, STP 762, 1982. 104. Accident Prevention Manual for Industrial Operations. National Safety Council (annual). 105. Miller, B. "Design-for-Assembly Slashes Cost of Hoover's All-Plastic Vacuum Cleaner." Plastics World, pp. 34-38, 1989. 106. Leaversuch, R. D. "Designers Create Whole New Categories of Housewares." Modern Plastics, pp. 44-47, Dec. 1988. 107. Applied Design Library: Pictorial Design Creative Books. Glen Cove, NY: PBC International. 108. Katz, S. Objects, Classical Designs, With a Collector's Guide. New York: Abrams, 1990. 109. Smith, P. G. "Winning the New Products Rat Race." Machine Design, May 1988. 110. "Plastics House of Disneyland." Plastics World, pp. 31-38, 1987. 111. "Monsanto Debuts Air Duct Manual." Plastics News, p. 11, Mar. 13, 1989. 112. "Living Environments Concept House Explores New Ideas, Materials and Building Processes." GE Plastics News Release, Nov. 1989. 113. "Plastics House Previews Future." Design News, pp. 28-30, Dec. 1989. 114. "Composites in Construction: GE House to Demonstrate, Explore New Applications for Composites." PPG Industries, Reinforcement Digest, pp. 20--27, Jan. 1990. 115. "Guide to Component Selection." Medical Device & Diagnostic Industry, pp. 30--36, July 1989. 116. Murr, L. E. Shock Waves for Industrial Applications. Park Ridge, NJ, Noyes Pub!., 1989. 117. "Engineering Plastics: More Products, More Competition." C&E News, pp. 21-46, Aug. 18, 1986. 118. Dibble, M. A. "Polymer Balloon Breaks Altitude Record." Machine Design, pp. 62-66, Aug. 1988. 119. Wigotsky, V. "Creativity and Economy-Automotive Plastics." Plastics Engineering, pp. 23-30, Sept. 1989. 120. Rowand, R. "Rogers Corp. Acquires Rights to Polymotor Plastic Engine." Plastics Today, p. 2, Sept. 18, 1989. 121. Kirkland, C. "New Routes to the All-Plastic Car." Plastics Technology, pp. 46-49, June 1986. 122. Kandt, A. "From Design to Production-Vehicle Fuel Tanks." European Plastics News, pp. 31-34, Mar. 1986. 123. "GM Minivans." Plastics Today, pp. I, 3, Sept. 23, 1989. 124. "Corvette Advanced Bumper Technology News." Design News, p. 24, Jan. 1989. 125. Hayward, T. C. "Rigid PVC for the Building Construction and Transportation Industries." SPE ANTEC, pp. 718-21, May 1989. 126. Bakkar, M. Encyclopedia of Packaging Technology. New York: Wiley, 1988. 127. Lund, P. R. "Advances in Food Packaging with Nitrile Polymers." SPE ANTEC, pp. 85659, May 1989.

REFERENCES 941

128. "New PET Yields Better Blisters." Packaging, p. 75, July 1989. 129. Canning and Related Processes. Book 1: Basics on Canning; Book 2: Packaging, Aseptic, Ingredients; Book 3: Processing Procedures for Canned Food Products. Baltimore: CTI Publ., 1986. 130. Packaging Encyclopedia. Packaging. Annual. 131. Weiss, M., and Freiherr, G. "The New Era of Medical Devices." Medical Device & Diagnostic Industry, pp. 45-48, Sept. 1989. 132. Holstein, H. M. "Designing Quality In." Medical Device & Diagnostic Industry, pp. 32-35, Oct. 1988. 133. Woo, L., and Chenng, Y. W. "Medical Plastics Compatibility Requirements and Testing." SPE ANTEC, pp. 1292-95, May 1989. 134. Wilson, T. W. "Dental Composites: A Relationship Between Glass Transition Temperature and Wear." SPE ANTEC, pp. 1287-89, May 1989. 135. Szycher, M. Biocompatible Polymers; Metals and Ceramics. Technornic, 1983. 136. _ _ . and Robinson, W. J. Synthetic Biomedical Polymers. Technornic, 1980. 137. Machlis, S. "Miracle Materials for Medicine." Design News, pp. 100-105, Oct. 1989. 138. Lynch, W. Implants: Reconstruction of the Human Body. New York: Van Nostrand Reinhold, 1982. 139. Lee, H. and Neville, K. Handbook of Biomedical Plastics, Pasadena [CAl Technical Press, 1971. 140. Allbee, N. "Plastics in the Medical World." Plastics Compounding, pp. 36-46, Sept.-Oct. 1989. 141. "Composites at the Paris Air Show." Advanced Composites, pp. 66-70, July-Aug. 1989. 142. McConnell, V. P. "Advanced Composites Make a Case in Space." Advanced Composites, pp. 34-46, Sept.-Oct. 1989. 143. Wigotsky, V. "Advanced Composites Are Spreading Their Wings." Plastics Engineering, pp. 25-32, Oct. 1988. 144. "DC-3's, The Gooney Bird Lives." Compressed Air, pp. 9-15, Sept. 1989. 145. Cullity, D., and Boller, T. "Development of Ballistic-Damage-Tolerant Flight Control Components Molded of a Short-Fiber Reinforced Composite Material." Goodyear Aerospace Corp., Sept. 1973. 146. Du Bois, J. H. "From Plastics to Composites: Aircraft, etc." Plastics World, pp. 44-49, Nov. 1968. 147. Plantema, F. J. Sandwich Construction. Vol. 3 of Airplane Missile and Spacecraft Structures. New York: Wiley, 1966. 148. Booda, L. "E-2A Compatible with Carrier Operations." Aviation Week & Space Technology, pp. 68-77, Mar. 1963. 149. Bragham, W. E. "Reinforced Plastics Aircraft Fuselage Structure. U.S. Navy, Neptune P2VZenith Plastics." Plastics World, Mar. 1954. 150. "Plastics Aircraft Wings; Experimental Flight Tested; AF T-6 Trainer." Plastics World, Mar. 1954. 151. "AT-6C Glass Outer Wing Panel Development." Army A.F. Tech. Report No. 5576, Apr. 23, 1947. 152. Reinfrank, G. B., Capt., and Norman, W. A., Lt. "Molded Glass Fiber Sandwich Fuselage for BT-15 Airplane." Army A.F. Tech. Report No. 5159, Nov. 8, 1944. 153. "Plastics in Aircraft." Army-Navy-Civil Industry Conference, Philadelphia, PA, 1943. Modern Plastics, June-Sept. 1943. 154. "Reinforced Plastics Vessels." NY: American Bureau of Shipping, 1978. 155. "Carbon/Nylon Makes a Fast Wheel for Less." Plastics Today, p. 10, Oct. 6, 1989. 156. "Heat Resistant Resin Makes Hot Selling Utensils." Plastics Today, p. 17, June 24,1989. 157. Basta, N. "Solving Our Environmental Crisis." Graduating Engineering, pp. 38-42, Sept. 1989. 158. Bemarde, M. A. Our Precious Habitat. New York: Wiley, 1989.

942 REFERENCES

159. Brewer, D. "The Disappearing Act of Degradable Plastic Packaging." Plastics Packaging, pp. 26-31, Sept.-Oct. 1988. 160. Chen, I. M., and Shiah, G. "Producing Tough PET/HDPE Blends from Recycled Beverage Bottles [in China]." Plastics Engineering, pp. 33-35, Oct. 1989. 161. "Clearing Away the Rubbish." Modern Plastics, p. 43, Sept. 1989. 162. Corley, J. "Incinerating PVC Is Not Harmful." Modern Plastics, p. 134, Dec. 1988. 163. "Food in Tubes: American Style Special Report." Tube Council of America, Tube Topic Marketing Newsletter. Vol. 23, no. 3, 1989. 164. "Hazardous Waste." Compressed Air, pp. 6-13, Jan. 1989. 165. Hunniball, K. "Aspects of Design Responsibility-Waste Source Reduction." Proceedings, IDSA Conference on Design Education, pp. 82-85, 1989. 166. Kaufer, H. "Product Design to Meet Recycling; Production by Instrinsic (Orientation) Strengthening." KunststoJfe, pp. 23-26, Apr. 1989, and pp. 45-47, Sept. 1989. 167. Keams, D. "PET Bottles Reduce Energy and Environmental Impacts-Study Shows." Plastics Today, p. 5, May 6, 1989. 168. Leonard, L. V. "Designing for Posterity-Are Plastics a Good Choice?" Plastics Design Forum, pp. 27-42, May-June 1988. 169. "Municipal Solid Waste." Report, Center for Recycling Research, Rutgers Univ., 1986. 170. O'Brien, J. "Humans: Do We Deserve Our Superior Status?" Marco Island Eagle, FL, p. 7, Oct. 4, 1989. 171. "Plastics Degrades Under Soil and Sea." Design News, p. 64, Sept. 1989. 172. Plastics Information File on Councils for Solid Waste Solutions, Keep America Beautiful, Plastics Recycling Foundation and Others." Plastics World, pp. 80--81, Sept. 1989. 173. "Proctor & Gamble Will Launch Use of Recycled PET Bottles." Plastics World, pp. 21-22, Dec. 1988. 174. RealI, J. "Plastic Land Fill Waste Overstated-Expert." Plastics Today, p. 4, Sept. 25, 1989. 175. "Recycling." Plastics Technology, p. 105, June 1989. 176. Recycling directories, plastic recycling update newsletter. SPI: Plastic Bottle Institute, 1989. 177. "Recyclying: Do It Today for Tomorrow." Chicago: Amoco Chemicals, 1989. 178. "Recycling Is Not Always Easy; Neither Is Life." Plastics Engineering, June 1988. 179. "Recycling Technology Proliferates." Plastics Technology, pp. 45-49, July 1989. 180. Rosato, D. V. "Plastics and Solid Waste." Rhode Island School of Design Lecture, Oct. 25, 1989. 181. Savage, D. "Waste Not." Amoco Torch, p. 21, Sept.-Oct. 1989. 182. Shannon, J. "Recycling-High Energy Value of Plastics Stressed." Plastics Today, pp. 21, 23, July 8, 1989. 183. Short, H. "Dow Chief Says Firms Must Beware of Anti-Plastics Efforts." Plastics News, p. 5, Oct. 16, 1989. 184. _ _ . "Germans Propose Waste Disposal Plan." Plastics Today, pp. 1,24, Sept. 11, 1989. 185. Silander, L. "Turning the Tide on Environmental Waste." Rhode Island School of Design Views, pp. 1-2, Sept. 1989. 186. "Solid Waste: The Way Out." Plastics World Special Issue, pp. 12-85, Sept. 1989. 187. Stacy, N., et al. "Biodegradable Natural-Synthetic Polymer Graft Copolymers." SPE ANTEC, pp. 1362-64, May 1989. 188. "The Word Is Out; Plastics Are Recyclable." Plastics Today, p. 5, July 22, 1989. 189. Valney, S. A. "Biodegradability Is Not the Solution." Modern Plastics, p. 8, Oct. 1989. 190. Vogler, J. "Small Scale Recycling of Plastics." Intermediate Technology, England, 1984. 191. "PVC Enters the Recycling Arena." Plastics Today, p. 3, Oct. 21, 1989. 192. "Material, Process Changes Begin As Waste Redution Gains." Plastics Today, pp. 1,3, Oct. 21, 1989. 193. "Yeas and Nays for Degradable Bags." Plastics Today, pp. 1,4, Sept. 2,1989. 194. "Plastics Industry & Solid Waste Management." SPI Bulletin, 1980. 195. "Plastics Bottle Recycling Directory & Reference Guide." SPI Plastic Bottle Institute, 1986.

REFERENCES 943

196. Carrier, K. "Technology for Recycling Post-Consumer and Industrial Waste Plastic into New High Quality Products." SPE ANTEC, pp. 1807-8, May 1989. 197. Alter, H. "The Greatly Growing Garbage Problem." U.S. Chamber of Commerce, Publications Fulfillment, 1615 H St., Washington, DC 20062, 1989. 198. Dibble, M. A. "Notice: Plastics May Be Vital to Your Health." Machine Design, pp. 5063, June 1988. 199. Rosato, D. V. "Plastic Replaced Aorta Permits Living Normal-Long Life." Newton-Wellesly [MAl Hospital, Mar. 1987. 200. Safety and Health Reference Guide for Plastics Industry Safety Handbook. SPI Plastips, 1984. 201. Billmeyer, F. W. Textbook of Polymer Science. 3 ed. New York: Wiley, 1984. 202. Lange, A., and Ezrin, M. Plastics Analysis Guide. Munich: Hanser, 1983. 203. Dealy, J. M. Rheometrics for Molten Plastics. New York: Van Nostrand Reinhold, 1981. 204. Bruins, P. F. Polyblends and Composites. New York: Wiley 1970. 205. Ceresa, J. A. Block Graft Copolymerization. New York: Wiley, 1973. 206. Mascia, L. The Role of Additives in Plastics. New York: Wiley, 1974. 207. Gacher, R. Plastics Additives. Hanser, 1987. 208. Burns, R. Polyester Molding Compounds. New York: Marcel Dekker, 1982. 209. Katz, H. S., and Milewski, J. Handbook of Fillers & Reinforcementsfor Plastics. New York: Van Nostrand Reinhold, 1978. 210. Hall, I. H. Structure of Crystalline Polymers. New York: Elsevier, 1984. 211. Matthews, G.A.R. Polymer Mixing Technology. New York: Elsevier, 1982. 212. McCullough, R. L. "Anisotropic Elastic Behavior of Crystalline Polymers, Treatise on Material Science and Technology." New York: Academic, pp. 453-540, 1977. 213. Bralla, J. G. Handbook of Product Design for Manufacturing. New York: McGraw-Hill, 1986. 214. Passaglia, E., and Knox, J. R. "Viscoelastic Behaviors & Time-Temperature Relationship." Engineering Design for Plastics. E. Baer, Ed. Chapter 3. New York: Van Nostrand Reinhold, 1964. 215. Schroder, E. Guide to Polymer Characterization. Munich: Hanser, 1988. 216. Struic, L.C.E. Physical Aging in Amorphous Polymers and Other Materials. New York: Elsevier, 1978. 217. Tobolsky, A. V. Properties and Structures of Polymers. New York: Wiley, 1960. 218. Woodward, A. E. Atlas of Polymer Morphology. Munich: Hanser, 1988. 219. Turi, E. A. Thermal Characterization of Polymer Materials. Technornic, 1981. 220. Tanner, R. I. Engineering Rheology. Munich: Hanser, 1986. 221. Mitchell, T. Applied Polymer Analysis and Characterization. Munich: Hanser, 1987. 222. Kreith, F. Principles of Heat Transfer, Serration. Pa. International Textbook Co., 1965. 223. Kampf, G. Characterization of Plastics by Physical Techniques and Practical Applications. Munich: Hanser, 1986. 224. Bell, R. G. "Microtorning: An Emerging Tool for Analyzing Polymer Structure." Plastics Engineering, Aug. 1979. 225. Thakkar, B. S., and Broutman, L. J. "Residual Stress Measurements." Polymer Science, 11, 18, 1980. 226. Schnabel, W. Polymer Degradation: Principles and Applications. Munich: Hanser, 1981. 227. Titomanlio, G.; Drucato, V.; and Kamal, M. R. "Mechanism of Cooling Stress Build-Up in Injection Molding of Thermoplastic Polymers." International Polymer Processing. Vol. 1, pp. 55-59, 1987. 228. Weeks, R. F. "Plastics Optics." Optics News, Sept. 1975. 229. Fantone, S. "A User's Guide to Plastics Optics." Proceedings, Society of Photo-Optical Instrumentation Engineers, Vol. 406, Apr. 1983. 230. Altman, R. M., and Lytle, J. D. "Optical Design Techniques for Polymer Optics." International Lens Design Conference, Optical Society of America, Society of Photo-Optical Instrumentation Engineers. Vol. 237, 1986.

944 REFERENCES

231. Benjamin, R. "Injection Molding of Precision Plastic Optics." Plastics Design Processing, Mar. 1979. 232. Feldman, G., et al. "Fluorinated Materials and The Boston Equalens." Contact Lens Forum, June 1987. 233. McLean, K. "Machinable Plastic Stock Shapes." Machine Design, pp. 125-28, Nov. 1987. 234. Berlinger, H. P., and Ludwigshafen, T. "Designing of Flanges on Plastic Components." KunststofJe, pp. 25-27, Sept. 1988. 235. Throne, J. L., and Progelhof, R. C. "Optimized Ribbed Plate Design." SPE ANTEC, pp. 1637-41, May 1989. 236. Opli, C., et al. "Product Design for Economical Injection Molding-The Influence of Part Shape on Cycle Time." SPE ANTEC, pp. 291-94, May 1989. 237. Medwin, S. J. "Forming Contoured Structural Parts." Plastics Today, pp. 15-16, Aug. 5, 1989. 238. Johnson, J. E., et al. "Reinforced Plastic Beam Shapes." ASCE Conference, Nov. 1972. 239. Wittfoht, A. M. Plastics Technology Dictionary. Munich: Hanser, 1987. 240. Whittington, L. R. Whittington's Dictionary of Plastics. Technomic, 1968. 241. Ezrin, M. Plastics Analysis Guide. Munich: Hanser, 1984. 242. Plastics Engineering Dictionary; German-Engli~h. Munich: Hanser, 1982. 243. Parker, S. P. Dictionary of Scientific & Technical Terms. New York: McGraw-Hill, 1989. 244. Braun, D. Simple Methods for Identification of Plastics. Munich: Hanser, 1987. 245. Pearson, C. E. Engineering Mathematics. New York: Van Nostrand Reinhold, 1986. 246. Belding, W. E. Handbook of Engineering Mathematics. ASM International, 1983. 247. Beakley, G. C. Engineering Design. Ne~ York: Macmillan, 1986. 248. Baumeister, T., and Marks, L. S. Standard Handbookfor Mechanical Engineers. New York: McGraw-Hill, 1978. 249. Krause, A., et al. Plastics Analysis Guide: Qualitative and Quantitative Methods. Munich: Hanser, 1983. 250. Hengl, R., and Gust, H. "Influence of Loading History on the Mechanical Properties of Polymeric Materials." KunststofJe, pp. 31-33, Mar. 1989. 251. Eshbach, O. W., and Souders, M. Handbook of Engineering Fundamentals. New York: Wiley, 1975. 252. Nielsen, L. E. Mechanical Properties of Polymers and Composites. Vols. 1,2. New York: Marcel Dekker, 1974. 253. Nielsen, L. E. Mechanical Properties of Polymers. New York: Van Nostrand Reinhold, 1962. 254. Sherman, R. "Anisotropic Properties in Injection Molding Fiber Reinforced Plastics Parts." Plastics Engineering, pp. 81-84, May 1988. 255. O'Neil, P. V. Advanced Engineering Mathematics. New York: Wadsworth, 1983. 256. Perry, R. H. Engineering Manual. New York: McGraw-Hill, 1976. 257. Blake, A. Handbook of Mechanics: Materials and Structures. New York: Wiley, 1985. 258. Horne, M. R. Plastic Theory of Structures. Cambridge, MA: MIT, 1971. 259. Jensen, A., and Chenoweth, H. H. Applied Strength of Materials. New York: McGraw-Hill, 1975. 260. Ashby, M. F. Engineering Materials. New York: Pergamon, 1986. 261. Green, W. A., and Micunovic, M. V. Mechanical Behavior of Composites and Laminates. New York: Elsevier, 1987. 262. Modern Plastics Encyclopedia. New York: McGraw-Hill, annual. 263. Ray, M. S. Engineering Design. New York: Prentice Hall, 1987. 264. Wendt, F. W., et al. Mechanics of Composite Materials. New York: Pergamon, 1970. 265. Tuma, J. J. Engineering Mathematics Handbook. New York: McGraw-Hill, 1987. 266. _ _ . Handbook ofNumerical Calculations in Engineering. New York: McGraw-Hill, 1989. 267. Taft, W. G. "Designing High Strength Light Weight Structures." Machine Design, pp. 99103, Jan. 1988. 268. Timoshenko, S. P. Strength of Materials. New York: Van Nostrand Reinhold, 1958.

