Acoustic Fluid Level Measurements-1

For distribution by Petroleum Extension-The University of Texas at Austin

Contents i
Acoustic Fluid Level

Measurements in Oil and Gas
Wells Handbook
First Edition
by A. L. Podio and
James N. McCoy
published by
2017
ii ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
Library of Congress Cataloging-in-Publication Data

Names: Podio, A. L. | McCoy, James N., 1933-
Title: Acoustic fluid level measurements in oil and gas wells handbook / by
A.L. Podio and James N. McCoy.
Description: First edition. | Austin : Petroleum Extension, The University of
Texas at Austin, 2017. | Includes bibliographical references.
Identifiers: LCCN 2017001999 | ISBN 9780886982799 (alk. paper)
Subjects: LCSH: Fluid dynamic measurements. | Oil wells. | Gas wells. |
Liquids—Measurement. | Fluids—Acoustic properties—Measurement. | Sound
waves—Measurement.
Classification: LCC TA357.5.M43 P63 2017 | DDC 622/.338--dc23 LC record avail-
able at https://lccn.loc.gov/2017001999
Disclaimer
Although all reasonable care has been taken in preparing this publication,
the authors, the Petroleum Extension (PETEX) of the University of Texas
at Austin, and any other individuals and their affiliated groups involved in
preparing this content assume no responsibility for the consequences of its
use. Each recipient should ensure he or she is properly trained and informed
about the unique policies and practices regarding application of the informa-
tion contained herein. Any recommendations, descriptions, and methods in
this book are presented solely for educational purposes.
©2017 by the University of Texas at Austin

All rights reserved
First Edition
Printed in the United States of America
This book or parts thereof may not be reproduced in any form and the com-
mercial use of this content is strictly prohibited without the express permission
of PETEX, the University of Texas at Austin. PETEX publications contain
copyrighted material. Users may not resell, assign, distribute, or transfer all
or part of this content to reuse in any way; or create derivative works from
this content with explicit permission from the copyright owner.
Brand names, company names, trademarks, or other identifying symbols

appearing in illustrations or in the text are used for educational purposes
only and do not constitute an endorsement by the author or the publisher.
Catalog No. 3.14010

Editor: Mary Lin ISBN 0-88698-279-0
Graphic Designer: 978-0-88698-279-9
Debbie Caples
Cover art and The University of Texas at Austin is an equal opportunity institution.
graphics: E. K. Weaver No state tax funds were used to print this book.

Contents iii
Contents
Figures viii
Tables xiv
Foreword xv
Preface xvii
Acknowledgments xix
About the Authors xxi
1. Visualizing Well Performance 1-1
Determining Optimum and Current Well Performance 1-1
Well Performance Analysis from Acoustic Fluid Level Measurements 1-4
What Must Be Known for Well Performance Analysis? 1-4
Well Tests 1-5
Static Bottomhole Pressure 1-6
Producing Bottomhole Pressure 1-6
Calculating the PBHP 1-7
Calculating the SBHP 1-9
Public Domain Acoustic Fluid Level Software 1-10
Inflow Performance Relations 1-10
Pressure above the Bubble Point 1-12
Pressure below the Bubble Point 1-14
Determining the IPR from a Multi-Rate Flow Test 1-17
Estimating the IPR Relation from a One-Rate Well Test 1-18
Other Inflow Performance Models 1-20
Multi-Rate Flow Test Monitored with Acoustic Fluid Level Records 1-21
Summary 1-22
References 1-24
Bibliography 1-25
2. Examples of Fluid Level Surveys in Producing and Static Wells 2-1
Guidelines for Acoustic Record Analysis 2-1
Guidelines for Quality Control of Acoustic Data 2-1
Acoustic Signal Acquisition Recommendations 2-3
Liquid Level Echo is Not Clearly Identifiable 2-3
Example Acoustic Records and Analysis 2-3
Example 1, Simple Wellbore with Uniform Casing and Tubing Diameters 2-5
Example 2, Deviated Wellbore 2-8
Example 3, Tapered Tubing and Casing Liner 2-10
Example 4, Well with Blast Joint Opposite Upper Perforations 2-14
Example 5, ESP Well Casing Shots: Producing and Static Well 2-15
Producing Fluid Level Record 2-18
Static Fluid Level Record 2-20
Example 6, ESP Well with Hole in the Tubing 2-20
Detailed Analysis of Multiple Echoes 2-24
Example 7, Surface-Controlled Subsurface Safety Valve (SCSSSV) Testing 2-25
Case A: Correctly Operating Safety Valve 2-25
Case B: Malfunctioning Safety Valve 2-25
Example 8, Corrosion Survey of Intermediate Casing 2-27
iii
iv ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
Example 9, Stratified Annular Gas Column 2-30

Variation of Gas Composition with Depth 2-34
Summary 2-35
References 2-36
3. Fundamentals of Acoustic Fluid Level Surveys 3-1
Sound Pulse Generation and Wave Propagation 3-1
General Solution of the Wave Equation 3-2
Characteristic Diagram in the z,t Plane 3-5
Reflection and Transmission of a Plane Wave 3-7
Reflection at the Discontinuity of Fluid Properties 3-7
Reflection and Transmission at a Geometric Discontinuity 3-11
Sound Pressure Wave Propagation in Pipes and Annuli 3-13
Acoustic Velocity in Gases 3-15
Effect of Gas Composition, Pressure, and Temperature on Acoustic Velocity 3-16
Acoustic Signal Amplitude Reduction 3-20
Summary 3-23
References 3-23
4. Acoustic Fluid Level Equipment and Procedures 4-1
Acoustic Pulse Generation and Signal Acquisition 4-1
Pulse Generation 4-5
Manual Pulse Generation 4-5
Explosion versus Implosion Pulses 4-7
Automatic Pulse Generation 4-8
Other Pulse Generation Methods 4-9
Acoustic Record Acquisition and Recording 4-10
Microphones 4-10
Signal Recording and Processing 4-11
Safety Considerations 4-11
Hazardous Locations in the Oilfields 4-12
API Classification of Drilling and Production Facilities 4-12
Recommended Operating Procedures 4-13
Installation of Sound Source 4-14
Well Preparation and Information 4-14
Acquisition and Recording 4-15
Random Acoustic Signals 4-16
Pumping-Related Noise 4-17
Gas Flow Noise 4-18
Calibration and Maintenance 4-18
Summary 4-19
References 4-19
5. Methods of Determining Distance to the Liquid Level 5-1
Converting Acoustic Pulse Travel Time to Distance 5-1
Collar Count Method 5-2
Stepwise Collar Echo Count 5-4
Automatic Digital Filtering 5-6
Distance to a Known Wellbore Anomaly (Downhole Marker Analysis) 5-8
Calculating Acoustic Velocity from Gas Gravity or Composition 5-10
Estimating Acoustic Velocity from Similar Wells or Past Acoustic Surveys 5-13
Recommendations for Background Noise 5-15
Summary 5-15
References 5-16

Contents v
6. Calculating Wellbore Pressure Distribution from Acoustic Fluid Level Surveys 6-1
Well Performance and Potential Analysis 6-1
Fluid and Pressure Distribution in Pumping Wells 6-2
Calculating Producing Pressure Distribution 6-4
Wellbore and Completion Classification 6-5
Pressure Calculation 6-6
Gas Gravity 6-7
Fraction of Liquid in a Gaseous Column 6-7
Liquid Level Depression Test 6-7
Procedure for Walker Test Acquisition and Analysis 6-10
Correlations for Determining the Gaseous Liquid Column Gradient 6-16
Annular Gas Flow Rate Determination 6-18
Quality Control of Pressure Calculations 6-21
Production Stabilization 6-21
Percentage of Liquid in the Annular Gaseous Column 6-22
Fluid Level Depth 6-22
Height of the Gaseous Liquid Column 6-22
Calculating SBHP in Pumping Wells 6-22
Summary 6-24
References 6-25
7. Applied Well Testing for Pressure Transient Data Acquisition 7-1
Programmed Fluid Level Surveys 7-2
Automatic Acoustic Fluid Level Survey 7-2
Special Requirements of Programmed Acoustic Data Acquisition and Processing 7-3
Surface Pressure Measurements 7-4
Wellbore Fluid Composition and Distribution 7-5
Pressure Distribution Calculation 7-6
Recommended Test Procedures and Practical Implicaions 7-6
Beam-Pumped Wells 7-8
ESP and PCP Wells 7-13
Gas Lift Wells 7-13
Gas Wells 7-13
Wells with Multiple Producing Zones 7-13
Example Field Tests 7-14
Well A 7-14
Well B 7-17
Well C 7-20
Well D 7-20
Well E 7-23
Summary 7-23
References 7-26
8. Applications of Fluid Level Measurements to Pumping Wells 8-1

Production Efficiency in Rod-Pumped Wells 8-2
Pump-Off: Excessive Pump Capacity 8-3
Gas Interference 8-4
Potentially Misleading Acoustic Fluid Level Surveys 8-5
Choked Pump 8-5
Annular Fluid Gradient Inversion 8-6
Liquid Level Depression Tests Confirm Gradient Inversion 8-10
Effect of Tubing Anchor on Well Performance 8-13

vi ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
Tubing Diagnostic Acoustic Surveys 8-13

Detection of Tubing Holes 8-13
Acquisition of Fluid Level Records in the Tubing 8-14
Determining Depth to Tubing Hole 8-15
Tubing Gas Flow and Wells That Kick 8-17
Recommended Troubleshooting Procedures 8-17
Operating Gassy Wells 8-18
Inefficient Pump Displacement 8-18
Documenting Fluid Movement 8-18
Paraffin Deposition 8-19
Fluid Level Surveys in ESP Wells 8-21
Acoustic Fluid Level Acquisition and Analysis 8-21
Presence of Gaseous Column 8-22
Comparing ESP Downhole Pressure Sensor Measurement and PIP from Acoustic Fluid Level 8-23
Pressure Distribution and Annular Fluid Level Gradient Discontinuity in Well with Multiple Producing
Zones 8-23
Fluid Level Surveys in PC Wells 8-26
Acoustic Record Quality 8-26
Recommended Procedures in Pumping Wells 8-29
Safety Considerations 8-29
Acquisition Workflow 8-29
Recommendations for Gas Gun Connection to the Wellhead 8-29
Quality Control 8-30
Summary 8-30
References 8-31
Bibliography 8-31
9. Fluid Level Measurement Applications for Gas Wells 9-1
Determining Liquid Loading of a Gas Well 9-1
Equipment Selection and Setup 9-3
Implosion Method 9-3
Explosion Method 9-3
Depth Determination to the Liquid Level 9-4
Analyzing Typical Gas Well Performance 9-5
Fluid Level Records for Gas Flow Above Critical Rate 9-5
Recommended Procedure 9-5
Example Gas Well Flowing Above Critical Rate 9-7
Description of the Acoustic Tests 9-7
Example Gas Well Flowing Below Critical Rate 9-9
Estimating BHP from Fluid Level Measurement in Tubing 9-11
Determining Static Bottomhole Pressure 9-15
Testing of Downhole Safety Valve Operation 9-18
Applications to Troubleshooting Gas Wells 9-20
Holes in Gas Well Tubing 9-25
Acoustic Survey in Packer-Less Gas Well 9-26
Summary 9-28
References 9-28
10. Fluid Level Measurement Applications for Gas Lift Wells 10-1
Benefits of Acoustic Measurements in Gas Lift Wells 10-2
Equipment Installation and Data Acquisition 10-2
Background Noise 10-2
Random Noise 10-4

Contents vii
Analyzing Gas Lift Well Fluid Level Records 10-6

Determining Acoustic Velocity 10-6
Using Mandrels as Markers 10-8
Troubleshooting Gas Lift Systems 10-14
Open Valves 10-14
Operating Pressure Distribution for Various Gas Lift Configurations 10-14
Static Bottomhole Pressure 10-14
Producing Bottomhole Pressure 10-16
Case A 10-17
Case B 10-17
Case C 10-17
Case D 10-17
Example Fluid Level and Pressure Survey 10-17
Background Information About Gas Lift Installations 10-19
Valve Operation 10-22
Valves and Mandrels 10-22
General Considerations for Gas Lift Design 10-24
Well Unloading 10-24
Unloading Sequence for Casing Pressure Operated Systems 10-26
Monitoring the Unloading Operation 10-31
Summary 10-34
References 10-35
Bibliography 10-35
11. Fluid Level Measurement Applications for Plunger Lift Wells 11-1
Plunger Lift System 11-2
Types of Plungers 11-3
Types of Controllers 11-3
Timers 11-3
Pressure Differential or Pressure Set Point 11-5
Programmable Logic Controllers 11-5
Plunger Lift Operation Cycle 11-5
Acoustic Fluid Level Monitoring of Plunger Well Operation 11-6
Determining Plunger Position and Velocity 11-6
Active Acoustic Monitoring of Plunger Position 11-7
Passive Acoustic Monitoring of Plunger Position 11-9
Data Acquisition and Recording for Passive Monitoring 11-11
Identifying and Annotating Key Events 11-11
Instantaneous Plunger Fall Velocity 11-13
Factors Affecting Plunger Fall Velocity 11-16
Determining Gas Properties 11-16
Detecting and Troubleshooting Operation Problems 11-17
Field Example 11-20
Summary 11-21
References 11-22
Bibliography 11-23
Appendix: Figure Credits A-1
Glossary G-1
Index I-1

viii ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
Figures
Frontispiece xxii
1.1 Typical oil lifting cost as a function of artificial lift efficiency and produced water cut 1-2
1.2 Calculation of the overall pumping efficiency from the pressure distribution in the annulus and tubing 1-3
1.3 Detailed analysis of the pumping well performance from the acoustic fluid level survey 1-5
1.4 Acoustic fluid level height during a liquid level depression test in an ESP well outfitted with a bottomhole
pressure sensor 1-8
1.5 Data input and output form for the well performance analysis 1-11
1.6 Schematic of the fluid and pressure distribution in a pumping well 1-11
1.7 Inflow performance based on the productivity index determined from a two-rate flow test 1-13
1.8 Schematic representation of gas saturation in an oil reservoir at a pressure lower than the bubble point
pressure 1-14
1.9 Producing bottomhole pressure is 200 psi for liquid and gas flow from perforations. 1-15
1.10 Inflow performance relations for single-phase (blue) and two-phase (red) flow in reservoir 1-16
1.11 Graphical representation of a multi-rate flow test for determining the well inflow performance 1-17
1.12 Vogel’s dimensionless curves for different stages of reservoir depletion and best fit reference curve 1-18
1.13 Schematic comparison of IPR from a multi-rate test and the corresponding Vogel relation based on single
well test flow rate q2 1-19
1.14 Comparison of IPR relations from Gallice and Wiggins 1-20
1.15 Acoustic fluid level and casing pressure acquired during a multi-rate test in a well pumped with a
variable-speed ESP 1-21
1.16 Variation of PBHP versus time during the variable rate test shown in figure 1.15 1-22
1.17 Inflow performance from a three-rate test 1-23
2.1 Example acoustic record with unexplained signals 2-2

2.2 Vertical well with a rod pump set at 5,115 feet. Average production is 27 bbl/day of oil, 60 bbl/day of
water, and 40 Mscf/day of gas. 2-5
2.3 Acoustic record with typical echoes from the liquid level and tubing collars 2-6
2.4 Detailed analysis of echoes from tubing collars recorded between 0.5 and 1.5 seconds from pulse
generation 2-6
2.5 Summary report including calculation of downhole pressures and production potential 2-7
2.6 Wellbore trajectory and pump location 2-8
2.7 Comparison of raw and filtered acoustic record 2-9
2.8 Overlay of records acquired at the start and the end of pump-down 2-10
2.9 Summary acoustic fluid level report with pressure distribution and potential analysis 2-11
2.10 Acoustic record showing multiple echoes 2-12
2.11 Detailed analysis of an acoustic record using the depth to the downhole marker at the echo where the
tubing tapers from 4H to 3H inches 2-12
2.12 Summary acoustic fluid level report, including the calculation of downhole pressures and production
potential 2-13
2.13 Well with multiple perforated intervals 2-14
2.14 Acoustic record with echoes from the blast joint and perforations 2-15
2.15 Detail of an acoustic record showing (A) the repeat echo from the top perforations and absence of an
identifiable liquid level echo and (B) the liquid level marker located at pump intake depth, based on
dynamometer determination of pumped-off condition 2-16
2.16 Well completion schematic 2-17
2.17 Wellbore trajectory, ESP location, and producing fluid level 2-17
2.18 Noisy record due to a resonating cavity or cable banding 2-18
2.19 Record in Figure 2.18 filtered with a low-pass filter 2-18
viii
Contents
Figures ix
2.20 Acoustic velocity determination using echoes from tubing collars after filtering the raw data 2-19
2.21 Producing fluid level analysis and pressure distribution summary 2-19
2.22 Acoustic fluid level record for a shut-in well 2-20
2.23 Summary report with producing and static fluid level analyses 2-21
2.24 Wellbore schematic and trajectory for Example 6 well with hole in the tubing. 2-22
2.25 First acoustic record acquired through the tubing string 2-22
2.26 Overlay of three acoustic records acquired in quick succession 2-23
2.27 Calculation of acoustic velocity using the known distance to the echo from the tubing taper 2-23
2.28 Determining the depth to the hole in the tubing 2-24
2.29 Identification of repeat echoes from multiple reflectors inside the tubing 2-24
2.30 Wellbore diagram of a flowing gas well with a downhole safety valve 2-26
2.31 Superposition of acoustic records acquired with an open (black) and a closed (blue) properly operating
safety valve 2-26
2.32 Comparison of acoustic traces for a malfunctioning subsurface safety valve 2-27
2.33 Wellbore diagram for a well undergoing a casing integrity test 2-28
2.34 Acoustic record acquired in the intermediate casing annulus and processed with a low-pass filter 2-29
2.35 Acoustic record acquired in the intermediate casing annulus, showing echoes from the couplings 2-29
2.36 Determining the acoustic velocity of nitrogen gas in a wellbore using echoes from the casing collars 2-30
2.37 Summary report for a casing integrity test 2-31
2.38 Well completion schematic and simplified directional survey for Example 9 well with stratified gas
column 2-32
2.39 Acoustic record acquired in a deep rod-pumped horizontal well 2-32
2.40 Dynamometer diagrams showing the range of pump liquid fillage during 63 strokes 2-33
2.41 Detailed results of the tubing joints count 2-34
2.42 Variation of acoustic velocity as a function of the RTTT 2-35
2.43 Determining average acoustic velocity from the known depth of the end of the tubing 2-36
2.44 Summary report for a horizontal well containing a stratified gas column 2-37
3.1 Acoustic pulse amplitude versus time, recorded at the gas gun where z = 0 3-3
3.2 Propagation of an acoustic pulse, as observed at different times by three recorders located at different
depths 3-4
3.3 Construction of the waveform observed at position Z1, applying the characteristic diagram to the
waveform generated at Z = 0 3-5
3.4 Observed amplitude versus time record at point Z1 3-6
3.5 Reflection and transmission at a medium discontinuity 3-7
3.6 Wellbore schematic, characteristic diagram, and synthetic acoustic record for a well with a gas/liquid
interface at 3,000 feet 3-9
3.7 Acoustic fluid level record with multiple echoes and an acoustic velocity of 1,129 ft/s 3-10
3.8 Reflection and transmission at an area discontinuity 3-11
3.9 Schematic wellbore diagram, characteristic diagram, and synthetic acoustic record for the liquid level
below the casing liner top. Solid blue lines represent down-kicks; dashed red lines represent up-kicks. 3-13
3.10 Echoes from tubing couplings, a tubing crossover, and the fluid level 3-14
3.11 Schematic PVT diagram for hydrocarbon fluids 3-16
3.12 Acoustic velocity in a hydrocarbon gas with a specific gravity of 0.6 as a function of pressure and
temperature 3-18
3.13 Sonic velocity in a hydrocarbon gas with a specific gravity of 1.2 3-19
3.14 Pulse amplitude attenuation and spreading in wellbore gas at 106 psi 3-21
3.15 Exponential amplitude decay of an acoustic pulse propagating inside coiled tubing 3-22
4.1 Pulse generation using a manually actuated gas gun with a quick-opening valve 4-5
4.2 Typical manually operated gas gun 4-6
4.3 Simplified schematic of the gas gun in figure 4.2 4-6

x ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
4.4 Comparison of records from explosion and implosion pulses acquired in the same wellbore. 4-7
4.5 Wireless remote control gas gun 4-8
4.6 Remotely controlled gas gun operation. 4-9
4.7 Typical laptop-based fluid level data acquisition, processing, and analysis systems 4-10
4.8 Strip chart recording of an acoustic trace 4-11
4.9 Hazardous areas classification and extension for a rod-pumping wellhead 4-13
4.10 Certified intrinsically safe fluid level recorder and gas gun 4-13
4.11 Connections to a well 4-14
4.12 Record showing decaying acoustic resonance in connection piping 4-15
4.13 Effect of chamber pressure on the amplitude of the liquid level echo. 4-16
4.14 Records showing random signals 4-17
4.15 Noise generated by pumping action 4-17
4.16 Acoustic record acquired in a well produced by gas lift 4-18
5.1 Acoustic record showing decay of the amplitude of echoes from tubing collars. 5-2
5.2 Digital processing of an acoustic record to detect echoes from tubing collars. 5-3
5.3 Identification of collar echoes in various segments of the acoustic record from figure 5.1, performed
visually and manually by the analyst 5-5
5.4 Variation of collar echo frequency as a function of depth 5-6
5.5 Detailed collar count analysis of the record in figure 5.1 5-7
5.6 Downhole marker analysis 5-8
5.7 Effect of temperature on the average acoustic velocity of gas (with a 0.6 gravity) in a pumping well 5-11
5.8 Temperature survey in a pumping well 5-12
5.9 Portable instruments for gas gravity (left) or gas analysis (right) 5-12
5.10 Isothermal gradient map 5-13
5.11 Variation of the acoustic velocity of the annular gas in a given well over three years 5-14
6.1 Pressure distribution in a pumping well 6-2

6.2 Fluid distribution in a stabilized pumping well 6-4
6.3 Classification of pumping wells by tubing depth and fluid distribution 6-5
6.4 Generalized wellbore configuration for tapered tubing and casing 6-6
6.5 Calculating the gaseous column gradient from a liquid level depression test. 6-8
6.6 Wellhead arrangement for performing a liquid level depression test 6-9
6.7 Typical fluid level and casing pressure recorded versus time during the Walker test 6-11
6.8 Height of the gaseous liquid column versus pressure at the gas/liquid interface during a liquid
depression test 6-12
6.9 Measured downhole pressure and gaseous liquid column height for varying annular pressures 6-13
6.10 Estimate of the gaseous column gradient from a liquid level depression test 6-14
6.11 Partial sequence of acoustic fluid level records for tests analyzed in figure 6.12 6-15
6.12 Analysis of the liquid level depression test 6-16
6.13 Effective oil fraction correlation from liquid level depression tests 6-17
6.14 Comparison of the effective oil fraction determined from liquid level depression tests in Venezuelan
heavy oil (10 to 11° API) wells and western Texas (32 to 43° API) wells 6-19
6.15 Annular pressure increase during the acoustic fluid level acquisition used to estimate the gas inflow
rate 6-20
6.16 Pressure balance and fluid distribution in a static well 6-23
6.17 Oil/water distribution for static well conditions 6-24
7.1 Acoustic velocity variation during a seven-day pressure transient test 7-4
7.2 Variation of casing pressure (circles) and wellhead temperature (triangles) during the transient test of
figure 7.1 7-5

Contents
Figures xi
7.3 (A) Pressure transient fluid level acquisition setup using wired instruments. (B) Pressure transient fluid
level acquisition setup using wireless standalone instrumentation. 7-7
7.4 Short-duration liquid level depression test showing the fluid level drop by 756 feet in 30 minutes 7-8
7.5 Dynamometer record acquired prior to initiating the pressure buildup test, showing pump displacement
of 110 bbl/day 7-9
7.6 Example of erratic pump operation caused by a damaged traveling valve 7-10
7.7 Thermal insulator placed on a pressure sensor attached to a gas gun 7-10
7.8 Example record indicating the correct selection of the liquid level echo in the presence of a repeat echo. 7-11
7.9 Examples of wellhead pressure variations observed in three different wells 7-12
7.10 Example of multiple sequential acoustic records with wellbore perforations. 7-14
7.11 Long-term pressure buildup test showing casing pressure (circles) and casinghead pressure transducer
temperature (triangles) as a function of elapsed time 7-15
7.12 Depth to liquid level (triangles) and casinghead pressure (circles) versus time 7-15
7.13 Computed BHP (triangles) and measured casinghead pressure (circles) versus time 7-16
7.14 Resulting log-log plot of delta pressure with derivative versus delta time in hours 7-16
7.15 Horner plot yields skin of 0.8 and P* of 2,018 psi. 7-17
7.16 Liquid level (triangles) rises 2,600 feet and casing pressure (circles) increases by 378 psi during the
4H-day test. The x points indicate bad data. 7-18
7.17 BHP (circles) levels off and liquid after-flow (triangles) tends to cease after about 48 hours. 7-18
7.18 Resulting log-log plot of delta pressure with derivative (triangles) versus delta time in hours 7-19
7.19 Horner plot shows skin of 8.9 and P* of 1,306 psi. 7-19
7.20 Variation of casing pressure and transducer temperature (triangles) during a 25-day and 17-hour buildup
test 7-20
7.21 Computed BHP (circles) increases as RTTT time to liquid decreases when liquid rises 7-21
7.22 Wellbore storage just beginning to be overcome and radial flow period starting in the well. 7-21
7.23 Casing pressure (circles) and BHP (triangles) during 3 days and 13 hours 7-22
7.24 Log-log plot shows boundary effect after radial flow period, as shown by increasing derivative (triangles). 7-22
7.25 Horner plot yields a skin of 1.7 and P* of 102 psia. 7-23
7.26 Liquid level (triangles) drops as casing pressure (circles) increases. 7-24
7.27 Casing pressure (circles) and BHP (triangles) versus time 7-24
7.28 Log-log plot shows beginning of radial flow period, indicated by stabilization of derivative (triangles). 7-25
7.29 Horner plot seems to indicate the presence of significant skin. 7-25
8.1 Example of an over-pumped well 8-3

