Well Testing

Introduction to Well Testing Log-Log

Derivative

Electric line section

Battery section

Pressure maintenance No-flow boundary

C-5 C-8

Microcontroller

UNIGAGE recorder section

Sensor sub section

CQG sensor

Customized Quartzdyne sensor

Sapphire sensor

Radial flow

Superposition Paris, July 1996

Schlumberger

Wireline & Testing

C-1 C-3 C-7

EEPROM data memory

Mode led re servo ir area

C-4

I n t r o d u c t i o n t o W e l l T e s t i n g

Paris, July 1996

Schlumberger Wireline & Testing

Introduction to Well Testing

Table of Contents

Introduction to Well Testing (Aug 1996)

TOC - 2

Schlumberger

Section 1.0

Basic Reservoir Engineering

1.1

Introduction - (Course Objectives)........................................................................... 1-3

1.2

History

.................................................................................................. 1-3

1.3

Geology Recap

.................................................................................................. 1-4

1.3.1 1.3.2 1.3.3

1.3.4

1.4

Interacting Forces, Saturation And Displacement...................................................... 1-9 1.4.1 1.4.2 1.4.3 1.4.4 1.4.5 1.4.6 1.4.7 1.4.8

1.5

General .................................................................................................. 1-9 Surface And Interfacial Tension....................................................................... 1-9 Wetting .................................................................................................. 1-10 Capillarity.................................................................................................. 1-10 Saturation .................................................................................................. 1-10 Irreducible Water Saturation........................................................................... 1-11 Residual Oil ............................................................................................... 1-12 Relations Between Permeability And Fluid Saturation........................................ 1-12

Fluid Properties .................................................................................................. 1-13 1.5.1 1.5.2 1.5.3

1.6

Rock Classification ...................................................................................... 1-4 Rock Properties........................................................................................... 1-5 1.3.2.1 Porosity....................................................................................... 1-5 1.3.2.2 Permeability................................................................................. 1-6 Hydrocarbon Accumulations .......................................................................... 1-7 1.3.3.1 Domes And Anticlines ................................................................... 1-7 1.3.3.2 Salt Domes And Plug Structures...................................................... 1-7 1.3.3.3 Faults.......................................................................................... 1-7 1.3.3.4 Unconformity................................................................................ 1-8 1.3.3.5 Lenticular Reservoirs...................................................................... 1-8 Reservoir Temperature And Pressure ............................................................... 1-8 1.3.4.1 Normal Pressure ............................................................................ 1-8 1.3.4.2 Abnormal Pressures ....................................................................... 1-9 1.3.4.3 Reservoir Temperature.................................................................... 1-9

Components Of Hydrocarbon ......................................................................... 1-14 Classification Of Hydrocarbons....................................................................... 1-15 Characteristics Of Formation Water ................................................................. 1-16

Phase Behaviour .................................................................................................. 1-17 1.6.1 1.6.2 1.6.3 1.6.4

1.6.5

1.6.6

Phase Behaviour Of A Single Component System ............................................. 1-17 Phase Behaviour Of A Multi-Component System .............................................. 1-17 Reservoir And Fluid Compressibility.............................................................. 1-19 Conversion Factors Between Surface And Downhole Volumes ............................. 1-19 1.6.4.1 Formation Volume Factor Of Gas, Bg ............................................... 1-20 1.6.4.2 Formation Volume Factor Of Oil, Bo................................................ 1-20 1.6.4.3 Formation Volume Factor Of Water, Bw ........................................... 1-21 Fluid Density Corrections ............................................................................. 1-21 1.6.5.1 Gas Density.................................................................................. 1-21 1.6.5.2 Oil Density .................................................................................. 1-21 1.6.5.3 Water Density............................................................................... 1-21 Viscosities.................................................................................................. 1-21 1.6.6.1 Gas Viscosity ............................................................................... 1-21 1.6.6.2 Oil Viscosity ................................................................................ 1-21 1.6.6.3 Water Viscosity ............................................................................ 1-22 1.6.6.4 Formation Compressibility ............................................................. 1-22

Introduction to Well Testing (Aug 1996)

TOC - 3

Table of Contents 1.7 Reservoir Drive Mechanisms .............................................................................................1-22 1.7.1 1.7.2 1.7.3 1.7.4 1.7.5 Section 2.0

Oil Reservoirs..............................................................................................1-22 Solution Gas Drive Reservoirs........................................................................1-23 Gas Cap Expansion Drive Reservoirs ...............................................................1-23 Water Drive Reservoirs..................................................................................1-24 Discussion Of Recovery Efficiency ...................................................................1-26 Completion Technology And Types

2.1

Introduction

2.2

Completion Components .........................................................................................2-3 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5

2.2.6 2.3

General 2.3.1.1 2.3.1.2

...................................................................................................2-12 Open Hole Completions..................................................................2-12 Cased Hole Completions.................................................................2-12

Completion Types ..................................................................................................2-12 2.4.1 2.4.2

2.4.3

2.5

Production Tubulars .....................................................................................2-4 Packers ...................................................................................................2-4 2.2.2.1 Permanent Packers..........................................................................2-5 2.2.2.2 Retrievable Packer ..........................................................................2-6 Flow Control...............................................................................................2-6 2.2.3.1 Landing Nipples ............................................................................2-6 2.2.3.2 Sliding Side Door..........................................................................2-7 Subsurface Safety Systems..............................................................................2-8 2.2.4.1 Subsurface Controlled Subsurface Safety Valves ...................................2-8 2.2.4.2 Surface Controlled Subsurface Safety Valves........................................2-8 Ancillary Components...................................................................................2-10 2.2.5.1 Wireline Entry Guide......................................................................2-10 2.2.5.2 Blast Joint ....................................................................................2-10 2.2.5.3 Flow Couplings.............................................................................2-10 General ...................................................................................................2-10

Completion Techniques ..........................................................................................2-12 2.3.1

2.4

...................................................................................................2-3

Natural Completions .....................................................................................2-14 Stimulated Completions................................................................................2-14 2.4.2.1 Hydraulic Fracturing.......................................................................2-14 2.4.2.2 Acidizing......................................................................................2-14 2.4.2.3 Extreme Overbalnce Perforating.........................................................2-15 2.4.2.4 Effects Of Perforation.......................................................................2-15 Sand Control Completions.............................................................................2-16 2.4.3.1 Production Rate Restriction .............................................................2-16 2.4.3.2 Gravel Packing ..............................................................................2-16 2.4.3.3 Sand Consolidation ........................................................................2-16 2.4.3.4 Resin Packs ..................................................................................2-16

The Effects Of Drilling On Completions..................................................................2-17 2.5.1 2.5.2

Pre-Formation Drilling..................................................................................2-17 Formation Drilling .......................................................................................2-17

Introduction to Well Testing (Aug 1996)

TOC - 4

Schlumberger

Section 3.0 3.1

Practical Well Testing Introduction 3.1.1 3.1.2

.................................................................................................. 3-3

Productivity Well Testing............................................................................. 3-3 Descriptive Well Testing .............................................................................. 3-5

3.2

Test Design

3.3

Tubing Conveyed Perforating - Tcp ........................................................................ 3-10 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5

3.3.6 3.4

General .................................................................................................. 3-10 3.3.1.1 Through Tubing Perforating ............................................................ 3-10 3.3.1.2 Casing Gun And High Shot Density Perforating (HSD)........................ 3-11 Benefits Of TCP.......................................................................................... 3-11 Equipment Selection .................................................................................... 3-12 Testing Requirements................................................................................... 3-13 Firing Heads ............................................................................................... 3-14 3.3.5.1 Differential Pressure Firing Head....................................................... 3-14 3.3.5.2 Hydraulic Delay Firing Head ........................................................... 3-14 3.3.5.3 Bar Hydrostatic Firing Head............................................................ 3-14 3.3.5.4 Trigger Charge Firing System ......................................................... 3-14 3.3.5.5 Redundant Firing Systems .............................................................. 3-16 Depth Control............................................................................................. 3-16 3.3.6.1 Procedure..................................................................................... 3-16

Drill Stem Testing Tools - Dst................................................................................. 3-19 3.4.1 3.4.2

3.4.3 3.4.4 3.5

.................................................................................................. 3-6

General .................................................................................................. 3-19 Basic Requirements...................................................................................... 3-19 3.4.2.1 Packer ......................................................................................... 3-19 3.4.2.2 Test Valve ................................................................................... 3-19 3.4.2.3 Reverse Circulation Valve ............................................................... 3-19 3.4.2.4 Slip Joint..................................................................................... 3-19 3.4.2.5 Hydraulic Jar ................................................................................ 3-20 3.4.2.6 Safety Joint .................................................................................. 3-20 3.4.2.7 Safety Valve ................................................................................. 3-20 3.4.2.8 Gauge Carrier................................................................................ 3-20 3.4.2.9 Sampling Chamber Tool ................................................................ 3-20 Types Of Drill Stem Tests ............................................................................ 3-22 3.4.3.1 Open Hole Drill Stem Testing......................................................... 3-22 3.4.3.2 Cased Hole Drill Stem Testing........................................................ 3-22 New Technology ......................................................................................... 3-25

Subsurface Safety Systems ..................................................................................... 3-26 3.5.1 3.5.2

3.5.3

General .................................................................................................. 3-26 Subsea Test Package .................................................................................... 3-26 3.5.2.1 Subsea Test Tree........................................................................... 3-26 3.5.2.2 Retainer Valve .............................................................................. 3-27 3.5.2.3 Lubricator Valve............................................................................ 3-27 New Developments ...................................................................................... 3-27

Introduction to Well Testing (Aug 1996)

TOC - 5

Table of Contents 3.6 Surface Testing Equipment ..............................................................................................3-28 3.6.1 3.6.2

3.7

Data Acquisition ...................................................................................................3-36 3.7.1 3.7.2

3.7.3

3.7.4 3.7.5 3.7.6

3.7.7 3.8

Section 4.0

High Pressure High Temperature - HPHT .........................................................3-46 High Flow Rates..........................................................................................3-47 Sampling Of Reservoir Fluids

Sampling Procedures Design ...................................................................................4-3 4.1.1 4.1.2 4.1.3 4.1.4

4.2

General ...................................................................................................3-36 Transducer Performance .................................................................................3-36 3.7.2.1 Static Parameters............................................................................3-36 3.7.2.2 Dynamic Parameters .......................................................................3-38 3.7.2.3 Calibration....................................................................................3-39 Pressure Transducer Technology .....................................................................3-39 3.7.3.1 Strain Gauge Sensors......................................................................3-39 3.7.3.2 Capacitance Sensors........................................................................3-40 3.7.3.3 Quartz Crystal Sensors....................................................................3-40 Surface Data Acquisition................................................................................3-41 Downhole Data Acquisition............................................................................3-43 3.7.5.1 Downhole Recording ......................................................................3-43 3.7.5.2 Surface Readout .............................................................................3-43 New Technology..........................................................................................3-44 3.7.6.1 Datalatch System ...........................................................................3-44 3.7.6.2 Wireless Telemetry ........................................................................3-44 3.7.6.3 Data Acquisition In Permanent Completions.......................................3-45 Formation Interval Testing.............................................................................3-45

Special Applications ...............................................................................................3-46 3.8.1 3.8.2

4.1

General ...................................................................................................3-28 Equipment ..................................................................................................3-29 3.6.2.1 Flowhead......................................................................................3-29 3.6.2.2 Choke Manifold.............................................................................3-31 3.6.2.3 Heater / Steam Exchanger ................................................................3-31 3.6.2.4 Test Separator ...............................................................................3-32 3.6.2.5 Gauge Tank ..................................................................................3-32 3.6.2.6 Pumps And Manifolds ....................................................................3-33 3.6.2.7 Burners ........................................................................................3-33 3.6.2.8 Emergency Shut Down Systems .......................................................3-33

Samples Representivity .................................................................................4-3 Producing Conditions ...................................................................................4-5 Well Conditioning .......................................................................................4-6 Hydrocarbon Sampling Methods .....................................................................4-7 4.1.4.1 Bottom Hole Sampling...................................................................4-7 4.1.4.2 Surface Sampling ...........................................................................4-7

Sampling Of Oil Reservoirs ....................................................................................4-8 4.2.1 4.2.2 4.2.3 4.2.4

Preliminary Conditions On Oil Reservoirs........................................................4-8 Pre-Job Required Data...................................................................................4-9 New Wells Or Wells In Undepleted Zones ........................................................4-9 4.2.3.1 Undersaturated Reservoirs ................................................................4-9 4.2.3.2 Saturated Reservoirs .......................................................................4-9 Producing Reservoirs Or Wells In Slightly Depleted Zones..................................4-10 4.2.4.1 Gor Is Equal To Gori......................................................................4-10 4.2.4.2 Gor Is Higher Than Gori .................................................................4-10

Introduction to Well Testing (Aug 1996)

TOC - 6

Schlumberger

4.3

Sampling Gas Reservoirs........................................................................................ 4-10 4.3.1 4.3.2

Preliminary Considerations On Gas Reservoirs.................................................. 4-10 Gas Reservoir Sampling Procedures ................................................................ 4-12 4.3.2.1 New Reservoirs Or Wells In Undepleted Zones................................... 4-13 4.3.2.2 Producing Reservoirs Or Wells In Depleted Zones............................... 4-13

4.4

Sampling Of Volatile Oil Reservoirs........................................................................ 4-13

4.5

Bottom Hole Sampling ........................................................................................... 4-15 4.5.1 4.5.2 4.5.3 4.5.4

4.6

Surface Sampling .................................................................................................. 4-19 4.6.1 4.6.2 4.6.3 4.6.4

4.6.5

4.7

Formation Water Sampling Methods............................................................... 4-29 4.7.1.1 Drillstem Tests............................................................................. 4-29 4.7.1.2 Surface Sampling........................................................................... 4-31

Sampling Equipment............................................................................................. 4-31 4.8.1

4.8.2 4.8.3 4.8.4 4.9

Well Conditioning For Surface Sampling......................................................... 4-19 Oil Surface Sampling Methods ....................................................................... 4-21 4.6.2.1 Piston Bottle Displacement Method.................................................. 4-25 Gas Surface Sampling Methods ...................................................................... 4-25 4.6.3.1 Vacuum Method ............................................................................... 4-25 Special Surface Sampling Cases...................................................................... 4-25 4.6.4.1 High Pressure Samples................................................................... 4-25 4.6.4.2 Multistage Separation System.......................................................... 4-25 4.6.4.3 Hydrogen Sulphide........................................................................ 4-26 Well Head Sampling Of Oil And Gas .............................................................. 4-29 4.6.5.1 Oil Sampling At Well Head............................................................ 4-29 4.6.5.2 Gas Sampling At Well Head ........................................................... 4-29

Sampling Of Formation Water ............................................................................... 4-29 4.7.1

4.8

Well Conditioning For Bottom Hole Sampling ................................................ 4-15 Bottom Hole Sampling Procedures ................................................................. 4-15 Bottom Hole Sample Transfer Procedures......................................................... 4-17 Checking Bottom Hole Sample Validity .......................................................... 4-18

Bottom Hole Sampling Equipment................................................................. 4-31 4.8.1.1 Open Hole Sampling From Formation Interval Testing........................ 4-31 4.8.1.2 Dst Sampling Tools ...................................................................... 4-31 4.8.1.3 Production Sampling Tools ............................................................ 4-32 4.8.1.4 Transfer Benches ........................................................................... 4-32 Surface Sampling Equipment ......................................................................... 4-35 Sample Containers....................................................................................... 4-35 4.8.3.1 Gas Sample Bottles ....................................................................... 4-35 4.8.3.2 Oil Sample Bottles ........................................................................ 4-35 New Developments ...................................................................................... 4-36

Fluid Analysis And Uses Of Data ........................................................................... 4-36 4.9.1

4.9.2

Field Estimation Of Reservoir Properties.......................................................... 4-36 4.9.1.1 Field Estimation Of Reservoir Properties Through Correlations ............. 4-36 4.9.1.2 Field Estimation Of Reservoir Properties With Portable PVT................ 4-37 4.9.1.3 Field Estimation Of Reservoir Properties With Fluid Properties Estimator..................................................................................... 4-37 PVT Laboratory Measurements Of Reservoir Properties ...................................... 4-38

Introduction to Well Testing (Aug 1996)

TOC - 7

Table of Contents Section 5.0 5.1

Basic Well Test Interpretation Introduction 5.1.1 5.1.2

5.2

5.2.3

5.2.4 5.2.5 5.2.6

Inverse And Direct Problem............................................................................5-5 Basic Model................................................................................................5-6 5.2.2.1 Homogeneous Reservoir ..................................................................5-6 5.2.2.2 Heterogeneous Reservoir..................................................................5-6 5.2.2.3 Radial Flow ..................................................................................5-6 5.2.2.4 Infinite Acting Radial Flow..............................................................5-7 Inner Boundary Conditions ............................................................................5-7 5.2.3.1 Wellbore Storage ...........................................................................5-7 5.2.3.2 Skin ............................................................................................5-8 5.2.3.3 Other Flow Regimes.......................................................................5-8 Outer Boundary Conditions............................................................................5-9 The Complete Model....................................................................................5-10 Various Phases During A Well Test ................................................................5-10

Model Recognition..................................................................................................5-11 5.3.1 5.3.2 5.3.3

5.3.4

5.3.5 5.3.6 5.4

Reservoir Model...........................................................................................5-3 Well Test Interpretation.................................................................................5-3

Defining The Reservoir Model .................................................................................5-5 5.2.1 5.2.2

5.3

...................................................................................................5-3

Log-Log Scale .............................................................................................5-11 Time Periods...............................................................................................5-11 Flow Regime Identification ............................................................................5-12 5.3.3.1 Radial Flow ..................................................................................5-13 5.3.3.2 Spherical Flow ..............................................................................5-13 5.3.3.3 Linear Flow ..................................................................................5-16 5.3.3.4 Bilinear Flow ................................................................................5-17 5.3.3.5 Compression / Expansion................................................................5-17 5.3.3.6 Steady State ..................................................................................5-18 5.3.3.7 Dual Porosity / Permeability ............................................................5-18 5.3.3.8 Slope Doubling .............................................................................5-19 Specialised Plots..........................................................................................5-19 5.3.4.1 Wellbore Storage ...........................................................................5-19 5.3.4.2 High Conductivity Fracture..............................................................5-19 5.3.4.3 Infinite Acting Radial Flow..............................................................5-20 The Complete System...................................................................................5-20 Additional Information For Model Recognition..................................................5-21

Parameter Estimation..............................................................................................5-21 5.4.1 5.4.2 5.4.3 5.4.4 5.4.5

Dimensionless Groups...................................................................................5-21 Type Curves................................................................................................5-22 5.4.2.1 Type Curve Matching.....................................................................5-22 Parameter Approximation From Type Curves ....................................................5-23 5.4.3.1 Basic Homogeneous Model..............................................................5-23 5.4.3.2 Basic Homogeneous - Specialised Plot...............................................5-25 Superposition ..............................................................................................5-25 Semi-Log Analysis For Parameter Estimation....................................................5-27 5.4.5.1 Horner (Or Superposition) Method ....................................................5-27 5.4.5.2 Skin Calculation ............................................................................5-27

Introduction to Well Testing (Aug 1996)

TOC - 8

Schlumberger

5.4.6

5.4.7

Other Key Type Curves ................................................................................ 5-27 5.4.6.1 Homogeneous Reservoir With Wellbore Storage And Skin................... 5-27 5.4.6.2 Homogeneous Reservoir With Wellbore And Infinite Conductivity Vertical Fracture ............................................................................ 5-28 5.4.6.3 Well With Wellbore Storage And Skin. Reservoir With Double Porosity Behaviour - Pseudosteady State Interporosity Flow.............................. 5-29 Parameter Refinement........................................ ........................................... 5-30

5.5

Verification Of Results........................................................................................... 5-33

5.6

Summary Of Interpretation Methodology ................................................................ 5-33

5.7

Gas Well Testing .................................................................................................. 5-34 5.7.1 5.7.2 5.7.3

5.8

Multiple Well Testing ............................................................................................ 5-39 5.8.1 5.8.2

5.9

Simplifications To The Pseudofunctions .......................................................... 5-34 Multi-Point Well Testing ............................................................................. 5-35 Types Of Gas Well Tests .............................................................................. 5-36 5.7.3.1 Flow-After-Flow Tests ................................................................... 5-36 5.7.3.2 Isochronal Tests............................................................................ 5-36 5.7.3.3 Modified Isochronal Tests............................................................... 5-36

Interference Testing ...................................................................................... 5-39 Pulse Testing.............................................................................................. 5-39

Other Specialised Testing Types ............................................................................. 5-40

Section 6.0

Nodal Systems - (Systems Analysis)

6.1

Introduction

6.2

Inflow Performance Curves .................................................................................... 6-3 6.2.1 6.2.2 6.2.3

6.3

Example Problem 1...................................................................................... 6-11 Example Problem 2...................................................................................... 6-12 Example Problem 3...................................................................................... 6-15 Example Problem 4...................................................................................... 6-17 Discussion Of Results To Problems 1 Through 4.............................................. 6-18

Changes In Flow Conduit Size................................................................................ 6-20 6.4.1

6.5

Tubing Intake Curves................................................................................... 6-4 Horizontal Flow Curves................................................................................ 6-6 Other Considerations.................................................................................... 6-6

The Nodal Concept................................................................................................ 6-8 6.3.1 6.3.2 6.3.3 6.3.4 6.3.5

6.4

.................................................................................................. 6-3

Example Problem 5...................................................................................... 6-20

Functional Nodes.................................................................................................. 6-23 6.5.1 6.5.2 6.5.3

Surface Wellhead Choke................................................................................ 6-23 Example Problem 6...................................................................................... 6-24 Example Problem 7...................................................................................... 6-28 6.5.3.1 Effects Of Separator Pressure ............................................................ 6-29 6.5.3.2 Effects Of Flowline Size .................................................................. 6-30 6.5.3.3 Effects Of Tubing Size .................................................................... 6-30

Introduction to Well Testing (Aug 1996)

TOC - 9

Table of Contents 6.6

General Discussion On The Effects Of The Variables ...............................................6-33

6.7

Graphical Representation Of The Total Producing System For One Well....................6-34

6.8

Summary

Section 7.0

...................................................................................................6-35

Work Sessions

Work Session For Section 1...............................................................................................7-3 Work Session For Section 2...............................................................................................7-10 Work Session For Section 3...............................................................................................7-12 Work Session For Section 4...............................................................................................7-17 Work Session For Section 5...............................................................................................7-18 Section 8.0

Work Session Answers

Section 9.0

References

Introduction to Well Testing (Aug 1996)

TOC - 10

Section 1 Basic Reservoir Engineering

Schlumberger

Introduction to Well Testing (Aug 1996)

1-2

Schlumberger

1.0

Basic Reservoir Engineering

1.1

Introduction and Course Objectives Every hydrocarbon bearing reservoir is a valuable asset. To ensure the best possible return, not just in terms of the commercial factors, although this is clearly the driving factor, it is important to understand as much as possible about the reservoir. This always presents a conceptual problem as we cannot physically see the reservoir in question. Fortunately the pioneering instinct and inquisitive minds of mankind have developed some ingenious techniques to help gain valuable data which in turn develops a descriptive picture or simulated model of the reservoir in question. Techniques such as; Seismic Data Acquisition, Electric Line Logging, Core Analysis, PVT Analysis, and Well Testing to name a few have become sciences within themselves and all produce valuable data which help build the simulated reservoir model and thus help in developing the most cost effective strategy to manage the asset. Well Testing is different from most techniques as it requires the reservoir to be in a dynamic state as opposed to a static state in order to trigger the responses needed for mathematical modelling. The objective of this course, as the title suggests, is to develop a basic understanding of well testing methodology and the associated interpretation techniques. To achieve this, a review of some of the fundamental basics of petroleum engineering is necessary in order to develop the essential principles required later on. The course will cover the necessary hardware requirements for different test scenarios and the objectives of testing with an emphasis on data acquisition and representative fluid sampling. Introduction to Well Testing (Aug 1996)

The introduction to well test interpretation will be kept in a simple form, avoiding complicated mathematical analysis where possible. Finally, the student will be given an introduction to the techniques of nodal analysis which complement the interpretation process. Well testing is a costly operation involving significant resources and logistics. As such, management require detailed justification before giving approval to any testing budget and it is often critical to highlight a return on the investment. Accurate well testing data can reveal extremely valuable information which in turn leads to efficient reservoir management. Consequently, the course will place significant emphasis on data quality. 1.2

History

Fossil fuels have been used by man for generations. The industrial revolution of the early 19th century created a large demand for fuel to supply the emerging industries, and the corresponding social changes demanded fuel and lighting for the homes of the wealthy. Initially oil was mainly obtained through whaling and other animal fats however the availability of surface oil deposits was creating specific excitement among the scientific community. The first commercial oil well was drilled near Oil Creek, Pennsylvania, by Colonel Edwin L. Drake in 1859 which marked the birth of the modern petroleum industry. Well testing techniques were first applied in the early 1920’s whereby a simple pressure measurement was taken using a Bourdon tube and stylus arrangement.

1-3

Schlumberger

Early analysis techniques were presented by Muskat in the 1930’s and later in the 1950’s with the classical work of Miller, Dyes, Hutchinson and Horner. More advanced instruments and testing techniques were gradually developed however, the onset of electronic recording devices in the early 1970’s and computerised systems in the early 1980’s set new standards of data acquisition and interpretation techniques. 1.3

Geology Recap

1.3.1 Rock Classification Rocks are broadly classified into three groups: • Igneous. • Sedimentary. • Metamorphic. Igneous rocks comprise 95% of the earth’s crust. They originate from the solidification of molten material emanating from below the earth’s surface. Volcanic igneous rocks are often glassy in texture as rapid cooling of magma does not allow time for the formation of crystals. Plutonic igneous rocks are formed Calstic Rock - Formed From Debris of Older Rock Rock Type

Particular

Diameter

Conglomerate

Pebbles

-

Sandstone

Sand

-

2 to 64 mm 0.06 to 2 mm

Siltstone

Silt

-

0.003 to 0.06 mm

Shale

Clay

-

Less than 0.003 mm

Fig. 1-1a Sedimentary rock classification.

in slower cooling intrusive magmas which allows the time needed for the atoms to arrange themselves into a crystalline grain structure. Sedimentary rocks are formed from the materials of older up-lifted formations which have broken down by erosion and transported by the elements to lower elevations where they are deposited. Consolidation of sands, silts, pebbles and clays by pressure of many thouIntroduction to Well Testing (Aug 1996)

sands of feet of overlying sediments, and cementation by precipitates from percolating waters act to convert these materials into sandstones, siltstones and conglomerates. Sedimentary rocks are classified into two groups;

• Clastic (the rocks of detrital origin or debris from older rocks) such as sandstone, siltstone and shales.

• Non-clastic (rocks of biochemical or chemical precipitate origin) such as limestone, dolomite and clays. Nonclastic - Mostly of Chemical or Biochemical Origin Rock Type

Composition

Limestone Dolomite Salt Gypsum Chert Coal

Calcite Dolomite Halite Gypsum Silica Chiefly Carbon

-

CaCO3 CaMg(CO3)2 NaCI CaSO4.2H2O SiO2

Fig. 1-1b Sedimentary rock classification.

Metamorphic rocks are formed from other sedimentary deposits by alteration under great heat and/or pressure. Examples of metamorphic rocks are;

• Marble - metamorphosised limestone. • Hornfeld - converted from shale or tuff. • Gneiss - similar to granite but has been metamorphicaly consolidated.

Oil and gas are not usually found in igneous or metamorphic rocks as both are so nonporous that hydrocarbons can not accumulate or be extracted from them. The few exceptions are when hydrocarbons have seeped from near-by sedimentary formations through cracks and fractures.

1-4

Schlumberger

1.3.2 Rock Properties To form a commercial reservoir of hydrocarbons, a geological formation must exhibit three essential characteristics;

• Contain sufficient void space to contain hydrocarbons.

• Possess adequate connectivity of these pore spaces to allow transportation over large distances.

• Possess a capacity to trap sufficient quan-

Fig. 1-2 Accumulation of oil and gas into a reservoir.

tities of hydrocarbon to prevent upward migration from the source beds.

1.3.2.1 Porosity The void spaces in the reservoir rocks are, for the most part, the intergrannular spaces between the sedimentary particles. Porosity is defined as a percentage or fraction of void to the bulk volume of the rock. While the proportion of void can be calculated from regular arrangements or uniform spheres (see Fig. 13), the arrangements within actual reservoirs is a much more complex picture and effected by many different parameters. In this case measurements are either done in the laboratory on core samples whereby actual conditions are simulated as closely as possible prior to measurement or in-situ via suites of electric logs such as Neutron, Density and Sonic Logs. Processes after sedimentation (cementation, re-crystallisation, weathering, fracturing etc.) can modify substantially the proportion and distribution of pore space. In reservoir engineering, only the interconnected or effective porosity is of interest since this is the only capacity which can make a contribution to flow. Pore spaces initially present but subsequently sealed off by cementation or recrystallisation effects are of no interest.

Introduction to Well Testing (Aug 1996)

Fig. 1-3 Intergranular porosity.

Primary Porosity refers to the void spaces remaining after sedimentation of the granules in the matrix and hence is a matrix porosity. Secondary Porosity is the contribution from pits, vugs, fractures and other discontinuities in the bulk volume of the matrix. The contribution of secondary porosity to the overall bulk porosity is generally small yet it can lead to dramatic increase in the ease with which hydrocarbons flow through the rock. From the reservoir engineering point of view, the distinguishing factor between primary and secondary porosity is not the mode of occurrence but the very different flow capacity where an interconnected secondary porosity system is present. This is known as a dual porosity system. In the real world of the reservoir this is often the case and one can easily see how quickly our simulated models can be made complex. Fortunately in the world of mathematical modelling certain prac1-5

Schlumberger

tical assumptions are made to help unbundle this complex approach and best fit the real world to a workable model. 1.3.2.2 Permeability Permeability is a measure - under nonturbulent flow conditions of the ease with which fluid flows through a porous rock and is a function of the degree of interconnection between the pores. To illustrate this Fig. 1-4 shows a volume of rock with the same effective porosity. It is clear that fluid will flow more rapidly through sample A than through sample B where the flow is restricted.

ential equations after subjecting the reservoir to a dynamic condition and monitoring the corresponding pressure and temperature response. While grain size has a negligible effect on the porosity of a rock,

Fig. 1-5 Definition of a Darcy.

( a)

( b)

Fig. 1-4 Source “Principles of Oil Well Production” T.E.W. Nind.

Permeability is measured in darcy units or more commonly millidarcy (md - one thousandth of a darcy) after Henry Darcy who carried out some pioneering work on water flow through unconsolidated sand stones. A practical definition of a darcy is as follows; A rock has a permeability k of 1 Darcy if a pressure gradient of 1 atm/cm induces a flow rate of 1 cc/ cm2 of cross sectional area of a liquid viscosity 1 cp. See Fig. 1-5. Permeability, like porosity can be measured in the laboratory from core samples. There is to date no instrument which measures permeability directly in-situ, but permeability can be calculated via solving complex differIntroduction to Well Testing (Aug 1996)

this parameter has a predominant effect on permeability. This is so because as we are dealing with flow, we are also dealing with friction of the fluid against the surface area of the rock grains. Each rock grain has a wetted surface surrounding it where fluid velocity is always zero by definition, thus shearing friction is formed between this zero velocity layer and any passing fluids. Thus more frictional forces are encountered passing the same fluid through a fine granular pack than through a coarse granular pack of equal porosity. Similarly one can understand that the permeability will also be dependent on the density of the fluid flowing through the rock and this plays an important part in the interpretation of different hydrocarbon bearing reservoirs. Permeability is denoted in three different ways, absolute permeability ka is derived in the laboratory by flowing a known quantity of fluid through a core while its pore spaces are 100% saturated with the same fluid. Absolute permeability will not change with varying fluids as long as the pore space configuration remains constant. Effective permeability is the permeability of a flowing phase which does not saturate 100% of the rock. The effective permeability 1-6

Schlumberger

is always less than the absolute value of k for the rock. Relative permeability is a dimensionless number which is the ratio of effective permeability (to a fluid) to absolute permeability of the same rock.

Fig. 1-7b Oil and gas accumulation in an anticline.

Fig. 1-6 Effect of grain size on permeability.

1.3.3 Hydrocarbon Accumulations Petroleum deposits will be found only in those areas where geological conditions combine to form and trap them. Hydrocarbons, being less dense than water, migrate upwards from the source beds until they escape at surface or an impervious barrier is encountered. The principle classifications of petroleum reservoir forming traps are as follows; 1.3.3.1

1.3.3.2 Salt Domes and Plug Structures This commonly occurring geological structure is caused by the intrusion from below of a salt mass, volcanic material or serpentine. In pushing or piercing through the overlying strata, the intrusion may cause the formation of numerous traps in which petroleum may accumulate.

Domes and Anticlines

Domes and anticlines are formed by uplifting and folding of the strata. When viewed from above the dome is circular in shape, whereas the anticline is an elongated fold.

Fig. 1-8 Hydrocarbon accumulation associated with a piercement salt dome.

1.3.3.3 Faults Reservoirs may be formed along the fault plane where the shearing action has caused an impermeable bed to block the migration of oil and gas through a permeable bed. Fig. 1-7a Dome structure. Oil and gas migrate upward from source beds until trapped by the impermeable cap rock.

Introduction to Well Testing (Aug 1996)

1-7

Schlumberger

Fig. 1-9 Trap formed by a fault.

1.3.3.4 Unconformity This type of structure can be formed where more recent beds cover older, inclined formations that have been planed off by erosion. A reservoir may be formed where oil and gas is trapped by an impermeable overlying layer.

Fig. 1-10 Oil and gas trapped under an unconformity.

1.3.3.5 Lenticular Reservoirs Oil and gas may accumulate in pockets of porous permeable beds or traps formed by pinch-outs of the porous beds within an impermeable bed. Lens type reservoirs are formed where sand was deposited along an irregular coastline or by filling in an ancient river bed or delta. Similar productive zones occur in various porous sections in thick impermeable limestone beds. Pinch-outs may occur near the edge of a basin where the sand progressively “shales out” as the edge of the basin is approached. In river deposited sand bars, shale-out frequently occurs within a few hundred feet.

Introduction to Well Testing (Aug 1996)

Fig. 1-11 Upper bounds of the reservoir formed by change in permeability of a sand.

1.3.4

Reservoir Temperature and Pressure

1.3.4.1 Normal Pressure As previously mentioned, hydrocarbon accumulations occur in partially sealed structures where the upward migration of oil and gas from the source beds is blocked by an impermeable barrier. As hydrocarbon accumulates, formation water is expelled from the porous reservoir rock. Unless subsequent tectonic movements completely seal the reservoir, the underlying waters are contiguous and pressures in the aquifer will approximate to some local or regional hydrostatic gradient. In a water column, the pressure at any depth is approximated to; p = h x Gw where: h is the depth and Gw is the pressure gradient. Although ground waters are saline, temperatures increasing with depth tend to reduce the water density and a common “normal” value of Gw is 0.433 psi/ft (0.1 kg/cm2/m), which is approximately a fresh water gradient. Gradients within the range 0.43 to 0.5 psi/ft are considered normal. Pressures at the top of a hydrocarbon bearing structure, higher than the hydrostatic gradient extrapolated from 1-8

Schlumberger

the hydrocarbon/water contact, is expected because of the lower density of hydrocarbon compared with water. Even in thick gas bearing zones this situation does not lead to dangerously abnormal pressures.

Fig. 1-13 Estimation of formation temperature.

1.4

Fig. 1-12 “Normal” pressure distribution from surface through a reservoir structure.

1.3.4.2 Abnormal Pressures Under certain depositional conditions or because of tectonic movements which close the reservoir structure, fluid pressures may depart substantially from the normal range. Abnormal pressures can occur when some part of the overburden load is transmitted to the formation fluids. Abnormal pressures corresponding to gradients of 0.8 to 0.9 psi/ft and approaching the geostatic gradient (generally taken as approximately equivalent to 1.0 psi/ft) may occasionally be encountered and can be considered dangerously high. 1.3.4.3 Reservoir Temperature Reservoir temperatures will conform to the regional or local geothermal gradient, a normal value being 1.6 °F/100ft. Because of the large thermal capacity of the rock matrix which comprises in the order of 80% of the bulk reservoir volume and the very large area for heat transfer, conditions within the reservoir may be considered isothermal in most cases.

Introduction to Well Testing (Aug 1996)

Interacting Forces, Saturation and Displacement

1.4.1 General A brief discussion on these topics is necessary as they play a crucial role in the development of the mathematical understanding of dynamic fluid movement within reservoirs. When two or more fluids exist within a reservoir (multiphase system) the number of interacting forces increases and thus the complexity of the simulated model. 1.4.2 Surface and Interfacial Tension The apparent film which separates two immiscible fluids, such as air and water is caused by unequal attractive forces of molecules at the interface. The work required to move a molecule of water across this barrier gives rise to surface tension. When the fluids are water and oil the phenomena is known as interfacial tension. Surface tension and interfacial tension are commonly measured in dynes per centimetre. Surface tensions between some common fluids and air at 20° C are given below: Water Benzene Cyclohexen

72.6 dynes/cm 28.9 dynes/cm 25.3 dynes/cm

1-9

Schlumberger

the interfacial tension between water and oil at 20° C ≈ 30 dynes/cm. 1.4.3 Wetting The adhesion tension , which is a function of the interfacial tension, determines which fluid will preferentially wet a solid. As an example, water will spread out on the surface of a sheet of glass whereas mercury will bead up and not adhere to the glass. Thus for water the adhesive forces between liquid and solid are greater than the cohesive forces holding the liquid molecules together, the opposite is true for the mercury. In the formation, water will adhere to water wet

Fig. 1-14 Apparent surface film caused by imbalance of molecular forces.

rock and bead-up on oil wet rocks. The tendency of one fluid to displace another from a solid surface is determined by the relative wettability of the fluids to the solid.

1.4.4 Capillarity When liquid wets the surface of a fine bore glass capillary tube, surface tension around the circumference of the contact pulls the liquid interface up the tube until an equilibrium is reached with the downward force due to the liquid column height. In the reservoir, although the pore spaces do not form the uniform capillary tubes of the laboratory, nevertheless they do interconnect to form a complex maze of capillary systems which in turn gives rise to capillary forces. These forces can be measured under laboratory conditions for a given rock - fluid system and in turn, the capillary height can be calculated if the density difference of the fluid system is known. 1.4.5 Saturation During deposition, reservoir rocks are completely saturated and water wet. As hydrocarbons migrate and accumulate in the reservoir rock, a portion of this water (connate water) is displaced. Both silica and calcite have a strong tendency to remain water wetted which means even after hydrocarbon percolation some connate water will always remain within the rock structure. Water saturation Sw represents the percentage of water occupying effective pore space and is expressed as a fraction of the pore volume. Likewise similar definitions exist for oil So and gas Sg. These values can be measured in the laboratory using cores or derived from electric logs.

Fig. 1-15 Contact angle as a measure of wetting.

Introduction to Well Testing (Aug 1996)

1-10

Schlumberger

the centre of the larger pore channels leaving oil behind in the smaller channels.

Fig. 1-16 Comparison of fluid rise in a capillary tube bundle of varying diameters illustrates the distribution of saturation in the transition zone above an oil/water contact.

1.4.6 Irreducible Water Saturation The minimum saturation that can be induced by displacement is one in which the wetting phase becomes discontinuous. To understand this, imagine a packing of uniform spheres, this would then correspond to the wetting phase remaining as pendular rings at the sphere contacts. The minimum saturation corresponds to the smallest mean radius of curvature of these rings and the maximum available capillary pressure. However since the wetting phase will become discontinuous at some finite capillary pressure there will always be some irreducible water saturation, a saturation which cannot be reduced by displacement by a non-wetting phase no matter how great a pressure is applied to the system. It follows that the size of the grains will also impact the irreducible water saturation as shown in Figs. 1-17 & 1-18.

Fig. 1-17 Shape of the capillary pressure vs. saturation curve.

Fig. 1-18 Shape of capillary curve through the transition zone is strongly affected by the distribution of grain size.

During the accumulation of hydrocarbon in the reservoir, some threshold pressure had to be overcome in order to permit the nonwetting oil to enter the water saturated pores. These same forces now aid the expulsion of oil from the tight and dead-end pores by inhibition of water along the surface of the grains

At higher wetting phase saturations, the mean radius of curvature increases and the capillary pressure decreases. Water tends to displace oil in a piston like fashion, moving first close to the rock surface where it is aided by capillary forces in squeezing oil from the smaller channels. Gas being more mobile tends to move easily along Introduction to Well Testing (Aug 1996)

Fig. 1-19 Natural displacement of oil by water in a single pore channel. (Courtesy Journal of Petroleum Technology - June, 1958).

1-11

Section 1

Fig. 1-20 Natural displacement of oil by gas in a single pore channel. (Courtesy of Journal of Petroleum Technology - June, 1958).

Basic Reservoir Engineering

Fig. 1-23 Capillary pressure gradient causes oil to move out and water to move into a dead.end pore channel when sand is water-wet. (Courtesy Journal of Petroleum Technology - June, 1958).

1.4.7 Residual Oil Residual oil is left in the smaller channels when interfacial tension causes the thread of oil to break leaving behind small globules of oil (See Fig. 1-24). Fig. 1-21 Gas displaces oil first from high permeability pore channels. Residual oil occurs in lower permeability pore channels. (Courtesy Journal of Petroleum Technology - June, 1958).

Fig. 1-24 Water drive leaves residual oil in sand because surface films break at restrictions in sand pore channels. (Courtesy Journal of Petroleum Technology - June, 1958).

1.4.8

Fig. 1-22 Capillary forces cause water to move ahead faster in low permeability pore channel (A) when water is moving slow through high permeability pore channel (B). (Courtesy Journal of Petroleum Technology - June, 1958).

Introduction to Well Testing (Aug 1996)

Relations between Permeability and Fluid Saturation The effective permeability of a fluid is a function of the saturation. In complex porous media it is not a unique function and depends upon the capillary structure of the rock and the wetting characteristics as well as the saturation history.

1-12

Schlumberger

A typical set of oil/water relative permeability curves are shown together with the corresponding capillary pressure relations (Fig. 125). Both are plotted versus water

Fig. 1-25a Water Saturation - Relative Permeability.

the matrix. At higher displacement pressures increasing amounts of water are drained from the core until at point (2) the irreducible water saturation is reached. At point (2) the relative permeability to water becomes zero. Note that the relative permeability to oil at point (2) can not reach 1.0 as the irreducible water reduces the amount of pore volume to oil flow. When water is imbibed into the core which now contains oil and water, the saturation of water increases up to the point (3) where the residual oil saturation is reached. The relative permeability to water can not reach a value of 1.0 because of the residual oil remaining in the pores. In spite of the wide variety of pore structures in reservoir rocks, preferential wettabilities between fluids and rock surfaces, and fluid properties, normalised plots of relative permeabilities against saturation exhibit general similarities. It has thus become possible to formulate theoretical, semi or even purely empirical relationships to assist in obtaining values of effective permeabilities. 1.5

Fig. 1-25b Water Saturation - Capillary Pressure.

saturation to illustrate their relationship. Starting with a 100% water saturated core at point (1) the threshold pressure must be exceeded before the non-wetting fluid (oil) can displace the wetting phase fluid, water from Introduction to Well Testing (Aug 1996)

Fluid Properties

Nearly all naturally occurring petroleum accumulations are made up of an extremely large number of organic compounds, primarily hydrocarbons, all mixed together. Seldom are two crude oil samples found that are identical and certainly never are two crude oils made up of the same proportions of the various compounds. The fact that carbon atoms have the ability to form long branching and cyclic chains allows an almost limitless diversity in the molecular composition of petroleum deposits. Different specialists involved in petroleum exploration and production have different reasons for wishing to examine and characterise the hydrocarbon fluids and water found together in petroleum reservoirs. A chemical engineer may be interested in a crude oil’s composition only as it relates to 1-13

Section 1 the amount of commercial products the oil will yield after refining. An exploration geochemist might have an interest in an oil or reservoir water’s composition insofar as it sheds light on the origin, maturation and degradation of the oil or helps point way toward a better geological interpretation. The petroleum engineer is particularly concerned with the analysis of hydrocarbons in order to determine their behaviour under varying conditions of pressure and temperature that occur in the reservoir and piping systems during the production process. 1.5.1 Components of Hydrocarbon Since hydrocarbon molecules have specific ratios of hydrogen and carbon atoms, hydrocarbon compounds making up petroleum can be grouped chemically into a few series, although each series may have thousands of members. The most common hydrocarbon compounds are those of paraffin or alkane series which include methane, ethane, propane, butane, etc. They are straight chain or branched configurations of carbon and hydrogen atoms that follow the general formula CnH2n+2. The alkanes are saturated, that is, the carbon atoms are connected with single bonds.

Fig. 1-26 Models for methane and propane, showing the tetrahedral nature of the carbon-hydrogen configuration.

Figure 1-26 shows the models used to visualise the structure of these hydrocarbons, and their shorthand formulae. As longer chains are built, it becomes possible to arrange the carbon atoms in either linear or branched fashion without changing the relative number of carbon and hydrogen atIntroduction to Well Testing (Aug 1996)

Basic Reservoir Engineering oms. These different arrangements are called isomers and possess different physical properties. All straight chain alkanes form CH4 (methane) to C40H82 (tetrocontane) have been identified in crude oil. Typically they amount to 15% to 20% of the oil. The possible isomers for these alkanes range from two for butane to 6.2 x 1013 for tetracontane. The alkanes are characterised by their chemical inertness, which probably accounts for the stability over long periods of geological times. The first four members of the series (methane, ethane, propane, butane) exist as gases under standard conditions of pressure and temperature. Those from C5H12 (pentane) to about C17H36 are liquids and C18H38 and higher are wax-like solids. Paraffin is a mixture of these solid members of the series. Saturated hydrocarbons that form closed rings rather than chains belong to a series known as cycloalkanes (also called cycloparaffins or naphtenes). These hydrocarbons follow the general formula CnH2n. Being saturated, they are relatively stable and possess chemical properties similar to those of the alkanes. Unsaturated hydrocarbons are compounds that contain a carbon-carbon double bond. These compounds can add hydrogen to their structures under appropriate conditions, and are therefore said to be unsaturated (with hydrogen). One class of hydrocarbons that contains carbon-carbon double bonds is the arene series (also called aromatic because many of them have fragrant odours). This group is made up of derivatives of benzene, whose formula is C6H2n and whose unique structure allows it to be relatively stable and unreactive. Arene hydrocarbons are either liquids or solids under standard conditions and are common constituents of crude oil.

1-14

Schlumberger

Other unsaturated compounds include alkenes or cycloalkenes (also known as olefins) and the acetylene series (triple carboncarbon bond). Olefin compounds are very uncommon in crude oils and the acetelyne series is virtually absent. This is undoubtedly due to their high degree of reactivity and their tendency to become saturated with hydrogen forming alkanes. Of the eighteen different possible hydrocarbon series, therefore, alkanes, cycloalkanes and arenes are the common constituents of most crude oils. 1.5.2 Classification of Hydrocarbons Common oilfield classification of oil and natural gas rely on observed producing characteristics and easily measured specific gravity. The gas-oil ratio (GOR), gas gravity and oil gravity are used to categorise reservoir hydrocarbons.

Fig. 1-27 Structure of several members of the cycloalkane series of hydrocarbons. Cyclohexane has an actual three-dimensional geometry that is slightly puckered, not planar as the formula suggests.

Gas-oil ratio (GOR) in this case refers to the cubic feet of gas produced per barrel of liquid (or cubic meters per cubic meter), with both volumes measured at standard conditions of temperature and pressure. Specific gravity is the ratio of the density of a substance to the density of some reference substance. Introduction to Well Testing (Aug 1996)

• For gases, the standard reference is dry air

at the same temperature and pressure as the gas in question.

• For liquids, the reference is pure water at 60°F and one atmosphere (14.7 psia).

• For hydrocarbon liquids, the API gravity

scale is most commonly used in the oil industry. It expands and inverts the range of numerical values for oil specific gravity. Water has an API gravity of 10.0 and the relationship between API and specific gravity is given as;   141.5 ° API =  - 131.5 (Specific G ravity at 60° F) 

Fig. 1-28 Structure of several members of the arene, or aromatic, series. The formula convention used here to represent benzene is called a Kekulè. The compounds shown are common constitutes of crude oil.