REFERENCES 945

269. _ _ , and Young, D. H. Elements of Strength of Materials. New York: Van Nostrand Reinhold, 1962. 270. Faupel, J. H. Engineering Design, A Synthesis of Stress Analysis and Materials Engineering. 2 ed. New York: Wiley, 1981. 271. Isayev, A. I., and Crouthamel, D. L. "Residual Stress Development in Injection Molding Polymers." Polymer Plastic Technical Engineering. Vol. 22, pp. 177-232, 1984. 272. Haag, E. "Polycarbonate Stress Crazing: Some Practical Experiences." SPE ANTEC, 1982. 273. Champoux, R. L. "Analytical and Experimental Methods for Residual Stress Effects in Fatigue." ASTM, STP 1004, 1989. 274. Hsu, T. H. Stress and Strain Handbook. Houston: Gulf Pub., 1986. 275. Wetton, R. E., et al. "Comparison of Dynamic Mechanical Measurements in Bending, Tension and Torsion." SPE ANTEC, pp. 1160-62, May 1989. 276. Williams, J. H. Stress Analysis of Polymers. 2 ed, London: Ellis Hardwood Ltd., 1980. 277. _ _ . Stress Analysis of Polymers. New York: Wiley, 1973. 278. Roark, R. J., and Young, W. C. Formula for Stress and Strain. New York: McGraw-Hill, 1976. 279. Reinhart, J. W. "Long-Time Hydrostatic Strength Characteristics of Thermoplastic Pipe." Polymer Engineering & Science, Oct. 1966. 280. Boyer, H. E. Atlas of Stress-Strain Handbook. Houston: Gulf Pub., 1986. 281. Bigg, D. M., and Leininger, R.1. "Stress Cracking Behavior ofPMMA and PMMAIAcryiate Copolymer." SPE ANTEC, 1981. 282. Powell, P. C. "Design of Load-Bearing Plastics" in Materials Engineering Applications. Vol. 1, pp. 190-202, June 1979. 283. Richardson, M. "Integration of Dynamics Testing into Product Design Cycle." Proceedings, Design Engineering Conference, Apr. 1981. 284. Hibbeler, R. C. Engineering Mechanics: Dynamics. New York: Macmillan, 1974. 285. Lindholm, U. S. Mechanical Behavior of Materials Under Dynamic Loads. New York: Springer-Verlag, 1968. 286. Griffith, A. A. "The Phenomena of Rupture and Flow in Solids." Philos. Transactions, Royal Society of London, Ser. A: 221, pp. 163-98, 1921; and "The Theory of Rupture." Proceedings, First International Congress-Applied Mechanics, p. 55, 1924. 287. Maxwell, J. C. Philos. Transactions, Royal Society, London. Vol. 157, p. 49, 1867. 288. Altshuler, T. L. "Fatigue: Life Predictions for Materials Selection and Design." ASM Software, 1988. 289. Little, R. "Tables and Estimating Medium Fatigue Limits." ASTM, STP 327, 1962. 290. Boyer, H. E. Atlas of Fatigue Curves. Metals Park, OH: American Society of Metals International, 1986. 291. Hertzberg, R. W., and Manson, J. A. "Fatigue Testing-Flaws Make It Better." Plastics World, pp. 50-53, May 1977. 292. _ _ . Fatigue of Engineering Plastics. New York: Academic Press, 1980. 293. Hertzberg, R. W. Deformation and Fracture Mechanics ofEngineering Materials. New York: Wiley, 1976. 294. "Vibration Testing with Computers." Machine Design, p. 50, June 1988. 295. Deutschman, A. D., et al. Machine Design: Theory & Practice. New York: Macmillan, 1975. 296. Findley, W. N. "Mechanism and Mechanics of Creep Plastics." SPE Journal, Jan. 1960. 297. Binder, K. "Non-Isothermal Long Term Creep (17 yrs.). Behavior of Thermoplastics Outdoors." KunststofJe, pp. 23-25, Nov. 1988. 298. Manson, S. S., and Haferd, A. M. "A Linear Time-Temperature Relation for Creep and Stress Rupture Data." NACA, TN 2890, 1953. 299. Larson, F. R., and Miller, J. A. "A Time-Temperature Relationship for Rupture and Creep Stresses." Transactions, ASME, Vol. 175, 1952. 300. Orth, F., and Ehrenstein, G. W. "Selection of Materials Under Dynamic Loading." KunststofJe, pp. 43-45, Aug. 1989.

946 REFERENCES

301. Findley, W. N., et al. Creep and Relaxation of Nonlinear Viscoelastic Materials. North Holland Pub., 1976. 302. Fritch, L. W. "The IZod-Orientation Concept for Impact Characterization. Plastics Engineering, pp. 43-45, Sept. 1989. 303. Kessler, S. I. "Instrumented Impact Testing of Plastics and Composite Materials." ASTM, STP 936, 1987. 304. Karaener, Z., etal. "Natural Aging of Impact-Resistance Styrene Copolymers." Kunststoffe, pp. 33-34, Mar. 1988. 305. Bruckway, G. S. "Resisting Impact Fractures." Plastics Design Forum, Oct. 1982. 306. Ku, C. C., and Liepins, R. Electrical Properties of Polymers. Munich: Hanser, 1987. 307. Wessels, B. M. Advances in Electronic Materials. ASM International, 1986. 308. Sarwell, J. A. Electronic Packaging. Metals Park, OH: ASM International, 1986. 309. Frisch, D. C., and Eby, J. "Molded 3D Circuitry: A Commercialized Technology." SPE RETEC, Oct. 27-28, 1988. 310. Rosato, D. V. "Radomes and Antennas." Reiriforced Plastics, pp. 10-12, Jan.-Feb. 1964. 311. Munstedt, H. "Electrically Conductive Polymers." Kunststoffe, pp. 11-14, June 1989. 312. Wolfer, D. "Electrically Conducting Silicone Rubber and Its Uses." Kunststoffe, pp. 14-16, June 1989. 313. "Conductive ABS Sheet Fills Growing Needs in Electronic Packaging." Plastics Today, p. 6, Aug. 5, 1989. 314. Kraus, H. "Battery Separators Made of Sinterable PVC." KunststofJe, pp. 31-32, Nov. 1988. 315. "Piezoelectric Materials." Machine Design, p. 166, May 1988. 316. Bak, D. J. "Foam Sheet Joins Chip Carrier, Circuit Board." Design News, pp. 112-14, Feb. 1989. 317. Mason, J. "Renaissance in Switch Design." Design News, pp. 81-84, Nov. 1989. 318. Wolf, M. "Injection Molding Three-Dimensional Circuit Boards." KunststofJe, pp. 20-22, Feb. 1989. 319. "UHMW-PE Protects and Combines Efficiency." Design News, p. 148, Jan. 1989. 320. "Bearings and Lubricants." Machine Design, pp. 127-49, Oct. 1988. 321. Dvorak, P. "Specialty Bearings Fill Nearly Every Niche." Machine Design, pp. 75-81, June 1988. 322. Timoshenko, S. P. History of Strength of Materials. New York: McGraw-Hill, 1953. 323. "Internally Lubricated Reinforced Thermoplastics and Fluoropolymer Composites." Lubricomp, ICIILNP Engineering Plastics & Design, Bulletin 254-986, 1986. 324. Stefanides, E. J. "Polyethylene Wear Plate Keeps Semi-Truck Trucking." Machine News, pp. 322-23, Nov. 1988. 325. Watson, K. R.; Wu, Y. T.; and Riewald, P. E. "Reinforced Thermoplastics for Wear-Resistant Applications." Plastics World, pp. 83-85, Mar. 1989. 326. Wu, Y. T. "How Short Aramid Fiber Improves Wear Resistance." Modern Plastics, pp. 8996, Mar. 1988. 327. Bayer, R. "Wear Tests for Plastics: Selection and Use." ASTM, STP 701, 1980. 328. Bartener, G. M., and Larrenter, V. V. Friction and Wear of Polymers. New York: Elsevier, 1981. 329. Dudley, D. W. Handbook of Practical Gear Design. New York: McGraw-Hill, 1985. 330. Adams, C. E. Plastics Gearing: Selection and Application. Metals Park, OH: ASM International, 1986. 331. Alban, L. E. Systematic Analysis of Gear Failures. Metals Park, OH: ASM International, 1985. 332. Smith, H. V. "Plastics and the Environment: Some Misconceptions." SPE ANTEC, pp. 1800-01, May 1989. 333. Boyer, H. E. "Selection of Materials for Service Environments." Source Book, Metals Park, OH: ASM International, 1987. 334. Kelen, T. Polymer Degradation. New York: Van Nostrand Reinhold, 1983.

REFERENCES 947

335. Schweitzer, P. A. Corrosion Resistance Tables. New York: Marcel Dekker, 1986. 336. _ _ . Corrosion and Corrosion Protection Handbook. New York: Marcel Dekker, 1989. 337. Munger, C. G. Corrosion Prevention by Protective Coatings. Houston: National Association of Corrosion Engineers, 1984. 338. Fontana, M. G. Corrosion Engineering. New York: McGraw-Hill, 1986. 339. Hell, J. et al., "Optimizing Production Parameters by the Stress Crack Testing Method." KunststofJe, pp. 40-41, Sept. 1989. 340. Albrecht, P., and Yamada, K. "Rapid Calculation of Stress Intensity Factors." ASCE Journal of the Structural Div., Feb. 1977. 341. Lustiger, A., et al. "Environmental Stress Cracking As a Tool in Evaluating the Long Term Strength of Medium Density PE." SPE ANTEC, 1981. 342. Hekeler, W. R. "Resistance of Thermoplastics Against Chemically Induced Stress Cracking." KunststofJe, pp. 29-30, Feb. 1989. 343. Fitz, H. "Weathering Resistance Fluorocarbons for Outdoor Applications." KunststofJe, pp. 17-19, June 1989. 344. Davis, A., and Sims, D. Weathering of Polymers. New York: Elsevier, 1983. 345. Bradley, R. Radiation Technology Handbook. Marcel Dekker, 1984. 346. Ely, J. L. "FDA Regulations and Policy Regarding New Biomaterials." FDA, Washington, DC, 1987. 347. Mertzel, L., et al. "Gamma Sterilization of Rigid and Flexible PVC." SPE ANTEC, Apr. 1988. 348. Lundy, C., et al. "Methods of Color Stabilization in Gamma-Radiation-Sterilized Polycarbonate." SPE-ANTEC, Apr. 1988. 349. Williams, J. L. "Radiation Stability of Polypropylene." SPE-ANTEC, May 1987. 350. Sacharow, S. "Radiation Sterilization Grows in Medical Packaging." Paper, Film & Foil Converter, pp. 125-29, June 1988. 351. Meyer, B. "Radiation Sterilization." SPE-ANTEC, May 1987. 352. Kline & Co., "Irradiation-resistant Packaging." Report, 1988. 353. Rosato, D. V. "Polymer Resistance to Hot Water and Steam Sterilization," Medical Device and Diagnostic Industry, July 1985. 354. Mazouch, R. "Engineering Polymers in Medical Applications." KunststofJe, pp. 28-30, Sept. 1988. 355. Stubstad, J. A. "Preventive Plastic Part Failure." Medical Device and Diagnostic Industry, pp. 50-59, Sept. 1989. 356. Giacin, J. R. "Additive and Monomer Migration/Diffusion in Plastics." Medical Device and Diagnostic Industry, pp. 35-38, Feb. 1983. 357. Volkmann, T., and Braunschweig, H. W. "Oxiftuorination of Polyethylene," KunststofJe, pp. 51-52, Aug. 1989. 358. Schaper, E. "Computer-Aided Barrier Design Yields Retortable Food Package." Proceedings, SPE International Coextrusion Conference 1989; also in Modern Plastics, pp. 94-99, Dec. 1989. 359. "New Polycarbonate Stands Up to Gamma Radiation." Plastics Today, p. 8, Feb. 18, 1989. 360. "New Acetal Fights Sun's Attack." Plastics Today, p. 10, July 22, 1989. 361. Nelson, G. L. "Flammability: Understanding Requirements and Test Procedures." Proceedings, SPI Structural Plastics Div. Annual Conference, 1989. 362. Wilson, E. L. "Flammability & High Temperature Characteristics of Composites. Flame Retardance of Polymer Materials, Vol. 3. New York: Marcel Dekker, 1975. 363. Warren, L. M. "Combustion Toxicity: The Burning Issue." Plastics Technology, pp. 54-63, July 1988. 364. Mark, H. F. "Combustion ofPoymers & Its Retardation." Proceedings, National Symposium on Fire Safety Aspects of Polymeric Materials. Washington, DC: Carnegie Institute, June 68, 1977.

948 REFERENCES

365. Hilado, C. J. "Flammability Handbook for Plastics." Fire & Flammability Series, 2 ed. Technomic, 1974. 366. "Fire Safety Aspects of Polymeric Materials." In Vol. 1: Afaterials-State of the Art. Washington, DC: National Materials Advisory Board Report Pub. NMAB 318-1, National Academy of Sciences, 1977. 367. _ _ . Vol. 3: Smoke & Toxicity. Washington, DC: National Materials Advisory Board Report Pub. NMAB 318-1, National Academy of Sciences, 1977. 368. Hilado, C. J., Ed. "Flammability of Cellular Plastics. " In Vol. 8: Fire & Flammability Series, Technomic, 1974. 369. "Life Safety Code." Boston: National Fire Protection Assoc., 1976. 370. Nelson, G. L. "Plastics Flammability." National Symposium on Fire Safety Aspects of Polymeric Materials. Washington, DC: Carnegie Institute, June 6-8, 1977. 371. "Standard Test Method for Measuring the Smoke Generated by Solid Materials." NFPA 258. Boston: National Fire Protection Assoc., 1976. 372. Miller, B. "Fire Code Standard." Plastics World, p. 31, Nov. 1989. 373. "UV-Stable Acetal Copolymer Slated for Auto Use." Plastics Design Forum, pp. 74-78, July-Aug. 1989. 374. Benjamin, B. S. Structural Design with Plastics. New York: Van Nostrand Reinhold, 1982. 375. Mehta, K. S. "Identifying and Correcting Part Design Problems." Plastics Design Forum, pp. 35-46, Nov.-Dec. 1986. 376. Bernhardt, E. C. "Cavity Dimensioning Using Computerized Shrinkage Evaluation Software." SPE ANTEC, pp. 1262-64, May 1989. 377. Paro, T., et al. "Eliminating Warp in Molded Plastics." Machine Design, pp. 69-73, Aug. 1986. 378. Liu, C. F., and Armeniades, C. D. "Cure Shrinkage Control of Polymerization Systems at Ambient Temperatures." SPE ANTEC, pp. 834-37, May 1988. 379. Iwakura, K. "A Basic Study of Warpage and Heat Transfer in Rotational Molding." SPE ANTEC, pp. 558-62, May 1989. 380. Menges, G., and Hoven-Nievelstein, W. B. "Studies on the Shrinkage of Thermoplastics." SPE ANTEC, 1983. 381. Desai, C. S., and Abel, J. F. Introduction to the Finite Element Method. New York: Van Nostrand Reinhold, 1972. 382. Matsco, M. M. "Nonlinear Finite Element Analysisin Plastic Design." SPE ANTEC, pp. 651-54, May 1989. 383. Haessly, W. P., and Ryan, M. E. "Finite Element Modeling of the Injection Blow Molding Process." SPE ANTEC, pp. 934-37, May 1989. 384. Taylor, C. A. "Blow Molding-Process Characterization and Finite Element Modeling." SPE ANTEC, pp. 931-33, May 1989. 385. Schoumacher, B. Engineers and the Law. New York: Van Nostrand Reinhold, 1986. 386. "Hazardous Materials" [applicable to designs]. OSHA Gov't Regulation PB89-210199/AS. Washington, DC: National Technical Information Services. 387. Canning almanac-reviews: FDA law/stds.; statistics/food packaging; labeling specs; USDA quality grades, agricultural data; imports/exports, etc. Westminster, MD: E. E. Judge & Sons. 388. "Packaging Pharmaceuticals--Obtaining Regulatory Clearance for Plastics." Publication No. AL-120. SPI, 1989. 389. "Polyimide Material Replacement Gives Longer Life-Pressure Controls." Plastics Design Forum, Mar.-Apr. 1989. 390. Mark-Bikales-Overberger-Menges. Encyclopedia of Polymer Science & Engineering. 2 ed., 15 vols., 1987. 391. Haut, D. "Success by Design." Medical Device & Diagnostic Industry, pp. 48-60, Sept. 1988. 392. Felton, L. P. "Structural Index Method in Optimum Design." Proceedings, ASME annual meeting, (winter) New York, 1974.