8.2 Example of gas interference during pump operation 8-4
8.3 Example of incomplete liquid fillage due to choked pump intake 8-6
8.4 Distribution of fluids and pressure versus depth in a stabilized pumping well 8-7
8.5 Schematic representation of a well completion exhibiting fluid gradient inversion due to the TAC 8-8
8.6 Dynamometer records in a well exhibiting annular gradient inversion 8-9
8.7 Casing pressure and liquid level depth as a function of time, showing a rapid drop in the fluid level past
the tubing anchor 8-11
8.8 Results of a liquid level depression test in a well with the tubing anchor set above the perforations 8-12
8.9 Fluid and pressure distribution caused by the presence of the tubing anchor 8-12
8.10 Frequency of rod couplings echoes in tubing (19.46 jts/s at 25.0 ft/jt) 8-14
8.11 Example tubing record showing an echo from a deep hole 8-15
8.12 Comparison of a hole-in-tubing echo overlay of low-pass filtered casing shot to a raw tubing shot 8-15
8.13 (A) Acoustic traces acquired in the casing annulus while the pump is stopped and while the pump is
operating, showing an inversion of polarity of echo from the tubing leak. (B) Overlay of acoustic records
acquired when the pump was operating and when the pump was stopped. 8-16
8.14 Tubing fill-up rate as a function of pump displacement and SPM 8-19
8.15 Successive acoustic records taken during 32 minutes of pump operation 8-20

xii ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
8.16 A PAP (grooved steel) plunger 8-20

8.17 (A) Fluid level record acquired in an ESP. (B) Detail of an acoustic signal showing echoes from cable bands 8-21
8.18 Productivity analysis for an ESP well 8-22
8.19 Typical annular fluid level record for the well in figure 8.20 8-23
8.20 Acoustic survey summary report displaying the computed annular pressure distribution 8-24
8.21 Pressure versus depth traverse showing the measured and computed pump intake pressure for the well
in figure 8.19 8-25
8.22 Analysis of pressure and flow distribution for a well producing from three perforated intervals 8-27
8.23 Examples of fluid level records acquired in wells producing with PC pumps 8-28
8.24 Recommendations for connecting a gas gun to a wellhead 8-29
9.1 Example of a critical rate diagram for different tubing sizes 9-2
9.2 Examples of gas gun connections to gas wells 9-4
9.3 Acoustic records acquired during a gas well shut-in 9-6
9.4 Position of the gas/liquid interface as a function of time 9-8
9.5 Height of a gaseous liquid column as a function of tubing pressure 9-8
9.6 Pressure traverse in a liquid loaded gas well 9-9
9.7 Pressure versus depth traverses 9-10
9.8 Gaseous column gradient changes in time 9-11
9.9 Sequence of acoustic records 9-13
9.10 Comparison of annular and tubular “S” curves 9-14
9.11 Offshore gas wellbore schematic and corresponding acoustic trace acquired down the tubing after
long shut-in time 9-16
9.12 Detail of the wellbore completion 9-17
9.13 Acoustic record corresponding to the wellbore in figure 9.12 9-18
9.14 Detail of the echo from the 3H- to 4H-inch crossover 9-19
9.15 Fluid and pressure distribution in the shut-in well 9-19
9.16 Acoustic records acquired in the tubing in a shut-in gas well with a properly operating SCSSV 9-20
9.17 Acoustic records acquired in the tubing of a shut-in gas well with a SCSSV stuck open 9-21
9.18 Examples of identification of downhole features from the polarity of the acoustic echoes 9-22
9.19 Hole in a gas well’s tubing string caused by corrosion 9-23
9.20 Acoustic record acquired inside the tubing of a gas well with a shallow corrosion hole 9-23
9.21 Characteristic wave path diagram corresponding to the presence of a hole in the tubing above the liquid
level 9-24
9.22 Production history of a gas well with a hole in the tubing 9-25
9.23 Annular fluid level in a liquid loaded gas well operating at stabilized conditions 9-26
9.24 Pressure distribution in a gas well without a packer 9-27
9.25 Sequence of acoustic records in a gas well with a hole in the tubing 9-29
1 0.1 Typical continuous injection gas lift well 10-1

10.2 Acoustic fluid level instrument connection to the casing or tubing of a gas lift well. 10-3
10.3 Comparison of acoustic records for different connection methods and gas guns 10-4
10.4 Acoustic record acquired while injection gas is flowing (top) and acoustic record acquired after stopping
the flow of injection gas (bottom) 10-5
1 0.5 Repeat acoustic record (blue) that does not show a random signal at about 10.4 seconds. 10-6
10.6 Chart depicting acoustic velocity as a function of pressure and temperature for 0.6 gravity hydrocarbon
gas 10-7
10.7 Acoustic record acquired down the tubing of a shut-in well, showing a mismatch of echoes and gas lift
mandrels due to varying acoustic velocity with increasing depth 10-8
1 0.8 Mandrel design determines the polarity of the echoes. 10-9
10.9 Complex acoustic record in a deep gas lift well 10-10

Figures
Contents xiii
10.10 Acoustic data showing a problem identifying multiple echoes because of internal reflections down the
tubing 10-11
10.11 Depth profile of the acoustic velocity of injection gas from acoustic records in figure 10.8 10-12
10.12 Correlation function of acoustic velocity versus the RTTT 10-13
10.13 Acoustic record analyzed automatically to generate a depth scale 10-13
10.14 Comparison of records acquired in the tubing and casing, showing an echo from the hole in the tubing 10-15
10.15 Gas lift well configurations 10-16
10.16 Casing survey: Liquid level at 5,602 feet above fourth gas level valve 10-18
10.17 Pressure traverses for the casing and tubing during shut-in. 10-18
10.18 Acoustic records acquired down the tubing after shut-in 10-19
10.19 Pressure distribution in the tubing and annulus of a gas lift well 10-20
10.20 Typical side-pocket mandrels 10-23
10.21 Typical conventional mandrels with external valves 10-24
10.22 Determination of the injection depth (6,500 feet) for the desired formation flow rate and operating
gas/liquid ratio. PBHP = 2,450 psi 10-25
10.23 Pressure in the tubing and casing when the kill fluid reaches the surface by U-tubing through the operating
valve 10-26
10.24 Depth and pressure when the first valve was unloaded 10-27
10.25 Unloading of the liquid to the flow line after the first valve was uncovered. This assumes sufficient gas
volume to achieve the minimum gradient. 10-28
10.26 Depth and pressure when the second valve was uncovered and gas was injected into the tubing 10-29
10.27 Depth and pressure when the third valve was uncovered 10-30
10.28 Depicting the depth and pressure when the operating valve was uncovered and gas was injected at
400 GLR 10-31
10.29 Portable pressure recorder monitoring the casing and tubing pressure during gas lift well unloading 10-32
10.30 Two-pen pressure-recorder chart illustrating continuous-flow gas lift unloading operations with choke
control of the injection gas 10-33
10.31 Gas lift injection manifold 10-33

1 1.1 Schematic diagram of a typical plunger lift installation 11-2

11.2 Surface components of the plunger installation 11-3
11.3 Various types of plungers 11-4
11.4 Three phases of plunger lift: shut-in, unloading, and after-flow 11-6
11.5 Typical installations of acoustic hardware used for plunger lift monitoring 11-7
11.6 Chart depicting the depth of the plunger versus time obtained from manually acquired acoustic records 11-8
11.7 Record showing the expanded portion of an acoustic signal and the tubing pressure versus the time
recorded during plunger fall 11-10
11.8 Acoustic recorder and pressure sensor installation schematic 11-10
11.9 Example acoustic record showing pulses generated by the plunger as it falls through the gas and through
the liquid at the bottom of the tubing 11-11
11.10 Acoustic and pressure record for a complete plunger cycle 11-12
11.11 One-minute section of an acoustic record of pulses generated at the tubing couplings, and the method
for determining the plunger fall velocity 11-13
11.12 Graph of acoustic pulses recorded digitally during plunger fall 11-14
11.13 Record showing the plunger depth and velocity as a function of time elapsed since the start of plunger fall 11-15
11.14 Determining the acoustic velocity from repeat echoes of the collar pulses 11-17
11.15 Acoustic and pressure record in a plunger well with dry tubing 11-18
11.16 Tubing head pressure increase observed after the plunger drops below a hole in the tubing 11-19
11.17 Graph showing recorded pressure changes corresponding to the plunger falling or stopping 11-19
11.18 Record showing the effect of a hole in the tubing on the plunger fall velocity 11-20
11.19 Complete recording of pressures and acoustic data during an entire plunger cycle 11-21
11.20 Graph depicting the plunger fall velocity and position for the well in the field example 11-22

xiv ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
Tables
2.1 Typical examples of acoustic records and corresponding analyses 2-4

5.1 Comparison of distance to liquid level computed by different methods 5-9
5.2 Acoustic velocity of annular gas in wells producing from same reservoir 5-14
6.1 Results of pumping well test with differing casinghead pressures 6-12
6.2 Effects of unstabilized flow on computed PIP 6-21
9.1 Test summary table 9-12

Contents xv
Foreword
The Petroleum Extension (PETEX) at the University of Texas at Austin is proud to present Acoustic Fluid Level
Measurements in Oil and Gas Wells Handbook. This technical manual provides a comprehensive overview of the
importance of acoustic fluid level measurements in the operation and analysis of oil and gas wells. During the pro-
duction of any well, one of the most important questions has been and will always be, “Is this well performing at
its maximum potential?” We hope that this book, and the acoustic fluid level tools, procedures, calculations, and
guidelines therein, will help well operators and professionals answer that question.
As career educators and professionals in the field of petroleum engineering, both authors offer specialized
insights from their experience in utilizing and analyzing acoustic fluid level measurements. With several chapters
on the background, mathematics, and chemistry involved in acoustic records and calculations, followed by specific
applications in different types of wells and equipment, the knowledge presented throughout this book should prove
useful and valuable both within the classroom and out in the field.
Many of the products we offer, of which the Acoustic Fluid Level Measurements Handbook is only one, are
constantly evolving in an effort to keep pace with customer interests and demands. With that in mind, we encourage
you to visit petex.utexas.edu to see what’s new and explore the ways in which we can help support your career.
Petroleum Extension (PETEX)

The University of Texas at Austin
Texas Extended Campus
For distribution by Petroleum Extension-The University of Texas xv

at Austin
xvi ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK

Contents xvii
Preface
Thousands of acoustic fluid level measurements are performed each day in oil and gas wells with the objective of
monitoring and optimizing their production performance. For many years, the information necessary to verify the
accuracy of acoustic fluid level measurements and to justify their use despite expensive operational changes has
been scattered in numerous publications that are practically inaccessible to the majority of practicing engineers and
technicians. Furthermore, the evolution from analog measurements to computer-based, sophisticated digital tools,
as well as the recent development of wells from shallow, simple, vertical wellbores to very deep, complex, multi-
tubular and deviated configurations, has resulted in acoustic records that are difficult to analyze. This difficulty does
not exonerate the analyst from the responsibility of verifying that automatic data processing and presentation can
yield reasonable results.
The objective of this handbook is to provide a comprehensive guide for both new and experienced practitioners
of fluid level measurements on how to properly acquire and analyze acoustic records in all types of well applica-
tions. The handbook is organized so that the reader can easily access the information that is most relevant to his or
her field applications based on the type of well and artificial lift method applied. The introductory chapters cover
the basic principles of sound propagation essential to acoustic fluid analysis. Numerous examples are provided to
illustrate the most common interpretation problems and techniques.
Source material for this volume comes from industry publications, as well as the writings and notes compiled
by the authors and their associates for teaching seminars to industry personnel on the art and science of well per-
formance analysis.
We wish to thank the staff of the PETEX for taking an interest in publishing this book, and for the enormous
amount of patient and skillful editing and artistic talent in preparing graphics for the text.
During the drafting of this work, several chapters were reviewed by experienced well performance analysts. In
addition to constructive criticism, the most frequent comment was “I wish this book had been around when I first
started shooting fluid levels!” The authors sincerely hope that this will be the feeling of everyone who reads this
handbook.
A. L Podio, PhD, PE
October, 2016
For distribution by Petroleum Extension-The University of Texasxvii

at Austin
xviii ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK

Contents xix
Acknowledgements
The authors extend their sincere and heartfelt appreciation to the engineers and technicians at Echometer Company
for their numerous contributions to the technology of acoustic fluid level measurements in oil and gas wells.
In particular, we thank Dieter Becker, Lynn Rowlan, Ken Skinner and Carrie-Anne Taylor for applications of
fluid level records to pumping well troubleshooting. In particular, we thank Dieter Becker for applications to pressure
transient acquisition and analysis, Lynn Rowlan and Dieter Becker for applications of fluid level records to gas well
troubleshooting, Carrie-Anne Taylor for her contributions to the technology of acoustic fluid level measurement and
analysis applied to gas lift wells, and Lynn Rowlan and Ken Skinner for applications to detailed analysis of plunger
well performance and troubleshooting. Thanks also go to Dr. Pierre Lichaa for his thorough review of the earliest
manuscript draft of the handbook and to David Bishop for his many constructive comments.
Accurate and efficient acquisition and analysis of acoustic records requires access to user friendly and sophisti-
cated computer applications. The outstanding creativity and technical ability of Kay Capps and his team of software
and hardware developers is gratefully acknowledged.
In addition, PETEX would like to thank the following individuals whose hard work, knowledge, and meticu-
lous attention to detail made this publication possible: Deborah Caples, Senior Graphics Designer; E.K. Weaver,
Graphics Designer; Mary Lin, Dewey Badeaux, and Virginia Dosher, Editors; and Leah Lehmann, Proofreader and
Digital Librarian.
For distribution by Petroleum Extension-The University of Texas xix

at Austin
xx ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK

Contents xxi
About the Authors
A. L. Podio is a petroleum engineering consultant specializing in artificial lift. Formerly,

he was a professor in the Petroleum and Geosystems Engineering Department at the
University of Texas at Austin, where he taught and directed research in the areas of
Drilling and Production. He holds a Bachelor of Science in Mining and Petroleum
Engineering from the University of the Andes in Bogota, Colombia and BS, MS
and PhD degrees in Petroleum Engineering from the University of Texas. He is
a registered Professional Engineer in Texas and has been a distinguished speaker
and technical editor for SPE, has published numerous articles in international
journals, holds seven U.S. patents, and is a co-author of The Beam Lift Handbook
published by the Petroleum Extension at the University of Texas at Austin (PETEX).
James McCoy was born in Wewoka, Oklahoma. He graduated from the University
of Oklahoma with a B.S. in Petroleum Engineering and from Penn State University
with an M.S. in Petroleum and Natural Gas Engineering. He worked for a consult-
ing firm and a major oil company and purchased and operated properties. In 1962,
he acquired Echometer Company and expanded the acoustic liquid level instrument
capabilities into a full Well Analyzer System that operates worldwide. He holds sev-
eral U.S. and foreign patents. He was instrumental in establishing the McCoy School
of Engineering at Midwestern State University and is a member of the Society of
Petroleum Engineers and other petroleum engineering organizations.
For distribution by Petroleum Extension-The University of Texas xxi

at Austin
xxii ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK

Visualizing Well Performance 1-1
1
Visualizing Well Performance

In this chapter:
• Principles of well production
• Relation between flow rate and pressure drawdown
• Characteristic performance of flowing wells and pumping wells
• Calculation of pressure distribution in the wellbore from fluid level surveys and casing pressure measurements
• Determining present operating conditions in relation to well potential
This introductory chapter addresses the widespread need An accepted rule of thumb is that the producing
for oil and gas field operators to continually verify that bottomhole pressure (PBHP) should be less than 10%
wells are being produced close to their optimum capacity of the static bottomhole pressure (SBHP) to ensure
and in the most cost-effective manner. The analysis is to that the maximum production is being obtained from
be made based on data obtained at the surface without the well. This requires measurement or calculation of
entering the wellbore and must yield an accurate rep- both the producing and the static bottomhole pressures.
resentation of conditions that exist within the wellbore, The PBHP must be obtained while the well is being
at the bottom of the well, at the sand face, and within produced under normal conditions, and the SBHP
the reservoir. As such, it is not an easy task, since fairly must be obtained when the well has been shut-in long
complicated processes are involved in the flow of gas, enough that the surface and bottomhole pressures in the
oil, and water mixtures in wellbores. Operators are often wellbore have stabilized and inflow from the reservoir
confused by the apparently contradictory evidence that has practically ceased.
may be obtained. Well performance is defined as the relationship between
The objective of this chapter is to present in simpli- the fluid flow rate and the pressure drawdown between
fied terms some of the basic concepts of well performance the wellbore and the formation pressure. This relation
analysis and to recommend a procedure to be followed may take several forms, all of which are approximations
in obtaining, organizing, and analyzing data acquired of the actual behavior. The most common forms are:
with acoustic fluid level instruments to visualize the • The productivity index (PI), defined as flow
performance of oil and gas wells. rate/drawdown, expressed in barrels per day
(bbl/day) per psi1,2
DETERMINING OPTIMUM AND CURRENT • The inflow performance relationship (IPR),
WELL PERFORMANCE defined as a functional relationship between
flow rate, flowing BHP, and static BHP, the most
The principal question that must be answered is: “Is the
common of which is Vogel’s relation3
well producing all the fluid that it is capable of producing
without problems and within the guidelines for optimum In order to answer the principal question stated
reservoir management?” If the answer is negative, then earlier, it is thus necessary to be able to determine the
additional questions must be answered to pinpoint the current well performance and to compare it to what is
reason(s) why the well is operating below its potential. considered optimum for the particular well.
For distribution by Petroleum Extension-The University of Texas 1-1

at Austin
1-2 ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
In any artificial lift well, particularly a pumping However, as the water cut increases, it becomes impera-
well, the overall system efficiency is a good indicator tive that the well be operated as efficiently as possible,
of performance because it expresses the relationship considering that the oil lifting cost rises exponentially
between the energy supplied and the effective work as the efficiency drops below 35%.
done by the pumping system in lifting a volume of fluid The overall efficiency of the artificial lift system
from a given depth. (Note that the energy supplied is can be estimated from a comparison of the useful work
directly related to operating cost, and that the effective that the system is developing—that is, lifting a certain
work performed by the pumping system is directly mass of material from a certain depth during a given
related to income.) time—to the power supplied to the system via the
Figure 1.1 shows the effect of the overall efficiency pump prime mover. The acoustic fluid level survey in a
on the lifting cost per barrel of oil for a typical situation pumping well provides a tool to determine from surface
in a mature oilfield. Notice that, as long as the water cut measurements the pressure distribution in the well and
is low—say, less than 60%—an inefficient system may be the net lift, as shown in figure 1.2 and as described in
tolerated since the oil lifting cost stays below $1.50/bbl. detail in chapter 6.
Figure 1.1 Typical oil lifting cost as a function of artificial lift efficiency and produced water cut

Pumping System Efficiency
Pwellhead PRESSURE
HHP
Efficiency = ———––
INPUT HP
SG x BPD x (Net Lift)
HHP = ————–––––––––
135,800
INPUT HP = Motor Power
SG = Specific Weight of Fluid

N
Pcasing Ptubing
PUMP
DISCHARGE
PRESSURE
Dp THROUGH PUMP
PUMP INTAKE
PRESSURE
DRAWDOWN
DEPTH Pwf Ps
Figure 1.2 Calculation of the overall pumping efficiency from the pressure distribution in the annulus and tubing
The useful power developed by the pump can be fluid column to determine the equivalent liquid pump
expressed in either of two ways. The first option is the submergence and the net lift.
work per unit time done by lifting a specific mass of The second option is the hydraulic horsepower devel-
material (oil, water, and gas), expressed as: oped by the pump moving the fluid at a given rate and
providing an increase in pressure from intake to discharge:
HHP = SG × BPD × Net Lift ÷ 135,800 Eq. 1.1
HHP = Q × Dp × 0.000017 Eq. 1.2
where
HHP = hydraulic horsepower (hp) where
SG = specific gravity of the fluid lifted Q = fluid rate (bbl/day)
BPD = fluid production during 24 hours (bbl/day) Dp = pressure increase through pump (PDP – PIP)
Net Lift = depth of annular gas-free liquid interface (psi)
(feet)
This relation requires that the pump discharge pres-
This relation requires that the depth to the fluid sure (PDP)—that is, the pressure at the bottom of the tub-
level be corrected for the presence of gas in the annular ing and above the pump discharge—be computed from

the surface tubing pressure and the gradient of the fluid of this handbook, the detailed analysis of acoustic fluid
present in the tubing (oil, water, and gas). In addition, level records can yield accurate values of producing
the pump intake pressure (PIP)—that is, the pressure in (PBHP) and static bottomhole pressures (SBHP). When
the annulus at the depth of the pump intake—must be combined with the production flow rate determined
computed from the surface casing pressure plus the gas from a recent well test, these values allow the operator
column pressure plus the gaseous liquid column pressure. to determine the percentage of maximum production
The first method of estimating the power de- that is being obtained from the well. The operator can
veloped by the pumping system should be preferred. It also estimate the maximum rate that could be achieved
involves fewer uncertainties to obtain the actual mass from the well when all the available reservoir pressure
of material lifted in 24 hours, provided that an accurate is expended while flowing fluids to the wellbore and
production well test is performed. when the resulting PBHP is reduced to zero.
The power provided to the pumping system is gen- Figure 1.3 illustrates a typical report that can be
erally represented by the electrical power consumption generated from detailed analysis of an acoustic fluid
when using electrical motors. In systems operated with level survey, showing the broad range of performance
engines, it can be estimated by the fuel consumption per information that is obtained. The amplitude versus time
unit time converted to an equivalent power value (1 hp of the acoustic signal is plotted at the left of the figure,
= 2,546 BTU/hour). starting at time zero at the top, and shows the echo
In all cases, the operator should strive to establish received from the fluid level at a depth of 1,408 feet.
a representative value of the overall efficiency of the The wellbore diagram (drawn with the same vertical
artificial lift system. The operator should then use the scale as the acoustic record) shows the casing, tubing,
representative value in conjunction with the well inflow sucker rods, perforations, and so on, and visualizes the
performance or productivity as guidelines to identify distribution of fluids in the annulus: gas in the upper
those wells that are not performing optimally. Operators part, gaseous oil between the fluid level and the pump
often approach the complicated problem of visualizing intake, and gaseous brine below the tubing intake down
well performance without enough information on exist- to the perforations. Details of pressure distribution, gas
ing producing conditions to perform a correct analysis. flow rate, and fluid composition are tabulated at the right
As seen by the efficiency relations presented above, and include an estimate of pressures and the potential
information about the well’s test rate and producing and production of the well.
static pressures must be secured in order to make an
intelligent assessment of the well’s current production What Must Be Known for Well Performance
potential. A well may be producing at a significant rate Analysis?
(say, over 100 bbl/day), but this rate could be 20%, 50%, It would be ideal to have a complete description of the
or 80% of the maximum flow rate. The only way to de- well’s history, characteristics, reservoir fluid properties,
termine the percentage is to establish the well’s IPR or reservoir performance, and geology. In practice, however,
PI based on measurements of pressures and flow rates. it is necessary to have at least an accurate description
of the wellbore and artificial lift system: depths, sizes,
WELL PERFORMANCE ANALYSIS FROM completion hardware, operational parameters, and so on.
ACOUSTIC FLUID LEVEL MEASUREMENTS This information should preferably be summarized and
Acoustic fluid level measurements are the basis of per- presented as a wellbore diagram, which would provide us
formance analysis of wells operated by artificial lift and, with the ability to visualize the wellbore conditions and
in some cases, flowing wells. They provide the principal the relationship between various elements. The diagram
technique to establish the fluid and pressure distribu- would answer the following questions:
tion inside the wellbore without the need to introduce • Is the pump set above the perforations?
wireline tools into the well. As presented in chapter 6 • Does liquid exist above the pump?

Figure 1.3 Detailed analysis of the pumping well performance from the acoustic fluid level survey
• Is gas flowing up the annulus? If so, at what rate? The analysis of the acoustic records acquired in these
• What is the probability of gas interfering with types of wells is greatly simplified and improved with
the pump operation? the knowledge of the actual wellbore configuration, as
• What is the percentage of liquid present at the explained in chapter 5.
pump intake?
• What are the pressures at the pump intake, the WELL TESTS
gas/liquid interface, and the perforations?
Since the analysis is based on the present well perfor-
• What is the magnitude of the pressure drawdown
mance, it is vital to have accurate well test information
from the reservoir?
that corresponds to the time frame of the acoustic records
The well completion information should be up to analysis. It is not uncommon to see operators trying to
date and as accurate as possible, especially for wellbores analyze the performance of their wells using month-old
that exhibit complicated geometries including liners, production test data or erratic test data. Such a practice
multiple perforations, and deviated trajectory, and so on. will not yield useful results.