Figure 1-29 shows the classifications of reservoir fluids based on GOR and fluid gravities. In general, low GOR, low API oils have lesser amounts of the light paraffinic hydrocarbons, while dry gases are composed almost entirely of these compounds. Sampling and analysing the behaviour of dry gas or black oil systems are relatively straight forward procedures. Condensate and volatile oil systems, on the other hand, can be much more complex in terms of their physical chemistry. 1-15

Section 1

Basic Reservoir Engineering

Reservoir fluid

Surface appearance

GOR range

Dry Gas

Colorless gas

Wet gas

Colorless gas with small amount of clear or straw-colored liquid Colorless gas with significant amounts of light-colored liquid Brown liquid with various yellow, red or green hues

Essentially no liquids Greater than 100 MSCF/bbl ( m3/ m3)

Condensate “Volatile” or high shrinkage oil “Black” or low shrinkage oil Heavy oil Tar

Gas specific gravity 0.60-0.65

API gravity

C1

C2

C3

C4

C5

C6+

96

2.7

0.3

0.5

0.1

0.4

0.65-0.85

60o

3 to 100 MSCF/bbl (900-18000 m3/ m3)

0.65-0.85

50o-70o

87

4.4

2.3

1.7

0.8

3.8

About 3000 SCF/bbl (500 m3/m3)

0.65-0.85

40o-50o

64

7.5

4.7

4.1

3.0

16.7

Dark brown to black viscous liquid

100-2500 SCF/bbl (20450 m3/m3

30o-40o

49

2.8

1.9

1.6

1.2

43.5

Black, very viscous liquid Black substance

Essentially no gas in solution Viscosity>10000 cp

10o-25o

20

3.0

2.0

2.0

2.0

71

<10o

90+

Fig. 1-29 General categories of reservoir hydrocarbons. There are no definite boundaries between these classifications and usage may vary depending on location. Gravities and GOR are also dependent on separation conditions.

1.5.3

Characteristics of Formation Water In addition to liquid and gaseous hydrocarbons, petroleum reservoirs always contain a third fluid, water. Formation water is termed connate water if it is believed to be a remnant of the original water in which the sediment was deposited. Meteoric water refers to formation waters that originate as rainfall and are carried into the ground via outcrops, fractures or permeable sediments. Interstitial water is the preferred term for the formation water that shares the pore space of the hydrocarbon reservoir with oil and gas, regardless of origin. Interstitial water saturations in petroleum reservoirs usually range from 10% to 50% of the pore space and water saturations can vary throughout a reservoir depending on the pore structure of the rock. Formation waters are most commonly distinguished by their varying degrees of salinity, that is, the amount of dissolved ions present in the water. The ionic composition of formation water is usually measured and expressed Introduction to Well Testing (Aug 1996)

in milligrams per litre (mg/l) or parts per million (ppm). A unit volume of solution is taken to have a million parts of weight, and the number of parts of weight contributed to the solution by a given ionic component is the parts by weight per million. Parts per million is only equivalent to milligrams per litre at a solution density equal to that of pure water. Since most formation waters have densities close to this value, the difference between numerical values is therefore not great. However, for an extremely salty water the difference could be as great as 20%. Salinity is the most important characteristic that can be measured for a formation water sample. The ways in which formation water and hydrocarbon mixtures change with pressure and temperature depend upon the degree of salinity. For example, the solubility of natural gas in a formation water with 150,000 ppm total dissolved solids is only about half the solubility in pure water. However, even in pure water, the solubility of natural gas is not great; it is only about 10 to 30 SCF/bbl (2 to 5 m3/m3) under most reservoir conditions.

1-16

Schlumberger

Determination of the formation water salinity is more important for correct log interpretation. The resistivity of a formation water is related to its salinity, and a value of formation resistivity is necessary for the quantitative evaluation of resistivity logs. Although this value can be estimated from log data, the best results are obtained when a sample of formation water is retrieved and analysed. 1.6

Phase Behaviour

A phase is a definite portion of a system which is homogeneous throughout and can be separated from other phases by distinct boundaries. Solids, liquids and gases are phases of matter which can occur, depending on pressure and temperature. Commonly, two or three different fluid phases exist together in a reservoir. Any analysis of reservoir fluids depends on the relationships between pressure, volume and temperature of the fluids commonly referred to as the PVT relationship. It is customary to represent the phase behaviour of hydrocarbon reservoir fluids on the P-T plane showing the limits over which the fluid exists as a single phase and the proportions of oil and gas in equilibrium over the two phase P-T range. 1.6.1

Phase Behaviour of a Single Component System Single component hydrocarbons are not found in nature, however it is beneficial to observe the behaviour of a pure hydrocarbon under varying pressures and temperatures to gain insight into more complex systems. As an example, the PVT cell shown at the upper left of figure 1-30 is charged with ethane at 60° F and 1000 psia. Under these conditions, ethane is in a liquid state. If the cell volume is increased while holding the temperature constant, the pressure will fall rapidly until the first bubble of gas appears. Introduction to Well Testing (Aug 1996)

This is called the bubble point. Further increases of cylinder volume at this stage does not reduce the pressure provided the temperature is held constant. The gas volume increases at this constant pressure until the point is reached where all the liquid is vaporised. This is called the dew point. At this stage, further increase of cylinder volume at constant temperature results in a hyperbolic reduction in pressure as the ethane gas expands. A series of similar expansions at varying temperatures produces a three dimensional chart (Figure 1-31). The locus of bubble points obtained at various temperatures projected on the pressure-temperature plane is a line called the vapour pressure curve. At pressures above the vapour pressure curve, ethane exists in the liquid phase and beneath it in the gaseous phase. The vapour pressure curve for single component systems terminates at the critical point. As the critical point is approached the properties of the gas and the liquid phases approach each other, and they become identical at the critical point. 1.6.2

Phase Behaviour of a MultiComponent System Consider the phase behaviour of a 50:50 mixture of two pure hydrocarbon components on the P-T plane shown in figure 1-32. The vapour pressure and bubble point lines do not coincide but form an envelope enclosing a broad range of temperatures and pressures at which two phases (gas and oil) exist in equilibrium. The dew and bubble point curves at that temperature and pressure at which liquid and vapour (gas) phases have identical intensive properties, density, specific volume etc. Similarly phase diagrams can be produced for different reservoir fluids which highlight the large variety and effects of reservoir fluid properties and the compressibility of gases.

1-17

Section 1

Basic Reservoir Engineering

Fig. 1-30 Phase behaviour of a single-component hydrocarbon.

Fig. 1-31 Three-dimensional diagram of single-component system.

Fig. 1.32 Vapor pressure curves for two pure components and phase diagram for a 50:50 mixture of the same components.

Introduction to Well Testing (Aug 1996)

1-18

Schlumberger

1.6.3

Reservoir and Fluid Compressability Boyle’s, Charles’s and Avogadro’s laws showed that for an ideal gas pV = nRT where: p V n T R

= = = = =

absolute pressure. volume. moles of gas. absolute temperature. constant.

At very low pressures, many gases exhibit ideal behaviour and the equation is valid. At higher pressures and temperatures such as those found in oil and gas reservoirs the equation does not hold. The petroleum industry has found that a simple correction factor can be added to the ideal gas law which describes the behaviour of gas mixtures under oilfield conditions. The factor is called the compressibility factor or z factor and varies with pressure, temperature and composition. The gas equation of state is thus modified to: pV = znRT where z is the dimensionless compressibility factor. 1.6.4

Conversion Factors between Surface and Downhole Volumes Conversion of downhole volumes (or volumetric rates) of oil, gas and water to equivalent volumes at surface conditions is made so frequently that it is convenient to use conversion factors which account for the overall changes which effect the fluids as they pass from one set of conditions to another. (Typically - compressibility and solubility).

Oil Volume is measured in the API barrel or cubic meter at stock tank conditions - conventionally 60° F (15.5°C) and 14.7 psia (1 atmosphere). Gas Rate is measured in standard cubic feet scf or in cubic meters at the same reference conditions 60° F (15.5°C) and 14.7 psia (1 atmosphere).

Fig. 1-33 Relationships between surface and downhole volumes - dissolved gas systems.

Units are standard cubic feet per stock tank barrel scf/B or cubic meters of oil per cubic meter of oil m3/m3. (as there are 5.6 cubic feet in a barrel then; scf/b ÷ 5.6 = m3/m3). The basic surface to downhole relationships are shown diagramatically in Fig. 1-33. Formation volume factors are designated by the letter B with a suffix denoting the fluid phase concerned. Formation volume factor is a function of fluid composition and the pressure and temperature difference between the downhole and reference state;

The following units are commonly used to measure hydrocarbon volumes; Introduction to Well Testing (Aug 1996)

1-19

Section 1

Basic Reservoir Engineering

Formation Volume Factor is equal to; Volume at downhole conditions Volume at reference conditions Figure 1-33 above shows that the total gas at surface is the sum of the solution gas evolved from the downhole liquids plus free gas produced independently from the oil. The gas volume factor accounts for expansion of free gas. Oil shrinks in volume between downhole and surface conditions primarily as a result of the solution gas evolved. A typical range of the formation volume factor for oil Bo is 1.2 for low GOR oil to 1.4 or higher for more volatile oils. Solubility of gas in water is low and the combined effects of reduction in temperature, pressure and loss of solution gas has a small (but for some purposes important) effect on the formation volume factor for water Bw. 1.6.4.1 Formation Volume Factor of Gas, Bg The gas formation volume factor may be calculated from PVT measurements on a gas sample, or it may be calculated using the gas equation of state; p1 V1 then

z 1 T1

Bg =

=

1.6.4.2 Formation Volume Factor of Oil, Bo The formation volume factor of oil is best determined by PVT measurement on a reservoir fluid sample. Figure 1-34 is a typical plot of PVT data for an undersaturated oil. From 8000 psig to the bubble point pressure at 6350 psig, the Bo increase is due to expansion of the undersaturated oil. On expansion from 6350 psi to atmospheric pressure increasing amounts of gas are liberated. Note that the Rs, the amount of gas in solution is almost directly proportional to the system pressure. Bo being dependent essentially on the amount of dissolved gas increases with pressure up to the bubble point, where all of the available gas is dissolved and then decreases at a a rate determined by the liquid compressibility. Solubility of natural gas in oil is dependent on the composition of the hydrocarbons, the temperature and pressure applied. As for gas, charts are available to estimate the values required in determining Bo.

p 2 V2 z2 T2

V 2 p1 T 2 z 2 = V1 p 2 T1 z 1

where conditions 1 are standard and conditions 2 are bottom hole. Charts based on the relationship and correlations between z and specific gravity, pressure and temperature have been established to simplify calculations.

Introduction to Well Testing (Aug 1996)

Fig. 1-34 Typical PVT data for differential vaporization of an undersaturated oil at constant temperature (305oF).

1-20

Schlumberger

1.6.4.3 Formation Volume Factor of Water, Bw The formation volume factor of water exists due to small amounts of dissolved gases and salts which make the water compressible. The solubility of natural gas in water and brine is small and the effect on compressibility can be neglected for small changes of pressure and temperature and Bw can be approximated as 1.0. For large changes in pressure and temperature, Bw may be approximated from charts. 1.6.5

Fluid Density Correlations

1.6.5.1 Gas Density Gas specific gravity γg, is widely used in the oil industry to characterise natural gases. Gas specific gravity is defined as the ratio of the density of gas to the density of air, both at standard conditions.

γg =

ρgsc

(ρ air )sc

The weight of any volume of a gas can be determined by multiplying the volume of gas times γg times ρair. The density of air at standard conditions is 0.001223 gm/cc or 0.0762 lb/cu ft. The density of gas at any temperature and pressure can be found from the gas formation volume factor, Bg 1

Bg

=

ρ gwf

ρ gsc

by rearranging ;

ρgwf = γ g (0 .001223)

1

Bg

(gm/cc)

Introduction to Well Testing (Aug 1996)

1.6.5.2 Oil Density At standard conditions, the density of oil is equal to the weight divided by the volume; or in equation form; ρ osc = W osc V osc At well flowing conditions, the density of oil is still equal to the weight divided by the volume. It is, however, not quite straight forward because the weight of the oil has been increased by dissolved gas, and the volume of the oil has been increased by the oil formation volume factor: ρ owf = or ρ owf =

W osc + W dis. gas V osc x Bo ρ osc + ρ air γ g R s Bo

1.6.5.3 Water Density The density of gas-free water is a function of temperature, pressure and water salinity. This value is typically determined from charts. 1.6.6

Viscosities

1.6.6.1 Gas Viscosity At elevated temperatures and low pressures, low gravity gases closely resemble “perfect” gas in their behaviour, while at low temperatures and high pressures, the heavier gases resemble liquids. 1.6.6.2 Oil Viscosity The viscosity of a crude oil decreases with a temperature increase and with an increase of dissolved gas. Heavier oils are generally more viscous than lighter oils of the same hydrocarbon base. 1-21

Section 1 1.6.6.3 Water Viscosity The viscosity of water is primarily a function of temperature and salinity. Charts are used to gain approximations of viscosity.

Basic Reservoir Engineering

• Invasion of the original oil bearing reservoir by the expansion of the water from an adjacent or underlying aquifer.

1.6.6.4 Formation Compressibility In addition to the compressibility of fluids which can play an important role in determining an accurate picture of the simulated reservoir model, it is also just as important to have a measure of the formation rock compressibility. Formation compressibility is best measured in the laboratory although again correlations and charts do exist which will give good approximations under certain conditions. 1.7

Reservoir Drive Mechanisms

1.7.1 Oil Reservoirs Oil can be recovered from the pore spaces of a reservoir rock, only to the extent that the volume originally occupied by the oil is invaded or occupied in some way. There are several ways in which oil can be displaced and produced from a reservoir, and these may be termed mechanisms or “drives”. Where one replacement mechanism is dominant, the reservoir may be said to be operating under a particular “drive”. Possible sources of replacement for produced fluids are:

• Expansion of undersaturated oil above the bubble point.

• Expansion of rock and of connate water. • Expansion of gas released from solution in the oil below the bubble point.

• Invasion of the original oil bearing reser-

voir by the expansion of the gas from a free gas cap.

Introduction to Well Testing (Aug 1996)

Fig. 1-35 Dissolved gas drive reservoir.

Since all replacement processes are related to expansion mechanisms, a reduction in pressure in the original oil zone is essential. The pressure drops may be small if gas caps and aquifers are large and permeable, and, under favourable circumstances, pressure may stabilise at constant or declining reservoir offtake rates. The compressibilities of undersaturated oil, rock and connate water are so small that pressures in undersaturated oil reservoirs will rapidly fall to the bubble point if there is no aquifer to provide water drive. So these expansion mechanisms are not usually considered separately, and the three principal categories of reservoir are:

• Solution gas drive (or depletion drive) reservoirs.

1-22

Schlumberger

• Gas cap expansion drive reservoirs. • Water drive reservoirs. Frequently two or all three mechanisms (together with rock connate water expansion) occur simultaneously. 1.7.2 Solution Gas Drive Reservoirs If a reservoir at its bubble point is put on production, the pressure will fall below the bubble point pressure and gas will come out of solution. Initially this gas may be a disperse, discontinuous phase, but, in any case, gas will be essentially immobile until some minimum saturation - the equilibrium, or critical gas saturation, is attained. The actual order of values for critical saturation are in some doubt, but there is considerable evidence to support the view that values may be very low - in the order of 1% to 2 % of the pore volume. Once the critical gas saturation has been established gas will be mobile, and will flow under whatever potential gradients may be established in the reservoir - towards producing wells if the pressure gradient is dominant - segregating vertically if the gravitational gradient is dominant. Segregation will be affected by vertical permeability variations in layers, but is known to occur even under apparently unfavourable conditions. Initially the gas-oil ratio of a well producing from a closed reservoir will equal solution GOR. At early times, as pressure declines and gas comes out of solution, but cannot flow to producing wells, the producing GOR will decline. When the critical gas saturation is established and if the potential gradients permit, gas will flow towards producing wells. The permeability to oil will be lower than at initial conditions, and there will be a finite Introduction to Well Testing (Aug 1996)

permeability to gas so that the producing gas oil ratio will rise. As more gas comes out of solution, and gas saturations increase, permeability to gas increases, permeability to oil diminishes and this trend accelerates. Ultimately, as reservoir pressure declines towards abandonment pressure, the change in gas formation volume factor offsets the increasing gas to oil mobility ratio and the gas oil ratio trend is reversed; i.e. although the reservoir GOR may continue to increase, in terms of standard volumes, the ratio standard cubic ft/stock tank barrel may decline. In addition to the effect of gas on saturation of, and permeability to, oil, the loss of gas from solution also increases the viscosity of the oil and decreases the formation volume factor of the oil.

Fig. 1-36 Production data dissolved gas drive reservoir. (Courtesy API, Drilling and Production Practicles-1943).

1.7.3

Gas Cap Expansion Drive Reservoirs The general behaviour of gas drive reservoirs is similar to that of solution gas drive reservoirs, except that the presence of free gas retards the decline in pressure. By definition the oil must be saturated at the gas oil contact, so that decline in pressure will cause the release of gas from solution, but the rate of release of gas from solution, and the build up of gas saturation and of gas permeability, will be retarded. At higher prevailing pressures, oil viscosities are lower, and provided that 1-23

Section 1

Basic Reservoir Engineering

the free gas phase can be controlled, and not produced directly from producing wells, better well productivities and lower producing gas oil ratios can be maintained.

Fig. 1-38 Production data-gas cap drive reservoir. (Courtesy API, Drilling and Production Practices-1943.)

1.7.4 Water Drive Reservoirs If a reservoir is underlain by, or is continuous with a large body of water saturated rock (an aquifer) then reduction in pressure in the oil zone, will cause a reduction in pressure in the aquifer. Although the compressibility of water is small (± 3 x 10 -6 psi -1) the total compressibility of an aquifer includes the rock pore compressibility. Fig. 1-37 Gas cap drive reservoir.

Under residual conditions the stock tank oil left in place is So/Bo and the smaller this factor the greater will be the oil recovery. Consequently the higher the pressure at abandonment, the greater the value of Bo, and the smaller this term becomes. In addition abandonment of wells and reservoirs depends primarily upon an “economic limit” - the rate of production required to pay for operating costs, and direct overheads - and an oil flow rate, which depends upon Ko/µo, which will be greater at any given saturation (and so given Ko ) under pressure maintenance conditions due to the lower oil viscosity than under depletion conditions.

Introduction to Well Testing (Aug 1996)

(± 5 x 10 -6 psi -1) making the total compressibility in the order of; 8 x 10 -6 .psi -1. The apparent compressibility of an aquifer can be substantially greater if some accumulation of hydrocarbons exist in small structural traps throughout the aquifer. An efficient water driven reservoir requires a large aquifer body with a high degree of transmissivity allowing large volumes of water to move across the oil-water contact in response to small pressure drops.

1-24

Schlumberger

As with gas cap drive reservoirs, the maintained pressures lead to lower viscosities and higher Bo values at any given saturation, reducing the saturation and minimising the term So/Bo hence the stock tank oil left at any given economic limit. While reservoir drive mechanisms may be classified into the three categories we have discussed, most often two or more of these mechanisms act simultaneously in a combination drive.

Fig. 1-39 Water drive reservoir.

This replacement mechanism has two particular characteristics - first there must be pressure drops in order to have expansion, and secondly, the aquifer response may lag substantially, particularly if transmissivity deteriorates in the aquifer. A water drive reservoir is then particularly rate sensitive, and so the reservoir may behave almost as a depletion reservoir for a long period if off-take rates are very high, or as an almost complete pressure maintained water drive reservoir if off-take rates are low, for the given aquifer. Because of the similarity in oil and water viscosities (for light oils at normal depths) the displacement of oil by water is reasonably efficient, and provided that localised channelling, fingering or coning of water does not occur, water drive generally represents the most efficient of the natural producing mechanisms for oil reservoirs.

Introduction to Well Testing (Aug 1996)

Fig. 1-40 Production data-water drive reservoir. (Courtesy API, Drilling and Production Practices-1942).

Fig. 1-41 Combination drive reservoir.

1-25

Section 1 1.7.5

Discussion of Recovery Efficiency (including gravity segregation) One mechanism, only briefly referred to, but which has an important role in several aspects of reservoir behaviour is that of gravity segregation the movement of phases countercurrent to each other, (generally of gas and oil) under the influence of the gravitational potential g x ∆p. Considering the solution gas drive reservoir, the behaviour described earlier assumes essentially that gas saturations build up uniformly throughout the oil zone without any saturation gradients in the vertical direction. (Saturation gradients existing as a result of horizontal pressure gradients, i.e. the pressure drops near the well bore). Under these conditions the expected recovery efficiency will depend on the economic limit for wells and could be as low as 2% - 3% for low permeability reservoirs with high viscosity, low gas oil ratio oils, and up to about 15% or so for high permeability reservoirs, normal GOR low viscosity oils, but will rarely exceed this range.

Basic Reservoir Engineering grating to structurally high positions, with oil counterflowing downwards. This mechanism has two effects. Firstly the oil saturation in the lower parts of the reservoir is maintained at a value higher than the average oil saturation - so that permeability to oil is higher, and permeability to gas lower than for the “pure” solution gas drive case. The producing gas oil ratio is then lower than for solution gas drive alone. Secondly, the lower producing gas oil ratio involves smaller gross fluid withdrawals than would otherwise be the case, so that the pressure decline at any given oil cumulative production will be smaller. The segregated gas may form a secondary gas cap, and the later life of a reservoir may then be similar to that of a primary gas cap drive reservoir. Under these conditions the recovery efficiency will be higher if the economic limit is low - possibly very much higher and may approach or even exceed the range 20% to 40% of oil in place. Gravity drainage plays its greatest role in dual porosity systems with great contrast where almost complete segregation can take place in the secondary porosity system, and the producing wells produce throughout at solution gas-oil ratio. Gravity drainage is then the predominant mechanisms in draining oil to residual saturation in the secondary gas cap.

Fig. 1-42 Reservoir pressure trends for reservoir under various drives. (Courtesy API, Drilling and Production Practices-1943).

If the vertical permeability to gas is non-zero however, there will be a vertical component of gas flow, under the gravitational potential, and gas will “segregate” in the reservoir, miIntroduction to Well Testing (Aug 1996)

The recovery efficiency of water drive reservoirs will be governed by an economic limit, the limit in this case being dictated by water handling problems. Provided that water can be controlled reasonably, efficiencies of 30% to 40% would be expected and sometimes under extremely favourable conditions recovery efficiencies up to 50% might be achieved. 1-26

Schlumberger

(Ultimately, of course, calculating a recovery efficiency depends on knowing the initial oil in place, and an apparently high recovery factor might be the result of under-estimating oil in place).

Fig. 1-43 Reservoir gas-oil ratio trends for reservoirs under various drives. (Courtesy API, Drilling and Production Practices-1943).

Introduction to Well Testing (Aug 1996)

1-27

Section 1

Introduction to Well Testing (Aug 1996)

Basic Reservoir Engineering

1-28

Section 2 Completion Technology

Section 2

Introduction to Well Testing (Aug 1996)

Completion Technology

2-2

Schlumberger

2.0

Completion Technology and Types

2.1

Introduction

The completion phase of a well development is a term generally used for a well that is prepared for production after drilling is completed. It essentially consists of installing selected tubulars and associated components into a pre-conditioned wellbore to ensure that production can be performed in a safe, controlled and efficient manner. It follows therefore that during the exploration, appraisal and development testing of wells, a temporary completion (often referred to as a Drill Stem Test String) is used as the well is temporarily produced to gain valuable reservoir data for the future development of the field. There are many different completion types and methods. The technique, like all others within the industry, is continually evolving especially with the onset of new drilling methodology such as horizontal wells and multi-lateral well development which all need cost effective completion methods while ensuring productivity is enhanced rather than impaired. The completion design engineer has to tailor each completion, not just to a specific reservoir but often to a specific well. Completion design is a complex science influenced by numerous factors such as temperature, pressure, effluent type/characteristics, reservoir type /characteristics, expected productivity and so on and so forth. It is therefore no surprise that the completion has an enormous impact on the productivity potential of a well and its life throughout the production cycle. This section aims to review the basic completion components and review the different completion types so that an understanding of their impact on the productivity of a well and Introduction to Well Testing (Aug 1996)

their involvement in the well testing operation is appreciated. 2.2

Completion Components

Completion is generally associated with the tubulars and general hardware which are installed into the wellbore, however there are two important aspects to consider when designing a completion which are very closely related; • Completion Technique “Communication between the reservoir and the wellbore”. • Completion Components “Transmission of the produced fluids to surface”. Whereas the natural progression of produced fluids warrants the order given above, we will first look at the essential completion components as an understanding of the different completion techniques requires a knowledge of basic completion components. The completion components will differ depending upon whether the well is eruptive or non-eruptive. Non-eruptive wells require “artificial lift” methods to give the extra energy required to drive the effluent to surface. These methods include the familiar “nodding donkey pump” or sucker-rod pump usually associated with on-land low producing wells, gas lift or electrical submerged pumps for the higher rate wells.

2-3

Section 2

Completion Technology count by the completion design engineer. Apart from the specialised equipment associated with the artificial lift technique chosen, there are still many components common to both systems. We will categorise these components into the following basic groups; • • • • •

Production Tubulars Packers Flow Control Subsurface Safety Systems Ancilliary Components

2.2.1 Production Tubulars The main component of any completion is the production tubing, as it is the one element required in considerable quantity (depth dependent). The choice of tubing is not a straight forward exercise of ordering enough length of pipe for the well concerned, the wrong choice of tubing can lead to disastrous consequences later on during the life of a well and so the economies of scale need to be properly scrutinised. In general the tubing type will mainly depend on the following; • Reservoir Fluid. • Presence of corrosive agents - H2S, CO2 etc. • Well Deliverability. • Pressure/Temperature These factors will govern the grade of steel required, the tubing size and the pressure seal or connection between the tubing joints. Fig. 2-1 Basic - Single String Completion for Oil or Gas Production.

Eruptive wells which have sufficient bottom hole pressure to provide the energy required to transmit the effluent to surface. It is also possible for wells which were initially eruptive to become non-eruptive as the well depletes and this must also be taken into acIntroduction to Well Testing (Aug 1996)

2.2.2 Packers The primary function of a packer is to create a seal between the production tubing and the casing. It also serves to isolate the casing from corrosive reservoir fluids and to enable selective production in the case when several producing zones are available. A packer also requires a seal between the packer and the tubing. 2-4

Schlumberger

There are two main types of packers; There are two main types of packers; • Permanent. • Retrievable. 2.2.2.1 Permanent Packers Permanent packers are generally used in high pressure differential or extreme environments. The packer consists of a dual slip system which locks the packer in both directions. It is termed “permanent” as once it has been set, it requires milling (or in extreme cases drilling) to remove it as opposed to manipulation of the tubing string. The packer is set either on wireline-electrically or on drillpipe-hydraulically. This depends to a certain extent on the deviation of the well or the weight of the packer and associated tailpipe assembly. A permanent packer contains a seal bore into which a seal locator on the production tubing will sting thus creating the seal between the packer and the tubing. This seal locator may be of the type that latches onto the packer or of sufficient length to allow for tubing movement which occurs due to forces created within the tubing by temperature differences (expansion/contraction), pressure differential, piston and buckling effects. It also contains a nipple profile such that the packer can be plugged off in the event of wanting to isolate the formation below during workovers. A variation on the permanent packer is the polish bore receptacle, this is the case whereby the female seal bore is integral to the casing usually at the liner top. The main advantage of these systems are their large bores. Fig. 2-2 Camco - “HSP-1” Hydraulic Set Permanent Packer.

Introduction to Well Testing (Aug 1996)

2-5

Section 2

Completion Technology

2.2.2.2 Retrievable Packers Retrievable packers are installed as part of the production tubing. They are designed to be released and removed without destroying the packer especially during workover campaigns. The packer can be set via tubing manipulation or more commonly - hydraulically. A packer can either be a tension or compression set packer depending on the application encountered. Modern retrievable packers have dual slip systems holding them both from the top and the bottom thus giving some of the advantages of the permanent packer while remaining retrievable. As the tubing is an integral part of the packer body, there is no movement between the packer and the tubing therefore to allow for tubing movement an expansion joint would be required to be installed in the tubing string usually one or two joints above the packer. Permanent and retrievable packers are available in a multitude of sizes and shapes depending on the application required, further information can be obtained from the promotional catalogues of the major completion component suppliers. 2.2.3 Flow Control Flow control in completions is achieved via numerous ancillary components strategically placed within the production tubing string. 2.2.3.1 Landing Nipples Landing nipples provide known depth seats into which various types of flow control devices can be located and landed. They have a honed bore for pressure sealing and a recess for the locking mechanism of the mandrel which accompanies the device being set. Fig. 2-3 Camco - “HRP-1” Hydraulic Set Retrievable Packer.

Introduction to Well Testing (Aug 1996)

2-6

Schlumberger

through the nipple. The nipple is either bottom no-go or top no-go. They are usually placed at the bottom of a series of similar sized selective nipples to prevent the loss of slickline tools. Selective landing nipples are installed at various strategic depths within the tubing string. They are profiled internally such that only certain setting devices can land in them or special procedures are required to locate them otherwise the setting device will pass straight through. Landing nipples are used as, among other things, a means to install plugs to seal off unwanted zones or temporarily create a pressure barrier, set chokes to restrict flow and to hang instruments during routine pressure/temperature surveys 2.2.3.2 Sliding Side Door A sliding side doors or SSD is designed as a means of providing communication between the tubing string and the annulus. It consists of a ported nipple with a moveable sleeve which can open or close the ports. The process of opening or closing is normally done by using a mechanical shifting tool run on slickline although there are other variations which operate under pressure. The main applications are; Fig 2-4a

Fig 2-4b

Fig. 2-4a and Fig. 2-4b “Camco - “D-Type” No-Go Landing Nipple and associated “C-Type” Wireline Lock”.

There are several types of landing nipples but the most common types are No-Go (or non-selective) and Selective (or universal). No-Go landing nipples, as the name suggests, are designed such that the setting device with the corresponding profile can not pass Introduction to Well Testing (Aug 1996)

• A means of circulating the well either to kill it prior to workover or for circulating completion fluids. • Selective testing or treating in a single string multi-zone well. • Gas lifting in the absence of proper gas lift mandrels. • Use in multi-string completions for selective production. 2-7

Section 2

Completion Technology

2.2.4 Subsurface Safety Systems A subsurface safety system is installed as a means to shut in the well in the event of an uncontrolled event at the surface. They are categorised into two groups; • Subsurface Controlled Subsurface Safety Valves - (SSCSSV) • Surface Controlled Subsurface Safety Valves - (SCSSV) 2.2.4.1 Subsurface Controlled Subsurface Safety Valves Subsurface controlled subsurface safety valves are the old generation of safety valves. They are mechanically operated valves usually plug valves (although flapper types are available) which are set to react to significant changes in flowing pressures causing the valves to close. They are not as reliable as surface controlled systems and they would certainly not be considered for use in high profile applications such as the offshore environment or under severe operating conditions. 2.2.4.2 Surface Controlled Subsurface Safety Valves

• Surface controlled subsurface safety valves are the modern safety valves installed in higher profile completions. (Their use is compulsory in certain countries) They are categorised into two groups; • Wireline Retrievable • Tubing Retrievable.

Fig. 2-5 “Camco - “DB-1” Sliding Side Door”.

Introduction to Well Testing (Aug 1996)

2-8

Schlumberger

Fig 2-6a Camo - “TRDP.5” Tubing, Retrievable SCSSV. Introduction to Well Testing (Aug 1996)

Fig. 2-6b Camco - “WRP-1A” Wireline Retrievable SCSSV. 2-9

Section 2 The common means of operation is via applying hydraulic pressure through a 1/4” stainless steel control hose to a ported nipple which forces the valve to open. (All valves should be fail-safe closed such that in the event of loss of pressure they will close). Modern systems are incorporating electrically operated valves. There are many types of valves on the market offering a wide variety of advantages and disadvantages. The main difference being that tubing retrievable valves offer larger internal diameters but can not be removed to be fully serviced (although some manufacturers offer them with removable critical parts) while wireline retrievable are the converse. 2.2.5

Ancillary Components

2.2.5.1 Wireline Entry Guide This is placed at the bottom of the tubing string to assist entry back into the tubing for slickline or wireline tools. 2.2.5.2 Blast Joint This is used mainly in multiple string completions. It is an extra thick joint of tubing designed to withstand the external erosion forces or blast from adjacent perforations.

Completion Technology 2.2.6 General The most common form of completion is the single string with a single packer. The beauty of this system is its simplicity which in turn leads to reliability as there are less components which could potentially have problems. Leading on from the various components discussed above, it can readily be seen that even the basic single string system could have one or several packers each isolating hydrocarbon bearing zones which can be selectively produced through sliding side doors. The next progression is a dual string completion, whereby two production tubing strings are run together with one string producing a lower zone and the other, an upper zone with a packer isolating the two. Other types of completions include concentric string completions, triple string completions etc. Within these various completion types, the available components open up a variety of combinations which can be tailored to each reservoir.

2.2.5.3 Flow Couplings These are placed above and below safety valves or restrictions within the tubing. They are extra thick short joints of tubing to withstand internal erosion created by turbulent flow from restrictions.

Introduction to Well Testing (Aug 1996)

2-10

Schlumberger

Fig. 2-7a Dual String - Gas Production.

Fig. 2-7b Dual String - Oil Production.

Introduction to Well Testing (Aug 1996)

Fig. 2-7c Single String - Gas Injection.

2-11

Section 2 2.3

Completion Technology

Completion Techniques

2.3.1 General Communication between the formation and the wellbore is an essential phase of the completion process as it can directly impact the productivity of the formation. Factors such as hydrocarbon saturation, porosity, permeability, fluid properties and geometry can be measured or inferred from the measurements but they cannot usually be controlled. By contrast, completion can be controlled and thus affect well performance. During the drilling, logging and testing phase of the well, valuable information will have been gained and the relevant completion technique chosen. There are two main categories of completions to consider, with of course the usual variations; • Open Hole Completions

• Cased Hole Completions.

2.3.1.1 Open Hole Completions An open hole completion is when the well is drilled to the top of the target formation and the casing is cemented at this stage. Drilling is continued across the target formation, the drilling mud removed, any necessary stimulation or preparation performed and then the well is completed and produced. Open hole completions are only possible in “competent” rocks that will hold their form and not cave in or crumble - so called hard rock environments. This technique is generally associated with older, cheaper methods of drilling and completing wells and today would only be used in very low profile applications, if at all. Variations on the straight forward open hole completion include gravel packing with slotted liners used to contain the pack. Whereas this technique offers the least restriction to flow from formation to wellbore and as mentioned is an economical Introduction to Well Testing (Aug 1996)

completion, it has many apparent disadvantages; • No possibility for selectively producing or treating different zones.

• Limited control of water or gas encroachment. These two factors alone can play a significant part in the future management of the well and this coupled with safety issues has lead the industry down the road of cased hole and perforated completions. 2.3.1.2 Cased Hole Completions A cased hole completion is when the well has been cased and cemented across the target formation and requires shaped charge perforation to achieve communication between the formation and the wellbore. This is the most common form of completing wells today and our discussion will centre around this technique. 2.4

Completion Types

Completions can be broken down into three main categories; • Natural • Stimulated

• Sand Control In all three the objective is to maximise production through enhancement of some aspect of reservoir performance modelled by the radial flow equation. Of particular importance in the analysis of well productivity is the change in radial flow geometry near the wellbore caused by flow convergence, wellbore damage (from drilling and filtrate invasion), 2-12

Schlumberger

Wellhead

Wellhead

Tubing Casing Packer

Production zone Large shapedcharge gun Mud, oil and salt water Tubing-Conveyed Completion pw < pf

Extreme Overbalance Completion pw > pf

Derrick

Pressure control equipment

Packer

Tubing

Casing

Cable to truck

Cable Casing Production zone

Large shapedcharge gun

Small throughtubing gun

Salt water

Mud, oil and salt water Wireline Casing Gun Completion pw > pf

Wireline Through-Tubing Gun Completion pw < pf

Fig. 2-8 Examples of Different Perforating Techniques.

Introduction to Well Testing (Aug 1996)

2-13

Section 2 perforations (debris), partial penetration and deviation. This damage is known as skin (S) and will be discussed in more detail in section 5, but for the time being can be considered as an induced pressure drop across the completion which effects productivity.

Completion Technology 2.4.2 Stimulated Completions These fall into two broad categories; • Hydraulic Fracturing • Acidizing Occasionally the two are combined in an “acid frac” job.

Fig. 2-9 Pressure Distribution in a Reservoir with Skin.

The aim of the completion design engineer and reservoir engineer is to reduce the influence of skin as much as possible. 2.4.1 Natural Completions The natural completion is usually chosen for sandstone reservoirs with permeabilities above 10 md and porosities above 9 p.u. These reservoirs typically have small damaged zones and limited skin, good transmissibility and stable rock mechanics. They generally do not require stimulation or sand control during primary completion. The objectives of the perforation in this case would be depth of penetration and effective shot density, the perforation diameter is generally unimportant if it is larger than 0.25” (0.5 cm). The deepest penetration with the greatest phase distribution is desirable for production enhancement.

Introduction to Well Testing (Aug 1996)

2.4.2.1 Hydraulic Fracturing Hydraulic fracturing is performed to enhance the effective wellbore radius rw and is usually indicated for reservoirs with small permeabilities (k < 1 md). This is accomplished by injecting fluids and propant at high pressure, which alters the stress distribution in a formation creating a fracture or crack in the rock. Hydraulic fracturing is generally a five step process. • Pre-fracturing treatment. • Fracture initiation and breakdown. • Fracture extension. • Proppant injection. • Cleanout and production. 2.4.2.2 Acidizing Acidizing is a stimulation process used to repair formation damage caused by the drilling or perforating operation. This type of damage is usually associated with plugging of the pore throats around the wellbore. Acidizing removes this damage from the matrix rock by injecting acid into the naturally porous rock at sub-fracturing rates, allowing the acid to dissolve the plugs.

2-14

Schlumberger

posed to this pressure by perforating the casing. The compressed nitrogen drives the stimulation fluid across the perforations into the formation. rs

Prefracturing Treatment

Fracture Initiation

Fracture Extension

ks

h

k p

rw

re Proppant Injection

Cleanout and Production

Fig. 2-10 Hydraulic Fracturing Process.

Acid frac jobs are used to etch the surface of the hydraulically induced fracture. After the fracture closes, the etched surface significantly improves the effective wellbore radius, rw. Acid frac jobs are operationally less complicated because no proppant is used thus eliminating the potential for premature frac termination that may be caused by screenout or problems of proppant flowback. The principle disadvantage of this technique are the expense of the acid fluids and non uniform leak-off resulting in “wormholes”. Acid frac jobs are usually performed on carbonate reservoirs. 2.4.2.3 Extreme Overbalance Perforating Extreme overbalance perforating (EOP) is a new technique developed to stimulate the formation in the region near the wellbore. It involves building up very high pressures in the wellbore by pressuring up with nitrogen in the tubing - much higher than the formation breakdown pressure. When the desired pressure has been reached, the formation is exIntroduction to Well Testing (Aug 1996)

Zone of altered permeability

Fig. 2-11 Well and Zone of Altered Permeability After Acidizing.

2.4.2.4 Effects of Perforation In the stimulated completion, perforation is critical to the success of the completion. In long intervals or multi-zone treatments, the proppant or acid may cover only part of the interval or enter only one zone because of permeability variations. Limiting the number and diameter of perforations can increase the pressure in the casing to a point where intervals of higher stress may be fractured or zones of lower permeability penetrated. This technique is called “limited entry”. The perforation diameter and uniformity are of primary importance for this type of operation because these become the limiting factors in creating pressure restrictions in the well and providing a sealing surface for ball sealers, if they are used.

2-15

Section 2

Completion Technology

For both limited and non-limited entry stimulations, completion success is influenced by three perforation effects; • Perforation Erosion Whereby the flow erodes the perforation thus reducing the pressure differential across it which is important for the limited entry technique. • Perforation Bridging Whereby the perforations bridged with sand particles. ments have shown that there is perforation diameter to reduce fect.

become Experian ideal this ef-

• Perforation Phasing Crucial for fracturing. 2.4.3 Sand Control Completions Oil and gas reservoirs producing through sandstones that are not structurally competent often produce sand along with the formation fluids. Fluid movement through sandstone reservoirs produces stress on the sand grains because of fluid pressure differences, fluid friction and overburden pressure. If these stresses exceed formation restraining forces, sand is produced, and the near wellbore permeability is significantly altered. The consequences include; • Sand plugging of casing, tubing or surface facilities. • Casing collapse resulting from changes in overburden stress.

• Costly disposal of produced sand. There are four general methods of sand control; 2.4.3.1 Production Rate Restriction Sand production can sometimes be prevented by restricting the flow of the fluids through the formation to a rate that avoids collapse of the stable arch that forms around the perforation. This method can be enhanced by utilising high shot densities, uniform perforations and controlled clean-up. 2.4.3.2 Gravel Packing Gravel packing is the oldest technique for controlling sand production. It consists of placing a sand pack between the formation and a wire screen to prevent migration of formation fines into the wellbore. Gravel packing may be accomplished externally in openhole or internally in cased hole. 2.4.3.3 Sand Consolidation It may be possible to consolidate sand by injecting plastic resins into the formation. This process binds the sands together while leaving the pore spaces open. Sand consolidation is normally limited to intervals of 30 ft (9 m) or less. 2.4.3.4 Resin Packs Resin coated gravel may be placed both outside and inside the perforations and casing. As the resin cures, it forms a permeable filter in the perforation tunnels by binding the sand grains together. The result is a strong, highly permeable synthetic sandstone. After curing, the excess material can be drilled out of the casing leaving an open wellbore. This method may be used with or without a screen.

• Destruction of downhole and surface equipment. Introduction to Well Testing (Aug 1996)

2-16

Schlumberger

The success of sand control treatments in cased hole is affected by perforation density, hole diameter and damage. If perforation density is too low or hole diameter too small, large pressure drops occur across the pack introducing large skin effects, reducing well performance and damaging the pack. Perforation plugging caused by perforation damage prevents gravel deposition in tunnels increasing the chance of premature pack failure and reducing well productivity. Perforation phasing is important in maintaining uniform flow patterns around the wellbore. This reduces fluid velocities and subsequent formation sand movement. 2.5

The Effects of Drilling on Completions

There are two main areas where drilling related problems can effect the success of a completion; • Pre-Formation Drilling • Formation Drilling Casing

Cement

Dynamic forces Fluid flow Stable arch

Fig. 2-12 Stable Arch and Forces Formed Around a Perforation in an Unconsolidated Formation.

Introduction to Well Testing (Aug 1996)

2.5.1 Pre-Formation Drilling During pre-formation drilling of the well down to the target formation(s), there are many problems which may be encountered which can effect the planned completion. These generally involve the need to set additional liners because of casing wear from the drill string, poor cement bonds or the need to set an intermediate casing string higher than anticipated due to unexpected over-pressures requiring higher mud weights to control the well. The inclusion of a liner influences the size of the final completion which in turn may directly effect the production potential of the well. Other factors such as the need to side track or irregular hole shapes will also cause complications with completions. 2.5.2 Formation Drilling During the formation drilling of the well, problems encountered can have significant effects later on. Formation damage can be caused by excessive invasion of fluids and solids from the mud systems, over-pressuring the formation can cause fracturing and breakdown of weak zones. Mud quality and washed out zones can further impair the quality of log and core data which in turn will effect the evaluations performed on the reservoir and the choice of the proposed completion. Poor cement bonds between casing and formation can result in communication between different zones. Incorrect perforating programmes can have a major influence on the performance of the completion as discussed in the previous section. Fortunately the age of information technology and the onset of new computer modelling techniques gives the modern engineer all the tools he requires to plan, execute and evaluate his formations as accurately and efficiently as possible.

2-17

Section 2

Completion Technology

yyyyyy yyyyyy Washpipe Tubing string

After perforation the perfs are washed with a wash tool.

Tubing string

The zone is squeeze packed.

Liner or producing screen

Liner or producing screen

Circulating screen on set shoe

Circulating screen on set shoe

The liner is gravel packed.

The gravel pack is completed.

Fig. 2-13 Gravel-packing process.

Introduction to Well Testing (Aug 1996)

2-18

Section 3 Practical Well Testing

Section 3

Introduction to Well Testing (Aug 1996)

Practical Well Testing

3-2

Schlumberger

3.0

Practical Well Testing

3.1 Introduction Tests on oil and gas wells are performed at various stages of drilling, completion and production. The test objectives at each stage range from simple identification of produced fluids and determination of reservoir deliverability to the characterisation of complex reservoir features. Most well tests can be grouped either as productivity testing or as descriptive/reservoir testing.

• Productivity well tests are conducted to;

− Identify produced fluids and determine their respective volume ratios. − Measure reservoir pressure and temperature. − Obtain samples suitable for PVT analysis. − Determine well deliverability. − Evaluate completion efficiency. − Characterise well damage. − Evaluate workover or stimulation treatment.

• Descriptive tests seek to; − − − −

Evaluate reservoir parameters. Characterise reservoir heterogenities. Asses reservoir extent and geometry. Determine hydraulic communication between wells.

Whatever the objectives, well test data are essential for the analysis and improvement of reservoir performance and for reliable predictions. These, in turn are vital to optimising reservoir development and efficient management of the asset. Well testing technology is evolving rapidly.

Introduction to Well Testing (Aug 1996)

Integration with data from other reservoir related disciplines, constant evolution of interactive software for transient analysis, improvements in downhole sensors and better control of the downhole environment have all dramatically increased the importance and capabilities of well testing. 3.1.1 Productivity Well Testing Productivity well testing, the simplest form of testing, provides identification of productive fluids, the collection of representative samples and determination of reservoir deliverability. Formation fluid samples are used for PVT analysis, which reveals how hydrocarbon phases coexist at different pressures and temperatures. PVT analysis also provides fluid physical properties required for well test analysis and fluid flow simulation. Reservoir deliverability is a key concern for commercial exploitation. Estimating a reservoir’s productivity requires relating flow rates to drawdown pressures. This can be achieved by flowing the well at several flow rates (different choke sizes) and measuring the stabilised bottomhole pressure and temperature prior to changing the choke. The plot of flow data verses drawdown pressure is known as the inflow performance relationship (IPR). For monophasic oil conditions, the IPR is a straight line whose intersection with the vertical axis yields the static reservoir pressure. The inverse of the slope represents the productivity index of the well. The IPR is governed by properties of the rock-fluid system and near wellbore conditions. Examples of IPR curves for low and high productivity are shown in Figure 32. The steeper line corresponds to poor productivity, which could be caused either

3-3

Section 3

Practical Well Testing

(a) QT4 QT3 QT2

Wellhead flow rate QT1

(b)

P1 Bottomhole pressure P2

P3 P4 Time

Fig. 3-1 Relations between flow rates and drawdown pressures used for estimating reservoir productivity. A stepped production schedule during a productivity test (a) is achieved by flowing the well at several flow rates. Associated (stabilized) bottomhole pressure (b) is measured before changing the choke. 4200

Sandface pressure (psia)

3800

3400 C

3000 A

B

2600 0

20,000

40,000

60,000

80,000

Flow rate at surface conditions (B/D)

Fig. 3-2 Typical inflow performance curves showing low (a) and high (b) productivity. For gas wells, IPR curves exhibit certain curvature (C) due to extra inertial and turbulent flow effects in the vicinity of the wellbore and changes of gas properties with with pressure. Oil wells flowing below the bubblepoint also display similar curvature, but these are due to changes in relative permeability created by variations in saturation distributions. Introduction to Well Testing (Aug 1996)

3-4

Schlumberger

by poor formation flow properties (low mobility-thickness product) or by damage caused while drilling or completing the well (high skin factor)

Production changes, carried out during a transient well test, induce pressure disturbances in the wellbore and surrounding rock. These pressure disturbances extend into the formation and are effected in various ways by rock features. For example, a pressure disturbance will have difficulty entering a tight reservoir zone, but will pass unhindered through an area of high permeability. It may diminish or even vanish upon entering a gas cap. Therefore, a record of wellbore pressure response over time produces a curve whose shape is defined by the reservoir’s unique characteristics. Unlocking the information contained in pressure transient curves is the fundamental objective of well test interpretation. To achieve this objective, analysts display pressure transient data in three different coordinate systems - log-log (for model recognition), semi-log (for parameter computation) and Cartesian (for model / parameter verification). Introduction to Well Testing (Aug 1996)

Pressure and pressure derivative (psi)

3.1.2 Descriptive Well Testing Estimation of the formation’s flow capacity, characterisation of wellbore damage and evaluation of a workover or stimulation treatment all require a transient test because a stabilised test is unable to provide unique values for mobility-thickness and skin. Transient tests are performed by introducing abrupt changes in surface production rates and recording the associated changes in bottomhole pressure. The pressure disturbance penetrates the reservoir much faster than in the near wellbore region, to such an extent that pressure transient tests have evolved into one of the most powerful reservoir characterisation tools. This form of testing is often called descriptive or reservoir testing.