REFERENCES 949

393. Stefanides, E. J. "Bent Axis Design-Grooving-Improve Conveyor Belt Seal." Design News, pp. 116-17, Nov. 1989. 394. Goldfein, G. "General Formula for Creep & Rupture Stresses in Plastics." Modern Plastics, Apr. 1960. 395. Kraus, H. "Creep Analysis." New York: Wiley, 1980. 396. Janzen, W., and Ehrenstein, G. W. "Hysteresis Measurements for Characterizing the Cyclic Strain & Stress Sensitivity of Glass Fiber Reinforced PBT." SPE-ANTEC, pp. 659-63, May 1989. 397. Orthwein, W. C. "Adding Ribs for Maximum Strength." Machine Design, pp. 113-16, Feb. 1981. 398. Redfam, C. A. A Guide to Plastics. London: I1iffe, 1958. 399. Timoshenko, S. P. History of Strengths of Materials. New York: McGraw-Hili, 1953. 400. Wendle, B. C. Engineering Guide to Structural Foam. Technomic, 1976. 401. Schwartz, V. V.,. et al. Illustrated Dictionary of Mechanical Engineering. Hingham, MA: Kluwer Boston, 1984. 402. Crandall, S. H., et al. An Introduction to the Mechanics of Solids. New York, McGrawHill, 1978. 403. Marks, L. S. Marks Handbook of Mechanical Engineering. New York: McGraw-Hili, 1951. 404. Diehl, G. M. "Stress Analysis & Mohr's Circle of Stress." Compressed Air Magazine, pp. 20-27, Aug. 1983. 405. "Advanced Composite Design Guide." MIL-HDBK-5C. Washington, DC: GPO, 1980. 406. William N. Touz, Correspondence, Collapsible Bottle of America, June 12, 1986. 407. Sheridan, T. W. "The Role of Imagination in Expanding Markets." Boston: SPE RETEC, Oct. 27-28, 1989. 408. Hoechst Celanese Corp. "Designing with Acetal-Celcon." Manual, 1985. 409. "A Design Manual: Snap-Fit Joints in Plastics." Mobay Engineering Thermoplastics Bulletin. Mobay Chemical Corp., 1990. 410. Hoechst Celanese. "Design & Production of Gears in CeIcon Acetal Copolymer." Manual, 1984. 411. Du Pont. "Designing with F1uoropolymers." Manual, 1981. 412. Lilley, D. T. "Grommets Tame Noise Polution." Design News, pp. 100-103, Nov. 1980. 413. Bucksbee, J. H. "The Use of Bonded Elastomers for Energy & Motion Control in Construction." Erie, PA: Lord Corp., Industrial Products Div., 1988. 414. Tuschak, P. A. "Designing for Vibration." Du Pont Engineering Design Magazine, pp. lO11, no. 2, 1989. 415. Gildenston, R. F., and Bucksbee, J. H. "How Vibration Design Requirements for Direct Fixation Fasteners Affect Procurement Practices." Report LL-7200. Erie, PA: Lord Corp., 1989. 416. Schmueser, D. W., and Johnson, N. L. "Structural Design, Fabrication & Vibrational Response of a Composite Auto Space Frame." GM Report GMA-6086 and SAE VII International Conference in Vehicle Structural Mechanics. Detroit: Apr. 11-13, 1988. 417. :'Architectural Shock Absorbers Reduce Damaging Effects of Earthquake Motion." Compressed Air, pp. 26-33, June 1988. 418. Rasmussen, S. "Shrink Data & Mold Error." Camas, Wash.: Hewlett-Packard, 1989. 419. Janover, J. S. and Sebring, J. F. "Finite Element Analysis of Plastics Gear Transmission." Plastics Design Forum, pp. 69-81, Nov.-Dec. 1989. 420. Debski, D. "Prototyping on Your Computer-FEA Products." Du Pont Engineering Design Magazine, pp. 14-15, Fall 1986. 421. Weckman, R. L. "Diagonal Ribs Increase Torsional Rigidity." Du Pont Engineering Design Magazine, pp. 14-15, Summer 1988. 422. Eiden, G., and Hess, J. "CAD/CAE Dimensioning of a Bumper Made of PP Elastomer Blend." Kunststoffe, pp. 38-39, July 1989.

950 REFERENCES

423. "Fiberglas & Design-Designing." Publications 5-PL-13920 and 5-PL-13236. Owens-Coming Fiberglas Corp., 1986. 424. Tartolano, F. W. "New Plastics Products." Design News, pp. 55-62, May 1989. 425. "FEA Solves Critical Packaging Problem." Design News, p. 36, Sept. 1989. 426. Mullens, B. "PEA: Predicting vs. Actual Performance." Plastics Design Forum, pp. 71-76, Mar./Apr. 1989. 427. "FEA-B1ow Molding." Plastics Engineering, p. 33, Apr. 1989. 428. Solecki, 1. S. "Optimizing a Truss Bridge." Machine Design, pp. 97-98, Mar. 1988. 429. Rebelo, N. "FEA of Forming." Machine Design, pp. 121-23, June 1988. 430. Klein, A. J. "How to Design With PEA." Plastics Design Forum, pp. 29-37, Sept.-Oct. 1987. 431. Rieger, N. F., and Steele, J. M. "The Basics of PEA Moldeling." Machine Design, pp. 16570, Apr. 1981. 432. Blair, J. A. "Codes and Standards Update." Plastics Today, p. 20, Jan. 7, 1989, and p. 58, Sept.-Oct. 1989. 433. Sherwink, K. Engineer Design for Performance. New York: Wiley, 1982. 434. "Materials Reference Guide." Machine Design, Annual. 435. "Materials Engineering 1980 Materials Selector." PentonlIPC, Dec. 1979. 436. Thermoplastic and Thermoset Plastics. 9 ed. San Diego: D.A.T.A., Inc., 1987. 437. Goodman, S. H. Handbook of Thermoset Plastics. Park Ridge, N.J.: Noyes Pub., 1986. 438. "Performance Temperature." Durez Molder Newsbriefs. Occidental Chern. Corp., winter 1989. 439. Guide to Materials Engineering Data and Information. Metals Park, OH: ASM International, 1986. 440. Wiebaden, W. S. "Improving the Properties of Plastics Film by Orientation." KunststoJfe, pp. 52-54, Oct. 1988. 441. Kraut, G. P. Statics and Strength of Materials. Reston, VA: Wheeler, 1984. 442. Seymour, R. B. Engineering Polymer Source. New York: McGraw-Hili, 1989. 443. _ _ . Polymers for Engineering Applications. Metals Park, OH: ASM International, 1987. 444. Source Book of Materials. Vol. 2, Sec. XV: Plastics. Metals Park, OH: ASM International, 1977. 445. Mascia, L. Thermoplastics: Materials Engineering. New York: Elsevier, 1980. 446. Hall, C. Polymer Materials: An Introduction for Technologists and Scientists. New York: Macmillan, 1981. 447. Munstedt, H., et al. "Polyaryletherketones-New Changes for Thermoplastics." KunststofJe, pp. 75-76, Oct. 1989. 448. Thomas, D. A. Plastics Polymer. Vol. 37, p. 485, 1969. 449. Houwink, R. "Synthetics." Proceedings, World Chemical Engineering Congress Meeting, Mexico City, Nov. 1965. 450. Hoechst Celanese Corp. Technical Literature. Annual updates. 451. Brown, D. W., and Harris, W. T. An Introduction to Engineering Plastics. Murray Hill, NJ: 1947. 452. Culbertson, B., and Pittman, C. New Polymers. New York: Plenum, 1984. 453. "Amorphous Fiuoropolymer." Plastics Today, pp. 1,4, July 22, 1989. 454. "New Nylon 6/6 Withstands Stress of Everyday Use." Plastics Design Forum, p. 74, Sept.Oct. 1989. 455. Burns, R. "Nylon for the Nineties." Plastics Today, pp. 60-65, Feb. 1988. 456. Crate, J. H. "50 Years Later; Nylon Remains at the Forefront of Engineering Design." Plastics Design Forum, pp. 113-16, May-June 1988. 457. "Nylon 4/6." Modern Plastics, pp. 34-35, June 1988. 458. Engineering with Rigid PVC: Processability and Applications. New York: Marcel Dekker, 1984. 459. Titow, W. V. PVC Technology. 4 ed. New York: Elsevier, 1984.

REFERENCES 951

460. Iwen, E. D. Degradation and Stabilization of pvc. New York: Elsevier, 1984. 461. Gorham, W. F. "Parylene Polymers." In Modern Plastics Encyclopedia, pp. 230-34. New York: McGraw-Hill, 1969. 462. Oertel, G. Polyurethane Handbook. Munich: Hanser, 1985. 463. "Reinforced PEEK Launches Rocket Cylinder." Design News, p. 35, July 1989. 464. Kohlhepp, K. "Properties and Applications of a New Linear PPS." Kunststojfe, pp. 27-29, July 1989. 465. Lo, L. Y., and Boulier, P. R. "A New Styrenic/Polycarbonate Alloy Family of Engineering Resins." SPE ANTEC, pp. 707-9, May 1989. 466. Klein, A. J. "High Temperature Imides." Advanced Composites, pp. 45-57, July-Aug. 1988. 467. _ _ . "Liquid Crystal Polymers Gain Momentum." Plastics Design Forum, pp. 49-56, Jan.-Feb. 1989. 468. Chapoy, L. Recent Advances in Liquid Crystal Polymers. New York: Elsevier, 1985. 469. Walker, B. M. Handbook of Thermoplastic Elastomers. New York: Van Nostrand Reinhold, 1979. 470. Klein, A. J. ''TPE's: An Expanding List of Types and Grades." Plastics Design Forum, pp. 31-44, Mar.-Apr. 1989. 471. Legge, N. R., and Holden, G. Thermoplastic Elastomers. Munich: Hanser, 1987. 472. Harper, C. A. Handbook of Plastics and Elastomers. New York: McGraw-Hill, 1975. 473. Heineck, D. W. Blow Molded Thermoplastic Elastomers. SPE Blow Molding Div., spring 1988. 474. Heineck, D. W., and Rader, C. P. "TPE's: Economical Rubber Products for the Plastics Processor." Plastics Engineering, pp. 87-91, Mar. 1989. 475. Burns, R. "Thermoplastic Elastomers: A New Picture Emerges." Plastics Technology, pp. 44-51, Aug. 1988. 476. "New Thermoplastic Elastomer Ready for Testing." Machine Design, p. 6, Apr. 1988. 477. "Comparison of Natural Rubber and Synthetic Rubbers." Natural Rubber Technology. Hertford, England: Malaysian Rubber Producers' Research Assoc., vol. 19, no. 1, pp. 14-18, 1988. 478. Margolis, J. M. Engineering Thermoplastics: Properties and Applications. New York: Marcel Dekker, 1985. 479. Sommer, J. G. "Molding of Rubber and the Design of Rubber Molds." Proceedings, Seminar, Univ. of Wisconsin-Milwaukee, Nov. 29-Dec. 1, 1989. 480. Hofmann, W. Rubber Technology Handbook. Munich: Hanser, 1989. 481. Bacchetti, J. "Thermoplastic Elastomers Gain Ground in Medical Applications." Plastic Trends, pp. 16-18, 1989. 482. "High Performance Plastic Films, 1989-1994." A. D. Little Report, Cambridge, MA, 1989. 483. Rosato, D. V. "Overview of Composites." In Lubin, G., Handbook of Composites. New York: Van Nostrand Reinhold, 1982, pp. 1-18. 484. Richardson, T. Composites: A Design Guide. Metals Park, OH: ASM International, 1987. 485. Composites Annual Conference Preprint Books. SPI, annually since 1944. 486. Wilson, D., et al. Composites Design Guide. Vol. 3: Processing and Fabrication Technology. Univ. of Delaware, 1980. 487. Strong, A. B. Fundamentals of Composites Manufacturing Materials, Methods and Applications. Metals Park, OH: ASM International, 1989. 488. Brown, G. "Fiber Glass Scores with Designers." Design News, pp. 258-61, Nov. 1988. 489. Sih, G. C., and Hsu, S. E. "Advanced Composite Materials and Structures." In Dorgham and Rosato, Designing with Plastics Composites. Geneva: Interscience Enterprises, 1986. 490. Schwartz, M. M. Fabrication of Composite Materials Source Book. Metals Park, OH: ASM International, 1985. 491. _ _ . Composite Materials Handbook. Metals Park, OH: ASM International, 1983. 492. Schwartz, R. T., and Schwartz, H. S. Fundamental Aspects of Fiber Reinforced Plastics. Cahners, 1968.

952 REFERENCES

493. Gaylord, M. W. Reinforced Plastics Theory and Practice. Newton, MA: Cahners, 1974. 494. Dietz, A.G.H. Composite Engineering Laminates. Cambridge, MA: MIT, 1969. 495. Agarwal, B. D., and Broutman, L. 1. Analysis and Performance of Fiber Composites. Reading, MA: Addison-Wesley, 1980. 496. Carlsson, L. A., and Pipes, R. P. Experimental Characterization of Advanced Composites Materials. New York: Prentice Hall, 1987. 497. Boutman, L. J., and Krock, R. H. Modern Composites Materials. Reading, MA: AddisonWesley, 1967. 498. Fiberglas Plus Design: A Comparison ofMaterials and Processesfor Fiber Glass Composites. Owens-Coming Corp., July 1985. 499. Rogers, J. K., and Gabriele, M. C. "Composite Resins, Fibers Machinery Debut at SAMPE Conference." Plastics Technology, pp. 25-31, July 1989. 500. Rossa, E. P. "Reinforced Fibers Give Phenolics High-Tech Uses." Plastics Engineering, pp. 39-41, Nov. 1988. 501. Sanders, B. "Short Fiber Reinforced Composite Materials." STP 772. ASTM, 1982. 502. Ward, S., and Bailey, R. "Low Carbon Composites Offer Stiff Competition." Machine Design, pp. 52-58, Feb. 1989. 503. Sutker, B. J. "Hollow Spheres for Engineering Thermoplastics." Plastics Compounding, pp. 61-66, Sept.-Oct. 1989. 504. Adams, D. F., et al. "Properties of Polymer-Matrix Composite Incorporating Allied Spectra 900 Polyethylene Fiber." Proceedings, SAMPE National Conference, Mar. 1985. 505. Ward, S. "Advances in Reinforced Thermoplastic Composites." Boston: SPE RETEC, Oct. 1988. 506. Strassel, A. "PVDF-ABS Composites: Properties and Applications." Kunststoffe, pp. 20-22, Sept. 1988. 507. "New Polyamide Makes Tougher High-Temperature Composites." Plastics Today, p. 7, Feb. 18, 1989. 508. "Polyimide Compounds." Advanced Composites, pp. 45-50, July-Aug. 1988. 509. Van Krevelen, D. W. Properties of Polymers, 2 ed. New York: Elsevier, 1976. 510. Ifwarson, M. "Life-Time of Polyethylene Pipes Under Pressure and Exposure of High Temperatures." Kunststoffe, pp. 20-22, June 1989. 511. Tortolano, F. W. "Rising Star in the Plastics Lineup Le., Design News, pp. 54-62, Nov. 1989. 512. "Springy Finite Element Model Elastomers." Machine Design, May 1986. 513. "Foam Makers Head Toward CFC-Free Urethanes." Plastics World, pp. 12-14, Nov. 1989. 514. Wilson, D. "Understanding Foam Density Crucial for Best Cushion." Packaging, pp. 9192, May 1987. 515. Murray, C. J. "Optical Film Transports Light" [Pipe LightJ. Design News, pp. 354-55, Nov. 1988. 516. Netolicky, W. "Processing Clear Resins." Plastics Today, pp. 7-9, Sept. 2, 1989. 517. Yeaple, F. "Aramid Fibers Reinforce Lightweight Plastic Sails." Design News, p. 128, Nov. 1989. 518. Rabinovitch, E. B., and Booth, P. C. "Use of Regrind in the PVC Extrusion Process." SPE ANTEC, pp. 761-66, May 1989. 519. Plastics. International Plastics Selector. San Diego: Cordura Pub., 1990. 520. Bower, H. E. Selection of Materials for Component Design-Source Book. ASM International, 1986. 521. MacDermott, C. Selecting Thermoplastics for Engineering Applications. New York: Marcel Dekker, 1984. 522. Stumpek, E. S. "A Resin Supplier's Approach to Resolving Plastics Problems." SPE ANTEC, pp. 1616-19, May 1989. 523. ASCE. Structural Plastics Selection Manual. No. 66, 1984.

REFERENCES 953

524. Kusy, P. F. Plastics Materials Selection Guide. No. 760663. Society of Automotive Engineers, Sept. 1976. 525. Budinski, K. G. Engineering Materials Properties and Selection. Reston, VA: 1979. 526. Brady, G. S., and Clauser, H. R. Materials Handbook. New York: McGraw-Hill, 1985. 527. Trade Name Directory. MP Encyclopedia, 1989. 528. Oberbach, K. "Fundamental Data Tables & Database (CAMPUS)-A Challenge & Opportunity." KunststoJfe, pp. 34-38, Aug. 1988. 529. Schmitz, J. "Kunststoffe-Datenbank CAMPUS, Einheitliche Software der Groben Vier." Plastverarbeiter, pp. 50-58, Apr. 1988. 530. Diekmann, D. "Computerized Databases for Plastics Materials." Plastics Engineering, pp. 29-37, Nov. 1989. 531. Herrlich, N., and Wierer, J. "Delegate the Material Selection to the Database." KunststojJe, pp. 106-9, Oct. 1989. 532. Engineered Materials Abstracts-Data Base. Metals Park, OH: ASM International, annual. 533. Datapro Directory of Software-Mainframe. New York: McGraw-Hill, updated annually. 534. The Software Eneyclopedia-Guide to Microcomputer Software. New York: R. R. Bowker, 1990. 535. "Computer Literature." Design News, pp. 74-75, Sept. 1989. 536. Kennedy, J. S., et al. "CAMPUS 2-Practical Applications." SPE ANTEC, May 8, 1990. 537. Grandilli, P. A. Technician's Handbook of Processing. New York: Van Nostrand Reinhold, 1981. 538. Pearson, J. R. Mechanical Polymer Processing. New York: Elsevier, 1985. 539. Manufacturing Handbook and Buyers Guide. Plastics Technology, annual. 540. Wright, R. E. Thermoset Molding Manual. Manchester, CT: Rogers Corp., 1984. 541. Harper, C. A. Handbook of Materials and Processes for Electronics. New York: McGrawHill, 1970. 542. Throne, J. L. Plastics Process Engineering. New York: McGraw-Hill, 1977. 543. Middleman, S. S. Fundamentals of Polymer Processing. New York: McGraw-Hill, 1977. 544. McKelvey, J. M. Polymer Processing. New York: Wiley, 1962. 545. "Hotbed of New Process Technology for Manufacturing Large Parts." Plastics Technology, pp. 15-17, Aug. 1989. 546. Trantina, G. G., and Ysseldyke, D. A. "An Engineering Design System for Thermoplastics." SPE ANTEC, pp. 635-39, May 1989. 547. "Redesign Calls for PPS; Insert Molding." Plastics Today, p. 17, June 24, 1989. 548. "Reinforced PVCs Can Be Extruded." Plastics World, p. 72, Apr. 1987. 549. Directory of Polyurethane Processors (Also Material & Equipment Suppliers). Glen Ellyn, IL: Polyurethane Mfgrs. Assoc., 1989. 550. Owen, E. D. Degradation and Stabilization of pvc. New York: Elsevier, 1984. 551. Stoeckhert, K. Mold Making Handbook. Munich: Hanser, 1983. 552. Sors, L.; Bardocz, L.; and Radnoti, I. Plastics Molds and Dies. New York: Van Nostrand Reinhold, 1981. 553. Rao, N. S. Designing Machines and Diesfor Polymer Processing with the Computer. Munich: Hanser, 1981. 554. Stanley, G., and WeIch, C. The Molding of an Industry. British Plastics Federation, 1983. 555. Brall, J. G. Handbook of Product Design for Manufacturing: A Practical Guide to Low-Cost Production. New York: McGraw-Hill, 1986. 556. "Compounding Lines." lmagineering News. RTP Co., Vol. 2, no. 1, June 1987. 557. Design Handbook for Du Pont Zytel Nylon Resin. Du Pont, 1988. 558. Moore, G. R., and Kline:D. E. Properties and Processing of Polymers for Engineers. New York: Prentice Hall, 1984, pp. 175-80. 559. Boller, W. A. "Specifying the Optimum Number of Cavities." Modern Plastics, p. 74, Nov. 1974.