Well tests should reflect the steady state performance example, workover, repair, equipment failure) and to
of the well. The producing flow rate has to be stabilized, interpret these static fluid level measurements in terms of
and the gas/oil ratio (GOR) and water/oil ratio (WOR) bottomhole pressure. What is required for performance
should be consistent with past performance. It is thus analysis is an estimate within 10% to 15% of the actual
important to maintain accurate historical records—ide- fully stabilized reservoir pressure. Better accuracy is not
ally, spreadsheets—of well test data, annotated with any necessary, considering the uncertainties present in the
changes in operating parameters or unusual conditions. other commonly available data. Such an estimate of static
Otherwise, erroneous conclusions are sometimes reached bottomhole pressure would be far better than the values
in this regard, based on brief production tests whose that are commonly guessed by production personnel.
accuracy is impaired by changes in the fluid content of
the casing or transient changes in fluid concentrations Producing Bottomhole Pressure
within the immediate vicinity of the wellbore4. The producing bottomhole pressure is a key element
in determining the flow behavior of the well. A steady
Static Bottomhole Pressure value of PBHP is an indication of stabilized conditions. A
The static bottomhole pressure represents the energy varying value of PBHP is an indication that the behavior
available to move the fluid from the reservoir to the of the well is changing and its current performance may
wellbore. It is one of the key factors in the analysis of not be representative of the “normal behavior” of the
well performance. If the amount of energy (pressure) well. This is especially important in reservoirs produced
available from the reservoir is unknown, then production by secondary recovery methods. A variation in PBHP
efficiency cannot be determined. Unfortunately, SBHP may also indicate that the acquired data is inaccurate.
is one of the least known quantities and is commonly The need for precise measurement of casing pressure
ignored by production operations personnel, as it is and casing pressure buildup rate cannot be overstressed.
supposed to be a value pertinent only to reservoir perfor- A stabilized value of PBHP combined with the SBHP
mance, which can be estimated only from complicated is required to calculate the well drawdown and estimate
and expensive transient pressure tests. the PI or IPR of the well. If the PBHP is small compared to
Chapter 7 of this handbook discusses in detail how the SBHP—say, 10% to 15%—then the well is probably
pressure transient test pressures are obtained with the being produced at 80% to 90% of its maximum rate. If
use of acoustic fluid level systems that are programmed the PBHP is only 50% of the SBHP, then fluid could be
for automatic and unattended acquisition of surface produced at a greater rate from the reservoir, provided
pressure and fluid level depth during the shut-in period. that the artificial lift system is adequately designed and
The fluid level records are analyzed by the software to has the additional displacement capacity.
yield accurate values of bottomhole pressure versus time. Thus, in order to judge how efficiently a well is being
The data is then exported for detailed pressure transient produced, it is necessary to know both the static and the
analysis using specialized software to estimate formation producing bottomhole pressures and have the results of an
damage, permeability, and SBHP. accurate well test. If any one of these three parameters is
A realistic estimate of SBHP is necessary for well grossly inaccurate, the wrong conclusion will be reached.
performance analysis. Every effort must be made to obtain Virtually all oil operators undertake fluid level
representative values that can be used in estimating PI measurements in pumping wells. Unfortunately, a large
or IPR relations. Otherwise, these relations are totally number of these measurements are only used to determine
meaningless. One way to obtain realistic SBHP data, at a if there is fluid above the pump intake. This ignores the
minimum cost and without additional loss of production, potential of converting these measurements into well
is to institute a policy to routinely make static acoustic performance visualization and analysis to answer the
fluid level measurements any time that any well in a field key question: What is the well’s actual performance in
is shut in for several days due to operational reasons (for relation to its potential?

Despite the advent of portable digital instrumenta- • In a well producing gas from the casing annulus
tion that incorporates well performance analysis calcula- into a flow line at constant pressure, the fluid
tion and display in real time at the well site, numerous column below the gas/fluid interface consists of
operators still monitor wells using strip-chart type a gas-liquid mixture with gas bubbling through
instruments, as discussed in chapter 4. Therefore, the the liquid all the way from the producing perfo-
following discussion is based on the assumption that rations to the gas/fluid interface. This is defined
all operators have access to at least a strip-chart type as a gaseous liquid column5.
of fluid level instrument that is in good condition, as • The liquid in the gaseous column above the pump
well as a pressure gauge or digital pressure sensor with intake consists of 100% oil, regardless of the
sufficient accuracy to measure well pressure within 1 WOR measured during a production well test.
psi. These instruments allow determining the depth to • The liquid below the pump intake will contain
the fluid level, the wellhead pressure, and its change as a percentage of water greater than the well test
a function of time during the fluid level test. These are WOR due to water holdup. For practical purposes,
the basic variables required for visualization of the exist- the density of this liquid can be assumed to be
ing conditions in the well and calculation of wellbore the density of the produced brine.
pressure distribution. • In a well producing with a closed casinghead
valve and exhibiting a constant casinghead pres-
Calculating the PBHP sure, the fluid column in the annulus below the
gas/liquid interface consists primarily of liquid,
The following relates to calculation of producing bottom-
because free gas is not entering the wellbore
hole pressure in wells that have an open annulus from the
through the formation perforations.
casinghead to the perforations. In these cases, a bottomhole
• The flow rate of gas bubbling through the liquid,
pressure calculation can be based on a measured casing-
the annular area, and the liquid properties deter-
head pressure plus the calculated gas column pressure
mine the percentage of liquid that is present in a
and the fluid column pressures extending to the bottom
gaseous liquid column. The larger the gas flow
of the wellbore. This method is probably applicable to at
rate, the smaller the percentage of liquid present
least 95% of all wells produced by pumping (rod pump,
in the gaseous liquid column. For high gas rates,
electrical submersible pump, or ESP, and progressing
both liquid and gas can be produced out of the
cavity, or PC, pump); the remaining 5% corresponds to
casing valve. As a result, there is not a definite
those wells that are completed with packers that isolate
gas/liquid interface in the annulus. This is known
the upper annulus from the formation perforations. as a flumping well.
Calculations require knowledge of the distribution of • Closing the casing valve in a well producing gas
fluids (oil, water, and gas) in the annulus or tubing depend- from the casinghead will result in an increase in
ing on the flow path. In a well producing at steady-state casinghead pressure. The casing pressure increase
conditions (stable flow rate, GOR, WOR, and casinghead versus time is a measure of the gas flow rate
pressure) through the tubing suspended in an open annulus, entering the wellbore through the perforations
the following basic concepts have been established from and percolating through the liquid. The faster
many years4 of experience and are generally accepted: the increase in surface pressure, the greater the
• A stable fluid level and a stable casinghead pres- gas flow rate. Over short periods of time (less
sure are an indication of a stable PBHP. than 10 minutes), a plot of the casing pressure
• The wellbore of a well producing at stabilized increase versus time should be a straight line.
conditions behaves as a fluid separator with the • Keeping the casing valve closed in a pumping
gas in the upper section, overlying a gaseous well that normally produces gas from the casing
liquid column extending to the depth of the will cause the fluid level in the annulus to be
producing perforations. depressed as the casinghead pressure increases.

In cases where the flow of free gas from the for- • Very accurate pump intake pressure (PIP) can
mation continues, the fluid level will drop to the be calculated from measurement of casinghead
pump intake if the casing valves remain closed pressure by installing a back-pressure regulator in
for a sufficient period of time. During fluid level the casinghead flow line and causing a controlled
depression, the PBHP may increase significantly increase in pressure until the fluid level stabilizes
above its normal value if tubing liquid flow is just above the pump intake. Once stabilized
limited by pump capacity. conditions have been reestablished, the measured
• A high fluid level in a well that is producing gas casinghead pressure will then constitute the major
from the annulus is not necessarily an indication component of the pump intake pressure, and the
of a high producing BHP. Correct estimation of the PBHP can be calculated with sufficient accuracy
PBHP can be made only by taking into account the to evaluate the well performance. This procedure
concentration of gas present in the annular gaseous is defined as a liquid level depression test, also
liquid column and computing an equivalent gas- known as a Walker test.6
free fluid level. Measurements have been made
in pumping wells where the annular fluid column Figure 1.4 presents the result of a liquid level depres-
consisted of only 10% to 15% liquid. sion test undertaken in a well producing with an electrical
• In a pumping well, the fact that a well is pumped submersible pump (ESP) outfitted with a bottomhole
down does not necessarily indicate that the maxi- pressure sensor that provides continuous surface reading
mum production is being obtained. of the pressure at the pump intake. The graph shows that,
3,000
2,500
GASEOUS LIQUID COLUMN HEIGHT, ft.
2,000
GAS/LIQUID PRESSURE
ESP SENSOR
1,500 BEST FIT
WALKER PIP
1,000
500
0
0 100 200 300 400 500 600 700 800 900 1,000
GAS/LIQUID INTERFACE PRESSURE, psig
Figure 1.4 Acoustic fluid level height during a liquid level depression test in an ESP well outfitted with a bottomhole
pressure sensor

as the surface casing pressure is increased in steps and requirements, surface pressure measurements from
allowed to stabilize, the height of the annular gaseous injectivity tests and, of course, bottomhole pressure
liquid column decreases in proportion to the pressure surveys run with wireline-conveyed sensors.
increase. The four measured points (the round dots) yield Just as in the case of PBHP calculation, in order to
a linear relation that can be extrapolated to zero height calculate the SBHP, it is necessary to know the distribu-
to estimate of the pump intake pressure of 940 psi (the tion and composition of wellbore fluids. This involves
green square). In this example, the pressure measured applying the following concepts that have been developed
by the downhole sensor (the orange triangle) showed a through experience8,9:
reading of 890 psi. The difference between measured • When a pumping well is shut in at the surface,
and computed pressure corresponds to about 5.6% of formation fluids will continue to flow into the
the measured value. well and accumulate in the annulus. The rate of
It is important to verify that, during the test, the pump this after-flow will decrease as the BHP increases
continues to operate at the same volumetric efficiency and eventually will stop. This effect contributes
when the fluid level stabilizes at a given wellhead pressure, to what is known as the wellbore storage effect.
so that the producing flow rate and bottomhole pressure • It is not practical to accurately predict the WOR
remain nearly constant during the duration of the test. during the after-flow period. Normally it is as-
After completion of the test, the surface pressure should sumed that the WOR will remain unchanged from
be returned slowly to the original value so that surface the WOR determined from well tests.
facilities are not overloaded by a surge in gas flow. • Fluid level and casing pressure measurement as a
function of time during the shut-in period generate
Calculating the SBHP good estimates of liquid and gas after-flow, which
Estimates of the static formation pressure can be obtained can be used in pressure transient interpretation.
from fluid level measurements in wells that have been • In general, the liquid that was present in the an-
shut in for a period of time sufficient for the wellbore nulus when the well was producing will remain
pressure to approach the pressure at the drainage radius in the annulus during a pressure buildup test and
of the well. Ideally, such pressure data is best obtained will be added to the after-flow fluid. Occasion-
by continuous measurement of fluid level and casing ally, it has been observed in high GOR wells
pressure during a pressure buildup test7. Automatic that, during the buildup test, the liquid level in
acoustic fluid level instruments are used routinely for the annulus decreases during the shut-in period.
this purpose and yield results comparable to those ob- This is caused by the gas after-flow generating
tained from bottomhole pressure gauges, as discussed a sufficient pressure increase to displace some
in detail in chapter 7. of the liquid into perforations connected to rela-
The reality is, however, that pressure buildup tests tively low-pressure stringers. This is especially
are performed rarely, since they may involve significant the case when dealing with long or multiple-zone
loss of production, which in the eyes of management perforated intervals.
is not justified by the improved knowledge of pressure • In order to accurately calculate the pressure at
at the drainage radius of the well or of the extent of static conditions, it is necessary to have informa-
wellbore damage. tion regarding the producing fluid level, produc-
However, all available indications of formation ing casing pressure, and well test. In other words,
pressure magnitude should be used to allow accurate a SBHP can be computed more accurately when
estimates of the SBHP for wells in a given reservoir. a PBHP survey was performed prior to shutting
These indications may include static fluid level surveys in the well.
in wells shut in or scheduled for workover, liquid level • Since most of the uncertainty in the SBHP calcula-
inferred during swabbing operations, kill fluid density tion is related to the composition of the gaseous

liquid column in the wellbore at stabilized flow Public Domain Acoustic Fluid Level Software
conditions, the most accurate results will be For the benefit of those pumping well operators that have
obtained when the height of that column is at a access only to strip-chart acoustic fluid level instruments,
minimum. Thus, if a programmed acoustic pres- the well performance calculations have been coded into
sure buildup test is to be performed on a well, it a simple application that can be downloaded from the
is recommended that, in preparation for the test, Echometer website: Analyzing Well Performance 2000,
the producing fluid level be depressed to near or AWP 2000, which is executable on most Windows®
the pump intake by means of a back-pressure operating systems.
regulator, as described in chapter 7. If a single- The program is public domain software applicable
point SBHP test is to be performed, improved to analysis of well performance using surface pres-
accuracy will be obtained if the casing valves sure and fluid level data acquired with any existing
are closed until the liquid level is depressed to instrumentation. The application was designed with
the pump. Then the pump is stopped, the well a user-friendly graphical interface, which uses a data
is shut in, and the pressure and liquid levels are input form organized to represent a wellbore diagram to
allowed to stabilize. (It may take several days, facilitate visualization of the fluid distribution and the
depending on the rate of gas inflow.) As a result, depth relationship between pump intake, perforations,
the computed SBHP will be more accurate, since and fluid level. The user is warned when important data
there will be a minimum of liquid accumulated is missing and offered the ability to use default values
in the wellbore. if the necessary information is unknown.
• Since most reservoirs are inhomogeneous and
The program allows the user to select the most
often faulted, it is most likely that wells producing
appropriate method for determining the distance to the
from the same rock formation will exhibit differ-
liquid level from the measured round trip travel time
ent SBHP depending on their location within the
(RTTT) of the acoustic echo, as discussed in chapter 5.
reservoir and their position relative to injectors
Figure 1.5 shows the resulting analysis obtained
or other producers. Thus, the acoustic SBHP is
by manually inputting the wellbore description, fluid
to be considered a well property reflecting the
properties, well production test, static reservoir pressure,
pressure conditions within the drainage area of the
acoustic fluid level depth, casing pressure, and casing
specific well. A byproduct of performing acoustic
pressure buildup. This particular well is being produced
static fluid level measurements whenever wells
efficiently at about 97.4% of its potential, although there
stop pumping (as suggested earlier) because of
is a 300-foot column of fluid above the pump intake.
mechanical failures is obtaining sufficient data
Although the principal application of the AWP 2000
for generation of a map of pressure distribution
program is for interpreting data obtained with strip-chart
in the reservoir that can also be very useful for
fluid level instruments, it is also useful for verifying the
reservoir management.
analysis obtained with digital fluid level recorders, such
These facts have been applied to develop accurate as the Total Asset Monitor (TAM) software11.
methods for calculation of PBHP. These methods are
implemented in various modern digital fluid level instru-
ments. The calculations take into account the pressure- INFLOW PERFORMANCE RELATIONS
volume-temperature (PVT) properties of the gas, oil, The flow performance of a well is defined by the
and water for computing the densities of the fluids and relationship between the producing flow rate and the
the corresponding gradients. The PBHP calculations corresponding pressure loss or drawdown that occurs
also account for the effect of gaseous liquid columns between the reservoir’s drainage radius and the well-
by correlating the casing pressure buildup rate with the bore. Figure 1.6 is a schematic representation of the
liquid fraction at every point in the wellbore5,10. pressure and fluid distributions in a pumping well and

Figure 1.5 Data input and output form for the well performance analysis
Mechanism of Flow into Wellbore

SURFACE FACILITIES
LIFT SYSTEM
FLOW RATE Q
Pt 100 bbl/d
PRESSURE
1,000 psi
SBHP
Dp/Dr1
STABILIZED
FLUID LEVEL DRAWDOWN
Dp/Dr2
PBHP
500 psi
re
rw DISTANCE
PBHP FLOW SBHP
PRODUCING DRIVING PRESSURE

PRESSURE 500 psi 1,000 psi
Figure 1.6 Schematic of the fluid and pressure distribution in a pumping well

in the reservoir when the production has stabilized at a or in reservoirs consisting primarily of water, where
rate of 100 bbl/day. free gas is not present in the rock’s pores. When there
The production rate of 100 bbl/day is set by the speed, is single-phase flow, the relation between the flow rate
characteristics, and volumetric efficiency of the pump and the pressures is given as a function of the oil (or
and is considered to be a constant when the casinghead water) and rock properties by the radial form of Darcy’s
pressure and fluid level are stabilized. Correspondingly, Law12 for single-phase flow, as shown in equation 1.4.
within the reservoir, the pressure distribution from the
wellbore (rw) to the drainage radius (re) also stabilizes koh (pe – pwf)
qo = –————————— Eq. 1.4
along the exponential (red) curve in the graph of pressure re
141.2Bo μo ln ––– +s
versus distance. The drainage radius is related to well rw
spacing and other geometric variables, but for radial
where
flow it can be approximated as:
qo = oil flow rate (stock tank barrels of oil
re = 118 × SQRT(Well_Spacing) Eq. 1.3 [STBO]/day)
where re is calculated in feet when the Well_Spacing is ko = permeability to oil (md)
given in acres. h = reservoir thickness (feet)
pe = reservoir pressure at the drainage
Assuming the well is drilled on a 20-acre spacing,
boundary (psig)
the drainage radius would be about 530 feet where the
pwf = flowing pressure at the bottom of the well
reservoir pressure is assumed to be 1,000 psi. When (psig)
the pump in this well is operating at a steady rate with Bo = oil formation volume factor (bbl/STB)
an effective liquid displacement of 100 bbl/day, the μo = oil viscosity at reservoir conditions (cp)
pressure at the wellbore radius (rw) opposite the perfora- re = drainage radius (feet)
tions stabilizes at 500 psi. Thus the pressure difference rw = well bore radius (feet)
between the reservoir and the wellbore, defined as the s = skin (dimensionless)
drawdown, corresponds to 500 psi.
Depending on the pressure existing in the reservoir In practice and from the standpoint of day-to-day
and the fluid flowing within, the formation is considered production operations, it is rare to have an up-to-date
to be primarily saturated with liquid (single-phase) or knowledge of reservoir and fluid properties (ko, Bo, and
a mixture of gas and liquid (two-phase). The presence μo). However, it is not necessary to know all the reservoir,
or absence of free gas is a function of the pressure and wellbore, and fluid properties shown on the right side of
temperature levels and the composition of the fluids. equation 1.4; they are implicit in the well test data. Thus,
These factors define what is known as the bubble point equation 1.4 can be rewritten in terms of field-measurable
pressure (Pb) of a given hydrocarbon composition. At a quantities (qo, pe, and pwf ) as equation 1.5. This relation
certain temperature, when the mixture is subjected to a is the definition of the productivity index (where J is
pressure lower than Pb, the fluid is present as a mixture the standard symbol for the productivity index, or PI)
of gas and liquid. When the pressure is greater than Pb, for the specific well at the specific stabilized flow rate:
the fluid consists only of liquid. J = q0 ÷ ( pe – pwf) Eq. 1.5
The relations used for the calculation of the flow
performance of the well have to take into account the where
presence or absence of free gas in the reservoir rock, as J = productivity index (STBO/day/psig)
discussed in the following sections. q0 = oil flow rate (STBO/day)
pe = reservoir pressure at the drainage
Pressure Above the Bubble Point boundary (psig)
Single-phase liquid flow exists in reservoirs where the pwf = flowing pressure at the bottom of the well
pressure is above the bubble point of the hydrocarbon (psig)

As long as the reservoir and fluid properties in 250 100

J = ———– = ————
the right side of equation 1.4 remain constant, the PI pe – 600 pe – 1,020
will remain constant as well. Note that the reservoir pe = 1,300 psig
pressure (pe) is required to calculate the PI. Ideally, 250
this value should be determined by shutting in the J = ———–— = 0.357 STBO/day/psig
1,300 – 600
well and performing a pressure buildup test. Gener-
ally, pe is considered to be a constant over relatively The well performance for single-phase liquid flow
short time intervals—over months, for example—but can be plotted as liquid flow rate versus flowing BHP.
it will change depending on the types of reservoir drive This plot is called the inflow performance relation (IPR).
mechanisms. Chapters 6 and 7 discuss in detail the An IPR graph for the previous example is shown in figure
use of acoustic fluid level measurements to determine 1.7 with the two well test points shown.
values of reservoir pressure with sufficient accuracy The slope of the line (psi/bbl) is the reciprocal
to analyze well performance. of PI. Either the graph or equation 1.5 can be used to
In wells outfitted with pumps operating with variable determine the flowing BHP for a given flow rate assum-
speed drives where changing the pump displacement rate ing that all variables (fluid and rock properties) remain
is relatively simple, it is possible to estimate both the PI constant—that is, that PI is a constant independent of
and the reservoir pressure, pe, by producing the well at the flow rate. The maximum flow rate that the reservoir
two different rates and determining the corresponding can provide to this well corresponds to the intercept of
producing bottom flowing pressures. the productivity line with the horizontal axis. At this
Consider the production well tests performed point, the PBHP is equal to zero, and the maximum flow
at two different rates, shown below: rate of about 464 bbl/day is defined as the potential of
the well. The term “potential” implies that it may not
qo1 = 250 STB/day at pwf1 = 600 psig be possible or practical to achieve this maximum rate
qo2 = 100 STB/day at pwf2 = 1,020 psig depending on the characteristics of the completion, of
1,400
SBH = 1,300 PSI
BOTTOMHOLE PRESSURE, Pwf (psiag)
1,200
1,000
SLOPE = 1/J = 2.8 PSI/BBL
800
600
POTENTIAL = 464 BBL/DAY
400
200
0
0 100 200 300 400 500
OIL RATE (STBO/Day)
Figure 1.7 Inflow performance based on the productivity index determined from a two-rate flow test

the reservoir, and of the artificial lift system. The linear available from well tests. The productivity index concept
nature of the productivity index performance, as plot- is applicable to a reservoir that is producing liquids that
ted in figure 1.7, indicates that a given increase in flow are largely gas-free. This is the case for wells that are
rate is accompanied by a corresponding increase in producing from reservoirs where the pressure is above
drawdown, regardless of the magnitude of the flow rate. the bubble point or for reservoir fluids that do not contain
Therefore, doubling the flow rate causes the doubling significant gas in solution.
of the drawdown. This linear behavior does not apply When the pressure in an oil reservoir decreases suf-
when both gas and liquid are flowing in the reservoir ficiently, a free-gas phase will develop in the rock pores
rock, as discussed in the following section. at every point where the pressure is less than the bubble
point pressure of the oil. As pressure decreases, the volume
Pressure Below the Bubble Point of free gas that occupies the rock’s pore space increases.
The productivity index is an elegant and powerful tool The volume of free gas, expressed as a percentage of the
that can be used to describe the flow capabilities of a pore volume, is defined as the gas saturation. As the gas
well at a particular time in the life of the reservoir. It volume within the pore space increases, and the pressure
includes implicitly all the influences of the reservoir decreases further below the bubble point pressure, the gas
and the completion outside of the wellbore. The data begins to compete with the oil and the water for the flow
required to establish the productivity index are readily path to the well, as shown in figure 1.8.
Depletion Causes Increase

in Gas Saturation
Initially only oil with connate water
Pressure reduction
SAND SALT WATER below the bubble
GRAIN point causes light GASEOUS
hydrocarbons (C1, HYDROCARBONS
C2, C3) to gasify.
SAND
GRAIN
OIL
OIL
Gas in the rock pores makes it harder

for oil to flow.
Figure 1.8 Schematic representation of gas saturation in an oil reservoir at a pressure lower than the bubble point
pressure

As the gas saturation increases, the permeability pressure of 800 psi (indicated as rbp in the figure by the
to gas increases and the permeability to oil decreases. vertical line). From this point to the wellbore, the rock
For a given liquid flow rate, the pressure drop per unit pores are partially filled with gas, which increases the
distance traveled by the liquid is greater. Thus, when pressure loss because of the flow of the liquid.
the fluid reaches the wellbore, the resulting producing Thus, at the wellbore (rw), the resulting produc-
bottomhole pressure is less than the value corresponding ing BHP is reduced to 200 psi instead of the 500-psi
to the same liquid flow rate for single-phase flow (one BHP shown in figure 1.6. The pressure drawdown has
with no free gas). This mechanism is illustrated in figure increased to 800 psi for the production of 100 bbl/day.
1.9, where it is assumed that the bubble point pressure A greater pressure drop in the vicinity of the wellbore
of the oil is 800 psi. that contains a higher gas saturation results in lower
As long as the pressure in the formation stays above PBHP at a given liquid production rate compared to a
800 psi, the pore space is occupied only by liquid. At situation where only liquid flows when above bubble
the stabilized flow rate of 100 bbl/day, the pressure de- point pressure. When the liquid flow rate is increased
clines from 1,000 psi at re along the same exponential further, the additional pressure drop will cause an ad-
pressure loss curve as in figure 1.6, until the point in ditional increase in the gas pore volume and a further
the reservoir where the pressure equals the bubble point increase of the overall pressure drawdown.
Gas and Liquid Flow into Wellbore
SURFACE FACILITIES
LIFT SYSTEM
FLOW RATE Q
100 bbl/d
Pt
Qoil = 100 bbl/day
Qgas = 25 MSCF/D
PRESSURE
rbp
1,000 psi
SBHP
BUBBLE Dp/Dr1
POINT Dp/Dr2
PRESSURE
Pb = 800 psi
DRAWDOWN
500 psi
200 psi
PBHP
re
rw DISTANCE
SBHP
DRIVING PRESSURE
PRODUCING GAS AND LIQUID ONLY 1,000 psi
PRESSURE 200 psi LIQUID
Figure 1.9 Producing bottomhole pressure is 200 psi for liquid and gas flow from perforations.