Homogeneous reservoir

Double porosity reservoir

Impermeable boundary

Elapsed time (hr)

Fig. 3-3 Characteristic pressure transient plots showing the types of responses that might occur due to various reservoir characteristics.

Typical pressure responses that might be observed with different formation characteristics are shown in Figure 3-3. Each plot consists of two curves presented as log-log graphs. The top curve represents the pressure changes associated with an abrupt production rate perturbation, and the bottom curve (termed the derivative curve) indicates the rate of pressure change with respect to time (refer to section 5). Its sensitivity to transient features resulting from well and reservoir geometries (which are virtually too subtle to recognise in the pressure change response) makes the derivative curve the single 3-5

Section 3

Practical Well Testing

most effective interpretation tool. However, it is always viewed together with the pressure change curve to quantify skin effects that are not recognised in the derivative response alone.

in) on a producing well, a drawdown test is performed by putting a well into production. Other well tests, such as multi-rate, multiwell, isochronal and injection well fall-off are also possible.

Pressure transient curve analysis probably provides more information about reservoir characteristics than any other technique. Horizontal and vertical permeability, pressure, well damage, fracture length, storativity ratio and interporosity flow

Mathematical models are used to simulate the reservoir’s response to production rate changes. The observed and simulated reservoir response can then be compared during well test interpretation to verify the accuracy of the model. By altering model parameters such as permeability or the distance from the well to a fault, a good match can be reached between the real and modelled data. The model parameters are then regarded as a good representation of those of the actual reservoir. Today’s computer generated models provide much greater flexibility and improve the accuracy of the match between real and simulated data. It is now possible to compare an almost unlimited number of reservoir models with the observed data.

Pressure and pressure derivative (psi)

101

100 Before acid buildup

10–1 After acid buildup

10–2 10–3

10–2

10–1

100

101

102

Elapsed time (hr)

Fig. 3-4 Derivative curves showing features of outer boundary effects. The effects of damage removal are clearly seen in the after-treatment pressure response curve.

coefficient are just a few of the characteristics that can be determined. In addition pressure transient curves can indicate the reservoir’s extent and boundary geometry. The shape of the curve, however, is also affected by the reservoir’s production history. Each change in production rate generates a new pressure transient that passes into the reservoir and merges with previous pressure effects. The observed pressures at the wellbore will be a result of the superposition of all these pressure changes. Different types of well tests can be achieved by altering production rates. Whereas a buildup test is performed by closing a valve (shutIntroduction to Well Testing (Aug 1996)

3.2

Test Design

Design and implementation of a well testing program can no longer be conducted under standard or traditional rule-of-thumb guidelines. Increasingly sophisticated reservoir development and management practises, stringent safety requirements, environmental concerns and a greater need for cost efficiency require that the entire testing sequence, from program design to data evaluation, be conducted intelligently. Proper test design, correct handling of surface effluents, high performance gauges, flexible downhole tools and perforating systems, wellsite validation and comprehensive interpretation are keys to successful well testing. The importance of clearly defined objectives and careful planning cannot be overstated. Design of a well test includes development of 3-6

Schlumberger

a dynamic measurement sequence and selection of hardware that can acquire data at the wellsite in a cost effective manner. Test design is best accomplished in a software environment where interpreted openhole logs, production optimisation analysis, well perforation and completion design and reservoir test interpretation modules are all simultaneously available to the analyst.

Selecting the instrumentation and equipment for data acquisition is the final step of the test design process. Surface and downhole equipment should be versatile to allow for safe and flexible operations. Key factors to consider include;

The first step in test design involves dividing the reservoir into vertical zones using openhole logs and geological data. The types of well or reservoir data that should be collected during the test are then specified. The data to be collected drive the type of well test to be run. (See Figure 3-5).

• Using combined perforating and testing techniques to minimise rig time.

Once the type of test is determined, the sequence changes in surface flow rate that should occur during the test are calculated. The changes in flowrate and their duration should be realistic and practical so they generate the expected interpretation patterns in the test data. This is best achieved by selecting an appropriate reservoir model and simulating the entire test sequence in advance. Test sequence simulation allows the range of possible pressure and flow rate measurements to be explored. Simulation also helps isolate the types of sensors capable of measuring the expected ranges. Diagnostic plots of simulated data should be examined to determine when essential features will appear, such as the end of wellbore storage effects, the duration of infinite acting radial flow and the start of total system response in fissured systems. The plots can also help anticipate the emergence of external boundary effects, including sealed or partially sealed faults and constant pressure boundaries.

• Controlling the downhole environment to minimise wellbore storage.

• Running ultra high precision gauges when test objectives call for detailed reservoir description. • Choosing reliable downhole recorders to ensure that the expected data will be retrieved when pulling the tools out of hole. • Selecting surface equipment to safely handle expected rates and pressures. • Environmentally sound disposal of produced fluids. Whatever the choice, it is important to ensure that all data is acquired with the utmost precision. To do this a good understanding of the available hardware options is necessary along with its prospective impact, if any, on the data quality.

The next step is to generate sensitivity plots to determine the effects of reservoir parameters on the duration of different flow regimes.

Introduction to Well Testing (Aug 1996)

3-7

Section 3

Practical Well Testing

Fig. 3-5 Summary of Current Test Types. Introduction to Well Testing (Aug 1996)

3-8

Schlumberger

10,000

Pressure (psia)

8000

6000

4000 0

1

2

3

4

Elapsed time (hr)

Fig. 3-6 Simulated pressure response for a test sequence of flow periods followed by shut-in-periods.

Pressure and pressure derivative (psi)

106 Pressure Derivative 105

10 4 Wellbore storage

Limits

103 Radial flow 10

2

Double porosity behavior 101 10 – 4

10–2

100

10 2

10 4

Elapsed time (hr)

Fig. 3-7 Test design flow identification plot. 0 – 60

k = 50

–120

Pressure (psia)

–180 –240 k = 100 –300 –360 –420 –480

k = 500 Flow period no. 3 Skin = 0 Tzero = 12

–540 – 600 0

1

2

3

4

5

6

7

8

9

10

11

12

Elapsed time (hr)

Fig. 3-8 Sensitivity analysis plot for different permeability valves.

Introduction to Well Testing (Aug 1996)

3-9

Section 3 3.3

Practical Well Testing

Tubing Conveyed Perforating TCP

3.3.1 General Shaped charge perforating is a critical part of the well completion process and development of the technique has been driven by the need for better well productivity, operating efficiency, safety and lower costs. Two basic perforating techniques are available to the completion design engineer;

• Through Tubing Perforating - The

guns are lowered into the well through the production string (or drill pipe / test string). The guns may be conveyed with wireline or coiled tubing.

• Casing Gun and High Shot Density

Perforating (HSD) - Large diameter guns are lowered into a cased well before the production string is run (or in some cases as part of the bottom-hole assembly especially during drillstem testing and certain completion techniques). The guns may be conveyed with wireline or the tubing string. Modern techniques include conveying some gun systems on coiled tubing or even slickline.

3.3.1.1 Through Tubing Perforating Through tubing perforating guns offer the following features;

• The wellhead and completion string are in place and tested before the casing is perforated.

• The underbalanced differential from the reservoir into the wellbore provides perforation clean-up.

Introduction to Well Testing (Aug 1996)

Fig. 3-9 2 1/8-in. 60o phased Scallop gun.

Fig 3-10 2 1/8-in. Enerjet gun.

3-10

Schlumberger

• Perforations may be made as required

to high downhole temperatures for a shorter time than with TCP. This is an important consideration in high temperature wells.

• Operating times are low, providing

3.3.2 Benefits of TCP Tubing conveyed perforating has however many benefits and the flexibility of the modern systems makes it an attractive prospect especially for well testing operations. Some of these benefits include;

over the life of the well, with or without a rig on site. good job efficiency and use of rig time. Maximum perforated interval per run is limited by the surface set-up and is typically 30 ft (9 m).

3.3.1.2 Casing Gun and High Shot Density Perforating (HSD) Casing or HSD guns offer the following features;

• Gun size is limited only by the casing

internal diameter, allowing the highest performance deep penetrating or big hole charges to be used at optimal shot density and perforating pattern.

• When guns are conveyed on wireline

the overbalanced differential from the wellbore into the formation allows the use of longer guns than with through tubing perforating. Typically 60 ft (18 m) can be readily achieved. Only simple wellhead control equipment is needed.

• Compared to expendable through tub-

ing guns, carrier type guns significantly reduce the amount of perforating debris introduced into the wellbore during the perforating process.

The choice between wireline and tubing conveyed perforating should be made on the completion objectives and operational considerations. From an operational viewpoint, wireline perforating operations are usually faster when there are a few short intervals to perforate. TCP operations are more efficient for long, multi-zone perforation intervals. Because of the faster operating speeds of wireline perforating, explosives are exposed Introduction to Well Testing (Aug 1996)

• TCP combines the advantages of the through tubing gun systems with those of the HSD systems.

• Large guns may be fired in an underbal-

anced condition with the full well control equipment and production string or drill pipe in place.

• Long intervals may be efficiently perfo-

rated in one run with a kill string in place if required.

• The programmed underbalance is ap-

plied to all perforated intervals, evenly and in a controlled fashion.

• A variety of firing systems and accessories accommodates a wide range of well conditions and completion techniques.

• After firing, expended guns may be

dropped to the bottom of the well allowing future through tubing operations.

3-11

Section 3

Practical Well Testing 3.3.3 Equipment Selection Accurate assessment of the operating environment, completion objectives and testing requirements are necessary to determine the safest and most efficient TCP technique. Non-standard or unusual conditions such as high bottomhole temperatures and pressures, long exposure times, high well deviations, small restrictions or an H2S environment may call for special equipment and techniques. The reservoir and type of completion form a unique set of conditions that determines the selection of various options;

• Casing size and formation characteris-

tics usually dictate the gun size and charge type. • uch as “SPAN” analysis help in selecting the best gun / charge option.

• Selection of a suitable explosive package is based on the anticipated maximum exposure time of the gun string at or near bottomhole temperature.

• Well testing requirements affect the

choice of TCP firing heads and accessories and the size of perforating guns in some cases.

• Formation characteristics, together with

safety and economic considerations, determine the amount of underbalance and how it is established.

• Accurate knowledge of the internal diFig. 3-11 7-in., 14 spf, 140o/20o High Shot Density Gun.

Introduction to Well Testing (Aug 1996)

ameters of all string components and thus possible restrictions is essential to choosing the firing system and to planning for possible fishing tool, positioning tool or cutter runs.

3-12

Schlumberger

3.3.4 Testing Requirements Combining perforating and testing operations provides a powerful means to evaluate the effectiveness of the completion itself as well as to gain valuable information about the reservoir. The annulus pressure firing system is ideally suited for impulse and drillstem tests in which the guns are run below a downhole valve. The system operates at any deviation, is practically immune to debris and does not require a fullbore string. The firing system is usually a differential pressure firing system. With this firing system, differential pressure between the annulus above the packer and the rat-hole actuates the firing head. Therefore, guns cannot be fired before the packer is set. Annulus pressure from above the packer is ported to the firing head through the packer and slotted pipe via a packer conversion kit. The packer conversion kit is either a non-fullbore universal packer conversion kit for standard retrievable packers or a fullbore packer conversion kit which is also suitable for hydraulic set retrievable and permanent packers as used in permanent completions. As annulus pressure is increased via the surface pumps, the test valve opens. At this time a pressure difference is created across the packer with the cushion pressure below it and the annulus plus pump pressure above it. The firing head is set to fire at this pressure or a few hundred psi pressure higher to allow packer seal confirmation before the guns are fired. Opening the test valve while simultaneously firing the guns is a possible option for applications in which open perforations exist in the well.

Radioactive marker sub

Reversing valve

Testing valve

Annulus pressure transfer sub

Retrievable packer

Slotted tail pipe

Differential pressure firing head Extension housing Safety spacer HSD gun

HSD gun

Fig. 3-12 Testing / TCP system.

Introduction to Well Testing (Aug 1996)

3-13

Section 3 3.3.5 Firing Heads Apart from the differential pressure firing system discussed above, there are many other options available to fit a variety of completion and testing needs. Each system has unique features that provide benefits for specific applications, the main difference between firing heads is the method of actuation. 3.3.5.1 Differential Pressure Firing Head The differential pressure firing head (discussed above) is actuated by the differential pressure between the annulus above the packer and the rat-hole pressure below. The main features are;

• A safety spring disables the firing pin when hydrostatic pressure is below 600 psi (250 psi optional), making the gun string safe on and near the surface.

• The packer must be sealed and the test

Practical Well Testing • A minimum of 150 to 300 psi hydrostatic pressure is required to activate the firing pin, making the gun safe on and near the surface. • The adjustable firing delay makes the head suitable for operations with nitrogen. • Pressure equalisation prevents firing flooded guns. 3.3.5.3 Bar Hydrostatic Firing Head The bar hydrostatic firing head is a drop bar actuated device. Once actuated, hydrostatic pressure drives the firing pin into the detonator. The main features are;

• A minimum of 150 to 300 psi hydro-

static pressure is required to activate the firing pin, making the gun safe on and near the surface.

valve must be open before the guns can be fired.

• The device is uncomplicated in design

• The system is specifically designed to

• Pressure equalisation prevents firing

• Pressure equalisation prevents firing

3.3.5.4 Trigger Charge Firing System The trigger charge firing system adapts either the absolute pressure, drop bar or jar down firing mechanism to a transfer assembly that is run into the well on slickline (or electric line) after the string and guns have been run, tested and positioned. The main feature is;

be combined with test tools. flooded guns.

3.3.5.2 Hydraulic Delay Firing Head The hydraulic delay firing head, an absolute pressure firing head, is actuated by tubing pressure shearing calibrated pins when a preset pressure level is reached, initiating a time delay period during which underbalance pressure is established before the guns are fired. Once the delay has expired, pressure at the firing head drives the firing pin into the detonator. The main features are;

Introduction to Well Testing (Aug 1996)

and operation. flooded guns.

• Heads containing primary explosives are run into the well latched and then retrieved independently of the gun string.

3-14

Schlumberger

Note: Delay time is dependent upon orifice range and downhole temperature and pressure.

Air chamber Orifice Oil chamber

Piston moves upward.

Traveling piston

Pump pressure is applied.

Shear pins

Ball retainer moves upward.

Shear pins break.

Tubing/ rathole pressure

Ball retainer

Ball retainer uncovers balls; firing pin is released.

Locking balls Firing pin Detonator

Firing pin is propelled onto detonator.

1–Running in

2–Actuating firing head

3–Delay

4–Firing

Fig. 3-13 Hydraulic Delay Firing System.

Primary high explosives are very sensitive and easily detonated by shock, friction or heat. For safety reasons, primary high explosives are used only in electrical or percussion detonators in Schlumberger gun systems.

• The firing head is connected after the

Secondary high explosives are less sensitive and require a high energy shock wave to initiate detonation (usually provided by primary high explosives - detonators). Secondary high explosives are used in all other elements of the ballistic chain (detonating cord, boosters and shaped charges). PETN, RDX, HMX and HNS are secondary high explosives used in oilwell perforating.

• The choice of firing head may be made

Introduction to Well Testing (Aug 1996)

guns are on depth, which improves the level of safety for the entire operation.

• The firing head is disconnected before gun retrieval.

after the guns have been run.

• Drop bar, jar down and hydraulic delay

versions require a minimum of 150 300 psi hydrostatic pressure to be activated.

3-15

Section 3 3.3.5.5 Redundant Firing Systems Redundant firing systems are available that allow the primary firing heads to be combined with one another as required. Both firing heads are located at the top of the gun string, allowing the guns and heads to be made up safely, and both heads retain their full safety features. Redundant systems are an excellent contingency which can save valuable rig time in the unforeseen event of any problems. 3.3.6 Depth Control Depth control is vital to ensure that the formation is perforated in the correct place and corresponding to the analysed log data. Off depth perforating can result in poor well performance and have severe financial implications. There are four main techniques to verify that the guns are at the correct perforating depth;

• Run a through tubing GR/CCL (gamma-

ray / casing collar locator) log to locate a reference point in the string and tie into on depth logs.

• Set the packer with electric wireline at a

known depth using GR/CCL for correlation and sting the guns and completion string through the packer.

• Set the packer and guns with electric wireline at a known depth using GR/CCL for correlation and completion string through the packer.

• Tag a fixed and accurate point such as a bridge plug.

Introduction to Well Testing (Aug 1996)

Practical Well Testing Special techniques are used on floating rigs derived from GR/CCL correlation. A reference point in the string (radioactive marker sub) is tied to the openhole logs, taking into account the movement of various pieces of equipment in the string (Slip-joints, jar, packer) after the packer has been set and weight slacked off. 3.3.6.1 Procedure a) Run in hole with TCP/DST string and subsea hanger, land off in subsea blowout preventer (BOP) stack, and run GR/CCL correlation log. b) Locate radioactive marker at a depth corresponding to the top shot, minus the length of assembly from top shot to radioactive marker measured in tension (including D+J). c) Pull out of hole to subsea hanger, and add or remove tubing/drillpipe below hanger as required in step b. Run back in hole with subsea hanger and add subsea tree assembly. d) When landed off, the top shot is D+J below desired top shot location. e) Pull up D+J+P, rotate right to set packer and start to slack off weight; at this point the top shot is P above its desired final location. f) As the weight is further slacked off, the setting stroke P of the packer brings the top shot to the desired location. Confirmation with GR/CCL is possible after packer setting.

3-16

Schlumberger

Fig. 3-14 Through Tubing Correlation Log.

Introduction to Well Testing (Aug 1996)

3-17

Section 3

Practical Well Testing

a+b

c

d

e

f

D+J+P

Slip joints fully open

Slip joints half closed

T

Radioactive marker sub

Reference tool closed

Reference tool stroked open

Jar closed

Jar open Packer set

P

Interval

D+ J

Pull up D + J + P. Packer ready to be set.

Adjust space out below hanger and add tree. Run gamma ray correlation log with subsea hanger landed off.

Set packer, and slack off weight to have subsea tree landed off slip joints in midstroke and guns at shooting depth.

Land subsea tree.

Fig. 3-15 TCP Depth Control on Floating Rigs.

Introduction to Well Testing (Aug 1996)

3-18

Schlumberger

3.4

Drill Stem Testing Tools - DST

3.4.1 General A set of drill stem test tools is a complex array of downhole hardware used for the temporary completion of a well. They are run as a means of providing a safe and efficient method of controlling a formation during the gathering of essential reservoir data in the exploration, appraisal and even development phase of a well, or to perform essential preconditioning or treatment services prior to permanent completion of the well. Many components are similar in function to those of permanent completions although the temporary nature of the string require some additional functionalities normally not associated with permanent completions. This can be better understood by realising that DST tools are designed for a wide range of operating environments and multi-use as in they can be redressed between runs while permanent completion components are designed for specific installations and long life. 3.4.2 Basic Requirements Drillstem tests are affected by three different pressures;

• Hydrostatic Press

Ph

• Formation Pressure

Pf

• Cushion Pressure

Pc

During testing Ph must be isolated from Pf and Pc to allow a formation to flow to surface. Three primary tools are used;

Introduction to Well Testing (Aug 1996)

3.4.2.1 Packer This provides a seal and isolates Ph from Pf much the same as for permanent completions. 3.4.2.2 Test Valve A test valve, run above the packer, isolates Pc from Ph while running in the hole. It also helps reduce the effects of wellbore storage which is an important element of interpretation (see section 5). After the packer is set and the test valve opened, flow to surface occurs. 3.4.2.3 Reverse Circulation Valve A reverse circulation valve provides a means of removing produced fluids before pulling out of the hole. For redundancy, two reversing valves with different operating systems are normally run. In addition, reversing valves are used to spot cushion and acid treatments. Additional tools may also be run to enhance string efficiency, safety and versatility. Some of the more common components are; 3.4.2.4 Slip Joint A slip joint is an expansion/contraction compensation tool. It accommodates any changes in string length caused by temperature and pressure during the DST. The tool is hydraulically balanced and insensitive to applied tubing pressures. Slip joints have a stroke of 5 ft (or occasionally 2 ft), the total number of slip joints depends on well conditions e.g. for a standard test at 10,000 ft. two or three slip joints are normal.

3-19

Section 3 3.4.2.5 Hydraulic Jar A hydraulic jar provides the means of transmitting an upwards shock to the tool string in the event that the packer and lower assembly become stuck. The tool has a time regulated action as transferring rapid movement in a long string is not a simple means. An upward pull activates a regulated oil flow until the hammer section is released thus giving a rapid upward movement and generating the relevant shock. 3.4.2.6 Safety Joint A safety joint is actuated only if a jar cannot pull stuck tools loose. By manipulating the tool string (usually by a combination of reciprocation and rotation), the safety joint, which is basically two housings connected by a course thread, can be unscrewed, and the upper part of the string removed from the well. 3.4.2.7 Safety Valve A safety valve independent of the main test valve is often required to provide additional well control, especially if the main downhole valve fails. 3.4.2.8 Gauge Carrier When run with a test string, both mechanical and electronic gauges must be placed in a carrier for support and protection. Carriers can either be of the above or below packer type. 3.4.2.9 Sampling Chamber Tool A sampling chamber tool is used to trap a downhole sample anytime during the test without shutting in the well.

Practical Well Testing

Upper Slips (Retracted) Floating Piston

Bypass Seal (Closed) Gauge Ring

Rubber Elements

Antiextrusion Rings

Gauge Ring

Setting Mandrel

Fig. 3-16 Packer.

Ball Seal Ball Valve Ball Valve Operator

Bias Area Hydrostatic Ports Operator Mandrel

Nitrogen Reference Pressure

Compensating Piston Pressure Reference Ports Rupture Discs

Rupture Discs

Closed

Open

Fig. 3-17 Test Valve. Introduction to Well Testing (Aug 1996)

3-20

Schlumberger

Top Sub

Slots/Dog

Index Section

Piston Mandrel

Stroke 2 ft or 5 ft

Spline

Spring

Reverse Circulation Through Ports

Reverse Ports

Internal Pressure Chamber

Seal Mandrel

V-Packing Seals

Annulus Pressure Chamber Closed

Cycling (Mandrel Pumped Down)

Open

Fig. 3-18 Reverse Circulation Valve.

Fig. 3-19 Slip Joint.

Timeregulating nut Anvil Shock

Oil

Shock Absorber

Hammer Brass valve Rapid oil transfer

Gauge

Buffer Tube

Metering section

Jars Closed

Jars Tripped

Fig. 3-20 Hydraulic Jar.

Introduction to Well Testing (Aug 1996)

Fig. 3-21 Gauge Carrier. 3-21

Section 3

Practical Well Testing

Drain Valve

Floating Piston

Sample Chamber

Oil Chamber

Sample Mandrel Lock

Flow Restrictor

Atmospheric Chamber

Operator Mandrel Rupture Disc

Before Sampling

After Sampling

Fig. 3-22 Sampling Chamber Tool.

3.4.3 Types of Drill Stem Tests As with permanent completion components there are a variety of different drill stem test tools designed for a range of operating conditions and to perform different functionalities. There are however two main categories of drill stem tests;

• Open Hole Drill Stem Tests • Cased Hole Drill Stem Tests 3.4.3.1 Open Hole Drill Stem Testing If hydrocarbons are detected in either cores or cuttings during drilling or indicated by the logs, an open hole DST provides a rapid, economical means to quickly assess the production potential of the formation. However the technique requires the hole to be in very good condition and highly consolidated as the packer elements actually seal on the rock face. The open hole sections also limit the application of pressure on the annulus therefore special strings are designed which are operated by pipe reciprocation and/or rotaIntroduction to Well Testing (Aug 1996)

tion. The Multiflow Evaluator System MFE is a self contained open hole drill stem test string. If drilling is not halted to allow testing when a potential hydrocarbon bearing zone is encountered, an alternative test method is to wait until the well is drilled to total depth and then use straddle packers to isolate the zone of interest. The introduction of inflatable packers allows the effective isolation and testing of individual zones pinpointed using wireline logs. Open hole drill stem tests gather important early information, but reservoir testing requires more data over a longer period. The extent of reservoir investigated increases with test duration. A key factor governing the duration of an openhole test is wellbore stability. At some point the well may cave in on top of the packer and the string may get permanently stuck downhole, calling for an expensive sidetrack. These hazards of wellbore stability have been eliminated by testing after the casing has been set and in many sectors particularly offshore, cased hole testing has replaced traditional open hole drill stem testing. 3.4.3.2 Cased Hole Drill Stem Testing As offshore drilling increased, floating rigs became common, increasing the potential for vessel heave to accidentally cycle traditional weight set tools and even un-set the packer. In addition, deeper more deviated wells make reciprocal tools more difficult to operate and control and thus jeopardise the safety of the operation. A new pressure controlled system was designed specifically for these applications which eliminates the need for pipe manipulation after the packer has been set and has eventually become the new standard in drill stem test operations. 3-22

Schlumberger

Drillpipe or tubing

Drillpipe or Tubing

Drill collars

Drill Collars

Pump-out reverse tool

Pump-Out Reverse Tool

Drill collars (1 stand)

Drill Collars (1 Stand)

Break-off plug reverse tool

Break-Off Plug Reverse Tool

Drill collars (1 stand)

Drill Collars (1 Stand)

Bar catcher

Bar Catcher

Multiflow evaluator valve

Multi-Flow Evaluator (MFE)

Openhole bypass

Pressure Recorder (Inside Reading)

Pressure recorder (inside reading)

TR Hydraulic Jars

TR hydraulic jars

Rotary Pump Safety joint

Multi-Stage Relief Valve

Safety seal

Safety Joint Openhole packer (conventional or bobtail)

Upper Inflate Packer

Perforated anchor

Blank Spacer Pipe or Drill Collars Pressure recorder (outside reading) Drill collar(s)

Ported bullnose

Fig. 3-23 Typical MFE Open-Hole String.

Lower Inflate Packer

¢¢¢ @@@ €€€ ÀÀÀ QQQ @@@ €€€ ÀÀÀ QQQ ¢¢¢

Deflate Drag Spring Tool

Pressure Recorder (Inflate Pressure) Bullnose

Fig. 3-24 Typical MFE Inflate Open-Hole String.

Introduction to Well Testing (Aug 1996)

3-23

Section 3 The Pressure Controlled Test System - PCT is a self contained cased hole drill stem test string. The main test valve and other key tools are operated by specific signatures of annulus and/or tubing pressure which is why a thorough understanding of the different pressures and potential differentials is important in the design of the cased hole DST string. In the specific case of the PCT, the valve opens when pressure above a certain threshold usually 1500 psi - is applied on the annulus, and closes when this pressure is bled off. It uses the same annular pressure threshold regardless of depth, hydrostatic pressure and temperature (provided the design specifications of the tool are not exceeded). To do this, a chamber in the tool is pre-charged at the surface with nitrogen. A compensating piston ensures that the nitrogen acquires hydrostatic pressure as the tool is run in the hole. (See Figure 3-17) Most pressure controlled systems provided today are termed fullbore which means that a minimum internal diameter of 21/4" is maintained throughout the string from top to bottom, which is essential for running wireline tools or coiled tubing inside the string to access the producing zone and hence enhance the flexibility of the test program. Services such as through tubing perforating, wireline or slickline conveyed sampling, pressure/temperature and production logging can readily be programmed into the test sequence either as main parts of the program or contingency measures. The flexibility of this type of system allows it to be run with most specialised systems;

• Permanent production packers or cement retainers. • TCP systems. • Surface Pressure Read Out Systems.

Introduction to Well Testing (Aug 1996)

Practical Well Testing The system is specifically useful in horizontal well applications, and offers almost unlimited testing, treating and stimulation operations in this technically demanding arena. Drillpipe or tubing Slip joint (open)

Slip joint (half open) Slip joint (closed)

Drill collars Pump-out reverse tool

Drill collars (1 stand)

Break-off plug reverse tool Drill collars (1 stand)

Bar-catcher Overpressure Safety Valve and Sampler (OSVS) PCT Pressure-Controlled Tester

Hydrostatic Reference Tool (HRT)

Pressure recorder (inside reading)

TR hydraulic jars Safety joint

PosiTest cased hole packer

Perforated/slotted tailpipe or tubing

Pressure recorder (outside reading) Ported bullnose

Fig. 3-25 Typical Fullbore PCT String. 3-24

Schlumberger

3.4.4 New Technology Increased sophistication in testing demands additional tools, creating the need for a complex sequence of distinguishable pressure pulses. The annular pressure has to supply not only a discrete signal to one of a number of tools, but also the power to operate it. For example, opening a single shot reversing valve at the end of a test can typically require 2000 to 3000 psi above the hydrostatic pressure. This creates significantly higher pressures in the annulus and great care must be taken not to exceed the tubing collapse or casing burst pressures. There is thus a limit to the number of discrete annular pressure signals that can be safely employed to command and power downhole equipment. A recent development addresses this limitation by employing much lower annular pressure variations as command signals to the downhole tools. The signals are analysed by the tool’s controller which uses field proven, reliable electronics to control the downhole test valve and circulating valve. Batteries power the electronics, annular hydrostatic pressure supplies the energy to operate the valves. The system is called the “Intelligent Remote Implementation System - IRIS”. Low intensity coded pulses of at least 250 psi are sent down the annulus using the rig mud pumps. The key recognition factor for the IRIS system’s pressure sensor is the shape of the pressure pulse. A threshold pressure has to be achieved, sustained and bled off within specific time and pressure variation constraints. The duration that a plateau pressure is sustained distinguishes one command from another. In the tool, a microprocessor reads the coded pressure pulses, compares them to pre-set Introduction to Well Testing (Aug 1996)

operating instructions and opens or closes solenoid valves to direct hydraulic fluid from chambers at annular hydrostatic pressure into chambers at atmospheric pressure. This fluid movement is used to operate the tool’s valves, closing them with a high intensity force driven by the differential pressure rather than by just the force of a spring, as in conventional systems. In addition, the fact that clean hydraulic fluid rather than mud is operating the tools, reliability is also enhanced.

Circulating valve

Test valve

Atmospheric chamber

Hydrostatic chamber

Pilot and solenoid valves Pressure sensor Electronics

Battery

Fig. 3-26 Intelligent Remote Implementation System-IRIS.

3-25

Section 3

Applied annular pressure (psi)

1200

Practical Well Testing stallation package. The basic floating installation package consists of; • Subsea Test Tree • Retainer Valve • Lubricator Valve

TCP guns fired using annular pressure TCP guns fired using drop bar

1000 TCP guns fire 800 Drop bar TCP guns fire

Main flow period

600 Test valve closes

Test valve opens 400

200

0 Test valve opens, sequential mode enabled

Time

Disable sequential mode

Circulating valve opens

Fig. 3-27 Annular pressure pulses needed to control the IRIS dual valve in conjunction with either pressure or drop bar operated TCP guns.

Since the tool functions through electrohydraulics, its mechanical construction is simplified. The 20 ft. (6 m) IRIS dual-valve replaces conventional fullbore test strings measuring up to 40 ft. (12 m). Elimination of pressurised nitrogen chambers also enhances the safety aspects of the tool. The equipment is compatible with conventional pressure operated test equipment including TCP systems. 3.5

Subsurface Safety Systems

3.5.1 General When testing from semi-submersible or floating drilling vessels, the well is controlled by landing a subsea control valve tool (subsea test tree) in the blow-out preventers. This tool is designed to allow the rig to shut-in the well, quickly disconnect from the main landing string and safely move off location in the event of unforseen weather patterns causing excessive heave. The additional safety featues of these tools have been incorporated into fixed and jack-up type drilling rigs. 3.5.2 Subsea Test Package The subsea test package varies between floating and fixed drilling installations. The fixed installation package is merely a simplified version of the more complex floating inIntroduction to Well Testing (Aug 1996)

3.5.2.1 Subsea Test Tree The subsea test tree is designed to provide a seabed master valve to close the drill string and allow disconnection during testing from a floating vessel. Disconnection may be necessary due to rough weather, loss of anchors or failure of the dynamic positioning system. It consists of a combination of a dual valve assembly and a hydraulic operator and latch assembly incorporated in the landing string and tailored to space out in the BOP stack. The hydraulic operator is powered by a surface hydraulic power unit connected via a hydraulic hose bundle. Underneath the complete assembly a slickjoint is connected to a fluted hanger which lands in the wear bushing of the BOP stack and is spaced such that during testing the BOP pipe rams are closed around the slick joint thus sealing off the well annulus. If it becomes necessary to unlatch, the hydraulic latch assembly disconnects from the valve assembly leaving the well shut-in and under control. The valve assembly is designed such that the blind rams can be closed above it. The main features of the tree are;

• Valves are normally closed and hydrau-

lic pressure is required to keep them open. The valves close when pressure is released or lost.

• The hydraulic mechanism is contained

in the latch. There is no communication of hydraulic fluid to the valves and therefore no danger of contaminating the hydraulic fluid by mud or well fluids. The entire operating mechanism 3-26

Schlumberger

can be serviced by pulling the latch, leaving the valves and string landed in the BOP stack.

• A safety piston in the downhole hy-

draulic assembly and a four way valve in the hydraulic console provide dual safety against accidental unlatching of the hydraulic assembly.

• The valve assembly is designed such

that it is possible to close two pipe rams on the slick joint and still close the blind rams above the valves after disconnecting the latch. Re-latching takes place deep in the BOP stack, where good centralisation is maintained and not in the area of the riser or ball joint.

• Pressure applied through one of the hydraulic lines, may be used to assist in closing the ball valve and thus cut slickline, electric-line or coiled tubing.

• The trees are designed to allow the in-

jection of chemical inhibitor to prevent hydrate formation.

In deep water (depths greater than 1000 ft ) a hydraulic control pod can be added to the system to allow faster response times for well closure and unlatching (less than 20 seconds at 3000 ft.) 3.5.2.2 Retainer Valve The retainer valve isolates the well fluids under pressure in the pipe above the subsea test tree and prevents communication to the riser should it become necessary to disconnect. This is particularly important in deep water because it prevents pollution and eliminates the dumping of high pressure gases into the riser.

Introduction to Well Testing (Aug 1996)

Fig. 3-28 Subsea Test Tree.

3.5.2.3 Lubricator Valve During offshore testing where slickline or wireline operations are performed, it may be advantageous to use a valve that avoids having a lubricator above the flowhead and thus high up from the drill floor. The lubricator valve, usually located about 30 m below the flowhead, enables the upper part of the string to be used as a lubricator. It is a balanced valve as in it remains in the position in which it was last placed (open or closed). It holds pressure from above and below and contains an equalising device to remove any differential prior to opening. 3.5.3 New Developments Modern subsea test packages are becoming shorter and faster acting with more stringent pre-service testing requirements, especially for use in high profile applications. Hydraulic operating systems can be combined with electric systems ensuring the fast operating responses. This becomes even more important in high pressure high temperature 3-27

Section 3

Practical Well Testing

(HPHT) applications and with exploration drilling in much deeper water.

3.6

Surface Testing Equipment

3.6.1 General On the surface, the fluids produced during a test are normally handled using temporary equipment or a surface testing package. This equipment needs to be assembled and designed to safely and reliably fulfil a wide range of operations;

• Provide a means of quickly controlling the pressure and shutting in the well.

• Separate produced fluids into gas, oil

and water phases, allowing the constituents to be metered, and record key data such as temperature and pressure.

• Allow representative samples of the produced fluids to be taken.

• Dispose of produced fluids in an environmentally acceptable manner.

Fig. 3-29 Lubricator Valve.

The onset of horizontal wellheads (whereby vary large external diameter service tools are required to be run through the subsea tree) and testing high rate gas wells calls for very large bore subsea test trees. Traditional size has been 3" internal diameter. Special applications today are producing subsea test trees with internal diameters of 5″, 6″ and 65/8″.

Introduction to Well Testing (Aug 1996)

The traditional safety philosophy seeks to maintain a minimum of two independent pressure retaining barriers between the surface equipment and the formation. These may be located at three levels; downhole, subsurface and surface. Downhole barriers include the DST test valve itself or a special safety valve used only in emergencies. Subsurface barriers are not universally employed on fixed rigs or onshore. In some cases, particularly high pressure gas wells or in hostile environments, an additional means of shutting in the well is required. A typical offshore well testing set-up, including all standard equipment, is shown in Figure 3-30. Although the surge tank is recommended and compulsory in the presence of H2S, both surge and gauge tanks are shown in this set-up. The surge tank is used as a second stage separator, and the connection of 3-28

Schlumberger

both tanks is shown. The grey area shows the permanent piping, which is now being used frequently on drilling rigs. For safety reasons, compressed air to supply the burners must be totally independent from the rig air supply. 3.6.2 Equipment A brief description of the main components now follows, but it must be emphasised that each well test set-up requires detailed planning and design related to expected operating conditions and customer objectives. 3.6.2.1 Flowhead Surface shut-in is usually provided by a flow control head or flowhead, which functions as a temporary christmas tree. The flowhead comprises of four valves, the master valve, the swab valve, the flow valve and the kill valve. The master valve is isolated from the

Introduction to Well Testing (Aug 1996)

other three valves by a swivel which allows rotation of the landing string without disconnecting the kill lines. The master valve allows isolation of the surface equipment from the downhole string and as such connects directly to the top of the string. The swab valve allows introduction and retrieval of slickline, wireline or coiled tubing. The flowline valve is equipped with an automatic hydraulic actuator which is linked into the emergency shut down system, thus automatically closing the valve in the event of a pre-programmed event (such as too high pressure etc.). An actuated valve can also be installed on the kill valve if required. In certain applications an additional valve called a stand alone valve will be placed immediately downstream of the flowhead to provide another safety barrier (specifically in HPHT applications).

3-29

Section 3

Practical Well Testing

Gauge tank

Separator

Heater or steam exchanger

Surge tank or second stage

Water from separator Diesel supply for diesel-fired heater

Optional for gauge tank Steam Gas Transfer pump

Laboratory cabin

Oil manifold Fluids from well

Lowpressure gas Steam generator

Flowhead

Choke manifold

Coflexip or Chiksans

Air compressor

Piping usually provided by client or permanent on the rig Seadragon* or mud burner (3 or more heads)

Safety valve

Water pump

Boom

Boom

Seadragon or mud burner (3 or more heads) Air Water

Air Water Oil

Oil

Gas Gas manifold Propane bottles Fig. 3-30 Typical Offshore Testing Setup. Introduction to Well Testing (Aug 1996)

3-30

Schlumberger

choke do not affect pressures and flow rates upstream of the choke. 3.6.2.3 Heater / Steam Exchanger It is often required to raise the temperature of well effluents to prevent hydrate formation, to reduce viscosity and to break down emulsions, thereby making separation of oil and water easier. To do this some form of heater is placed downstream of the choke manifold. There are two main types, an indirect fired heater or a steam exchanger. Indirect fired heaters are fuelled by diesel or gas and for that reason are often not permitted by regulations in many operating environments, particularly offshore. The main functions of a heater are as follows;

• Hydrate Prevention

Fig. 3-31 Flowhead.

3.6.2.2 Choke Manifold After the flowhead comes the choke manifold, which controls the produced fluid by reducing the flowing pressure and providing a constant flow rate. A choke is simply a restriction to flow; the choke manifold consists of valves and fittings arranged so that flow can be directed in one of two directions (through one of two choke boxes). Each box accepts either a fixed or variable choke. On the fixed side of the choke manifold, calibrated choke beans are used. Each bean is a specific size usually in graduations of 1/64" and is screwed into the choke box. The configuration allows a flow rate and specific choke size to be specified at the end of a test. A variable choke fitted to the other side of the choke manifold enables the fixed choke to be changed without interrupting flow. It also allows prediction of flowrates and pressure drops across the choke manifold during cleanup. The aim is to impose critical flow across the choke. When this has been achieved, changes in pressure made downstream of the Introduction to Well Testing (Aug 1996)

Natural gases contain water vapour. Under certain choked flow conditions, sufficient expansion occurs to lower the temperature of the flow and cause hydrate formation; that is particles of water and some light hydrocarbons in the gas become solid. This is a serious problem; if these particles freeze in the surface equipment, valves and flowmeters become inoperative and chokes blocked. Natural gas hydrates appear as hard snow and consist of chemical compounds of hydrocarbon and water. They form at temperatures above the freezing point of water when certain hydrocarbons are dissolved in water and under some low temperature and high pressure conditions. Using a heater helps to maintain the temperatures above the point where hydrates form. • Viscosity Reduction High viscosity impairs the flow of an effluent through a pipe. This is not usually a problem in well testing, however, combined effects of changes in composition as the reservoir fluid is brought to surface ambient temperature, the viscosity occasionally becomes high and effects separation efficiency. Because vis3-31

Section 3 cosity is temperature dependent, a heater may be used to lower the viscosity and prevent the problem. • Emulsion Breakdown With the inevitable production of water from a reservoir, it is necessary to separate the oil from the water. Under certain conditions, the oil and water emulsify and will not separate unless chemicals are injected or the effluent temperature is raised by the use of a heater. 3.6.2.4 Test Separator To accurately measure flow rate and take samples, the produced fluid must be separated into oil, gas and water. Test separators are capable of handling all types of output; gas, gas condensate, light oil, heavy oil, foaming oil, water and spent stimulation fluids such as acid. The possibility that hydrogen sulphide (H2S) can be produced necessitates special equipment and enhanced safety precautions. Designed for such versatility they are not expected to achieve as perfect a separation as production station separators, however separation efficiency is essential for accurate metering of the separated phases and as much care as possible must be taken to pre-condition the fluid for ideal separation. Test separators are generally very compact to facilitate easy installation in limited space environments such as offshore installations. They are available in a variety of types to cope with different operating environments. The main elements of a test separator are;

• The vessel - including essential internal

components to separate the phases and ease the separation process. The dixon plates or coalescent plates prevent droplets larger than 15 microns from

Introduction to Well Testing (Aug 1996)

Practical Well Testing being carried into the outlet gas stream, and the mist extractor, which is the last obstacle the gas encounters before leaving the separator vessel, blocks fine liquid droplets still in the gas stream. The blocked droplets coalesce and fall back into the oil phase.

• Instrumentation for level regulation and

retention time setting - separators are such flexible pieces of equipment because the level of liquid can be adjusted between two values: plus or minus 6 in. of the centre line of the vessel. A pneumatic liquid level controller with a long vertical float provides a vast range of oil levels.

• Metering manifold - Oil metering is

done via positive displacement meters for low rates and a turbine meter for higher rates. Gas metering uses a calibrated orifice plate.

• Piping and valves. • Skid and protective frame. 3.6.2.5 Gauge / Surge Tank The separated liquids then pass into either a gauge tank, which vents to the atmosphere via a flame arrestor or when H2S is expected, a pressurised surge tank. In these, volume is also measured to help calibrate the flowmeters and because the pressure is further reduced, additional gas comes out of solution causing shrinkage which can also be measured solution causing shrinkage which can also be measured.

3-32

Schlumberger

Rupture disk

Safety valve

Effluent inlet

Gas outlet

Deflector plate

Mist extractor

Baffle

Dixon plates

Weir

Water outlet

Vortex breaker

Manhole

Oil outlet

Fig. 3-32 Horizontal Three-Phase Separator.

3.6.2.Pumps and Manifolds The gas is usually directed via a gas manifold to the gas flare where it is burned. The oil is then redistributed via transfer pump and oil manifold to be burned through the oil burner. 3.6.2.7 Burners As environmental constraints tighten, acceptable disposal of produced fluids presents an increasing challenge. Gas and oil are generally burned. Onshore this usually occurs in flare pits. Offshore, the primary concern is to avoid dropping oily or carboniferous residue into the sea. Within a typical burner, oil flows into a chamber where it is atomised by compressed air. The mixture is then ignited. Water sprayed into the flame creates high turbulence, improves the efficiency of the burning and prevents the formation of carbon black smoke. High efficiency burners have been introduced in the early 1990’s improving the efficiency of the burners by a factor of 10. Using modern technology the atomisers have been computer modelled such that atomisation is improved shearing the oil into extra fine droplets. The velocity of the jet leaving the nozzle creates turbulence in the Introduction to Well Testing (Aug 1996)

surrounding air which creates a pressure profile so that it sucks extra air, essential for combustion, into the stream. Other essential parameters needed for enhanced burners can be continuously monitored and adjusted during the course of the test. While this method can be used with available rig equipment, additional air compressors will be required and for best results additional equipment is needed. 3.6.2.8 Emergency Shut Down Systems Emergency shut down systems are used to control the process chain in the event of an unforeseen hazard. High/low pilots initiate well closure when the pressure either rises above a high level threshold (choke blockage) or falls below a specified level (flowline leak / rupture). These pilots are installed at the upstream choke manifold, upstream heater or upstream separator. They are pneumatically operated and if triggered actuate a hydraulic actuator which will close down the wing valve on the flowhead or an additional stand alone valve upstream of the choke manifold.

3-33

Section 3

Practical Well Testing

P2

Gas metering valves (GMV)

PGV2

Air supply

Relief valve 1

PCV2 PCV7

Gas outlet valves (GOV) T3 4 3

2 4

3

1

T2 PGV3 P3

5

6

2

Farris safety valve Nonreturn swing valve

GOV7 P1 T1

GOV6 5 3

1

4

T5

PCV1

PCV3

PCV4

PGV1

5 V3

1

6

2

Shrinkage tester valves (SHV)

7

3 5

7

V1

6 2

4

WOV1 Water outlet valves (WOV)

8

Liquid levels valves (LLV)

PCV5

3

WOV2 PCV6 V4

2

2 4 1

Oil metering valves (OMV)

P1: P2: P3: T1: T2: T3: T4: V1: V2: V3: V4:

V2

Separator pressure gauge Pressure gauge Pressure gauge Thermowell Thermowell Thermowell Thermowell Separator inlet valve Bleedoff valve Separator bypass valve Separator bypass valve

1 3 T4 5 Oil outlet valves (OOV)

GMV: Gas metering valves GOV: Gas outlet valves LLV: Liquid levels valves OMV: Oil metering valves OOV: Oil outlet valves PCV: Pilot circuit valves PGV: Pressure gauge valves SHV: Shrinkage tester valves WOV: Water outlet valves

Fig. 3-33a Separator Flow Sheet.

Introduction to Well Testing (Aug 1996)

3-34

Schlumberger

Fig. 3-33b Separator Flow Sheet (Index).

Modern systems are electrically operated and computer controlled. This opens up a whole new range of safety options. For example, in addition to the high/low pressure pilots, the system can handle high/low levels in any of the relevant vessels, high/low temperatures, high /low flowrates, activate closure of the

Introduction to Well Testing (Aug 1996)

subsea test tree etc. The electrical activation significantly speeds up the reaction time and the computer input makes pre-programming of settings a formality. These systems are specifically important for HPHT and hostile environment operations.

3-35

Section 3

3.7

Practical Well Testing

Data Acquisition

3.7.1 General During well testing, there are two main places where data is acquired; at the surface and downhole. In the early days of well testing, only mechanical instruments were available to perform the measurements required. In many places this is still often the case especially for surface measurements. Today, however, sophisticated electronic circuitry with high performance strain, capacitance or quartz transducers are available coupled with automatic computer acquisition, control and display. Whatever the instrumentation chosen for measuring the various parameters, the end user requires accurate, valid data to ensure that the best interpretation of the data can be made and used in the future development of his asset. The old adage “rubbish in , rubbish out” is certainly relevant to data acquisition. Before looking at the different components and techniques it is important to emphasise two main factors;

• The severe difference in environment

between a pressure/temperature transducer required for surface measurement and that of one required for downhole measurement. Whereas this may seem obvious, the effects of pressure and temperature on the associated electronics can severely affect the measurement and this needs to be understood and accounted for in the design of the instrument.

• The performance of pressure / temperature transducers is paramount to producing accurate and reliable data. There are many factors which affect a gauge’s

Introduction to Well Testing (Aug 1996)

performance and these are important to understand when selecting the correct transducer for the task required. 3.7.2 Transducer Performance Manufacturers of pressure sensors most commonly depict their products and their performance through technical specifications. It is important to understand these specifications and what they apply to so that a thorough understanding of the product is realised. Typical pressure measurement parameters can be split into the following two main classes;

• Static Parameters. • Dynamic Parameters. 3.7.2.1 Static Parameters These parameters describe the transducer performance in static conditions. The main factors influencing performance are;

• • • •

Accuracy Resolution Stability Sensitivity

♦ Accuracy This is considered to be the algebraic sum of all the errors influencing the pressure measurement. These errors are due to; Mean Quadratic deviation (MQD) is a measure for the quality of the mathematical fit of the sensor response at one constant temperature. This parameter is a function of the transducer linearity (i.e. the closeness of a calibration curve to a specified straight line) 3-36

Schlumberger

and of the calibration procedure (i.e. coefficient grid and polynomial function used).

Fig. 3-36 Repeatability.

Fig. 3-34 M.Q.D.

Hysteresis is the maximum discrepancy of the output of the transducer signal between increasing and decreasing pressure (or temperature) excursions.