954 REFERENCES 560. Miller, B. "Predicting Part Shrinkage Is a Three-Way Street." Plastics World, pp. 47-52, Dec. 1989. 561. Bikales, M. Extrusion and Other Plastics Operation. New York: Wiley, 1971. 562. "High-Tech Applications Spur Metallizer's Growth." Paper, Film & Foil Converter, pp. 5657, Apr. 1988. 563. Rogers, J. K., et al. "Design Meets Manufacturing." Plastics Technology, pp. 44-55, Nov. 1988. 564. Paulson, D. C. "Plastics Part Design for Injection Molding." Videocassette Education Course, Paulson Seminars Program, 1982. 565. "PLASPEC: Manufacturing Handbook & Buyer's Guide: Material & Machinery." Plastics Technology, annual. 566. Dowler, C. A. "Automatic Model Building Cuts Design Time, Costs." Plastics Engineering, pp. 43-45, Apr. 1989. 567. "Making Prototypes or Models in Hours." Plastics Design Forum, p. 14, July-Aug. 1987. 568. Gorjup, D. M. "Stereo Lithography; System for Rapid Turnaround of Plastics Models." Boston: SPE RETEC, Oct. 27-28, 1989. 569. Beal, G. L. "A Listing of Mold Making and Design Teaching Institutions." SPE, 1989. 570. Du Bois, J. H., and Pribble, W. I. Plastics Mold Engineering Handbook. New York: Van Nostrand Reinhold, 1978. 571. Menges, G., and Mohren, P. How to Make Injection Molds. Munich: Hanser, 1986. 572. Pye, R.C.W. Injection Mold Design. London: Plastics & Rubber Institute, 1980. 573. Colby, P. "Screw and Barrel Technology." Bulletin. Youngstown, OH: Spirex Corp., 1988. 574. "Everything You Ever Wanted to Know about Electric Heaters and Control Systems." St. Louis: Watlow Hq., annual. 575. Shigley, J. E., and Mischke, C. R. Standard Handbook 0/ Machine Design. New York: McGraw-Hill, 1986. 576. Orberge, F., and Jones, A. Machinery Handbook. Industrial Press, 1981; Deutschman, A. D., et al. Machine Design: Theory and Practice. New York: Macmillan, 1975. 577. Dym, J. B. Injection Molds and Molding. New York: Van Nostrand Reinhold, 1979. 578. Hull, J. L. Injection Molding Thermoplastics and Thermosets. Hatboro, PA: Hull Corp., 1986. 579. Manzione, L. T. Applications o/Computer-Aided Engineering in Injection Molding. Munich: Hanser, 1987. 580. "Injection Molding Operations." Bolton, Quebec, Canada: Husky 1M Systems Ltd., 1980. 581. Rubin, I. I. Injection Molding Theory and Practice. New York: Wiley, 1972. 582. Brown, J. Injection Molding 0/ Plastic Components. New York: McGraw-Hill, 1979. 583. Hume, W. J. "Injection Molding Rigid PVC-Hot Runners Recommended." SPE ANTEC, pp. 767-69, May 1989. 584. Beaumont, J. P. "Use of Flow Analysis for Predicting Flash and Clamp Tonnage" [in Thermosets]. SPE ANTEC, pp. 813-16, May 1989. 585. Olmsted, B. A. "Solving the Shot-to-Shot Variation Dilemma" [in 1M]. Plastics Machinery & Equipment, p. 38, Dec. 1987. 586. Austin, C. "Modelling the Effects of Filling and Loading History on the Structural Performance of Injection Molded Products." SPE ANTEC, pp. 619-20, May 1989. 587. "Stereolithography: Liquid Gold to U.S. Auto Makers." Plastics News, pp. 1,20; Nov. 13, 1989. 588. "Same Day Prototyping Blossoms." Plastics World, pp. 11-13, Dec. 1989. 589. Bay, R. S., et al. "Effect of Processing on Fiber Orientation in Simple Injection Moldings." SPE ANTEC, pp. 539-42, May 1989. 590. Taggert, D. G., and Pipes, R. B. "Processing Induced Fiber Orientation in Transfer and Injection Molding." Univ. of Delaware, 1979. 591. Denault, J., and Vu-Khauh, T. ''Tensile Behavior of Long Fiber Thermoplastic Composites Processed by Compression and Injection Molding." SPE ANTEC, pp. 891-95, May 1989.

REFERENCES 955

592. Meyer, W. "Coinjection Can Benefit Automotive Parts Manufacturers." Plastics Today, p. 21, Sept. 2, 1989. 593. Koch, P. E., et al. "Design Constraints of Injection Molding Foaming Thermoplastics Using Gas Counter-Pressure Molds." SPE ANTEC, pp. 231-34, May 1989. 594. Rezayat, M. "Numerical Computation of Cooling-Induced Residual Stress and Deformed Shapes for Injection Molded Thermoplastics." SPE ANTEC, pp. 341-43, May 1989. 595. _ _ , and Burton, T. E. "Combined Boundary-Element and Finite-Difference Simulation of Cooling & Solidification in Injection Molding." Proceedings, Fifth International Conference on Numerical Methods in Thermal Problems, Montreal, July 1987. C. Taylor, et al., Eds. 596. "Composite Auto Wheels Roll Out at Last." Modern Plastics, p. 18, Dec. 1989. 597. Galli, E. "Designing for Structural RIM." Plastics Design Forum. pp. 53-57, Nov.-Dec. 1989. 598. Hebel, T. E. "Vibrating Steel to Relieve Residual Stresses." Plastics Machinery & Equipment. pp. 43-44, Jan. 1990. 599. "Processing Opportunities." Plastics Machinery & Equipment, pp. 29-41, Jan. 1990. 600. Wehrenberg, R. H. "Plating Plastic in the Mold." Manufacturing Engineering. Nov. 1982. 601. Schott, N. R. Proceedings, Plastics Extrusion Processing Technology Seminar. Univ. of Lowell [MA], 1989. 602. Hensen, F. Plastics Extrusion Technology. Munich: Hanser, 1988. 603. Rauwendaal, C. Polymer Extrusion. Munich: Hanser, 1986. 604. Levy, S. Plastics Extrusion Technology. Industrial Press, 1981. 605. Nichols, R. J., and Kruder, G. A. "Extrusion." SPE ANTEC, May 1974. 606. Richardson, P. N. Introduction to Extrusion. SPE, 1974. 607. Tadmor, Z., and Klein, J. Engineering Principles of Plasticating Extrusion. New York: Van Nostrand Reinhold, 1970. 608. Schenkel, G. Plastics Extrusion Technology and Theory. New York: Elsevier, 1966. 609. Fisher, E. G. Extrusion of Plastics. London: Plastics & Rubber Institute, 1976. 610. Duska, J. J., and Kruder, G. A. "Extrusion Developments." SPE ANTEC, May 1976. 611. Bikales, M. Extrusion and Other Plastics Operations. New York: Wiley, 1971. 612. McKelvey, J. M. "Extrusion." SPE ANTEC, May 1978. 613. Kruder, G. A. "Venting: Critical for Quality Extrusion." Plastics Machinery & Equipment, p. 2, May 1982. 614. Janssen, L. P. Twin Screw Extrusion. New York: Elsevier, 1978. 615. Gomez, I. L. Engineering with Rigid PVC: Processing and Applications. New York: Marcel Dekker, 1984. 616. Michaeli, W. Extrusion Dies: Design and Engineering Computations. Munich: Hanser, 1984. 617. Keller, R. "Comparing the Cast Film and Blown Film Processes." Plastics Engineering, pp. 31-34, Aug. 1989. 618. Leonard, L. V. "Engineering Films Stretch the Imagination." Plastics Design Forum. pp. 17-22, Sept.-Oct. 1989. 619. Hensen, F., and Braun, S. "Manufacture of Film Tapes." Industrial & Production Engineering. Mar. 1978. 620. Keller, R. "Guidelines for Comparing the Blown and Cast Film Processes." SPE ANTEC, pp. 39-43, May 1989. 621. Chan, T.W.D., and Lee, L. J. "Orientation and Residual Stress Distribution in Plastic Sheet Extrusion." SPE ANTEC, pp. 31-34, May 1989. 622. Smith, D. J. Proceedings. "Blown Stretch Film Processing Parameters." pp. 163-65. Technical Assoc. of Pulp & Paper Industry Laminations and Coatings Conference, 1987. 623. Cramm, R. H., and Sibbach, W. R. Coextrusion Coating and Film Fabrication. TAPPI, 1983. 624. Drennan, W. C. "Tie-Layer Resins: More Than Ties to 'Bind.' " Plastics Packaging. pp. 32-35, Sept.-Oct. 1988.

956 REFERENCES

625. Martelli, F. "Calculating Extruder Energy Efficiency." Plastics Compounding, Apr.-May 1980. 626. Galli, E. "Designed for Blow Molding." Plastics Design Forum, pp. 21-28, Jan.-Feb. 1989. 627. King-Liestman, C. "Structural Blow Molding Techniques Require a New Pinch." Plastics Machinery & Equipment, pp. 30-32, July 1989. 628. Miller, J. C. "The Influence of Screw Wear on Blow Molding Processability." Plastics Engineering, pp. 37-41, Oct. 1989. 629. Menges, G., etal. "Optimization of the Cooling Process in Blow Molding." Kunststojfe, pp. 36-38, Jan. 1989. 630. Stoeckhert, I. K. "Evaluation of Interior Cooling of Blow Molding Products." Industrial & Production Engineering, Feb. 1981. 631. Farrel, T. J. "Intricate Fuel Tank Graces 'Car of the Year.' " Plastics Today, p. 19, Mar. 18, 1989. 632. Kurella, E. "HDPE Fuel Tanks Gain; Process Options Expand." Plastics Today, p. 5, Mar. 18, 1989. 633. "Progress in Blow Molded Auto Fuel Tanks Is Tracked in Reports at SAE Conference." Modern Plastics, p. 31, 1989. 634. "Big-Part Blow Molding Invades Big Markets." Piastics World, pp. 65-68, Oct. 1985. 635. Goldsworthy, W. B. "Composite Fabrication." In Encyclopedia ojComposites, S. Lee, Ed. New York: VCH, 1990. 636. Owens-Coming Fiberglas. ''The Hand Lay-Up Spray-Up Guide." Grandville, OH: OwensComing. 637. Mahlke, M., et al. "New Structures to Preheat Glass Mat Reinforced Thennoplastics" [GMT]. SPE ANTEC, pp. 873-75, May 1989. 638. "Development of an Automated Lay-Up System for Monofilament Fiber Structures." Report per Contract No. DAAJO 1-70-C-0049 (PIG) by Boeing Vertol Co. for Anny Aviation Systems Command, St. Louis, 1977. 639. Castro, J. M., and Tomlinson, G. "Predicting Molding Forces in SMC Compression Molding." SPE ANTEC, pp. 583-88, May 1989. 640. Lam, R. C., and Kardos, J. L. "The Penneability and Compressibility of Aligned and CrossPlied Carbon Fiber Beads During Processing of Composites." SPE ANTEC, pp. 1408-12, May 1989. 641. Molnar, J. A., et al. "Mold Filling in SRIM and RTM: Controlling a Critical Processing Parameter." Modern Plastics, pp. 120-26, Sept. 1989. 642. Michaeli, W., et al. "The RTM Process-Technology and Areas of Applications." Kunststojfe, pp. 17-19, July 1989. 643. Brownbill, D. "What Is New in RP, PUR: New Equipment & Improved Fibers." Modern Plastics International, p. 41, Oct. 1981. 644. "Experimental Polycarbonate Pultrudes Like a Thennoset." Modern Plastics, pp. 13-14, Sept. 1988. 645. Han, C. D., et al. "Development of a Mathematical Model for the Pultrusion Process." Polymer Engineering Science. Vol. 26, No.1. New York: Wiley, 1986. 646. Sanjana, Z. N., and Testa, J. H. "Toughened Epoxy Resins for Filament Winding." Symposium, SAMPE, Mar. 1985. 647. Morgan, R. J., et al. "Development of Epoxy Matrices for Filament Wound Graphite Structures." SAMPE, pp. 1209-19, Mar. 1985. 648. Wackerle, P. M., et al. "Applications in Filament Winding." Industrial & Production Engineering, pp. 37-45, Oct. 1986. 649. "Design, Tooling and Manufacturing with Technopolymer Structures from Azdel Inc." GE Plastics, Productivity Workbook, Series 872. 650. Bigg, D. M., et al. "High-Strength Stampable Thennoplastic Composites." Plastics Engineering, pp. 51-54, Mar. 1988.

REFERENCES 957

651. McConnell, W. K. "Fundamentals of Thermoforming." Plastics Trends, pp. 18, 29, JulyAug. 1989. 652. Throne, J. L. "Thermoforming." Munich: Hanser, 1987. 653. Muzzy, J. D., et al. "Thermoforming of High Performance Thermoplastic Composites." SPE ANTEC, pp. 1465-70, May 1989. 654. Rogers, J. K. "Engineered Thermoforming Heats Up." Plastics Technology, pp. 44-47, Aug. 1989. 655. Smoluk, G. R. "Quiet Thermoforming Evolution Begins to Make Its Impact." Modern Plastics, pp. 50-56, Dec. 1988. 656. "Thermoformer Lands Biggest Twin-Sheet Job." Plastics World, pp. 15-16, Jan. 1989. 657. Titus, J. B. "Solid Phase Forming/Cold Forming of Plastics." Plaspec Rep. R-42, Plaspec. Dover, NJ: Picatinny Arsenal, 1976. 658. Murphy, C. "Radiant Heater Panels Improve Thermoforming." Modern Plastics, pp. 78-82, May 1987. 659. Hylton, D. C., et al. "Thermoforming Polypropylene on Conventional Equipment." Plastics Engineering, pp. 55-57, Apr. 1988. 660. Galli, E. "Thermoforming XPLE Foams." Plastics Design Forum, pp. 73-76, Mar.-Apr. 1987. 661. "Europe Moves Towards High Pressure Forming." European Plastics News, p. 10, May 1988. 662. Beal, G. L. "Designers' Guide to Pressure Forming." Plastics Design Forum, pp. 44-49, Sept.-Oct. 1985. 663. McDermott, J. "The Structure of the Advanced Composites Industry." Advanced Composites, pp. 9-22, 1990 Bluebook. 664. Macosko, C. Fundamentals of Reaction Injection Molding. Munich: Hanser, 1988. 665. Sweeney, F. M. Introduction to Reaction Injection Molding. Technomic, 1979. 666. Becker, W. Reaction Injection Molding. New York: Van Nostrand Reinhold, 1978. 667. Chen, J. H., et al. "Influence of Heat Transfer Phenomenon in Nylon Block Copolymer Reaction Molding." SPE-ANTEC, pp. 577-82, May 1989. 668. Baker, T. E., et al. "Characterization of Vapor-Deposited Paraxylene Coatings." Journol of Electrochemical Society, pp. 879-900, June 1977. 669. Longo, F. N. Thermal Spray Coatings: New Materials, Processing and Applications. Metals Park, OH: ASM International, 1985. 670. Dominik, M. "Coating of Web-Like Carrier Materials." KunststoJfe, pp. 82-84, Oct. 1988. 671. Union Carbide Chemical Co. "Parylene Conformal Coatings." Bulletin No. F-43427A, July 1~4. • 672. Shell Chemical Co. "Eponol Resins for Surface Coatings." Bulletin SC: 73-29; "High-Tech Applications Spur Metallizer's Growth." Paper, Film & Foil Converter, pp. 56-57, Apr. 1988. 673. Skiest, I. Radiation Curing. Skeist Labs., NJ, 1984. 674. Margolis, J. M. Decorating Plastics, Munich: Hanser, 1987. 675. Ward, W. "Flocking: Decorative and Useful Fiber Finishes." Plastics Engineering, pp. 3133, July 1987. 676. Heck, W., and Weissach, H. T. "Painting of Plastics in the Outer Skin of the Car Body." KunststoJfe, pp. 84-89, Oct. 1988. 677. Darmstadt, N. V. "Advances in the Electrostatic Flocking of Plastics Moldings." KunststoJfe, pp. 89-90, Oct. 1988. 678. Ballway, B. "Finishing Plastics." Plastics Design Forum, pp. 77-80, Sept.-Oct. 1988. 679. Podhajny, R. M. "Waterborne Inks Proven As Acceptable Alternative." Paper, Film & Foil Converter, pp. 49-53, July 1988. 680. Hanau, G. B. "Vacuum Coating of Plastics." KunststoJfe, pp. 6-7, Sept. 1988. 681. Wark, D. "Electroplated Plastics." Plastics World, p. 108, Feb. 1987.

958 REFERENCES

682. "Fastening, Joining and Assembly Components." Machine Design. Reference issue, Nov. 1988. 683. Allbee, N. "Aspects of Mechanical Fastening Advanced Composites." Advanced Composites, pp. 54-60, Nov.-Dec. 1988. 684. Weston, J. "Aircraft Builders Use More Specialty Fasteners." Machine Design, pp. 121-24, Jan. 1988. 685. Watson, M. N., and Murch, M. G. "Optimizing Hot Plate Welding of Thermoplastics." Plastics Engineering, pp. 47-50, June 1989. 686. Tappe, F., and Potente, H. "New Findings in the Spin Welding of Plastics." SPE ANTEC, pp. 710--15, May 1989. 687. Benatar, A., and Gutowski, T. G. "Ultrasonic Welding of Thermoplastic Composites." SPE ANTEC, pp. 507-10, May 1989. 688. Giessen, E. S. "High Strength FRP-Threaded Joints." KunststoJfe, pp. 32-34, Feb. 1989. 689. Klein, A. J. "Joining Plastics." Plastics Design Forum, pp. 39-52, Sept.-Oct. 1988. 690. Johnson, N. L., and Kardomateas, G. "Structural Adhesive Joints for Applications to a Composite Space Frame: Analysis and Testing." Warren, MI: GM Research Labs., Nov. 19, 1987. 691. Dom, L., et al. "Design of Plastics to Metal Adhesive Joints." KunststoJfe, pp. 3-6, July 1989. 692. "Adhesives." In Packaging Encyclopedia, pp. 33-36, Annual 1987. 693. Adams, R. D., and Wake, W. C. Structural Adhesive Joints in Engineering. New York, Elsevier, 1984. 694. Pimputkar, S. M. "The Dependence of Butt Fusion Bond Strength on Joining Conditions for Polyethylene Pipe." SPE ANTEC, pp. 455-57, May 1989. 695. ASTM Plastics Annual Standards: Vol. 8.01 Test Methods; Vol. 8.02, Materials; and other annual volumes. Philadelphia: ASTM. 696. Industry Standards. Northbrook, IL: Underwriters Laboratories. 697. ISO Plastics Standards. 2 vols. New York: ANSI, 1984. 698. Chamis, C. C. Test Methods and Design Allowables for Fibrous Composites. Vol. 2. STP 1003. ASTM, 1987. 699. Brown, R. P. Handbook ofPlastics TestMethods. London: Plastics & Rubber Institute, 1981. 700. Manual on Presentation of Data and Control Chart Analysis. STP 15D. ASTM, 1980. 701. "Testing the Limits of Plastics." Design News, pp. 152-81, Nov. 1988. 702. Evans, R. "Physical Testing of Plastics." STP 736. ASTM, 1981. 703. Crosby, D. "Quality-The.Causes of Error." Rubber World, pp. 14-15, Dec. 1985. 704. Leonard, L. V. "Compression Testing Results-A Tough Nut to Crack." Advanced Composites, pp. 57-63, July-Aug. 1989. 705. Stefanides, E. J. "PC-Based System Automates Testing." Design News, pp. 68-69, Sept. 1989. 706. Sib, G. C., et al. Analytical and Testing Methodologiesfor Design with Advanced Materials. New York: Elsevier, 1989. 707. Gauer, U., et al. "Measuring FiberlMatrix Adhesion in Thermoplastic Composites." Plastics Engineering, pp. 43-45, Oct. 1989. 708. Hauk, V., et al. "Determining Residual Stresses in Polypropylene Pipes by X-Ray Diffraction." KunststoJfe, pp. 36-37, Feb. 1989. 709. Redner, A. "Automatic Optical System Measures Residual Strains." Plastics Engineering, pp. 37-39, Feb. 1987. 710. Williams, J. B. "Use of DSC as a Quality Control Tool." SPE-ANTEC, pp. 734-38, May 1989. 711. Wolcott, J. "Recent Advances in Ultrasonic Technology." SPE-ANTEC, pp. 502-6, May 1989. 712. Green, A. "Quality Assurance of Polymer Materials and Products." STP 846. Philadelphia: ASTM,1985.