Figure 1.10 is a plot of the producing bottomhole when only liquid was flowing. As a result, at each flow
pressure (PBHP) as a function of the liquid flow rate rate Q1, Q2, and Q3, the corresponding PBHP is lower, as
(Q) for a well producing by means of a pumping system. indicated by the red curve. The difference in the drawdown
When the well is shut in (Q = 0), the value of PBHP between the blue and the red curves increases as the flow
is equal to the static formation pressure (SBHP). The rate increases, to the point that it would be impossible
blue straight line represents the PBHP, assuming the to produce the well at the rate Q4, since all the available
formation produces only liquid. Operating the pump at reservoir pressure is expended at the rate where the line
a displacement rate of Q1 results in a stabilized produc- intersects the horizontal axis (Qmax = 357 bbl/day).
ing pressure of PBHP1 and a corresponding pressure The red curve is the actual performance of the well
drawdown of ∆P1. When the flow rate is doubled to Q2, when both gas and liquid are produced at the perfora-
the producing bottomhole pressure decreases to PBHP2 tions. Acoustic fluid level records acquired in conjunc-
and the drawdown doubles to ∆P2, and so on for Q3 and tion with measurement of casinghead pressure provide
∆P3, until the maximum flow rate, Qmax (potential), of a tool to determine whether free gas is flowing from the
the well is achieved. formation by monitoring the change in wellhead pressure
The red line in figure 1.10 shows the inflow per- during the acquisition of the acoustic record. Details
formance relation (IPR) for the well when free gas is presented in chapter 6 indicate that if the casinghead
produced in conjunction with the oil. The presence of pressure increases during the time the flow of gas from
the gas phase means that, at a given pumping rate, the the casing to the flow line is stopped, then free gas is
pressure loss due to flow in the reservoir is greater than entering through the perforations and, in that particular
Inflow Performance (IPR)
PRESSURE
STABILIZED FORMATION PRESSURE

SBHP
∆p1
PBHP1 ∆p2 INCREASING THE FLOW

RATE CAUSES ADDITIONAL
∆p3 DRAWDOWN.
PBHP2 ∆p4
PBHP3
The slope is a function
of flow rate, defining a
PBHP4 curve known as the
inflow performance
relation (IPR).
FLOW RATE, Q
0
0 Q1 Q2 Q3 Qmax(IPR) Qmax(PI)
357 Bbl/D 500 Bbl/D
Figure 1.10 Inflow performance relations for single-phase (blue) and two-phase (red) flow in reservoir

well, the red curve should be used to establish the well’s for beam pumps, progressing cavity pumps, and elec-
inflow performance. If, on the other hand, the casing- trical submersible pumps. The advent of linear motion
head pressure remains constant, then no gas is flowing pumping units (hydraulic lift and linear drive pumping
through the perforations and the blue line (PI) may be units) has simplified the process of changing the surface
used to describe the well’s performance. stroke. However, despite these new developments, it is
The comparison of the performance relations in not very common to perform a variable rate flow test,
figure 1.10 demonstrates that using the productivity because of the loss in production when reducing the
index when free gas flows from the reservoir can lead to pumping speed and the personnel time requirement for
a substantial overestimation of the maximum production closely monitoring the progress of the test.
rate and well potential. This, of course, would lead to As illustrated in figure 1.11, the test involves oper-
an overdesign of the pumping equipment. ating the pumping system at a constant rate (Q1) for a
sufficient length of time for the producing bottomhole
Determining the IPR from pressure to stabilize at P1. Stabilization is verified with
a Multi-Rate Flow Test repeated acoustic fluid level surveys to calculate and
Accurately determining the inflow performance relation plot the PBHP versus time and with well production
requires performing a multi-rate flow test in conjunction tests to measure each flow rate accurately. The pumping
with a pressure buildup test. In a pumping well, this speed is then increased, and the flow rate is stabilized
implies changing the speed of the pump or, if the well at Q2 and the pressure at P2. The process is repeated for
is produced with a rod pump, changing the length of the an additional rate Q3 and P3, and then the pump may be
surface stroke. The advent of variable speed drives has stopped to undertake a pressure buildup test. This test
simplified the process of changing the pumping speed can also be conducted by means of programmed fluid
Measured IPR from Multi-Rate Test
PRESSURE
IPR FROM MULTI-RATE TEST
Ps P Ps
P1 P1
P2
P3
t
P2
Q MULTI-RATE TEST
q3
q2
P3 q1
SHUT-IN SHUT-IN
t
q1 q2 q3 FLOW RATE
Figure 1.11 Graphical representation of a multi-rate flow test for determining the well inflow performance

level surveys, as discussed in detail in chapter 7. The For the great majority of wells operated with artificial
buildup test will yield a more accurate estimate of the lift systems, the pressure in the reservoir is below the
static formation pressure (SBHP) in addition to forma- bubble point and a significant volume of free gas flows
tion permeability and skin effect. with the liquid from the reservoir. Thus, the productivity
The time required to perform the complete test may index model is not sufficiently accurate to determine the
vary from a few days to several weeks, depending on overall performance over a wide range of flow rates.
how rapidly flow and pressure stabilization are achieved. A more realistic inflow performance model for two-
This time period is one of the main reasons that these phase flow, known as Vogel’s Equation3, was developed
tests are seldom performed. through numerical simulation of well performance,
producing from a solution gas drive reservoir for sets of
Estimating the IPR Relation from wide-ranging, pressure-dependent fluid properties and
a One-Rate Well Test porous media flow effects, such as oil and gas relative
The need to define a well’s inflow performance to evalu- permeability, skin effect, geometric parameters, and
ate the efficiency or to design an appropriate artificial depletion level. The result of numerous simulations with
lift system spurred the development of mathematical all possible combinations of important parameters was
models that could approximate the IPR from data that the development of a generalized correlation equation
are commonly available as part of routine production that describes, with a minimum deviation, a dimension-
surveillance: less reference curve applicable to most wells producing
• Production well test flow rate in reservoirs below bubble point pressure, as illustrated
• PBHP derived from fluid level surveys by the red curve in figure 1.12.
Vogel IPR
1.0
2
Qf Pwf Pwf
= 1.0 – 0.2 – 0.8
(pwf/pR), FRACTION OF RESERVOIR PRESSURE
Qmax Pr Pr
0.8
BOTTOMHOLE WELL PRESSURE
Np/N = 0.1%, 2%, 4%
6%, 8%
0.6
BEST FIT
10%
0.4 12%
14%
0.2
0
0 0.2 0.4 0.6 0.8 1.0
PRODUCING RATE (qo /(qo)max), FRACTION OF MAXIMUM
Figure 1.12 Vogel’s dimensionless curves for different stages of reservoir depletion and best fit reference curve3

The vertical axis represents the ratio of the PBHP to always be remembered that the Vogel relation is only
the average SBHP, while the horizontal axis represents an approximation of the actual performance of the well.
the ratio of the liquid flow rate (q) to the maximum flow To produce the well at a rate close to its potential,
rate (qmax), or well potential. The dimensionless curves the pressure at the bottom of the well must be kept as
that correspond to reservoir depletion levels from 0.1 to low as possible by doing the following:
14% of oil in place are grouped in a narrow band that • Sizing the artificial lift system to produce fluid
is described by a generalized reference curve known as at a rate slightly greater than the potential of the
Vogel’s Equation (as shown as equation 1.6), where qmax well (thus preventing a liquid or a gaseous liquid
is the potential of the well, p is the average reservoir column from forming and stabilizing above the
pressure generally considered to be equal to the SBHP, perforations)
and pwf is the PBHP at a specific flow rate q less than • Locating the pump intake at or below the depth of
or equal to qmax. perforations and maintaining the surface pressure
pwf pwf 2 on the casing-tubing annulus as low as possible
q = qmax 1 – 0.2 —– – 0.8 ––– Eq. 1.6
p p This is known as keeping the well “pumped off.” This
STB mode of operation can be damaging to the pumping
qmax = maximum liquid rate —–– at pwf = 0 psi.
day system, unless “pump-off controller” systems are in
place to stop the pump when there is insufficient fluid
As with the productivity index, the data necessary for
to keep the pump full of liquid or to provide necessary
applying Vogel’s Equation can be easily obtained from
lubrication or cooling to downhole hardware.
a single-rate well test. The average reservoir pressure
may change in the long term, depending on the reservoir
Other Inflow Performance Models
drive mechanism, but generally is considered to be fairly
constant for determining present inflow performance. The need to describe the nonlinear characteristic of the
Based on a single well test rate q2 and the corre- inflow performance function in a practical and simple
sponding producing bottomhole pressure pwf (PBHP) manner has spurred the development of other relations
obtained from analysis of an acoustic fluid level survey, such as the Fetkovich’s approximation13, the Jones equa-
equation 1.6 is used to calculate the maximum flow rate tion14, and others, as described in the paper by Gallice and
qmax. The other variable that is required to define the IPR Wiggins15, where a comparison of the different relations
of the well is the static bottomhole pressure (SBHP), is presented as shown in figure 1.14. A unique, actual
which also can be obtained from the fluid level record multi-rate test with independently measured flowing
when the well is shut-in for a long time. Having these bottomhole pressures is used to compare the various
two quantities, the flowing bottomhole pressure that methods of computing the performance and the potential
corresponds to any given flow rate (between zero and of the well. Computed values, based on the single test
qmax) can be estimated by using the reference curve in point at a flow rate of 1,250 bbl/day and flowing bot-
figure 1.12 or computed by solving the quadratic equation tomhole pressure of 1,285 psi, allow extrapolating the
1.3 for the flowing pressure, as shown in equation 1.7: performance to higher rates and determining the potential
using various IPR relations.
pwf = p × {0.2 – SQRT[0.04 + 3.2 × (1 – q/qmax)]} Comparing the field-measured values (green
/(-1.6) Eq. 1.7 squares) to the values computed with the various equa-
This is valid only for values of q between zero and qmax. tions shows that none of the approximations that are
Once the relation between flow rate and producing based on a single flow test are able to forecast an ac-
pressure is established, it is used to correctly design a curate performance curve and generally yield an overly
new artificial lift system or to evaluate an existing sys- optimistic well potential (Vogel, Fetkovich for n = 1).
tem’s characteristics in relation to the potential of the A better match of the measured performance can be
well. As shown schematically in figure 1.13, it should achieved by using the relations that include at least two

Vogel IPR from One-Rate Well Test
IPR FROM
P MULTI-RATE TEST
WELL 2
TEST PBHP PBHP
= 1.0 – 0.2 – 0.8
Ps SBHP Qmax SBHP SBHP
P1
PBHP P2
P3
WELL
TEST
q1 q2 q3 9
Figure 1.13 Schematic comparison of IPR from a multi-rate test and the corresponding Vogel relation based on
single well test flow rate q2
1,600
1,400
FLOWING BOTTOMHOLE PRESSURE, psi
FIELD
TEST POINT VOGEL
1,200 FETKOVICH n=1
JONES
1,000
KLINS
SUKARNO
800
FETKOVICH
600
400
200
0
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000
FLOW RATE, STB/D
Figure 1.14 Comparison of IPR relations from Gallice and Wiggins

flow test points and the reservoir pressure to describe the an electric submersible pump (ESP). The casing pressure,
inflow performance, such as the Fetkovich approxima- which was varying slightly at about 5 psi during the test,
tion and the Jones quadratic relation. and the fluid level were monitored every 5 minutes during
Extensions of the Vogel IPR relation to three-phase 1.6 days to compute the producing PBHP. Figures 1.15
flow (oil, water, and gas) and methods applicable to fore- and 1.16 show how the fluid level and producing pressure
casting future well performance as the reservoir depletes change with rate as a function of time and include 469 data
are also available in the literature16-18. A study by the points obtained from acoustic fluid level surveys. Flow
National Institute for Petroleum and Energy Research rate was decreased from the initial value of 18,000 bbl/day
(NIPER) showed that the dimensionless Vogel relation to 9,000 bbl/day by reducing the power frequency from
can also be used to characterize the flow performance 61.8 Hz to 49 Hz. Correspondingly, the fluid level rose
of wells with inclined and horizontal wellbores19. from a depth of 775 feet and stabilized at about 650 feet.
Industry experience indicates that the Vogel’s rela- The subsequent increase in flow rate caused a decrease in
tion is widely accepted because of its simplicity, but fluid level to 700 feet and then back to the original depth
the reader should note that it is just an approximation. of 775 feet when the initial flow rate was reestablished.
To obtain an accurate definition of the well’s inflow Each data point of fluid level and casing pressure was
performance, every effort should be made to obtain an converted to producing bottomhole pressures and plotted
accurate SBHP and perform at least a two-rate flow test versus time in figure 1.16.
with flow stabilization monitored through acquisition The pressure endpoints of each flow period were
of acoustic fluid level records. averaged and used to plot the producing bottomhole
pressure versus flow rate in figure 1.17, which shows
Multi-Rate Flow Test Monitored with Acoustic the corresponding IPR and PI. The PI was calculated to
Fluid Level Records be 161 bbl/day-psi total fluids and the static BHP was
The following is an example of a three-rate inflow perfor- estimated at 3,122.2 psi by the intercept of the linear
mance test undertaken using a variable frequency drive for correlation with the vertical axis.
ESP Well—Multi-Rate Test

Acoustic Fluid Level and Casing Pressure
Select Left Axis: Depth to Liquid Select Right Axis: NONE
(Symbol: Circle) (Symbol: Triangle)
3,070.00
9,000 BFPD (49 Hz)
3,060.00
13,500 BFPD (54 Hz)

FLUID LEVEL DEPTH (feet)
3,050.00
3,040.00
Select Left Axis: Casing Pressure Select Right Axis: NONE
3,030.00
7.50
3,020.00
6.25
3,010.00
CASING PRESSURE (psig)
5.00
3,000.00
0 500.0 1,000.0 1,500.0 2,000.0 2,500.0
TIME (min)
3.75
2.50
18,000 BFPD (61.8 Hz)

96.64% bsw 1.25
0
0 500.0 1,000.0 1,500.0 2,000.0 2,500.0 3,000.0
469 DATA POINTS AT 5-MINUTE INTERVALS TIME (min)
Figure 1.15 Acoustic fluid level and casing pressure acquired during a multi-rate test in a well pumped with a
variable-speed ESP

Select Left Axis: BHP Select Right Axis: NONE

3070.00
3060.00 Q Pwf
9000 3065
13000 3042
3050.00
18000 3008
18000 3010
BHP (psia)
3040.00
3030.00
3020.00
3010.00
3000.00
0 500.0 1000.0 1500.0 2000.0 2500.0
TIME (min)
Figure 1.16 Variation of PBHP versus time during the variable rate test shown in figure 1.15
PI from Multi-Rate Test—ESP Well

Y = -0.0063X + 3122.2
Pl = 161 bbl/day IPR RELATION R2 = 0.998
3,070
3,060
3,050
BHP, psia
3,040
3,030
3,020
3,010
3,000
0 5,000 10,000 15,000 20,000
FLOW RATE, bbl/day
Figure 1.17 Inflow performance from a three-rate test

Variable speed drives are also currently available for This is particularly important when the producing water/
beam pumps and progressing cavity pump installations. oil ratio increases.
As a result, multi-rate flow tests are being performed Realistically determining well potential requires that
more often, especially in wells with good productivity, the value of SBHP be determined as accurately as possible
with the objective of optimizing the artificial lift system. from static acoustic fluid level surveys. A policy should
be established to measure fluid level and casing pressure
SUMMARY in all wells that are shut in due to mechanical problems
Analyzing well performance is an important step toward in order to determine realistic values of SBHP that can
increasing profits by improving production techniques. be used in performance calculations without suffering
Generally, the analysis is made by field observation and additional losses in production. These values of SBHP,
examination of well data. Acoustic fluid level measure- although still approximate, generally are better estimates
ments are an essential tool for operators to continually than those commonly “guessed.”
verify that wells are being produced as close to their Use the AWP program to accurately calculate pro-
optimum capacity as possible in the most cost-effective ducing pressures from fluid level surveys when other ap-
manner. Downhole pressures can be determined from the plication programs are not available to calculate pressure
measurement of surface pressure and depth to the liquid distribution and well potential.
without introducing sensors into the wellbore. In all cases, Productivity index (PI) assumes that productivity
the operator should strive to establish a representative stays constant as drawdown increases. A given increase
value of the overall efficiency of the artificial lift system. in flow rate is accompanied by a corresponding increase
The operator can then use this value in conjunction with in drawdown regardless of the magnitude of the flow
the well inflow performance or productivity as guidelines rate, doubling the flow rate causes the doubling of the
to identify those wells that are not performing optimally. drawdown. PI can result in over-prediction of a well’s
Well performance visualization and analysis requires potential when free gas is produced from the formation.
determining static and producing BHPs that are coor- When reservoir pressure is below bubble point and
dinated with accurate and recent production well tests. free gas is present in the rock pores, it is harder for the
Well performance is defined as the relationship between liquid to flow and thus the well develops a higher pressure
fluid flow rate and the pressure drawdown between the drawdown at a given liquid flow rate. Oil relative perme-
wellbore and the static formation pressure. ability decreases, gas relative permeability increases, and
A generally accepted rule of thumb states that, when produced GOR increases. When more gas is produced, it
the producing bottomhole pressure (PBHP) is less than is harder to pump and pump volumetric efficiency is lower.
10% of the static formation pressure (SBHP), the well is Pumping system design and installation should consider
producing at about 95% of the maximum rate or potential. methods to reduce gas volume that reaches the pump intake.
Accurately determining well performance requires Inflow performance relations (IPR) such as Vogel’s
that wellhead casing pressures and fluid levels be relation describe the decrease in productivity with increas-
determined at stabilized flowing conditions, so that the ing drawdown, which is characteristic of the performance
assumptions about fluid distribution in the wellbore inher- of reservoirs producing below bubble point where both
ent in the bottomhole pressure calculations are satisfied. liquid and free-gas enter the wellbore.
The overall efficiency of the artificial lift system can When the only data available are a single-rate flow test
be estimated from a comparison of the useful work that and producing BHP, the well potential can be estimated
the system is developing—that is, lifting a certain mass using the Vogel approximation. A better approximation
of material from a certain depth during a given time—to can be obtained using at least two stabilized flow rates
the power supplied to the system via the pump prime and fluid levels.
mover. The operating cost of artificial lift can be reduced Artificial lift design and analysis requires knowledge
by maintaining high overall pumping system efficiency. of accurate well performance. Accurately determining the

inflow performance relation requires doing a multi-rate 13. M. J. Fetkovich, “The Isochronal Testing of Oil
flow test in conjunction with a pressure buildup test. Wells,” SPE 4529, 1973.
An operator can remove the oil and gas present in 14. L. G. Jones, E. M. Blount, and O. H. Glaze, “Use
the wellbore but cannot control what enters the wellbore. of Short Term Multiple Rate Flow Tests To Predict
However, a drop in the producing liquid level at any Performance of Wells Having Turbulence,” SPE
producing rate without an associated change in the static Annual Fall Technical Conference and Exhibition,
liquid level would indicate wellbore damage, which could 1976.
be improved by some method of well stimulation. 15. F. Gallice and M. L. Wiggins, “A Comparison of
Two-Phase Inflow Performance Relationships,” SPE
REFERENCES Production and Facilities, 2004.
1. M. L. Haider, “The Productivity Index,” Transactions 16. M. L. Wiggins, “Generalized Inflow Performance
of AIME, 1937. Relationships for Three-Phase Flow,” SPE Reservoir
2. M. Muskat, et al.: “Use of Data on the Build-up of Engineering, 1994.
Bottom-hole Pressures,” Transactions of AIME, 1937. 17. M. L. Wiggins, et al., “Analytical Development Of
3. J. V. Vogel, “Inflow Performance Relationships for Vogel-Type Inflow Performance Relationships,” SPE
Solution Gas Drive Wells,” Journal of Petroleum Journal, 1996.
Technology, January 1968. 18. M. B. Standing, “Concerning the Calculation of Inflow
4. J. N. McCoy, “Analyzing Well Performance,” SPE Performance of Wells Producing from Solution Gas
337, SPE Secondary Recovery Symposium, 1962. Drive Reservoirs,” Journal of Petroleum Technology,
5. J. N. McCoy, A. L. Podio, and K. L. Huddleston, 1971.
“Acoustic Determination of Producing Bottom Hole 19. A. M. Cheng, “Development of an Inflow Perfor-
Pressure,” SPE Formation Evaluation, September mance Relationship (IPR) for a Slanted/Horizontal
1988. Well,” NIPER Report No. 458, March 1990.
6. C. P. Walker, “Determination of Fluid Level in Oil
Wells by the Pressure-wave Echo Method,” Transac-
BIBLIOGRAPHY
tions of AIME, 1937.
H. H. Evinger and M. Muskat, “Calculation of Theoreti-
7. Directive 005: “Calculating Subsurface Pressure via
cal Productivity Factor,” Transactions of AIME, 1942.
Fluid-Level Recorders,” Alberta Energy Regulator,
Alberta, Canada, 1978. J. J. Jakosky, “Bottom-hole Measurement in Pumping
Wells,” Transactions of AIME, 1939.
8. K. Huddleston, J. N. McCoy, A. L. Podio, and S.
Weeks, “Analyzing Well Performance from Acous- J. N. McCoy, et al., “Advanced Techniques for Acoustic
tic Surveys,” Petroleum Society of Canada, Annual Liquid Level Determination,” Proceedings of the South-
Technical Meeting, 1986. west Petroleum Short Course, 2002.
9. J. N. McCoy, A. L. Podio, K. L. Huddleston, and B. J. N. McCoy, A. L. Podio, and D. Becker, D., “Pressure
Drake, “Acoustic Static Bottom Hole Pressures,” Transient Digital Data Acquisition and Analysis From
Production Operations Symposium, 1985. Acoustic Echometer Surveys in Pumping Wells,” SPE
10. J. N. McCoy, A. L. Podio, and K. L. Huddleston, Permian Basin Oil and Gas Recovery Conference, 1992.
“Analyzing Well Performance XV,” Artificial Lift H. C. Miller, et al., “Well Behavior Based on Pressures
Symposium, 1987. and Production Data,” Transaction of AIME, 1937.
11. Echometer Co., Total Asset Monitoring (TAM) C. P. Walker, “Method of Determining Fluid Density, Fluid
Software, 2014. Pressure and the Production Capacity of Oil Wells,” U.S.
12. AAPG, “Fluid Flow Fundamentals,” AAPG Patent 2,161,733, June 1939.
Wi k i , h t t p : / / w i k i . a a p g . o rg / F l u i d _ f l o w _ G. Weber, “Fluid Level Indicator Useful in East Texas
fundamentals#Radial_flow, 2014. Field,” Oil and Gas Journal, December 1938.

Examples of Fluid Level Survey in Producing and Static Wells 2-1
2
Examples of Fluid Level Surveys in
Producing and Static Wells

In this chapter:
• Acoustic data analysis and quality control
• Recommendations for optimizing acoustic signal records
• Examples of different well types, acoustic fluid level records, and analyses
• Summary reports and comparisons of multiple records
• Testing for safety valve position and casing integrity, and determining gas composition using acoustic records
This chapter presents a series of examples of acoustic reference distance is the average length of pipe joints
records that illustrate some of the most likely cases (in feet per joint) or the distance to a known wellbore
encountered in practice. The objective is to provide cross-sectional area anomaly that creates a detectable
guidance to the reader for interpretation of the acoustic echo in the acoustic record. It is important that these
records acquired in wells with similar characteristics. reference lengths be as accurate as possible, or else the
The assumption is that the reader has access to acoustic distances computed from acoustic signal travel times
analysis software that is similar to that used to process can have very significant errors.
these records1, provides graphical representation of the The distances to specific points in the wellbore may
acoustic signal, and includes tools to determine the travel have been measured either relative to the wellhead or
time to specific echoes. relative to the rotary table of the completion or workover
rig. The difference in distance between these values cor-
GUIDELINES FOR ACOUSTIC RECORD responds to what is defined as the KB correction (kelly
bushing correction) and can be a significant quantity,
ANALYSIS
especially when dealing with wells on offshore platforms.
The acoustic record analyst should have a clear un- It is important to verify which reference point is used
derstanding of the wellbore configuration with all its when depth information is provided for input into the
geometric details. This will allow the analyst to visual- acoustic analysis software.
ize all features that may generate echoes of the acoustic
pulse transmitted from the surface. In wells that have a
complicated wellbore (multiple casing or tubing sizes, GUIDELINES FOR QUALITY CONTROL
liner, multiple perforations, and so on), it is advantageous OF ACOUSTIC DATA
to overlay the wellbore diagram onto the acoustic record To determine the position of the liquid level, it is important
using a distance scale based on the average acoustic to obtain a clear indication of the corresponding echo and
velocity in the wellbore. an accurate measurement of the round trip travel time
The conversion of round trip travel time (RTTT) to (RTTT) of the acoustic pulse. This also assumes that the
a distance implies that an accurate reference distance can moment of generation of the transmitted pulse is identified
be used as the basis of the conversion. Generally, this correctly and is used as time zero for the record.

at Austin
The presence of echoes that cannot be explained, as (stopping the pump or varying its speed). These changes
shown in figure 2.1, with knowledge of the wellbore ge- cause the fluid level to move within the wellbore until
ometry may be caused by random events. These could be it is possible to determine the echo that moves in time.
mechanical or electrical interference or acoustic signals. This echo is the most likely liquid level echo.
A second acoustic record acquired in quick succession Downhole pressure calculations from the analysis
will either show those random echoes located at differ- of the acoustic record are based on certain assumptions
ent times or not show them at all. Measurements should about the distribution of the fluids in the wellbore when
also be repeated whenever excessive acoustic noise is the flow rate is stabilized. Verify that the fluid level and
present and fluid level echo is not clearly identifiable. surface pressure are steady by repeating the acquisition
When the liquid level echo is deep, it is recommended at various times.
that a second acoustic record be acquired for a length Surface pressure measurements (of casing and
of time sufficient to observe the repeat of the primary tubing) are the basis of calculation of pressures within
liquid level echo. The acoustic signal travel time of the the wellbore. The pressure gauges or sensors used for
repeat echo should be approximately twice the time of acquisition of these values should be properly calibrated
the primary echo. and have adequate precision and sensitivity.
It is also strongly recommended to verify that the Before finalizing the analysis and generating a report,
acoustic velocity that is used in calculations or displayed verify that the most recent well test production informa-
by the software is within the range that corresponds to tion and well configuration data are used in the calcula-
the type of well and the pressure level. The acoustic tions. In particular, the following quantities should be
velocity of the gas in a given well or in wells producing checked and updated as needed:
from the same reservoirs is fairly constant over time,
• Tubing and casing diameters and depths
as long as the pressure in the well does not change sig-
• Representative average tubing joint length
nificantly. It is good practice to refer back to previous
acoustic records to verify that the current velocity value • Wellbore deviation survey’s measured and true
is within the expected range. vertical depths, which are required even when
In deep pumping wells with low wellbore pressure the pump may be located in the vertical section
where the echo from the liquid level exhibits a low of the wellbore
amplitude, it is recommended to take multiple records • Oil and water densities
while changing the mode of operation of the pump • Top and bottom depth of all perforated intervals
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
WHAT IS AT THIS DEPTH? RANDOM SIGNAL?