♦ Resolution This is the minimum pressure change that is detected by the sensor. When referring to a gauge resolution, it is important to take into account the associated electronics as the two are always used together. It is also important to measure the resolution with respect to a specific sampling time. The gauge resolution is equal to the sum of three factors;

• The sensor Resolution. • The digitizer resolution. • The electronic noise induced by the amplification chain.

Fig. 3-35 Hysteresis.

Repeatability is defined as the discrepancy between two consecutive measurements of a given pressure. As shown below, the repeatability is not affected by the hysteresis because both measurements are performed using the same procedure.Temperature Sensitivity or dP/dT is the ratio of the temperature sensitivity (of the pressure signal) to the pressure sensitivity. Example - a gauge with a dP/dT of 5 psi/°C and an associated temperature sensor of 0.1°C accuracy will induce the following error:

In the case of tools equipped with strain gauge transducers this last factor is by an order of magnitude the predominant parameter. In addition, mechanically induced noise may also be a factor that limits a gauge resolution. This is because some gauges behave as microphones or accelerometers. This may be an important consideration during tests when there is fluid movement or tool movement downhole.

5 psi/°C x 0.1°C = 0.5 psi.

Introduction to Well Testing (Aug 1996)

3-37

Section 3

Practical Well Testing 3.7.2.2 Dynamic Parameters These parameter describe the transducer performances in dynamic conditions. Under this classification we find; • Transient response during temperature variation. Fig. 3-37 Resolution.

♦ Stability This is the ability of a sensor to retain its performance characteristics for a relatively long period of time. The stability gives the sensor mean drift in psi per day obtained at a given pressure and temperature. Three levels of stability can be defined;

• Short term stability for the first day of a test.

• Medium term stability for the following six days.

• Long term stability for a minimum of one month.

♦ Sensitivity This is the ratio of the transducer output variation induced by a change of pressure to this change of pressure. In other words, the sensitivity represents the slope of the transducer output verses the pressure curve.

The sensor response is monitored under dynamic temperature conditions whilst the applied pressure is kept constant. The peak error represents the maximum discrepancy between the applied pressure and the stabilised sensor output. By general consensus, the stabilisation time represents the time needed to be within 1 psi of the stabilised pressure. The offset represents the difference between the initial and final pressure. This parameter provides for a given temperature variation, the time required to get a reliable pressure measurement. • Transient response during pressure variation. The sensor response is recorded before and after a pressure variation whilst the temperature is kept constant. Peak error and stabilisation time are measured as previously described for a temperature variation.

Fig. 3-39 Transient response during temperature variation. Fig. 3-38 Sensitivity.

Introduction to Well Testing (Aug 1996)

3-38

Schlumberger

3.7.2.3 Calibration The calibration of a pressure transducer provides a set of coefficients which allow the mathematical description of the transducer function between the pressure input and the sensor output at static temperature. The sensor is subjected over its full operating range to a series of extremely precise known pressures (generated by a highly accurate precision Dead Weight Tester) under stabilised temperatures. By calibration it is generally understood as a static calibration as under dynamic conditions few sensors have their pressure responses modelled. A recent technology breakthrough using quartz technology has produced a sensor whereby a mathematical representation of the sensors response under dynamic conditions has been produced thus allowing a dynamic correction to be applied and thus enhance performance - the sensor is called the CQG™ - Crystal Quartz Gauge. The calibration of gauges is extremely important to ensure accurate, reliable data.

Fig. 3-40 Transient response during pressure variation.

There are generally two levels of calibration performed. The master calibration which is performed in tailor designed laboratories whereby the gauge is scrutinised over its entire operating range. A master calibration can take several days to perform. Pre and post job calibration checks are performed locally Introduction to Well Testing (Aug 1996)

to verify the response of a gauge over a specific range as anticipated in the forthcoming test. Both these checks are essential procedures in the process of pressure and temperature data acquisition. 3.7.3 Pressure Transducer Technology Modern pressure transducers operate on the same basic principle of converting a pressure into a mechanical displacement or deformation. The mechanical displacement or the deformation of the sensing element is then converted into an electrical signal which can be processed by the measuring system.. There are three basic types of modern transducers as well as a combination of the different types;

• Strain Gauge • Capacitance • Quartz Crystal 3.7.3.1 Strain Gauge Sensors The strain gauge consists of a strain sensitive resistor directly attached to the measuring sensor. When the sensor is subjected to a force (pressure) it undergoes a displacement which in turn changes the resistor’s geometry (length) and therefore imparts a change in resistance. A mathematical relationship between this change in resistance and the applied force at a given temperature is the key to strain gauge technology. There are many types of strain gauge sensors; The Paine sensor is a bonded wire transducer; two sets of strain wires are wrapped around a tube sensing member (active legs). As pressure increases, the tube bore is stretched causing a change in the wire resistance. Two sets of strain wires are wrapped on the upper part of the tube which is not exposed to the pressure (passive or reference legs). The four 3-39

Section 3

Practical Well Testing

sets of strain wires form a Wheatstone bridge as shown in Figure 3-41(a). BONDED WIRE SENSOR SCHEMATIC R1 R2 A

R3

P

-

A OUTPUT

POWER

P

R4

+

3.7.3.2 Capacitance Sensors Capacitance sensors consist of a variable gap capacitor sensing element whereby the sensing element is formed by two metallic or quartz plates. As the external pressure increases the deflection in the sensing plate creates a change in capacitance which can be mathematically related to the applied pressure. THIN FILM TRANSDUCER SCHEMATIC

Fig. 3-41a Bonded Wire Sensor Schematic.

Resistors A and P are respectively the active and passive legs of the Wheatstone bridge. R1 and R2 are the balance adjustment and compensation resistance’s of the bridge. R4 is the thermal compensation resistor and R3 the sensitivity adjustment resistor. The thin film sensor consists of a resistor pattern which is vapour or sputter deposited onto a force summing element. An ingenious development of this technology is the Sapphire strain gauge whereby the stable deformation properties of a sapphire sensing element is used to transfer the force to the sensing resistors. BONDED WIRE TRANSDUCER Tube sensing member

Balance,adjustement and thermal compensation resistors

PRESSURE

Active strain wires

Reference strain wires

Fig. 3-41b Bonded Wire Transducer.

Strain gauges are in general rugged, low cost and have good dynamic behaviour. they are however susceptible to drift and have a limited resolution. Introduction to Well Testing (Aug 1996)

Compensation components Sputtered resistors

Pressure

Measuring Chamber

Measuring Diaphragm

Fig. 3-42a Thin Film Transducer Schematic (Pressure Gauges Review).

Capacitance gauges have good stability up to temperatures of around 150°C, they respond poorly to temperature transients and are sensitive to acceleration, orientation and mechanical noise. 3.7.3.3 Quartz Crystal Sensors Quartz crystal pressure sensors are built around a vibrating quartz crystal sensing element. The crystal is forced by external electronic circuits to oscillate at its resonate frequency. Pressure induced stress is applied to the quartz crystal resonator causing its frequency to vary in a precise manner. Quartz has been chosen as a vibrating crystal because of its excellent elastic properties, long term stability and ease of vibrational excitement. The way the quartz is cut (i.e. the orientation of the crystal faces) determines its resonant frequency and its sensitivity to pressure and temperature. Crystal gauges have historically been built around variations of the HP or Hewlett 3-40

Schlumberger

Packard quartz sensor, recent technology is almost exclusively using the Quartzdyne quartz sensor however a recent development by Schlumberger has produced the CQG STRAIN GAUGE ON SAPPHIRE* SENSING ELEMENT Sputtered strain gauges RTD Strain gauge Vacuum Four arms precision active bridge

Neutral fluid

Sapphire capsule

Elastic membrane Pressure

Bonding

Electrical feedthrough

* Schlumberger (ALPHA)

Fig. 3-42b Strain gauge on Sapphire (Pressure Gauges Review).

quartz sensor. It is based on a single quartz structure in which a resonator coupled with a dual mode oscillator generates two simultaneous bulk acoustic waves. The frequency of the first wave is mainly pressure dependent while the frequency of the second wave is almost totally temperature dependent.

sor design leads to the following two main improvements over previous crystal transducer technology; • From a static point of view, the main advantage is to have only one sensor giving both the pressure and the temperature therefore suppressing all the problems linked to the non-uniform ageing of the pressure and temperature transducers. (Normally separate sensors.)

• From a dynamic point of view, the

major advantage is the very small peak error induced by transient conditions. Furthermore these errors can be almost completely suppressed by using a dynamic compensation algorithm. The mathematics of the compensation algorithm are based on the equations of the resonator and are very efficient due to the simplicity of the thermal/time model of the CQG.

Quartz gauges are the most accurate on the market, they have excellent resolution and good long term stability. They are however very expensive. Fig. 3-44 below shows a comparison of the CQG, HP and strain gauge when subjected to a thermal shock.

Fig. 3-43 CQG™ Dual Quartz Gauge.

3.7.4 Surface Data Acquisition Modern systems use a series of strain gauge transducers for measuring the pressures, temperatures and differential pressures located at strategic places on the essential surface equipment. The key measurements required at the surface are as follows:

The CQG design brings a major breakthrough by the fact that pressure and temperature measurements are done at the very same location in the quartz material. The senIntroduction to Well Testing (Aug 1996)

3-41

Section 3

Practical Well Testing

• Flowhead:

• Tanks:

Pressure, temperature of tubing and casing.

• Choke Manifold:

Pressure, temperature upstream downstream of the choke.

Temperature, shrinkage.

• Burners:

High Efficiency burners require the pressure, temperature and flow of compressed air, oil and water to be monitored for correct burning.

and

• Heater:

Pressure, temperature upstream and downstream of the choke. Steam pressure and temperature.

The sensors are specifically designed for use on the surface and are linked into a computerised acquisition system. New sensors are constantly being brought into the market as technology progresses, most are 4-20 ma type sensors compatible with most major service companies acquisition systems. Electronic sensors are now available for the majority of the measurements required.

• Separator:

Pressure, temperature, differential pressure across the gas orifice meter (required for gas flow rate), flow rate of oil, gas and water, oil shrinkage, basic sediment and water- BS&W, oil and gas gravity, H2S, CO2, and other gas contents.

THERMAL SHOCK RESPONSE 8010

Thermal shock from 120ºC to 130ºC at 8000 psi. 8005

Peak error Strain gauge 4 psi

Strain gauge + 1 psi 8000

Pressure signal in psi

- 1 psi

CQG

Peak error CQG 2.5 psi

HP

7995

7990

Peak error HP 15 psi

7985 Stabilization time S.G 17 minutes

7980 0

4

8

12

Stabilization time CQG 18 minutes

16

20

24

28

32

Stabilization time HP 39 minutes

36

40

44

48

52

56

60

64

68

Time in minutes

Fig. 3-44 Thermal Shock Response Comparison.

Introduction to Well Testing (Aug 1996)

3-42

Schlumberger

3.7.5 Downhole Data Acquisition Downhole data gathering is performed using pressure/temperature sensors attached to either battery powered recorder modules or surface readout modules. The complete unit is often referred to as a pressure gauge. 3.7.5.1 Downhole Recording Downhole recording is when the complete gauge (sensor, recorder plus battery) are installed in a gauge carrier in the DST string or hung off in a nipple profile and used to record, process and store in its memory the pressure and temperature data during a well test. After the test is completed, the string is pulled out of hole and the data retrieved from the gauges. 3.7.5.2 Surface Readout Surface Readout usually involves the same type of sensors as downhole recorders except that the data is retrieved in real time usually via an electric line which is attached to the gauge. Whatever the technique it is clear that the gauges are subjected to extreme conditions of pressure and temperature and also have to survive shocks as associated with TCP guns firing or the turbulence of high flowrates. This is important to understand, as one of the key objectives of well testing is bottomhole pressure and temperature data needed for essential interpretation and the last thing a client wants is to find out that his pressure gauge failed to work properly or even at all. Reliability and ruggedness are therefore essential features of any modern downhole pressure gauge.

of the test objective, ranging from high accuracy, high resolution CQG/Quartzdyne gauges to rugged high temperature Sapphire/Strain gauges. The gauges are normally available in recording or real time readout mode. In recording mode the gauges are pre-programmed to record at various sampling rates over periods corresponding to the type of test and the amount of data required from the test objectives. There is clearly a limit to the amount of data they can store and the duration of the battery life although most gauges today have sufficient capacity for standard tests.

Battery Section

Electric Line Section

Microcontroller UNIGAGE Recorder Section

EEPROM Data Memory

ASIC

Sensor Sub Section

CQG

Customized Sapphire Quartzdyne

Fig. 3-45 Downhole Recorder and Surface Readout.

Modern downhole recorders incorporate field proven ruggedised electronics suitable for different ranges of pressure and temperature. They will also have a selection of pressure sensors available to accommodate the needs Introduction to Well Testing (Aug 1996)

3-43

Section 3 3.7.6 New Technology Two of the main advantages of surface readout during well testing is that it confirms functionality of the downhole equipment and also it confirms that sufficient data has been acquired for the relevant interpretation. One of the major disadvantages is the presence of cable in the well during critical phases of the operation (i.e. during flowing periods). Service companies have developed various techniques over the years to overcome these problems but usually some part of the test string’s functionality is lost or they involved complex lengthy procedures Two recent technology developments have overcome these hazards. 3.7.6.1 DataLatch System This system combines the advantages of a fullbore pressure and temperature recorder system with optional surface read-out capabilities. Throughout the test, the system records pressure and temperature above and below the downhole valve and in the annulus. But at any time, wireline may be used to interrogate the recorder memory and reprogram it. The system allows surface read-out during critical phases of the operation i.e. during build-up and then the wireline is once again removed prior to flow. Furthermore, the wireline can be run at a relatively convenient time during build-up and used to determine that the design for the rest of the well test is appropriate and to check that acquired data meet the test objectives. At the heart of the system is an innovative connection that links the wireline and the recorder. The Latched Inductive Coupling tool uses electromagnetic induction principle for the two way passage of electronic information and therefore eliminates the need for direct electrical contacts adding to the reliability of the system. The tool can also be powered directly from the surface in the Introduction to Well Testing (Aug 1996)

Practical Well Testing event of low battery levels and can take up to 4 CQG™, Quartzdyne or Sapphire™/Strain recorders.

Power LINC running tool

LINC Inductive coupling

DLWA

DLWA electronic board Independent gauge batteries (up to 4 gauges can be run)

DGA

Up to 4 independent gauges

Selective porting

Test valve

Fig. 3-46 DataLatch system with its two main components: The LINC Latched Inductive Coupling section on top and a fullbore MSRT MultiSensor Recorder / Transmitter below.

3.7.6.2 Wireless Telemetry This new technique permits downhole data to be transmitted electronically without wireline in the hole. The technique uses low frequency electromagnetic transmission through the formation for both open and cased hole. Two antennae, one at the surface and the other downhole are used as transmitting and receiving devices. The downhole antennae is located in the DST string, using an insulating gap to create the dipole. The technique relies on the resistivity of the formation as the higher the resistivity the greater the dissipation of the signal. The technique only works in certain operating environments.

3-44

Schlumberger

3.7.6.3 Data Acquisition in Permanent Completions In permanent completions, apart from routine pressure/temperature surveys or production logging another technique is to have a pressure/temperature sensor permanently installed as part of the completion string. The gauge is mounted on the outside of the tubing in a special gauge holding mandrel and connected to a cable which is strapped to the tubing all the way to surface. Ingenious termination technology is used to feed the cable through the wellhead (specifically for subsea applications) where they are linked in to surface signal processing systems. Long life and stability are clearly the key elements of such installations. 3.7.7 Formation Interval Testing An alternative method of gaining a rapid evaluation of a formation’s deliverability is via the use of formation interval testing tools. These tools are normally run as part of the open hole logging suite. They are conveyed on large heptacables and consist of precision engineered electro-hydraulic tools capable of performing a variety of tasks. The tools were initially designed as a means of obtaining a reservoir fluid sample at an early stage of the drilling process. Later tools however developed into high accuracy pressure/temperature monitoring tools designed to perform “mini-tests” at precise locations in an openhole section and thus gain an understanding of the reservoir. The tool is capable of making an unlimited number of “mini-tests” and catching up to two samples of formation fluid. The tool is accurately positioned relative to the openhole logs using a gamma-ray correlation log. A powerful hydraulic section forces a packer against the rock wall face and drives a probe into the sealed off area. Two pre-set chambers are then opened via pistons which helps Introduction to Well Testing (Aug 1996)

suck the formation fluid into the tool creating a two tiered drawdown. The well is then allowed to build-up pressure completing the “mini-test”. At the end of the test, the two pistons force the extracted fluid back into the wellbore and the tool is moved to the next point. Naturally the limitation on the amount of fluid withdrawn means that the depth of investigation is limited and the technique can not replace true well testing but is an excellent tool taking samples, localized reservoir evaluation, and also confirming the need for a test. A recent advancement in this technology has produced the Modular Dynamic Tester (MDT), this tool has significantly more capabilities than the old Repeat Formation Tester’s and incorporates all the latest high technology pressure sensors (CQG). The tool is totally modular in design and includes a number of different modules which allow the tool to be configured for a variety of operations. The key modules are; • Single probe Module Similar to the Repeat Formation Tester but the tool has a variable pre-set chamber which is programmable during the “mini-test” to allow the operator to adjust it for tight formations. It also has a resistivity cell which measures the flowline fluid resistivity. • Sample Chambers The tool allows the use of up to 12 samples to be taken if required. There is also a multisample module which contains six 450 cc sample chambers designed for PVT quality sampling.

• Multi-Probe Module

Contains two probes mounted back to back. This coupled with the single probe module allows vertical permeability analysis or anisotropic permeability determination. 3-45

Section 3

Practical Well Testing

• Flow Control Module Allows larger volumes of fluid to be withdrawn from the formation for enhanced permeability determination. • Pump-Out Module

Allows unwanted fluid to be removed from the near wellbore area of the formation to help capture virgin formation fluid during sampling.

• Packer Module

Employs two inflatable packers used for zonal isolation. Also increases the cross sectional area of the interval to be tested (as compared to a probe) thus allowing much greater depths of investigation during testing.

• Optical Fluid Analysis Module

Uses infrared spectroscopy to analyse the flowline fluid to differentiate between oil and water. Reflective/refractive techniques also allow gas and liquids to be differentiated.

Electrical power module

Electrical power module

Hydraulic power module

Probe module Sample modules

Probe module

Hydraulic power module

Sample module

Probe module

Sample module

Fig. 3-47 MDT™.

Introduction to Well Testing (Aug 1996)

3.8

Special Applications

As has been emphasised throughout this section, well testing is not a straight forward operation and there are many variations which can effect the equipment selection in order to perform the test in a safe, controlled and efficient manner while keeping the test objectives firmly in mind. Two special applications warrant some discussion; • High Pressure High Temperature HPHT • High Flowrate 3.8.1

High Pressure High Temperature - HPHT Testing wells with wellhead pressure above 10,000 psi (in fact above 8500 psi would probably fall into the category of HPHT) requires special equipment, precautions and procedures. It is certainly not a domain for the novice. The very nature of these tests require thorough pre-planning and usually involve HAZOP or Hazard and Operability studies way in advance of the actual test. A HAZOP study is essentially a process where each phase of the test sequence is broken down into segments and a safety analysis study is performed to cover all eventualities and contingencies required to handle them (a type of risk analysis). For example, if a safety analysis study was being performed on the flowhead, a possible event could be “surface temperature rises above safe acceptable limit of operation” the contingency would be to set a temperature sensor linked to the electrical shut down system which would close the well in. By analysing the sequence of potential events in this manner, the idea is to capture all potential hazards ahead of time and set in place procedures to eliminate them. Equipment needs to be adequately rated to handle the extreme conditions (TCP, 3-46

Schlumberger

DST, Subsea, Surface and Gauges), and special seals and flanges are used, designed for such conditions. In addition extra shut down and safety devices are required to give total protection to personnel and the surface equipment. It goes without saying that this process takes time and a great deal of experience to perform competently.

Introduction to Well Testing (Aug 1996)

3.8.2 High Flow Rates In a similar manner to HPHT, high flow rate testing while not necessarily having the same levels of pressure and temperature create problems in that the equipment needs to be adequately sized to handle large gas or oil flow rates. Again these type of tests require adequate pre-planning and involve new developments such as large bore DST strings, large bore subsea test trees, large internal diameter flowlines, high capacity separators or several separators linked in parallel and high capacity burners to dispose of the produced hydrocarbons.

3-47

Section 3

Introduction to Well Testing (Aug 1996)

Practical Well Testing

3-48

Section 4 Sampling of Reservoir Fluids

Section 4

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids

4-2

Schlumberger

4.0

Sampling of Reservoir Fluids

The purpose of sampling is to obtain a representative sample of reservoir fluid identical to the initial reservoir fluid. This condition is absolutely essential because reservoir engineering studies, which are performed using PVT analysis data, are always made on the basis of the reservoir at its initial conditions. For this reason, sampling operations should ideally be conducted on virgin reservoirs (having not yet produced) or in new wells completed in undepleted zones, containing fluids identical to the initial reservoir fluids.

fluid is altered by the selective loss of light or heavy hydrocarbons. While the liquids in a gas condensate reservoir may never reach a saturation where they can flow, the gas saturation in an oil reservoir will almost certainly reach the point where gas flow occurs. Because of the relatively low viscosity of gas, this flow of gas will increase rapidly, exhibiting the typical performance trend of a solution gas drive reservoir.

Sampling Procedures Design

Even if these phenomena are not reservoirwide, the pressure drawdown associated with flow will often be sufficient to drop the pressure of the fluid in the immediate vicinity of the wellbore below its bubble point or dew point pressure and into the two-phase region, as illustrated in Figure 4-1. A sample of such fluid will not be representative of the original fluid existing farther out in the reservoir. Steps must be taken to determine the reservoir pressure, temperature, and the general category of the reservoir fluid. If the relationship between reservoir pressure and bubble point or dew point pressure can be estimated, steps can be taken to ensure that the sampled fluid is representative.

4.1.1 Samples Representivity In designing a sampling procedure, we must consider the effect producing conditions will have had on the reservoir fluids we are sampling. When the pressure in an oil reservoir drops below the bubble point, gas comes out of solution and forms a separate phase. Similarly, when the pressure in a gas condensate reservoir drops below the dew point, liquid begins to condense in the reservoir. In either case, the minor phase must build up to a certain critical saturation within the reservoir rock before it will begin to flow. In the meantime, the composition of the produced

Another concern in obtaining a representative sample is the degree of variation in the original reservoir fluid throughout the reservoir. Large reservoirs having thick, vertical oil columns have been known to exhibit variations in fluid properties with depth. Variations such as these cannot be accounted for in a specific sample. A pattern must be established from several samples or producing characteristics, from various wells, completed at different intervals. In such cases, proper sampling procedures can ensure that the sample obtained is representative of the reservoir fluid at the sampling depth and

Nevertheless, if sampling cannot be performed at initial conditions, or if for any special reason samples have to be taken in a well that has already produced a sizeable amount of oil, the reservoir and production data should be analysed carefully. If the production fluids are still identical to the initial fluids, the sampling procedure will be very similar to that of new wells. On the other hand, if the produced fluid is not identical to the fluid initially in place in the reservoir, one cannot hope to obtain representative samples. 4.1

Introduction to Well Testing (Aug 1996)

4-3

Section 4

Sampling of Reservoir Fluids

Fig. 4-1 Diagram of the pressure distribution within the formation (Reservoir Fluids Sampling Fundamentals).

sampling time. Timing is an important consideration in obtaining a representative sample of the original reservoir fluid. Obviously, it makes sense to sample as early in a reservoir’s producing life as possible. Once production creates significant volumes of free gas on a reservoir-wide basis, obtaining a sample of the original fluid may be impossible. Often, a reservoir fluid sample will be part of a well testing procedure that immediately follows the completion of the first well in a reservoir. An example would be a newly discovered field where development plans may rely on the early determination of expected reserves and production rates. In such cases, it is important that the new well be cleaned up

Introduction to Well Testing (Aug 1996)

before sampling to remove all traces of drilling fluid from the well and wellbore area. Considerable thought and planning are often needed to co-ordinate fluid sampling with other testing procedures so that one does not adversely influence the other. For instance, in modern offshore development situations there is often an emphasis on accelerating production. Drilling, completion, production, and testing activities may be occurring simultaneously from a single platform. This may affect the time, space, or money allocated to fluid sampling. On the other hand, accurate fluid samples are necessary for the decisionmaking behind these development activities.

4-4

Schlumberger

Estimates of fluid properties can be helpful. For example, correlations of bubble point pressure can be employed with early test data (perhaps from a drillstem test) to determine if a reservoir fluid is understaturated. If it is, a well in that reservoir might be produced for some time without fear of a free gas phase forming and sampling of such a well could be deferred while more critical testing is done. 4.1.2 Producing Conditions The producing conditions and surface or subsurface equipment can be important considerations in designing a sampling procedure. The most important of these are; • The type of fluid being sampled. • The stability and accuracy of gas rate, oil rate, and GOR measurements. • The proximity of gas-oil or oil-water contacts to the productive interval. • Whether the well is a flowing or pumping well. • The dimension of downhole equipment. • The well location. Dry gas reservoirs and highly undersaturated oil reservoirs, where the produced fluids remain in a single phase under any flowing conditions (including surface conditions), are relatively easy to sample on surface. An oil reservoir at or slightly above the bubble point will undoubtedly yield free gas at bottom hole flowing pressures and require conditioning prior to sampling. Conditioning is a procedure whereby the production rate is gradually reduced, resulting in successively higher flowing bottom hole pressures. This simultaneously removes the altered fluid from near the wellbore and moves fresh, unaltered reservoir fluid into the pores. If samples of oil and gas are taken at the surface, it is vital that the producing rates and gas-oil ratio be accurately determined in order that Introduction to Well Testing (Aug 1996)

the fluids may be recombined in the correct ratios to formulate a representative sample. If the well Is not producing with stable GOR’s, or if separation facilities are not adequate for accurate measurements, a surface recombination sample should not be considered. Water production can be troublesome, even in small amounts. If possible, no well that is producing water should be considered for obtaining a representative hydrocarbon sample. If necessary, a water-producing well may be sampled if precautions are taken to obtain the sample from above the oil-water contact in the well or separator. Wells that have (or may have) gas coning into the production interval should be avoided as candidates for sampling. Flowing wells are the best candidates for fluid sampling. Production rates are more easily controlled, and measuring the bottom hole pressure is practical. In contrast, subsurface sampling on a pumping well requires the removal of the pump and rods. For obvious reasons, wells on continuous gas lift are unsuitable for surface sampling procedures. However, if a gas lift well will flow at low rates on its own, it may be conditioned and sampled the same as any flowing well. The wireline bottom hole sampler is not extremely large, but may be unsuitable in wells with tubing restrictions (subsurface safety valves, downhole chokes, and the like), or twisted tubing. Any completion equipment that prohibits the sampler from reaching the producing interval will complicate the bottom hole sampling procedure.

4-5

Section 4 4.1.3 Well Conditioning The objective of well conditioning is to replace the non-representative reservoir fluid located around the wellbore with original reservoir fluid by displacing it into and up the wellbore. A flowing oil well is conditioned by producing it at successively lower rates until the non-representative oil has been produced. The well is considered to be conditioned when further reductions in flow rate have no effect on the stabilised gas-oil ratio. • If the GOR remains constant, the flow into the wellbore is monophasic, unsaturated oil and the well can be considered ready for sampling. • If the GOR decreases, the presence of a free gas saturation is indicated. This gas may be present due to coning (the drawing down of the free gas cap into the producing interval) or due to the flowing bottom hole pressure being less than the bubble point pressure. Correlations can be used to determine the normal gas-oil ratio to be expected without any free gas production. • If the GOR increases, simultaneous production of a gas and oil zone may be indicated. The lower drawdown allows less oil and relatively more gas to flow from separate intervals. Such wells should not be sampled, because it is very difficult to determine when they are adequately conditioned. At low flow rates, some wells will “head”, or produce slugs of liquid followed by gas. This irregular flow makes it difficult to measure the GOR accurately. Some wells may have such low productivity that even a low flow rate requires a large drawdown. Reducing the Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids drawdown enough to bring the flowing bottom hole pressure above the bubble point pressure may result in “heading”, or it may take an impractically long time. Pumping oil wells are conditioned in the same general manner. If preliminary correlations show the reservoir fluid to be saturated, the pumping rate should be reduced in order to allow the pressure at the formation face to increase. After the GOR stabilises, the well should be pumped for several days before taking surface samples. If bottom hole sampling is to be done, the pump must be stopped after the well is conditioned and the rods and pump pulled. The well can then be swabbed at a low rate to ensure a representative sample in the bottom of the well before the bottom hole sampler is lowered to the sampling depth. A gas condensate well is also conditioned by flowing it at successively lower flow rates and monitoring the GOR. The GOR should generally decrease as the rate is decreased. This is because the lower rate results in a lower drawdown, which brings the wellbore pressure back out of the two-phase region. The heavier hydrocarbons will be produced rather than condensed in the reservoir, thus increasing the liquid volume at the surface and decreasing the GOR. When the GOR stabilises, the well has been conditioned for sampling. The duration of the conditioning period depends upon the volume of reservoir fluid that has been altered as a result of producing the well below the bubble point pressure, and how quickly it can be produced at low rates. Most oil wells that have not been produced for a long period of time require little conditioning; however, some wells may require up to a week of conditioning to achieve stable GORs. 4-6

Schlumberger

During the conditioning process, careful records should include; • Flowing bottom hole pressure and temperature (if possible). • Flowing tubing pressure and temperature. • Oil and gas flow rates. • Separator pressure and temperature. • Stock tank oil production rate. • Water production rate. Any auxiliary data should be noted, such as radical surface temperature changes, equipment malfunctions, and measurement methods. 4.1.4 Hydrocarbon Sampling Methods After conditioning the well, samples may be taken with a bottom hole sampling device, or individual samples of oil and gas may be taken at the surface and recombined to obtain a representative reservoir fluid sample. The choice of sampling technique is influenced by: • The volume of samples required. • The type of reservoir fluid to be sampled. • The degree of reservoir depletion. • The surface and subsurface equipment. 4.1.4.1 Bottom Hole Sampling Bottom hole sampling is the trapping of a volume of fluid in a pressurised container suspended on wireline inside the well to the productive interval. This method is used when • Only a small volume of fluid is required. • The oil to be sampled is not so viscous that it impairs sampler operation. Introduction to Well Testing (Aug 1996)

• The flowing bottom hole pressure is known to be greater than the reservoir oil saturation pressure. • The subsurface equipment will not prevent the sampler from reaching the appropriate depth or make its retrieval difficult. 4.1.4.2 Surface Sampling Surface sampling involves the taking of samples of separator oil and gas, along with accurate measurements of their relative rates, and reconstructing a representative sample in the laboratory. This method is often used when • A large volume of both oil and gas are required for analysis (as in the case of gas condensate fluids). • The facilities for separating oil and gas and measuring their rates are in excellent condition and operated by thoroughly competent people. • The fluid at the bottom of the well is not representative of the reservoir fluid (i.e., gas condensate reservoirs and oil reservoirs producing large quantities of water). The main difficulty, while sampling on surface, arises from the fact that liquid and gas are in dynamic equilibrium inside the separator. Any drop in pressure or increase in temperature of the separator liquid, which is at its bubble point, will result in the formation of gas. For the separator gas, which is at its dew point, any increase in pressure or decrease in temperature will result in the condensation of heavy components. In such a case, when a fluid becomes diphasic during the sampling operation, it is probable that disproportionate quantities of the two phases will be collected and the sample will not be representative. Also, before any surface sampling is attempted, the sampling point should be checked to ensure there is no 4-7

Section 4 possibility of contamination (oil or gas condensate carry-over for a gas sampling point; water or sludge for a liquid sampling point). If the well is under chemical injection (glycol, methanol, inhibitors...) upstream of the separator, the injection must be stopped and ample time must be allowed for the chemicals to be purged from the separator. If it is impossible to operate without chemical injection, then the chemical used and injection rate must be recorded. Generally, a bottom hole sample is preferred if gas and oil surface measurement capabilities are in question. However, if they are reliable, the surface sampling technique can give a statistically valid value of GOR measured over a long period of time. Whenever possible, separator liquid and gas samples should be taken simultaneously in order to have the same sampling conditions for both fluids. 4.2

Sampling of Oil Reservoirs

4.2.1

Preliminary Conditions on Oil Reservoirs In an oil reservoir, the saturation pressure or bubble point pressure may either be equal to the initial static pressure (saturated reservoirs) or below the initial static pressure (undersaturated reservoirs). If a gas cap is found above the oil, the oil is always saturated. In undersaturated oil reservoirs, it is possible to produce the well on a small enough choke size to ensure a flowing bottom hole pressure higher than the bubble point pressure. There is no gas liberation and the flow in the reservoir is monophasic. On the contrary, in saturated oil reservoirs the flowing pressure is always below the bubble point pressure. Due to this fact, the gas in solution in the oil is

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids liberated and may flow through the reservoir along with the oil (two-phase flow). It is important to point out that when gas and oil flow together through the reservoir, the amount of produced gas is always higher than the gas initially in solution in the oil. The total surface gas-oil ratio is given by;    µ   k rg  GOR = R s +  Bo   o     Bg   µ g   k ro 

Where: GOR Rs Boand Bg µo and µg krg/kro

Production gas oil ratio. Gas in solution in oil. Oil and gas volume factors. Oil and gas viscosities.

Gas/oil-relative permeability-ratio (proportional to the amount of free gas in the reservoir).

This equation shows that in a monophasic flow, when there is no free gas and krg/kro is equal to zero, the GOR is equal to Rs and the well stream is identical to the reservoir fluid. This is the case of undersaturated reservoirs with Pwf > Pb and new wells (even in saturated reservoirs producing with small drawdowns) where there is no free gas and initial production has a GOR equal to Rs. In a two-phase flow, free gas exists and krg/kro is different from zero, GOR is greater than Rs and the well stream is different from the reservoir fluid. This is generally the case of saturated reservoirs. These considerations show that the first condition that must be met to obtain representative samples is a monophasic flow in the reservoir.

4-8

Schlumberger

To summarise: In oil reservoirs, samples are representative when the sampled oil contains in solution exactly the same amount of gas as was in solution in the initial reservoir fluid. 4.2.2 Pre-Job Required Data In order to determine whether the fluid flow in the reservoir is monophasic and that the reservoir is saturated or undersaturated, it is necessary to estimate the bubble point pressure and compare it with the reservoir static and flowing pressures. For this purpose the following data is required; • Initial or present static reservoir pressure (pwsi or pws).

Reservoir temperature. Oil gravity. Gas gravity. Flowing reservoir pressure at one or several flow rates (pwf). • Initial and present gas oil ratio (or production history for producing wells) at one or several flow rates. • • • •

The above data in conjunction with the STANDING correlation (see Figure 4-2) enable the estimation of the bubble point pressure. 4.2.3

• A possible estimation of the bubble point pressure pb, by the STANDING correlation and its value pb < pwsi will confirm that the reservoir is undersaturated. Bottom hole and surface sampling can be done with the well flowing at stabilised conditions at any flow rate which pwf > pb 4.2.3.2 Saturated Reservoirs (pwsi = pb) These reservoirs are characterised by; • A GOR only equal to Rs during a very short production period. Then, the gas oil ratio increases slightly if the well is produced at a constant and low flow rate, and considerably if the drawdown is increased (due to the higher gas liberation in the reservoir). • A possible estimation of the bubble point pressure by the STANDING correlation using GORi = Rsi and its value should be close to pwsi: • A bubble point pressure pb always equal to the initial reservoir pressure pwsi, if a gas cap was proved from geological studies.

New Wells or Wells in Undepleted Zones

Bottom hole sampling can be done as follows;

4.2.3.1 Undersaturated Reservoirs (pwsi > pb) These reservoirs are characterised by;

• The flow rate should be decreased progressively and then the well closed. During the flow rate reduction period, the flowing pressure pwf increases and free gas dissolves in the oil. When the well is finally closed and initial static conditions are reached, the reservoir fluids will be very close to their initial conditions i.e. pb = pwsi.

• A constant GOR equal to Rs. At very high drawdowns, the GOR may increase because the flowing pressure p wf could be lower than the bubble point pressure pb. Introduction to Well Testing (Aug 1996)

4-9

Section 4 • At these conditions, the well can be sampled and it will be opened at the smallest possible flow rate (for example 1/ ″ choke) for 10 or 15 minutes and 16 then shut in just before the sampler is actuated to fill it with fresh reservoir oil. During this short flowing period, the drawdown will be practically zero and gas liberation will be too small to affect the validity of samples. Surface sampling can be done only if, at minimum stabilised flow, the GOR is very close to the initial GOR, but for more security, bottom hole sampling should be done at the same time. 4.2.4

Producing Reservoirs or Wells in Slightly Depleted Zones

4.2.4.1 GOR is equal to GORi

In such cases, the flow is monophasic and surface and bottom hole sampling can be done as indicated in section 4.2.3.1 for undersaturated reservoirs. In this case pws > pwf > pb. 4.2.4.2 GOR is higher than GORi In such cases, the flowing conditions are in the two phase region, the bubble point should be determined using the initial gas oil ratio GORi and compared with the present static pressure p ws and flowing pressure p wf . a) If pws > pb > pwf:

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids Bottom hole sampling can be carried out as for saturated reservoirs, but the time necessary to reach static stabilised conditions could be very long and depends on the depletion of the reservoir. Surface sampling is only available if it is possible to reach production conditions where the GOR is very close to GORi. Nevertheless, if surface samples have been taken with GOR > GORi, they can be recombined in the laboratory and then adjusted in order to get a reservoir fluid having a specific bubble point pressure (for instance equal to pwsi). This procedure is advisable only when the real bubble point pressure is known, but representative samples cannot be taken. b) If pb > pws: The reservoir is very depleted and under two phases. The initial reservoir fluid no longer exists and there is no way to obtain representative samples. For surface samples, the bubble point pressure can be adjusted in the laboratory. 4.3

Sampling of Gas Reservoirs

4.3.1

Preliminary Considerations on Gas Reservoirs Gas reservoirs can be classified in three different categories; dry gas, wet gas and gas condensate reservoirs. In a dry gas reservoir, the gas always remains entirely in gas phase, whether at reservoir or separator conditions.

4-10

Schlumberger

Fig. 4-2 Prediction of minimum gas flow rate required for liquid removal from gas wells.

A depletion from initial to abandonment pressure will not affect its state and the composition of the well stream will be constant. In a wet gas reservoir, the gas also remains entirely in gas phase in the reservoir (at reservoir temperature). A depletion from initial to abandonment pressure will not affect the state of the reservoir fluid, or its composition, which remains constant. However, at separator conditions, the well stream will be in two phases, liquid and gas. The temperature drop between the reservoir and the separator causes the heavier gas components to condense as a liquid. In a gas condensate reservoir, as the pressure drops below the dew point pressure, there is condensation of the gas’s heavier components in the reservoir. The volume of condensed liquid is generally very small with reIntroduction to Well Testing (Aug 1996)

spect to the reservoir’s porous volume and it will not flow, the well stream composition will vary with the pressure. At separator conditions, the production is always in two phases. Very often, wet and gas condensate reservoirs have a very similar behaviour and it is sometimes not possible from well testing data alone to decide which type of reservoir it is. In gas condensate reservoirs, the dew point pressure or saturation pressure may either be equal to the initial static pressure (saturated reservoirs) or below the initial static pressure (undersaturated reservoirs). In undersaturated reservoirs, it is theoretically always possible to produce a well with pwf > pd (pd = dew point pressure), in order to avoid liquid condensation in the reservoir, and to have a well stream identical to the initial reservoir fluid (production with GOR = GORi = constant). 4-11

Section 4 On the contrary, in saturated reservoirs, the production is always with p wf < pd. Liquid deposit is condensed in the reservoir, and the well stream is different from the initial reservoir fluid. The liquid deposit being trapped in the reservoir, the separator gas liquid ratio increases proportionally to the difference between the flowing pressure pwf and the dew point pressure pd. This shows that sampling operation will require flowing conditions with practically no liquid condensation i.e.: GOR = GORi or pwf higher than or very close to pd. To summarise; In gas condensate, wet and dry gas reservoirs, samples are representative when the sampled gas contains the total amount of heavier components contained in the initial reservoir. 4.3.2

Gas Reservoir Sampling Procedures Since in gas reservoirs it is impossible to establish from well testing data whether it is a dry gas, wet gas or gas condensate reservoir, and since in addition, for a gas condensate reservoir, the dew point pressure cannot be previously estimated, sampling should always be done while assuming the most unfavourable conditions; gas condensate with a dew point pressure equal to the initial static pressure. Furthermore, in gas reservoirs, sampling should always be done at the surface, the separator liquid and gas being recombined in the laboratory. Surface sampling methods are the same as for oil reservoirs.

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids Bottom hole sampling is unsuitable for the following reasons; a) In the case of bottom hole gas sampling, the liquid condensed in the bottom hole sampler, when removed from the well, can never be completely transferred from the sampler to the shipping cylinder. Very often, the amount of this condensate is very small (only a few drops wetting the walls of the sampler) and during transfer at atmospheric temperature, part of it will remain in the sampler, the transferred sample thus not being representative. Even if the sampler is heated to reservoir temperature, complete liquid revaporisation could take a very long time and be impossible to check at the wellsite. In such a case, the only solution is to send the sampling chamber, well secured inside a special transportation container, to the PVT laboratory. The laboratory will then be able to transfer and check the sample. b) From the commercial aspect, the liquid phase is of great interest and its analysis requires a volume of about one litre which can easily be obtained at the separator, but never by bottom hole sampling. In addition to proper reservoir sampling conditions, the surface sampling of gas wells will require another condition; the liquid condensed in the production string, between the bottom of the well and the surface, should be completely removed from the well and produced in the separator. This condition will be satisfied if the gas velocity is high enough to carry the liquid phase. The charts in Figure 43 gives such minimum gas flow rates versus well head pressure for different tubing sizes.

4-12

Schlumberger

4.3.2.1 New Reservoirs or Wells in Undepleted Zones At initial reservoir conditions, gas wells can always be sampled because the initial static pressure is very close or equal to the dew point pressure and the gas liquid ratio is very close or equal to the initial gas liquid ratio. Therefore, the well stream contains the total amount of the reservoir fluid’s heavier components. Surface separator sampling should be done with the well producing at the lowest possible flow rate, in order to have the minimum drawdown, but compatible with; • Separator and well stability. GOR and well head pressure should be constant. • Homogeneous flow in the tubing. Liquid deposit should be removed from the well bore by sufficient gas flow velocity. Well conditioning is the same as for surface sampling of oil wells. Under these conditions, even when the dew point pressure is equal or very close to the initial static pressure, and when the flowing pressure is slightly lower than the dew point pressure, samples will be acceptable. In wells having a very low permeability and a very great drawdown, the flowing pressure may be so much lower than the dew point pressure, that samples taken will be considered modified. In this case, representative sampling is impossible. 4.3.2.2 Producing Reservoirs or Wells in Depleted Zones The only data to be analysed is the gas-liquid ratio:

Introduction to Well Testing (Aug 1996)

a) If GOR = GORi, then the well is producing at monophasic conditions in the reservoir and sampling can be done as explained in section 4.3.2.1. b) If GOR > GORi, then the flowing pressure pwf is below the dew point pressure pd, but there is no possibility of establishing whether the static pressure p ws is higher or lower than pd. The research of conditions required for proper sampling when GOR ≠ GORi, is too long to be advised as standard procedure. Validity of any sampling will only be confirmed when p d is known, thus after measurement in the laboratory. 4.4

Sampling of Volatile Oil Reservoirs

A volatile oil is a very light oil with a very high Rs in relation to the bubble point pressure. In some cases, it can be confused with a gas condensate reservoir, mainly due to; • High API gravity of separator liquid. • High GOR. The STANDING correlation cannot be used for estimating the bubble point pressure, since the correlation is only valid for GOR’s less than 2000 scf/bbl. Due to these difficulties, these reservoirs should be sampled as gas condensate reservoirs and the PVT analyses will show what kind of fluid it is. If proved to be an oil reservoir, then bottom hole sampling can also be done as for oil saturated reservoirs.

4-13

Section 4

Sampling of Reservoir Fluids

Fig. 4-3 Prediction of Minimum Gas Flow Rate Required for Liquid Removal from Gas Wells.

Introduction to Well Testing (Aug 1996)

4-14

Schlumberger

4.5

Bottom Hole Sampling

4.5.1

Well Conditioning for Bottom Hole Sampling a) Sampling of a flowing undersaturated well (when GOR = GORi = constant) • The well should be flowing for at least 24 hours, at a minimum stable flow rate ensuring a maximum column height of monophasic fluid. • The pressure at sampling depth must be at least 100 to 200 psi higher than the saturation pressure in the fluid column. A good figure is 500 psi. • The well should be clean - in order to eliminate traces of contaminated oil or water, the stable flowing period should be preceded by a production period equal to 5 to 10 times the total volume of the tubing string. • The flowing stability can be checked by; ♦ Stabilised gas flow rate, oil flow rate and GOR. ♦ Stabilised well head pressure. ♦ Stabilised p wf (best way to ensure flowing stability). b) Sampling of a shut-in saturated well (when GOR ≥ GORi)

• The flow rate should be progressively reduced and then the well shut-in until a stabilised static pressure is reached. A minimum of 12 hours should elapse before sampling is made, with a good figure being 24 hours. The shut-in period, however,

Introduction to Well Testing (Aug 1996)

can be established according to the buildup data or according to the stability of the wellhead pressure. • A static pressure gradient will be very helpful in detecting a possible presence of water. • When the sampler is at sampling depth, the well should be opened on the smallest possible choke only to fill the casing around the sampler with fresh reservoir oil, and then shut-in. 4.5.2

Bottom Hole Sampling Procedures

• The well must have been conditioned beforehand to insure that a single phase representative reservoir fluid is in the wellbore at sampling depth. • A pressure and temperature survey should be run to determine fluid levels and pressures. This will help select the sampling point and confirm the validity of the well conditioning. • Pressures and temperatures should be monitored during the taking of the samples. This will insure that the well conditioning remains valid during the time required for filling the sampling chamber(s). Real time surface read out of downhole recorders is the best option, but it requires the availability of a mono-conductor wireline unit. (In addition the availability of insitu fluid density measurements is also beneficial.) Otherwise, a pressure and temperature recorder should be included in the tool string. The bottom hole sampler, is run in hole in the same way as any downhole production tool.

4-15

Section 4

Sampling of Reservoir Fluids

Fluid Sampling Sampling Possibilities and Procedures Produced Well Fluids Position

Reservoir and Flow Characteristics

New reservoir or undepleted zones Producing reservoirs or depleted zones New reservoir

Gas Reservoirs Volatile Oil or Doubtful Cases

F

Well flowing with pwf > pb

Ri > Rpi pwsi = pb Saturated reservoirs

Progressive reduction of flow rate; well closed until stabilized condition; sampling with well producing at minimum possible flow rate

Flow rate reduction in order to get Rp close to Rpi ; stabilized flow with minimum drawdown

Rp = Rpi = constant pws > pwf > pb

Stabilized flow with pwf > pb

Stabilized flow with pwf > pb

Rp > Rpi pws > pb > pwf

Progressive reduction of flow rate; well closed until stabilized condition; sampling with well producing at minimum possible flow rate

Flow rate reduction in order to get Rp close to Rpi ; stabilized flow with minimum drawdown

Well conditioning could be very long; depends on the depletion.

Rp > Rpi pws < pb

No sampling possibility

Representative sampling is impossible.

Surface samples can be recombined in the lab in order to have pb = pwsi .

Rp = Rpi = constant or Rp very close to Rpi

Not advisable

Smallest possible flow rate but compatible with homogeneous flow in the tubing, separator stability

pd cannot be estimated but measured only in the lab on recombined surface samples.

Rp = Rpi

Not advisable

Smallest possible flow rate but compatible with homogeneous flow in the tubing, separator stability

Rp > Rpi

Not advisable

Smallest possible flow rate but compatible with homogeneous flow in the tubing, separator stability

Validity of sampling will be known after pd measurement.

No possibility of obtaining reservoir characteristics from well test data

Not advisable

Smallest possible flow rate but compatible with homogeneous flow in the tubing, separator stability

Whether sample is representative will only be known after PVT study.

G

H

I

Remarks

Stabilized flow with pwf > pb

D

E

Surface Sampling

Rp = Rpi = constant pwsi > pb Undersaturated reservoirs

B

C Producing reservoirs or depleted zones

Oil Reservoirs

New reservoir or undepleted zones

A

Bottomhole Sampling

Reservoir with a gas cap

Fig. 4-4 Summary of reservoir fluid sampling possibilities and procedures.