REFERENCES 959

713. Feigenbaum, A. V. Total Quality Control. New York: McGraw-Hill, 1983. 714. Eiden, H. P. "Automatic PC with Quality Data Systems." Industrial & Production Engineering, p. 114, Mar. 1985. 715. Ishakawa, K. "Guide to Quality Control." Tokyo: Asian Productivity Organization, 1976. 716. Juren, J. M. Quality Control Handbook. New York: McGraw-Hill, 1974. 717. Charboneau, H. C., and Webster, G. L. Industrial Quality Control. New York: Prentice Hall, 1978. 718. Wu, Y. Introduction to Off Line Quality Control. Tokyo: Central Japan QC Assoc., 1979. 719. Besterfield, D. H. Quality Control. New York: Prentice Hall, 1979. 720. Brostow, W., and Corneliussen, R. Failure of Plastics. Munich: Hanser, 1986. 721. Wulps, D. J. Understanding How Components Fail. ASM International, 1985. 722. Marshall, I. H. Composite Structures. Vols. 1, Analysis and Design Studied; 2, Damage Assessment and Material Evaluation. New York: Elsevier, 1987. 723. Failure Analysis and Prevention. Vol. 11, Metals Park, OH: ASM International, 1986. 724. Burgess, J. A. Design Assurance for Engineers and Managers. New York: Marcel Dekker, 1984. 725. Plastics: Problems and Promise. D. L. Babson & Co.; Investment Council, June 1, 1967. 726. Witzler, S. "Composite Failure Analysis." Advanced Composites, pp. 23-33, Nov.-Dec. 1986. 727. Forbes, G., et al. ''The Use of Computer Simulation in Product Designs to Prevent Failure of Plastics Parts." SPE ANTEC, pp. 1592-93, May 1989. 728. Tormala, S. "Automatic Statistical Process/Quality Control System for Plastic Containers." SPE ANTEC, pp. 941-43, May 1989. 729. "Succeeding with Statistical Process Quality Control." Plastics Technology, pp. 95-101, Nov. 1984. 730. Abbott, W. H. "Statistics Can Be Fun." Chesterlan, OH: Abbott Pub., 1985. 731. Deming, W. E. Quality, Productivity and Competitive Position. Cambridge, MA: MIT Center for Advanced Engineering Studies, 1982. 732. _ _ . On Some Statistical Aids Toward Economic Production Interfaces. Providence, RI: Institute of Management Science, 1987. 733. Warner, S. A., et al. "Ford Molds a Better Bumper via SPC." Modern Plastics, pp. 68-74, Feb. 1986. 734. Grant, E. U., et al. Statistical Quality Control. 5 ed. New York: McGraw-Hill, 1980. 735. Wenskus, J. J. "Statistical Process Control for Injection Molding." Rochester, NY: Eastman Kodak, Document EK No. 248026V, Sept. 7, 1988. 736. Burr, I. W. Elementary Statistical Quality Control. New York: Marcel Dekker, 1979. 737. Statistical Quality Control Handbook. Indianapolis: I.D.C. Commercial Sales, Western Electric, 1956. Note: This is the classic review of SQC. 738. Gammons, M. P. "SPC in Fiber Glass Manufacturing." PPG Industries Reinforced Digest, No. 43, 1986. 739. See "Failure Papers" throughout ANTEC and SPE ANTEC, 1988, 1989. 740. "Failure Analysis of Composite Laminates." Seminar notes. San Diego: Technomic, 1986. 741. Bever, M. B. Encyclopedia of Materials Science & Engineering. New York: Pergamon, MIT Press, 1986. 742. International Journal of Materials & Product Technology. Geneva: Interscience Enterprises, Ltd., 1986. 743. International Journal of Fracture. Netherlands: Kluwer, 1990. 744. [Polymer NDE]. Proceedings, European Workshop on Non-Destructive Evaluation of Polymers & Polymer Matrix Composites. K. Ashby, Ed. Portugal: Technomic, 1984. 745. Engel, L. et al. An Atlas of Polymer Damage; Surface Examination of Fractures & Other Types of Damage by Scanning Electron Microscopy. San Diego: Technomic, 1981. 746. Zeuch, N. "Machine Vision." Plastics World, pp. 103-7, May 1985. 747. ASTM Standards & Literature References for Composite Materials. ASTM, 1989.

960 REFERENCES

748. 749. 750. 751. 752.

753. 754. 755. 756. 757. 758. 759. 760. 761. 762.

763.

764. 765. 766. 767. 768. 769.

no. 771.

772. 773. 774. 775. 776. 777. 778. 779. 780. 781.

ASTM Composites. ASTM Vol. 15.03 (annual). "The Heat of the Matter." World Plastics & Rubber Technology, pp. 79-81, 1990 Annual. "Put to the Test." World Plastics & Rubber Technology, pp. 72-73, 1990 Annual. Maloney, L. "Essential Tools of the Trade." Machine Design, p. 11, Sept. 1989. Tuschak, P. A. "Non-Linear Stress Analysis with the Abacus Finite Element Program." pp. 14-15, Du Pont, Engineering Design Magazine, Winter 1987. Throne, J. L., and Kmetz, M. "Computer-Aided Design in Thermoforming." Plastics Engineering, pp. 35-38, Sept. 1989. Bernhardt, E. C. "Using CAE Systems for Successful Design." Plastics Design Forum, pp. 61-66, Sept.-Oct. 1989. "All in a Day's Work." World Plastics & Rubber Technology, pp. 183-86, 1990 Annual. Dietz, C., and Bingen, K. F. "CAD-CAM in Mold and Model-Making." KunststofJe, pp. 7-13, July 1989. McCarthy, L. "Plastics Software Comes of Age." Plastics World, pp. 49-53, Apr. 1989. Rockett, D. "CAD Libraries End Design Redundances." Design News, pp. 93-95, Jan. 1990. Fallon, M. "New Dimensions in Mold Analysis." Plastics Technology, pp. 74-83, Apr. 1989. "Composite Manufacturing Takes a Giant Leap Forward." Plastics World, pp. 46-48, Jan. 1990. Leonard, L. V. "Dynamic Duo for the '90s: Plastics and DFMA." Plastics Design Forum, pp. 21-32, Jan.-Feb. 1990. Easy CAD: Computer-Aided Drafting. Metals Park, OH: T&W Systems, ASM Software, 1988. Giesecke, F. E., et al. Engineering Graphics with Computer Graphics. New York: Macmillan, 1985; Dowler, C. A. "Automatic Model Building Cuts Design Time Costs." Plastics Engineering, pp. 43-45, Apr. 1989. "En Plot: Coordinate Engineering Graphics." Metals Park, OH: PSI Systems, ASM Software, 1988. "Beams and Frames." Compu-Tech Engineering, Metals Park, OH: ASM Software, 1988. Thomas, T. K. Computer Source Book. Metals Park, OH: ASM International, 1986. Basu, S., and Rauwendaal, C. J. "Computer Program." SPE ANTEC, 1982. Underwriters Laboratories Data Services. Melville, NY: Underwriters Laboratories. "New Computer Aids for Tool Design and Process Modeling." Plastics Technology, pp. 2127, Aug. 1986. Bernhardt, E. C. Computer-Aided Engineeringfor Injection Molding. Munich: Hanser, 1984. _ _ , and Bertacchi, G. "Quality Molding by Design." SPE ANTEC, pp. 299-302, May 1989. Bernhardt, E. C. Processing of Thermoplastics. New York: Van Nostrand Reinhold, 1986. _ _ , and Bertacchi, G. "New Tool for Mold Design: Computerized Shrinkage Analysis." Plastics Technology, Jan. 1986. _ _ . "The Practical Application of Computerized Shrinkage Analysis. " SPE ANTEC, 1986. _ _ . "TMC Plastics Molding and Cost Optimization." SPE ANTEC, May 1983. Tucker, C. Fundamentals of Computer-Aided Engineering in Injection Molding. Munich: Hanser, 1988. Groth, K. B. "Computer-Aided Mold Cooling Design for Plastic Parts." In DorghamJRosato, Designing with Plastics and Composites. Geneva: Interscience Enterprises, 1986. "C-Flow Injection Mold Filling Analysis Hardware." Ithaca, NY: Advanced CAE Technology, 1987. Reddy, K. H. "The Development/Implementation of ACTFLOW, a CAE Environment Designed for Polymer Flow Analysis." SPE ANTEC, pp. 1766-68, May 1989. Schuder, D., and Caren, S. "Mold Design with CAD/CAM/CAE." SPE ANTEC, pp. 125861, May 1989. Palit, K., and Chen, I-J. "How to Assess CAE Injection Mold Design Software for Production Applications." SPE ANTEC, pp. 356-59, May 1989.

REFERENCES 961

782. Beaumont, J. P. "Computerized Analysis Can Straighten Out Warped Plastic Parts." SPE RETEC, Oct. 27-28, 1988. 783. Rakos, R., et al. "Computer Design Aids for Non-Axis-Symmetric Profile Dies." SPE ANTEC, May 1989. 784. Zimny, I. C. "Blow Molding Tools Up With CAD/CAM." Plastics Technology, pp. 69-71, Dec. 1984. 785. Szajna, J. L. "Traditional and CAD/CAM Plastic Bottle Tool Design." SPE ANTEC, pp. 868-70, May 1985. 786. Bonds, T. W. "Plant Layout and Design Utilizing 3-Dimensional Solid Modeling in CAD." SPE Blow Molding Div. Tech. Conference, Nov. 11-13, 1985, pp. 221-25. 787. Fisher, T., and Grosse, C. "Automation in Forming Using CAD/CAM." Industrial & Production Engineering, pp. 60-63, Apr. 1984. 788. Pearson, J., and Richardson, S. Computational Analysis of Polymers. New York: Elsevier, 1983. 789. Callari, J. "Resin Data Bases Springing Up Fast." Plastics World, pp. 14-16, July 1989. 790. Dieckmann, D. "Roundtable on Computerized Plastics Materials Data Bases." SPE ANTEC, pp. 1784-89, May 1989. 791. "Advanced Materials Data Base System." Costa Mesa, CA: PDA Engineering. 792. Computer-Aided Material Preselection by Uniform Comparability of Data From Different European Plastics Producers and Operations [CAMPUS]. Includes material producers BASF, Bayer, Hoechst, Huls, etc., and operations Dow Chemical, GE, Ciba-Geigy, Exxon, etc.; KU-A-KFC, Geb.B207, D-5090. Leverkusen, Germany. 793. ASTM Preparatory for an Equivalent to the European CAMPUS via Committees D-20, E49, etc.; Develop Standard Formats for Computerized Data Bases. ASTM, 1989. 794. CENBASE, Computer-Aided Materials Database. Garden Grove, CA: Infodex/CENCAD. 795. Engineering Design Database [EDD]. Pittsfield, MA: GE Plastics. 796. Engineering Properties on Screen [EPOS] Malvern, PA: ICI Advanced Engineering, LNP Div. 797. International Plastics Selector, Div. of D.A.T.A., San Diego, CA. 798. Materials Properties Data System [MAT.DB.] Metals Park, OH: ASM International. 799. Materials Engineering Center Data Base [MEC]. Midland, MI: Dow Chemical. 800. Materials Property Data [MPD] Provides Automatic Scan of All Data Bases. Columbus, OH: National Materials Property Data Network. 801. Plastic Materials Selection System [PLASCAMS]. Shrewsbury, England: RAJ>RA Technology, Ltd. 802. PLASPEC. "Engineering and Marketing Databank Technology." Plastics Technology. New York: Plaspec. 803. PLASTEC. "Plastics Technical Evaluation Center." Dover, NJ: U.S. Army, Picatinny Arsenal. 804. POLYDATA. "Computer-Aided Polymer Selection" [CAPS]. Dublin: Polydata Ltd. 805. POLYFACTS. "Selection and Work with Engineering Plastics." Wilmington, DE: Du Pont. 806. "Thermofile-Computer Database Program." Brighton, MI: Thermofil, Inc. 807. Beall, G. "Designing Wodd Class Plastics Products." Kunststojfe, pp. 95-98, Oct. 1989. 808. Marks, W. F. "Stress/Strain Polynomials: Their Use Could Result in More Accurate Design Calculations." Du Pont Engineering Magazine, pp. 14-15, winter 1986. 809. Alvord, L. J. "Engineering Blow Molded Part Design." Paper presented at GE Plastics SPI Structural Plastics Conference, 1989. 810. Peterson, R. E. Stress Concentration Design Factors. New York: Wiley, 1980. 811. Schonewald, H. "Snap Joints in Use." Kunststojfe, pp. 8-11, Aug. 1989. 812. "Silicone Engineering Data and Processing Bulletins": 01-133, 01-206, 08-317, 22-222a, 22290,23-251, 23-262a, etc. Midland, MI: Dow Coming. 813. Galli, E. "Design Guidelines for Structural Blow Molding." Plastics Design Forum, pp. 3744, Sept.-Oct. 1989.

962 REFERENCES

814. Chamberlain, G. "New Design Battleground: Auto Interiors." Design News, pp. 56-64, Jan. 1990. 815. Corbo, V. J. "Advanced Materials Geared for the 90's." Design News, pp. 69-70, Jan. 1990. 816. Financial Management Manuals for Plastics Processors. Vols. 1: Mfgr. Cost Control; 2: Pricing and Estimating, 3: Planning, 4: Materials Management. SPI, 1979. 817. Ozan, T. R. Financial Management Seminars. New York: Ernest & Whinney, 1989. 818. Tobin, W. J. "Practical Part Costing for Custom Molders." Plastics Machinery & Equipment, pp. 37-39, Dec. 1986. 819. Kroh, W. "The Economical Application of Modem Technologies in Small and Medium-Sized Plants." Industrial & Production Engineering, pp. 86-90, Oct. 1986. 820. Newnan, D. G. Engineering Economics, Analysis. San Jose: Engineering Press, 1983. 821. Fulmer, G. E. "Price Is the Most Important Physical Property of a Polymer." SPE ANTEC, 1967. 822. Chow, W. M. Engineering Economy. New York: Van Nostrand Reinhold, 1978. 823. Facts and Figures of the US Plastics Industry. SP!. Annual. 824. Machlis, S., "Working Smarter." Design News, pp. 56-64, Sep. 1989. 825. "Plastics Market, a Great Growth Market That Won't Stop." Industry Management, 1965. 826. Moore, M. "Jarvik Heart Recall Will Not End Research." Plastics News, p. 6, Jan. 1990. 827. Wadsworth, H. M. Handbook of Statistics/Methods for Engineers & Scientists. New York: McGraw-Hill, 1990. 828. "Ski Pole's Color Changes with Temperature." Design News, p. 41, Jan. 1990. 829. Haldenwanger, H. G., and Schaper, S. "Lightweight Construction in Automobiles Using Plastics Means the Conservation of Resources." Kunststojfe, p. 5, Nov. 1989. 830. Roney, c., and Becka, L. "For Straight Shooters: A Super Tough Nylon Casing." Du Pont Engineering Magazine, pp. 6-7, spring 1987. 831. Goldsberry, C. A. "Aqua Culture Cultivates Plastics Gardens." Plastic News, p. 11, Jan. 1990. 832. Klein, A. J. "Plastics Materials and Design: Improving and Saving Lives." Plastics Design Forum, pp. 25-37, Nov.-Dec. 1988. 833. "Paragon Grows As Contact Lens 'Button' Holder." Plastics News, pp. 1,20, Jan. 1990. 834. Gray, E. K., and Hopf, G. "Nylon 12 in the Automotive Fuel System Environment." Plastics Engineering, pp. 37-41, Dec. 1989. 835. "Composite Wheel Cuts Wind Drag for Bikes." Design News, p. 41, Jan. 1990. 836. Bacchetti, J. A. "Electronic Connector Demands." Plastic Trends, pp. 15, 20, Nov.-Dec. 1989. 837. "Forming-Syntactic Foam Plugs." Plastic Trends, p. 19, Nov.-Dec. 1989. 838. Rees, D. V. "Verton Shines in Halo Applications" [MRI]. Plastics Today, p. 15, Dec. 1989. 839. Yeaple, F. "Composites Extend Life of Seawater Pump." Design News, pp. 72-73, Jan. 1990. 840. "Polyester Flush Tank Works Under Pressure-Hoechst Celanese." Plastics Today, p. 13, Dec. 1989. 841. Wood, A. S. "In the Powertrain, Where It Really Counts-Plastics." Modern Plastics, pp. 46-52, Oct. 1989. 842. Vernyl, B. "Using Plastics to Buy Plastic: Bank Machines to Get Stamps." Plastics News, pp. 1,20, Dec. 1989. 843. Ritz, G. J. Total Engineering Project Management. New York: McGraw-Hill, 1990. 844. "Sioux Indians Manufacturing Plastic Helmets for U.S. Army" [Allied Spectra]. Plastics News, p. 4, Jan. 1990. 845. "Copolymer Sparkles Like Crystal." Design News, p. 26, Jan. 1990. 846. "Composite Boom Stabilizes New Helicopter." Plastics News, pp. 1,5, Jan. 1990. 847. "Isometric Trisaryloxycyclotriphos-Phazene Polymer Precursors." NASA Technical Briefs, p. 42, Jan. 1990. 848. Vernyi, B. "Composites Look Strong Into 1990's." Plastics News, pp. 7,14, Jan. 15, 1990.

REFERENCES 963

849. 850. 851. 852. 853. 854. 855. 856. 857. 858. 859. 860. 861. 862. 863. 864. 865. 866. 867. 868. 869. 870. 871. 872. 873. 874. 875. 876. 877. 878. 879. 880. 881. 882. 883. 884.