1.0 mV
LIQUID LEVEL
BACKGROUND
NOISE
Figure 2.1 Example acoustic record with unexplained signals

Acoustic Signal Acquisition Recommendations • Shut down a noisy pumping unit or other surface
The following recommendations apply to all types of equipment.
wells where acoustic fluid measurements are undertaken.
Additional recommendations are given in chapters deal- Liquid Level Echo is Not Clearly Identifiable
ing with specific applications. In some wells produced by artificial lift, it may not
• The quality of the acoustic record is greatly be possible to immediately and confidently identify
improved when the gas gun is connected to the the echo from the liquid level in the newly acquired
well with the shortest possible distance through acoustic record. This difficulty may arise from one of
a fully opening valve of the same diameter as the following conditions:
the gun’s connection. • Excessive random noise
• As a general rule, for best results, the gas gun • A liquid level that is below the liquid entry point
should not be connected to piping that has an • A liquid level that is obscured by a partial annular
internal diameter smaller than H inch. or tubular obstruction
• When the distance between the gun and the • A liquid level that is below a large increase in
wellhead increases, the probability of generating the wellbore diameter (as at a cased hole to open
ringing signals that may interfere with the correct hole transition)
analysis also increases.
• Adaptors and pipe size reducers may be used if In these cases, it is recommended that multiple
necessary, but they will result in reduced signal acoustic records be acquired using different pulse am-
amplitude. plitudes to improve the signal-to-noise ratio and verify
• Shooting through a needle-type valve can obstruct the repeatability of the signals.
or block the signal and is not recommended. Once the operator is satisfied that the record has
Replace the needle valve with a fully opening been optimized, he or she should initiate a change in
ball valve. the operating conditions of the artificial lift system that
would result in a change in the position of the liquid
• Shooting through a chemical pot will cause a
level and acquire additional acoustic records. For ex-
resonating signal and will give poor results.
ample, stopping the pumping system for an extended
When inspecting the acoustic trace, if an insufficient period of time will cause the liquid level to rise, and
liquid level kick is obtained, the shot was not detected, shutting in the outflow of gas from the annulus will
or the data is otherwise unsatisfactory, then it will be depress the liquid level. These changes of the liquid
necessary to increase the pressure difference between level should be reflected by displacement in time of
the wellbore and the volume chamber and repeat the the signal corresponding to the liquid level echo in the
acquisition. It is also recommended to inspect the well acoustic record.
noise before shooting. If excessive noise exists:
• Close the valve between the gas gun and the EXAMPLE ACOUSTIC RECORDS
wellbore and determine if noise was generated AND ANALYSIS
downhole or at the surface. Table 2.1 shows at a glance some acoustic fluid level
• Use a larger pressure differential in the gas gun. records that correspond to fairly common well comple-
• Increase the casing/tubing pressure to improve tion configurations. Each case is analyzed in detail to
collar echoes and the liquid level response, there- illustrate some of the difficulties encountered in con-
by improving the signal-to-noise ratio. ducting an analysis and the methods for achieving fairly
• Inspect connections and microphone cable, and confident results. Discussion of specific analysis tools
check electronic circuits. and techniques is expanded in further chapters.

Table 2.1
Typical Examples of Acoustic Records and Corresponding Analyses
Vertical Well Uniform Wellbore Hole in the Tubing, ESP Well

Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26
WHAT IS AT THIS DEPTH? RANDOM SIGNAL? 4

HOLE IN THE
TUBING?
2
1.0 mV
mV
LIQUID LEVEL
BACKGROUND
NOISE
Deviated Wellbore ft 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
HORIZONTAL DEPARTURE (ft)

Sec 0 5 10 15 20 25 30
0 2,000 4,000 6,000 8,00010,000 12,00014,000
0
10
Subsurface Safety Valve Test

CASING PRESSURE = 845 psi
02/17/14
5
VERTICAL DEPARTURE (ft)
2,000
0
mV
-5 20
4,000
SAFETY VALVE CLOSED
-10
0
6,000 PUMP -15
CASING PRESSURE = 176 psi
02/17/14
-20
mV
-40
Well with Tapered Tubing and Casing Liner

-60 SAFETY
VALVE
CLOSED
Sec 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
15 ft 0 2,000 4,000 6,000 8,000

MIRROR IMAGE
OF LINER ECHO
ABOUT LL
10
5
TAPERED
TUBING Corrosion Survey Casing Integrity
0
mV
-5
-10
LINER TOP
-15
ft 0 500 1,000 1,500 2,000 2,500
Well with Blast Joint and Upper/Lower Perforations Stratified Annular Gas Column
Sec 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
15
TOP PERFORATION 4
AT 1,903 FT COLLAR COUNT
10 STOPS HERE
2
5
0 0
mV
mV
REPEAT OF
-5 BLAST JOINT
AND TOP -2
PERFORATIONS
AT 7,137 SEC.
-10
BLAST JOINT
AT 1,870 FT -4
-15
ft 1,000 1,500 2,000 2,500 3,000 3,500 4,000 ft 0 2,000 4,000 6,000 8,000 10,000 12,000

0 2,000 4,000 6,000 8,000 10,000 12,00014,000
0
Sec 0 2 4 6 8 10 12 14 16 18 20 22
40
2,000
LL:2,347 ft
20
4,000
mV
-20
6,000
PUMP: 6,192 ft -40
ESP Well Casing Shots
ft
Producing and Static

0 2,000 4,000 6,000 8,000 10,000 12,000

Example 1 The acoustic record shown in figure 2.3 exhibits a

Simple Wellbore with Uniform Casing clear echo from the fluid level at a round trip travel time
and Tubing Diameters (RTTT) of 5.329 seconds. The background acoustic noise
This type of well represents the large majority of rod- level (recorded before time zero) is minimal compared to
pumped wells drilled and completed before the advent the amplitude of echoes from the tubing collars, which
of horizontal drilling and the practice of commingling are clearly visible and automatically counted almost to
production from multiple zones. An example of this the liquid level.
type of well is shown in figure 2.2. Most of these wells The acoustic velocity of the gas in the section of the
produce from partially depleted reservoirs. As a result, wellbore from 300 to 900 feet averages 1,157 ft/s, as
the annular fluid level at producing conditions is gener- shown in figure 2.4, while the acoustic velocity of the
ally found in the lower section of the well. gas within the annulus from the surface to the depth of
Casing Stroke Length Tubing

OD 5.500 in Length 100.00 in OD 2.375 in
Weight 15.50 lb/ft Weight 4.60 lb/ft
Top 0.00 ft Top 0.00 ft
Bottom 5,221.00 ft Bottom 5,115.00 ft
Avg Joint
Length
31.700 ft
Rod String
# Tapers 2
VIEW
Pump
Plung Dia 1.250 in
SN Depth 5,115.00 ft
Length 12.00 ft
Top Hold Down

Tubing Anchor Bot Hold Down
Depth 5,035.00 ft Tubing Pump
Traveling Barrel
Perforation
Top 5,200.00 in
Plugback
Bottom 5,221.00 ft
Depth 5,300.00 ft
Figure 2.2 Vertical well with a rod pump set at 5,115 feet. Average production is 27 bbl/day of oil, 60 bbl/day of
water, and 40 Mscf/day of gas.

Figure 2.3 Acoustic record with typical echoes from the liquid level and tubing collars
Figure 2.4 Detailed analysis of echoes from tubing collars recorded between 0.5 and 1.5 seconds from pulse
generation

the liquid level averages 1,175 ft/s. This overall average annular gaseous column consists of only 37.75% liquid
value is used to compute the distance to the liquid level because of the gas flow of 27.6 Mscf/day, as evidenced
as 3,131 feet from the wellhead. by a casing pressure buildup of 1.3 psi in 2 minutes that
In the annulus, there is 1,984 feet of gaseous fluid was recorded during the test.
above the pump. This corresponds to an equivalent As shown in figure 2.5, the producing bottomhole
gas-free liquid pump submergence of 749 feet, since the pressure of 437 psi computed at the depth of 5,221 feet in
Fluid Above Pump 1,984 ft

Liquid Level 3,131 ft Equivalent Gas Free Above Pump 749 ft
Gun Production Modify...

Date Entered 02/24/95
Current Potential
Oil 27 31 BBL/D
Water 60 69 BBL/D
Gas 40.0 46.1 Mscf/D
IPR Method Vogel
Producing Efficiency 86.72%
Casing Pressure
Pressure 127.9 psi (g)
Annular Gas Flow

Gas Flow 27.6 Mscf/D
1 2
Fluid Properties
% Liquid Above Pump 37.75%
% Liquid Below Pump 52.08%
Fluid Properties
PIP 411.9 psi (g) @ 5,115 ft
PBHP 437.0 psi (g) @ 5,221 ft
SBHP 1,485.3 psi (g) @ 5,221 ft
Gas/Liq Interface 140.9 psi (g) @ 3,131 ft
Depths
Pump Intake Depth 5,115 ft
Formation Depth 5,221 ft
Sensor Serial No.

WG Unknown
Collar Analysis (Automatic) Casing Pressure Buildup

131
130
129
128
0 0.5 1 1.5 2
Delta Time, Minutes
Casing Pressure 127.9 psi (g)
Acoustic Velocity 1,175 ft/s Buildup 1.3 psi (g)
Joints per Second 18.53 jts/sec Buildup Time 2 min 0 sec
Joints to Liquid 98.77 jts Gas Gravity from Acoustic Velocity 0.7888 Air = 1
Figure 2.5 Summary report including calculation of downhole pressures and production potential

relation to the static bottomhole pressure of 1,485.3 psi After running the rod string and completing pump
indicates that the well is producing at 87% of its poten- installation, several acoustic records were acquired
tial based on Vogel’s approximation. Liquid production at different times during the pump-down sequence to
could be increased from 87 to 100 bbl/day, assuming monitor the progress of the well drawdown until flow
that currently the pumping system is not operating at stabilization was reached. Pump-down was performed
capacity and the casinghead pressure could be reduced at a constant pumping speed while keeping the casing-
to a minimum. to-flow line valve closed. The casing pressure increased
from an initial value of 116 to 845.8 psi by the time
Example 2 the last acoustic record was acquired. High-frequency
noise, probably caused by a resonating manifold where
Deviated Wellbore
the gun was attached, is present in the acoustic record.
This example acoustic record is from a deep horizontal Although the liquid level echo is visible in the raw data,
well produced with a rod pump that is landed in the as shown in figure 2.7, proper analysis requires applying
vertical section at 10,415 feet, as shown in figure 2.6. a low-pass filter to highlight the liquid level echo. The
0 2,000 4,000 6,000 8,000 10,000

0
2,000
4,000
6,000
LL 6,099 ft
8,000
10,000
PUMP: 10,416 ft
Figure 2.6 Wellbore trajectory and pump location

Sec 0 5 10 15 20 25 30 35
40
RAW DATA
20
REPEAT OF
LIQUID LEVEL
0
mV
LIQUID LEVEL
AT 6,093 FT
-20
-40
ft 0 5,000 10,000 15,000 20,000
Sec 0 5 10 15 20 25 30 35
15
10 LOW-PASS FILTERED DATA
5
REPEAT OF
LIQUID LEVEL
0
mV
-5
LIQUID LEVEL
AT 6,093 FT
-10
-15
ft 0 5,000 10,000 15,000 20,000
Figure 2.7 Comparison of raw and filtered acoustic record
acoustic velocity was obtained from the frequency of Acquiring multiple records as a function of time
the echoes generated at the tubing collars and showed is a good way to monitor changes in fluid level that
an increase from 1,125 to 1,162 ft/s, corresponding to correspond to changing conditions in the well. Subse-
the casing pressure increase. quent positions of the liquid level are easily compared

by overlaying the traces on a graph with a common Whether the potential total liquid rates of 155 bbl of
depth axis, as shown in figure 2.8. The blue record was oil and 251 bbl of water are achievable depends on the
acquired at the start of the pump-down and shows the ability of the pumping system to displace the liquid from
liquid echo at a shallow depth of 1,245 feet. Following the wellbore into the tubing at the rate of 406 bbl/day.
the liquid level echo, at least five repeats of the liquid
level echo exhibit decreasing amplitude due to sound Example 3
attenuation as the wave travels longer distances. The Tapered Tubing and Casing Liner
black trace was acquired after approximately 24 hours This vertical well in a heavy-oil reservoir is completed
of continuous pumping at a constant casinghead pressure with a perforated liner and a tubing string that tapers from
and shows the liquid level at 6,093 feet with one repeat a 4H-inch OD to 3H inches from a depth above the liner
of the liquid level echo. to the pump setting depth. The purpose of the tapered tub-
The summary report shown in figure 2.9 indicates ing is to minimize the frictional loads and pressure losses
that the liquid level is stabilized with about 1,219 feet of due to the high viscosity of the produced oil. Figure 2.10
liquid submergence of the pump. Gas is flowing up the shows the acoustic record acquired in this relatively shal-
casing annulus at a rate of 79.8 Mscf/day. The annular low well that includes echoes from the change in tubing
gas flow causes aeration of the liquid and generates a diameter, the top of the casing liner, and the liquid level
gaseous fluid column with a liquid concentration of located within the perforated interval.
29.97% that extends 4,067 feet above the pump. The analysis of the acoustic record and the calcula-
The pressure at the pump intake is computed as tion of the depth to the liquid level is performed using the
1,397.3 psi and the producing bottomhole pressure (at the echo from the tubing taper as the depth reference point to
heel of the horizontal) is 1,543 psi. Based on the Vogel compute an average acoustic velocity in the annular gas
inflow performance relation, the formation is producing of 1,450 ft/s. This is a typical value normally observed in
at about 84% of the well’s potential, assuming the static heavy oil wells (whose oil has API gravities from 10 to
formation pressure is 4,542.9 psi. 15°) with low casinghead pressure. The liquid level depth
Sec 0 5 10 15 20 25 30
10
END
Casing Pressure = 845 psi
02/18/14
5
0
mV
-5
-10
-15
START
Casing Pressure = 176 psi
02/17/14
ft 0 2,000 4,000 6,000 8,000 10,000 12,000
Figure 2.8 Overlay of records acquired at the start and the end of pump-down

Liquid Level 6,350 ft Fluid Above Pump 4,067 ft

Equivalent Gas Free Above Pump 1,219 ft
Gun—WG 3837 Production Modify...

Current Potential
Oil 130 155 BBL/D
Water 210 251 BBL/D
Gas 285.0 339.8 Mscf/D
IPR Method Vogel
PBHP/SBHP 0.34
Casing Pressure
Annular Gas Flow

Fluid Properties
% Liquid Below Pump Modify... 45.11%
Fluid Properties
PIP 1,397.3 psi (g) @ 10,416 ft
PBHP 1,543.0 psi (g) @ 11,070 ft
SBHP 4,552.0 psi (g) @ 11,070 ft
Gas/Liq Interface 1,041.0 psi (g) @ 6,350 ft
Depths
Casing Pressure Buildup

Acoustic Velocity
847
846.5
Acoustic Velocity 1,162 ft/s
846
Joints per Second 18.33 jts/sec
Joints to Liquid 200.30 jts 845.5
Gas Gravity 0.8084 Air = 1 0 0.5 1 1.5 2

Delta Time, Minutes
Entered from Known Acoustic Velocity Buildup 0.9 psi (g)
Buildup Time 2 min 0 sec
Gas Gravity from Acoustic Velocity 0.8084 Air = 1
Figure 2.9 Summary acoustic fluid level report with pressure distribution and potential analysis
is very near the pump intake, as verified by the dyna- tubing diameter change and includes an estimated gas
mometer record that shows a pumped-off characteristic. gravity of 0.6, corresponding primarily to methane.
Figure 2.11 presents the details of the calculation of The down-kick echo from the top of the liner could
acoustic velocity based on the up-kick echo from the also be used as the reference point. However, in this

Sec 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5

15
MIRROR IMAGE
OF LINER ECHO
ABOUT LL
10
TAPERED
TUBING
5
0
mV
-5
-10
LINER TOP
-15
ft 0 500 1,000 1,500 2,000 2,500
Figure 2.10 Acoustic record showing multiple echoes
Figure 2.11 Detailed analysis of an acoustic record using the depth to the downhole marker at the echo where the
tubing tapers from 4½ to 3½ inches

case, it is very close to the change in tubing diameter most likely they may be partially plugged by viscous
and thus would yield essentially the same results. It heavy oil deposits.
should also be noted that there is not a clear echo (up- The summary report shown in figure 2.12 indicates
kick) from the top of the perforations, indicating that that the well is producing at 96% of its potential, based
Fluid Above Pump 4 ft

Liquid Level 2,113 ft Equivalent Gas Free Above Pump 4 ft
Gun Production Modify...

Current Potential
Oil 90 94 BBL/D
Water 1 1 BBL/D
Gas 2.5 2.6 Mscf/D
IPR Method Vogel
Casing Pressure
Pressure –0.0 psi (g)
Annular Gas Flow

Gas Flow 0.1 Mscf/D
Fluid Properties
Fluid Properties
PIP 2.3 psi (g) @ 2,117 ft
PBHP 30.6 psi (g) @ 2,180 ft
SBHP 326.0 psi (g) @ 2,180 ft
Depths
Sensor Serial No.

WG Unknown
Marker Analysis Casing Pressure Buildup

3
2
Marker Used Inner Tubing OD Change
Depth 1,820 ft 1
0
Joints per Second 23.50 jts/sec 0 0.5 1 1.5 2 2.5 3
Delta Time, Minutes
Joints to Liquid 68.51 jts
Casing Pressure -0.0 psi (g)
Buildup 0.0 psi (g)
Gas Gravity 0.6026 Air = 1
Figure 2.12 Summary acoustic fluid level report, including the calculation of downhole pressures and production potential

on the calculated producing and static bottomhole pres- height, these will absorb the majority of the acoustic
sures and the fact that the liquid level is below the top pulse and make it difficult to observe the echo from the
of the perforated interval and the casinghead pressure liquid interface.
is zero (atmospheric), resulting in a low PIP of 2.3 psi. The well is vertical with the pump intake located
near the bottom of the perforated interval, as shown
Example 4 in figure 2.13. The completion string includes a set of
heavy-wall tubing joints (blast joints) located opposite
Well with Blast Joint Opposite the upper zone to provide resistance to erosion that
Upper Perforations
can be caused by the gas and liquid inflow through the
This example illustrates that, when the liquid level is perforations. The top of the blast joints is located at a
below upper and lower perforated intervals of significant depth of 1,270 feet.
KB Offset 15.0 ft Production Data

Avg Joint
Length
30.960 ft
Rod String
# Tapers 2
VIEW
Pump
Plung Dia 2.000 in
BLAST JOINT @ 1,270 ft
Length ft
Top Hold Down

Tubing Anchor Bot Hold Down
ft Tubing Pump
Traveling Barrel
Perforation
Top 1,903.00 ft
Plugback
Bottom 3,528.00 ft
Depth 3,661.00 ft
Figure 2.13 Well with multiple perforated intervals

The larger outside diameter of the blast joints causes The round trip travel time (RTTT = 5.496 seconds) to
an annular area restriction that generates a distinct the echo generated at the top of the lower perforated
echo with a downward deflection, as seen in figure interval (2,825 feet) is used to calculate the average
2.14. This echo is almost immediately followed by the acoustic velocity of 1,028 ft/s.
echo from the top of the upper perforated interval and After the echo from the deep perforations, one
then the echo from the top of the deeper set of perfora- would expect to observe the echo from the liquid level
tions, both exhibiting upward deflections (up-kicks). at a depth somewhere above the end of the tubing. No
Sec 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8

15
TOP PERFORATION
AT 1,903 FT
10
0
mV
REPEAT OF
-5 BLAST JOINT
AND TOP
PERFORATIONS
-10 AT 7,137 SEC.
BLAST JOINT
AT 1,870 FT
-15
ft 1,000 1,500 2,000 2,500 3,000 3,500 4,000
Figure 2.14 Acoustic record with echoes from the blast joint and perforations
such echo is observed, even when amplifying the scale least two acoustic records: the first just before the pump-
of the graph and zooming in, as shown in figure 2.15A. off controller starts the pump and the second just after
Since the well is producing by means of a rod pump the pump is turned off. A comparison of the two records
operated with a pump-off controller, it was possible to would show the variation of the signal corresponding to
acquire dynamometer data that indicates that the pump the movement of the gas/liquid interface. Unfortunately,
liquid fillage decreases from 95% (almost a full pump) most lease operators do not perform this check because
immediately after starting the pump and then decreases of lack of time or adequately trained personnel.
to and stabilizes at 50% fillage after about 8 minutes This example also illustrates the usefulness of dy-
of operation when the pump shuts down. Therefore, it namometer data for verifying the analysis of acoustic
is clear that the liquid level was very near the depth of records acquired in rod-pumped wells that exhibit
the pump intake when the acoustic record was acquired. complex wellbore geometries.
This information is used to manually locate the liquid
level (LL) reference line, as shown in figure 2.15B, by Example 5
aligning it at the crossing of a downwards deflection ESP Well Casing Shots: Producing
and the zero mV line. and Static Well
A method to verify that this choice of the liquid level The following example acoustic records were acquired
position is reasonably accurate would be to acquire at in a horizontal well with an electrical submersible pump

B
Sec 4.8 5 5.2 5.4 5.6 5.8 6 6.2 6.4 6.6 6.8 7 7.2 7.4 7.6 7.8 8
4
LIQUID LEVEL ESTIMATED
FROM DYNAMOMETER
0
mV
-2
REPEAT OF TOP
PERFORATIONS
AT 7.2 SECONDS
-4
ft 2,600 2,800 3,000 3,200 3,400 3,600 3,800 4,000
Figure 2.15 Detail of an acoustic record showing (A) the repeat echo from the top perforations and absence of an
identifiable liquid level echo and (B) the liquid level marker located at pump intake depth, based on dynamometer
determination of pumped-off condition
(ESP) installed near the bottom of the vertical wellbore of 600 bbl/day of oil plus 2,800 bbl/day of water and
at a depth of 6,192 feet, as shown in figures 2.16 and again several days later after stopping the pump for a
2.17. Fluid level surveys were performed with the pump sufficient length of time to determine the static bottom-
operating and the well producing at a steady state rate hole pressure (SBHP) in the formation.

Casing Tubing
OD 7.000 in OD 2.875 in

Avg Joint
Length
32.300 ft
ESP
Depth 3,661.00 ft
Perforation
Top 9,136.00 in
Bottom 21,604.00 ft
Plugback
Depth 21,630.00 ft
Figure 2.16 Well completion schematic
HORIZONTAL DEPARTURE, ft
0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
0
VERTICAL DEPARTURE, ft
2,000
LL: 2,347 ft
4,000
6,000
PUMP: 6,192 ft
Figure 2.17 Wellbore

trajectory, ESP location,
and producing fluid level

PRODUCING FLUID LEVEL RECORD The frequency of this signal is of the order of 25 Hz,
As discussed in chapter 8 of this handbook, acoustic which is larger than the normal collar echo frequency
records acquired in ESP wells typically exhibit signifi- in the range of 15 to 22 Hz.
cant high-frequency noise, which is created primarily The liquid level echo and its repeated echoes are
by the presence of equally spaced steel bands used to visible, but they can be visualized more clearly after
clamp the power cable to the tubing in the middle of applying a low-pass filter to reduce the amplitude of
each tubing joint. The change in annular cross-sectional the high-frequency noise, as seen in figure 2.19. The
area due to the banding causes additional echoes that acoustic velocity is determined using the echoes from
can interfere with the detection of the echoes from the the tubing collars by using digital filtering with a high
tubing collars and with clear identification of the liquid. band-pass filter that minimizes interference caused by
Figure 2.18 displays the recorded acoustic signal with echoes from the cable clamps, as seen in figure 2.20.
multiple echoes from the relatively high liquid level. The detailed analysis of the acoustic record shown in
Note the high-frequency signal with decaying amplitude figure 2.21 yields a liquid level depth of 2,347 feet and a
that exists during the time between successive echoes. PBHP of 1,058 psi at the current stabilized production rate.
Sec 0 2 4 6 8 10 12 14 16 18 20 22
40
20
mV
-20
-40
ft 0 2,000 4,000 6,000 8,000 10,000 12,000
Figure 2.18 Noisy record due to a resonating cavity or cable banding
Sec 0 2 4 6 8 10 12 14 16 18 20 22
40
20
mV
-20
-40
ft 0 2,000 4,000 6,000 8,000 10,000 12,000
Figure 2.19 Record in Figure 2.18 filtered with a low-pass filter

Figure 2.20 Acoustic velocity determination using echoes from tubing collars after filtering the raw data
Gun
Production Modify...
Current Potential
Oil 600 739 BBL/D
Water 2,800 3,447 BBL/D
Gas 825.0 1,015.6 Mscf/D
IPR Method Vogel
Casing Pressure
Annular Gas Flow

Fluid Properties
Fluid Properties
PIP 839.7 psi (g) @ 6,207 ft
PBHP 1,057.8 psi (g) @ 15,370 ft
SBHP 2,844.0 psi (g) @ 15,370 ft
Depths
Figure 2.21 Producing fluid level analysis and pressure distribution summary

