Std-124

OISD-124 Amended edition FOR RESTRICTED CIRCULATION No.

PREDICTIVE MAINTENANCE PRACTICES

OISD RECOMMENDED PRACTICES-124 First Edition, March 1990 Amended edition, August, 1999

Oil Industry Safety Directorate Government of India Ministry of Petroleum & Natural Gas

1

OISD RP-124 First Edition, March, 1990 Amended edition, August, 1999 FOR RESTRICTED CIRCULATION No.

PREDICTIVE MAINTENANCE PRACTICES

Prepared by

COMMITTEE ON INSPECTION OF ROTARY EQUIPMENT

OIL INDUSTRY SAFETY DIRECTORATE 2ND FLOOR, “KAILASH” 26, KASTURBA GANDHI MARG NEW DELHI-110 001

2

NOTE

OISD publications are prepared for use in the oil and gas industry under Ministry of Petroleum and Chemicals. These are the property of Ministry of Petroleum and Chemicals and shall not be reproduced or copied and loaned or exhibited to others without written consent from OISD. Through every effort has been made to assure the accuracy and reliability of the data contained in these documents, OISD hereby expressly disclaims any liability or responsibility for loss or damage resulting from their use. These documents are intended only to supplement and not to replace the prevailing statutory requirements.

Note 1 in superscript indicates the modification/ changes/addition based on the amendments approved in the 17th Safety Council meeting held in July, 1999.

3

FOREWARD The oil industry in India is 100 years old. As such a variety of practices have been in vogue because of collaboration/association with different foreign companies and governments. Standardisation in design philosophies and operating and maintenance practices at a national level were hardly in existence. This, coupled with feed back from some serious accidents that occurred in the recent past in India and abroad, emphasised the need of the industry to review the existing state of art in designing, operating and maintaining oil and gas installations. With this in view, the then Ministry of Petroleum & Natural Gas in 1986, constituted a State Council assisted by Oil Industry Safety Directorate (OISD) staffed from within the industry, in formulating and implementing a series of selfregulatory measures aimed at removing obsolescence, standardising and upgrading the existing standards to ensure safer operations. Accordingly, OISD constituted a number of Functional Committees of experts nominated from the industry to draw up standards and guidelines on various subjects. The present document on "Predictive Maintenance Practices" was prepared by Functional Committee on Inspection of Rotary Equipment. This document is based on the accumulated knowledge and experience of Industry members various manuals, national and international codes of practices. This document is meant to be used as a supplement and not as a replacement for existing codes, standards and manufacturers' recommendations. It is hoped that provision of this document if implemented objectively may go a long way to improve safety and reduce accident in the oil and gas industry. Suggestions for amendments, if any, to this document should be addressed to: The Coordinator, Committee on "Inspection of Rotary Equipment", Oil Industry Safety Directorate (OISD), 2nd Floor, “Kailash”, 26, Kasturba Gandhi Marg, New Delhi-110001. This document in no way supersedes the statutory regulations of CCE, Factory Inspectorate, or any other statutory body which shall be followed as applicable.

4

COMMITTEE ON INSPECTION OF ROTARY EQUIPMENT List of Members ____________________________________________________________ Name Designation & Status Organisation ____________________________________________________________ 1. Sh.K.Gopalakrishnan

Sr. Maint. MGR-CRL

Leader

2. Sh. B.P.Sinha

Chief Project MGR-MRL

Member

3. Sh. Chotey Lal

Chief Engineer-ONGC

Member

4. Sh. R.C.Chaudhary

Office Engg. MGR-BPCL

Member

5. Sh. K.M. Bansal

Chief Maint. MGR-IOC

Member

6. Sh. Ehsanuddin

Director-OISD

Member

7. Sh. R.M.N. Marar

Joint Director-OISD

Member Coordinator

__________________________________________________________________ In addition to the above, several other experts from industry contributed, in the preparation, review and finalisation of this document.