"A New Pitch for Business." Du Pont Magazine, pp. 24-25, Jan. 1990. "BCB's: New High-Heat Thermosets for Electronics." Plastics Technology, p. 13, Jan. 1990. "DSM Bringing New Nylon to U.S. Market." Plastics News, p. 15, Jan. 1990. Kreisher, K. R. "Maintenance by Modem: Processing Problems Can Be Solved Remotely." Modern Plastics, p. 44, Jan. 1990. "Gas Injection Makes Better Thick Parts for Less." Plastics Today, p. 12, Dec. 1989. "Snail-Shell Molding: New Way Makes Hollow Shells." Plastics Today, p. 13, Dec. 1989. "Recycling Technology." World Plastics & Rubber Technology, pp. 218-20, 1990 Annual. "CSWS Reports on Solid Waste Legislation." Plastics Engineering, p. 5, Dec. 1989. "Voluntary Guidelines, Rigid Plastic Container Material Code System: Mold Modification Drawings." SPI Technical Bulletin, RPCD-13, 1989. "Landfill-Plastics Volume Figures Keep on Piling Up." Modern Plastics, p. 12, Dec. 1989. "New Ventures in Degradable Films." Plastics Technology, p. 122, Dec. 1989. Lindsay, K. F. "Freedom from CFCs Draws Nearer As Global Efforts Grow More Intensive." Modern Plastics, pp. 132-41, Jan. 1990. "Improving Packaging Resins, Recycling Highlight Materials." Plastics World, pp. 16-20, Jan. 1990. "Degradable Boycott Disputed." Plastics News, p. 2, Jan. 1990. "Growth Rates for Recycling, 1989-1994." Modern Plastics, p. 143, Jan. 1990. Earle, J. H. Engineering Design Graphics. 5 ed. Reading, MA: Addison-Wesley, 1987. "Cyclics Technology May Make Thermoplastic Composites Common." Modern Plastics, pp. 10-12, Nov. 1989. "Pyrogaf II: New Process Produces Vapor-Grown Carbon Fibers." GM Research Labs. Search. Vol. 24, no. 4, Sept.-Oct. 1989. Haut, D. "Success by Design." Medical Device & Diagnostic Industry, pp. 48-60, Sept. 1988. Murray, C. J. "Toronto Skydome: Power Transmission Showcase (Million Dollar Model)." Design News, pp. 52-57, Jan. 1990. Scott, R. R., and Straw, R. B. "FDA's 51O(K) Review Procedure for Medical Devices." Medical Device & Diagnostic Industry, pp. 62-65, 152-62, Jan. 1990. Murray, C. J. "Foam Exhibits Negative Poisson's Ratio." Design News, pp. 8-9, Dec. 1989. Earle, J. H. Graphics for Engineers. Reading, MA: Addison-Wesley, 1985. Bauer, E. "Knowledge Based Plastics Application Databases." SPE ANTEC, May 8, 1990. The Software Catalog. New York: Elsevier. Annual. Buckley, C. P., and Resen, A. S. "Characterization of Creep of Plastics Under Multiaxial Stress." SPE ANTEC, May 7, 1990. Brueller, O. S., and Steiner, H. "Nonlinear Characterization of Relaxation Behavior of Plastics by Using Creep Data." SPE ANTEC, May 7, 1990. Malone, J. G. Using Parallel Processing for Finite Element Analysis. GM Research Laboratories. Vol. 25, no. 1, Feb. 1990. Arnold, F. E. "Structural Materials for the Future: Molecular Composites." SPE ANTEC, May 7, 1990. Woelfel, J. "Composite Automotive Wheel." SPE ANTEC, May 7, 1990. Connon, H. "The Push for Alternatives: Chlorofluorocarbons." Du Pont Magazine, pp. 13, Mar.-Apr. 1990. Beaumont, J.; Koch, P.; and Farrell, R. "Frictional Heating in Runners for Enhanced Thermoset Molding." SPE ANTEC, May 7, 1990. Potsch, G., and Michaeli, W. "Prediction of Linear Shrinkage and Warpage for Thermoplastic Injection Moldings." SPE ANTEC, May 8, 1990. "Success by Design." Medical Device & Diagnostic Industry, pp. 48-59, Sept. 1988. Middendorf, W. H. Design of Devices and Systems. New York: Marcel Dekker, 1986. Callari, J. J. "New Equipment Recipe: Quality + Control + Versatility." Plastics World, pp. 43-46, Mar. 1990.

964 REFERENCES

885. McConnell, V. P. "Designing for Noise, Vibration and Shock Control." Plastics Design Forum, pp. 29-49, Mar.lApr. 1990. 886. "Special Report: Intertwined Medical and Polymer Technologies." Modern Plastics, pp. 5586, Mar. 1990. 887. Naitove, M. "Needed: Better Materials Data." Plastics Technology, p. 7, Dec. 1989. 888. Marotta, C. D. "Predicting the Shelf-Life of Packaging Materials." Medical Device & Diagnostic Industry, pp. 22-26, Aug. 1984. 889. Ellis, J. R. "Plastics/Packaging: PVC-New Products Keep Coming." Medical Device & Diagnostic Industry, pp. 90-92, Mar. 1990. 890. "Plastic Fuel Tank Bladder Cuts Explosion Hazard." Plastics World, p. 22, Mar. 1990. 891. Rosato, D. V. "Elastomers Perform and Prosper." Plastics World, Apr. 1967. 892. Miller, B. "FR's on Track for Higher Physicals, Lower Smoke." Plastics World, pp. 4853, Mar. 1990. 893. Sharp, G. R. "Lightweight, Thermally Conductive Composite Material." NASA Technical Briefs Report LEW-14814, p. 66, Mar. 1990. 894. Barenberg, S. A. "Special Report: Biomaterials." Medical Device & Diagnostic Industry, pp. 81-87, Mar. 1990; Pawar, P. "Flame-Retardant Polyolefins Don't Need Halogen." Plastics Trends, pp. 75-79, Mar. 1990. 895. "New TP Elastomers Challenge Fluoropolymers in Plenum Cable." Plastics Technology, p. 13, Mar. 1990. 896. Niederstadt, G., et al. "Viscoelastic and Damping Behavior of Composites Made from Carbon Fiber Reinforced Prepregs." KunststoJfe, pp. 27-29, Jan. 1990. 897. Das, S. et al. "Phenolic-Triazine Resins Yield High-Performance Thermoset Composites." Modern Plastics, pp. 72-79, Feb. 1990. 898. Ellis, J. R. "Packaging/Sterilization: EtO-Does It Have A Future?" Medical Device & Diagnosticlndustry, pp. 50-51, Feb. 1990. 899. Timoshenko, S. P. History of Strength of Materials. New York: McGraw-Hill, 1953. 900. Redfam, C. A. A Guide to Plastics. London: Iliffe, 1958. 901. Schwartz, M. M. Composite Materials Handbook. New York: McGraw-Hill, 1984. 902. Evans, B. "The SMC Plant of the '90's." Plastics Technology, pp. 64-70, Mar. 1990. 903. Chamas, D. "Product Development Cycle Time: Design to Manufacture." Plastics News, pp. 1, 20, Mar. 26, 1990. 904. Anton Malburg, H. "Carousel Plant for Producing Foam-Encapsulated Steering Wheels." Industrial & Production Engineering, pp. 76-77, Apr. 1989. 905. "PVC Makes Progress in Hot Runners, Structural Blow Molding." Plastics Technology, pp. 15-17, Mar. 1990. 906. Ohta, Y., et al. "Warpage in Rotational Molded Parts of Polyethylene." KunststoJfe, pp. 4244, Dec. 1989. 907. Tormala, S. "Measuring Layer Thickness in Coextrusions by Interferometry." Plastics Engineering, pp. 75-77, Mar. 1990. 908. "Spectroscopy." Compressed Air, pp. 26-31, Mar. 1990. 909. Powers, J. R., and Jenkins, S. R. "Cutting Costs, with No Loss, of Making Shelf-Stable Packaging." Plastics Engineering, pp. 71-73, Mar. 1990. 910. "B.F. Goodrich Establishes PVC Recycling Laboratory." Plastics Technology, p. 105, Mar. 1990. 911. "Recyclying Ventures Proliferate." Plastics Technology, pp. 119-20, Mar. 1990. 912. Nosker, T. J., et al. "Recycle Polystyrene, Add Value to Commingled Products." Plastics Engineering, pp. 33-36, Feb. 1990. 913. "Design-Integrated Manufacturing Solves Solid-Waste Problem." Plastics World, pp. 13-14, Feb. 1990. 914. "All-Starch Polymer Enters Biodegradable Derby." Plastics World, p. 12, Mar. 1990. 915. "Plastic Starch Injection Molds Degrades." Design News, p. 35, Mar. 1990.

REFERENCES 965

916. Stover, D. "Recycling Technology Ready for the '90s." Plastics Machinery & Equipment, pp. 35-36, Mar. 1990. 917. Machlis, S. "How to Bring CAD to Life." Design News, pp. 54-60, Mar. 1990. 918. Feldmann, P. F. "User Experiences with PC-Based CAD/CAE System." Kunststojfe, pp. 810, Jan. 1990. 919. Noller, R. "Understanding Tight-Tolerance Design." Plastics Design Forum, pp. 61-73, Mar.-Apr. 1990. 920. Stanley, G., and Welch, C. The Molding of an Industry. British Plastics Federation, 1983. 921. Haut, D. "Success by Design." Medical Device & Diagnostic Industry, pp. 48-69, Sept. 1988. 922. Kraus, H. Creep Analysis. New York: Wiley, 1980. 923. Goldfein, S. "General Formula for Creep and Rupture Stresses in Plastics." Modern Plastics, Apr. 1960. 924. Sheridan, T. W. "Creative Design with Thermoplastic Elastomers." Plastics Design Forum, pp. 21-26, Mar.-Apr. 1989. 925. Janzen, W., and Ehrenstein, G. W. "Hysteresis Measurements for Characterizing the Cyclic Strain and Stress Sensitivity of Glass Fiber Reinforced PBT." SPE ANTEC, pp. 659-63, May 1989. 926. Felton, L. P. "Structural Index Method of Optimum Design." Proceedings, AS ME winter annual meeting, New York, 1974. 927. Rogers, J. K. "Design Meets Manufacturing." Plastics Technology, pp. 44-55, Nov. 1988. 928. Klein, A. J. "Plastics Materials and Designs: Improving and Saving Lives." Plastics Design Forum, pp. 25-37, Nov.-Dec. 1988. 929. Murray, C. J. "Design with Lives in Mind." Design News, pp. 100-2, Feb. 1990. 930. Stephenson, G. M., and Freiherr, G. "Cost-Effectiveness: New Yardstick for Medical Device Evaluation." Medical Device & Diagnostic Industry, pp. 71-73, Mar. 1990. 931. Bregar, B. "Architect's Concept Would Clad Buildings in Plastic." Plastics News, p. 3, Feb. 19, 1990. 932. Goldsworthy, B. "Composites for Roads and Bridges." Plastics Trends, p. 12, Jan.-Feb. 1990. 933. McConnell, V. P. "Launch Systems: Advancing Composites into Space." Advanced Composites, pp. 32-44, Jan.-Feb. 1990. 934. Robinson, G. M. "Tomorrow's Technologies at Your Doorstep." Design News, pp. 103-7, Mar. 1990. 935. Vernyi, B. "Composites Prove Worthy; GM Testing Use in Luxury Cars." Plastics News, pp. 1,6, Feb. 16, 1990. 936. Snyder, M. "Composites Market Plateaus." Plastics Machinery & Equipment, pp. 31-32, Apr. 1990. 937. "A Glimpse at Future Auto Designs." Design News, pp. 23-24, Mar. 1990. 938. "Composites at Sea." Advanced Composites, pp. 38-58, Mar.-Apr. 1990. 939. Grolik, F. W., and Naumann, H. G. "Reinforced Polypropylene for Motor Vehicle Interior Fittings." Kunststo.IJe, pp. 34-38, Dec. 1989. 940. Rosato, D. V. "Product Design: Plastic Selection Guide." SPE ANTEC, May 1990.

Index

Ablation, 154,299. See also Heat shield Abrasion, 239, 244, 415 ABS, 74, 622 Absorption, 273 Accumulator head, 652 Acetal, 321,414, 883 Acetate plastic, 417 Acidproof brick, 421 Acid resistance, 438 Acoustic, 380 Acoustic test, 748 Acrylate elastomer, 470 Acrylic plastic, 272, 415 Acrylonitrile-butadiene-styrene plastic, 406 Actual versus theoretical values, 97 Additive, 63, 77, 287 Additive migration, 277 Adhesive, 362, 432, 714, 869 Adiabatic heat, 201 Advantages. See Plastic advantages and disadvantages Aerodynamic shape, 235 Aerospace, 18, 196, 300, 309, 361, 740, 885 Aesthetic, 51, 230, 303, 780 Aging at elevated temperature, 95, 385 Aircraft. See Aerospace Air entrapment, 614, 686, 814 Alkyd plastic, 416, 435 Alloy, 72 Aluminized film, 434 Aluminum, I, 127, 246 Alvin depth vehicle, 295 American Society for Testing and Materials, 734, 736 Amino plastic, 416, 423 Amorphous plastic, 66, 166 Anisotropic, 71, 122, 329. See also Directional properties; Cbientation

Anti-icing, 235 Antilittering, 402 Aorta tube, 26 Apparent creep modulus. See Modulus of elasticity, creep-apparent Appliance, 38, 233, 371, 403, 902 Approach engineering, 46 practical, 46, 763, 888 Aquacycle, 648 Aramid fiber, 82, 196, 241 Area under the stress-strain curve, 141, 208. See also Stress-strain Arrhenius plot, 260 Assembling, 340, 345, 714, 869 Athletic equipment, 436 Autoclave, 274 Autoclave molding, 678 Automobile, 29, 371, 681, 885, 904 Automotive bumper, 154, 220 Auxiliary equipment, 643, 711 Barcol hardness, 105, 245 Barrel,628 Barrier resistance, 276, 328 BASF,534 Bayer, 534 Beam, 312, 321, 366, 521 Beam equation, 146 Bearing, 240, 380, 420, 424 Belleville washer, 402 Bellow structure, 317 Bending modulus. See Modulus of elasticity,

flexural Biaxial direction, 118. See also Cbientation Bidirectional, 122 Biodegradable, 45, 891 Biological degradation, 273, 281 Birefringence, 99

967

968 INDEX

Black box, 297 Bleed out, 418 Blend, 72, 406 Blow molding, 17,318,434,638,646,847 Blown film, 638 Boat. See Ship Bolt, 163, 362, 400, 833 Boltzman superposition principle, 167 Bond. See Adhesive Boron fiber, 82, 196 Bosses, 342, 790, 823 Bottle, 433, 646, 659, 747, 770 Bottle cap, 832, 841 Bottle, collapsible, 318 Box beam, 319, 368 Brackish water, 883 Brass, 1,246 Brinell hardness, 105 Brittle fracture. See Fracture Brittleness, 106, 201 Buckling strength, 328, 371 Building and construction, 18, 273, 283, 371, 684 Bulk molding compound, 39, 416, 418, 494 Bumper. See Automotive bumper Buoy, 289 Buoyance, 295 Burn cleanly, 893 Burning, 889 Butadiene plastic, 406 Butyl elastomer, 470 Butyrate plastic, 417 Calendering, 693 Cam, 230 Campus database, 534 Canoe, 10 Cantilever beam, 21, 148, 315 hinge, 350 snap fit, 349, 402 spring, 322 Carbon black, 81, 385. See also Filler Carbon fiber, 196, 241, 885 Casting, 424, 693 Catastrophic failure, 153, 198, 285, 310 Cathode sputtering, 230 Cauchy-Riemann differential equation, 395 Cause-effect relationships, 56 Cavitation erosion, 295 Cellulose acetate butyrate plastic, 417 Cellulose acetate plastic, 417 Cellulose acetate propionate plastic, 417 Cellulose nitrate plastic, 417 Cellulose plastic, 72, 417 Cement, 421. See also Adhesive Center of gravity, 385 Charpy impact test, 212. See also Impact Chemical characteristic, 116 process, 71 reaction, 276

resistance, 264, 417, 561 separation system, 886 Chimney liner, 47 Chlorinated polyether plastic, 417 Chlorinated polyethylene plastic, 418, 470 Chlorinated polyvinyl chloride plastic, 452 Chlorosulfonated polyethylene elastomer, 470 Chrome plate, 412 Circuit board, 233, 623 Classifying plastic, 525 Clearance. See Tolerance Coating, 230, 301, 420,475,646 Coefficient of diffusion, 276 of friction, 242, 420 of linear thermal expansion, 89, 337, 362, 388, 798

of torsion, 401 Coextrusion, 18,280,635 Coining, 349, 693 Coinjection molding, 623 Cold forming, 105, 355, 668 Cold work, 351, 794 Collapsible container, 316, 663 Collapsible mechanism, 355 Color, 78, 100, 790 Combustion. See Flammability Commodity plastic, 82, 254. See also Engineering Plastic Communication protocol, 711, 888 Comonomer, 71 Compatibilizer, 72 Competition, 592, 762 Composite. See Reinforced plastic, definition; Plastic, definition Compound, 72 Compressible melt, 116 Compression molding, 668, 679 Compression modulus. See Modulus of elasticity, compression Compression properties, 145, 375, 382 Compressive recovery rate, 145 Compressive strength, 135,474 Compressive yield strength, 207, 400 Compromise, 303 Computer image processor, 753 in design, 52, 776 modeling, 763, 680 program, tolerances, 399 technologies, 53 Computer-aided chemistry, 741 design, 53, 615, 757 design drafting, 53 engineering, 615, 778 manufacturing, 53, 652, 758 testing, 54 Computer-integrated manufacturing, 53, 606, 785 Computerized software and databases, 528. See also Appendix D

INDEX 969

Concrete, 1, 152, 328 Conductive plastic, 230, 239 Conservation, 402 Constrains, 789 Construction. See Building and construction Conversions. See Appendix B Cooking utensil, 420 Cooling stresses. See Stresses, cooling Copolyester elastomer, 472 Copolymerization, 204 Copolymer, 71, 406, 414 Copper, 1 Core. See Foam; Honeycomb; Sandwich Cored metal casting, 693 Coring, 804, 824 Comer, 799 Corrosion, 241, 264, 372,417,420 Corrugation, 49, 838 Cost, 51, 58, 133,249,276,338,373,524,593, 650, 706, 888, 898 Cost modeling, 702 Crack, 142, 154, 191,221,270,309 Crazing, 142, 159, 270 Creativity, 51, 125, 914 Creep, 114, 155,247,401 guidelines, 179, 189 modeling, 174 modulus, 175 rupture, 156, 158, 167, 273 theory, 182 Creep-gasket, 360 Cross-bond, 65 Cross-head rate. See Test rate Cross-link, 64, 97, 282, 416, 440 Crystal clear, 415 Crystalline plastic, 66, 166, 195 Cushioning, 475, 488 Cutting plastic guidelines, 728 Cyclic chemistry, 884 Cyclic load. See Fatigue Damping, 154,201,361 Data theoretical, 97 Databases, 528, 758, 773, 782, 929. See also Computer and Appendix D Decay, 161 Decomposition temperature, 95, 284 Decorating, 714, 790 Defect, melt flow. See Melt flow defects Deflection, 315, 338, 372, 402, 820 Deflection temperature under load, 94, 258 Deformation, 107, 153, 163,203,381 Degradable, 894 Degradation, 117,213,281 Density, 84, 98, 112 Density reduction, 365 Design allowables, 311 analysis, 46, 48, 126, 129, 696, 789, 910 bending of bar, 184, 521 bolt preload, 400

building-fire, 285 building frame, 282 challenge, 52, 914 compromises, 2, 5, 48, 900 consideration, 789, 898 criteria, 125, 132, 910 cushioning, 488. See also Design, dynamic load databases, 782. See also Databases deficiencies, 754, 789 definition, 9 dynamic load, 371, 540. See also Design, cushioning fail-safe, 194 failure analysis. See Failure analysis features, 789 flow diagram, 4 for environment, 385 gears, 359 graphic. See Graphic graphic uncertainties, 310 optimization, 1, 789, 780 overdesign, 126. See also FALLO approach parameters, 4, 789, 910 pipe-heavy wall, 189 pipe joint, 339 pipeline on supports, 185 pipe-reinforced plastic, 326. See also Pipe pipe, thin wall, 187 procedures, 48, 54, 696 procedures, flow diagrams, 4 products, 2, 703, 789, 895, 929 pseudo-elastic method, 304 redesign, 3, 5, 903. See also FALLO approach reinforced plastic theory, 495 rib, 835 safety factor, 309, 332. See also Safety factor sandwich structure, 368 screw (for processing plastic), 614, 628 shape, 49 snap fit, 346, 402 spring, 321 structures, 303, 835 success, 789, 917 thermal expansion, 362 tolerance guide, 802, 846 torsional spring, 323 with computers. See Computers in design with success, 51, 900 Design concept, 4, 231, 247, 904 blow molding, 659, 847 extrusion, 631, 844 injection molding, 617, 796 intermittent. See Intermittent behavior long term. See Long-term behavior reinforced plastic, 856 rotational molding, 865 thermoforming, 854 Designer and computer, 52 Designer and regulations, 402. See also Regulations Designer ethics. See Ethics