STATIC FLUID LEVEL RECORD pressure is computed as 2,844.2 psi relative to the datum
depth of 7,289 feet TVD. This is an approximate value,
The pump was stopped and the well was shut in for
since all flow from the reservoir has not stopped, as
several days to obtain an approximate value of the static
indicated by the still increasing casinghead pressure
bottomhole pressure from the fluid level, assuming that
at a rate of 0.55 psi per minute. The well should be
formation pressure buildup was near stabilization by
shut in for a longer time to obtain a better estimate of
that time. A more precise method for acoustic mea-
the SBHP.
surement of the SBHP in pumping wells is discussed
This computed value of the estimated static bot-
in chapter 7. Specifically, chapter 7 describes how the
tomhole pressure is then used to determine the Vogel
acquisition of fluid level records can be automated to
inflow performance relation and calculate the well po-
obtain a detailed description of bottomhole pressure
tential production of 739 bbl/day of oil and 3,447 bbl/
buildup for long periods of time. This data can be used
day of water that is displayed at the top right of figure
in pressure-transient analysis to obtain good estimates
2.21. The current production corresponds to 81% of
of the static reservoir pressure, wellbore skin, and
the potential production, indicating that the pumping
formation permeability.
system is properly operated and sized.
As seen in figure 2.22, the liquid level echo RTTT
is now measured at about 2 seconds, compared to 3.7
Example 6
seconds when the pump was operating, as seen in figure
2.18, and the repeat echoes are spaced much closer ESP Well with Hole in the Tubing
together. This corresponds to the rise of the liquid level The following example of a well, operated with an
during the shut-in period from an initial depth of 2,347 electrical submersible pump installed in the deviated
feet to a current depth of 1,240 feet. section of the wellbore, reported a significant decline in
The analysis of the static fluid level acoustic re- liquid production that prompted the operator to perform
cord is combined with the information from the last fluid level measurements, which showed a significant
producing fluid level to calculate the fluid and pressure increase in annular liquid level. The pump was stopped
distribution in the well, as seen in the pressure versus and the well shut in to monitor fluid distribution in the
true vertical depth (TVD) depth graph presented in the tapered tubing string (3H to 2M inches) at 7,949 feet,
summary report in figure 2.23. The static bottomhole as shown in figure 2.24.
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34
2
mV
-2
-4
ft 0 5,000 10,000 15,000 20,000
Figure 2.22 Acoustic fluid level record for a shut-in well

Producing Shot Static Bottomhole Pressure Static Shot

03/13/2014 12:38:48 am 2,844.2 psi (g) @7,289 ft 03/28/2014 09:54:03 am
Static Liquid Level 1,230 ft
Oil Column Height 2,476 ft
Water Column Height 3,583 ft
ft
0
2,000
4,000
Casing Pressure Buildup Casing Pressure Buildup

6,000
447 406
446 405
445 404
444 8,000 403
0 0.5 1 1.5 2 0 500 1,000 1,500 2,000 2,500 3,000

0 0.5 1 1.5 2
Delta Time, Minutes Delta Time, Minutes
Casing Pressure 443.9 psi (g) Casing Pressure 403.2 psi (g)
Well Test, 03/13/2014
Buildup 0.8 psi (g) Buildup 1.1 psi (g)
Buildup Time 2 min 0 sec Oil 600 BBL/D Buildup Time 2 min 0 sec
Gas Gravity 0.6749 Air = 1 Water 2,800 BBL/D Gas Gravity 0.6604 Air = 1
Casing Pressure Comments and Casing Pressure

Recommendations
Pressure 443.9 psi (g) Pressure 403.2 psi (g)
Annular Gas Flow 900 psi Annular Gas Flow

Gas Flow 88.2 Mscf/D Gas Flow 12.2 Mscf/D
Fluid Properties
Gas Free Above Pump 1,128 ft
Figure 2.23 Summary report with producing and static fluid level analyses

KB Offset 0.0 ft Production Data HORIZONTAL DEPARTURE (ft)

Casing Tubing 0 500 1,000 1,500 2,000 2,500
0
# Tapers 4 # Tapers 2
VIEW VIEW 2,000

4,000
6,000
8,000
ESP LL: 9,038 ft
Depth 11,330.22 ft
10,000
PUMP: 11,146 ft
Perforation
Top 14,393.04 in 12,000
Bottom 14,888.45 ft
14,000
Plugback
Depth 14,888.45 ft
Figure 2.24 Wellbore schematic and trajectory for Example 6 well with hole in the tubing.
The deviated wellbore has an approximate “S” To verify that the echo at about 5,000 feet is a real
profile with the bottom plugged back to 14,888 feet echo from within the wellbore and not a random signal
and the ESP located at 11,330 feet. The first fluid level caused by a spurious event, additional fluid level records
record—acquired down the tapered tubing string by were acquired at intervals of about 10 minutes and are
attaching the gas gun to the swab valve—indicates a plotted as overlays in figure 2.26.
possible hole at a depth of about 5,000 feet, as seen by The overlay shows that the echo at 5,000 feet and
the large-amplitude up-kick echo in figure 2.25. the echoes from the change in tubing diameter and
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26
4
HOLE IN THE
TUBING?
2
mV
ft 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
Figure 2.25 First acoustic record acquired through the tubing string

-2
mV
-4
-6
ft 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
Figure 2.26 Overlay of three acoustic records acquired in quick succession
from the liquid level are observed at the same depths gas in the tubing is computed as 1,202 ft/s using the
in all three records. This indicates they are caused by round trip travel time of the down-kick echo from the
wellbore changes in cross-sectional areas and not by tubing ID reduction in area at 7,949 feet, as shown in
random events. The average acoustic velocity of the figure 2.27.
Figure 2.27 Calculation of acoustic velocity using the known distance to the echo from the tubing taper

Based on this value of the acoustic velocity, the must be repeats of the echoes already received by the
distance to the hole in the tubing is computed as 5,019 microphone. This analysis is described in more detail
feet, as seen in figure 2.28. in chapter 5.
The detailed analysis of this acoustic record, shown
Detailed Analysis of Multiple Echoes in figure 2.29, indicates that—following the primary
Whenever multiple echoes are observed in an acoustic echoes generated at the hole in the tubing, the tubing
record, a detailed analysis should be performed to change in diameter, and the liquid level—one can ob-
explain the origin of each echo. Remember that once serve repeat echoes from internal reflection between the
an echo has been identified as a possible liquid level, liquid level and the tubing taper at 16.366 and 17.862
any additional echoes recorded at a later travel time seconds and a repeat of the liquid level at 29.541 seconds.
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
4
HOLE IN THE
TUBING
2
mV
ft 0 2,000 4,000 6,000 8,000
Figure 2.28 Determining the depth to the hole in the tubing
Sec 8 10 12 14 16 18 20 22 24 26 28 30
4
HOLE IN MIRRORING OF
THE TUBING TUBING ID CHANGE REPEAT OF TUBING
ID CHANGE
2
mV
-2
SECOND REPEAT OF
MIRRORING ECHO LIQUID LEVEL
-4
ft 6,000 8,000 10,000 12,000 14,000 16,000 18,000
Figure 2.29 Identification of repeat echoes from multiple reflectors inside the tubing

Example 7 to a low-amplitude echo caused by the small differ-

ence in cross-sectional area of the interior of the valve
Surface-Controlled Subsurface Safety Valve
compared to the internal diameter of the tubing. This
(SCSSSV) Testing
signal is used as the reference depth to calculate the
Flowing oil and gas wells producing on offshore acoustic velocity of the gas, since it is not possible to
platforms and many such wells producing on land are identify echoes from the tubing collar recesses of this
outfitted with subsurface downhole safety valves that tubing string.
automatically shut off the flow of fluids from within The lower record (blue) was acquired after the
the production string if the wellhead is damaged or an safety valve was actuated to close. Now, a series of
accident causes an uncontrolled outflow at the surface. repeat echoes is visible, with the first echo in line with
In offshore wells, the safety valve is located inside the the depth of the safety valve and the multiple repeat
tubing at a depth below the mud line and its internal echoes spaced evenly at constant time intervals. This
diameter is different from the diameter of the tubing record verifies that the valve did change position from
string, so that an acoustic echo is generated at a travel open to closed. However, this does not verify whether
time corresponding to the valve’s depth. the valve is providing a good seal. The high-frequency
Usually, the operation of the safety valve is con- signal that is seen at about 4,200 feet is random noise
trolled from the surface by a hydraulic or electrome- generated by some activity taking place on the platform.
chanical actuator. Because it is required to periodically
verify that the valve operates properly, the safety valve Case B: Malfunctioning Safety Valve
must be actuated, its open/closed position checked, and
its tightness verified by creating a differential pressure Figure 2.32 illustrates the acoustic records that were
across its seals. Typically, these tests are undertaken obtained in a well with a downhole safety valve that
when well flow is already stopped, when the wells have failed to operate upon receiving the “close” command
been shut in for periodic maintenance of the surface from the surface controller. A series of four records
processing facilities. were acquired for each valve position. All eight records
showed that, regardless of the surface controller indica-
tion, the subsurface valve did not change position, as
Case A: Correctly Operating Safety Valve the character of the echoes generated at the bottom of
The following example of an acoustic record is from the well and at the depth of the safety valve remained
an offshore well that has a surface-controlled subsur- unchanged.
face safety valve (SCSSV) installed inside the tubing Random high-frequency signals caused by platform
below the mud line at a measured depth of 1,178 feet, noise and vibrations can also be observed at various
as illustrated in the schematic completion diagram in times. To properly identify these as random events and
figure 2.30. not as echoes from wellbore features, it is recommended
After the well had been shut in for some time, that multiple records be acquired in quick succession
several acoustic records were acquired when the safety and compared to each other by overlaying. Echoes from
valve was open and then after the safety valve was ac- actual downhole anomalies will always be recorded at the
tuated to close. Comparison of typical records, shown same travel time and will exhibit the same characteristic
as overlays in figure 2.31, permits evaluating whether frequency content and amplitude. Random events will
the valve operated correctly or did not change position. be recorded at variable travel times and generally will
The upper record (black), acquired with the safety exhibit different characteristics.
valve in the open position, shows a clear echo from the This example illustrates one of the most useful
liquid level inside the tubing at about 4,450 feet and applications of acoustic fluid level measurement in
its repeat echo at 8,900 feet. This record also shows a analyzing the operation and performance of flowing
high-frequency signal at 1,178 feet that corresponds wells.

Casing
OD 9.625 in Subsurface Valve
Weight 47.00 lb/ft
Depth 1,178.00 ft
Top 0.00 ft
Bottom 12,880.00 ft
Tubing
OD 5.500 in
Weight 15.50 lb/ft
Landing Nipple
Top 0.00 ft ft
Bottom 12,000.00 ft
Avg Joint
Length
31.700 ft
Packer
Depth 11,900.00 ft
Perforation
Top 12,050.00 in
Plugback
Bottom 12,850.00 ft
Depth 12,880.00 ft
Figure 2.30 Wellbore diagram of a flowing gas well with a downhole safety valve
20
SAFETY VALVE OPEN
-20
mV
-40
-60 SAFETY
VALVE
CLOSED
ft 0 2,000 4,000 6,000 8,000
Figure 2.31 Superposition of acoustic records acquired with an open (black) and a closed (blue) properly operating
safety valve

20
CLOSED SSSV
0
mV
-20
-40
OPEN SSSV
-60
ft 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000
Figure 2.32 Comparison of acoustic traces for a malfunctioning subsurface safety valve
Example 8 sively overloaded. Thus, a small pressure dif-

Corrosion Survey of Intermediate Casing ferential between the gas gun and the wellbore
(of the order of 20 to 50 psi) is normally used
This example of a record was acquired in the interme- for these tests as a starting point. Then it may
diate casing string of an offshore gas lift well that is be increased in subsequent shots until a clear
undergoing an integrity test that must show the outer record is obtained.
annuli are not pressured and are filled with liquid near • The very first echo from the liquid level may
the top of the well. The test consists of monitoring the not be visible because the recorded signal could
casinghead pressure and acquiring acoustic fluid level be driven off-scale by the initial pressure pulse.
records to verify the gas/liquid interface is near the top In this case, the liquid level depth may be com-
of the well. Occasionally, the pressure in the annulus is puted starting with the first visible repeat echo
found to be nearly a vacuum and consequently, the echoes and calculating the difference of the round trip
in the acoustic record are very low-amplitude and dif- travel times between successive repeat echoes
ficult to interpret, requiring the injection of nitrogen gas of the liquid level.
to slightly pressurize the wellbore and thereby improve • Depending on conditions that exist at the well-
sound transmission and minimize acoustic attenuation. head, especially in wells on offshore platforms,
Figure 2.33 shows a schematic diagram of the well- it may be necessary to eliminate background or
bore. The annulus being tested is between the outer 13K vibration noise by applying digital filters.
intermediate casing and the 10I production casing.
Normally during these tests, the liquid level is Figure 2.34 shows an example of a record acquired
found very close to the surface, so special procedures in the implosion mode with a pressure differential of
for acquisition and analysis of the acoustic record must about 70 psi, which drives the signal off-scale for about
be followed: 0.6 seconds so that the first echo is not clearly visible
• The quality of the acoustic record is improved at the location of the “LL” black dashed line. On the
when the recording microphone is not exces- other hand, the first repeat of the liquid level echo

Casing Tubing
OD 10.750 in OD 5.500 in
Bottom 705.38 ft Bottom 6,553.00 ft
Avg Joint
Length
39.750 ft
Gas Lift Valves

# Vlvs 6
VIEW
Packer Landing Nipple
Depth 6,233.60 ft ft
Perforation
Top 6,553.00 in
Bottom 6,879.92 ft
Plugback
Depth 7,053.81 ft
Figure 2.33 Wellbore diagram for a well undergoing a casing integrity test
(at 1.070 seconds) and the subsequent additional repeats known to be 1,139 ft/s at the pressure and temperature
are easily identifiable to the end of the recording. The existing in the annulus, which gives a depth of 305 feet
analysis in this case is done by locating a reference line to the liquid interface.
at the time of arrival (1.070 seconds) of the first repeat A second record (fig. 2.35) was acquired, also in
and manually adjusting the liquid level marker (LL) at implosion mode, and resulted in visible echoes from
a round trip travel time of 1.070/2 = 0.535 seconds. In the couplings of the 10¾-inch casing string that has
this case, the acoustic velocity of the nitrogen gas is an average joint length of 40.2 feet per casing joint.

Figure 2.34 Acoustic record acquired in the intermediate casing annulus and processed with a low-pass filter
Figure 2.35 Acoustic record acquired in the intermediate casing annulus, showing echoes from the couplings

The echoes from the couplings are seen clearly in the Example 9
record sections between the repeat echoes from the Stratified Annular Gas Column
liquid level.
The manual determination of the frequency of the The following example of an acoustic record was ac-
echoes from the couplings, shown in figure 2.36, yields quired in a horizontal well producing by means of a rod
an acoustic velocity of 1,121 ft/s and a depth to the pump set near the bottom of the vertical section of the
liquid of 303 feet from the wellhead. This value is in wellbore at 9,624 feet, shown schematically in figure
good agreement with the depth determined earlier (305 2.38. The end of the lateral (toe) is located at 21,186
feet) using the acoustic velocity of nitrogen computed feet measured depth and 10,250 feet TVD.
at the pressure and temperature of the wellbore. The acoustic record shown in figure 2.39 indicates
This example, summarized in figure 2.37, illustrates that there may be a discrepancy between the wellbore
the special processing that is required to analyze the information and the automatically analyzed liquid level
acoustic record whenever the liquid level is very shallow. depth, since the computed distance to the liquid level
Figure 2.36 Determining the acoustic velocity of nitrogen gas in a wellbore using echoes from the casing collars

Fluid Above Casing Shoe 402 ft

Liquid Level 303 ft Gas Free Above Casing Shoe 402 ft
Gun
Current
Oil 0 BBL/D
Water 100 BBL/D
Gas 0.0 Mscf/D
Casing Pressure
Pressure -0.0 psi (g)
Depth
Tubing Depth 6,553 ft
Sensor Serial No.

1 2 CG 5000

Acoustic Velocity

Joints per Second 14.32 jts/sec
No Pressure Acquired
Joints to Liquid 7.62 jts
Entered from Known Acoustic Velocity Casing Pressure 1 min 0 sec

Gas Gravity from Acoustic Velocity 0.8602 Air = 1
Figure 2.37 Summary report for a casing integrity test


0 2,000 4,000 6,000 8,000 10,000
0
Avg Joint 31.880 ft
Length
2,000

Rod String
# Tapers 4
VIEW
Pump 4,000
Plung Dia 1.250 in
Length 26.00 ft
Top Hold Down 6,000
Bot Hold Down
Tubing Anchor Tubing Pump
Depth 9,528.00 ft Traveling Barrel
8,000
Perforation
PUMP
Top 10,250.00 in
Plugback 10,000
Bottom 21,100.00 ft
Depth 21,100.00 ft
Figure 2.38 Well completion schematic and simplified directional survey for Example 9 well with stratified gas column
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28
4
COLLAR COUNT
STOPS HERE
2
mV
-2
-4
ft 0 2,000 4,000 6,000 8,000 10,000 12,000
Figure 2.39 Acoustic record acquired in a deep rod-pumped horizontal well
echo at 10,573 feet exceeds the depth of the tubing string significant liquid filling the pump barrel in amounts that
at a measured depth of 9,674 feet. vary periodically from 30% to 95% during acquisition
This well is being operated with a rod pump, and of the record, as shown in figure 2.40, where the red
fortunately a long dynamometer record that includes 63 diagram corresponds to a pump barrel full of liquid and
strokes was acquired almost simultaneously with the the blue diagram to the lowest observed pump fillage.
acoustic record. The pump dynamometer diagrams show This variation in pump fillage is an indication of a type

Klb
AVG. STROKE LENGTH = 168.00 in.
25
Wrf + Fo Max = 23.66 Klb
20 Wrf = 19.16 Klb
15
10
UNANCHORED Kt
SV OPEN SV CLOSE
5
Fo Max = 4.50 Klb
TV OPEN
0 TV CLOSE
AVG EPT = 91.97 in AVG MPT = 141.19 in
0 20 40 60 80 100 120 140 160 in
Figure 2.40 Dynamometer diagrams showing the range of pump liquid fillage during 63 strokes
of two-phase flow defined as liquid slugging, which is of 10,573 feet and the known distance to the end of the
a characteristic of many horizontal wells. In sections tubing at 9,674 feet. The liquid level depth value was
of the “horizontal” lateral where the inclination of the obtained by automatically counting the echoes from tub-
wellbore is oscillating a few degrees about the horizon- ing collars and determining an average acoustic velocity
tal, pockets of gas and liquid can form. When sufficient of 936 ft/s. Note in figure 2.39 that the counting of the
liquid has accumulated, it is displaced by the gas and echoes from the tubing joints ends early at about 11.2
flows to the pump intake as a liquid slug, thus filling the seconds. Details of the collar echo analysis are shown
pump with liquid during the duration of the slug. The in figure 2.41, and the tabulation of the frequency of the
pump fillage then decreases when the gas slug pushing collar echoes (values at the right side of each two-second
the liquid arrives at the pump intake. interval) shows that there is a significant variation of the
Based on the dynamometer analysis, it can be con- acoustic velocity (from 16.67 to 12.92 jts/s) from the
cluded that the liquid level had to be located at or near upper to the lower part of the wellbore. Further details
the pump intake, since liquid was entering the pump. about this figure are discussed in chapter 5.
This conclusion requires checking that the wellbore in- Several important features are observed in the col-
formation regarding tubing and pump depth is accurate lar analysis:
and reflects the actual well construction. • Only 206 collar echoes are counted, out of 303
Since the accuracy of the well completion data tubing joints that are present in the well.
was verified, the problem becomes how to resolve the • The number of joints per second at the top of the
discrepancy between the computed liquid level value well is 16.67 and this value decreases to 13.01

Figure 2.41 Detailed results of the tubing joints count
at the end of the automatic count. This indicates Variation of Gas Composition with Depth
a significant decrease of the acoustic velocity of
A more detailed analysis of the variation of the frequency
the gas as a function of depth from about 1,063
of collar echoes determined for each one-second interval
to 830 ft/s.
of the acoustic record yields the corresponding acoustic
• The average acoustic velocity of 936 ft/s cor-
velocity variation from the surface to the point where
responds to an average jts/s value of 14.68.
the echoes from the collars become indistinguishable
• The average value of 936 ft/s is not representative
from the background noise, at a time of about 17 sec-
of the acoustic velocity of the gas in the upper
onds. These values are plotted in figure 2.42 and range
or lower third of the wellbore.
from 1,071 ft/s at the top of the well to 811 ft/s at 17
Since the average acoustic velocity is used to calcu- seconds into the acoustic record, which corresponds to
late the number of additional tubing joints from the time about 7,000 feet. The acoustic velocity averages 936
where the automatic counts stops (14.652 seconds) until ft/s over this interval.
the time when the liquid level echo is detected (21.668 The decrease in acoustic velocity with increas-
seconds), an excessively large value of total tubing joints ing depth is probably caused by a combination of the
is computed. As a result, the liquid level appears to be stratification of the gases in the wellbore, with the light
located below the end of the tubing. components accumulating near the top of the annulus

Variation of Acoustic Velocity vs. RTTT
1,100
1,050
ACOUSTIC VELOCITY, ft/sec
1,000
950
900
850
800
750
700
0 2 4 6 8 10 12 14 16 18
ROUND TRIP TRAVEL TIME (RTTT), seconds
Figure 2.42 Variation of acoustic velocity as a function of the RTTT
and the effects of increased temperature toward the must be undertaken to reduce errors due to gas proper-
bottom of the well. The liquid level depth, calculated ties such as inhomogeneity. Whenever echoes from
using the average velocity of 936 ft/s, yields a depth in known downhole markers (such as perforations, liners,
excess of the end of the tubing string, which is errone- and tubing anchor) are visible in the acoustic record,
ous, as demonstrated by the pump fillage observed in they should be used to verify the accuracy of the depth
the dynamometer records. scale determination.
In this case, the liquid level depth calculation should
be undertaken using the end of the tubing as the refer-
ence depth (a downhole marker) to estimate a more SUMMARY
representative average value of the acoustic velocity of This chapter illustrates a few of the many diverse char-
the annular gas. This is illustrated in figure 2.43, where acteristics of acoustic fluid level records and the need
the liquid level echo detected at 21.633 seconds is cor- to apply all the analysis tools available to the user to
related with the end of the tubing at a depth of 9,674 ensure an accurate determination of the distance to the
feet. This yields an average acoustic velocity of 893 ft/s liquid level. The majority of wells where acoustic fluid
for the total column of gas in the annulus, as shown in levels are acquired would be considered “simple” cases
figure 2.44. Note that this value is 32 ft/s lower than the of a vertical well with a uniform wellbore. The analysis
average value used for the initial analysis. of corresponding acoustic records is straightforward.
This example shows that, whenever the automatic Automatic analysis by software is likely to be accurate
liquid level depth calculation is based on a partial count as long as the user inputs and verifies the accuracy of
of tubing joints, a more detailed analysis of the record all the necessary wellbore and fluid information.

Figure 2.43 Determining average acoustic velocity from the known depth of the end of the tubing
Whenever records are acquired in complex wellbores REFERENCES

exhibiting multiple perforations, changes in cross- 1. Echometer Co., Total Asset Monitoring (TAM)
sectional area, anomalies such as holes in the tubulars, Software, http://echometer.com/Software/TotalAs-
inhomogeneous gas composition, or significant changes setMonitor/tabid/123/Default.aspx, 2015. (Note to
in temperature or pressure with depth, then interpreta- the reader: This software can be downloaded free
tion will require the user to study the record carefully of cost. Installation includes varied examples of
and apply his or her knowledge and experience to verify acoustic records that are documented to illustrate
that the results are accurate and reliable. the correct analysis. Viewing these examples is a
very good method to gain experience and learn how
to properly analyze the acoustic records.)

Fluid Above Pump *.* ft

Liquid Level 9,674 ft Equivalent Gas Free Above Pump *.* ft
Gun
Current Potential
Oil 22 *.* BBL/D
Water 3 *.* BBL/D
Gas 28.0 *.* Mscf/D
IPR Method Vogel
Casing Pressure
Annular Gas Flow

Fluid Properties
Fluid Properties
PIP 83.2 psi (g) @ 9,324 ft
PBHP 194.0 psi (g) @ 21,000 ft
SBHP *.*
Depths
Sensor Serial No.