Introduction to Well Testing (Aug 1996)

4-16

Schlumberger

The running speed, when the sampler is run with an open chamber, should be between 100 to 200 ft/mn according to the well conditions • When running a circulation type bottom hole sampler, upward and downward movements at sampling depth should be made over 20 to 30 feet in order to drain the sampling chamber completely. • The sampling depth should be as close as possible to the perforated zone to avoid having a large pressure difference between the reservoir and the sampling depth. • A clock operated system should be at the sampling depth about half an hour before the closing time and removed about fifteen minutes after closing. • A minimum of three representative samples should be taken and sent to the laboratory for a complete PVT analysis. It is important to point out that a bottom hole sampler, in which the valves are kept shut due to the internal pressure being higher than the external pressure, is not suitable for sampling undersaturated fluids or formation waters. When the sampler is pulled up hole, the internal pressure will drop very quickly due to the drop in temperature and the risks of opening are very high. In such cases, it is most advisable to use a bottom hole sampler in which valves are kept shut by positive mechanical means. 4.5.3

Bottom Hole Sample Transfer Procedures The sampling chambers of bottom hole samplers are generally not designed for transporting the pressurised reservoir fluids or for storing them. As soon as the sampler is out of hole, the sample, which is trapped in the Introduction to Well Testing (Aug 1996)

sampling chamber, must therefore be transferred to a safe and certified shipping/storing container (sampling bottle, cylinder, receptacle ...). This is done using an apparatus known as a transfer bench. This operation is as delicate as the collection of the fluids downhole and the procedure must insure that the representivity of the sample is not lost between the sampler and the bottle. Since, at surface temperature, the fluid trapped in the sampling chamber is almost always under two phase conditions, there could be a danger of trapping one or part of the phases in the circuits of the transfer bench during the transfer operation. Therefore, it is necessary to displace the collected sample in an homogeneous and monophasic state. This is achieved by re-pressuring the sample, with the transfer bench, to about 1000 psi above its bubble point pressure or static bottom hole pressure (when p b could not be estimated). The transfer will then be made by displacement at this constant high pressure. Forcing gas back into solution, by re-pressurisation only, is not a rapid and simple process. It takes time for an hydrocarbon mixture to reach its phase equilibrium under a given set of pressure, volume and temperature conditions, and the time factor is always a constraint at the wellsite. It has been discovered, however, that an agitation of the sample will speed up the equilibrium process and it is a requirement for any sample transfer bench to feature some kind of sample agitation facility during transfer or bubble point determination. The transfer procedure will thus be as follows; • Re-pressurisation of the sampling chamber 1000 psi above the expected bubble point pressure or static bottom hole pressure.

4-17

Section 4 • Agitation of the chamber and verification that the pressure is stable. • Determination of the apparent bubble point inside the chamber with agitation and pressure stability checks in between each sample volume step. • Displacement of the sample into the shipping bottle at constant pressure and with constant agitation. In case of very viscous oil, it may be necessary to heat up the sampler and the transfer circuitry to the bottom hole temperature. Sample displacement at constant pressure has historically been done with mercury due to its high cohesive forces and thus low wettability properties (making it an ideal medium to totally displace fluids from sample chambers). However, the use of mercury is prohibited in certain areas for health and safety reasons and the use of mercury free sampling equipment is now becoming the norm. This new technology involves special designed equipment using pistons and/or membranes to separate the transfer medium from the sample (to avoid contamination). This technique however, does introduce a new problem as in there is always a certain amount of dead volume which cannot be removed. It is important to check that the composition of this dead volume is as close as possible to the composition of the fluid inside the shipping bottle. A very important safety issue is to ensure that when transfer is completed and the bottle sealed, that a 10% gas cap is created within the bottle to allow for fluid expansion versus temperature during transportation or storage. This insures that the internal pressure will never increase beyond the pressure rating of the shipping bottle. Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids 4.5.4

Checking Bottom Hole Sample Validity Theoretically, the best way to insure the validity of a sample is to measure its bubble point pressure and compare it with the bubble point pressure measured on a duplicate sample taken at the same conditions. The best field procedure is to measure the pressure in the sampling chamber, prior to transfer, and establish a Pressure-Volume curve from which a field bubble point pressure can be determined. For several samples correctly taken in a well which was properly conditioned for sampling, these figures should be within 2%. Consequently, at least three samples should always be taken and compared. As previously mentioned, it is of the utmost importance that the sample be agitated while measuring the pressure changes as no agitation will result in a lack of sharp compressibility change and therefore in arbitrary bubble point measurement with an error which could be as much as 50%. It will also affect the reproduction of the pressure curve and therefore the validity of comparing a duplicate sample. Figure 4-5 shows the pressure-volume plot of a sample in which diphasic fluid was recompressed to 4000 psi. The pressure is recorded together with the cumulative volume of hydraulic oil (transfer fluid) that was displaced from the system at each step. No agitation of the sample was performed. Figure 4-6 corresponds to the same procedure as in Figure 4-5, except that the sample was rocked at 90 degrees for 4-5 times at each step. In this case, the mass of the displaced oil was recorded using a high precision electronic balance. 4-18

Schlumberger

Figure 4-7 illustrates the same procedure as in Figure 4-6, except that the volume of the displaced fluid was recorded instead of the mass (as would happen in the field) This oil sample example clearly shows how the lack of agitation can result in wrong and arbitrary field apparent bubble point pressure estimation. 4.6

Surface Sampling

4.6.1

Well Conditioning for Surface Sampling The stable flow period, during which the samples are to be taken, should be preceded by a cleaning-up period long enough to eliminate the drilling, completion or stimulation fluids. The well should then be flowed through the separator. The flow stability should be determined for the lowest possible flow rate. The choice of such flow rate will depend on the productivity of the well. In high productivity wells, there will be no problem. In average or low productivity wells, or when the productivity is unknown, the choice of a flow rate giving regular flow of the two phases, liquid and gas, to the separator might be difficult. In such cases, the flow must be maintained at the minimum steady flow rate. When the GOR is steady between two flow reductions, the well is producing fluids representative of the reservoir. Flowing stability can be checked by; • Stabilised gas flow rate, oil flow rate and GOR, the temperature and pressure of the separator remaining unchanged.

Introduction to Well Testing (Aug 1996)

• Stabilised well head pressure and temperature. Stabilised pwf (best way to ensure flowing stability). When stability is reached, the well should be produced for at least 12 hours in order to obtain accurate measurements of the gas-oil ratio. The GOR stability should normally be better than 5%, but in difficult well conditions, variations of up to 10% may be unavoidable. In practical terms, time constraints and therefore financial constraints may not permit such ideal conditions but nevertheless less, the longer the flow period, the better the flow stability and the quality of the samples. Accurate knowledge of the GOR is essential for proper recombination at the PVT laboratory of the fluid and gas samples in order to reconstruct a representative reservoir fluid. In the case of a gas well, the liquid condensed in the production string, between the bottom of the well and the surface, should be removed from the well and produced in the separator. This condition will be satisfied if the gas velocity is high enough to carry the liquid phases as previously discussed in section 4.3.2. On the separator side, the pressure must be adjusted to minimise any liquid carryover at the gas outlet. This can be checked against the monogram of Figure 4-8 which gives the theoretical capacity of horizontal separators.

4-19

Section 4

Sampling of Reservoir Fluids

4000 3500

Pressure (psig)

3000

2500

2000

1500

1000

500 0 0

5

10

15

20

25

30

35

40

45

50

55

60

65

Volume of recovered oil (cm3) Fig. 4-5 Pressure-Volume Plot for an Oil Bottomhole Sample, Without Sample Agitation.

4000 3500

Pressure (psig)

3000 2500 2000 1500 1000 500 0 0

5

10

15

20

25

Mass of recovered oil (g) Fig. 4-6 Pressure-Mass Plot for an Oil Bottomhole Sample, With Sample Agitation. Introduction to Well Testing (Aug 1996)

4-20

Schlumberger

4000 3500

Pressure (psig)

3000 2500 2000 1500 1000 500 0 0

5

10

15

20

25

Volume of recovered oil (cm3) Fig. 4-7 Pressure-Volume Plot for an Oil Bottomhole Sample, With Sample Agitation.

4.6.2 Oil Surface Sampling Methods All oil surface sampling methods attempt to keep the separator liquid at or above its bubble point pressure (if possible) until the sample is transferred inside the sample bottle (by keeping the sample at separator pressure and below separator temperature). The sampling bottle must be maintained at or below the separator temperature to prevent gas liberation which would interfere with the filling operation. In cases where the separator temperature is below the ambient temperature, the sampling bottle should be cooled in a water/ice or water/salt/ice bath.

Introduction to Well Testing (Aug 1996)

Separator liquid and gas samples should be taken simultaneously in order to have the same sampling conditions for both fluids. At least two liquid samples of 600 cc should always be taken. As with bottom hole sampling transfer, and for safety reasons, a gas cap of 10% must always be formed in the liquid sample bottle.

4-21

Section 4

Sampling of Reservoir Fluids

Fig. 4-8 Theoretical Gas Capacity of Horizontal Separators.

Introduction to Well Testing (Aug 1996)

4-22

Schlumberger

Fig. 4-9 Surface Oil Sampling by Using a Piston Type Bottle. Introduction to Well Testing (Aug 1996)

4-23

Section 4

Sampling of Reservoir Fluids

Oil or Condensate Sampling Methods

Decreasing Sample Validity

Method

Advantages

Disadvantages

Field of Application

Equipment

Mercury displacement

Liquid sample under monophasic conditions

Mercury safety

No limits

Mercury flask Stainless steel bottles Manifold, valve and pressure gauge

Displacement using a piston or membrane type (Recommended Technique)

No mercury

Vacuum pump and gauge needed Dead volume

No limits

Vacuum pump and gauge PSR-F or membrane type bottle Flasks Hydraulic oil and pump Manifold, valve and pressure gauge

Displacement and equilibrium with separator gas

No mercury

Slight modification of Liquid of liquid composition low viscosity from gas-cap must be reported

Flasks Stainless steel bottles Manifold, valve and pressure gauge

Gas or air displacement

No mercury Easy sampling

Risk of slight No limits modification of liquid composition from gas-cap formation technique

Flasks Stainless steel bottles Manifold, valve and pressure gauge

Water displacement

No mercury Liquid sample under monophasic conditions

Possible solution and reaction of CO2 and H2 S

Flasks Stainless steel bottles Manifold, valve and pressure gauge

Not to be used if CO2 or H2S present

Fig. 4-10 Oil Condensate Sampling Methods.

Introduction to Well Testing (Aug 1996)

4-24

Schlumberger

4.6.2.1 Piston Bottle Displacement Method The mercury displacement method is the most reliable method for liquid surface sampling, however, as previously mentioned, the health and safety aspects associated with mercury is slowly removing it from sampling procedures in uncontrolled field environments. The new mercury free sampling bottles (piston or membrane type) have slightly different but essentially the same procedures. Such bottles contain a very low dead volume above the piston (or membrane) which will have to be evacuated before sampling (1 to 2 mm Hg vacuum) A non compressible liquid replaces the mercury below the piston and the procedure, as illustrated in Figure 4-9 becomes similar to the mercury displacement method. The 10% gas cap is formed by removing enough liquid from below the piston. Other methods of oil sampling are also available but not recommended. A summary of the different types is given in Figure 4-10. 4.6.3 Gas Surface Sampling Methods When sampling gas at surface, enough gas sample volume should be collected in order to allow recombination at reservoir conditions by the PVT laboratory. The total volume of gas sample to be taken per oil sample can be defined as follows; VG > 2.5 * (GOR/P) where VG

=

GOR = P

=

Minimum volume of gas in litres under separator sampling conditions. Separator gas oil ratio in Std cu. ft./bbl. Separator pressure in psi.

4.6.3.1 Vacuum Method This involves filling a container which has previously been evacuated. This eliminates any condensation, which could be due to gas circulation, since once the bottle has been filled, it is not re-circulated. In the laboratory, any condensation is vaporised. The minimum vacuum (maximum pressure) allowed is 10 mm Hg (10 Torr), but the recommended vacuum of 1 to 2 mm Hg should normally be obtained before sampling is attempted. It takes 1/2 hour to 1 hour to evacuate a 20 litre bottle to this recommended void. When sampling, no heating of the bottle is necessary, nor should purging or recirculating of gas be attempted. Poor vacuum will lead to air contamination of the sample. Figure 4-11 illustrates this method. Other methods of oil sampling are also available but not recommended. A summary of the different types is given in Figure 4-12. 4.6.4

Special Surface Sampling Cases

4.6.4.1 High Pressure Samples In most cases of high pressure samples, gas and liquid can be readily distinguished at surface conditions. However, if doubt exists while sampling a high pressure monophasic fluid, the mercury displacement method should be considered. 4.6.4.2 Multistage Separation System In the case of multistage separation (more than one separator in series), gas and liquid must be taken form the first (high pressure) stage separator. In exceptional circumstances, liquid samples could be taken from the lower

In all cases, an extra gas bottle must be filled per set of surface samples. Introduction to Well Testing (Aug 1996)

4-25

Section 4 stage separators, but only if samples of gas are taken from the same lower stage separator and all higher stage separators. All gas flow rates must also be noted. 4.6.4.3 Hydrogen Sulphide Hydrogen sulphide, H2S concentrations in a sample can change due to reaction, adsorp-

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids tion or solution. Laboratory analyses frequently give reduced H2S concentrations due to these phenomena. Thus in such cases where hydrogen sulphide is present in a reservoir fluid, on-site analysis (even by Draeger reactive tubes) is highly recommended. Concentrations in all produced fluids should be determined.

4-26

Schlumberger

Fig. 4-11 Surface Gas Sampling by the Vacuum Method.. Introduction to Well Testing (Aug 1996)

4-27

Section 4

Sampling of Reservoir Fluids

Gas Sampling Methods

Decreasing Sample Validity

Method

Advantages

Disadvantages

Field of Application

Equipment

Filling under vacuum (recommended technique)

No heating Fast High volume sampled

Vacuum pump and gauge needed

No limits

Vacuum pump Manifold including 2 valves, 1 vacuum gauge and 1 pressure gauge

Mercury displacement

No heating No vacuum pump

Only stainless steel bottles can be used, may be low volume Risk of chemical reaction between mercury and H2S Large volume of mercury needed

No limits

Mercury flasks Stainless steel containers Manifold including valve and pressure gauge Safety equipment

Purging

High volumes sampled No vacuum pump

Risk of condensation from cooling Heating advisable

No limits

Flowmeter Manifold including valve and pressure gauge

Circulation

High volumes sampled No vacuum pump

Risk of condensation from cooling

No limits

Flowmeter Manifold including valve and pressure gauge

Separator water

High volumes sampled No vacuum pump

Long duration

No limits

Manifold including valve and pressure gauge Separator water Flasks

Salt water

High volumes sampled No vacuum pump

Long duration Possible change of composition

Preferably with no H 2S or CO2

Manifold including valve and pressure gauge

Fresh water

High volumes sampled No vacuum pump

Long duration Possible change of composition

Preferably with no H 2S or CO2

Manifold including valve and pressure gauge

Air displacement

Water displacement

Fig. 4-12 Sampling and PVT Analysis Sampling Techniques and Recommendations. (Surface Testing Services).

Introduction to Well Testing (Aug 1996)

4-28

Schlumberger

4.6.5

Well Head Sampling of Oil and Gas

4.6.5.1 Oil Sampling at Well Head Oil sampling at the well head is possible only when the well head pressure is higher than the bubble point pressure at well head temperature. For this condition to be achieved, it may often be necessary for samples to be taken at a low flow rate, typically less than 10 tubing total volumes per day. Clearly, a good idea of the bubble point is needed. It is recommended that separator samples be taken at the same time to act as a back-up in case of unexpected two phase flow at the well head or other causes of invalid sampling. Normal liquid sampling methods should be used but great care should be taken to make sure sample bottles and associated equipment have a working pressure rated above the well head pressure. 4.6.5.2 Gas Sampling at Well Head Since most gas condensate wells produce two phases at the surface, gas sampling at the well head will only be possible on the rare occasions when monophasic well head flow is expected, with a well head pressure higher than the dew point pressure at well head temperature. As in oil sampling at the well head, separator back-up samples should be taken. Most applications will be for dry gas wells where no liquid is formed in the separator. In this case, a well head sample will be identical to a separator gas sample. Normal gas sampling methods should be used but great care should be taken to make sure sample bottles and associated equipment have a working pressure rated above the well head pressure.

Introduction to Well Testing (Aug 1996)

Fig. 4-13 Example of DST Recovery of Formation Water. Changes in Ionic Composition are Apparent as Produced Water Becomes More Representative of Formation Fluid in the Lower Portion of the Drillpipe.

4.7

Sampling of Formation Water

The goal of formation water sampling is also to obtain a representative sample. The composition of formation water is not generally as dependent on the temperature and pressure changes, and sampling procedures are in most cases simpler than for oil and gas. Small volumes of formation water retrieved from a drillstem test or formation tester can often be used for analysis of dissolved salts. Samples are more often taken from the separator or wellhead. 4.7.1

Formation Water Sampling Methods

4.7.1.1 Drillstem Tests During a drillstem test, or DST (as discussed in Section 3), fluids may or may not flow to surface, depending on the reservoir pressure, the productivity of the formation and the design of the test. If formation fluids flow to the surface, sampling may proceed just as in the case of a production well test, as long as care is taken to produce all of the drilling mud filtrate that has invaded the pore space in the near wellbore area. This cleaning-up process may take some time, depending on how much filtrate loss or lost circulation occurred during drilling.

4-29

Section 4

Sampling of Reservoir Fluids

DST’s are not generally reliable means of obtaining oil and gas samples, unless all the requirements for a representative sample are met. Water samples taken form DST’s are generally reliable if tests are made on site to ensure a representative sample. For example, when water samples taken at intervals from the produced fluid column within the drill pipe on a well that did not flow to surface

a

b

were analysed, they showed how errors can be caused by incorrect sampling. In Figure 413, an analysis is shown of top, middle and bottom samples taken from a 50 ft column of fluid. The data shows an increase in salinity with depth in the drillpipe, indicating that the first water was contaminated by mud filtrate.

c

d

e

Electric

Electric

Electric

Electric

Electric

Sample

Pumpout

Pumpout

Pumpout

Pumpout

Hydraulic

Sample

Sample

Sample

Sample

Single probe

Packer

Hydraulic

OFA module

Multisample

Single probe

Hydraulic

OFA module

Single probe

Hydraulic

Single probe

Fig. 4-14 MDT tool configurations for sampling. (a) Basic tool. (b) Inflatable packer for low-permeability and difficult conditions. (c) Pumpout module may be used to remove filtrate. (d) OFA module differentiates among gas, oil and water. (e) The multisample module contains six individually controlled 450-cm3 PVT containers.

Introduction to Well Testing (Aug 1996)

4-30

Schlumberger

If it is desirable to measure the dissolved gas content of the formation water, the DST must be shut in to allow the pressure at the bottom of the well to build up. A sample of the formation water at formation conditions may then be taken with a bottom hole sampler run through the drillpipe. Some DST tools also contain a subsurface sampling device within the tool string assembly that catches a sample at bottom hole flowing conditions. 4.7.1.2 Surface Sampling Two very basic methods exist; in the first, a plastic or rubber tube is used to fill a sample bottle from the bottom. Several volumes of fluid are displaced before the tube is slowly removed and the container sealed. An alternative method is to place the sample container within a larger container, filling from the bottom of the inner one until the brine overflows both containers. The sample is then capped under water to prevent air contamination. 4.8

Sampling Equipment

There are a variety of sampling equipment available on the market for both surface and bottom hole sampling techniques. Different equipment is designed to meet specific needs, and each service company will have its preferred tools. Whatever the equipment selected, the end objective of obtaining representative samples must be at the front of ones mind and it should be clear that if the well has not been properly conditioned and/or proper procedures are not followed; then no matter what equipment is used the samples are not going to be of the quality required.

Introduction to Well Testing (Aug 1996)

4.8.1

Bottom Hole Sampling Equipment

4.8.1.1 Open Hole Sampling from Formation Interval Testing Although primarily an open hole logging device for confirmation of formation fluid and indications of productivity and formation pressure, the formation interval testing tool can also collect fluid samples for laboratory analysis. The Repeat Formation Tester (RFT) and Modular Formation Dynamics Tester (MDT) as discussed in section 3 are two such tools which can be used for this purpose. The main advantage of this technique is that it gives an early set of samples giving some preliminary estimates before the well is cased and the reservoir produced. The main drawbacks are that the samples collected are generally from the near wellbore formation and may not contain fluids representative of the actual hydrocarbon reservoir. The MDT’s flexible operational mode coupled with its resistivity and optical analyser sensors allows greater quantities of fluid to be removed from the near wellbore and the fluid to be monitored thus increasing the prospects of obtaining a more representative sample See Fig. 4-15 and 4-16. 4.8.1.2 DST Sampling Tools The fluid sample trapping features of DST strings have a lot in common with the Formation Interval Testers. The one main important difference is that the DST sampler can take larger samples and after significant quantities of fluid have been produced from a zone and/or the well conditioned for sampling. This being said most tools on the market, while capable of catching large samples, even if the well does not flow to surface, are generally not capable of trapping samples suitable for full PVT analysis. One exception 4-31

Section 4 is the Fullbore Annular Sample Chamber Tool (FASC).(Discussed in section 3.4.2.9 Figure 3-22) The tool is activated by a specific annulus over-pressure at any time during the test. A sample regulator ensures that the sample is acquired in a controlled manner to avoid gas being drawn out of solution. The sample is trapped in a detachable chamber which can be sent directly to the laboratory or alternatively the sample can be transferred to a shipping bottle as previously described. 4.8.1.3 Production Sampling Tools Through tubing production sampling tools, whether run on electric wireline or slickline, are best suited for bottom hole sampling of reservoir fluids since they can be accurately positioned in front or just above the perforations. Thus, fluids sampled should not be contaminated by fluids coming from higher levels. Downhole pressure and temperature measurements (either from downhole recorders or in real time) can be easily combined with the sample procedure which gives another level of control in ensuring correct well conditioning and suitability for sampling. Production samplers are small and light enough so as to allow simple field operations, especially concerning bubble point pressure measurement and sample transfer. The internal design of the sampling chamber and the transfer circuit is such that dead volumes are kept to a minimum. Circulation type samplers are run in hole open and the open sampling chamber must be cleaned up with downhole fluid once the sampling depth is reached. While, open, the tool must never go deeper than the sampling depth and the clean up is achieved by moving the tool up and down slowly for at least 5 minutes above the sampling depth. The main advantage of this type of sampler is the removal of the risk of any gas coming out of Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids solution during the sampling process, the main disadvantage is the risk of the sample chamber not being clean. Admission type samplers are run in the hole closed. Once at the required sampling depth, they are activated either by battery and down hole clock (pre-set at surface for operations on slickline) or via an electrical signal sent from surface. This in turn either sets off a small detonator which shears a plug thus allowing the sample to enter the chamber or operates a solenoid valve. The sampling is done via the sample pushing on a piston which forces clean hydraulic oil through a regulated orifice to avoid gas breaking out of solution. When sampling is complete a sealing system is automatically activated. Admission type samplers are the most common used in the oilfield and many variations on their design exist. 4.8.1.4 Transfer Benches Transfer benches are designed, as the name suggests, to transfer a sample from the sampling chamber of the bottom hole sampler into a bottle that can safely be shipped to the PVT laboratory. In order to maintain the representivity of the transferred sample, the transfer must be performed at constant pressure and in single phase above the bubble point or dew point. This entails that transfer benches must first enable a correct measurement of the bubble point pressure before any transfer may actually be started. Modern transfer benches are designed for use with mercury free systems.

4-32

Schlumberger

Gas detector Lamp

€À@Q¢ ¢QÀ€@ @@@@ €€€€ ÀÀÀÀ QQQQ ¢¢¢¢ ¢QÀ€@ @@@@ €€€€ ÀÀÀÀ QQQQ ¢¢¢¢

Fluid flow

Flowline

Liquid detector

Water

Oil

Gas

Fig. 4-15 The OFA Module With its two Sensor Systems; One for Liquid detection and Analysis and the Other for Gas Detection.

4 Visible Near infrared Crude A

Optical density

3

Crude B

Oil-base mud filtrate

Water

2

Condensate

1

Diesel 0 500

1000

1500

2000

Wavelength (nm)

Fig. 4-16 The Absorption Spectrum of Water and Several Oils.

Introduction to Well Testing (Aug 1996)

4-33

Section 4

Sampling of Reservoir Fluids

Battery pack Clock

Air chamber

Break plug

Hammer pin Break plug Regulator choke

Detent assembly Braking system

Rod

Hydraulic oil

Sample fluid

Floating piston Port holes

Sealing piston Mercury

Fig. 4-17 Battery-Operated Sampler.

Introduction to Well Testing (Aug 1996)

4-34

Schlumberger

4.8.2 Surface Sampling Equipment Surface sampling equipment is a collection of special fittings couplings and hoses required to connect the oil or gas sample bottles to the relevant points on the separator. In addition, vacuum pumps and hydraulic pumps are required for bottle preparation prior to sampling. Modern systems can be provided which consist of self contained modules complete with automatic flow control and monitoring equipment to confirm the quality of the sampling process. With such systems the sampling process is practically automated.

Fig. 4-18 Multipurpose Transfer Bench

4.8.3 Sample Containers Sample containers are “mobile pressure vessels” containing hydrocarbon gases and liquids together with other hazardous subIntroduction to Well Testing (Aug 1996)

stances. As such, they are subjected to stringent regulations in terms of design, manufacturing, testing, certification, operation and transportation. They should only be handled by fully trained personnel. 4.8.3.1 Gas Sample Bottles Gas sample bottles are defined not only by working and test pressure, but also by a maximum sampling pressure verses temperature. This is a direct consequence of the general gas law “PV/ZT = constant”. Since a gas sample bottle may be subjected to high temperature after it has been filled, the initial filling pressure at initial filling temperature must be such that the bottle’s internal pressure will never be allowed to exceed the safe maximum working pressure for which the bottle was certified. (specifically during operations in hot climates) There are many types of designs for gas bottles each with their own merit. The most practical feature being ease of handling and transportation. 4.8.3.2 Oil Sample Bottles Oil sample bottles are designed to contain liquid samples which can be taken under very high pressures. They are therefore much more rugged in design than gas sample bottles but nevertheless still subjected to stringent regulations. The one key factor with oil sample bottles is to ensure that a safety gas cap is made in the bottle after sampling as the internal pressure can very quickly rise with temperature for a fluid under monophasic conditions. As with gas bottle there are many different designs available the key points to note, are the ease with which an oil bottle can be cleaned and the elimination of as much dead space as possible. New bottles are almost entirely mercury free; either piston displacement types or membrane types.

4-35

Section 4 4.8.4 New Developments Recent developments have seen the introduction of Monophasic Bottom Hole Samplers whereby the sample is taken and maintained and transferred under monophasic conditions. Whereas this technique solves many of the disadvantages discussed throughout this section, extra care must be taken as the equipment uses some form of nitrogen pre-charge which requires highly trained personnel to operate it. Another technique currently under development, is that of Inert Bottom Hole Sampling - this involves the use of sophisticated metals which will not allow adsorption of gases from the sample into the body of the sampling tool. 4.9

Fluid Analysis and Uses of Data

The values of the physical properties of a reservoir fluid constitute an integral part of the data required for a comprehensive study of the reservoir and the optimal design of the oil recovery and production schemes. More specifically, PVT data, corresponding to the fluid under study, are needed to validate the well test and to provide meaningful interpretation. The optimum design of the well completion and the surface facilities is possible only after determining the type of fluids which flow through the well bore and are produced through the separator, as well as their volumetric properties. In addition, the accurate estimation of the reservoir reserves and the design of an optimum depletion strategy is feasible when realistic fluid properties are available. It is well known that an underestimation of the oil formation volume factor by 20%, not unusual when using current correlations, is enough to reduce the expected reserves by the same percentage.

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids The report of the full PVT study on a representative reservoir fluid sample answers all of these questions and provides the necessary data. Nevertheless, in most cases, due to backlogs, expedition and transportation problems, this report only becomes available several months after the well test. Meanwhile, crucial decisions concerning the management and the planning of the reservoir have to be taken and are based on estimated values of reservoir fluid physical properties. 4.9.1

Field Estimation of Reservoir Properties

4.9.1.1 Field Estimation of Reservoir Properties through Correlations Physical properties of reservoir fluids may be derived from correlations. These correlations can provide reasonable estimates given that the chemical nature of the fluid under test is similar to the one of the fluids that have been used for developing the correlation. The rapidity in obtaining answers at no cost is the advantage of this approach. There is however, a considerable risk concerning the accuracy of these predictions as is shown in the example in Figure 4-19 where the estimations for p b (bubble point pressure in psia) and Bo (oil formation volume factor in reservoir barrels per standard conditions barrel) given by three of the best known correlations are compared with the values measured in the PVT laboratory.

4-36

Schlumberger

The three correlations used were: • STANDING Correlation, which has been derived from the study of 105 Californian crudes. • LASATER Correlation, which has been derived from the study of 137 Canadian, Western, Mid-Continental and South American crudes.

Alascan crude 1 Alascan crude 2 Angola Nigeria

Pb Bo Pb Bo Pb Bo Pb Bo

• GLASO Correlation, which has been derived from the study of 45 North Sea crudes. In addition correlations can only provide data limited to certain conditions (e.g. bubble point) and since the compositional analysis of the fluid is not taken into consideration, they fail to predict the evolution of each property as the main reservoir parameters change during depletion.

PVT Lab

STANDING

LASATER

GLASO

1802 1.20 1159 1.16 4210

1505 1.20 955 1.14 6308

1504

4400

1645 1.12 1011 1.12 4775

5768 3.45

8785 4.20

7500 3.80

6367 4.40

918

Fig. 4-19 Example Data.

4.9.1.2 Field Estimation of Reservoir Properties with Portable PVT Another approach for estimating the physical properties of reservoir fluids on the well site is through the use of a portable PVT laboratory. This approach, although it offers good measurements, involves considerable investment in high pressure and high temperature equipment and requires the presence of well trained, experienced personnel. As the PVT study should always be performed under conditions of perfect equilibrium, it takes several hours, if not days, of rig time to be completed. In addition, the weight and volume limitations imposed on the equipment to be transported to the rig dictates that the PVT equipment used on-site has consideraIntroduction to Well Testing (Aug 1996)

bly reduced volume than that used in the laboratory. The reduction of the sample volume inevitably leads to reduced accuracy of measurements. 4.9.1.3 Field Estimation of Reservoir Properties with Fluid Properties Estimator As compared to an estimation from correlations, a more reliable prediction of phase behaviour and physical properties can be achieved at the well site using a Fluid Properties Estimator. This service can be performed on downhole reservoir samples as well as on the surface gas and liquid samples taken form the separator. Simple measurements of key physical properties, specially selected to 4-37

Section 4

Sampling of Reservoir Fluids

characterise the reservoir fluid, are performed on-site with easily operated portable equipment. These measurements are used as calibration points to tune an Equation-of-State (EOS) based simulator which runs on the field computer. The tuned thermodynamic model is used subsequently to predict the phase behaviour of the fluids at reservoir, well and surface conditions, and thus to generate the principal PVT data necessary for the preliminary reservoir and production engineering calculations. Hence by providing predictions based on measurements, the fluid properties estimator combines the simplicity and speed of the correlation approach with the accuracy of the portable PVT. The service takes less than 3 hours to perform once a sample has been brought to surface. The accuracy of the predictions is better than 5 % and the range of the possible deviations for each of the main sets of estimations is expected as follows; • Bubble point pressure (pb): 0-3%

4.9.2

PVT Laboratory Measurement of Reservoir Properties As soon as the samples arrive at the PVT laboratory the first thing that is done is to check their validity, in other words to ensure they are usable and have not been contaminated or destroyed by bad sampling / transfer techniques. Compositional analysis of the reservoir fluid is a key component of a PVT analysis and has several applications in reservoir and production engineering. The most important application is establishing how much gasoline, kerosene, fuel oil and bitumen will be extracted when refining a barrel of crude oil. It also dictates how and where the oil will be refined. Another application is detecting corrosive compounds that require special consideration when defining production and transportation equipment. Composition is also required as an input to equation-of-state simulators used for reservoir description. Techniques used include, gas chromatography, distillation and micro distillation and mass spectrometry.

• Oil volume factor at pb:

0-1%

• Oil density at pb:

0-2%

• Black-oil PVT Studies.

• Oil viscosity at pb:

5-10%

• Volatile or near-critical Fluid PVT Studies.

• Total GOR:

0-5%

• Compositional analysis

0-3%

Figure 4-20 shows the basic flow chart for the fluid properties estimator for a down hole sample. While Figure 4-21 shows the same for a surface sample.

Introduction to Well Testing (Aug 1996)

Other studies available include;

• Gas-condensate Reservoir Fluid PVT Studies. • Enhanced Oil Recovery Studies. • Stock Tank Analysis. • Formation Water Analysis.

4-38

Schlumberger

Reservoir Fluid Composition C7+ GCGas Analyzer (y)

Flash GOR

Gas Reservoir Fluid

Flash Kit (Vg, mo)

Res. Fluid Sample

Liquid

Tuning Kij CH4 - C7+ Pc, Tc - C7+

Gas Comp. C7+

GC Liquid Analyzer (x)

Liquid Comp. C7+

Refractometer (n)

Liquid Mol. Mass

Densimeter (ρ)

Liquid Density

Viscometer (µo)

Liquid Viscosity

Transfer Bench (PVT)

Res. Fluid pb (T amb.)

Equation-of-State Simulator Match Points

pb at res. conditions Volume factor Compressibilities Viscosities Densities Separator GOR Tank GOR Shrinkage factors Relative volumes Res. fluid comp. C7+

Fig. 4-20 FPE Flow Chart - Reservoir Sample.

Separator Gas

Separator

Separator Liquid

Separator Flowmeters (qg, ql)

Separator GOR

Pycnometer (density)(ρ)

Shrinkage Factor

(1) + (2) + (3)

Sep. Gas Comp. C7+

GCGas Analyzer (y)

(1) Tank Gas Comp.

Reservoir Fluid Composition C7+

Tuning Kij CH4 - C7+ Pc, Tc - C7+

(2) Tank GOR

Tank Gas

Flash Kit (Vg, mo)

Tank Liquid (STO)

GC Liquid Analyzer (x)

STO Comp. C7+

Refractometer (n)

STO mol. Mass

Densimeter (ρ)

(3) STO Density

Viscometer (µo)

STO Viscosity

Transfer Bench

Res. Fluid pb (T amb.)

(Recomb) (PVT)

Equation-of-State Simulator Match Points

pb at res. conditions Volume factor Compressibilities Viscosities Densities Separator GOR Tank GOR Shrinkage factors Relative volumes Composition C7+ Sep.streams comp. C7+

Fig. 4-21 FPE Flow Chart - Separator Samples. Introduction to Well Testing (Aug 1996)

4-39

Section 4

Introduction to Well Testing (Aug 1996)

Sampling of Reservoir Fluids

4-40

Section 5 Basic Well Test Interpretation

Section 5

Introduction to Well Testing (Aug 1996)

Basic Well Test Interpretation

5-2

Schlumberger

5.0

Basic Well Test Interpretation

5.1 Introduction Comprehensive interpretation of acquired data contributes to efficient reservoir development and management by quantifying parameters that characterise the dynamic response of the reservoir. The objectives of any well test interpretation is to obtain the most self consistent and correct results that correspond to and confirm the pre-defined reservoir model. 5.1.1 Reservoir Model The field development strategy is usually established with the help of models. Two key models are the; • Economic Model • Reservoir Model

RESERVOIR MODEL

ECONOMIC MODEL

FIELD DEVELOPMENT STRATEGY

Fig. 5-1 Example Model.

The economic model takes into account present and projected economic information such as; Price of Oil and Gas. Interest Rate. Return on Investment. Cost Factors. Exchange Rates. Political Climate. etc.

Introduction to Well Testing (Aug 1996)

The reservoir model is used to predict the physical behaviour of the field, in terms of production and fluid recovery, for different operating conditions such as;

• Primary Production. • Secondary Production. • Tertiary Recovery. In order to perform realistic predictions, the reservoir model should represent the actual reservoir as closely as possible. It is built up using information available from geology, geophysics and data previously collected from nearby wells. A reservoir model is never definitive, and must be adjusted as new information becomes available when additional wells are drilled. The parameters used for the construction of the reservoir model are obtained from;

• Direct Measurements On cores or cuttings while drilling. On PVT samples.

• Results of Interpretations Seismic data. Electric log data. Well test data. 5.1.2 Well Test Interpretation Well testing is a powerful tool for providing data for the model as it requires the reservoir to be in a dynamic situation. Different types of well tests and the type of parameters measured or calculated were discussed in section 3, to recap, a well test, depending on its design may provide; 5-3

Section 5

Basic Well Test Interpretation

Geophysics

Electric Logs

Well Test

P.V.T. Core

INTERPRETATION

Geophysical Model

LogM odel

PRESSURE

MEASUREMENT

Well TestModel

RESERVOIR ENGINEERING

TIME

RESERVOIR MODEL

Fig. 5-2 Typical Input to Reservoir Model.

• Permeability. • Initial or average pressure. • Near wellbore conditions (damage, • • • •

stimulation). Reservoir flow behaviour. Reservoir size. Inflow performance response. Communication between wells.

A well test essentially consists of recording the downhole pressure response due to changes in flow rate. A disturbance is created in the reservoir by changing the flowrate; (Fig. 5-3)

Fig. 5-4

The pressure response, which is a function of reservoir characteristics and the production history can be interpreted by analysis of; Change in pressure, ∆p versus change in time ∆t; The analysis of this response and the matching of the characteristics of the response to a known model is the basis of well test interpretation. It is based on the solutions of complex partial differential equations which model the fluid through the porous media in the formation. This type of analysis is often termed transient well test analysis.

FLOW RATE

FLOWRATE PRESSURE

∆p

∆t q (t)

TIME

TIME

Fig. 5-3

Fig. 5-5 Change in Pressure Versus Time.

and measuring the associated pressure response; (Fig. 5-4).

Introduction to Well Testing (Aug 1996)

5-4

Schlumberger

There are two main types of transient analysis tests;

• Pressure Drawdown Test

After the well has been shut in for a long enough period to establish static pressure conditions, the well is opened and produced at a steady rate while the pressure fall off is observed with a bottom hole gauge.

• Pressure build-up Test

After flowing the well for a long enough period to establish quasi-steady state conditions, the well is closed in while the pressure build-up is observed with a bottom hole gauge.

The analysis of build-up data is preferred because achieving the stable flow rates required for drawdown analysis can be difficult and or lengthy to achieve. The generation of drawdown curves is simpler because of the use of actual flow rates within the diffusivity equations. Unfortunately, any discussion of well test interpretation invariably involves complex looking equations which require a good understanding of mathematics to manipulate and solve. A detailed understanding of the equations associated with the different models is beyond the scope of this document but reference will be required to some of the basic equations which form the core of the mathematical models. Fortunately, modern well test interpretation techniques have developed structured approaches to identifying common reservoir models and this coupled with user friendly software helps unbundle the complexity of the science and gives access to a basic understanding.

Introduction to Well Testing (Aug 1996)

5.2

Defining the Reservoir Model

5.2.1 Inverse and Direct Problem The principles governing the analysis of well tests are more easily understood when one considers well test interpretation as a special pattern recognition problem. In a well test, a known signal I (the constant withdrawal of reservoir fluid) is applied to an unknown system S (the well and reservoir) and the response O of that system (the change in reservoir pressure) is measured during the test.

I

S

O

Fig. 5-6 Inverse and Direct Problem.

This type of problem is known in mathematics as the inverse problem; O/I → S Its solution involves finding a well defined theoretical system, whose response to the same input signal is as close as possible to that of the actual reservoir. The purpose of well test interpretation is to identify the system knowing only the input and output signals and possibly some other reservoir characteristics, such as initial or boundary conditions, shape of drainage area, etc. The response of the theoretical reservoir is computed for specific initial and boundary conditions, that must correspond to the actual ones. This is called the direct problem; IxS→O Interpretation relies on models, whose characteristics are assumed to represent the characteristics of the actual reservoir. If the wrong model is selected, then the parameters 5-5

Section 5 calculated for the actual reservoir will not be correct. On the other hand, the solution of the inverse problem is not always unique. It is possible to find several reservoir configurations that would yield similar responses to a given input signal. But the number of alternative solutions decreases as the number and the range of output signal measurements increase. Models used for analysis are always built in the same manner. They include:

Basic Well Test Interpretation been identified and mathematical models built to define them, such as;

• Dual Porosity The double porosity (or dual porosity) reservoir consists of two homogeneous porous media of distinct porosity and permeability that interact. They may be uniformly distributed or segregated but only one medium can produce fluid to the well; the other acts as a source.

• A Basic Model

• Dual Permeability

• Inner Boundary Conditions

The double permeability reservoir refers to two distinct porous media, as in a double porosity system, but, in this case, each medium can produce into the wellbore.

• Outer Boundary Conditions 5.2.2 Basic Model Most basic models in the oil industry have impermeable upper and lower boundaries, and are of infinite lateral extent. Initially, the pressure is uniform. Under these assumptions, flow eventually becomes radial in the reservoir. We may have either a homogeneous or a heterogeneous acting porous medium. 5.2.2.1 Homogeneous Reservoir A homogeneous acting reservoir is one that, with respect to flow, acts as though it has identical properties throughout. This condition is even diagnosed for a reservoir which is so randomly heterogeneous, it acts as though it is a single homogeneous reservoir. Many petroleum reservoirs have been found to be homogeneous and, in the early years, all reservoirs were considered to be homogeneousacting for purposes of analysis. 5.2.2.2 Heterogeneous Reservoir Heterogeneous acting reservoirs have been the subject of many recent developments in well test analysis. Many different types have Introduction to Well Testing (Aug 1996)

• Multi-Layered In multi-layer reservoir models, only one layer delivers fluids to the wellbore. The other layers act as sources of fluids.

• Composite Composite reservoir models, consist of a mixture of the above systems. 5.2.2.3 Radial Flow Flow through non-fractured formations is approximately radial, at least for a few hundred feet surrounding the well, therefore an idealised cylindrical model may be used to calculate flow rates and describe pressure distribution with good accuracy. This is the basic flow regime used and assumed as a starting step in the basic model. Other flow regimes do exist and the model can be updated to account for them. Such as; spherical flow, linear flow, bi-linear flow, etc.

5-6

Schlumberger

5.2.2.4 Infinite Acting Radial Flow The reservoir production is established when all the fluid flowing at surface comes from the reservoir. p (t)

q (t)

5.2.3 Inner Boundary Conditions To be useful for practical applications, the basic model must be associated with Inner boundary conditions. Inner boundary conditions that are most common in practice are; Wellbore storage and skin. Fractures. Partial penetration.

Fig. 5-7 Infinite Acting Radial Flow Regime.

An apparent radius of drainage, rd, can be defined beyond which no pressure drop is measurable. (This is, however, dependent on the resolution of the pressure gauge used and emphasizes the importance of gauge selection when analyzing well test data.) Within the drainage area the pressure drops. The minimum pressure is in the wellbore. The pressure front, defined by rd, propagates radially. This configuration is called Infinite acting radial flow regime. It represents the response from the reservoir as if the pressure wave was spreading radially through an infinite extent and the pressure response will not change (assuming the resolution of the gauge has been reached) unless some outer boundary features are encountered.

5.2.3.1 Wellbore Storage When a well is opened, initially the production at surface is due only to the decompression of fluid in the wellbore. The reservoir contribution is negligible. This is the wellbore storage effect. p (t)

Fig. 5-9 Model of Wellbore Storage Effect.

The plot below gives a simplified description of the pressure distribution versus the distance, r, from the well;

p

p

p

q (t)

r i

p

r i

SKIN, s = 0 rd pressure profile

Fig. 5-8 Apparent Radius of Drainage.

Introduction to Well Testing (Aug 1996)

pressure profile

Fig. 5-10 Simplified Pressure Distribution.

5-7

Section 5

Basic Well Test Interpretation

The reservoir is still at initial pressure, p i.

p

Only the pressure in the wellbore has dropped. The wellbore storage effect prevents the sand-face flow rate from instantaneously following the surface flow rate.

p

rw

r

i

∆p s

SKIN, s > 0

pressure profile

FLOW RATE

Fig. 5-12 Pressure Profile for a Damaged Well.

surface flow rate sand-face flow rate ∆t elapsed time from opening of well

Fig. 5-11 Wellbore Storage Effect - Pressure Profile.

Similarly when the well is shut in, the reservoir will continue to deliver fluid into the wellbore as the pressure at the shut-in point builds up due to the fluids compressing and the height of fluid in the wellbore. Down hole shut in valves are used during well testing to significantly reduce the effects of wellbore storage during build-up periods. 5.2.3.2 Skin The communication between the well and the reservoir is affected by:

This causes an additional pressure drop around the wellbore which is quantified by a skin factor, s. For a damaged well , s > 0 After stimulation, the ability to flow into the wellbore is improved. Thus the pressure drop measured in the well is smaller. This is seen as a reduction in the skin factor which may even become negative. For a stimulated well , s < 0 In the literature, the concept of apparent wellbore radius, rwa, has been used to represent the skin. p

p

rw rwa

r

i

∆p

s

SKIN, s < 0

• Presence of mud cake. • Invasion of drilling fluids. • Insufficient perforation density. • Partial penetration.

Introduction to Well Testing (Aug 1996)

pressure profile

Fig. 5-13 Apparent Wellbore Radius.

5.2.3.3 Other Flow Regimes Sometimes infinite acting radial flow is not established immediately after the end of the wellbore storage flow regime. This can be the case if:

5-8

Schlumberger

• The well is fractured. • The well partially penetrates the reservoir.

• The reservoir is fissured.

5.2.4 Outer Boundary Conditions In a test of sufficiently long duration, another flow regime will occur at late time due to the presence of outer boundaries, such as;

• No flow outer boundary (sealing faults…).

• The reservoir is multi-layered.

It may take some time before the equivalent homogeneous behavior of the whole system is reached. In a fissured reservoir, the fissure response is much faster than the matrix response. During the first period of production only the fissure system produces, the matrix blocks still being at initial pressure. Thus two pressure profiles exist in the reservoir. pf for the fissure system.

• Constant pressure outer boundary (water drive…).

• Closed system. For example, if there is a sealing fault near the well, after rd has reached the fault, the shape of the pressure profile is changed.

pm for the matrix blocks.

Once the matrix blocks start to produce into the fissures, pm, drops from pi to p f. When both systems (fissures and matrix blocks) have the same pressure the behavior appears homogeneous See Fig. 5-14). p (t)

p (t)

q (t)

sealing fault

q (t)

Fig. 5-15 Model of Sealing Fault.

p r

pi

rd pressure profile in the matrix pressure profile in the fissures

Fig. 5-14 Fissured Reservoir.

Introduction to Well Testing (Aug 1996)

5-9

Section 5

Basic Well Test Interpretation

p

TIME t1 pressure front has not reached the fault

p

i

TIME t2 > t1 pressure front has reached the fault and is reflected but no effect is seen at the wellbore pressure profile without fault

- Impermeable Upper and Lower Boundaries - Uniform Initial Pressure • Outer Boundary Conditions - No Flow - Constant Pressure 5.2.6

Various Phases during a Well Test The various phases occurring during a well test are illustrated on a schematic reservoir map.

p

p

i

pressure profile without fault reflected pressure profile

wellbore storage

actual pressure profile fracture flow radial flow (infinite acting reservoir)

p

TIME t3 > t2

boundary effect

pressure reflection has reached the wellbore, the effect of the fault is seen in the measured signal

p

i

pressure profile without fault reflected pressure profile actual pressure profile

Fig. 5-16 Pressure Profile for Sealing Fault.

Fig. 5-17 Schematic Reservoir Map.

5.2.5 The Complete Model The complete configuration of an interpretation model would therefore be constructed in a similar manner to the following; • Inner Boundary Conditions

The example is a fractured well located near a sealing fault. First the wellbore storage acts and there is no pressure change in the reservoir. Then the reservoir starts to produce. Initially the flow is linear and normal to the fracture. As the area of drainage expands, the anisotropy due to the fracture disappears. Radial flow is established and the fracture is only seen as a negative skin factor. Eventually the pressure front is reflected back by the sealing fault. When the reflection reaches the well, an additional pressure drop will be observed indicating the presence of a boundary.