970 INDEX Designer responsibilities. See Responsibilities Designing classic approach, 46 dies, 629 fundamentals, 1,910 heat generating products, 902 molds, 393, 758. 799, 934 molds with "holes", 819 with creep data. 175 with databases, 4, 52, 773, 932. See also Databases with elastomers, 458 with fatigue data, 201 with impact data, 220 with materials, 932 with screw insert, 831 Detractor, 789 Deutsches Instut-Normung, 732 Diallyl isophthlate plastic, 418 Diallyl phthlate plastic, 418 Diaphragm, 420, 841 Dichroism, 99 Die, 391, 591, 599, 629 Die shear rate, 109 Dielectric strength, 229, 282, 420 Differential scanning calorimetry, 737 Diffusion, 276 Digitizing, 775 Dimensional change. See Tolerance Dimensional characteristic, 87, 265, 634, 834, 869 Directional properties, 118, 121. 333, 700 Disadvantages. See Plastic, advantages and disadvantages Dispersion, 100 Disposable labware, 18 Disposal,403 Distortion temperature. See Deflection temperature under load Dow Chemical, 470 Downtime, 889 Draft angle, 790, 810 Drain-waste-vent, 412 Drawing. See Graphics Droop test, 262 Drug release implant, 45 Drying plastic. See Hygroscopic plastic Ductile, 105 Ductile fracture. See Fracture Ductile-to-brittle transition temperature, 203 Ductility, 102, 160,220 DuPont, 400, 472 Durometer, 385, 460 Dynamic load, 153, 348, 371,540 Dynamic mechanical analysis, 387, 746 Dynamic mechanical behavior, 153, 590, 872 Dynamic modulus. See Modulus of elasticity, dynamic Dynamic sensitivity, 238 Dynatup test, 216. See also Impact

Eagle wing, 301 Earthquake, 373 Eastman Chemical, 472 "Ecdel", 472 Ecology, 402 Economic analysis, 877, 897. See also Cost Educational information, 53, 896. See also Training Ejector, 814 Elastic deformation, 107, 220, 387 Elastic limit. See Tensile elastic limit Elastic modulus. See Modulus of elasticity Elastic turbulence, 114, 154 Elasticity, 102, 112 Elastomer, 61, 70, 233, 301, 371,458, 472, 527 Electric iron, 39 Electric power tool, 38 Electrical, 223, 260, 418, 450 Electromagnetic compatibility, 225 interference, 225 radiation, 235, 282 welding, 727 Electronic, 12, 232, 434, 534 Electronic radiation, 228 Elongation creep, 157 fatigue, 191 impact, 207 tensile, 137, 385 Empathy, 51 Endurance limit, fatigue, 194 Energy absorbed, 372, 590 conservation, 713 consumption, 628 control, 371, 781 dissipation, 195, 196, 540 efficiency, 373 impact, 141 kinetic. See Kinetic energy ratio of recovery, 590 to failure, 210 to manufacture, 58 via incineration, 891. See also Incineration Engine, 10, 427, 880 Engineering approach. See Approach, engineering equation, 125,304,314 plastic, 82, 234, 366. See also Commodity plastic stress. See Stress, engineering Environmental concept house, 25 conditions, 253, 429 enbrittlement, 154 influence, 201, 385 Environmental Protection Administration, 891 Epoxy plastic, 263, 419, 427 Equipment improvements, 887

INDEX 971 Ergonomic, 51 Erosion, 295, 300 Ethics, 52 Ethyl cellulose plastic, 417 Ethylene acrylic elastomer, 471 Ethylene chiorotrifluoroethylene plastic, 421 Ethylene oxide, 274 Ethylene propylene elastomer, 470 Ethylene tetraftuoroethylene plastic, 421 Ethylene vinyl acetate plastic, 419 Euler equation, 371 European market, 530 EVAL Company of America, 276 Exothermic reaction, 474 Expandable polystyrene plastic, 449, 690 Extensometer, 142 Extruded hinge, 355 Extrusion, 610, 624, 642, 844 Extrusion blow molding, 651 Extrusion tandem complex, 888 Fabric, 82, 510, 584 Fabric, three-dimensional. See Three-dimensional woven fabric Fabricating. See Processing Fact and fiction. See Statistics, fact and fiction Factor of ignorance, 55, 309 Factor of safety. See Safety factor Facts and myths, 785 Fail-safe, 194 Failure analysis, 387, 754 Falling weight test, 415. See also Impact FALLO approach, 3, 589 Fastener, 340, 364, 400, 723, 870 Fatigue, 153, 190, 249, 385 crack, 197 endurance limit, 194, 588 heat generation, 195 rate dependence, 163, 195 strength, 265, 359 strength versus tensile strength, 193 testing, 191 transparent, 196 Fatigue-reinforced plastic, 191, 195 Fatigue-reinforced plastic film interlayer, 196 Fiber orientation, 118 Fick's law of diffusion, 279 Filament winding, 295, 326, 684 Filler, 63, 77, 287, 727 Fillet, 402, 795 Film, 473, 638, 888 Film orientation. See Orientation Financial. See Cost Fine tuning, 40 Finish, 38, 714 Finite element analysis, 55, 399, 401 Finite element modeling, 773 Fire, 282, 287, 404, 879 Firefighter, 286, 432 Fire Research Center, 288, 403

Fire Safety, 404 Fire, small and large scale, 285 Fit. See Press fit; Snap fit Flame dripping, 287 Flame retardant, 285, 420 Flammability, 9, 282, 422 Flash,524 Flexural modulus. See Modulus of elasticity, flexural Flexural properties, 146 Flexural strength, 135, 146 Flotation, 292 Flow, nonlaminar, 108, 114 Flow performance, melt. See Melt flow Fluorinated ethylene propylene plastic, 421 Fluorine gas, 276 Fluorocarbon, 472 Fluoroplastic, 39, 358, 420 Fluorosilicone elastomer, 471 Foam, 49, 295,475, 590, 624 Foam reentrant, 887 Foam reservoir molding, 679 Foam, structural. See Structural foam Foamed-in-place, 475 Foaming processes, 690 Foldable container, 317 Folded membrane, 232, 317 Forrest Products Laboratory, 122 Forming, 665 Fracture, 153, 197, 221, 387 Fracture mechanics, 196 Fracture resistance, 201, 204 Friction, 239 Frozen stress, 638 Fumes, 286, 431 Fungal deterioration, 282 Furan plastic, 421 Fusible core molding, 693 Future high performance plastic, 36, 97, 133 Garbage, 889 Gardner test, 414. See also Impact Gas injection molding, 624 Gas permeability, 104, 276 Gasket, 360,420 Gasohol, 414 Gasoline storage tank, 265 Gate location, 217, 808 Gauge length, 136 Gear, 40, 240, 359, 398 Gear pump, 628 General purpose polystyrene plastic, 449 Geometry factors, 196, 304, 346, 360, 368, 402, 699, 749, 837 German database, 530 Glass, 276, 293 Glass fiber, 66, 196, 241 Glass transition temperature, 70, 83, 195,431, 638 "Glasshopper," 45

972 INDEX

Golf course, 879 Gossamer Albatross airplane, 32 Government groups. See Appendix A Grafting, 75 Granulate, 524, 886 Granule, 71, 591 Graphics, 310, 783, 930. See also Appendix D Graphite fiber, 30, 82, 196,241,389,884 Gravitational force, 292 Grommet, 360 Guided load, 306 Hairline crack, 309. See also Crack Hand lay-up molding, 674 Handle, 649, 662 Hardness, 105, 239,244 Haze, 98, 100 Health care, 18 Heat capacity, 84, 88 condition, 201 dissipated, 695 distortion temperature. See Deflection temperature under load exchange, 116, 614 generation, fatigue, 195, 359 history, 602 release, 282, 359, 555 shield, 253. See also Ablation test, 260 welding, 727 Heterogeneous, 73, 122 Hevea Brazilienst, 458, 467 High-density polyethylene plastic, 276, 444 High molecular weight HOPE plastic, 444 High pressure laminate, 82 Highway, 372 Hinge, 346,349,841 Hoechst Celanese, 358, 472, 534, 883 Hole-blind, 819 Hollow channel, 319, 666 Homogeneous, 73 Homopolymer, 414 Honeycomb, 31, 49 Hooke's law, 139, 495 Hoop stress, 335, 342, 355 Hopper rail car, 45 Hot air duct, 235 Hull-deep submergence, 293 Humidity, 273, 388 Hydrocarbon plastic, 420 Hydrodynamic force, 292 Hygroscopic plastic, 201, 602 "Hypalon," 470 Hypervelocity impact. See Impact, strength Hysteresis effect, 140, 201, 359, 540, 565 "Hytrel," 472 I-beam, 314, 368 Identify plastic, 525, 547 Ignition temperature, 284

Ignorance factor. See Factor of ignorance Impact detector, 238 factors influencing, 153, 217 load, 210, 361 rapid load, 153, 201, 205, 209 strength, 96. See also Area under the stressstain curve tensile, 119, 210, 215 Implant, 45 Impregnated, 82 Improvements, 57 Incineration, 58, 889, 892 Indentor, 105 Induction welding, 727 Industrial pollutant, 273 Inert, 422 Infrared spectroscopy, 746 Initial modulus. See Modulus of elasticity, initial tangent Injection blow molding, 657, 749 Injection molding, 450, 610, 685, 712, 796, 808 circuitboard, 233 gas, 624 liquid,687 Insert, 340, 719, 797, 825, 854 Inspection, 606 Insulator, 20, 225, 275, 902 Integral hinge. See Hinge Intermittent behavior, 74, 130, 164 Interrelating product-resin-process performance, 55 Interrelating properties, plastics, and processing, 116,758 Intuitiveness, 125 Ionic bonding, 421 Ionomer plastic, 421 Ironing clothes, 38 Irradiation, 274 Isochronous graph, 157, 181,256 Isometric graph, 157, 181,256,769 Isoprene rubber, 70 Isostatic, 280 Isothermal heat, 201 Isotropic, 71, 121, 141, 368, 509. See also Orientation Iwd impact, 208,212, 594,797. See also Impact Joining, 841, 714, 869 Joint, 336, 339, 838 "Kelvar," 82 Ketone plastic, 422 Keyboard, 238 Kinetic energy, 210 Kinked molecular chain, 118 Knoop indentor, 105 Labeling, 319, 721 Landfill, 58, 891 Law, 402 Leaching, 891

INDEX 973

Leaf spring, 321 Life cycle, 908 Light scattering, 98 Light transmissability, 99 Linear thennal expansion. See Coefficient of linear thennal expansion Liquid crystal polymer, 68, 263,422 Liquid injection molding, 687 Liquid penetrant test, 748 Liquid plastic, 591 Living hinge. See Hinge Load analyzed, 305, 335, 565 Load bearing product, 303 Load deflection, 370, 380 Load, dynamic. See Dynamic load Load pipe externally, 326 Load relaxation. See Relaxation load Loading, impact. See Impact, load Loading, intermittent. See Intermittent behavior Loading, long-tenn. See Long-tenn behavior Loading mode, 130,201,362 Loading short duration, rapid. See Impact, rapid load Loading, short-tenn. See Short-tenn behavior Loading, static, 153 Locking fastener, 402 Longitudinal stress, 336 Long-tenn behavior, 134, 153, 241, 247 Lord Corporation, 372 Loudspeaker, 361 Low-density polyethylene plastic, 441 Low pressure laminate, 81 Lubricity, 107 "Lubricomb," 239 Luminous transmittance, 98 Luscher and Hoeg equation, 333 Machine conditions, 55 Machine improvements, 590, 887 Machining plastic, 727 Machining plastic guidelines, 594, 728 Magnetic resonance imaging, 882 Manufacturing. See Fabrication Manufacturing via computer. See Computer-aided manufacturing Markets, 9, 877 Mar resistance, 244 Material handling, 713 Material selection. See Plastic selection Materials comparison, 7, 8, 35, 97, 740, 931 Mathematical analytical solution, 48 Mathematical model, 614, 776 Matte finish, 38 Maxwell model, 175 Mechanical behavior, dynamic. See Dynamic mechanical behavior blending, 72 damping. See Damping interface, 342 loading, 195,488,870 properties, 86, 96, 127, 133, 930

properties, superior, 97 Medical, 45, 228, 882 Melamine fonnaldehyde plastic, 82, 416,423 Melt compressibility. See Compressible melt flow, 113, 123,209,591,617,630, 778,782 curve, 397 defects, 113, 618 trends, 116 fracture, 114 index, 110, 440 orientation, 117. See also Orientation pump, 628 shear rate, 107 structure, 115 temperature, 65, 83, 611, 614 viscosity, 653 Melting point, 68, 425 Membrane technology, 232 Memory in plastic, 117 Metal, 153, 203, 342, 368 Metallization, 238, 717. See also Coating Microbial degradation, 281 Microcracking, 272 Microtoming, 750 Microwave oven, 233 Migration, 276 Mining in the sea, 289 "MinIon," 400 Missile, 44, 235, 453 Mixing, 591 Mobay Chemical, 349 Modeling, 596, 763 Modeling cost, 898 Modulus of elasticity, 136, 510 compression, 145 creep, 155, 162 creep-apparent, 161, 175 dynamic, 154, 195, 385 flexural, 146, 835 initial tangent, 139, 304 secant, 140, 304, 349 shear, 149, 369 soil, 330 static, 385 temperature, 108 tension, 119, 139 theoretical, 97 viscoelastic, 175, 209 Modulus of rupture. See Aexural strength Mohr's circle, 308 Moisture, 273, 282, 416 Mold, 348, 391, 591,599, 615,758,808,934 Mold action, 841 Mold cooling analysis, 780, 934 Mold quotation guide, 600 Mold, time to produce, 601 Molded-in hinge. See Hinge Molded-in insert, 340, 825 Molded-in stresses, 116 Mold operation, 605

974 INDEX

Molding area diagram, 609 Molding variables versus performances, 616 Molding volume diagram, 609 Molecular motion and mechanical damping, 154 Molecular orientation, 118. See also Orientation Molecular weight, 109, 281 Molecular weight distribution, 109, 116 Molecule, 66 Molecule, kinked, 118 Molybdenum disulfide, 241 Moment of interia, 49,312,319,332,367,521, 835 Moment of inertia, polar, 324 Monomer, 71, 277 Morphology, 66, 84 Motion control, 371 Motor housing, 690. See also Engine "Mylar," 196 Myths and facts, 785, 889 Natural rubber, 70, 467 Neat plastic, 89, 147, 246 Necking, 136, 158, 203 Neoprene elastomer, 469 Netting die, 636 Newtonian liquid, 109 Nitrate plastic, 417 Nitrile elastomer, 469 Nitrogen, 276 Noise, 360, 371 Nomography, 102 Nonburning, 284 Nonconductive, 228 Noncrystalline plastic. See Amorphous plastic Nondestructive testing, 55, 746 Nonlarninar flow. See Flow, nonlarninar Nonlinear effect, 130 Non-Newtonian liquid, 109, 630 Nonplastication, 115 Notched, 191,208,212 Notch sensitivity, 106 Nuclear magnetic resonance spectroscopy, 747 Nuclear radiation, 282 Nylon plastic, 38, 119, 289, 400, 423, 873 O-ring seal, 292 Ocean environment, 288 Offset strain. See Strain, offset Oil resistance, 385 Olefin plastic. See Polyolefin plastic Optical analysis test, 98, 750 Optical plastic, 491 Orientation, 118, 166, 209, 268, 515, 616, 620, 638,658,734,791,797,888 Orthotropic, 122, 510 Oscillatory rheometer, 149 Osmosis, 883 Overdesigned, 126 Oxifluorination, 276 Ozone, 273, 283, 385 Packing, 12, 274, 432, 880

Paper, 82 Parabolic reflector, 239 Parison thickness control, 653 Parting line, 805, 843 Parts handling, 713 Parylene plastic, 425 Pascal. See Appendix B Pattern making, 596 Pellet, 591 Perfection, 403 Performance prediction, 55, 97, 125 Performance versus molding variables, 616 Permanent set, 136, 402 "Permasep," 883 Permeability, 277, 424, 732. See also Gas permeability; Vapor permeability Perms, 104 Phenol-formaldehyde plastic, 426 Phenolic plastic, 82, 426 Phenoxy plastic, 427 Photoelastic stress analysis test, 748 Photopolymerization, 599 Physical properties, 98 Piezoelectric, 237 Pinch-off, 654 Pipe, 47,269,326,358,432,644,845. See also Design pipe Pipe orientation. See Orientation Pipe thread. See Thread Plant overhead, 60 Plastic advantages and disadvantages, 56 and the world, 9, 879 behavior, 209. See also Intermittent behavior; Short-term behavior; Long-term behavior brittle, 209 classification, 525 comparisons, 535 consumption, 405, 591, 877, 916 database. See Computerized software and databases; Appendix D definition, 61, 109 development, I diversification, 878 future, 97, 915 geometry, 66 high performance, 97 hygroscopic. See Hygroscopic plastic identification guide, 525, 527 intermittent behavior. See Intermittent behavior long-term behavior. See Long-term behavior misapplications, 754 myths, 785, 889 processing. See Processing properties, 6, 61, 129, 276,405, 931 selection, 53, 126, 166, 525, 535 short-term behavior. See Short-term behavior structure, 66 thickness, 388. See also Thickness to metal, 358, 368 to metal wear, 246