5

PREDICTIVE MAINTENANCE PRACTICES CONTENTS

SECTION 1.0 Introduction 2.0 Scope 3.0 Vibration 3.1 Vibration parameters 3.2 Criteria for Measurement 3.3 Filter out and filter in 4.0 Vibration measurement and severity standards 4.1 Vibration measurement 4.2 Vibration recording 4.3 Vibration severity standards 5.0 Vibration analysis 5.1 Data acquisition 5.2 Data interpretation 6.0 Shock pulse measurement 6.1 General 6.2 Unit and pick-up points 6.3 Terminology 7.0 Frequency of vibration & shock pulse measurement 7.1 Classification 7.2 Frequency of measurement 8.0 References

6

PREDICTIVE MAINTENANCE PRACTICES

1.0

INTRODUCTION The Rotary equipment plays a vital role in hydrocarbon processing industry. Timely inspection and maintenance of Rotary equipment will go a long way in ensuring safer operations of the installations in Oil Industry. 2.0

SCOPE

This document covers the recommended practices and procedures for carrying out predictive maintenance of rotating machinery. Predictive maintenance covers monitoring of vibration and shock pulse measurement levels, comparing them with standard values and analyzing the readings taken to find out the real cause of machinery problem. 3.0

b) PERIOD: The time required for completing one cycle of vibration is called the period of vibration. c) CPM (CYCLES PER MINUTE): The number of cycles repeated in a given interval of time normally minute (because the rotating speed is expressed in RPM), is the frequency of vibration, which is expressed in the abbreviated form as CPM. d) VIBRATION DISPLACEMENT: Displacement at any instant of the cycle is the distance traveled by the vibrating part from one extreme limit of travel to the other extreme limit of travel and is selected for measurement which is referred to as the 'peak to peak' value of displacement. It is expressed in microns in metric units and in mils. in British Units. e) VELOCITY: The velocity of the vibrating part is constantly changing as displacement changes. The peak value, which represents the most severe condition during a cycle, is selected for vibration measurement. It is expressed in mm/sec. and inch./sec. in Metric units and British units respectively. f) ACCELERATION: Acceleration is another important characteristic of vibration and peak value is measured which is normally expressed in 'g's. g) PHASE: Phase is another important characteristic of vibration which is defined as the position of vibrating part at a given instant with reference to a fixed point or another vibrating part.

VIBRATION

Vibration is the motion of a machine or machine part back and forth, from the position of rest. The cause of vibration must be a force which is changing in either its direction or amount. It is the force which causes vibration and the resulting characteristics will be determined by the manner in which the forces are generated. Hence each vibration has its own typical characteristics. 3.1

VIBRATION PARAMETERS

Referring to figure 1 where the movement of weight is plotted against time, various parameters can be defined as: a)

CYCLE: The motion of the weight from its neutral position to the top limit of travel back through the neutral position to the bottom limit of travel and its return represents one cycle of motion. Hertz (Hz) is a unit of measurement for vibration frequency. One Hz is equal to one full vibration cycle of oscillation per second. Note 1

7

The machinery vibration is mostly complex, consisting of components at many frequencies, i.e. the machinery does not vibrate at a single frequency but vibrate at many frequencies. (The frequency is decided by the troubles causing vibration).

3.2 CRITERIA FOR MEASUREMENT The displacement, velocity and acceleration of vibration are referred to as the Amplitude of vibration. Displacement, velocity and acceleration of vibration are directly related. Vibration velocity is directly proportional to the displacement and the frequency as shown in Equation 1 and Vibration acceleration is directly proportional to the displacement and frequency squared as shown in Equation 2.

The total amplitude of vibration measured is the vector sum of vibrations at different frequencies. This is termed as the ‘Filter out' amplitude. When the vibration is complex, we will have to analyse the vibration to know the amplitude at different frequencies of interest. For this purpose, vibration analysers are made use of. With this, by tuning the filter, vibration amplitude at different frequencies can be measured. This is termed as ‘Filter in' amplitude.

V Peak =52.30D (F/1000) x 10 -3 ....(1) g peak =5.6 D (F/1000) 2 x 10 - 4 ....(2) V peak = Vibration velocity in mm/s peak g peak = Vibration analysis peak D= “Peak to peak" displacement in microns F = Frequency in CPM

4.0

The forces, which cause vibration, are generated through the rotating motion of the machine parts and these forces change in amount and direction and the rotating part changes its position with respect to rest of the machine. Hence the frequency of the vibration produced would be related to the rotating speed of the part which has the trouble. Because of this, it is essential to know the vibration frequency for analysis.

4.1

VIBRATION MEASUREMENT AND SEVERITY STANDARDS VIBRATION MEASUREMENT

Electronic instruments for measuring machinery vibration are generally classified as meters, monitors and analysers. a) A vibration meter is a portable device and used for periodic vibration checks on machinery to determine the overall machine vibration level. b) A vibration monitor is similar in function to a vibration meter, but is permanently installed to provide continuous monitoring of equipment vibrations. Note 1 c) A vibration analyser includes a tunable filter for separating the individual frequencies of complex vibration. This can measure and record all vibration amplitudes at different frequencies.