- Wellbore Storage - Skin - Fractures - Partial Penetration • Basic Model - Infinite Lateral Extent

Introduction to Well Testing (Aug 1996)

5-10

Schlumberger

Model Recognition

As a starting point it is required to identify a theoretical reservoir model with pressure trends that resemble those observed in the acquired data. It is important to note at this stage that before thorough and accurate interpretation can be performed, the data should be validated. This procedure must be performed at the wellsite and preferably in real time. The techniques involved require a knowledge of interpretation similar to those described in this section. 5.3.1 Log-Log Scale To select the appropriate theoretical model and identify the flow regimes the pressure data is analysed. This is greatly simplified by a knowledge of straight line pressure derivative response trends associated with the formation flow geometry. The change in pressure during the test, Dp, and the pressure derivative, Dp'Dt, are plotted versus time, Dt, on a log-log scale. This scale is chosen over other types of scales such as straight Cartesian plots or semi-log plots as the mathematical representations highlight the fundamental characteristic shape of a response without being distorted by the effect of its magnitude. In addition the derivative value has proven to be even more sensitive and highlights other features which would previously not have shown on semi-log or Cartesian scales. The derivative value is obtained by taking the slope at the point required of the test data plotted on a pressure versus superposition time plot. The algorithm takes a point before (left) and a point after (right) the point (B) considered, calculates the two corresponding derivatives and places their weighted mean at the point of interest. Data may be noisy, so the two points are normally chosen to be adequately distant Introduction to Well Testing (Aug 1996)

from the point considered, in order to smooth the scatter in the differentiated data. Attempts to over smooth noisy data may distort the actual pressure response. Smoothing may be increased up to the point when distortion starts. Y Y2

B Y1

X2 X1

X

dY

()

Y1 X2 X1

dX B

+

Y2 X1 X2

X1 + X 2

Fig. 5-18 Example of Derivative Mathematics.

5.3.2 Time Periods The change in pressure versus elapsed time is plotted on a log-log graph for an idealised drawdown test. Four different time periods can be identified. 1 C

LOG Dp & LOG Dp' Dt

5.3

2 x f, h d, w, l , k

3 kh, s, p*

4 r, A, p

no flow boundary

wellbore storage

fracture partial penetration fissures multilayers

radial homogeneous flow

constant pressure boundary closed system

LOG Dt

Fig. 5-19 Time Periods.

Period 4 includes late time data, where boundary effects are predominate. This period was the first to be investigated by well testing, in the 1920's and 1930's. 5-11

Section 5 Production wells were shut in at regular intervals and spot downhole pressure measurements were taken to obtain the average reservoir pressure. A material balance was then used to estimate the reserves. As all closed systems have the same pseudo-steady state behaviour, no other information can be extracted from the data. It was then realised that the validity of the spot pressure measurements was dependent upon the duration of the shut-in period. The less permeable the formation, the longer the shut-in period necessary to reach average reservoir pressure. Transient pressure testing was thus developed in the 1950's and 1960's with the classical work of Horner and Miller, Dyes and Hutchinson (MDH). This corresponds to period 3 in Figure 5-19. Data from period 3 are analysed to obtain;

• Permeability - thickness product, kh. • Skin, s. • Extrapolated pressure, p*.

The kh and s values from period 3 represent only a gross reservoir behaviour, and cannot be used to describe the system in greater detail. The kh value could represent a homogeneous, a multi-layered, or a fissured reservoir. In the same way, a positive skin could indicate either a damaged well, or an undamaged well with partial penetration; and a negative skin characterises a stimulated well (as a result of acidizing or hydraulic fracturing) or a well in a fissured reservoir.

Detailed information is only obtained from period 2 and has been the subject of many recent developments. It describes the specific flow characteristics of the system before the gross radial homogeneous behaviour is reached. If period 2 is present, parameters detailing the well and reservoir structure are then accessible; Introduction to Well Testing (Aug 1996)

Basic Well Test Interpretation

• • • •

Partial penetration. Fracture. Natural fissures. Multi-layers.

Period 1 always appears first. It is the wellbore storage effect. It may obscure the presence of period 2 data. The duration of period 1, proportional to the wellbore volume, can be advantageously reduced with downhole shut-in tools. The test duration is then shortened as the subsequent periods, which give parameters of interest are reached earlier. 5.3.3 Flow Regime Identification Identifying flow regimes, which show up as characteristic patterns displayed by the pressure derivative data, is important because each regime is a streamlined geometry for flow in the formation being tested. Thus for each flow regime identified, there is a set of well and/or reservoir parameters that can be computed, using only that portion of the transient data exhibiting the characteristic pattern behaviour. Presently there are only eight flow regime patterns commonly observed in well test data. These are;

• • • • • • • •

Radial. Spherical. Linear. Bilinear. Compression/Expansion. Steady State. Dual Porosity/Permeability. Slope Doubling.

A commonly used tool for discriminating between these eight identifiable subsurface flow regimes on log-log plots, called the Flow Regime Identification Tool (FRID) is useful for understanding downhole and reservoir conditions.

5-12

st at e

Schlumberger

Ps eu do st ea dy

n) w do near w Li ra rd f( o

Radial Sp her ica l

W el lb or e

st or ag e

Radial

ear Lin ar Biline

Radial

ear Lin

Fig. 5-20 Flow Regime Identification Tool - FRID.

5.3.3.1 Radial Flow The most important flow regime is radial flow, which is recognised as an extended constant or flat trend in the derivative. Radial flow geometry is described as streamlines converging to a circular cylinder. With fully completed wells, the cylinder may represent the portion of the wellbore intersecting the entire formation as in Figure 521(b). With partially penetrated formations or partially completed wells, the radial flow may be restricted in early time to only the fraction of the formation thickness where there is flow directly into the wellbore, Figure 5-21(a). When wells have been stimulated, Figure 5-21(c) or horizontally completed, Figure 5-21(e), the effective radius for the radial flow may be enlarged. Horizontal wells may also exhibit early time radial flow in the vertical plane normal to the well, Figure 5-21(d). Finally, if the well is located near a barrier to flow, such as a fault, the pressure transient response may exhibit radial flow to the well, followed by radial flow to the well plus its image across the boundary, Figure 521(f).

5.3.3.2 Spherical Flow Spherical flow occurs when flow streamlines converge to a point. This flow regime occurs for the partially completed well in Figure 523 (a) and the partially penetrated formation in Figure 5-23(b). For partial completion or partial penetration near the upper or lower bed boundary, the nearer bed imposes a hemispherical flow regime. Both spherical and hemispherical flow are seen on the derivative as a negative half slope. Whenever this appears, the spherical permeability can be determined, which in turn yields a value or the vertical permeability, kv, if the horizontal permeability, kh, can be quantified from a radial flow regime seen in another portion of the data. The importance of vertical permeability for predicting gas, water coning or horizontal well performance has highlighted the practical need for quantifying this parameter. A drillstem test conducted when only a small portion of the formation has been drilled (or perforated) may yield both vertical and horizontal permeabilities, thus enabling optimised completion engineering or providing a rationale as to whether a horizontal well should be drilled.

Whenever radial flow occurs, values for permeability and skin can be determined; when it occurs in late time, the average reservoir pressure can also be computed (classical Horner analysis).

Introduction to Well Testing (Aug 1996)

5-13

Section 5

Basic Well Test Interpretation

(b) Complete radial flow

(a) Partial radial flow

Top of zone

(c) Pseudoradial flow to fracture

Fracture

Bottom of zone

(d) Radial flow to horizontal well

(e) Pseudoradial flow to horizontal well

Fracture boundary

(f) Pseudoradial flow to well near sealing fault

Fig. 5-21 Different Types of Radial Flow Regimes.

Actual well

Image well

10 3

Pressure and pressure derivative (psi)

Well A, radial flow

10 2

Radial

101

= Pressure = Derivative

10 0 10–2

10–1

10 0

101

102

Fig. 5-22 Radial Flow Observed at late times. Elapsed time (hr)

(a) Spherical flow to partially completed zone

(b) Hemispherical flow to partially penetrated zone

Fig. 5-23 Spherical Flow Regimes. Introduction to Well Testing (Aug 1996)

5-14

Schlumberger (a)

10 3

I

Pressure and pressure derivative (psi)

Well B, single layer flowing

= Pressure = Derivative

10 2

101 I Radial

Sp her ica l

100

10 –1

10–2 10 –5

10–4

10–3

10–2 10–1 Elapsed time (hr)

100

101

(b)

10 3

I

Pressure and pressure derivative (psi)

Well B, two layers flowing II

10 2

= Pressure = Derivative

10 2

101

100 I and II Radial 10 –1

10–2 10 –5

10–4

10–3

10–2

10–1

100

101

10 2

Elapsed time (hr)

Fig. 5-24 Data Set Showing Spherical Flow Regime Indicated By The Negative Half Slope Trend. (Red Line in Top Plot).

Figure 5-24 is an example of a drillstem test that yielded vertical and horizontal permeabilities for the lower layer. These permeabilities were derived from the portion of the data exhibiting the spherical flow regime (negative half slope) trend shown by the red line in Figure 5-24(a). The reason for the spherical flow in early time is evident from the open hole logs, from the same well, which shows only a few feet of perforations into the middle of the lower layer.

Introduction to Well Testing (Aug 1996)

Frequently, negative half slope behaviour is observed in well tests that indicate a high skin factor. A complete analysis in this case may provide vertical permeability and decomposition of the skin into components that indicate how much of the skin is due to the limited entry, and how much to damage along the actively flowing interval. In turn, the treatable portion of the damage can be determined, and the cost effectiveness of the damage removal and/or re-perforating to improve well productivity can be evaluated.

5-15

Section 5

Basic Well Test Interpretation tion of the streamlines, and the flow area normal to the streamlines.

Effective Porosity

Oil 100

0

Corrected Core Porosity

Moved Hydrocarbons Depth (ft)

100

0

(a) Fracture linear flow

Shale Volume

Water 0

(b) Linear flow to fracture Fracture Fracture boundary

100

12,400

(c) Linear flow to horizontal well

Fracture

(d) Linear flow to well in elongated reservoir

II

12,425

Fig. 5-26 Linear Flow Regimes.

Perforations

12,450

I

12,475

OWC

12,500

Fig. 5-25 Open Hole Log Showing a Partially Completed Interval.

5.3.3.3 Linear Flow Linear flow streamline geometry consists of strictly parallel flow vectors. Linear flow is exhibited in the derivative as a positive half slope. Figure 5-26 shows why this flow regime is evident in vertically fractured and horizontal wells, or in a well producing from an elongated reservoir. Since the streamlines converge to a plane, the parameters associated with this flow regime are the formation permeability in the direcIntroduction to Well Testing (Aug 1996)

When formation permeability thickness is known from another flow regime, the width of the flow area can be determined. This provides the fracture half length of a vertically fractured well, the effective production length of a horizontal well, or the width of an elongated reservoir. The combination of linear flow data with radial flow data (in any order) can provide the principle values of kx and kv for directional permeabilities in the bedding plane. In an anisotropic formation, the productivity of a horizontal well is enhanced by drilling the well in the direction normal to the maximum horizontal permeability. Figure 5-27 shows a water injection well exhibiting linear flow. Although no radial flow is evident, the time of the departure from linear flow, as indicated on the figure, coupled with an analysis of the data that follows the half slope derivative trend, provides two independent indicators of both formation permeability and fracture half length, permitting quantification of both. The subtle rise in the derivative after the departure from linear flow suggests a boundary, which was interpreted as a fault. 5-16

Schlumberger

Pressure and pressure derivative (psi)

10 2 Well C, flow to vertical fracture

101 ear Lin

End of linear flow

10

0

10 –1 10–4

= Pressure = Derivative 10–3

10–2 10–1 Elapsed time (hr)

10 0

101

Fig. 5-27 Test Data Showing Linear Flow.

5.3.3.4 Bilinear Flow Hydraulically fractured wells may exhibit bilinear flow instead of, or in addition to, linear flow. This flow regime occurs because a pressure drop in the fracture itself accounts for parallel streamlines in the fracture, while at the same time, the streamlines in the formation become parallel as they converge to the fracture. Since the two linear flow patterns occur simultaneously in normal directions, this flow regime is termed bilinear. The derivative trend for this flow regime has a positive quarter slope. When the fracture half length and the formation permeability are known independently, the fracture conductivity, kwf, can be determined from this flow regime.

Fig. 5-28 Bilinear Flow Regime.

5.3.3.5 Compression / Expansion The derivative of a compression / expansion flow regime appears as a unit slope trend whenever the volume containing the pressure disturbance is not changing with time, and pressures at all points within this volume vary in the same way. This volume can be Introduction to Well Testing (Aug 1996)

limited by a portion or all of the wellbore, a bounded commingled zone, or a bounded reservoir. When the wellbore is the limiting factor, the flow regime is called wellbore storage, and when the limiting factor is the entire drainage volume for the well, this behaviour is called pseudosteady state. One or more unit slope trends preceding a stabilised radial flow derivative may represent wellbore storage effects. The transition from the wellbore storage unit slope trend to another flow regime usually appears as a hump. The wellbore storage flow regime represents a response that is effectively limited to the wellbore volume. Hence it provides very little information about the reservoir. Furthermore, a predominance of wellbore storage may mask important early time responses that would otherwise characterise near wellbore features, including partial penetration or a finite damage radius. This flow regime is minimised by shutting in the well near the production interval. This practice can reduce the portion of the data dominated by wellbore storage behaviour by two or more logarithmic cycles in time. In some wells tested without downhole shut-in, wellbore storage effects have lasted up to several days. After radial flow has occurred, a unit slope trend that is not the final observed behaviour may occur because of production from one zone into one or more other zones commingled in the wellbore, or vice versa. This behaviour is accompanied by cross flow in the wellbore, and occurs when commingled zones are differentially depleted. When the unit slope occurs as the last observed trend, as in Figure 5-29(a) the assumption is that this represents pseudosteady state conditions for the entire reservoir volume contained in the well drainage area. The late time unit slope behaviour due to pseudosteady state is ob5-17

Section 5

Basic Well Test Interpretation

served only during drawdown. When the unit slope is seen after radial flow, either the zone (or reservoir) volume or its shape can be determined. (a)

10 3

10 2 Buildup

Pseudosteady state

101

Radial

Drawdown

Wellbore storage hump

Buildup

10 0

10 –1 10– 4

10–3

10–2

10–1 Elapsed time (hr)

100

101

10 2

(b)

10 3

Pressure and pressure derivative (psi)

Steady state

10 2

(a)

10 3 Dual porosity

101

Steady state

Radial

Buildup

Wellbore storage hump Drawdown 10 0

10 –1 10– 4

10–3

10–2

10–1 Elapsed time (hr)

100

101

10 2

Pressure and pressure derivative (psi)

Pressure and pressure derivative (psi)

Psuedosteady state

5.3.3.7 Dual Porosity / Permeability Dual porosity / permeability behaviour occurs when reservoir rocks contain distributed internal heterogenities that have highly contrasting flow characteristics. Examples are naturally fractured or highly laminated formations. the derivative behaviour for this case may look like the valley shaped trend shown in Figure 5-30(a) or it may resemble the behaviour shown in Figure 5-30(b). This feature may come and go, during any one of the flow regimes already described, or during transition from one flow regime to another. From this flow regime, parameters associated with internal heterogeneity are determined, such as interporosity flow transmissibility, relative storativity of the contrasted heterogeneities, or geometric factors.

Fig. 5-29 Flow Regime Trends Exhibited by Wellbore Storage, Bounderies and Pressure Maintenance.

Radial: fractures

Radial: total system

101 Wellbore storage hump

Dual porosity valley

10 0

10–1 10–4

10–3

10–2

10–1 Elapsed time (hr)

10 0

101

10 2

(b)

10 3 Dual porosity or dual permeability

Pressure and pressure derivative (psi)

5.3.3.6 Steady State Steady state implies that pressures in the well drainage volume are not varying in time at any point, and that the pressure gradient between any two points in the reservoir is constant. This condition may occur for wells in an injection / production scheme. In buildup and fall-off tests, a steeply fallingderivative may represent either pseudosteady state or steady state.

10 2

10 2

Radial: total system 101 Radial: fractures Wellbore storage hump

Dual porosity transition

10 0

10–1 10–4

10–3

10–2

10–1 Elapsed time (hr)

10 0

101

10 2

Fig. 5-30 Characteristics Patterns Exhibited by Naturally Fractured and Highly Laminated Formations. Introduction to Well Testing (Aug 1996)

5-18

Schlumberger

5.3.3.8 Slope Doubling Slope doubling describes a succession of two radial flow regimes, with the second being at a level exactly twice that of the first. This behaviour is frequently explained by a sealing fault, but because of the similarity between figures 5-30(b) and 5-31, it can also be due to dual porosity / permeability heterogeneity, particularly in laminated reservoirs. When slope doubling is identified with a sealing fault, the distance from the well to the fault can be determined.

tions. Parameters estimated from a specialised plot may be used as a starting value for computerised refinement of the model for the transient response in the second interpretation stage. The following examples illustrate derivative responses with their specialised plots; 5.3.4.1 Wellbore Storage A unit-slope log-log straight line passing through early time Dp and Dp'Dtdata is usually indicative of wellbore storage.

10 3

Dp Dp' Dt LOG Dp & LOG Dp' Dt

Pressure and pressure derivative (psi)

Single sealing fault

10 2

Radial: single fault Radial: infinite-acting

101 Wellbore storage hump

Slope doubling transition

Wellbore Storage

10 0

LOG Dt 10 –1 10–4

10–3

10–2

10–1

100

101

10 2

Elapsed time (hr)

Fig. 5-31 Slope Doubling Caused by a Succession of Two Radial Flow Regimes (Sealing Fault).

5.3.4 Specialised Plots The dominant geometry for the flow streamlines in the formation determines which of the flow regimes patterns appears in the pressure transient response at any given time. The presence of one or more of the recognised derivative patterns marks the need to select a model that accounts for the implied flow regimes. Moreover, each of several easily recognised derivative trends has a specialised plot that is used to estimate parameters associated with that trend. In fact, the specialised plot for each straight derivative trend is a plot of the pressure change versus the elapsed time, raised to the same power as the slope of the derivative line on the log-log plot. Slopes and intercepts of these plots provide equations for parameter computaIntroduction to Well Testing (Aug 1996)

Fig. 5-32 Wellbore Storage.

Because Dp is proportional to Dt, the same data points must also be located on a straight line passing through the origin when Dp is plotted versus Dt in Cartesian co-ordinates. Such a plot, specific to a given flow regime, is called a specialised plot (See Fig. 5-33). 5.3.4.2 High Conductivity Fracture A high conductivity fracture communicating with the wellbore yields a log-log straight line with half slope (Dp is proportional to Dt ). (See Fig. 5-34). The specialised plot in this case is a plot of Dp versus Dt , which yields a straight line through the same points (See Fig. 5-35).

5-19

Section 5

Basic Well Test Interpretation The corresponding specialised plot is a semilog plot of Dp versus log Dt. Wellbore Storage

LOG Dp & LOG Dp' Dt

Dp

Dp Dp' Dt

0

Homogeneous System

Dt

0

Fig. 5-33 Wellbore Storage Cartesian Coordinates. Dp LOG Dp & LOG Dp' Dt

Dp' Dt

LOG Dt

Fig. 5-36 Infinite Acting Flow.

5.3.5 The Complete System The log-log behaviour of a complete model is simply obtained as the superposition of the individual log-log behaviour of each component of the model.

High Conductivity Fracture Flow

LOG Dt

Fig. 5-34 High Conductivity Fracture.

The illustrations below show how model components have been superimposed to give three typical examples: Well A Well with wellbore storage / Closed homogeneous reservoir.

Dp

High Conductivity Vertical Fracture

0 0

Dt

Well B Well with wellbore storage and a low conductivity fracture / Closed homogeneous reservoir. Well C Well with wellbore storage / Heterogeneous reservoir of infinite lateral extent.

Fig. 5-35 Specialised Plot for High Conductivity Fracture.

During infinite acting radial flow, Dp is a linear function of the logarithm of Dt. The value of the pressure derivative is constant (Horizontal Line) and equal to 0.5 (when scales are in dimensionless parameters - explained later). Introduction to Well Testing (Aug 1996)

LOG Dp & LOG Dp' Dt

5.3.4.3 Infinite Acting Radial Flow

Dp Dp' Dt Radial Flow

Closed System

Wellbore Storage

LOG Dt

Fig. 5-37 Well A. 5-20

Schlumberger

LOG ∆p & LOG ∆p' ∆t

∆p ∆p' ∆t

Radial Flow

Closed System

Fracture Flow Wellbore Storage

LOG ∆t

Fig. 5-38 Well B.

LOG ∆p & LOG ∆p' ∆t

∆p ∆p' ∆t

Transition Period

Radial Flow

ting in the well at the surface can mask important flow regime trends. Furthermore, late time trends frequently are distorted by superposition effects that could be minimised with adjustments in the test sequence or by inadequate pressure gauge resolution that could have been avoided with selection of a more sensitive gauge. Missing or incomplete late time trends may result from premature test termination that would have been avoided with real time surface acquisition and on site data validation. Even for well designed tests, flow regimes may be difficult to discern, but this is relatively rare. 5.4

Wellbore Storage

LOG ∆t

Fig. 5-39 Well C.

5.3.6

Additional Information for Model recognition Reservoir information collected from geoscientists helps in the selection of a reservoir model. Occasionally, the distinctions among various model options consistent with the transient test data are not clear cut, and two or more models may give similar responses. In these cases the analyst may rule out most model options by consulting colleagues working with other independent data. When the flow regime responses are poorly developed or non-existent, interdisciplinary discussion may suggest selection of an appropriate model and reasonable starting values for the parameter estimation stage of the interpretation. Often, flow regime responses may be difficult to recognise because of a problem or procedure that could have been addressed before starting the test. This underscores the need for careful test design. For example, excessive wellbore storage resulting from shutIntroduction to Well Testing (Aug 1996)

Parameter Estimation

Once the reservoir model has been identified, it is necessary to compute the model parameters. This process essentially involves matching the acquired data response to that of the best fitting theoretical model and thus the key parameters of this best fit theoretical model present a best approximation solution to the actual reservoir. Today this is done with the aid of computer processing as the equations and their solution require powerful iteration techniques which if done manually would take an inordinate amount of time. 5.4.1 Dimensionless Groups The theoretical model responses are represented in terms of dimensionless groups to simplify their solution, the most common forms are;

• Dimensionless Pressure, pD • Dimensionless Time, tD These dimensionless groups define universal pressure and time scales independent of the actual magnitude of the physical parameters involved (q, kh, ø…). During a test, for example, the system response would be doubled if a given drawdown were to be repeated 5-21

Section 5

Basic Well Test Interpretation

at twice the original flow rate. The values of p D and tD , however, remain unchanged. They characterise the system behaviour.

Dimensionless groups are combinations of certain system variables. They are used to vastly simplify the solutions to the partial differential equations which are mapped graphically for given theoretical models and displayed as type curves. Other groups may also be defined as deemed necessary. The dimensionless pressure is usually written as; p D =A Dp - where A is a function ƒ(kh, m, …) The dimensionless time is usually written as; tD = B Dt

- where B is a function g(k, ø, ct, …) In log-log co-ordinates; log pD = logADp

= log Dp + log A

log tD = logBDt

= log Dt + log B

Consequently, when the proper model is being used, real and theoretical pressure versus time curves are identical in shape, but translated one with respect to the other when plotted on the same log-log graph. The translation factors for both pressure and time axes, are proportional to the well and reservoir parameters. They can, therefore, be used to compute these parameters. 5.4.2 Type Curves The model responses plotted in dimensionless form on log-log scale are know as “Type Curves”. Type curves are designed to give a Introduction to Well Testing (Aug 1996)

global description of the pressure response. A theoretical interpretation model can only be used if the corresponding log-log curve matches all the data from very early time to the last recorded point. The various components participating in the pressure response are acting at different times in a well defined order. Log-log diagnosis allows each successive flow regime to be identified. The specialised analysis relevant to each flow regime can then be performed. With the introduction of sophisticated interpretation models, log-log diagnosis has become a powerful tool and is now considered essential in interpretation. For a given theoretical model, not all type curves are equivalent. The ease with which a given type curve can be used is dependent on the choice of dimensionless groups. The various flow regimes must be clearly indicated, with limits computed from realistic approximation criteria, so that appropriate specialised analysis methods can be applied to the corresponding test data. This last point is particularly important, as specialised analysis methods, which use the slope of a straight line on the specialised plots, usually provide more accurate results than quantitative log-log analysis. 5.4.2.1 Type Curve Matching The steps involved in manual type curve matching are outlined below; 1. Select the appropriate type curve suitable to the identified model. 2. Overlay the type curve with tracing paper. Using tracing paper guarantees that the plot of the data and the type curve have the same scale. 3. Trace the major grid lines. 5-22

Schlumberger

4. Label the axes.

5. Using the type curve grid showing through the tracing paper as a guide, plot the measured test data. Ignore the curves and scale on the type curve during this plotting phase; use only the base grid. 6. Now slide the tracing paper with the plotted test data, keeping the grids parallel, until the data points match one of the type curves. This step determines the translation factor.

7. After the match is completed pick a convenient “match point” on the data plot, such as an intersection of major grid lines. Record values at that point on the data plot and the corresponding values lying beneath that point on the type curve grid. This step measures the translation factor. Modern methods permit this step to be done directly with the computer. An understanding of the manual technique is a useful exercise in understanding the basic solution process. 5.4.3

Parameter Approximation from Type Curves

5.4.3.1 Basic Homogeneous Model To understand the basics of interpretation techniques it is relevant to start with the simplest model which is the basic homogeneous model. The assumptions for this model, which can be considered as the starting point from which more general and complex models are formed are as follows;

• The reservoir consists of homogeneous

porous media with constant thickness (h), permeability (k), porosity (φ) and is infinitely large.

Introduction to Well Testing (Aug 1996)

• The reservoir is fully saturated with a single phase fluid with constant viscosity and compressibility.

• The reservoir is fully penetrated with a well that has infinitesimally small radius (line source) rw→0.

• The well is flowed at constant rate and reservoir pressure initially is at pi.

Using Darcy’s law, it is possible to generate a basic equation for radial flow of an incompressible fluid in a homogeneous reservoir. Starting from this basic equation and incorporating the law of conservation of mass and the equation of state it can be shown that the pressure at any given time and radius is given by the following equation; p( r ,t ) = p i − where ; q = k = h = µ = P(r,t) = Pi = B = c = φ = Ei =

. qµB  1  − φµ c t r 2   1412   − Ei  kh  2  4(0.000264)kt  

Volume rate of flow, bbl/day. Permeability, md. Thickness, ft. Viscosity, centipoise Pressure at point r and time t, psi. Initial reservoir pressure, psi. Formation volume factor, dimensionless Compressibility, vol/vol/psi. Porosity, fraction. ∞e − u − x = du Ei ( ) ∫ u x (The exponential integral).

This equation is known as the “Line Source Solution” or “Exponential Integral Solution”. It is possible to solve this mathematically and plot it on a log-log scale to produce the Line Source Type Curve or Theis Type Curve. 5-23

Section 5

Basic Well Test Interpretation

rD =

φµ c t r w

2

r

1

-1

10

-1

10

1

2

10

10

10

3

4

10

2

t D /rD

Fig. 5-40 Basic Homogeneous Type Curve.

Now rearranging these equations 1 −r 2 p D = − Ei D 2 4t D

It can therefore be seen that there are two governing groups, namely; pD and tD/rD2 The above equation can be approximated to; pD ≈

D

tD /r D2 p D'

10-2

rw

⇒

p t D /rD2 pD'

0.0002637 k∆t

10

and

tD =

t D > 2.4. 2 rD

D

kh( p i − p w ) pD = 1412 . qµB

used when

p

In this form the equation is very difficult to use because of the many variables, therefore the following dimensionless groups are selected;

 1 t . log e D2 + 080907  2   rD

This equation is called the log approximation and is valid when t D > 100. 2 rD However; when t D2 > 10, the maximum error is only rD about 1%. when t D2 > 5, the maximum error is only rD about 2%. It is usually accepted that the solution can be Introduction to Well Testing (Aug 1996)

This type curve describes the pressure response at any point , r in the reservoir. By looking at the pressure response at a distance, r from the wellbore, the effects of the inner boundary conditions (skin and wellbore storage) can usually be neglected. If we now return to the Basic Homogeneous Model and the Theis Type Curve. From the pressure match, transmissibility (kh/µ) can be calculated as follows; p  kh = 1412 . qB D  match µ  ∆p  From the time match, storativity (φhct) can be calculated as follows; φh c t =

kh 00002637 . . µ r 2  t / r D2   D   ∆t  Match  

Type curves will always state the dimensionless groups which exist for the general model they cover and from these relationships actual first pass parameter estimation can be determined. 5-24

Schlumberger

5.4.3.2 Basic Homogeneous - Specialised Plot It can be shown that a solution to the line source equation can be expressed as follows; pw = pi −

  k∆t  141.2 qµB   − 7 .43049  log e   φµc t r2  2 kh  

changing from natural log to base 10, gives; pw = p i −

 162. 6 qµB   k  3 2275 − .  log 10 ∆t + log 10   kh  φµ c t r 2   

If we compare this to the equation of a straight line y = mx +c, it can be seen that a plot of pw versus log ∆t will yield a straight line, when the log approximation is valid, with slope m where m=

162.6 qµB kh

or rearranging for transmissibility kh 162.6 qB = . µ m This type of semi-log plot is called a specialised plot. Values obtained from the specialised plot should have good agreement with those obtained from the log-log plot. The Theis curve describes the response of one active well in a homogeneous reservoir. It is therefore used for the analysis of interference tests. 5.4.4 Superposition The principle of superposition as used in mathematics implies that if we have two wells draining the one reservoir and we would like to calculate the pressure drop at a point in the reservoir, we can calculate the pressure drop at that point due to each well as if it were the only one producing. We then add Introduction to Well Testing (Aug 1996)

the two pressure drops to obtain the combined effect. This is known as superposition in space and it enables us to handle multi well situations by generalising the approach to any number of wells. Another form of superposition, known as superposition in time, helps us handle rate variations. Assume that a well produced at rate q1 staring at t = 0 and after t = t1, the rate changed to q2 and we are interested in the pressure change at time t which is greater than t1. The superposition implies that we first calculate ∆p due to rate q1, as if it continued unchanged at time t. Then to account for the rate change we algebraically add the pressure change due to rate (q2 - q1) for the duration (t - t1). If q2 > q1 we add the change, if q2 < q1, then we subtract the change. A combination of superposition in space and time enables us to treat any number of wells with any arbitrary flow rate history for each well. Most type curves are generated on the assumption of a constant drawdown from the initial condition of constant reservoir pressure and a single step change of a single well producing from the reservoir. In reality this is seldom the case, as most transient periods have been preceded by some form of variable rate history. On an exploration well this could be a clean-up flow followed by a shutin period followed by a main flow period. In a production well, one can imagine the complexity of the previous flow rate history. The effect of these varying rates causes perturbations to the reservoir pressure and to ideally model for transient behaviour these effects need to be accounted for. Consider the following example; The pressure response during an ideal, simple test in an infinite reservoir is shown in Fig. 5-41.

5-25

PRESSURE

Section 5

Basic Well Test Interpretation

∆p

Dd

∆p

tp

BU

TIME

Fig. 5-41 Pressure Response for Ideal Simple Test.

The bottomhole pressure is initially at pi. During a time t p , the well is flowed at constant rate and the pressure drops. The well is then shut in for a pressure build-up. After an infinite shut-in time the pressure will be back to pi. In terms of pressure change ∆pBU (∆t=∞)

=

≤

∆pDd (∆t=tp)

∆pDd (∆t)

drawdown LOG ∆p

pD = pD[(∆t)D] + pD[(tp)D] - pD[(tp + ∆t)D]

Similarly the general equation drawdown type curve is; pD

therefore at the same elapsed time, ∆pBU (∆t)

be used, computed for the actual production time, t p. The same is true for drawdown type curves. We stated earlier that build-up analysis was preferred due to the requirement of achieving a steady state during drawdown and the problems associated with achieving this. To use a build-up analysis on drawdown data can however be achieved if certain criteria are met. Superposition in time is used to obtain a build-up solution from a constant draw down solution. Using this approach it can be shown that the general equation of a build-up type curve is:

buildup

tp LOG ∆t

Fig. 5-42 Ideal Responses.

The illustration shows the drawdown and build-up responses on the same log-log scale. The build up curve deviates from the drawdown response and tends asymptotically towards ∆pDd(∆t=tp) for large elapsed time. This means that for log-log analysis of build-up data, build-up type curves should Introduction to Well Testing (Aug 1996)

=

for

a

pD [(∆t)D ]

Thus a drawdown type curve can be used to analyse a build-up only if ; pD[(tp)D] - pD[(tp + ∆t)D] is negligible. or ∆t is small compared with t p or the drawdown period was significantly long as compared to the shut-in period. When using drawdown curves to match build-up data, errors in the match can easily be made if the correct production time is not considered, this is emphasised in the example above. There are also specific cases when drawdown curves can not be used, such as closed systems at late times. The weakness inherent in analysis with published type curves can be avoided by constructing curves that account for the effects of flow changes during the test itself.

5-26

Schlumberger

Improved computing techniques have facilitated the development of these custom madecurves (multi-rate type curves) resulting in a major advance in well test interpretation.

5.4.5.2 Skin Calculation The general equation for skin s, can be derived as follows;

5.4.5

pi− p

Semi-Log Analysis for Parameter Estimation Specialised plots, as mentioned earlier are used to confirm the trends of the model identified through log-log analysis. In addition different models and or features will yield parameter estimates from the specialised plots. The analyst needs to become familiar with the various specialised plots and their features associated with the different models, although these are included by default in modern interpretation software models. 5.4.5.1 Horner (or Superposition) Method When the log-log analysis indicates that the radial flow configuration has been reached, the build-up and semilog analysis can then be performed. There are several methods available for doing this, the most common being the Horner method. Using superposition, the build-up equation can be written as:

p ws (∆t ) = p i −

t p + ∆t 162.6 qµB log10 kh t

This implies that a plot of the shut-in pressure p ws (∆t) at any shut-in time ∆t versus

(t+∆t)/∆t on semi-log paper should give a straight line through the infinite acting radial flow data. The straight line has a slope m and intercept p* at (tp+∆t)/∆t = 1. The slope can be used to calculate kh: kh =

ws (

  k ∆t = 0) = m log 10 t p + log 10 − 323 s . + 08686 . 2   φµ ct r w

It can be shown that by combining the buildup equations with the above general equation for skin and taking ∆t = 1 hour, then; p tp + 1 − p wf k  s = 11513 .  1hour − log 10 + 3.23 + log 10 m t p   φµc t r w 2

In the case that t p >> 1, then (tp+1)/ t p ≈ 1 and Log10 (1) = 0 and the equation simplifies to; p  − p wf k s = 11513 . − log 10 + .  1hour 3 23  2 m   φµct r w

5.4.6 Other Key Type Curves There are many published type curves covering a variety of models. As previously mentioned, modern techniques permit type curves to be tailor made to the model defined, however some key general type curves warrant discussion within the scope of this course. 5.4.6.1 Homogeneous Reservoir with Wellbore Storage and Skin This type curve was developed by Gringarten et al in 1979. It covers the basic homogeneous model but introduces the concept of wellbore storage and skin into the model.

162.6 qµB m

Introduction to Well Testing (Aug 1996)

5-27

Section 5

D

D

D

40 20

]

D

10 10

10

10 4

(

6 4 3 2

-1

pp

10 -1 10

-1

1

10

10

2

3

10

4

10

t /C

Fig. 5-43 Wellbore Storage and Skin. Reservoir with Homogeneous behavior.

There are four dimensionless groups defined as follows; kh∆p 1412 . qµB

pD = tD =

0.0002637 k∆t

CD = s

φµ c t r w 0.8936C

φ ct h r w

2

2

where C represent the wellbore storage and CD the dimensionless parameter for wellbore storage and s the skin. Three governing groups are; pD, tD/CD and Cde2s.

The shape of the curve is defined by the label Cde2s. If CD and s are changed, but the product CDe2s is kept constant, the curve is not changed. In addition;

0.000295 kh . C= µ t /C   D D  Match  ∆t   Introduction to Well Testing (Aug 1996)

Cde2s > 103 the well is damaged.

if

if 3 < Cde2s < 103 the well is not damaged. if Cde2s < 3 the well is stimulated. The period representing infinite acting radial flow can then be used for semi-log analysis as described earlier, using Horner. 5.4.6.2 Homogeneous Reservoir with Wellbore Storage and Infinite Conductivity Vertical Fracture This type curve was developed by Bourdet et al in 1985. 10 CDf

p

The type curve is used in a similar manner to that described in section 5.4.2.1 with the added point that the data is matched to one of the CDe2s curves. The following parameters can then be calculated; p  kh = 1412 . qBµ D  match .  ∆p  From the pressure match.

)

 C D e 2s Match . s = 05 . log e    CD   From the curve match.

D D

1

10

6 4 3 2

10 -1

10 APPROXIMATE ENDOF UNIT SLOPESTRAIGHT LINE

From the time match.

∆t tp 1

0.0 0.003 0.01 0.03 0.1 0.3

D

10

t /C p '

p '

D D

t /C

and

10

D

t Df p '

CDe 2s

p

1

and

2

D

10

Basic Well Test Interpretation

10 -1 p t

D

Df

p ' D

10 -2 10-4

10 -3

10 -2

t Df

10-1

1

10

Fig. 5-44 Wellbore Storage and Infinite Conductivity Vertical Fracture. Reservoir With Homogeneous behavior.

5-28

Schlumberger

t Df = C Df =

t /C p '

D

D

10 10

and

-6

10 10 -1 -2

D

λCD /1-ω 3x10 =

1

-1

-2

10 3x10

-2

-3

10 -4

3x10

102

10

-3

10 3

104

t D/CD

Fig. 5-45 Well with Wellbore Storage and Skin. Reservoir with Double Porosity Behavior.

0.0002637 k . φµ c t  t Df    Match  ∆t 

From the time match.

08936 . From the curve match.

10

10 λe -2s

10 -1

2

φ c t h x f 2(C Df )Match

40

-30

D

D

t /C

D

10

10 -1

p  kh = 1412 . qBµ D  match .  ∆p  From the pressure match.

C=

10

p

2

where xf represent the plane vertical fracture half length. Three governing groups are; p D , tDf and CDf. Quantitative log-log analysis yields;

xf =

D

λCD /ω (1-ω) 10 =

08936 . C

φ c t hx f

10

1

00002637 . k∆t φµ c t x f

CD e 2s

p

D

kh∆p pD = 1412 . qµB

10 2

p '

There are three dimensionless groups defined as follows;

.

5.4.6.3 Well with Wellbore Storage and Skin. Reservoir with Double Porosity Behaviour pseudosteady state interporosity flow This type curve was developed by Gringarten et al in 1980. It covers the basic model but with a heterogeneous matrix with double porosity and covers wellbore storage and skin.

Introduction to Well Testing (Aug 1996)

There are five dimensionless groups defined as follows; pD =

0.0002637 k∆t

tD =

φµ c t r w

CD = s

kh∆p 1412 . qµB 2

0.8936 C

φ ct h r w

2

λ = αr w2 km kf where λ is the interporosity flow parameter which characterises the ability of the matrix to flow into the fissures and α is a function of structural geometry such that; α=

12

h m2

for horizontal slab matrix blocks.

15

for spherical matrix blocks. r m2 Five governing groups are defined; α=

pD, tD/CD, Cde2s, ω and λe-2s.

5-29

Section 5

Basic Well Test Interpretation

ω is the storativity ratio which is in general related to the percentage of oil present in the fissures. ω=

(φVc t )f (φVc t )f + (φV c t )m

f denotes fissure and m matrix. There are three flow regimes modelled; 1) The fissures produce first, the pressurefollows one of the wellbore storage and skin type curves Cde2s.

( C D e 2s)f + m ω= (CD e 2s)f From the last Cde2s curve match.

λ e −2 s) (C D e 2s) ( f +m f +m λ= CD

From the λe-2s curve matched during the transition period.

C=

5.4.7 Parameter Refinement Once the reservoir model has been identified and the initial parameters calculated, the simulated and observed pressure responses usually differ slightly at this stage. Modern analysis, however, is assisted by nonlinear regression techniques which is a mathematical process that uses iteration to automatically fine tune the curve fit and therefore the parameters until the simulated model coincides with the observed data for the essential components of the transient response. It can readily be seen that the ease of this refinement process is very much dependent on the initial correct choice of the reservoir model. As such, the first stage model recognition represents the main challenge for the analyst. The following example emphasizes this point;

 C D e 2s Match  s = 05 . log e    CD  

Figure 5-46 shows a combined pressure / pressure derivative plot in Figure 5-46(a) and the specialized Horner plot in Figure 5-46(b). At first glance, these plots could be caused by five possible reservoir configurations or characteristics;

2) Transition period; the pressure follows a λe-2s curve. 3) The total system (fissures plus matrix) produces last; the pressure follows a lower wellbore storage and skin type curve (Cde2s)f+m. Quantitative log-log analysis yields; p  kh = 1412 . qBµ D  match .  ∆p  From the pressure match. 0.000295 kh . µ t /C   D D  Match  ∆t   From the time match.

(

)

From the curve match.

Introduction to Well Testing (Aug 1996)

5-30

Schlumberger

2. The well might lie between two faults, which are at right angles.

(a)

10 Pressure and pressure derivative (psi)

1. A single sealing fault might be indicated by the doubling of the slope in the generalized Horner plot.

1

0.1 1

3. The trough in the derivative plot may be the result of a dual porosity system.

10

100

1000

Elapsed time (hr)

(b) 6000

5. This may also be a composite system. The composite model was discarded since knowledge of the reservoir revealed that the existence of such a system was not feasible. Therefore, pressure derivative and pressure plots were computed by assuming the first four models.

5000 Pressure (psia)

4. This may be a dual permeability (twolayered) reservoir.

m 2 = 2m1 4000 m1 3000

2000

5

4

3

2

Log of (t + Dt)/Dt

1

0

Fig. 5-46 Pressure / pressure deriviate plot (a) and Horner plot (b) of measured data used in the model identification process.

The results are shown in figures 5-47(a) through (d). As suspected the single sealing fault, Figure 5-47(a), does not match the observed transient. Figure 5-47(b) represents the modeled results for a well located between two sealing faults, and again the observed and modeled data are in poor alignment. Figure 5-47(c) was derived assuming a dual porosity is a much better match than the two previous models, although it is still imperfect. Figure 5-47(d) highlights how the dual permeability or two layered reservoir model gives an extremely good fit with the observed pressure transient and derivative curves.

Introduction to Well Testing (Aug 1996)

5-31

Basic Well Test Interpretation

(a)

101 Well near a sealing fault

100

10–1 Pressure change Pressure derivative Multirate type curve 10–2 10–2

10

–1

10

0

1

10

10

2

10

3

Pressure and pressure derivative (psi)

Pressure and pressure derivative (psi)

Section 5

(b)

101 Dual porosity model (transient transition)

100

10–1

10–2 10–2

Pressure change Pressure derivative Multirate type curve 10–1

(c)

101 Dual porosity model (pseudosteady-state transition) 100

10–1 Pressure change Pressure derivative Multirate type curve 10–2 10–2

10–1

100

101

101

10 2

10 3

Elapsed time (hr)

10 2

10 3

Pressure and pressure derivative (psi)

Pressure and pressure derivative (psi)

Elapsed time (hr)

100

(d)

101 Dual permeability model

100

10–1 Pressure change Pressure derivative Multirate type curve 10–2 10–2

10–1

Elapsed time (hr)

100

101

10 2

10 3

Elapsed time (hr)

Fig. 5-47 Possible Solutions to Fig. 5-46.

Introduction to Well Testing (Aug 1996)

5-32

Schlumberger

5.5

Verification of Results

5.6

Several drawdown and build-up periods are often made during a well test, and it is common to interpret every transient and crosscheck the computed reservoir parameters. However, analysis of all the transients in a test is not always possible. In such situations, forward modeling may help confirm the validity of the reservoir model. Basically this involves simulating the entire series of build-ups and drawdowns using the selected reservoir model and parameters.

Pressure (psia)

4000

3000

Summary of Interpretation Methodology

The following basic steps summarises the basic interpretation process;

• Model Identification Model identification is performed by plotting real data as pressure change and pressure derivative versus elapsed time on log-log scale. Qualitative as well as quantitative information on the reservoir is obtained. Comparing the shapes of real and theoretical curves is essential for the selection of the most appropriate theoretical model, and for identifying the dominating flow regimes for which specialised analysis methods can be used.

• Parameter Estimation Measured Calculated 2000 0

100

200

300

400

Elapsed time (hr)

Fig. 5-48 Typical example of forward modeling in which the model is used to reproduce the entire test data set.

Because the simulation continues for much longer than an individual transient, the effects of reservoir boundaries are more likely to be noticed. If the simulation does not match the observed pattern, then the assumed model should be reconsidered. If, for example, an infinite acting reservoir model has been assumed from the analysis of a single transient, the forward modelling technique would show whether the model was correct. If the reservoir was assumed to be infinite acting when, in reality, it was closed, the simulation would not reveal realistic reservoir depletion.

Introduction to Well Testing (Aug 1996)

Quantitative information is obtained by matching the log-log plot of the test data against a type curve derived from a theoretical model that includes the various features identified in the actual data. The model not only allows the pertinent parameters to be calculated but also specifies their significance. For example, kh and s have a different significance in a homogeneous system than in a double porosity system. Specialised analysis is performed on the relevant data as defined from log-log diagnosis.

• Results Verification If an inconsistency is found in the checking procedures the whole process must be restarted.

5-33

Section 5

Basic Well Test Interpretation

RAW DATA

gas properties with pressure are accounted for by introducing the real gas pseudopressure function (from Al Hussainy et al., 1966). p

MODEL IDENTIFICATION log-log analysis

2p x dp p 0 µ(p ) x z(p)

m(p ) = ∫

and the real gas pseudotime function (from Agarwal, 1979) p

specialized analysis

WELL & RESERVOIR PARAMETERS

dt p 0 µ(p) x c t (p)

t a (p ) = ∫

then all the equations used for gas well testing analysis may be obtained from the liquid equations by replacing p with m(p) and t by ta(p). Consequently, all the liquid solutions can be applied, and the same techniques used for the analysis of oil well testing are applicable to gas well testing. 5.7.1

FINAL RESULTS Fig. 5-49 Interpretation Methodology - Flowchart.

There are two main differences between gas well testing and liquid well testing. First, because gas properties are highly pressure dependent, some of the assumptions implicit in liquid well testing theory are not applicable to gas flow. Second, high fluid velocities often occur near the wellbore and an additional pressure drop is caused by viscous inertial effects. This additional pressure drop is termed rate dependent skin. The variations of Introduction to Well Testing (Aug 1996)

/p

=

Gas Well Testing

µz

5.7

µz

µc t co ns ta nt

checking procedures

Simplifications to the Pseudofunctions Analysis based on pseudopressure may be used for all ranges of pressures. However, simplifications can be made for certain limits. Although these limits are approximate, apply to certain temperature ranges, and depend on gas properties, the following rule of thumb is usually valid. Refer to Figure 5-50(a).

µz = constant

(b)

(a)

2000

3000 Pressure (psia)

Fig. 5-50 Typical pressure dependency of the viscosity-real gas deviation factor product (a) and viscosity-total compressibility product (b). 5-34

Schlumberger

For pressures between 2500 and 3500 psi, no simplification can be made and use of m(p) is mandatory. For pressures over 3500 psi, the term mz/p is constant and m(p) is proportional to the pressure and the analysis can be performed using pressure instead of m(p).

Pressure and pressure derivative (psi)

101

100

10–1 Pressure change Pressure derivative Multirate type curve 10–2 10–2

10–1

100

101

10 2

10 3

Elapsed time (hr)

Fig. 5-51 Type-curve match for a gas Well Testing pressure data set (from Hegeman et al., 1993.

When pressure drawdowns are large, the changes in the product mct are important, see Figure 5-50(b), and pseudotime must be used. For small pressure variations, however, the effect of changing gas properties is minimal and real time may be used. For convenience, the pseudofunctions are normalised with reference to conditions at static reservoir pressure. Then, pseudopressure is expressed in dimensions of pressure, and pseudotime is expressed in units of time. Figure 5-51 shows a type curve matching for a gas well test, the log-log plot of the normalised pseudopressure variations versus normalised pseudotime changes is superimposed to p D versus tD/CD type curves, which include variable wellbore storage. The reservoir parameters are obtained the same way as for oil wells, but with the appropriate units and corresponding conversion factors.

Introduction to Well Testing (Aug 1996)

5.7.2 Multi-Point Well Testing The second problem posed by gas wells is addressed by multi-point well testing. In a conventional well, the additional pressure drop induced by high gas velocities, together with the one caused by formation damage, will show up as a high skin factor. To distinguish between these two effects, gas wells are usually tested with a sequence of increasing flow rates. Theoretically, two transients would be enough to separate these two skins, but in practise it is common to make a multipoint test. The skin factor is determined for each transient, and a plot similar to the one shown in Figure 5-52 yields the formation damage skin or true skin. Multi-point or back-pressure tests are conducted not only to estimate true skin but also to determine deliverability curves and absolute open flow potential (AOFP). The AOFP of a well is defined as the rate at which a well would produce at zero sandface pressure. Although this rate cannot be achieved, regulatory authorities use this to set the maximum allowable rates. Deliverability curves are used to predict flow rates against any particular back-pressure. For gas wells the relationship between rates and bottom hole pressures is given by the so called “back-pressure” equation;

(

q = C p2 - p2 ws wf

)

n

where C is called the performance coefficient and n is the inertial effect exponent. Deliverability curves can also be used for determining the number and location of wells in a given field, designing compressor requirements and establishing base performance curves for future comparisons.