WG Unknown

Marker Analysis
58
57
Marker Used End of Tubing
56
Depth 9,674 ft
55
Acoustic Velocity 893 ft/s 0 1 2 3 4
Joints per Second 14.01 jts/sec Delta Time, Minutes
Joints to Liquid 303.44 jts Casing Pressure 55.2 psi (g)
Buildup 1.3 psi (g)
Figure 2.44 Summary report for a horizontal well containing a stratified gas column


Fundamentals of Acoustic Fluid Level Surveys 3-1
3
Fundamentals of Acoustic Fluid Level Surveys
In this chapter:
• Information required for understanding acoustic fluid level records and analyzing surveys
• Propagation of sound and sound pressure waves in pipes and annuli
• Effect of composition, pressure, and temperature on acoustic velocity in gases and other fluids
• Reflection, attenuation, resonance, and interference
• Correlations for acoustic velocity calculations
The bottomhole pressure (BHP) corresponding to vari- SOUND PULSE GENERATION

ous rates of production allows for the determination of a AND WAVE PROPAGATION
well’s productivity potential (as discussed in chapter 1).
A wave is a disturbance or change from a preexisting
Thus, it is one of the most important measurements in
condition that moves in space from one point to another,
oil and gas well production studies. For pumping wells,
carrying the deviation information at a certain finite
especially rod-pumping wells, direct measurement us-
speed depending on the medium’s properties.
ing downhole pressure sensors is impractical and costly
Acoustic or sonic waves are generally caused by
because the rods must be pulled prior to installation of
pressure changes in a gas or liquid and propagate through
the sensor, which disrupts production and alters the pres-
the fluid at a speed defined as acoustic velocity, also
sure response. Permanently installed pressure sensors
known as sonic velocity. Propagation of a sonic wave
with surface readouts are not economically justifiable
requires the presence of a material medium: solid, liq-
for routine monitoring of pressure in rod-pumped wells,
uid, or gas. Sound cannot propagate in a vacuum and is
since most of these wells produce at low oil rates.
greatly attenuated when the pressure in the gas is lower
For these reasons, acoustic fluid level measurements
than atmospheric pressure. The shape or character of the
were introduced long ago with two objectives:
wave is arbitrary; it does not have to be oscillatory or
• Determining the distribution of fluids present
sinusoidal. It can be triangular, rectangular, bell-shaped,
in the wellbore1 (particularly the amount of
or spike-shaped, depending on how it is generated.
liquid above the pump intake, defined as pump
For many types of waves, their motion is described
submergence)
mathematically by the wave equation, which can be
• Estimating the dynamic and static pressures at written as:
the depth of the producing zone without the need
∂2u
to introduce any tools into the well2 c2∇2u – —– = 0 Eq. 3.1
∂ 2t
Over the years, this technology has been refined so where u is the physical property (for example, pressure in
that regulatory agencies in many states and countries a gas or strain in a solid) of the disturbance, the operator
accept the results of acoustic surveys for calculating is defined as the partial second derivatives with respect
well potentials and BHPs3,4. to rectangular xyz coordinates, t is time (in seconds),

at Austin
and c is a constant representing the speed (ft/s) at which For the conditions typically encountered in oil and gas
the wave travels. wells, with mixtures of hydrocarbon gases with impurities
For the specific case of our interest, which deals with such as hydrogen sulfide or carbon dioxide at elevated
acoustic waves propagating inside long pipes or annuli, pressure and temperature, determining the speed of sound
it is assumed that the wave propagates only along the will require considering the real behavior of the gas by
axis of the pipe (wellbore), and thus the wave equation using an equation of state that considers the phase behavior
is simplified: of the gas mixture.
∂2u ∂2u
c2 —– – —– = 0 Eq. 3.2
∂2z ∂2t GENERAL SOLUTION OF
In addition, the plane wave assumption is applied THE WAVE EQUATION
so that u is defined as the z component of the fluid’s
The solution of the one-dimensional wave equation must
particle velocity, which is a constant over the surface of
be a function of both time and distance. Experience
the front of the wave. Other assumptions include that:
tells us that many waves have a sinusoidal shape, so it
• Propagation of the wave is lossless—that is, the
may be assumed that functions that satisfy equation 3.2
amplitude of the disturbance remains constant
could be sinusoidal in nature. A possible solution could
as it travels through the medium.
be of the form:
• The velocity of propagation is constant and inde-
pendent of the pulse shape and form (frequency u = A × sin (az + bt) Eq. 3.4
content) of the disturbance.
where A, a, and b are arbitrary constants.
As discussed later, in the case of sound propagation Substitution in the wave equation yields:
in oil and gas wells, some of these assumptions may not
be applicable and the real behavior of the acoustic wave –c2a2A × sin (az + bt) + b2 A sin (az + bt) = 0 Eq. 3.5

must be considered in the analysis of some acoustic This relation can be satisfied if b =—+ac, thus giving a
records, as illustrated in chapter 2. possible solution:
The velocity of propagation of a sonic wave is a u = A × sin a(z – ct).
function of the gas’s thermodynamic properties, pres-
Similarly, many other solutions may be found, yielding
sure, and temperature. For an ideal gas, a simplified
the fact that arbitrary functions of the variables (z – ct)
relation between gas properties and acoustic velocity
or (z + ct) also satisfy the wave equation5.
can be expressed as equation 3.3:
Consider such an arbitrary function of distance and
γ (144) p (1 + α T ) time, P(z,t), to represent the amplitude of the pressure
c0 = g ———————– Eq. 3.3
ρ signal that is generated by an acoustic gas gun connected
to the casinghead of a well, as described in chapter 4.
where
The gun location is considered to be at z = 0.
c0 = speed of sound in an ideal gas (ft/s) For our discussion, P(z,t) is the waveform shown
γ = gas specific heat ratio (dimensionless)
in figure 3.1 that corresponds to a pressure pulse of a
p = pressure (lbf/in2)
duration of about one-tenth of a second (0.1 seconds) and
ρ = gas density (lbm/ft3)
a peak-to-peak amplitude of about 2 psi, as recorded in
α = constant (1/°F)
volts by the microphone in the gun as a function of time.
T = temperature (°F)
This type of wave is similar to the sound wave that
g = force units conversion [32.2 (lbm/lbf)(ft/s2)]
is generated by a balloon bursting inside a large room
This relation is useful for obtaining the acoustic where the ambient air is still. The bursting generates
velocity of single-component gases at low pressure and a pulse that increases the pressure momentarily at the
for temperatures above the critical temperature of the gas. location of the balloon, but the pressure change quickly

propagates to the adjacent air, creating a wave that travels down-kick. Conversely, when the microphone senses a
in all directions. At the origin, the short increase in pres- pressure decrease, the signal is positive and the recorded
sure is followed by a short decrease in pressure as the signal polarity is upward, forming an up-kick. As discussed
wave moves away, after which the pressure returns to the later in this chapter, this is an important feature that permits
original pressure level. Such a wave is also known as a the operator to interpret the acoustic record and detect the
free wave and its waveform approximates what is known presence of perforations, liners, landing nipples, or other
as an “N” type wave. When such a wave is generated irregularities of the wellbore cross-section.
at the wellhead, it is confined inside the wellbore and The point indicated on the figure as time zero is also
propagates only along the axis of the well. This allows known as the first break and is considered the point in
us to consider that such a wave inside the well can be time corresponding to the head of the wave. The maxi-
analyzed as a plane wave. mum value is known as a peak, and the minimum value
By industry convention, the microphone and elec- is known as a valley. The source of the pulse (the gas
tronics used in the fluid level instrument to detect the gun) is located at the origin or point in space where z = 0.
wave have been designed so that, when the pressure Let’s define the function P(z,t) as representing this
pulse corresponds to an increase in pressure, the signal particular pulse that has an amplitude value of zero at
displayed on the recorder is negative and the acoustic all times except during the tenth of a second of duration
record (amplitude versus time) deflects towards the of the pulse (that is, between t = 0 and t = 0.1 seconds)
bottom of the screen or chart in a movement called a where the amplitude is defined by the graph in figure 3.1.
TIME = 0.1 SECONDS

AMPLITUDE, volts
TIME = ZERO
TIME, t
Figure 3.1 Acoustic pulse amplitude versus time, recorded at the gas gun where z = 0

The pressure disturbance will propagate along the • Point at distance z = Z3: Amplitude at any time t
wellbore at the speed of sound of the gas. Assuming that is equal to P(Z3 – ct) is different from zero only
three more microphones are located at depths Z1, Z2, and when t is between Z3/c and (Z3 – 0.1)/c.
Z3 (in feet) in the well, they will detect the passage of
Figure 3.2 shows a graphical representation of the
the wave and generate corresponding signals. It would
signal conditions observed at each point, displaying
be useful to display what would be observed at each
three snapshots of the location of the pulse taken at
microphone location as a function of time. To illustrate
equal intervals in time:
this, the wave equation can be applied as follows:
• At time 1, the head of the pulse has passed by
• Point at distance z = Z1: Amplitude at any time t the microphone at point Z1 and the instantaneous
is equal to P(Z1 – ct). This value will be equal to signal amplitude is equal to the peak value. The
zero except at the times when t is such that the signals at points Z2 and Z3 indicate the undisturbed
quantity (Z1 – ct) is a value between zero and 0.1 background pressure level.
that is in the time interval (Z1 – 0.1)/c < t < Z1/c. • At time 2, the pulse head has arrived at point
• Point at distance z = Z2: Amplitude at any time Z2, while at points Z1 and Z3, the microphones
t is equal to P(Z2 – ct) and different from zero are indicating the undisturbed background pres-
only when t is between Z2/c and (Z2 – 0.1)/c. sure level.
Pulse Propagation with Velocity c

and Three Observers at Points Z1, Z2, and Z3
1
=
0
AMPLITUDE, u
:u
=
:u
AK
AD
PE
-4
TIME 3
HE
=
u
Y:
DISTANCE Z
LE
L
VA
TIME 2
AMPLITUDE, u
DISTANCE Z
TIME 1
AMPLITUDE, u
DISTANCE Z
Z2 Z3
Z1
Figure 3.2 Propagation of an acoustic pulse, as observed at different times by three recorders located at different
depths

• At time 3, the pulse is located between points Characteristic Diagram in the z,t Plane
Z2 and Z3. All microphones are indicating the A useful technique to present the propagation of a
undisturbed pressure level. wave is to plot the position of a specific point on the
Notice that the pulse is steadily translated, without waveform at a specific time on a graph with axes t and
change in amplitude or shape, along the direction of z, as shown in figure 3.3. Each point of the waveform
propagation. The time required by the pulse to travel falls on a line with slope 1/c and a z-intercept that cor-
the distance from Z1 to Z2 corresponds to the distance responds to the position of the specific point on the
(Z2 – Z1) divided by the acoustic speed c of the wave. waveform at time zero.
The pulse also exhibits a finite width in space, For example, a point corresponding to the head of
labeled w, that corresponds to its duration in time (0.1 the waveform will follow the line t = z/c, while a point
seconds, in this example) multiplied by the acoustic corresponding to the tail of the pulse will follow a line t =
velocity of the medium. Assuming the pulse is propa- (z + w)/c, since the tail is located at z = (-w) at time zero.
gating in air that has an acoustic velocity c = 1,100 ft/s, The characteristic diagram can be used to construct
the pulse width would be about 110 feet. the waveform that would be recorded as a function of
Characteristic Diagram
L
AD
I
TA
HE
TIME, t
AK
PE
Ttail = (Z1 + w)/c
PULSE IS GENERATED AT THE

ORIGIN (z = 0) AT TIME ZERO.
Thead = (Z1)/c PEAK TO PEAK AMPLITUDE = 5
PULSE WIDTH = w
SLOPE = 1/c
DISTANCE, z
0 Z1
PULSE
WIDTH, w
Figure 3.3 Construction of the waveform observed at position Z1, applying the characteristic diagram to the
waveform generated at Z = 0

time at a specific distance from the sound source—for inherent in the electronics used to process and display
example, at point Z1. To do this, one draws a vertical the acoustic signal.
line that intersects the characteristic lines at the times When an acoustic record from a fluid level survey
when that part of the waveform reaches the location performed at a well is analyzed, the first task is to
Z1. Thus, the head of the wave is detected at time Z1/c, identify a signal that corresponds to the echo from the
while the tail is detected at time (Z1 + w)/c, so that the liquid level. The second task is to measure accurately
time record of the microphone signal at point Z1 would the elapsed time from the moment of pulse generation to
be as shown in figure 3.4. the time corresponding to the first break of the recorded
Assuming that the acoustic velocity of the gas in signal. The next task is to convert that time interval to a
the wellbore is equal to 1,100 ft/s and the microphone distance, considering that—since the pulse is generated
is located at a measured depth of 300 feet from the by the gas gun at the surface and the echo is detected by
surface, the first break of the trace would be detected the microphone at the same location (the wellhead)—that
at T1 = 300 ÷ 1,100 = 0.273 seconds from the start of recorded time, from zero to the first break, corresponds
the recording. All signals observed before the arrival to the round trip travel time (RTTT) of the wave from
of the wave (head) and after the end of the pulse (tail) the surface to the liquid level and back to the surface.
correspond to the acoustic noise (background noise) Acoustic wave reflection and transmission are dis-
present at the location of the microphone plus any noise cussed in more detail in the following section.
Recorder Signal at Point Z1
PULSE IS GENERATED AT THE

ORIGIN (z = 0) AT TIME ZERO.
ACOUSTIC VELOCITY = c
T1 = Z1/c
AMPLITUDE, µ
0
TIME, t
PULSE DURATION = 0.1 SECONDS
Figure 3.4 Observed amplitude versus time record at point Z1

REFLECTION AND TRANSMISSION Reflection at the Discontinuity of Fluid

OF A PLANE WAVE Properties
A wave will travel in the propagation direction indefi- Consider a pressure pulse generated using a gas gun
nitely as long as the medium is uniform and exhibits installed at the top of a tubing string that is partially
constant physical or chemical properties. If this condi- filled with gas and partially filled with liquid, as shown
tion is not met, then the wave’s propagation direction, in figure 3.5. The pressure pulse propagates downward
velocity, and amplitude will change. When a plane until it reaches the gas/liquid interface. At this point,
wave propagating inside a well (such as the one shown two physical conditions must be satisfied:
in fig. 3.1) encounters a change in the properties of the • The pressure must be the same on either side of
fluid—for example, from gas to liquid—a portion of the interface.
the wave energy will be transmitted and a portion will • The normal component of the particle velocity
be reflected. must be continuous across the interface.
In addition, when the fluid properties are constant
but the wave encounters geometrical discontinuities, such These conditions give rise to the splitting of the
as changes in cross-sectional area, the discontinuities pressure wave into a wave that is reflected back to the
will cause a portion of the energy to be reflected. These surface in the gas and a wave that is transmitted into
conditions are discussed in the following sections. the liquid below the interface.
PRESSURE PULSE
GENERATED AT SURFACE
Preflected
REFLECTION COEFFICIENT =
Pincident
MEDIUM1
IMPEDANCE
Z1 = ρ1C1 Z2 – Z1
R=
Z2 + Z1
Pincident
Preflected
Z2 – Z1
R=
Z2 – Z1
Ptransmitted
TRANSMISSION COEFFICIENT =
Ptransmitted Pincident
2Z2
T=
Z2 + Z1
MEDIUM2
IMPEDANCE
Z2 = ρ2C2
Figure 3.5 Reflection and transmission at a medium discontinuity

The first condition can be expressed as: Pincident Considering the previously derived relation 1 + R
+ Preflected = Ptransmitted. = T, the values of the transmission and reflection coef-
ficients are expressed in terms of the impedances of
By definition, the reflection and transmission coef- the two fluids:
ficients are dimensionless quantities that represent the Z2 – Z1 2Z2
pressure levels of the reflected and transmitted waves as a R = ——— and T = ——— Eq. 3.11
Z2 + Z1 Z2 + Z1
fraction of the pressure of the incident wave, as follows:
• Reflection coefficient R = Preflected ÷ Pincident Applying this relation to the case illustrated in figure
• Transmission coefficient T = Ptransmitted ÷ Pincident 3.5, where the fluids consist of methane gas overlying
brine and the average pressure in the gas is atmospheric,
Substituting in the pressure relation gives: the specific impedances are as follows:
1+ R = T Eq. 3.6 • Brine density = 65.5 lb/ft3
• Acoustic velocity in brine = 4,925 ft/s
The second condition that the particle velocities must • Gas density = 0.045 lb/ft3
be continuous across the interface can be expressed as: • Acoustic velocity in gas = 1,400 ft/s
• Zgas = 1,400 × 0.045 = 63
U-> + U<- = Utr Eq. 3.7
• Zbrine = 65.5 × 4,925 = 322,600
where: • Reflection coefficient = (322,600 – 63.45) ÷
U-> = particle velocity of the incident wave 322,664 = 322,536 ÷ 322,664 = 0.9996
U<- = particle velocity of the reflected wave
Utr = particle velocity of the transmitted wave Thus, when the acoustic pulse reaches the gas/liq-
uid interface, almost all (99.96%) the sound energy is
The ratio of pressure at a point in a specific fluid reflected and the amplitude of the echo will be almost
to the fluid’s particle velocity is defined as the specific equal to the amplitude of the incident wave.
acoustic impedance and is expressed as: It should be noted that a similar situation is en-
countered when a solid obstruction is encountered by
Z = p ÷ u Eq. 3.8
the wave—for example, when the end of the pipe is
This quantity is a characteristic of the specific plugged, a packer has been installed, or a paraffin plug
fluid and is related to its thermodynamic properties. has been deposited in the annulus of the well. All these
The specific acoustic impedance is normally expressed features will generate an echo that will have similar
as the product of the fluid’s density multiplied by the characteristics of an echo generated at the liquid level.
acoustic velocity of the fluid at the existing pressure For reference purposes, the acoustic velocity of 30
and temperature. API crude oil is about 4,300 ft/s. A mixture of liquid
The particle velocity equation 3.7 can then be ex- and gas will exhibit an acoustic velocity that decreases
pressed in terms of the pressures and the characteristic rapidly from that of the liquid to that of the gas as the
acoustic impedances of the respective fluids as follows: gas concentration increases. Since a foam is a gas/liquid
Pincident Preflected Ptransmitted mixture with only 2% to 3% liquid, the acoustic velocity
––––– – –––— = –––––– Eq. 3.9 of a gas/oil foam is close to the acoustic velocity of the
Z1 Z2 Z3
gas; thus, an echo will not be generated when a foam is
where the minus sign accounts for the reversed direction present in the wellbore, as the reflection coefficient for
of motion of the reflected wave. a gas/foam interface is close to zero.
Introducing the definitions of the reflection and Figure 3.6 shows a synthetic acoustic record and the
transmission coefficients, the relation is rearranged as: characteristic diagram corresponding to reflection at a
Z1 gas/liquid interface for an average acoustic velocity in
1 – R = —– T Eq. 3.10
Z2 the gas of 1,000 ft/s. The gas/liquid interface is at 3,000

MICROPHONE
SIGNAL
TIME, SECONDS
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0
1,000
2,000
3,000
DEPTH, FT
4,000
5,000
6,000
7,000
8,000
Figure 3.6 Wellbore schematic, characteristic diagram, and synthetic acoustic record for a well with a gas/liquid
interface at 3,000 feet
feet from the surface. The incident pulse is generated by causing a decrease in amplitude of each repeat echo.
the gas gun at the wellhead and propagates downward. The amplitude attenuation of each repeat echo can be
Almost total reflection at the gas/liquid interface gen- clearly seen in the acoustic record acquired in an actual
erates an upward-traveling wave, so that the first echo oilwell, as shown in figure 3.7.
is received at the microphone after 6 seconds from the The record has two horizontal scales. The upper
start of the pulse. scale is the elapsed time in seconds measured from the
The returning wave is then reflected at the casing- start of acquisition of the microphone signal. The zero
head and travels downward, where it is again reflected corresponds to the time of pulse generation. The signal
at the gas/liquid interface, which generates a repeat between time zero and time -0.5 second represents the
of the echo at 12 seconds from the shot. The wave background acoustic noise, which in this case is very
continues, but the next repeat is not observed because low compared to the amplitude of the transmitted pulse
the acoustic record is acquired for only 16 seconds. As and the echoes. The lower scale is the actual distance of
the pulse travels through the gas, some energy is lost, the reflector (the liquid level, in this example) from the

Sec -0.5 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8
150
FIRST ECHO AT
1.679 SECONDS
100 SHOT FIRED
AT TIME = 0
50
0
mV
-50 THIRD
SECOND REPEAT
COLLAR FIRST
ECHOES REPEAT
-100 REPEAT
LL at 1,017 ft
-150
ft 0 1,000 2,000 3,000 4,000
Figure 3.7 Acoustic fluid level record with multiple echoes and an acoustic velocity of 1,129 ft/s
location of the pulse generator (the gas gun), computed A vertical dashed red line is positioned on the first repeat
using a constant average acoustic velocity of 1,129 ft/s. of the fluid level echo at 3.356 seconds and shows, based
In this case, the acoustic velocity has been determined on the distance scale, that during that time the wave has
from the average frequency of the collar echoes, or traveled downward for a distance of 2,034 feet (twice
echoes generated at the tubing joints, that can be clearly 1,017 feet). The actual distance traveled by the wave
observed as the high-frequency signals recorded after the from the surface is 4 × 1,017 feet, or 4,068 feet (two
pulse is generated. These signals decay in amplitude but round trip distances). Second and third repeats are also
are visible for nearly 6 seconds at the current scale of the visible in the record, which is cut off at 8 seconds. Ad-
graph. The average frequency of the collar signals from ditional repeats would be visible if the acoustic signal
time zero to the red vertical marker labeled “C” is used had been acquired for a longer period of time.
to determine the average acoustic velocity in the gas. Note that, contrary to our earlier assumption that the
The microphone signal goes off-scale at time zero propagating pulse remains unchanged in amplitude and
when the explosion pulse is generated, with a differen- shape as it travels within the wellbore, figure 3.7 shows
tial pressure of about 200 psi between the well and the that, in the actual field record, the amplitudes of the
chamber of the gas gun. The first echo from the liquid repeat echoes of the liquid level become smaller as the
level is detected at 1.679 seconds and marked by the signal propagates over a longer distance. This is due to
vertical dashed line labeled “LL.” Because the pulse several effects:
generator and the microphone are located at the wellhead, • Because of the change in diameter between each
the indicated time of 1.679 seconds corresponds to the tubing coupling and the pipe body, small echoes
round trip travel time (RTTT), defined as the two-way are generated and the amplitude of the transmitted
travel time of the wave from the surface to the liquid wave decreases correspondingly. This effect is
level and back to the surface. The distance to the liquid discussed in detail in the next section.
level is indicated as 1,017 feet from the surface. Repeat • As an acoustic pulse travels in a viscous, thermally
echoes are observed at 3.356, 5.035, and 6.714 seconds. conducting fluid, wave propagation is not loss-

less, even in tubing of uniform diameter with no the medium is uniform and its properties are the same
couplings (such as coiled tubing). Energy is used in both sections of the pipe.
because of friction and heat transfer, causing an To determine the wave reflection and transmission
attenuation of the signal. This effect varies with coefficients, it is necessary to apply the flow and pres-
the fluid properties and operating conditions. In sure equilibrium conditions at the junction of the two
general, amplitude losses are most important conduits. Figure 3.8 shows schematically that the pipe
when the ambient pressure is low (near atmo- is filled with a uniform medium. The upper part has a
spheric) and become progressively less severe cross-sectional area A1 and the lower part has a smaller
as the pressure increases above 10 psi. cross-sectional area A2. A typical field situation would
occur when the operator acquires an acoustic record in
Figure 3.7 shows the result of the combination of
a well completed with a casing liner when the tubing
these two effects: The peak-to-peak amplitude of the
has not been installed.
liquid level echo decreases from 270 mV to a peak-to-
Continuity of pressure at the junction requires that:
peak value of 180 mV for the first repeat, after having
traveled an additional 2,034 feet in the wellbore. Pincident + Preflected = Ptransmitted

Eq. 3.12
Reflection and Transmission at a Geometric Material balance requires that the volume velocity
Discontinuity of the fluid above the junction must equal the volume
velocity of the fluid below the junction, so that:
A sudden change in cross-sectional area of a pipe or
annulus will generate a reflection of sound, even when Qincident + Qreflected = Qtransmitted Eq. 3.13

PRESSURE PULSE
GENERATED AT SURFACE
Preflected
REFLECTION COEFFICIENT =
Pincident
AREA A1
A2 – A1
R=
Qincident Qreflected A2 + A 1
AT THE JUNCTION
Pincident + Preflected = Ptransmitted
Qtransmitted
Ptransmitted
TRANSMISSION COEFFICIENT =
Pincident
AREA A2
2A2
T=
A2 + A 1
Figure 3.8 Reflection and transmission at an area discontinuity

The flow in terms of particle velocities that are assumed This is illustrated schematically for the case shown
to be uniform over the cross-sectional areas can be in figure 3.9, which represents a well with:
expressed as:
• 2K-inch, 4.7-lb/ft tubing installed inside 5H-inch,
Qincident = A1 × Uincident
17-lb/ft casing
Qreflected = A1 × Ureflected • a 4-inch, 4.9-lb/ft liner hung at 3,000 feet

• an annular area that changes from A1 = 0.1 ft2 to
Qtransmitted = A2 × Utransmitted
A2 = 0.038 ft2
Thus, the material balance can be rewritten as:
The liquid level is 1,000 feet below the depth of the
(A1 × Uincident) – (A1 × Ureflected) = A2 × Utransmitted Eq. 3.14 liner hanger. The average acoustic velocity in the gas
is assumed to be 1,000 ft/s.
The negative sign accounts for the reversal in direction The characteristic diagram plots the depth of the
of the reflected wave motion. Dividing by the incident wave (head) measured from the wellhead as a function
values and applying the characteristic impedance rela- of time starting with the pulse generation. The charac-
tions yields: teristic line for the downward-propagating wave has a
1 – R = (A2/A1) × T Eq. 3.15 negative slope equal to 1,000 ft/s.
Combining with the relation 1+ R = T, solving for R The reflection and transmission coefficients at the
and T yields: top of the casing liner for the downward-propagating
wave are computed based on A1 = 0.1 square feet and
R = (A1 – A2)/(A1 + A2) and T = 2A1/(A1 + A2) Eq. 3.16 A2 = 0.038 square feet:
where the indices 1 and 2 refer to the areas viewed in R = (0.1 – 0.038) ÷ 0.138 = 0.45
the direction of propagation from 1 to 2.
T = (2 × 0.1) ÷ 0.138 = 1.45
Note that in the case described in figure 3.8, when the
wave is traveling from the top to the bottom of the well, so that about 145% of the pulse amplitude is transmitted
area A1 is larger than A2, so that R at this interface is a and 45% is reflected to the surface. The liner junction
positive number. The portion of the wave that is transmitted actually amplifies the wave that is transmitted downward.
then travels to the bottom of the well, where it is totally Note that the top of the liner is behaving as if it were a
reflected and then travels back up the wellbore. When it sound source that generates an upward-propagating wave
arrives at the junction where the area increases, it sees and a downward-propagating wave. In the characteristic
an enlargement, so that now area A1 is the small area and diagram, the upward wave is represented as a line with
A2 is the large area. Thus, (A1 – A2) is a negative number, a positive slope equal to 1,000 ft/s. Based on the known
yielding a negative reflection coefficient. A portion of the distance and acoustic velocity, the arrival of this wave
wave is reflected back towards the bottom of the well, but is detected by the microphone at 6 seconds and has an
now it has negative amplitude since R is negative. The amplitude of about 45% of the initial pulse.
wave’s polarity has been inverted or, more precisely, has The transmitted downward wave will have an am-
experienced a 180° phase shift. Physically speaking, if plitude of 145% of the incident wave and travels to the
the pulse that was generated at the surface was a down- liquid level, where it is totally reflected, generating an
kick, then the echo from the first passage of the wave by upward-propagating wave that reaches the top of the
the junction will also be a down-kick. The echo from the casing liner. The reflection and transmission coefficients
closed bottom of the well will also be a down-kick, but as for this upward-propagating wave are now computed
the wave returns and passes by the junction, an internally based on A1 = 0.038 and A2 = 0.1 square feet:
reflected inverted wave (up-kick) will propagate back
R = (0.038 – 0.1) ÷ 0.138 = -0.45
to the bottom, where it will be reflected unchanged and
eventually reach the surface as an up-kick. T = (2 × 0.038) ÷ 0.138 = 0.55

MICROPHONE
SIGNAL
TIME, SECONDS
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0
1,000
2,000
DEPTH, FT
3,000
4,000
5,000
6,000
7,000
8,000
Figure 3.9 Schematic wellbore diagram, characteristic diagram, and synthetic acoustic record for the liquid level
below the casing liner top. Solid blue lines represent down-kicks; dashed red lines represent up-kicks.
Note the negative sign of the reflection coefficient, which SOUND PRESSURE WAVE PROPAGATION
indicates the change in polarity of the wave traveling IN PIPES AND ANNULI
back to the liquid level.
The graphical presentation of the polarity of acoustic
The wave transmitted through the liner top maintains
its polarity and reaches the microphone at 8 seconds, records has been standardized by the industry; primary
where a down-kick is detected. Its amplitude will be echoes that correspond to reductions in area (restric-
(1.45 × 0.55 = 0.798), about 78% of the amplitude of the tions) are always depicted as downward deflections of
initial pulse. Note that in this case the amplitude of the the signal recorded in time, while echoes that correspond
echo from the liquid level is greater than the amplitude to increases in area (enlargements) are always depicted
of the echo from the top of the liner. (This may not be as upward deflections. Repeat echoes may be subject
true in practice, since this analysis does not consider to polarity reversals, as discussed earlier, and must be
viscous and thermal losses.) analyzed carefully in detail.