Vibration severity is a function of both the distance the vibrating part moves from its position of rest (peak displacement) and the number of times the vibrating part moves about its position of rest in unit time (frequency). Since vibration velocity is a function of the displacement and frequency, unfiltered vibration velocity should be recognised as a direct measure of vibration severity. Vibration acceleration is directly related to the force causing vibration in the machine. Since vibration acceleration is a function of the displacement and frequency squared, a very small displacement at very high frequency may be due to a large vibrating force present in the machine. Hence, vibration acceleration measurements are recommended for vibration frequencies above 60,000 CPM.

4.2 VIBRATION RECORDING Suggestive formats for recording vibration measurements are given in Table 1.1 and 1.2. Table 1.1 should be used for vibration recording and Table 1.2 should be used for vibration analysis purposes. 4.3 VIBRATION SEVERITY STANDARDS The aim of vibration measurement is to find out an acceptable limit for safe operation of a given machine.

3.3 FILTER OUT AND FILTER IN The vibration of a machine may not always generate harmonic motion as the weight suspended from the spring does.

For centrifugal pumps, centrifugal compressors and steam turbines acceptable

8

vibration values are as given in Table 1.3 5.0

reference mark under strobe light. c) Since many machine troubles have similar characteristics and several troubles may be present in a machine simultaneously, it becomes necessary to choose between several possibilities.

VIBRATION ANALYSIS

The purpose of vibration analysis is to identify the specific machinery problem. For this purpose, the initial reading taken during the commissioning of the equipment should be taken as the most ideal base line data". After that, whenever overall machinery vibration has revealed a significant increase in “base line data", vibration analysis should be carried out to pin-point the machinery problem. The vibration analysis procedure is divided into two steps:

The vibration and noise identification chart given in Table 1.6 gives a comprehensive listing of most of the common problems encountered and provide a relative probability rating number which provides an indication of which trouble is the most likely set of circumstances. If the analysis does not give an indication of the problem, signature analysis should be carried out.

a) Data acquisition b) Data interpretation

6.0

SHOCK PULSE (SPM) 6.1 GENERAL

5.1 DATA ACQUISITION Vibration data can be obtained for analysis by applying the techniques given below: a) b) c) d) e) f) g) h)

Amplitude vs frequency Amplitude vs time Amplitude vs frequency vs time Time wave form Orbits Amplitude vs phase vs rpm Phase analysis Mode shape interpretation.

5.2

DATA INTERPRETATION

Shock pulse measurement is based on monitoring the mechanical impacts caused by bearing damage and operating condition problems. 6.2 UNIT AND PICK-UP POINTS The sensitivity of the SPM method is such that the shock pulses generated by a typical antifriction bearing increase upto 1000 times from when the bearing is in good condition to the condition when the same is about to fail. For covering this large range, a logarithmic scale is used and the shock pulse values are expressed in decibels (db).

Once the data mentioned in 5.1 are obtained, the next step is to interpret the data thus obtained for identifying the machinery problem. This is done by comparing the reading with the characteristic vibration due to typical machinery troubles.

Shock pulses are generated mainly in the load zone of the bearing and spread spherically from the point of impact through the bearing, its housing and adjacent machine parts. The shock pulses are dempened when they pass an interface or are forced from their straight path.

a) The chart given in Table 1.4 lists the vibration frequencies normally encountered in terms of rpm and the possible cause of vibration. Referring to the chart, we can identify the part causing trouble. b) The vibration identification given in Table 1.5 lists most common of vibration together with its relation amplitude, frequency and position of

MEASUREMENT

6.3 TERMINOLOGY Following are the terminology associated with shock pulse measurement. i) Initial Shock Value (dbl) Even a newly installed and properly lubricated bearing generates shock pulses.

chart cause to the phase

9

10

11

12

13

This is known as Initial Shock Value (dbl). This value is primarily dependent on the rotating speed and the bore dia as shown in Table 1.7 ii)

Shock Value

The absolute strength of the shock pulses emanating from the bearing is known as Shock Value (dbsv) iii) Normalised Shock Value The increase of shock value (dbsv) above initial shock value (dbl) is defined as Normalised Shock Value (dbn). The normalised measuring scale is used for measuring shock pulse value in the shock pulse meter. The normalised measuring scale starts from dbl and shows only that part of shock value which is directly related to the condition of the bearing being monitored. iv)

Carpet Value

Surface roughs will cause rapid sequence of minor shock pulses which together constitute the shock carpet of the bearing. The magnitude of the shock carpet on the normalised measuring scale is expressed by the Carpet Value (dbc). This value helps to analyse the cause of reduced or bad operating condition. v)

Maximum Value

Any damage in the bearing will cause single shock pulses with higher magnitudes at random intervals. The highest shock pulse value measured on a bearing is called its Maximum Value (dbm). This determines the operating condition of bearing.