5-35

Basic Well Test Interpretation

Measured skin, s¢

Section 5

an isochronal test, the well is flowed at four or more different rates for periods of equal duration. Between flow periods, the well is shut in until static conditions are reached. The last flowing period is extended until stabilised flowing conditions are reached (See Fig. 5-54).

Slope = D s

s t = s + Dq 0 Flow rate, q

Fig. 5-52 Measured skin versus flow rate in a multirate transient test.

5.7.3 Types of Gas Well Tests Back-pressure tests are commonly conducted with an increasing rate sequence. However, gas well test sequences vary according to stabilisation times. 5.7.3.1 Flow-After-Flow Tests High productivity formations are usually tested with a four point back-pressure test, commonly called flow-after-flow test. In such a test, the well is flowed at four different stabilised flow rates for periods of equal duration. At the end of each flow period, the rate is changed without closing the well (See Fig. 5-53). 5.7.3.2 Isochronal Tests Because stabilisation times can be too long in low productivity formations, an isochronal test is preferred to a flow-after-flow test. In

Introduction to Well Testing (Aug 1996)

5.7.3.3 Modified Isochronal Tests In practice, the true isochronal test is often replaced by a modified test sequence having flowing and shut in periods of equal duration. This modified test sequence is faster because it is not necessary to wait for stabilisation. Like the isochronal test, however, the last flowing period is extended until stabilisation is reached. this test is called the modified isochronal test. The results of back-pressure tests are conventionally presented as log-log plots of (p2ws-p2wf) versus flow rate. The resulting straight line is used to obtain the exponent n, which varies between 0.5 (high inertial effects) and 1 (negligible inertial effects). For isochronal or modified isochronal tests, the resulting curve is termed the transient deliverability curve. The stabilised curve is drawn through the extended data point using a line parallel to the transient deliverability curve. The modified isochronal test does not yield a true stabilised deliverability curve but rather a close approximation.

5-36

Schlumberger

pwi pwf1 pwf2

Bottomhole pressure

pwf3 pwf4

Cleanup

Initial shut-in

T

T

T

T

Final shut-in

Elapsed time (hr)

Q4 Q3 Gas flow rate

Q2 Q1

Elapsed time (hr)

Fig. 5-53 Schematic of Rate Sequence and Pressure Variations in a Flow-After-Flow Multipoint Test.

Introduction to Well Testing (Aug 1996)

5-37

Section 5

Basic Well Test Interpretation

pR pwf1

Bottomhole pressure

Initial shut-in

Cleanup

pwf2 pwf3 pwf4 T

T

Final shut-in

T

T

Elapsed time (hr)

Q4 Q3 Gas flow rate

Q2 Q1

Elapsed time (hr)

Fig. 5-54 Schematic of Rate Sequence and Pressure Variations in an Isochronal Multipoint Test.

pwi

pwf2

pwf1

Bottomhole pressure Initial shut-in

Cleanup

pwf3

T

T

T

T

T

T

T

pwf4

Final shut-in

Elapsed time (hr)

Q4 Q3

Gas flow rate

Q2 Q1 Elapsed time (hr)

Fig. 5-55 Schematic of Rate Sequence and Pressure Variations in a Modified Isochronal Multipoint Test.

Introduction to Well Testing (Aug 1996)

5-38

Schlumberger 1000 pR

100

(p 2ws – p2wf) ¥ 10–6 (psia)2

1⁄ n

Stabilized deliverability curve (p2R – p2wf)

10

Transient deliverability curve (p2ws – p2wf) 1

Q1 Q 2 Q 3 Q 4

AOF

0.1 10

100

1000

10,000

100,000

Flow rate (Mscf/D)

Fig. 5-56 Log-Log plot of Modified Isochronal Test Data.

5.8

Multiple Well Testing

In single well testing, the primary target is the nearby well region. However, when there is a need to investigate the inter-well region, more than one well is directly involved in the test. In multiple well testing, the flow rate is changed in one well and the pressure response is monitored in another. These tests are conducted to investigate the presence or lack of hydraulic communication within a reservoir region. In addition they are often used to estimate inter-well reservoir transmissivity and storativity. Examples of multiple well testing are interference tests and pulse tests. Vertical interference tests are sometimes classified as multiple well tests. These tests are conducted to investigate vertical communication and estimate vertical permeability, and are performed between two sets of perforations or test intervals within the same well. Multiple well tests are more sensitive to reservoir horizontal anisotropy than single well tests. Therefore, multiple well tests are often carried out to describe reservoir anisotropy - directional permeabilities.

Introduction to Well Testing (Aug 1996)

5.8.1 Interference Testing These tests require long duration production/injection rate changes in the active well. The associated pressure disturbance, which is recorded in the observation well, yields valuable information regarding the degree of hydraulic communication within the inter-well region When monophasic conditions prevail within the investigated region of the reservoir, the pressure response can be analysed to estimate inter-well reservoir properties. The analysis technique uses the same type curve matching approach as for drawdown tests, but with a different type curve. This is because, unlike single well tests, the pressure response is observed at some distance from the location where the perturbation was originally created. 5.8.2 Pulse Testing Pulse testing is a special form of multiple well testing which may last from a few hours to a few days. The technique uses a series of short rate perturbations at the active well. Pulses are created by alternating periods of production or injection and shut-in. The pressure response to these pulses is measured at one or more observation wells. Since the pulses are of short duration, the pressure responses are small. therefore, high resolution gauges are usually required to measure these small variations. The advantages of pulse testing compared with interference testing derive from the relatively short pulse length; the reservoir pressure trends and noise are automatically removed using the appropriate analysis techniques.

5-39

Section 5

Basic Well Test Interpretation

Observation well r

rw

Rate at active well

Active well rate = q

Dt t1

Bottomhole pressure

Elapsed time

Observation well

Established trend Time lag Dt t1 Elapsed time

Fig. 5-57 Interference Test.

The example in Fig. 5-58 and Fig. 5-59 illustrates how pulse testing was used to understand the degree of hydraulic communication within a Middle eastern reservoir and to check a suspected fluid migration toward a nearby field. The test involved six wells including the active well. The pulses were created by an alternating sequence of injection and shut-in periods of 36 hours each. The

Introduction to Well Testing (Aug 1996)

resulting pressure pulses were monitored in the observation wells for 12 days. The observed pressure responses were analysed with history matching techniques. The analytical solution of the diffusivity equation for a homogeneous rectangular reservoir with mixed boundary conditions yielded an excellent match between measured and simulated pressure response. The test indicated good hydraulic communication within the area of investigation. It was also possible to determine the inter-well reservoir properties and geometry of the area. Moreover, the fact that constant pressure boundaries were needed to match the data implied that there was no leakage toward the neighbouring field as previously suspected. The small amplitude of the signal detected in two of the observation wells suggested the presence of free gas in the upper part of the structure. this was confirmed by other sources of information. This knowledge was useful in locating future water injection wells and in managing the reservoir more efficiently. 5.9

Other Specialised Testing Types

There are many other types of testing techniques such as layered reservoir testing, horizontal well testing, impulse testing, measurement while perforating and so on and so forth. The techniques involved in analysing the data from these tests are often complex and require a solid grounding in interpretation techniques.

5-40

Schlumberger Pressure maintenance No-flow boundary

Mode led re servo ir area

C-4 C-5 C-1

C-8

C-3 C-7

Fig. 5-58 Schematic of the field showing the configuration of producing and injection wells. The yellow rectangle delineates the area modeled by the reservoir study (from Torre et al., 1993.

15 Observed pressure variation (psi) Simulated pressure variation (psi) Test rate sequence (10,000 BWPD)

Pressure and pressure derivative (psi)

13.5 12 10.5 9 7.5 6 4.5 3 1.5 0 0

30

60

90

120

150

180

210

240

270

Elapsed time (hr)

Fig. 5-59 Test sequence and corresponding observation well pressure response (from Torre et at., 1993). Introduction to Well Testing (Aug 1996)

5-41

Section 5

Introduction to Well Testing (Aug 1996)

Basic Well Test Interpretation

5-42

Section 6 Nodal Analysis

Section 6

Introduction to Well Testing (Aug 1996)

Nodal Analysis

6-2

Schlumberger

6.0

Nodal Systems - (Systems Analysis)

6.1 Introduction A nodal approach is presented for applying systems analysis to the complete well system from the outer boundary of the reservoir to the sand face, across the perforations and completion section to the tubing intake, up the tubing string including any restrictions and downhole safety valves, the surface choke, the flow line and separator. Using a combination of; 1. Well inflow performance.

2. Downhole multipurpose flow conduit performance (vertical or directional conduit performance).

3. Surface performance (including choke, horizontal or inclined flow performance and separator). The performance of either a naturally flowing or artificial lift well will be determined. The effect of various changes in one component of the system has an over-all effect on the entire system. Typical wells are selected in order to show the effect of various changes, such as; 1. 2. 3. 4.

Separator pressure. Flowline size. Surface choke size. Tubing size.

Analysis shows whether or not the particular well is limited in its production rate by the reservoir’s ability to give up fluids or by the producing system. The selection of various parameters, such as separator pressure or size of flowline is related to economics. For example, the selection of the separator pressure in a gas lift system is extremely important in determining compressor horse-power. Introduction to Well Testing (Aug 1996)

Separator pressures from 40 to 120 psi may have very little effect on the flow rate from a low productivity well (perhaps 10 B/D), but may have a very decisive effect of the flow rate of high productivity wells (perhaps 500 B/D). A complete systems analysis shows the effect of varying the separator pressure on compressor horse power and, hence, the economic feasibility of buying more or less horse power. The various profit indicators such as pay-out, rate of return, net present value, etc., can be used to make the decision. In other cases, the changing out of the flowline may permit the same separator pressure but reduce the wellhead flowing pressure and, hence, increase production considerably. It may be that the downhole and horizontal conduits have not been properly sized. Too small a tubing size may retard the production rate as well as too large a tubing size. Also, low flow rates can be inefficient in large tubing sizes and undesirable heading conditions may exist. 6.2

Inflow Performance Curves

In performing a system analysis on a well, it is necessary to have good test data on the well so that the reservoir capability can be predicted. Referring to Figure 6-1, we note that the IPR (Inflow Performance Relationship) curve may be shown as; 1. A straight line (constant PI - Productivity Index denoted “J”). 2. A curve which shows that the PI is decreasing with rate. 3. A combination of 1 and 2. 6-3

Section 6

Nodal Analysis

The constant PI normally occurs for single phase liquid flow above the bubble point pressure, and the curved line shows the PI to be decreasing below the bubble point pressure because of two phase flow conditions in the reservoir (liquid plus gas). Figure 6-2 shows a series of IPR curves for future reservoir pressure conditions. The following methods and respective equations may be used to predict the shape of IPR curves. 1.

Constant PI q L = J(P r − P wf )

2.

Changing PI

a.

Vogel’s Equation1 for undamaged wells, FE = 1.0. (FE= Flow Efficiency). q0

P 2 P = 1 − 0.2 wf − 0.8 wf q 0 (max) Pr Pr

b.

Standing’s Modification to Vogel’s Equation for damaged or stimulated wells2, FE = 1.0. q0

P  P  = 1 − 0.2 wf  − 0.8  wf  q 0 (max)  Pr   Pr 

2

where

4.

Fetkovich’s Equation for oil wells4 n q 0 = J ′ 0  P r 2 − P wf 2

5.

Fetkovich’s Equation for predicting future inflow curves for oil wells4 n P  q 0 = J ′ 01 r 2   P r 2 − P wf 2  P r 1

where J′01 was determined at Pr1 and we are interested in making flow rate predictions when the static pressure is Pr2. 6.2.1 Tubing Intake Curves Figure 6-3 shows a tubing intake or vertical multiphase flow curve being added to the inflow performance curve in order to determine the flow rate. If the simplest case of a constant wellhead pressure is assumed, then this curve is easily constructed by assuming flow rates and determining the corresponding flowing bottomhole intake pressure for a set tubing size, GOR, wellhead pressure, depth and fluid properties. Any number of vertical multiphase flow correlations may be used with the most popular being those of Hagedorn and Brown5, Orkiszewski6, Duns and Ros7 and Beggs and Brill8. Modern software allow these individual curves to be generated directly with the computer as opposed to referring to the thick volumes of pregenerated curves.

Pwf = Pr-(Pr - Pwf) FE 3.

Standing’s Relationship for predicting future inflow curves3.

J *2 (K ro / µ o B o ) 2 = J *1 ( K ro / µ o B o) 1

Introduction to Well Testing

6-4

Schlumberger

2000

P1

Pwf (PSI)

P2

(1) C

(2) (3)

P3

1000

ONS

TAN

I.P.

T P.

R.

I.

CO

MB

INA

TIO

NI

.P.

R.

0 0

400

800

1200

1600

q0 (BOPD)

Fig. 6-1 Inflow Performance Curves.

2000

P1 P2 P3 I.P.

Pwf (PSI)

R. 1

I.P

.R.

1000

2

I.P

.R

.3

0 0

400

800

1200

1600

q0 (BOPD)

Fig. 6-2 Future Inflow Performance Curves.

Introduction to Well Testing (Aug 1996)

6-5

Section 6

Nodal Analysis

Tubing Size : 2.992 in. I.D. PI = 2.000 B/D/Psi

2000

Pwf (PSI)

1600

P.I. GOR = 600

1200

Tubing

Intake

800

Rate possible Obtained (The intersection of both curves)

400

200

400

600

800

1000

1200

1400

1600

1800

q0 (BOPD)

Fig. 6-3 Tubing Intake Curve in Combination with IPR Curve.

6.2.2 Horizontal Flow Curves. For the more practical oil field problem, a horizontal flowline is generally incorporated into the system. A constant separator pressure can be expected, but as the flow rate changes, the wellhead pressure will increase and, hence, the tubing intake pressure will also increase. The same type of plot can be made as in Fig. 6-3, except the wellhead pressure changes with each flowrate.

6.2.3 Other Considerations The effect on production rate of various restrictions, such as surface chokes, downhole chokes, safety valves, and completion restrictions, can all be properly accounted for. These can be adequately handled in the entire system, for any producing oil or gas well.

The manner of solution is to assume a flow rate, find the wellhead pressure by using a horizontal multiphase flow correlation, and then, using that wellhead pressure, find the tubing intake pressure. The wellhead pressures can be plotted on the same graph as noted in Figure 6-4, or the wellhead pressures may be used for an alternate solution as noted in Figure 6-5.

Introduction to Well Testing

6-6

Schlumberger

Tubing Size : 2.992 in. I.D. Flowline Size : 3.000 in. I.D. PI = 2.000 B/D/Psi

2000

Pwf (PSI)

1600

P.I.

1200

ation

bin Com

GOR = 600

and ontal

ke

g Inta

Tubin

Horiz

800

400

Rate Possible GOR = 600

re Wellhead Pressu

200

400

600

800

1000

1200

1400

1800

1600

q 0 (BOPD)

Fig. 6-4 Intake Curve from Combined Flowline and Tubing Multiphase Flow.

Tubing Size : 2.992 in. I.D. Flowline Size : 3.000 in. I.D. PI = 2.000 B/D/Psi

500

Pwf (PSI)

400

300

Rate Possible

Ver

tica

l

200

GOR = 600

ntal

Horizo

100

GOR = 600

200

400

600

800

1000

1200

1400

1600

1800

q 0 (BOPD)

Fig. 6-5 Alternate Solution to Exhibit 6-4.

Introduction to Well Testing (Aug 1996)

6-7

Section 6 6.3

Nodal Analysis

The Nodal Concept

Figure 6-6 is a schematic of a simple producing system. This system consists of three phases:

3. Flow through horizontal pipe. Figure 6-7 shows the various pressure losses that can occur in the system from the reservoir to the separator. Beginning from the reservoir these are noted as;

1. Flow through porous medium. 2. Flow through vertical or directional conduit. ∆ P1 = Pr - Pwfs ∆ P2 = Pwfs - Pwf ∆ P3 = PUR - PDR ∆ P4 = PUSV - PDSV ∆ P5 = Pwh - PDSC ∆ P6 = PDSC - PSEP ∆ P7 = Pwf - Pwh ∆ P8 = Pwh - PSEP

= = = = = = = =

Pressure loss in porous medium. Pressure loss across completion. Pressure loss across regular, choke or tubing nipple. Pressure loss across safety valve. Pressure loss across surface choke. Pressure loss in surface flow line. Total pressure loss in tubing string which includes ∆P3 and ∆P4. Total loss in surface flow line including surface choke. Gas Sales

Flowing Wellhead Pressure

Horizontal Flowline

Separator

Stock Tank

Vertical or Inclined Tubing

Intake

Flow Through Porus Media P, K, IPR

Fig. 6-6 Simple Producing System. Introduction to Well Testing

6-8

Schlumberger

The various well configurations may vary from the very simple system of Figure 6-6 to the more complex system of Figure 6-7, or any combination thereof, and present day completions more realistically include the various configurations of Figure 6-7. In order to solve the total producing system problem, nodes are placed to segment the portion defined by different equations or correlation’s.

Figure 6-8 shows locations of the various nodes. This figure is the same as Figure 6-7 except only the node positions are shown.

∆P8

Pwh

Gas Sales

∆P6 Psep

∆P5

PDSV

Separator

Stock Tank

∆P4 ∆P1 = (Pr - Pwfs) ∆P2 = (Pwfs - Pwf) ∆P3 = (PUR - P DR) ∆P4 = (PUSV - PDSV ) ∆P5 = (Pwh - PDSC) ∆P6 = (PDSC - Psep) ∆P7 = (Pwf - Pwh) ∆P8 = (Pwh - Psep)

PUSV ∆P7

PDR ∆P3

= Loss in Porus Medium = Loss across Completion = Loss across Restriction = Loss across Safety Valve = Loss across Surface Choke = Loss in Flowline = Total Loss in Tubing = Total Loss in Flowline

PUR

Pwf ∆P2

Pwfs

Pr ∆P1

Fig. 6-7 Possible Pressure Losses in Complete System.

Introduction to Well Testing (Aug 1996)

6-9

Section 6

Nodal Analysis

1A

2

3

1 4

1B

5

6

Node

Location

1 2 3 4 5 6 7 8 1A 1B

Separator Surface Choke Wellhead Safety Valve Restriction Pwf Pwfs Pr Gas Sales Stock Tank

Remarks Functional Functional Functional

8

7

Fig. 6-8 Location of Various Nodes.

∆P3-1 = Pwh - P sep)

3

1

∆P6-3 = (P wf - Pwh)

Node

Location

1 3 6 8

Separator Wellhead Pwf Pr

∆P8-6 = (Pr - P wf) 6

8

Fig. 6-9 Nodes for Simple Producing System. Introduction to Well Testing

6-10

Schlumberger

The node is classified as a functional node when a pressure differential exists across it and the pressure or flow rate response can be represented by some mathematical or physical function.

6.3.1

Node 1 represents the separator pressure, which is usually regulated at a constant value. The pressure at node 1A is usually constant at either gas sales lines pressure or gas compressor suction pressure. The pressure at node 1B is usually constant at 0 psig. Therefore, the separator pressure will be held constant at the higher of the two pressures needed to flow single phase gas from node 1 to node 1A or to flow single phase liquid from node 1 to node 1B. It will be assumed that the separator pressure is constant for any flow rate, and it will be designated as node 1.

Separator pressure: Flow line: WOR: Depth: GOR: Pr: IPR:

Notice that in the system there are two pressures that are not a function of flow rate. They are Pr at node 8 and PSEP at node 1. For this reason any trial and error solution to the total system problem must be started at node 1 (PSEP), or at node 8 (Pr) or both nodes 1 and 8 if and intermediate node such as 3 or 6 is selected as the solution node. Once the solution node is selected, the pressure drops or gains from the starting point are added until the solution node is reached. The following four examples show this procedure for the four possible nodes shown in Figure 6-9. Although all nodes are not shown, the same node numbers are maintained as in Figure 6-8. Note: Pr can be a function of flow rate or drainage distribution in the reservoir. However, for the flow rates investigated in this section, Pr is assumed to be constant.

1. Select flow rates for a trial and error procedure: Assume flow rates of 200, 400, 600, 800, 1000 and 1500 B/D.

Introduction to Well Testing (Aug 1996)

Example Problem 1 - Using Node 8 to Find the Flow Rate Possible (Node 8 = Pr) - Given data:

Flowing oil well

Tubing size:

100 psi 2", 3000 ft long 0 5000 ft mid perf. 400 scf/B 2200 psi PI = 1.0 B/D psi (assume constant) 2-3/8"

Find the oil flow rate using node 8 as the solution point? Procedure:

2. For each rate, start at PSEP = 100 and add all the pressure losses until reaching Pr at node 8. From Figure 6-9, we note that these losses would be ∆P3-1 + ∆P6-3 + ∆P8-6 or loss in surface flow line + loss in tubing string + loss in porous medium. These various losses for the assumed rates are noted in Table 6-1 3. Plot the created pressure vs flow rate (Figure 6-10). This represents the system performance from the separator to Pr. 4. Plot Pr at the given 2200 psi (Figure 610). 5. The intersection of the reservoir pressure line and the system performance line gives the presided flow rate (900 BOPD). 6-11

Section 6

q 200 400 600 800 1000 1500

Nodal Analysis

PSEP 100 100 100 100 100 100

Horiz. Multiphase Flow P3 ∆P3-1 115 15 140 40 180 80 230 130 275 175 420 1320

Vertical Multiphase Flow P6 ∆P6-3 750 635 880 740 1030 850 1220 990 1370 1095 1840 1420 Table 6-1

6.3.2

Example Problem 2 -Using Solution Node 6 to Find Flow Rate (Flowing Bottomhole Pressure) Given data: Same as Example Problem 1 For this solution, pressure drops must be added from node 1 to node 6 and subtracted from node 8 to node 6. Procedure: 1. Since the predicted flow rate is already known from Example 1, the same flow rates will be assumed: 200, 400, 600, 800, 1000 and 1500 B/D. 2. Determine the pressure loss from node 1 (separator) to node 6 (Pwf). For each assumed flow rate, start node 1 (PSEP) and add ∆P3-1 + ∆P6-3. Table 6-2 shows these results.

Introduction to Well Testing

IPR P8 950 1280 1630 2020 2370 3340

Total Loss

∆P8-6 400 600 800 1000 1500

∆P8-1 850 1180 1530 1920 2270 3240

3. Determine the pressure loss (∆P8-6) from node 8(Pr) to node 6(Pwf). For a constant PI assumption this can be calculated from the equation ∆P8-6 = Assumed Rate/PI. These values are noted in Table 6-3. 4. Plot P6 vs q from both step 2 and step 3 (Figure 6-11). Node 6 is called the intake node since this point is the intake from the reservoir into the production tubing. The intersection of the PI line and the socalled intake curve is the predicted flow rate for this system (900 BOPD, Figure 6-11). The presentation based on the selection of node 6 as the solution node is good if it is desired to evaluate changing Pr’s or different IPR curves. Notice the answer is the same as Example 1 and this is true regardless of the node selection.

6-12

Schlumberger

Assumed Rate 200 400 600 800 1000 1500

PSEP 100 100 100 100 100 100

Assumed Rate 200 400 600 800 1000 1500

Introduction to Well Testing (Aug 1996)

Horiz. Multiphase Flow Pwh ∆P3-1 115 140 180 230 275 420

Table 6-2.

Vertical Multiphase Flow P6 ∆P6-3

15 40 80 130 175 320

750 880 1030 1220 1370 1840

635 740 850 990 1095 1420

Pr

∆P8-6

P6= Pwf

2200 2200 2200 2200 2200 2200

200 400 600 800 1000 1500

2000 1800 1600 1400 1200 700

Table 6-3

6-13

Section 6

Nodal Analysis

2500

ep

2000

Pr psi

m

ce

to

Pr = 2200

Pr

Ps

fro

an

1500

m or

rf

m

Pe

te

s Sy

1000

900 BOPD

500

0 0

500

1000

1500

qo BOPD Fig. 6-10 Solution to Example Problem 1. 2500

2000

IPR

Cu

P wf , psi

rve

1500

ve

m

te Sys

Cur

rm

erfo

P ake

1000

e anc

Int

900 BOPD

500

0 0

500

1000

1500

qo , BOPD Fig. 6-11 Solution to Example Problem 2.

Introduction to Well Testing

6-14

Schlumberger

3. Determine the pressure loss from node 8 (Pr) to node 3 (Pwh). For each assumed rate, start at Pr and add ∆P8-6 + ∆P6-3 These values are tabulated in Table 6-5.

6.3.3

Example Problem 3 - Using Solution Node 3 to find the -Flow Rate (Flowing Wellhead Pressure) Given Data: Same as Example Problem 1.

4. Plot P3 vs q from both step 2 and step 3 (Figure 6-12). Node 3 is called the flowing wellhead pressure (Pwh).

For this solution we have selected the wellhead as the location of the solution node. Therefore, this is a common point at which we add the pressure losses from node 1 to node 3 and subtract pressure losses from node 8 to node 3.

5. The intersection of the flow line pressure drop line and the downhole performance curve is the predicted flow rate for the system (900 BOPD, Figure 6-12). The presentation based on the selection of node 3 as the solution node is good if it is desired to evaluate different flowlines or wellhead back pressure. Notice the predicted rate of 900 BOPD remains the same.

Procedure: 1. Assume the same flow rates as for the previous examples: 200, 400, 600, 800, 1000 and 1500 B/D. 2. Determine the pressure loss from node 1 (separator) to node 3 (wellhead). For each assumed rate and for PSEP = 100 psi we find ∆P3-1 and P3(Pwh). These values are tabulated in Table 6-4.

ca

rti

Ve

600

nd

la

Node 3 Solution

R

IP

500

rm

rfo

Pe e

c an

400

ve ur

Pt f , psi

C

300

e

urv

a

form

200

ste

l Sy

ta izon

er mP

C nce

Hor

900 BOPD

100

0 0

500

1000

1500

qo , BOPD Fig. 6-12 Solution to Example Problem 3.

Introduction to Well Testing (Aug 1996)

6-15

Section 6

Nodal Analysis

q

PSEP

200 400 600 800 1000 1500

100 100 100 100 100 100

q 200 400 600 800 1000 1500

Pr 2200 2200 2200 2200 2200 2200

Introduction to Well Testing

∆P3-1 for Horiz. Multiphase Flow 15 40 80 130 175 320

P3 = Pwh 115 140 180 230 275 320

Table 6-4

P6 2000 1800 1600 1400 1200 700

Table 6-5

∆P8-6 200 400 600 800 1000 1500

P3 610 440 450 330 180

∆P6-3 1390 1250 1150 1070 1020

6-16

Schlumberger

Example Problem 4 - Using Solution Node 1 to Find Flow Rate (Separator) Given Data: Same as Example Problem 1.

2. Plot P1 from Table 6 vs q (Figure 6-13).

In this example, the separator pressure is held constant at 100 psi and is designated as node Therefore, all pressure losses from node 8 (Pr) to node 1 (separator) are determined and then subtracted from node 8.

4. The intersection of the separator pressure line and the system performance line is the predicted flow rate (900 BOPD, Figure 613). The presentation based on the selection of node 1 as the solution node is good if it is desired to evaluate different separator or header pressures. Notice that the predicted rate of 900 BOPD remains the same.

6.3.4

3. Plot PSEP at the given 100 psi (Figure 613).

Procedure: 1. Assume flow rates of: 200, 400, 600, 800, 1000 and 1500 B/D. For each rate, start at Pr = 2200 psi and subtract ∆P8-6 + ∆P6-3 + ∆P3-1. This information is noted in Table 6-6.

em

st

Sy

600

rfo

Pe ce

an

rm

500

ve

ur

C Pr tp Ps

300

ep

Psep, psi

m

fro

400

200 900 BOPD Psep = 100 psi

100

0 0

500

1000

1500

qo , BOPD

Fig. 6-13 Solution to Example Problem.

Introduction to Well Testing (Aug 1996)

6-17

Section 6

Nodal Analysis

From IPR q 200 400 600 800 1000 1500

Pr 2200 2200 2200 2200 2200 2200

P6 2000 1800 1600 1400 1200 700

∆P8-6 200 400 600 800 1000 1500

From Vertical Multiphase Flow P3 ∆P6-3 610 1390 550 1250 450 1150 330 1070 180 1020 P3<0 Table 6-6

6.3.5

Discussion of Example Problems 1 through 4 It is important to notice that when starting at the reservoir (node 8), the slope of the resulting systems curve on the pressure-flow rate diagram at the solution node is zero or negative. This can be observed in figures 610 through 6-13. This is expected since any systems curve developed by starting at Pr (regardless of the solution node) includes reservoir performance in the form of PI or IPR. A pressure flow rate curve generated by starting at Pr actually displays the required pressure at the solution node for the reservoir to produce the stated flow rate. For example, the vertical and IPR curve shown on Figure 6-12 shows that if a flowing wellhead pressure of 100 psi could somehow be created, the reservoir and well would produce 1100 B/D. In contrast, notice that when starting at the separator pressure (node 1), the slope of the resulting systems curve on the pressure-flow rate diagram at the solution node is zero or positive. This is again shown clearly in figures 6-10 through 6-13. The pressure-flow rate curve generated by starting at the separator pressure displays the created pressure at the solution node for each flow rate. For ex-

Introduction to Well Testing

From Horiz. Multiphase Flow P1 595 524 412 255 P1<0 P1<0

∆P3-1 15 26 38 75 ∆P3-1>180

∆P8-1 1605 1676 1788 1945 ∆P8-1>2200 ∆P8-1>2200

ample, the flowline curve shown on Figure 612 shows that for a production rate of 1100 BOPD the created wellhead pressure is 300 psi. The total producing system will produce only where the created pressure at any node is equal to the required pressure at that node for the stated producing rate. This occurs where the two curves intersect as shown in figures 6-10 through 6-13. Notice on Figure 6-12 for 1100 BOPD, the required pressure is 100 psi at node 3 (wellhead pressure) and the created pressure is 300 psi. Therefore, this system will not produce 1100 BOPD. Obviously, the rate possible must be the same irrespective of the node selected to solve the problem. Different nodes are selected for convenience based on which system parameter is to be studied. For example, suppose in our example problem it is desired to know what this well will produce with a 3" ID flow line. A new flow line system curve could be generated and overlaid on Figure 6-12 as shown on Figure 6-14. Node 3 was selected for the solution node because of clarity of presentation showing the flow line pressure loss. Notice that the same vertical and IPR curve applies regardless of the flow line system.

6-18

Schlumberger

Ve

600

l ca rti an d R

IP

500

Pe rfo

" ,2

line

rm

w Flo

an ce

400

.

I.D

ur ve

Ptf, psi

C

300

. Flowline, 3" I.D

200 900 BOPD

1040 BOPD

100

0 0

500

1000

1500

qo , BOPD Fig. 6-14 Effect of Change in Flow Line Size.

3

∆P3-1 1

∆P5-3 2-7/8" or 3" Tubing 5

2-3/8" Tubing

Node

Location

1 3 5 6 8

Separator Wellhead Taper Connection Pwf Pr

∆P6-5

Liner

6

∆P8-6

8

Fig. 6-15 Tapered Strings. Introduction to Well Testing (Aug 1996)

6-19

Section 6 6.4

Nodal Analysis

Changes in Flow Conduit Size

Thus far, the discussion has pertained to the rather simple system shown in Figure 6-9. Notice that on this system there is only one flow line size and one tubing size. Of course, it is possible and sometimes advantageous to change one of these pipe sizes in the middle of the string. To evaluate a system of this nature, the solution node could be placed at the point where the pipe size changes. 6.4.1

Example Problem 5 - Tapered Tubing Strings Suppose that in the previous example for some reason it was necessary to set a liner from near 3500 feet through the producing zone at 5000 feet and this liner was of such ID that 2-3/8" tubing was the largest size tubing that could be installed. Let us investigate the possible production rate increases by installing larger than 2-3/8" tubing above the liner from 3500 feet to the surface. Refer to Figure 6-15. Given Data: Same as Example 1. The solution node (node 5) selected to solve this problem is located at the tubing taper (Figure 6-15). In this example, the pressure drops must be added from node 1 to node 5 and subtracted from node 8 to node 5. Maintaining the same nomenclature as in Figure 68, we have designated the tapered connection as node 5. Procedure:

2. Determine the pressure loss from node 1 (separator) to node 5 (tapered connection). For each assumed rate and starting with PSEP = 100 psi we add ∆P3-1 + ∆P5-3. Table 6-7 summarises these results, and both 2-7/8" and 3" tubing are evaluated above the tapered connection. 3. Determine the pressure losses from node 8 to node 5. for each rate, start at Pr = 2200 psi and subtract ∆P8-6 + ∆P6-5. These results are noted in Table 6-8. 4. Plot P5 vs q from both step 2 and step 3 (Figure 6-16). 5. The intersection of the two performance curves at the tapered connection predicts a flow rate of about 1020 BOPD for 2.5" ID tubing and 1045 BOPD for 3" ID tubing. Remember, for a 2.0" ID tubing string the predicted rate was 900 BOPD. Notice that the increase in rate from 2.0" ID to 2.5" ID is much more significant than the increase in rate from 2.5" ID to 3" ID. As pointed out previously, this problem could have been solved by placing the solution node at any point in the system. However, this approach can simplify the procedure depending on the manner in which the curves or computer programs available are formatted. This same procedure could be used if a change in flow line configuration occurs at some point along the path of the horizontal system.

1. Assume flow rates of 200, 400, 600, 800, 1000 and 1500 B/D.

Introduction to Well Testing

6-20

Schlumberger

Horiz. Multiphase Flow q 200 400 600 800 1000 1500

PSEP 100 100 100 100 100 100

P3 115 140 180 230 275 420

∆P3-1 15 40 80 130 175 320

Table 6-7a

Horiz. Multiphase Flow q 200 400 600 800 1000 1500

PSEP 100 100 100 100 100 100

P3 115 140 180 230 275 420

∆P3-1 15 40 80 130 175 320

Table 6-7b

From PI q 200 400 600 800 1000 1500

Pr 2200 2200 2200 2200 2200 2200

Introduction to Well Testing (Aug 1996)

P6 2000 1800 1600 1400 1200 700

Table 6-8

∆P8-6 200 400 600 800 1000 1500

(2-7/8" Tubing) Vertical Multiphase Flow P5 ∆P5-3 475 360 500 360 600 420 718 488 820 545 970 550

(3" ID Tubing) Vertical Multiphase Flow P5 ∆P5-3 420 305 475 335 560 380 660 430 780 505 900 480

From Vertical Multiphase Flow P5 ∆P6-5 1400 600 1300 500 1170 430 1000 400 820 380 360 340

6-21

Section 6

Nodal Analysis

2500 Tapered String Scenario 5000' to 3500' 3500' to 0' 3500' to 0'

Ptaper , psi

2000

1500

Belo

wT

ape

r Pe

rform

anc

eC

Tubing

urve

1000

2-7/8" 3"

rve

Above

500

2" 2-7/8" 3"

Taper

ce Cu forman

Per

1045 BOPD

1020 BOPD 0 0

500

1000

1500

qo , BOPD Fig. 6-16 Tapered String Solution (Example 5). 2 3 1 Horizontal Flowline

6

Node

Location

1 2 3 6 8

Separator Surface Choke Wellhead Pwf Pr

Remarks Functional

8

Fig. 6-17 Surface Choke Problem. Introduction to Well Testing

6-22

Schlumberger

6.5

Functional Nodes

In the previous discussion, it has been assumed that no pressure discontinuity exists across the solution node. However, in a total producing system there is usually at least one point or node where this assumption is not true. When a pressure differential exists across a node, that node is termed a functional node since the pressure flow rate response can be represented by some physical or mathematical function. Figure 6-8 shows examples of some common system parameters which are functional nodes. Of course, there are many other surface or downhole tools or completion methods which could create pressure drops with flow rates as those shown in Figure 6-8 (such as safety valves, perforations, etc.). It is important to notice that for each restriction placed in the system shown in Figure 68 the calculation of pressure drop across that node as a function of flow rate is represented by the same general form. ∆P ≈ qn That is, the pressure drop, ∆P, is proportional to the flow rate. In fact, there are many equations available in the literature to describe these pressure drops for common system restrictions. It is not the purpose of this section to discuss the merit of the different equations, but rather to show how to use them once the selection has been made, considering the entire producing system.

Introduction to Well Testing (Aug 1996)

6.5.1 Surface Wellhead Choke Refer to Figure 6-17 for a physical description of the well with a surface choke installed. The same nodes as set out in Figure 6-8 are maintained. Since the wellhead choke is usually placed at node 2, this will be the solution node selected to solve the problem. It is necessary to solve this problem in two parts. The first part of the solution is exactly the same as previously shown. Inspection of figures 6-17 and 6-12 show that the “vertical and IPR performance curve” actually represents the upstream pressure from node 2 (Pwh’, Figure 6-12) and the “horizontal system performance curve” actually represents the downstream pressure from node 2 (PDSC’, Figure 6-12). Thus far, we have considered no pressure drop across the node and, therefore, the predicted rate is where upstream pressure equals the downstream pressure (Pwh = PDSC). However, we know the wellhead choke will create a pressure drop across functional node 2 for each flow rate. This created ∆P can be calculated with one of many pressure drop equations for choke beans. Therefore, the solution procedure is to find and plot the required ∆P vs q from Figure 6-12 and overlay the created ∆P vs q from the choke bean performance calculations.

6-23

Section 6

Nodal Analysis

6.5.2

Example Problem 6 - Determine Effect of Surface Choke Sizes Using Node 2 as the Solution Node Given Data: Same as Example 1 Procedure: 1. Generate the total system analysis curve using node 2 as the solution node exactly as done in Example 3, (Figure 6-12). 2. Select arbitrary required pressure drops across node 2 (∆P = Pwh - PDSC) and determine the flow rate for each ∆P as shown in Figure 6-18. (Notice Figure 6-18 is the same as Figure 6-12 with ∆P’s displayed). These results are noted in Table 6-9. 3. From step 2, plot the required ∆P vs q as shown on Figure 6-19. 4. Calculate the created pressure drop vs flow rate for choke beans of interest. The equation used for these calculations is: Pwh =

C R0.5 q S2

(from Gilbert)

where Pwh =

Flowing wellhead pressure, psi.

R

=

GLR; MCF/STB.

q

=

Gross liquid rate, STB/D.

S

=

Choke bean size, 64th’s of an inch

C

=

Constant, assume 500 for this problem.

Gilbert noted that his formula is good when the downstream pressure (PDSC) is Introduction to Well Testing

less than 70% of the upstream pressure (Pwh) or PDSC/Pwh < 0.7. Suppose we are interested in investigating well performance for the following choke bean sizes: 16/64", 20/64", 24/64", 28/64". Table 6-10 shows these results. The ∆P’s calculated are unique to the example system since the downstream pressures were calculated for the example system. Notice that in each case a check was made to ensure PDSC/Pwh ≤ 0.7 so that Gilbert’s equation would apply. If this is not the case, a subcritical flow equation must be used to calculate ∆P across the choke. 5) From the tables generated, plot the choke bean performance as shown on Figure 620. 6) Overlay the results shown on Figure 6-19 and Figure 6-20 to produce Figure 6-21. Figure 6-21 displays the total system performance for different wellhead choke sizes. The system performance curve shows the required ∆P for various flow rates considering the entire system from reservoir to separator. The choke performance curves show the created ∆P for various flow rates considering choke performance for different choke sizes. The intersection points of the created and required ∆P’s represent the possible solutions. For example, the rate will drop from 900 BOPD to 715 BOPD with the installation of a 24/64" wellhead choke. Figure 6-22 shows another presentation that is often used to evaluate wellhead chokes. The presentation shows the entire system performance, which sometimes is advantageous. The same techniques discussed in this paper are used to generate this type of analysis. Notice that this solution gives the same answer. 6-24

Schlumberger

600

ps

U tre am

500

of od

N e )

∆P = 100

o

e str

ke

∆P = 200

ho

∆P = 300

2

o fN

wn

Do

C

∆P = 400

am

ad

300

e lh el (W

Ptf, psi

de

2

400

Pu = Pd

0

500

q = 800 BOPD

0

q = 690 BOPD

100

q = 560 BOPD

q = 410 BOPD

200

1000

1500

qo , BOPD Fig. 6-18 Surface Choke.

500

400

∆P, psi

300

200

q 0 = 900 BOPD at ∆P= 0

100

0 0

500

1000

1500

qo , BOPD

Fig. 6-19 Total Systems Performance Curve for Surface Choke Problem 6. Introduction to Well Testing (Aug 1996)

6-25

Section 6

Nodal Analysis

∆P = Pwh - PDSC 100 200 300 400 16/64

q BOPD 300 400 500 600

20/64

q 300 500 700 900

24/64

q

500 700 900 1100 28/64

q

800 1000 1200

q, B/D 800 690 560 410

Table 6-9

PDCS From Fig 6-18 128 140 160 180

Pwh From Gilbert 370 494 617 741

PDCS From Fig 6-18 128 160 200 250

Pwh From Gilbert 237 395 553 711

PDCS From Fig 6-18 160 200 250 300

Pwh, psi From Gilbert 274 384 494 603

PDCS From Fig 6-18 227 275 330

Pwh, psi From Gilbert 322 403 484

Introduction to Well Testing

Table 6-10

PDSC/Pwh

∆P=Pwh-PDSC

.35 .28 .26 .24

242 354 457 561

PDSC/Pwh

∆P= Pwh-PDSC

.54 .41 .36 .35

109 235 353 461

PDSC/Pwh

∆P = PDSC - Pwh

.58 .52 .51 .50

114 184 244 303

PDSC/Pwh

∆P = PDSC - Pwh

.70 .68 .68

95 128 154 6-26

Schlumberger

500 20/64

16/64

400

24/64

∆P, psi

300

200 28/64

100

0 0

500

1000

1500

qo , BOPD Fig. 6-20 Choke Bean Performance.

500 20/64

16/64

m

ste

Sy

400

rfo

Pe 24/64

a rm

300

ce

∆P, psi

e nc

an

m for

r

e ok

Pe

Ch

200

28/64

100

0 0

500

1000

1500

qo , BOPD Fig. 6-21 Systems Performance for Various Wellhead Chokes. Introduction to Well Testing (Aug 1996)

6-27

Section 6

Nodal Analysis

2500 16/64 Choke

IPR

20/64 Choke

Cur

24/64 Choke

28/64 Choke

st em

1500

No Choke

Sy

Pressure , psi

In ta ke

Pe rfo rm

an ce

ve

2000

1000 Vertica l IPR Perfor mance

16/64 Choke

20/64 Choke 24/64 Choke 28/64 Choke

500

ce Curve

erforman

lP Horizonta

No Choke

0 0

500

1000

1500

qo , BOPD

Fig. 6-22 Surface Choke Evaluation.

6.5.3 Example Problem 7 In order to further illustrate the effect of certain variables, such as tubing size, flowline size, etc., two problems will be worked. One is a high productivity well (high rate) and the other a low productivity well. The high rate well is designated well A and the low rate well as well B.

Flowline Size Casing Size Separator Pressure Well produces

Data for Well A:

Pr = 2300 psi, Pb = 2300 psi and one test shows 500 B/D for Pwh = 180

Depth Pr J GOR Tubing Size Flowline Length

= = = = = =

8000 ft. 2800 psi 15 B/D psi 400 SCF/STB 2-1/2 in. OD 4000 ft.

Introduction to Well Testing

= 2-1/2 in. ID. = 7 in. = 100 psi 100 % oil

Data for Well B: Same as Well A except:

Figure 6-23 shows the solution to these problems assuming no restrictions are present in the production conduits. Table 6-11 summarises the results:

6-28

Schlumberger Tubing Size: 2.441 in. I.D. Flowline Size: 2.500 in. I.D.

5000

GOR = 400

Pwf (psi)

4000

3000 Well A

2000

1000

Well B

Wellhead

qB

qA

500

1000

1500

2000

2500

3000

Pressure

3500

GOR = 400

4000

4500

qo (BOPD)

.

Fig. 6-23 Solution to Determine Possible Rate for Well A and Well B.

qo (BOPD) Pwf (psi) Pwh (psi)

WELL A 2320 2650 550

Table 6-11

6.5.3.1 Effects of Separator Pressure In analysing these wells, it is important to see the effect of different separator pressures while maintaining everything else constant. Several computer runs were made varying the Psep (psi) 40 60 80 100 120 140 160 180 200

Introduction to Well Testing (Aug 1996)

WELL B 850 1380 200

Well A 2350 2340 2330 2320 2305 2290 2270 2245 2215

separator pressure from 40 to 200 psi as shown in Figure 6-24. Note, that for the high productivity well, the change in separator pressure has a significant effect on the flow rate. Table 6-12 shows the results:

Table 6-12

qo (BOPD)

Well B 855 852 850 848 845 842 840 838 835

6-29

Section 6

Nodal Analysis

6.5.3.2 Effects of Flowline Size The flowline sizes were varied between 2" and 4" nominal sizes with everything else remaining constant. This information is plotted in Figure 6-25. Note that the flowline size

Flowline Size (in) 2 2-1/2 3 4

qo (BOPD) Well A 1480 2320 2950 3650

Introduction to Well Testing

Well B 690 850 890 925

Table 6-13

6.5.3.3 Effects of Tubing Size The tubing sizes were varied between 2" and 4" sizes and all other variables were held constant. Again, it is noted that for all Well A the larger diameter tubing sizes show a large

Tubing Size (in) 2 2-1/2 3 4

has a great deal more effect for Well A than for Well B. Table 6-13 summarises these results:

increase in rate, while Well B shows lower increases in flow rates. Refer to figures 6-26 and 6-27. Table 6-14 summarises these results:

Well A 1800 2320 2600 2700

Table 6-14

qo (BOPD)

Well B 730 830 910 950

6-30

Schlumberger

Tubing Size: 2.441 in. I.D. Flowline Size: 2.500 in. I.D. GOR: 400 scf/bbl

2600

qo (BOPD)

2400

Well A

2200

1000 Well B

800

600 40

80

120

160

200

240

280

Separator Pressure (psi)

Fig. 6-24 Effect of separator Pressure.

Tubing Size: P sep: GOR:

4000

2.441 in. I.D. 100 psi 400 scf/bbl

3500 Well A

qo (BOPD)

3000

2500

2000

1500

1000

Well B

500

0 1

2

3

4

Flowline Inside Diameter (Inches)

Fig. 6-25 Effect of Flowline Size.

Introduction to Well Testing (Aug 1996)

6-31

Section 6

Nodal Analysis

P sep : 100 psi Flowline Size: 2.500 in. I.D. GOR 400 scf/bbl

5000

2"

2.5"

Pwf (psi)

4000

3" 4"

3000 Well A Well B

2000

Wellhead

1000

500

1000

1500

2000

2500

3000

Pressure

3500

4000

4500

qo (BOPD)

Fig. 6-26 Effect of Tubing Size. Flowline Size: 2.5 in. I.D. P sep: 100 psi GOR: 400 scf/bbl

3000

Well A

qo (BOPD)

2500

2000

1500

1000 Well B

500

0 1

2

3

5

4

Tubing Inside Diameter (Inches)

Fig. 6-27 Effect of Tubing diameter .

Introduction to Well Testing

6-32

Schlumberger

6.6

General Discussion on the Effect of the Variables

Systems analysis can be used to include the effect of different variables on the production rate of a well.

ficient flow conditions, a situation which is analogous to small flow rates for a given large tubing size in lower productivity wells.

The effect of separator pressure is more significant for higher productivity wells than for lower productivity wells. This may be very important, for example, when gas lift is implemented on a well where the increase in production rate due to a decrease in separator pressure may be compared to the economics of the compressor horse-power.

Since vertical multiphase flow correlations differ considerably in predicting that rate at which flowing bottomhole pressure begins to increase for the next smaller rate, the example problem was worked with the Hagedorn and Brown, Duns and Ros, Orkiszewski and Beggs and Brill correlations, respectively, as shown in Figure 6-28. Table 6-15 summarises these results: In choosing a combination of tubing-flowline sizes, the nature of the inflow performance relationship should be considered under present and future conditions. As shown in Figure 6-29 an increase in tubingflowline sizes causes a significant increase in flow rates, especially for Well A, for which all possible rates occur at stable and efficient conditions. However, for Well B, all possible rates after the tubing-flowline increase occur under unstable pressure conditions very close to those at which the critical rate exists. If the GOR would decrease (GOR = 140 SCF/STB), then eventually Well B will not flow under the higher capacity tubingflowline configuration. Although the production rates would be smaller under the lower capacity configurations, this particular tubing-flowline configuration not only would ensure efficient and stable flow, but also would cause the well to flow for a longer period of time.