INDEX 975 to plastic wear, 246 types, 6, 62, 166 volume, 901, 916 with a memory, 117 Plastic Bottle Institute, 896 Plasticity, 102, 591 Plasticizer, 204, 692 Plastics are bad, 785, 889 Poisson's ratio, 141, 151,323,336,510 Poisson's ratio, negative, 887 Polarized light, 99 "Polimotor," 880 Polyallomer plastic, 428 Polyamide plastic, 423, 473 Polyamideimide plastic, 263, 428 Polyanhydride plastic, 45 Polyarylate plastic, 430 Polyarylether plastic, 420 Polyaryletherketone plastic, 420 Polyarylsulfone plastic, 431 Polybenzimidazole plastic, 431 Polyblend, 73, 166 Polybutadiene elastomer, 469 Polybutylene plastic, 432 Polybutylene terephthalate plastic, 433 Polycarbonate plastic, 273, 432 Polyester plastic, 433 Polyester thermoplastic, 433 Polyester thermoset plastic, 435 Polyehter chlorinated plastic, 417 Polyetheretherketone plastic, 263, 422, 438 Polyetherimide plastic, 439 Polyetherketone plastic, 422, 436 Polyetherketoneetherketoneketone plastic, 431 Polyethersulfone plastic, 263, 440 Polyethylene chlorinated plastic, 418 Polyethylene plastic, 289, 440 Polyethylene terephthalate plastic, 433 Polyimide plastic, 445 Polyisoprene elastomer, 467 Polymat database, 530 Polymer. See Plastic Polymerization, 599 Polymethylmethacrylate plastic, 415 Polymethylpentene plastic, 446 Polynomial, 872 Polyolefin elastomer, 473 Polyolefin plastic, 421, 440, 446 Polyperfluoroalkoxyethylene plastic, 421 Polyphenylene ether plastic, 446 Polyphenylene sulfide plastic, 239, 447 Polypropylene plastic, 447 Polypyrrole plastic, 239 Polystyrene plastic, 23, 449 Polysulfide elastomer, 471 Polysulfone plastic, 449 Polytetraftuoroethylene plastic, 241, 360, 420 Polyurethane elastomer, 469, 472 Polyurethane plastic, 361, 451, 588 Polyvinyl acetate plastic, 454 Polyvinyl alcohol plastic, 454

Polyvinyl butyral plastic, 454 Polyvinyl chloride plastic, 489, 452 Polyvinylidene chloride plastic, 455 Polyvinylidene fluoride plastic, 238, 421, 455 Postage stamp, 881 Postforming, 641, 670 Postshrinkage, 388. See also shrinkage Powder, 591 Powder coating, 692 Practical approach. See Approach, practical Pragmatic approach, 126 Preload, 400 Press fit, 40, 163, 342 Pressure bag molding, 678 Pressure, velocity, 241 Pressure vessel, 325, 883 Printed wiring board, 232 Problem solving, 914 Process analysis tools, 777 Process control, 392, 603, 614, 627, 653 Process selection, 166,591,695. See also Selection Processing, 107, 116, 209, 217,524,589,675, 759, 785, 934 Processing and tolerance, 391, 596, 640 Processing cycle, 595 Processing equipment improvement, 887 Processing guidelines, 2, 682, 703 Processing variables, 55 Product life cycle, 908 Product shape, 129 Product size, 699 Product specification, 398 Productivity, 611, 759,879,903 Professional groups. See Appendix A Profile, 475, 647. See also Extrusion Property, theoretical, 97 Propionate plastic, 417 Proportional limit. See Tensile proportional limit Protocol, communication, 711, 888 Protocol, fire test, 404 Prototype, 164,341,596,762 Pseudo-elastic method. See Design, pseudo-elastic method Publications. See Appendix A Pultrusion molding, 683 Pump, 295, 420 Punching, 105 Purging, 615 Pyrolyzing, 885 Quality auditing, 753, 761 Quality control, 591, 731, 754 Quotation, 600 Radiation, 274, 902 Radio frequency interference, 228 Radio frequency welding, 725 Radiography, 746 Radome, 13, 235, 301

976 INDEX Rain erosion, 300

Rapid transit vehicle, 371 Reaction injection molding, 682, 687 Reactive polymer, 76 Reactor, 71 Reciprocating screw, 610 Recovery rate, compression, 145 Recreation, 9 Recycling, 24, 58, 524, 602, 889, 892 Reduced cross section, 136, 158 Reentrant foam, 887 Refractive index, 98 Refrigerator, 42 Regrind. See Granulate Regulation, 402, 732, 897 Reinforced plastic, 81, 121, 265,293,299,310,

421,435,493,668,670, 856 definition, 81 493 fatigue testing, 191, 195 film interlayer, 196 hysteresis measurement, 584 pipe, 326 processing, 673, 698 spring, 321 weep, 334 Reinforcement, 63, 77, 501, 673 Relative thermal index, 261 Relaxation load, 114, 153, 164 Reliability control, 754 Residence time, 602 Residual deflection. See Deflection Residual monomer, 277 Residual stress, 791, 872 Resin. See Plastic, definition Resin-transfer molding, 679 Resource utilization, 762 Responsibilities, 51 Retardation, 140 Reynold's number, 395 Rheology, 107 Rheometer, oscillatory, 149 Ribs, 50, 310,316, 331,790,833 "Riteflex," 472 Rivet, 362 Robots, 888 Rockwell hardness, 245 Rotational molding, 691, 865 Roving, 510 Rubber, 458, 467. See also Elastomer Rupture strain. See Strain, rupture Safety factor, 54, 125, 332. See also Design, safety factor Safety factor guidelines, 309 Safety glass, 18 Sailboard, 9 Salt spray test, 273 Sandwich structure, 31, 49, 235, 295, 368 Scrap, 524 Scratch, 244, 415, 491 Screen pack, 627

Screw design, processing plastic. See Design, screw Screw thread. See Thread Sea environment, 288 Seal, 341, 360 Sealing. See Hot sealing Secant modulus. See Modulus of elasticity, secant Secondary operations, 711 Selection. See Plastic selection; Process selection; Test selection Selection worksheet, 535 Self-extinguishing, 284 Self-latching, 318 Self-tapping screw, 358 Sensor, 239 Sewing machine, 41 S-glass, 196 Shape, 49, 129, 235, 319, 663, 699, 789, 843 Shape factor, 360, 373, 591 Sharkskin, 114 Shear modulus. See Modulus of elasticity, shear Shear properties, 149, 382. See also Melt, shear Shear rate, melt. See Melt, shear rate Shear strength, 135, 151 Shear strength, direct shear, 150, 325 Shear stress. See Stress, shear Shear yield strength, 151 Sheet, 473, 645 Sheet molding compoond, 416, 494 Sheet orientation. See Orientation Shielding, 228 Ship, 11,289,371,879 Shock, 371 Shoe, 195,587,769 Shore durometer, 105, 245 Short-term behavior, 135,201,209,241,247, 355 Shot peening, 265 Shot-to-shot variation, 615 Shotgun shell, 881 Shrink allowance, 390, 393,596,607,616,656, 661, 791 Shrinkage, 70, 115, 123, 388, 758, 829, 934 Silicone elastomer, 471 Silicone plastic, 233, 241,455 Simple support load, 306 Sink mark, 624, 798, 804, 836 Sinusoidal deformation, 153 Sketches. See Graphics Skin, integral, 49 Skis, 9 Smoke generation, 264, 282, 286 Snap fit, 12,345, 402, 838 Snap switch, 232 S-N curve, 191 Software. See Computer; Databases; Appendix D Solar radiation, 272, 297 Sole of shoe. See Shoe Solid modeling, 772 Solid-phase forming, 668 Solid waste, 58, 889, 896

INDEX 977 Solubility, 66 Solvent, 429, 714, 869 Sound absorbing, 361 Space communication antenna, 15 Spadoning, 38 Space environment, 297, 371 Spangler Iowa equation, 331 Specific gravity, 98 Specific heat, 285 Specification, 398, 402, 594, 732 Spherical shape, 235 Spin welding, 725 Splay, 366, 624 Spray-up molding, 674 Spray-zinc arc, 230 Spring, 163, 195, 380, 400. See also Design, spring Staking, 325 Stamping, 666, 683, 717 Stamps, 881 Standard,732 Stapler, 321 Static loading, 153 Statistics, 126, 193, 223, 212 Statistics, fact and fiction, 890 Steel, I, 127,247,328 Stereolithography, 598 Sterilization, 274 Stiffness, %, 129, 319, 368, 380, 836 Stiffness factor, 331, 370, 521 Stiffness-to-weight ratio, 310 Storage tank, 265 Strain analysis, 164, 333, 339 compression, 145, 375 creep, 155 damping, 154 dynamic. See Dynamic load fatigue, 190 impact. See Impact induced load, 305, 336 initial, 158, 385 limit, 304 offset, 137 plastic, 136 rapid, 201, 215 recovery, 164 rupture, 159 shear, 151, 523 tensile, 136, 336 thermal contraction, 333 time dependent, 159 Strain-to-first crack, 334 Strapping, 474, 664 Strength, 91. See also Tensile strength; Compression strength; Flexural strength; Fatigue strength; Shear strength; Buckling strength Strength, regression, 161 Strength-to-weight ratio, 310 Stress, 872

allowable working, 178 amplitude, 191, 197 analysis, 304, 820, 872 beam bending, 312, 315 bending, 146, 359 compression, 147,400 concentration, 217, 794, 799 constant, 201 corrosion, 265, 295 cracking, 156, 270 crazing, 159, 270 creep, 155 damping, 154 deflection. See Deflection engineering, 136 fatigue, 190 frozen, 638 harden, 873 history, 153 hoop. See Hoop stress impact, 362 imposed, 167,825 longitudinal, 336 multiaxial, 308 nonrecoverable, 136 number of cycles curve, 335 peak,359 relaxation, 114, 118, 162, 360 residual. See Residual stress sandwich beam, 369 shear, 110, 149, 323, 325, 355, 387 spring, 321 tensile, 136, 147, 400 thermal. See Thermal stress time, creep, 157 true, 136 Whitening, 160, 272 zero, 1% Stress-strain behavior, 136, 147, 157, 181,207 curve/data, 57, 139, 141, 143,273,335,459 impact, 361 polynomial, 872 Stresses-molded-in, 820 Stretching. See Orientation Structural foam, 23, 295, 365, 474, 709 Structure, 31, 50, 293, 297, 303, 368, 380, 667, 837 Styrene, 418 Styrene-acrylonitrile plastic, 412, 883 Styrene butadiene elastomer, 468 Styrene copolymer elastomer, 473 Submarine, 289 Success by design, 51 Super polymer, 422 Support condition, 305 Surface finish, 700 Surface modeling, 767 Surfboard, 10 Suspension system, 378, 385 Swelling, 66, 630

978 INDEX Swimming pool, 436 Switch, 232, 416 Symbols, 143 Synergistic effects, 72 Syntactic foam, 295, 475 Tank,44,46, 265, 647, 883 Tape, 639, 643 Taper, 660, 810 Technical cost modeling, 898 "Techturf," 879 Telephone, 412 Temperature, 203,253, 297,903 Temperature, glass transition. See Glass transition temperature Temperature index, 95 Temperature range, 92, 428 Tennis, 9 Tensile bridging, 338 elastic limit, 136, 138, 387 elongation. See Elongation, tensile hysteresis. See Hysteresis, effect impact. See Impact, tensile modulus of elasticity. See Modulus of elasticity, tensile properties, 136 property symbols, 143 proportional limit, 137 strain. See Strain, tensile strength, 36, 135, 138, 2ll, 724 strength, theoretical, 97 strength at break, 138 strength versus fatigue strength, 193 stress. See Stress, tensil stress crazing. See Crazing yield strength, 137, 400, 402 Tentering frame, 638, 642 Terpolymer, 71 Test assumptions, 308 Test method, 734, 735 acoustic, 380 appliance safety, 403 artificial, 273 chemical, 272 classification, 525 coefficient of linear thermal expansion, 89 compressive properties, 145 creep-rupture properties, 156, 160 deflection temperature under load, 94 density, 98 drop-weight, 361 Dynatup, 215 elastomer, 458 electric, 225, 422 electromagnetic, 228 falling weight, 215 fatigue, 191 flame, 287, 403 flexable culvert, 331 flexural, 146

flow, 417 foundation, 332 friction and wear, 241 Gardner, 215 gas transmission, 279 gasket, 360 glass transition temperature, 83 hardness, 71, 105 heat capacity, 88 heat test, 260 hydrostatic, 334 impact, 209,212, 217 limiting oxygen index, 287 long-term behavior, 153 melt index, llO modulus, 175 pipe deflection, 330 plastic selection, 540 safety factor, 332 shear, 149 specific gravity, 98 spring, 322 stress cracking, 271 temperature index, 95 tensile, 136 tensile impact, 215 test rate, 143 thermal conductivity, 87 thermal diffusivity, 88 thermal insulation, 87 torsion, 152 wall thickness, 747 water absorption, 104, 272 wear. See Test method; Friciton; Wear weather, 273 Test rate, 143 Test selection, 755. See also Selection Test simplification, 308, 732 Test specimen cross-section, 136, 735 Testing, 272, 292, 334, 731 high speed, 209 nondestructive. See Nondestructive testing short-term, 134 Testing via computer. See Computer-aided testing Theoretical versus actual values, 97 Thermal aging, 260 conductivity, 84, 87 contraction, 333, 377 diffusivity, 84, 88 expansion, 84, 90, 362, 377 failure, 196, 201 index, relative, 261 insulation, 84 properties, 83, 903 stress, 90, 297, 362 Thermoanalytical test, 735 Thermodynanrics, 116, 201 Thermoforming, 666, 886 Thermoforming orientation. See Orientation Thermogravimetric analysis test, 740

INDEX 979

Thennomechanical analysis test, 744 Thennoplastic, 64, 253, 611 Thennoplastic elastomer, 70, 371,472 Thennoset, 64, 66, 253, 611 Thennoset elastomer, 70, 467 Thickness, 790 Thread strength, 355, 358, 680, 824, 831 Three-dimensional woven fabric, 122 Throwaway aspect, 890 Time dependent loading, 130 Titanium, I Toilet flushing, 883 Tolerance, 387, 398,416,596, 606,629,647, 660,761,791, 802,834,846,855,864,869 Tooling, 596, 701, 758 Torque, 832, 870 Torsion pendulum, 149 Torsion properties, 152, 323, 355, 380, 401 Toughness, 96, 105,203,415,421,432. See also Area under the stress-strain curve Towline, 292 Toxicity, 282, 286,431, 889 Trade association. See Appendix A Trade magazine. See Appendix A Trade name, 76,405, 533. See also Appendix C Training, 786. See also Educational infonnation Transducer, 237 Transfer molding, 679 Transmission, water vapor, 104 Transparency, 98, 415, 491 Transportation, 18 Trash compactor, 361 Trial-and-error, 399 Troubleshooting, 603 True stress. See Stress, true Turf, 879 Turnaround time, 762 Two-shot molding, 623, 719 ''Tyrin,'' 470 Ultrahigh molecular weight polyethylene plastic,

444 Ultrasonic test, 748 Ultrasonic welding, 272, 723 Ultraviolet radiation, 415 Uncertainties. See Safety factor Undercut, 348, 816, 843 Underground, 44, 265, 429 Underwater, 288, 293, 330 Underwriters Laboratory, 95, 732, 739 Uniaxial direction, 118 See also Orientation Unidirectional, 122. See also Orientation Unsaturated polyester plastic, 416 Unscrewing, 355 Urea fonnaldehyde plastic, 416, 423,456 Urethane. See Polyurethane Utensil, 420 Vacuum bag molding, 677 Vacuum metalization, 230 "Vamac," 471

Vapor barrier, 104 Vapor penneability, 104 Variables combined, 399,495 "Velcro" strip, 622 Vent, 814 Vessel, 325 Vibration, 239, 372 Vibration welding, 725 Vickers, 105 Vinyl dispersion, 692 Viscoelasticity, 107,114, 140, 155, 174,304 Viscosity, 870 Viscous, defonnation, 107, 174 Viscous modulus. See Modulus of elasticity, creep-apparent Vision system inspection, 749, 753 Void,217 Void content determination, 686 Volatile, 115 Volume resistivity, 229 Wall thickness, 187, 189, 201, 355, 598, 747, 780, 798, 801 Warpage, 70, 261, 510, 616, 624, 798, 804 Washer, 401 Waste, 883. See also Solid waste economic consideration, 889 public education, 890, 896 regulatory support, 891 Waste-to-energy incineration, 891 Water, 924 Water absorption, 104, 274 Water treatment, 883 Water vapor transmission, 104 Watertight assembly, 233, 292 Wear, 239, 420, 424, 628, 882 Wear, plastic to metal, 246 Wear, plastic to plastic, 246 Weather resistance, 272, 414 Weep, 334 Weight saving, 37, 365, 890 Welding, 336, 362, 723, 869 Weldline, 217, 618, 695, 734, 813 Wetability, 884 Wire frame modeling, 757, 765 Wood, I, 122, 161, 311 Woods Hole vehicle, 295 Wodd without plastic, 9 Wright-Patterson Air Force Laboratories, 122 X-ray spectroscopy test, 747 "Xydar," 263 Yield. See Tensile, yield strength; Compressive yield strength Young's modulus. See Modulus of elasticity Zero defect, 887 Zinc-arc straying, 230 "Zytel," 873, 882

~ PRCAlECT TEAM FEASIBILITY

-

H

STUDY

Dimensions. structural loads. gov'Vlndustry standards. service environments, etc__ 1

Identify specific functions

Target quantity. cost and production schedule

Apply available experience

I Manual approach

l-.{ COnceptual product layout

Apply design creativity

Computer approach

I

r1 Formulate plan'

R

PRELIMINARY DESIGN ANALYSIS

r-y

I

Dimensions & tolerances

Geometric drawings

or graphic analYSis

Engineering analysiS

Solid

4~

Image manipulation

r-

I

Surface

c-j

Structural Integrity

Environment

---I Physical integrity

Mechanical

simulatkm

I

J---@

Dynamic

Finite element modeling

J

Minimum weight/cost

n

Ae_c.etc.1

--1

rl OPTIMIZE DESIGN

r

Mok:Jfdie destgn

Material selection

I

Fabrication analysis

4

J

Shrinkage analysis Cycle timeJ

I

detai.led/individual orthotropic drawings

Set up safety factors to meet product functions

I Nonstandard parts I Standard parts

~ Provtde

-1

I

selection

---i ~

Material

I

----i ProcesSlequipmentltools _I -1

Produce service per10rmance

I

--I Plant personnel capability I

iii

rl H

~ 1l

Release tooling (mOld/die) for manufacture Set up processing specifications

I

I

rl Equipment I H Material handling I

'iii

15

I j

MANUFACTURING~

H

Automationtrobotics

Plant layout

H

n

ANALYSIS

Y

'12l

§

I

Safety procedures

W

Manufacturing processing specifications

I

I

Troubleshooting gUide Testing/quality control

Plant personnel capability evaluation

f rl

OPTIMIZE &

DOCUMENT

PRODUCTION

~

I

I I I

Ensure product quality Document good manufacturing practice (GMP) procedure

H

I

I

I

Preventive maintenance

Machine operations

H

Update plant personnel training

I

H DevelOp production cost models

I

Finalize testing & quality conlrol

I

p od hed I r uct output sc ue

~ ~:~~:ii~: :~=~~: Inventory schedule

Y Documentation for management to ensure meeting delivery schedules and profitability I J

I

PRODUCT RELEASE

Ensure meeting all product functions Set up value analysis (VA)

I

II

I

Cost analysis

rl Matt flow analysis H Tool thermal analysis I - Tool part I

-1

Databases andlor available information

Process selection

After start of production. analyze ccmplate design (again) to change design/production/safety factors in order to reduce costs

I

I