14

TABLE 1.4 VIBRATION FREQUENCIES AND LIKELY CAUSES Most Likely Other Possible Causes and causes Remarks

Frequency in terms of RPM ____________________________________________________________________________ 1xRPM Unbalance 1. Eccentric Journals, Gear or Pulleys 2. Misalignment or Bent shaft if high axial vibration 3. Bad belts if RPM of belt 4. Resonance 5. Reciprocating forces 6 Electrical problems ____________________________________________________________________________ 2 x RPM Mechanical 1. Misalignment-if high axial vibration looseness 2. Reciprocating forces 3. Resonance 4. Bad belts if 2x RPM of belts _____________________________________________________________________________ 3 x RPM Misalignment Usually a combination of misalignment and excessive axial clearances (looseness) _____________________________________________________________________________ Less than Oil Whirl 1. Bad drive belts 1 X RPM (less than 2. Background vibration half RPM) 3. Sub-harmonic resonance 4. "Beat" vibration ____________________________________________________________________________ Synchronous Electrical Common electrical problems include rotor bars, (A.C. line problems eccentric rotor, unbalanced phases in poly-phase frequency) systems, unequal air gap _____________________________________________________________________________ 2 x Synch. Torque Rare as a problem unless resonancce is excited. frequency Pulses ____________________________________________________________________________ Many times Bad gears, AeroGear teeth times RPM of bad gear, number of fan RPM(Harmo- dynamic forces/ blades times RPM, No. of impeller times RPM, may nically mechanical loose occur at 2,3,4 and sometimes higher harmonics if related ness, reciprocating severe looseness frequency) forces _____________________________________________________________________________ High freBad anti-friction 1. Bearing vibration may be unsteady-amplitude and quency (not bearings frequency. harmonically 2. Cavitation, recirculation and flow turbulence cause related) random, high frequency vibration 3. Improper lubrication of journal bearings (friction excited vibration) 4. Rubbing ___________________________________________________________________________

18

19

20

21

22

vi)

Instrument and Evaluation Chart ii)

The basic measuring equipment consists of the shock pulse transducer with the probe which picks up the shock pulses and the shock pulse meter which measures the magnitude. Once the readings are taken and tabulated as shown in Table 1.8., the intensity can be checked from the shock pulse diagram given in Figure 2.0

iii) iv) v) vi)

7.0

FREQUENCY OF VIBRATION AND SHOCK PULSE MEASUREMENT CLASSIFICATION

7.1

The entire equipment in oil industry should be classified into three categories: i) ii) iii)

xii)

The equipment that can cause unit shutdown and the failure of which will lead to release of hydrocarbons should be classified as Critical equipment. Mostly these equipment will not be having any spare equipment. The equipment that can cause only production loss should be classified as Semi-critical equipment. Rest of the equipment should be classified as Noncritical. 7.2 FREQUENCY OF MEASUREMENT The following frequency of vibration monitoring should be adhered to in order of criticality of equipment: Critical equipment: Once in a week Semi critical equipment : Once in two weeks Other Equipment : Once in a month The frequency of measuring shock pulse value depends on the magnitude of maximum shock value. The frequency recommended by manufacturer should be followed as given below : Frequency 1-3 months 1-2 months Daily

8.0 REFERENCES The following codes, standards, and publications have either been referred to or used in the preparation of this document and the same shall be read in conjunction with this document: i)

viii) ix) x) xi)

Critical Semi-critical Non-critical

dbm 0-20 20-35 35-65

vii)

An introduction to Machinery Analysis and

23

Monitoring BS 4675 : Part I : 1976 (ISO2372) Mechanical Vibration in rotating and reciprocating machinery API 610 : Centrifugal pump for general refinery service API 611 : General purpose steam turbine for refinery service API 612 : Special purpose steam turbines for refinery service API 617 : Centrifugal compressor for general refinery service Condition monitoring of roller bearings with shock pulse meter IRD Mechanalysis Advanced Training Manual Predictive maintenance Manual of Indian Oil Corporation Sawyer’s turbomachinery Handbook Instruction Manual of Hard bearing balancing machine Turbomachinery Handbook published by Hydrocarbon Processing

24