An increase in flowline size can represent significant additional production rate, especially in high productivity wells. The flowline should always be at least as large as the tubing size, or one size larger. At the same time, consideration should be given to the fact that too large a flowline may not be used for low flow rates, which is not normally indicated by the standard horizontal multiphase flow correlations as an adverse condition. The increase of tubing size can considerably affect the production rate of a well. Significant additional production rate may be achieved in high productivity wells after tubing size increase. However, there is a limit to the increase in tubing size with which the mechanism of two-phase flow will still be efficient. As the area for flow increases, the fluid velocities decrease, creating excessive fallback, which causes very unstable and inefCorrelation Hagedorn and Brown Duns and Ros Orkiszewski Beggs and Brill

Introduction to Well Testing (Aug 1996)

Table 6-15

Critical Rate (BOPD) 70 350 400 310

6-33

Section 6

6.7 Graphical Representation of the Total Producing System for One Well Refer to Figure 6-30, which shows most of the components and variables that exists in any one well. The most common way of plotting this information is to use pressure on the ordinate vs flow rate on the abscissa, although it may be reversed in some cases to permit a pressure depth plot on the same exhibit. Each curve or straight line in Exhibit 6-30 has very important significance in evaluating a complete system and each will be briefly discussed. 1. Static Pressure, Pr. The static pressure is the starting point for all systems graphs and is shown as a straight line. This would also represent an infinite productivity index line.

2. Sand Face IPR Curve, Pwfs. This line represents the flowing pressure that exists at the sand face for different flow rates. It is only the same as the measured flowing bottom hole pressure when no restrictions exist in the completion of the well, such as generally exist in the perforation or perhaps a gravel pack.

3. Flowing Bottom Hole Pressure, Pwf. This represents the flowing pressure that exists at the center of the perforated interval and is that pressure measured by running a pressure recording gauge in the well. 4. Tubing Intake Curve. This represents the pressure required at the bottom of the tubing string to allow certain production rates to enter the stock tank and, therefore, includes pressure losses in the flow line and tubing string, surface chokes, safety valves and any other restrictions. It should be noted that the tubing intake Introduction to Well Testing

Nodal Analysis

curve can be shifted to the right by removing any of the restrictions and is normally also shown without restrictions.

5. Choke Performance Curve. A Choke performance curve is shown on this plot. In this case, it is sized to give the flow rate of qL for P choke.

6. Horizontal Flowline Curve. This curve is prepared by assuming flow rates and, starting with the separator pressure, obtaining the pressure required on the downstream side of the choke.

7. The separator pressure is a reasonably constant value for all rates. 8. The stock tank pressure is a constant value at all rates.

Several changes or improvements may be made after studying an exhibit similar to Figure 6-30. The first question may be: “What dominates this well at the present time?” It appears to be the tubing intake curve which consists of pressure losses in the horizontal flowline, surface choke and tubing string. The curve can be immediately shifted to the right and, therefore, the rate can be increased by enlarging or even removing the surface choke. Larger flow lines and even larger tubing sizes will also shift this curve to the right and give higher production rates. One of our objectives would be to try to make the Pwf and Pwfs line coincide. There also appears to be quite a large ∆p across the completion, which suggests the need for greater shot density in perforating or possibly a restricted gravel pack or other completion problems. The separator pressure can possibly be lowered, which will also increase the rate. If the objective is lower producing rates, then this can easily be accomplished by installing a smaller surface choke. 6-34

Schlumberger

6.8

Summary

A nodal system has been presented in order to effectively evaluate a complete producing system. All of the components in the well, starting from the static pressure (Pr) and ending at the separator, are considered. This includes flow through the porous medium, flow across the perforations and completion, flow up the tubing string with passage through a possible downhole restriction and safety valve, flow in the horizontal flow line with passage through a surface choke and on to the separator. With the aid of multiphase flow correlations on computer, a systems graph can be quickly prepared. This graph can then be analysed to determine what dominated the well. If higher or lower rates are the objective, then the sys-

Introduction to Well Testing (Aug 1996)

tems graph shows immediately what can be done to accomplish this. Many wells may not have sufficient perforations, or pipe sizes may be inadequate. These types of problems can be easily detected. Frac or acid jobs may or may not be necessary. It may be that the well is tubing dominated and not reservoir dominated. In that case, a stimulation treatment is not necessary unless tubing sizes are increased. Good systems analysis can greatly improve the efficiency of a well or group of wells. This approach can dictate whether either well stimulation or decreases in certain components in the system may result in improving the production rate of a well, thus reducing excessive expenditures.

6-35

Section 6

Nodal Analysis

Tubing Size: 2.441 in. I.D. Flowline Size: 2.500 in. I.D.

5000

Pwf (psi)

4000

3000

2000

rill

dB

n gs a Beg os dR s an own Dun nd Br na dor e g Ha ski zew s i k Or

Well B

Well A

1000 Well B

500

1000

1500

2000

2500

3000

3500

4000

4500

qo (BOPD) Fig. 6-28 Critical Velocity Prediction with Different Correlations.

P sep

100 psi

GOR = 400

5000

GOR = 200 GOR = 140

Pwf (psi)

4000

GOR = 140

3000

GOR = 200 Well A GOR = 400

2000

2.441" Tubing + 2.5" Flowline

1000 Well B

1000

2000

3000

3.958" Tubing + 4.0" Flowline

4000

5000

6000

7000

8000

9000

qo (BOPD) Fig. 6-29 Effect of the Nature of IPR Curves on Tubing-Flowline Size.

Introduction to Well Testing

6-36

Schlumberger

1 Static Pressure ∆P Formation

2 IPR Curve fo

r Sand

∆P Completion 3 e tak

∆P Tubing

4

Flow

ing

In ing rve b Tu Cu

Pre

Face P

ressure

ssu

re in

We

llbo

re

e

urv

∆P Choke

rm

erfo

P oke

eC anc

5 Ch ∆P Horizontal Line

Surface ∆P Facilities

7 Separator Pressure

wline

l Flo izonta

Curve

6 Hor

Rate qL

8 Stock Tank Pressure Fig. 6-30 Graphical Representation of Total Producing System.

Introduction to Well Testing (Aug 1996)

6-37

Section 6

Introduction to Well Testing

Nodal Analysis

6-38

Section 7 Work Session

Section 7

Introduction to Well Testing

Work Session

7-2

Schlumberger

(Note: Section 6 has self contained work sessions in developing the Nodal Analysis concept) Work Session for Section 1 1. Define porosity and permeability, explain briefly how they can effect a reservoir’s performance?

2. As an example of the effect of grain size on a wetted surface, compare the wetted surface area of a 1 m3 rectangular conduit with a 1 m3 rectangular conduit filled with 0.1 mm diameter sand grains. (Porosity of the sand grain filled conduit is 20%) Hint:

Wetted surface = particle surface x number of particles. Number of particles = volume of cube divided by volume of sand grain.

Figure 7-1

Introduction to Well Testing

7-3

Section 7

Work Session

What conclusion do you draw from the result?

3. Define irreducible water saturation, discuss the relationship between this and permeability?

4. Explain the difference between saturated and unsaturated hydrocarbons?

Introduction to Well Testing

7-4

Schlumberger

5. Using the chart provided, figure 7-2, find Rsb, the solution gas-oil ratio at bubble point pressure, under the following conditions: pb

T wf γg γo

= 900 psia = 140 °F = 0.7

= 40 °API

Introduction to Well Testing

7-5

Section 7

Work Session

Figure 7-2 Introduction to Well Testing

7-6

Schlumberger

6. Given that the density of air at standard conditions is 0.001223 gm/cc or 0.0762 lb/cu ft, find the weight of 500 scf of gas with γg = 0.55.

7. Find the density of gas at standard conditions when γg = 0.7.

8. Find ρowf, the density of oil at well flowing conditions when; Bo Rs γg γo

= 1.21 = 350 cf/B = 0.75 = 30 °API

(Note a conversion factor of 1/(5.615) must be applied to the top right hand term of the equation for to convert from metric to English units)

Introduction to Well Testing

7-7

Section 7

Work Session

9. Repeat the same example with the chart provided, figure 7-3.

Introduction to Well Testing

7-8

Schlumberger

Figure 7-3

Introduction to Well Testing

7-9

Section 7

Work Session

Work Session for Section 2 1. Draw a typical completion diagram for a dual string oil producing well.? Explain how the well could be selectively produced from the upper zone only?

Introduction to Well Testing

7-10

Schlumberger

2. List the advantages disadvantages between a tubing retrievable and wireline retrievable surface controlled sub-surface safety valve?

Introduction to Well Testing

7-11

Section 7

Work Session

Work Session for Section 3 1. List several key objectives of well testing.

2. List the advantages and disadvantages of through tubing perforating and high shot density perforating.

Introduction to Well Testing

7-12

Schlumberger

3. Draw a typical DST string diagram for cased hole applications. Include TCP guns and an appropriate firing system.

Introduction to Well Testing

7-13

Section 7

Work Session

List 5 key points which should be noted from an operational viewpoint and/or from a data enhancement viewpoint.

4. The attached diagram, figure 7-4 shows a typical layout for a surface test onshore. Label the key items. List 5 key points which should be noted from an operational viewpoint and/or from a data enhancement viewpoint.

Introduction to Well Testing

7-14

Schlumberger

Water from ? Optional diesel supply for ? Optional gas supply from ? for a ? Fluids from well

Laboratory cabin

Tubing flare lines

Coflexip, Chiksans or fixed piping

Safety valve

Oil flare

Rig floor

Gas flare

Burning pit

Figure 7-4

Introduction to Well Testing

7-15

Section 7

Work Session

5. Discuss the merits between strain gauge and crystal gauge pressure sensors. List one application where you would use each type of sensor.

6. Discuss the difference between the resolution of a pressure transducer and that of a pressure gauge.

7. Discuss the importance of calibration with respect to data quality.

Introduction to Well Testing

7-16

Schlumberger

Work Session for Section 4 1. Outline the procedures required to condition a well prior to bottom hole sampling.

Why is this important? 2. Discuss some practical limitations to pre-conditioning prior to sampling.

Introduction to Well Testing

7-17

Section 7

Work Session

Work Session for Section 5 1. Re-write the following expressions; (a) Loge ab =

(b) Loge a/b =

(c) Loge xn =

2. Solve the following integrals; 1 (a) ∫ dx = x

(b) ∫ x n dx =

3. Starting from a basic cylindrical model for the flow of an incompressible fluid through a homogeneous system as shown in figure 7-5; and given that from Darcy’s law and the equation of state, it can be shown that; Q k dP = 2πrh µ dr Show that the radial flow equation can be derived as follows; Q =

Introduction to Well Testing

2πkh(Pe − Pw ) r  µ log e  e   rw 

7-18

Schlumberger

Figure 7-5

where

Q k h µ Pe Pw re rw

= = = = = = = =

volume rate of flow, cc/sec. permeability, darcys. thickness, cm. viscosity, centipoise. pressure at external boundary, atm. pressure at internal boundary, atm. radius to external boundary, cm. radius to internal boundary, cm.

Introduction to Well Testing

7-19

Section 7

Work Session

4. A homogeneous infinite reservoir is flowed for 48 hours at a constant stable rate of 4600 bbls/day. The initial static pressure prior to flowing was 4438 psi and the final flowing pressure was 2835 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

75 feet. 1.4 cp. 1.47

ct rw φ

= = =

15 x 10-6 psi-1. 4 inches. 22 %.

A strain gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) After Shut-In 0.00 0.17 0.33 0.50 0.67 1.00 1.50 2.00 2.50 3.00 3.50 4.00 5.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 Introduction to Well Testing

Pressure (psi) 2853 3430 3480 3529 3571 3636 3726 3840 3960 4070 4141 4180 4229 4247 4267 4283 4297 4307 4315 4323 4330 4335 4342 4345 4351 4354 4358 4362 4366 4368 7-20

Schlumberger

Using the Horner semi-log method, calculate the Horner time and then plot the data on a semi-log graph and calculate the following; (a) Initial Reservoir pressure (from the graph)

(b) Permeability.

(c) Skin Effect.

Use;

kh =

162.6 qµB m

Introduction to Well Testing

p  − p wf k s = 11513 .  1hour − log 10 + 3 . 23  m   φµc t r w2 7-21

Section 7

Work Session

5. A homogeneous infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a constant stable rate of 900 bbls/day. The initial static pressure prior to flowing was 2500 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

53 feet. 1.5 cp. 1.2

ct rw φ

= =

=

1.0 x E-05 psi-1. 0.29 feet.. 15 %.

A crystal gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) After Shut-In 1.27 E-03 1.69 E-03 2.54 E-03 3.38 E-03 4.23 E-03 5.08 E-03 6.77 E-03 8.47 E-03 1.21 E-02 1.69 E-02 2.54 E-02 3.38 E-02 4.23 E-02 5.08 E-02 6.77 E-02 1.27 E-01 1.69 E-01 2.54 E-01 3.38 E-01 4.23 E-01 5.08 E-01 6.77 E-01 8.47 E-01 1.27 E+00 1.69 E+00 2.52 E+00 3.38 E+00 4.23 E+00 5.08 E+00 6.77 E+00 8.47 E+00 Introduction to Well Testing

Pressure (psi) 2.72 E+01 3.57 E+01 5.25 E+01 6.85 E+01 8.39 E+01 9.88 E+01 1.27 E+02 1.53 E+02 2.13 E+02 2.65 E+02 3.50 E+02 4.16 E+02 4.69 E+02 5.13 E+02 5.79 E+02 7.06 E+02 7.54 E+02 8.13 E+02 8.50 E+02 8.76 E+02 8.97 E+02 9.29 E+02 9.53 E+02 9.95 E+02 1.02 E+03 1.06 E+03 1.09 E+03 1.11 E+03 1.13 E+03 1.16 E+03 1.18 E+03

Pressure Derivative 2.98 E+01 3.45 E+01 4.96 E+01 6.32 E+01 7.62 E+01 8.84 E+01 1.10 E+02 1.37 E+02 1.60 E+02 1.79 E+02 2.22 E+02 2.34 E+02 2.37 E+02 2.34 E+02 2.21 E+02 1.79 E+02 1.58 E+02 1.34 E+02 1.23 E+02 1.16 E+02 1.13 E+02 1.09 E+02 1.06 E+02 1.02 E+02 1.01 E+02 9.91 E+01 9.76 E+01 9.80 E+01 9.79 E+01 9.74 E+01 9.71 E+01 7-22

Schlumberger

Plot the data on a log-log graph of elapsed time versus pressure change (derivative) and using the appropriate type curve, figure 7-6, determine; (i) Pressure match

pD ∆p

=

(ii) Time match

t D / CD ∆t

=

(iii) Curve match

C De2 s

=

Then calculate the following using the equations listed below; (a) Permeability.

(b) Wellbore storage ratio.

Introduction to Well Testing

7-23

Section 7

Work Session

(c) Skin Effect.

Use;

p  kh = 1412 . qBµ D  match  ∆p 

(

)

 C D e 2s Match  s = 05 . log e    CD  

Introduction to Well Testing

C=

0.000295 kh µ t /C   D D  Match  ∆t  

CD =

0.8936C

φ ct h r w

2

7-24

Schlumberger

Figure 7-6

Introduction to Well Testing

7-25

Section 7

Work Session

6. A homogeneous infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a constant stable rate of 150 bbls/day. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

50 feet. 2.5 cp. 1.05

ct rw

= = =

φ

4.2 x 10-6 psi-1. 0.29 feet.. 25 %.

A crystal gauge recorder placed beneath the tester valve recorded data which produced the loglog plots and semi-log plot shown in figures 7-7 and 7-8.

Using the appropriate type curve determine; (i) Pressure match

pD ∆p

=

(ii) Time match

t D / CD ∆t

=

(iii) Curve match

C De2 s

=

Then calculate the following using the equations listed below; (a) Permeability.

Introduction to Well Testing

7-26

Schlumberger

(b) Wellbore storage ratio.

(c) Skin Effect.

Using the semi-log plot, determine; (d) Permeability.

Introduction to Well Testing

7-27

Section 7

Work Session

(e) Skin Effect.- Given that s can be determined to equal 2.43, discuss the values for K and s as determined by the different methods.

Use the following equations; p  kh = 1412 . qBµ D  match  ∆p 

(

)

 C D e 2s Match  s = 05 . log e    CD  

Introduction to Well Testing

C=

0.000295 kh µ t /C   D D  Match  ∆t   CD =

0.8936C

φ ct h r w

2

7-28

Schlumberger

Figure 7-7 Introduction to Well Testing

7-29

Section 7

Work Session

Figure 7-8 Introduction to Well Testing

7-30

Schlumberger

7. A double porosity infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a stable rate of 900 bbls/day. The initial static pressure prior to flowing was 4200 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

53 feet. 1.5 cp. 1.2

ct rw φ

= =

=

1.0 x E-05 psi-1. 0.29 feet.. 15 %.

A crystal gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) 1.00 E-03 1.25 E-03 1.50 E-03 2.00 E-03 2.50 E-03 3.00 E-03 4.00 E-03 5.00 E-03 7.50 E-03 1.00 E-02 1.25 E-02 1.50 E-02 2.00 E-02 2.50 E-02 3.00 E-02 4.00 E-02 5.00 E-02 7.50 E-02 0.100 0.125 0.150 0.200 0.250 0.300 0.400 0.500 0.750 1.000 1.250 1.500 2.000 Introduction to Well Testing

Pressure Change (psi) 3.47 4.24 4.99 6.43 7.77 9.05 11.41 13.55 18.15 21.92 25.06 27.71 31.92 35.08 37.50 40.88 43.05 45.87 47.09 47.70 48.07 48.53 48.88 49.19 49.80 50.39 51.78 53.07 54.26 55.36 57.34

Pressure Derivative 3.48 3.83 4.45 5.58 6.56 7.46 8.99 10.22 12.37 13.65 14.34 14.58 14.36 13.66 12.68 10.60 8.73 5.36 3.41 2.35 1.86 1.58 1.66 1.88 2.41 2.92 4.04 4.95 5.73 6.38 7.39 7-31

Section 7

Work Session

Time (Hours)

Pressure Change (psi)

Pressure Derivative

2.500 3.000 4.000 5.000 7.500 10.000 12.500 15.000 20.000 25.000 30.000 40.000 50.000 75.000 100.000

59.08 60.61 63.19 65.29 69.25 72.10 74.34 76.14 79.01 81.24 83.06 85.93 88.16 92.21 95.09

8.11 8.61 9.23 9.55 9.85 9.94 9.96 9.97 9.98 9.98 9.99 9.99 9.99 9.99 10.00

Plot the data on a log-log graph of elapsed time versus pressure change (derivative) and using the appropriate type curve determine;

(i) Pressure match

pD ∆p

=

(ii) Time match

t D / CD ∆t

=

(iii) Curve match

( C D e 2 s )f

=

(iv) Curve Match

( C D e 2 s )f+m

=

(v) Curve Match

λ e−2s

=

Introduction to Well Testing

7-32

Schlumberger

Then calculate the following using the equations listed below; (a) Permeability.

(b) Wellbore storage ratio.

(c) Skin Effect.

Introduction to Well Testing

7-33

Section 7

Work Session

(d)

The Storativity ratio ω.

(e)

The interporosity flow parameter λ.

Use the following equations; p  kh = 1412 . qBµ D  match .  ∆p  C=

0.000295 kh . µ t /C   D D  Match  ∆t  

(

)

 C D e 2s Match . s = 05 . log e    CD  

CD =

0.8936C

φ ct h r w

2

λ e −2 s) (C D e 2s) ( f +m f +m λ= CD

C De 2 s) ( f+m ω= 2 s (C De )f Introduction to Well Testing

7-34

Work Session Answers

Work Session Answers

Introduction to Well Testing

8-2

Schlumberger

(Note: Section 6 has self contained work sessions in developing the Nodal Analysis concept) Work Session for Section 1 1. Define porosity and permeability, explain briefly how they can effect a reservoir’s performance? Porosity is defined as the percentage or fraction of void to the bulk volume of a rock. Only the effective or interconnected porosity is usefull in reservoir engineering as this represents the ammount of void which can contribute to flow. Permeability is the ease with which, under non-turbulent flow conditions, a fluid can flow through a porus rock. It is therefore a function, among other things of the degree of interconnection between the pores of the rock. 2. As an example of the effect of grain size on a wetted surface, compare the wetted surface area of a 1 m3 rectangular conduit with a 1 m3 rectangular conduit filled with 0.1 mm diameter sand grains. (Porosity of the sand grain filled conduit is 20%) Hint:

Wetted surface = particle surface x number of particles. Number of particles = volume of cube divided by volume of sand grain.

Figure 7-1

Surface area = 4 m2 for the rectangular conduit. Surface area for the filled conduit = Surface area of spheres x number of spheres. (1 − φ ) = 3 (1 − φ ) 3 (0 .8) = 4πr2 x 4 3 0 .0001 r πr 3 = 24000 m2.

Introduction to Well Testing

8-3

Work Session Answers What conclusion do you draw from the result? Grain size plays a crucial role in a rock’s permeability as the smaller the grain size, the greater the surface area in contact with the reservoir fluid and thus the greater the friction to flow. 3. Define irreducible water saturation, discuss the relationship between this and permeability? Irreducible water saturation defines the amount of water saturation that cannot be reduced by displacement by a non-wetting phase no matter how great a pressure is applied to the system. (The amount of water that cannot be displaced from the system). Irreducible water saturation is effected by the grain size, as the smaller the gain size the greater the capillary forces in the system and also the greater the surface area of the gains for the water wet phase to be affected by (problem 2 above). As grain size is also directly related to permeability it follows that there is a direct relationship between the irreducible water saturation of a system and its permeability. This is a very important concept in reservoir engineering as this leads onto the concept of residual oil or irreducible oil saturation i.e. the amount of oil contained in the system which can not be displaced. 4. Explain the difference between saturated and unsaturated hydrocarbons? Saturated hydrocarbons exist when the carbon atoms in their molecular structure are connected with single bonds. Unsaturated hydrocarbons exist when the carbon atoms in their molecular structure form carbon to carbon double bonds. These compounds can add hydrogen to their structures under appropriate conditions and are therefor said to be unsaturated (with hydrogen).

Introduction to Well Testing

8-4

Schlumberger

5. Using the chart provided, figure 7-2, find Rsb, the solution gas-oil ratio at bubble point pressure, under the following conditions: pb

T wf γg γo

= 900 psia = 140 °F = 0.7

= 40 °API

Using the chart in figure 7-2, proceed as follows; (a) From right to left: draw a line through γo = 40 ° and γg = 0.7 to locate a point on line B. (b) From that point, draw a line through pb = 900 psia to line A. (c) From there, draw a line to Twf = 140 °F. (d) Read the answer which is Rsb = 220 scf/bbl.

Introduction to Well Testing

8-5

Work Session Answers

Figure 7-2 Introduction to Well Testing

8-6

Schlumberger

6. Given that the density of air at standard conditions is 0.001223 gm/cc or 0.0762 lb/cu ft, find the weight of 500 scf of gas with γg = 0.55. Weight

= Vg x ρair x γg. = 500 x 0.0762 x 0.55 = 20.95 lbs.

7. Find the density of gas at standard conditions when γg = 0.7.

ρg

= ρair x γg = 0.7 x 0.001223 = 0.000856 gm/cc.

8. Find ρowf, the density of oil at well flowing conditions when; Bo Rs γg γo

= 1.21 = 350 cf/B = 0.75 = 30 °API

(Note a conversion factor of 1/(5.615) must be applied to the top right hand term of the equation for to convert from metric to English units)

ρosc

ρowf

=

=

141.5 = 0.876 1315 . + 30.0

ρ osc +

ρ airγ g Rs

5.615 Bo

=

0.876 +

0 .001223 x 0.75 x 350 5.615 1.21

= 0.771 gm/cc.

Introduction to Well Testing

8-7

Work Session Answers 9. Repeat the same example with the chart provided, figure 7-3.

Using the chart in figure 7-3, proceed as follows;

(a) Draw a line from γg = 0.75 through Rs = 350 cf/B to line A.

(b) Draw a line from point a to γo = 30 °API and establish point b. (c) Draw a line from point b through Bo = 1.21 to the answer

ρowf = 0.77 gm/cc.

Introduction to Well Testing

8-8

Schlumberger

Figure 7-3

Introduction to Well Testing

8-9

Work Session Answers Work Session for Section 2 1. Draw a typical completion diagram for a dual string oil producing well.? Explain how the well could be selectively produced from the upper zone only? Refer to figure 2-7b. The lower string well could be shut-in at surface and only the upper zone produced, or a plug could be set in the landing nipple below the packer in the long string, the sliding sleeve opened and the well produced either up both strings from the upper zone or through either.

Introduction to Well Testing

8-10

Schlumberger

2.

List the advantages disadvantages between a tubing retrievable and wireline retrievable surface controlled sub-surface safety valve? A selection from the following; Tubing Retrievable: Advantages; Large internal diameter implies larger flow rates. Less turbulence therefore less erosion. Fullbore access for intervention work. Can contain a profile to accept a wireline retrievable valve in the event of problems. Disadvantages; Can not readily retrieve the valve for repair or maintenance (although some modern versions have critical parts removable) Cost. Large size make them difficult to handle and maintain. Wireline Retrievable: Advantages; Can readily be retrieved for repair or maintenance. Cost. Easy to handle and maintain. Disadvantages; Smaller internal diameters giving restriction to flow and creating turbulence and erosion. Needs to be removed for intervention work - especially the smaller sizes.

Introduction to Well Testing

8-11

Work Session Answers Work Session for Section 3 1. List several key objectives of well testing. Select any of the following;

• Productivity well tests are conducted to; ∗ ∗ ∗ ∗ ∗ ∗ ∗

Identify produced fluids and determine their respective volume ratios. Measure reservoir pressure and temperature. Obtain samples suitable for PVT analysis. Determine well deliverability. Evaluate completion efficiency. Characterise well damage. Evaluate workover or stimulation treatment.

• Descriptive tests seek to; ∗ ∗ ∗ ∗

Evaluate reservoir parameters. Characterise reservoir heterogenities. Asses reservoir extent and geometry. Determine hydraulic communication between wells.

2. List the advantages and disadvantages of through tubing perforating and high shot density perforating.

• Through Tubing Perforating ∗ The wellhead and completion string are in place and tested before perforation. ∗ The underbalanced differential from the reservoir into the wellbore provides perforation clean-up. ∗ Perforations may be made as required over the life of the well, with or without a rig onsite. ∗ Operating Times are low , giving good efficiency. ∗ Restricted gun size and length. ∗ More debris especially when carrier type guns not used.

• High Shot Density Perforating ∗ ∗ ∗ ∗ ∗

Large gun sizes can be used implying optimal perforation program. Long gun strings can be used. reduced Perforation debris. Requires a rig. Mis-runs are costly in time and therefore money.

Introduction to Well Testing

8-12

Schlumberger

3. Draw a typical DST string diagram for cased hole applications. Include TCP guns and an appropriate firing system. Use figure 3-12 or 3-25 as reference.

Introduction to Well Testing

8-13

Work Session Answers List 5 key points which should be noted from an operational viewpoint and/or from a data enhancement viewpoint. Many choices, some typical ones are; Ensure all components are correctly rated in terms of expected maximum pressure, differential pressure and temperature. Ensure internal diameters are sufficient for any wireline/slickline runs required throughout the test. Ensure correct explosives are chosen in relation to expected temperatures. Ensure gauges are calibrated over the expected temperature and pressure range Ensure gauge resolution and memory capacity is adequate for the test expected. 4. The attached diagram, figure 7-4 shows a typical layout for a surface test onshore. Label the key items. Use figure 3-30 as guidance. List 5 key points which should be noted from an operational viewpoint and/or from a data enhancement viewpoint. Many choices, some typical ones are; Ensure all components are correctly rated in terms of expected maximum pressure, and temperature. Ensure equipment is designed to handle any hostile effluents such as H2S, CO2 etc. Ensure equipment is designed to handle expected flowrates. Ensure surface piping is thoroughly flushed to clear of any debris. Ensure all measurement devices have recently been calibrated. Ensure gauge resolution and memory capacity is adequate for the test expected.

Introduction to Well Testing

8-14

Schlumberger

Gauge tank

Separator

Heater Water from separator Optional diesel supply for diesel-fired heater Optional gas supply from separator for a gas-fired heater

Oil manifold

Fluids from well

Choke manifold

Transfer pump Tubing flare lines

Laboratory cabin

Flowhead

Safety valve

Oil flare

Coflexip, Chiksans or fixed piping

Rig floor

Gas flare

Burning pit

Figure 7-4

Introduction to Well Testing

8-15

Work Session Answers 5. Discuss the merits between strain gauge and crystal gauge pressure sensors. List one application where you would use each type of sensor. Strain: in general they are rugged, low cost and have good dynamic behaviour, they are ideal for hostile environments where high pressures and high temperatures are expected, also can be placed close to TCP guns as they can handle the expected shocks. Crystal They are extremely accurate and do not drift as much as strain gauges. they are however costly and can be fragile. they are ideal for testing gas wells with high permeabilities where resolution is crucial, they are also ideal for permanent installations. 6. Discuss the difference between the resolution of a pressure transducer and that of a pressure gauge. The resolution of a transducer is covers the function of the transducer only, transducers are however, part of a more complete assembly namely the pressure gauge, which includes additional electronics. the resolution of interest is the resolution of the complete assembly as this is what is used to make the measurement. 7. Discuss the importance of calibration with respect to data quality. A pressure gauge has as a basic reference a master calibration, this verifies the response of the gauge over its entire operating range and provides a set of coefficients to mathematically model the gauge’s response to the actual value being measured. Master calibrations need to be up to date in order for the gauge to give reliable and meaningful data. Pre and post job calibrations can yield even better data as they are performed over the expected pressure and temperature range just prior to and after the job.

Introduction to Well Testing

8-16

Schlumberger

Work Session for Section 4 1. Outline the procedures required to condition a well prior to bottom hole sampling. a) Sampling of a flowing undersaturated well (when GOR = GORi = constant)

• The well should be flowing for at least 24 hours, at a minimum stable flow rate ensuring a maximum column height of monophasic fluid. • The pressure at sampling depth must be at least 100 to 200 psi higher than the saturation pressure in the fluid column. A good figure is 500 psi. • The well should be clean - in order to eliminate traces of contaminated oil or water, the stable flowing period should be preceded by a production period equal to 5 to 10 times the total volume of the tubing string. • The flowing stability can be checked by; ♦ Stabilised gas flow rate, oil flow rate and GOR. ♦ Stabilised well head pressure. ♦ Stabilised pwf (best way to ensure flowing stability). b) Sampling of a shut-in saturated well (when GOR ≥ GORi)

• The flow rate should be progressively reduced and then the well shut-in until a stabilised static pressure is reached. A minimum of 12 hours should elapse before sampling is made, with a good figure being 24 hours. The shut-in period, however, can be established according to the build-up data or according to the stability of the wellhead pressure. • A static pressure gradient will be very helpful in detecting a possible presence of water. • When the sampler is at sampling depth, the well should be opened on the smallest possible choke only to fill the casing around the sampler with fresh reservoir oil, and then shut-in. Why is this important? To ensure that the samples are representative. 2. Discuss some practical limitations to pre-conditioning prior to sampling. Time used to condition the well can often be too long and therefore expensive in rig time. A sensible compromise needs to be derived. Introduction to Well Testing

8-17

Work Session Answers Work Session for Section 5 1. Re-write the following expressions; (a) Loge ab = Loge a + Loge b

(b) Loge a/b = Loge a - Loge b

(c) Loge xn = n Loge x

2. Solve the following integrals; 1 (a) ∫ dx = Loge x + c x

x n+ 1 +c (b) ∫ x dx = n+1 n

3. Starting from a basic cylindrical model for the flow of an incompressible fluid through a homogeneous system as shown in figure 7-5; and given that from Darcy’s law and the equation of state, it can be shown that; Q k dP = 2πrh µ dr Show that the radial flow equation can be derived as follows; Q =

Introduction to Well Testing

2πkh(Pe − Pw ) r  µ log e  e   rw 

8-18

Schlumberger

Figure 7-5

where

Q k h µ Pe Pw re rw

= = = = = = = =

Starting from Pe

∫ dP =

Pw

volume rate of flow, cc/sec. permeability, darcys. thickness, cm. viscosity, centipoise. pressure at external boundary, atm. pressure at internal boundary, atm. radius to external boundary, cm. radius to internal boundary, cm.

Q k dP = separating the variables gives µ dr 2π rh

µQ re 1 ∫ dr 2πkh r r w

integrating between the limits gives, Qµ r  Pe − Pw = log e  e  2πkh  rw  re-arranging gives; Q =

2πkh( Pe − Pw) r  µ log e  e   rw 

Introduction to Well Testing

8-19

Work Session Answers 4. A homogeneous infinite reservoir is flowed for 48 hours at a constant stable rate of 4600 bbls/day. The initial static pressure prior to flowing was 4438 psi and the final flowing pressure was 2835 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

75 feet. 1.4 cp. 1.47

ct rw φ

= = =

15 x 10-6 psi-1. 4 inches. 22 %.

A strain gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) After Shut-In 0.00 0.17 0.33 0.50 0.67 1.00 1.50 2.00 2.50 3.00 3.50 4.00 5.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 28.00 30.00 32.00 34.00 36.00 38.00 Introduction to Well Testing

Pressure (psi) 2853 3430 3480 3529 3571 3636 3726 3840 3960 4070 4141 4180 4229 4247 4267 4283 4297 4307 4315 4323 4330 4335 4342 4345 4351 4354 4358 4362 4366 4368 8-20

Schlumberger

Using the Horner semi-log method, calculate the Horner time and then plot the data on a semi-log graph and calculate the following; (a) Initial Reservoir pressure (from the graph) From graph, extrapolated pressure p* = 4440 psi

(b) Permeability. m = 205 psi/cycle from the graph. So using kh = so k =

162.6 qµB 1626 . qµ B , k= m mh

1626 . x 4600 x 1.4 x 1.47 ≈ 100 md. 205 x 75

(c) Skin Effect. p  − p wf k .  1hour − log 10 + 3 . 23  Using s = 11513 m   φµc t r w2  4092 − 2853  100 s = 11513 .  − log 10 + 3 . 23  205   0.22 x 1.4 x 15x10 -6 x (0.33)2 s = -6.31

Use;

kh =

162.6 qµB m

Introduction to Well Testing

p  − p wf k s = 11513 .  1hour − log 10 + 3 . 23  m   φµc t r w2

8-21

Work Session Answers 5. A homogeneous infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a constant stable rate of 900 bbls/day. The initial static pressure prior to flowing was 2500 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

53 feet. 1.5 cp. 1.2

ct rw φ

= =

=

1.0 x E-05 psi-1. 0.29 feet.. 15 %.

A crystal gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) After Shut-In 1.27 E-03 1.69 E-03 2.54 E-03 3.38 E-03 4.23 E-03 5.08 E-03 6.77 E-03 8.47 E-03 1.21 E-02 1.69 E-02 2.54 E-02 3.38 E-02 4.23 E-02 5.08 E-02 6.77 E-02 1.27 E-01 1.69 E-01 2.54 E-01 3.38 E-01 4.23 E-01 5.08 E-01 6.77 E-01 8.47 E-01 1.27 E+00 1.69 E+00 2.52 E+00 3.38 E+00 4.23 E+00 5.08 E+00 6.77 E+00 8.47 E+00 Introduction to Well Testing

Pressure (psi) 2.72 E+01 3.57 E+01 5.25 E+01 6.85 E+01 8.39 E+01 9.88 E+01 1.27 E+02 1.53 E+02 2.13 E+02 2.65 E+02 3.50 E+02 4.16 E+02 4.69 E+02 5.13 E+02 5.79 E+02 7.06 E+02 7.54 E+02 8.13 E+02 8.50 E+02 8.76 E+02 8.97 E+02 9.29 E+02 9.53 E+02 9.95 E+02 1.02 E+03 1.06 E+03 1.09 E+03 1.11 E+03 1.13 E+03 1.16 E+03 1.18 E+03

Pressure Derivative 2.98 E+01 3.45 E+01 4.96 E+01 6.32 E+01 7.62 E+01 8.84 E+01 1.10 E+02 1.37 E+02 1.60 E+02 1.79 E+02 2.22 E+02 2.34 E+02 2.37 E+02 2.34 E+02 2.21 E+02 1.79 E+02 1.58 E+02 1.34 E+02 1.23 E+02 1.16 E+02 1.13 E+02 1.09 E+02 1.06 E+02 1.02 E+02 1.01 E+02 9.91 E+01 9.76 E+01 9.80 E+01 9.79 E+01 9.74 E+01 9.71 E+01 8-22

Schlumberger

Plot the data on a log-log graph of elapsed time versus pressure change (derivative) and using the appropriate type curve, figure 7-6, determine; (i) Pressure match

pD ∆p

= 0.52/100 = 0.0052

(ii) Time match

t D / CD ∆t

= 11/0.1 = 110

(iii) Curve match

C De2 s

= 100

Then calculate the following using the equations listed below; (a) Permeability. p  Using kh = 1412 . qBµ D  match  ∆p  k=

1412 . x 900 x 1.2 x 1.5 x 0.0052 53

k = 22.44 md. (b) Wellbore storage ratio. Using C =

C=

kh 0 .000295 gives µ t / C  D  Match  D   t ∆  

22 .44 x 53 x 0.000295 15 . x 110

∴ C = 2.13 x 10-3 bbl/psi.

Introduction to Well Testing

8-23

Work Session Answers (c) Skin Effect. Now C D =

0.8936C

0 .8936 x 0.00213 ∴ = = 284.68 C D 2 0.15 x 0.00001 x 53 x 0.29 x 0.29 φ ct h r w

(

)

 C D e 2s Match ⇒ . log e  Using s = 05   C D    100  s = 0 .5 loge    28468 . 

∴ s = -0.52

Use;

p  kh = 1412 . qBµ D  match  ∆p 

(

)

 C D e 2s Match  s = 05 . log e    CD  

Introduction to Well Testing

C=

0.000295 kh µ t /C   D D  Match  ∆t  

CD =

0.8936C

φ ct h r w

2

8-24

Schlumberger

Figure 7-6

Introduction to Well Testing

8-25

Work Session Answers 6. A homogeneous infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a constant stable rate of 150 bbls/day. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

50 feet. 2.5 cp. 1.05

ct rw

= = =

φ

4.2 x 10-6 psi-1. 0.29 feet.. 25 %.

A crystal gauge recorder placed beneath the tester valve recorded data which produced the loglog plots and semi-log plot shown in figures 7-7 and 7-8.

Using the appropriate type curve determine; (i) Pressure match

pD ∆p

= 5.8 x 10-3

(ii) Time match

t D / CD ∆t

= 200

(iii) Curve match

C De2 s

= 10000

Then calculate the following using the equations listed below; (a) Permeability. p  Using kh = 1412 . qBµ D  match  ∆p  k=

1412 . x 150 x 1.05 x 2.5 x 0.0058 50

k = 6.45 md.

Introduction to Well Testing

8-26

Schlumberger

(b) Wellbore storage ratio. Using C =

C=

kh 0 .000295 gives µ t / C  D  Match  D   ∆t  

6 .45 x 50 x 0.000295 2.5 x 200

∴ C = 1.9 x 10-4 bbl/psi. (c) Skin Effect. Now C D =

0.8936C

0 .8936 x 0.00019 ∴ = = 38.45 C D 2 0.25 x 0.0000042 x 50 x 0.29 x 0.29 φ ct h r w

(

)

 C D e 2s Match ⇒ . log e  Using s = 05   C D    10000  s = 0 .5 log e    38 .45 

∴ s = 2.78 Using the semi-log plot, determine; (d) Permeability. m = 204.5 psi/cycle from the graph. So using kh = so k =

162.6 qµB 1626 . qµ B , k= m mh

162.6 x 150 x 2 .5 x 1.05 ≈ 6.26 md. 204 .5 x 50

Introduction to Well Testing

8-27

Work Session Answers (e) Skin Effect.- Given that s can be determined to equal 2.43, discuss the values for K and s as determined by the different methods. The semi-log method is a good approximation to the results and can also be used as a control to the log-log method. We would assume that the log-log method gives more accurate approximations as it is based on a more complex model and more closely modelled function.

Use the following equations; p  kh = 1412 . qBµ D  match .  ∆p 

(

)

 C D e 2s Match  s = 05 . log e    C D  

Introduction to Well Testing

C=

0.000295 kh µ t /C   D D  Match  ∆t   CD =

0.8936C

φ ct h r w

2

8-28

Schlumberger

Figure 7-7 Introduction to Well Testing

8-29

Work Session Answers

Figure 7-8 Introduction to Well Testing

8-30

Schlumberger

7. A double porosity infinite acting reservoir with wellbore storage and skin is subjected to a constant drawdown test and build-up. The well was flowed at a stable rate of 900 bbls/day. The initial static pressure prior to flowing was 4200 psi. Data derived from core analysis, and the logging program revealed the following data; h µ Bo

= = =

53 feet. 1.5 cp. 1.2

ct rw φ

= =

=

1.0 x E-05 psi-1. 0.29 feet. 15 %.

A crystal gauge recorder placed beneath the tester valve revealed the following data; Time (Hours) 1.00 E-03 1.25 E-03 1.50 E-03 2.00 E-03 2.50 E-03 3.00 E-03 4.00 E-03 5.00 E-03 7.50 E-03 1.00 E-02 1.25 E-02 1.50 E-02 2.00 E-02 2.50 E-02 3.00 E-02 4.00 E-02 5.00 E-02 7.50 E-02 0.100 0.125 0.150 0.200 0.250 0.300 0.400 0.500 0.750 1.000 1.250 1.500 2.000 Introduction to Well Testing

Pressure Change (psi) 3.47 4.24 4.99 6.43 7.77 9.05 11.41 13.55 18.15 21.92 25.06 27.71 31.92 35.08 37.50 40.88 43.05 45.87 47.09 47.70 48.07 48.53 48.88 49.19 49.80 50.39 51.78 53.07 54.26 55.36 57.34

Pressure Derivative 3.48 3.83 4.45 5.58 6.56 7.46 8.99 10.22 12.37 13.65 14.34 14.58 14.36 13.66 12.68 10.60 8.73 5.36 3.41 2.35 1.86 1.58 1.66 1.88 2.41 2.92 4.04 4.95 5.73 6.38 7.39 8-31

Work Session Answers

Time (Hours)

Pressure Change (psi)

Pressure Derivative

2.500 3.000 4.000 5.000 7.500 10.000 12.500 15.000 20.000 25.000 30.000 40.000 50.000 75.000 100.000

59.08 60.61 63.19 65.29 69.25 72.10 74.34 76.14 79.01 81.24 83.06 85.93 88.16 92.21 95.09

8.11 8.61 9.23 9.55 9.85 9.94 9.96 9.97 9.98 9.98 9.99 9.99 9.99 9.99 10.00

Plot the data on a log-log graph of elapsed time versus pressure change (derivative) and using the appropriate type curve determine;

(i) Pressure match

pD ∆p

= 0.5 / 10 = 0.005

(ii) Time match

t D / CD ∆t

= 210

(iii) Curve match

( C D e 2 s )f

= 10

(iv) Curve Match

( C D e 2 s )f+m

= 0.3

(v) Curve Match

λ e−2s

= 8 x 10-3

Introduction to Well Testing

8-32

Schlumberger

Then calculate the following using the equations listed below; (a) Permeability. p  Using kh = 1412 . qBµ D  match  ∆p  k=

1412 . x 900 x 1.2 x 1.5 x 0.05 53

k = 215.8 md. (b) Wellbore storage ratio. Using C =

C=

kh 0 .000295 gives µ t / C  D  Match  D   ∆t  

215.8 x 53 x 0.000295 15 . x 210

∴ C = 0.011 bbl/psi. (c) Skin Effect. Now C D =

0.8936C

0 .8936 x 0.011 ∴ = C D 0.15 x 0.00001 x 53 x 0.29 x 0.29 = 1470.2 2 φ ct h r w

(

)

 C D e 2s Match ⇒ . log e  Using s = 05   CD    10  s = 0.5 log e    1470 .2 

∴ s = -2.5

Introduction to Well Testing

8-33

Work Session Answers (d)

The Storativity ratio ω.

(C D e 2 s)f +m Now ω = (C D e 2s)f ∴ω=

0 .3 = 0.03 10

(e) Using λ =

λ=

The interporosity flow parameter λ.

(λ e −2s)f +m (CD e2 s)f +m CD

0 .008 x 0 .3 = 1.6 x 10-6 1470 .2

Use the following equations; p  kh = 1412 . qBµ D  match .  ∆p  C=

0.000295 kh . µ t /C   D D  Match  ∆t  

(

)

 C D e 2s Match . s = 05 . log e    CD  

CD =

0.8936C

φ ct h r w

2

λ e −2 s) (C D e 2s) ( f +m f +m λ= CD

C De 2 s) ( f+m ω= 2 s (C De )f

Introduction to Well Testing

8-34

Schlumberger

REFERENCES Chapter 1 Source documents; 1. “Reservoir and Production Fundamentals” - Schlumberger 1980. 2. “Reservoir Fluids Sampling Fundamentals” - Schlumberger 1990. Chapter 2 Source documents; 1. “Perforating Services Manual” - Reference number SMP-7043-1 - Schlumberger 1995. Individual documents; 1. “Packers and Completion Accessories Catalog” - Camco 1986. 2. “Subsurface Safety Systems” - Camco 1991. Chapter 3 Source documents; 1. “Modern Reservoir Testing” - Reference number SMP-7055 - Schlumberger 1994. 2. “Perforating Services Manual” - Reference number SMP-7043-1 - Schlumberger 1995. 3. “Downhole Testing Services Catalogue” - Reference number SMP-7015 - Schlumberger 1991. 4. “Surface Testing Services Catalogue” - Reference number SMP-7042 - Schlumberger 1992. 5. “Pressure Gauges Review” - Reference number M-073596 - Schlumberger 1990. Chapter 4 Source documents; 1. “Reservoir Fluids Sampling Fundamentals” - Schlumberger 1990 2. “Surface Testing Services Catalogue” - Reference number SMP-7042 - Schlumberger 1992 3. “Wireline Formation Testing and Sampling” - Reference number SMP ?- - Schlumberger 1996

Introduction to Well Testing

Ref-1

References Chapter 5 Source documents; 1. “Modern Reservoir Testing” - Reference number SMP-7055 - Schlumberger 1994. Individual documents; 1. “Well Test Interpretation Course Manual” - written by Flopetrol, Melun France 1983 - Updated 1991.. Chapter 6 Source documents; 1. “A Nodal Approach for Applying Systems Analysis to the Flowing and Artificial Lift Oil or Gas Well” - Joe Mach, Eduardo Proano (Schlumberger) and Kermit Brown (University of Tulsa). 2. “Systems Analysis as Applied to Producing Wells” - Joe Mach, Eduardo Proano (Schlumberger) and Kermit Brown (University of Tulsa). Individual References; 1. Vogel, J.V., “Inflow Performance Relationships For Solution-Gas Drive Wells”, J. Pet. Tech. (January 1968) 83 - 92. 2. Standing, M.B., “Upflow Performance Relationships for Damaged Wells Producing by Solution - Gas Drive”, J. PET. Tech. (November 1970) 1399 - 1400. 3. Standing, M.B., “Concerning the Calculation of Inflow Performance of wells Producing From Solution Gas Drive Reservoirs”, J. Pet. Tech. (September 1971) 1141 - 1142. 4. Fetkovich, M.J., “The Isochronal Testing of Oil Wells” SPE Paper No. 4529, 48th Annual Fall Mtg. of SPE of AIME, Las Vegas, Nevada, (September 30 - October 3, 1973). 5. Hagerdorn, A.R. and Brown, K.E.: “Experimental Study of Pressure Gradients Occurring During Continuous Two Phase Flow in Small Diameter Vertical Conduits,” J. Pet. Tech. (April 1965) 475 484. 6. Orkiszewski, J., “Predicting Two-Phase Pressure Drops in Vertical Pipes”. J. Pet. Tech. (June 1967) 829 - 838. 7. Duns, H., and Ros, N.C.J.: “Vertical Flow of Gas and Liquid Mixtures in Wells,” Proc., 6th World Pet. Congress (1963) 451. 8. Beggs, H.D. and Brill, J.P.: “A Study of Two-Phase Flow in Inclined Pipes”, J. Pet. Tech. (May 1973) 607 - 617.

Introduction to Well Testing

Ref-2