Figure 3.10 illustrates these concepts using a typical Echoes are also generated as the wave propagates
acoustic record acquired in the annulus of a well that was past tubing joints in a well completed with tubing joints
completed with a tapered tubing string (3H inches in the that have external upset ends. The difference in diameter
upper part to 2M inches in the lower part) with external between the body of the tubing and the diameter of the
upset end (EUE) connections. The top panel shows the couplings is sufficient to generate distinct echoes at a
complete record from 1 second before the shot is fired (at relatively high frequency. These signals are defined as
time zero on the upper time scale) to about 16 seconds. collar echoes and are instrumental to converting the
The large echo with a downward deflection at 14.827 record’s time scale to a distance scale, as discussed in
seconds corresponds to the echo from the liquid level. the next section.
The small echo with an upward deflection at about The panel on the lower left side of the figure displays
6.2 seconds corresponds to the crossover point from the section of the acoustic record between 1.5 and 2.5
3H-inch to 2M-inch tubing. This section of the signal seconds after filtering the raw data with a high-pass
is amplified and expanded in the panel on the lower digital filter. The distinct down-kicks correspond to the
right, in which the vertical dashed line flags this point echoes from the tubing collars and are flagged with tick
at a depth of 4,017.43 feet. marks by the signal processing software.
Select Liquid Level Depth Determination Casing Pressure BHP Collars

Sec 0 2 4 6 9 10 12 14 16
C LL
TUBING
CROSS-OVER
100 mV
COLLARS FLUID
LEVEL
Explosion
0 1500 3000 4500 6000 7500 10000
JTS/sec 19.5312 Acoustic Vel. 1218.52 ft/s 14.827 sec 293.686 Jts 9161.24 ft
Filter Type: High Pass
<<< [ 1.5 to 2.5 (Sec) ] >>> • Coarse
Fine
|<--->|
->|-|<-
<--
--> Show Depth Reference Line
Apply Automatic Collar Count 4017.43 ft <-- -->
13.5 mV Scale: Up Down Rst Analysis Method: Automatic ? < Pg Up Pg Dwn >
Figure 3.10 Echoes from tubing couplings, a tubing crossover, and the fluid level

The objective of the digital filter processing is to The relation most frequently used is the isentropic (or
highlight these echoes from the tubing collar so that the adiabatic) gas law:
user can verify that the software has correctly identified P ρ γ
–– = –– Eq. 3.18
these signals and determined their frequency, which in P0 ρ0
this case is displayed as 19.531 jts/s.
where:
The distance scale (0 to 10,000 feet) presented on
the lower portion of the top panel is constructed by P0 and ρ0 = the undisturbed values of pressure and density
γ = the ratio of the gas-specific heats
calculating the average acoustic velocity based on the
frequency of the echoes from the tubing couplings, Applying this relation to the propagation of a pres-
considering the time interval from the beginning until sure disturbance p = P – P0 (noting that the disturbance
the position of the vertical dashed line labeled “C” at could be an increase or a decrease in pressure) that causes
about 9.6 seconds. The “C” marker indicates how far a change in density δρ = ρ – ρ0, the acoustic velocity
into the record the echoes from the tubing collars can be can be expressed as:
identified by the software. The closer this marker is to γp0
the liquid level echo, the more reliable the determination c0 = ––– Eq. 3.19
ρ0
of the liquid level depth will be.
where c0 is known as the “small signal” sound speed
that depends on the pressure and the density of the gas,
ACOUSTIC VELOCITY IN GASES
which in turn depends also on the temperature.
The relationship between the acoustic velocity and the Using the ideal gas law in terms of the density:
gas properties for an ideal gas is given in equation 3.3.
This relation is a good working relation for calculating
P = RρT Eq. 3.20
the speed of sound in gases at low pressures. Since The acoustic velocity is expressed as:
acoustic measurements are generally performed in oil
and gas wells over pressure ranges that can span several
c0 = γRT0 Eq. 3.21
thousand psi, it is necessary to apply a more rigorous Equation 3.21 shows how the acoustic velocity is de-
method to determine acoustic velocity. pendent on the square root of the temperature.
The one-dimensional wave equation for gases was The relation between gas density, pressure, and
derived by applying the laws of conservation of momen- temperature is very important for the analysis of acoustic
tum and conservation of mass in a cylindrical control records obtained from echometric surveys, from the
volume and considering that losses are negligible so standpoint that computing the acoustic velocity for the
that, in the process, entropy can be considered constant. gas at the measurement conditions allows calculation
The relationship between thermodynamic variables and of the physical distance to a specific echo recorded in
fluid properties is known as an equation of state. A well- time. The distance from the sound source to the feature
known example is the perfect gas law: causing the echo—the liquid level, for example—is
simply one-half of the product of the round trip travel
P × V = nRT Eq. 3.17
time (RTTT) multiplied by the acoustic velocity.
where: On the other hand, when the distance to the spe-
P = absolute pressure cific feature (reflector) causing the echo is known, then
V = volume measuring the round trip travel time to the echo allows
n = number of moles computing the acoustic velocity of the gas by dividing
R = the gas constant the distance to the feature by one-half the measured
T = absolute temperature round trip travel time. For instance, when the distance is
An equation of state that is used in acoustics is known to be the depth to the top of the liner, the acoustic
the relation between pressure, density, and entropy. velocity of the gas between the sound source and the

liner can be calculated by dividing the distance to the

reflector by one-half the round trip travel time to the echo.
The computed sound speed then can be used to 4,000 GAS
RESERVOIRS
estimate the in-situ density of the gas using the relation: RETROGRADE
CONDENSATE
P 3,500 RESERVOIRS
ρ0 = γ — 20 Eq. 3.22
c0
3,000 T CRITICAL
Pi
This requires measuring the pressure and estimating POINT DEW
POINT
PRESSURE, psia
BUBBLE
the value of γ, which for most hydrocarbon gases varies POINT
2,500
between 1.3 and 1.4. Pwf Twf
%
In an oil and gas well, knowing the density of the gas 2,000 80
CRICONDENTHERM
POINT
Ptf Ttf
present in the wellbore is necessary to compute the variation % %
40 20
of pressure as a function of depth from a measurement of 1,500
L UM
E
10
%
VO
the surface pressure and a known temperature distribution.
UID
LIQ
The estimate of the pressure at the bottom of the gas column 1,000
is then used to compute the producing bottomhole pressure 5%
0%
500
(PBHP) by adding the hydrostatic pressure due to the fluids 0 50 100 150 200 250 300 350
that are present between the depth of the liquid level and TEMPERATURE, °F
the formation. Calculation of the pressure distribution in
the wellbore is discussed in chapter 6. Figure 3.11 Schematic PVT diagram for hydrocarbon
fluids
Effect of Gas Composition, Pressure, and
Temperature on Acoustic Velocity
The gases found in the wellbore of oil and gas wells are The figure shows a two-phase region, bound by red
mixtures of numerous hydrocarbons and other gases such and green lines where both liquid and gas coexist. In this
as nitrogen, carbon dioxide (CO2), or hydrogen sulfide region are tie lines of constant liquid percentage that con-
(H2S). The behavior of this mixture as a function of pres- verge at the critical point, the pressure and temperature
sure and temperature is quite complex. Depending on the at which the gas and liquid are indistinguishable from
pressure and temperature, a given mixture could exist as one another. To the left of the critical point and above
a gas phase only, as a liquid phase only, or as both a gas the red line, the fluid exists as a liquid. The outside edge
phase and a liquid phase. The composition and relative of the region in this area is called the bubble point line,
amounts of each component in the liquid or gas phases the pressure level at which gas at a given temperature
can only be described accurately by taking a sample of begins to evolve from the liquid and continues to do
the gas from the well and performing detailed chemical so as the pressure is decreased below the bubble point
analyses and volumetric tests that are commonly known pressure. To the right of the critical point and outside
as PVT (pressure, volume, and temperature) tests. The the region, the fluid exists as gas. The outside line of
results of these laboratory tests include the composition the region in this area is called the dew point. At a given
of the overall mixture as a tabulation of the concentra- temperature, a portion of the gas condenses and forms
tion of each component and tables or charts that describe a liquid as the pressure is reduced below the dew point
the properties (such as density) of the gas phase and the pressure. This schematic diagram also shows a retrograde
liquid phase and their relative amounts as a function of effect common to hydrocarbon gases where, within the
pressure and temperature. two-phase region as pressure continues to be reduced,
A schematic phase diagram is shown in figure 3.11 some of the liquid vaporizes back into the gas.
as a plot of pressure versus temperature defining certain The large majority of acoustic fluid level records in
regions where gas or liquid can exist. oil and gas wells are acquired at pressure and temperature

conditions where both gas and liquid phases are present. Z = gas compressibility factor
The phase diagram provides information about the rela- γ = ratio of the specific heats Cp/Cv
tive amounts of gas and liquid present in the well but R = gas constant (1,545 ft-lbf/[lb-mole-°R]/
does not provide information regarding the composition molecular weight)
of the gas at a specific pressure and temperature condi- T = gas temperature (°R)
tion. This information has to be obtained separately, p = gas pressure (psia)
based on the knowledge of the composition of the gas The difficulty in applying this relation in practice
and liquid mixture at standard conditions, as well as a is that, in the majority of the cases related to acoustic
combination of laboratory tests and computations based surveys in oil and gas wells, it is very unusual to know
on an assumption of equilibrium between the phases. the composition of the wellbore gas from a detailed
As discussed in chapter 5, the ability to determine laboratory analysis. At best, it is possible to obtain an
the distance to echoes observed in acoustic records estimate of the gas’s specific gravity from a sample taken
depends on determining the acoustic velocity in the gas at the wellhead. Accordingly, additional correlations
present in the wellbore. When a direct calculation of are required that express the PVT properties in terms
the average acoustic velocity in the gas of the wellbore of gas gravity rather than gas composition.
is not possible because of the absence of echoes from The following figures, based on the results pre-
a reflector at a known depth or the presence of echoes sented in reference 6, illustrate the variation of acoustic
from the tubing couplings, the velocity must be com- velocity in two hydrocarbon gases with different com-
puted from gas properties and the measured pressure positions. The authors correlated the pseudo-critical
and temperature. temperature and pseudo-critical pressure with gas
The acoustic velocity of mixtures of gases depends gravity, using the data from the Phillips Natural Gas
on the composition of the gases as well as on temperature and Gasoline Department. Then they related the specific
and pressure. At the elevated pressure and temperature heat of gas at constant pressure and constant volume
conditions found in most wells, the relation between to the PVT behavior of the gas, as well as calculated
pressure, temperature, and volume does not follow the the acoustic velocities in natural gases with gas grav-
ideal gas laws discussed earlier. The real behavior of ity of 0.6, 0.8, 1.0, and 1.2. They concluded that the
the gas must be considered by including the effects of BWR equation of state can be used to calculate heat
the gas compressibility factor, and the equation of state capacity ratios and in turn the speed of sound with an
must include the ability to account for the presence average error of 0.71%.
of multiple hydrocarbon components and impurities. The information in figure 3.12 represents a gas that
Early solutions that applied to low and medium pressure has a specific gravity of 0.6 and, is mostly composed of
conditions6 were expanded to include a broad range of light hydrocarbons (methane and ethane), and it applies
pressures by introducing correlations for critical pres- to coal bed methane wells.
sure and temperature7 and the Benedict-Webb-Rubin At a pressure of 400 pounds per square inch absolute
(BWR) equation of state. For real gases, the dependence (psia) and a temperature of 108°F, the acoustic velocity
of the acoustic velocity on pressure and temperature is in figure 3.12 is estimated to be about 1,390 ft/s. Notice
expressed as: the significant effect of changing the temperature at
gcZγRT the same pressure, as compared with the lesser effect
cf = ––———— Eq. 3.23
p ∂z of varying pressure at a constant temperature, espe-
1–— —
z ∂p cially for pressures less than 1,000 psia. Increasing the
T
where: temperature to 215°F causes the acoustic velocity to
cf = speed of sound for a real gas increase to 1,535 ft/s at the same pressure of 400 psi,
gc = unit conversion constant (32.17 lb-mft/ while increasing the pressure to 800 psi only causes a
[lbf-sec2]) decrease in velocity to 1,370 ft/s.

The information in figure 3.13 represents a gas temperature conditions (400 psi and 107°F), the acoustic
mixture that has a specific gravity of 1.2 and is com- velocity for this heavier gas is much lower (785 ft/s) than
posed primarily of heavier hydrocarbons (such as for the gas in figure 3.12. The acoustic velocity is also
ethane, propane, and butane). At the same pressure and more sensitive to variations in pressure and temperature
PRESSURE, kg/cm2
0 20 40 60 80 100 120 140 160 180 200

2,000
600
ACOUSTIC VELOCITY
1,900 GAS GRAVITY = 0.6 580
(BASED ON AIR = 1.0)
560
1,800
TEMPERATURE
FAHRENHEIT CELSIUS 540
200º C
392º F
1,700 520
ACOUSTIC VELOCITY, meters/sec

161º C
321º F
500
1,600
250º F 121º C
480
215º F 101º C
1,500 460
144º F 62º C 440
1,400 108º F
42º C 420
73º F
23º C
55º F 400
1,300 13º C
37º F
19º F 3º C 380
-7º C
1,200
360
VELOCITY = 1,390 ft/sec
340
1,100
0 500 1,000 1,500 2,000 2,500 3,000
PRESSURE, psia
Figure 3.12 Acoustic velocity in a hydrocarbon gas with a specific gravity of 0.6 as a function of pressure and
temperature

compared with that of the gas with a specific gravity of • Coal bed methane wells
0.6. These two examples cover the ranges of acoustic
The accuracy of the acoustic velocity calculated from
velocities normally found in:
equations of state and PVT data has been established by
• Oilwells producing by artificial lift whose annular
comparison to laboratory measurements that have been
pressure is generally less than 600 psi and whose
the basis for development of improved correlations. In
annular gas gravity varies from 0.6 to 1.2
particular, the Gas Research Institute and the University of
• Low-pressure gas wells
Oklahoma present a highly accurate correlation for ideal
PRESSURE, kg/cm2
0 20 40 60 80 100 120 140 160 180 200

1,500
440
ACOUSTIC VELOCITY
1,400 GAS GRAVITY = 1.2
(BASED ON AIR = 1.0) 420
400
1,300
380
1,200 TEMPERATURE
ACOUSTIC VELOCITY, meters/sec

360
FAHRENHEIT CELSIUS
404
ºF
35 340
1,100 0º 207
ºC
32 F
293º F
268º F 17
7º 320
8 C
24 º F 16
2º
1,000 218º F 14 C
183º F 7º 300
8º 13 C
F 2º
16
11 C
1º 35º 00º F
7º 280
F
1
900 C
F
10
1
7º 7º
260
C
8
C
800 240
72
ºC
220
57
º C 42º C
700
VELOCITY = 785 ft/sec 200
600
0 500 1,000 1,500 2,000 2,500 3,000
PRESSURE, psia
Figure 3.13 Sonic velocity in a hydrocarbon gas with a specific gravity of 1.2

gas properties and a vapor phase equation of state that several days for a pressure buildup test or because of
are used to compute sound speed8. Results are compared mechanical problems), the gas is subjected to the action
to measurements on pure methane and simulated natural of gravity and has a tendency to stratify, with the lighter
gas mixtures. The uncertainty in the calculated values is components (such as methane) near the top of the well
approximately 0.1% for pressures lower than 1,500 psi and the heavier components (such as propane) at the bot-
and less than 1% for higher pressures. tom. The acoustic velocity of the gas at the top will be in
The National Institute of Standards and Technology the 1,200 to 1,400 ft/s range, while at the bottom it may
(NIST) has developed and made available to the public a be less than 900 to 1,000 ft/s, depending on the increase
set of software packages that allow very precise calcula- in temperature. The difference in velocity is even more
tion of sound speed and PVT properties for mixtures of significant when gases such as H2S or CO2 are also pres-
hydrocarbon and other gases9. Heriot-Watt University ent in the produced gas mixture.
recently published an extensive theoretical and experi-
mental study that culminated in the development of very ACOUSTIC SIGNAL AMPLITUDE
accurate correlations and software for determining sound REDUCTION
speed in reservoir fluids10. The acoustic survey record in figure 3.7 shows clearly
Knowledge of the acoustic velocity in the gas is how the amplitude of the echoes from the tubing col-
needed to convert the round trip travel time of an acoustic lars and from the liquid level decreases as a function of
pulse to the corresponding distance traversed by the sound time. This effect is the combination of various factors:
wave. The variation of acoustic velocity with pressure and • As the pulse reaches tubing couplings, a portion
temperature, presented in figures 3.12 and 3.13, illustrates of the energy is reflected so that the amplitude
the difficulty inherent in obtaining an accurate value of of the downward-propagating wave decreases
the distance from the echo travel time measurement. accordingly. The echo from the liquid level
In all oil and gas wells, the acoustic velocity of the then travels back to the surface and, as it travels
gas is slightly different at every point in the wellbore, upward through the same tubing couplings, it
since pressure, temperature, and gas composition are loses additional amplitude.
slightly different at every point in the wellbore. This • Propagation of sound in a real gas is not lossless,
problem was recognized by the early developers of liquid but is accompanied by absorption of sound due
level measurement systems1,11; it is currently addressed to viscosity, heat conduction, relaxation, and
by state-of-the-art technology2,12, as discussed in more other mechanisms. The longer the path traveled,
detail in chapter 5. the greater the reduction in amplitude caused by
The severity of this problem depends partly on well these loss mechanisms. Therefore, even when the
depth and partly on the gas flow rate. In a typical well tubing couplings are not present, the amplitude of
produced by artificial lift, oil, water, and gas are produced the returning echo will be decreased in proportion
by the formation and flow through the perforations at a to the distance traveled.
constant rate. The majority of the liquid flows through These combined effects on the amplitude are gener-
the tubing, while the majority of the gas flows through ally expressed in terms of an absorption coefficient that
the casing annulus to the surface. While flowing to the is dependent on the gas properties and conditions and
surface, the gas in the annulus is in constant motion, related to the frequency content of the propagating wave
causing mixing. Thus, the composition of the gas is fairly pulse. A detailed discussion of absorption is beyond the
uniform from the top to the bottom of the gas column, scope of this handbook (although an interested reader
and the acoustic velocity varies only slightly because of can review reference 5 for more information). In general
pressure and temperature changes. terms, though, it should be noted that:
On the other hand, in a well where the flow of gas • Absorption is inversely proportional to the density
is minimal or has stopped (such as a well shut in during of the gas and directly proportional to its viscosity.

• Absorption increases as the square of the fre- Figure 3.14 shows an example of the attenuation of
quency of a time-harmonic signal. an acoustic pulse as it travels in a gas at a pressure of
• Absorption causes the amplitude of an acoustic about 106 psi. On the left side, the graphs display the
signal to decay exponentially as a function of acoustic pulse echo after traveling a distance of 7,102
distance traveled. feet and the repeat echo that has traveled twice the
Amplitude Attenuation Absorption of High Frequencies
DISTANCE
TRAVELED
7,102 FT
DISTANCE
TRAVELED
14,204 FT
AMPLIFIED
VERTICAL
SCALE
DISTANCE
TRAVELED
21,306 FT
Figure 3.14 Pulse amplitude attenuation and spreading in wellbore gas at 106 psi

distance. The vertical scales of the graphs are identical. Figure 3.15 is a plot of the recorded amplitude of
The peak-to-peak amplitude of the pulse has decreased the repeated echoes showing an almost perfect cor-
from 69.2 to 6.61 mV. The second repeat echo (shown relation with an exponential absorption coefficient of
at the bottom right with an amplified vertical scale) has 2.85 × 10-4 nepers/ft.
a peak-to-peak amplitude of 1.21 mV after traveling an Since the absorption coefficient is inversely depen-
additional distance of 7,102 feet. The record was obtained dent on the gas density, as the pressure increases, the
in a well completed with coiled tubing; therefore, no corresponding increase in density will cause a reduc-
couplings are present. Thus, the loss mechanisms listed tion of the absorption coefficient. This is one reason
earlier are the only causes of the attenuation of the echoes. why in a well it is easier to obtain an acoustic record
The acoustic pulse was generated with a conventional that clearly shows a deep liquid level echo when the
gas gun, so it is not time-harmonic but contains a broad wellhead pressure is high (greater than 20 to 30 psi)
spectrum of frequencies. Thus, each component will at- than when the pressure is low (near atmospheric). In
tenuate differently: the high frequencies will attenuate wells operating with the gas flowing to the intake of
more than the low frequencies. As a result, as the pulse a compressor so that casinghead pressure is less than
travels in the gas, its width will increase and the pulse atmospheric, it may be necessary to shut in the compres-
will spread. The three graphs on the right of figure 3.14 sor and allow the casing pressure to increase in order
are plotted with identical horizontal (time) scales but with to obtain a recognizable echo from the liquid level. In
different vertical scales to visualize the increase of the some cases, it may even be necessary to pressurize the
pulse width as a function of the distance traveled. wellbore using nitrogen gas.
80
70
y = 473.918884e-0.000285x
60
R2 = 0.992661
50
AMPLITUDE, mV
40
30
20
10
0
0 5,000 10,000 15,000 20,000 25,000
DISTANCE, ft
Figure 3.15 Exponential amplitude decay of an acoustic pulse propagating inside coiled tubing

SUMMARY temperature. Within the wellbore, the acoustic velocity

of the gas will be different at every point since pressure,
The fundamental concepts presented in this chapter
temperature, and composition of the gas are varying as
should be the basis for an accurate analysis of acoustic
a function of depth.
records acquired in oil and gas wells. Sound pulses are
Knowing the pressure, temperature, and composi-
generated as pressure disturbances in the wellbore gas
tion of the gas, it is possible to estimate the acoustic
and propagate at a speed defined as the acoustic velocity.
velocity using correlations of the gas’ PVT properties
The wave propagating in a wellbore is considered to be
and an equation of state to describe the relation between
a plane wave propagating along the axis of the wellbore.
acoustic velocity and thermodynamic parameters.
A change of the medium where the wave is propagat-
Since information about gas composition from
ing from a low-density fluid (gas) to a high-density
a detailed gas analysis is not generally available, the
fluid (liquid or gaseous liquid) will cause the wave to
acoustic velocity of mixtures of hydrocarbon gases can
be reflected and return to the surface, creating an echo.
be computed using the gas’ specific gravity to represent
By convention, the microphone and electronics used
the gas composition.
in recording the pressure pulse will generate a graph
The known pressure and temperature of the gas and
that will deflect downward (below zero) when the pulse
the measured acoustic velocity allow the computation of
corresponds to an increase in pressure. The polarity will
its average gas gravity, using the correlations for PVT
be inverted, deflecting upward when the pressure pulse
properties and an equation of state.
corresponds to a decrease in pressure.
As the sound pulse travels in the wellbore, its ampli-
Echoes generated from the liquid level will have the
tude decreases because of absorption caused by viscosity,
same polarity (deflection) as the polarity of the pulse gen-
heat conduction, and other energy loss mechanisms.
erated at the surface. An echo will be generated when the
In deep, low-pressure wells, obtaining an accurate
propagating wave reaches a point where the cross-sectional
fluid level record with clearly recognizable echoes re-
area of the wellbore changes. When the propagating wave
quires generating a large amplitude signal at the surface
encounters a decrease in cross-sectional area, the polarity
by increasing the pressure or the volume of the gas gun
of the echo will be the same as the polarity of the incident
chamber.
wave. When the propagating wave encounters an increase
in area, the polarity of the echo will be the inverse of the
polarity of the incident wave. REFERENCES
Knowing the change in area at a point in the wellbore, 1. C. P. Walker, “Determination of Fluid Level in Oil
it is possible to estimate the amplitude of the reflected Wells by the Pressure-wave Echo Method,” Trans-
and transmitted pulses by computing the corresponding actions of AIME, 1937.
reflection and transmission coefficients. The greater the 2. J. N. McCoy, et al., “Method for Processing Echo
change in cross-sectional area is, the greater the ampli- Sounding Data Collected from Boreholes in the
tude of the reflected wave is. Total obstruction of the Earth,” U.S. Patent 5,200,894, April 1993.
wellbore by liquids or solids will generate the largest
3. J. N. McCoy, A. L. Podio, and K. Huddleston,
amplitude echo. The known distance to a discontinuity
“Acoustic Determination of Producing Bottom Hole
in wellbore area and the measured time of arrival of the
Pressure,” SPE Formation Evaluation, September
echo (RTTT) that it generates allow the calculation of
1988.
the average acoustic velocity of the gas present in the
wellbore between the surface and the discontinuity. 4. J. N. McCoy, et al., “Acoustic Static Bottom Hole
In other words, the acoustic velocity equals twice the Pressures,” SPE 1985 Production Operations Sym-
distance divided by the RTTT. posium, 1985.
Acoustic velocity in the gas present in the wellbore 5. D. T. Blackstock, Fundamentals of Physical Acous-
is a function of the gas composition, pressure, and tics (John Wiley & Sons, 2000).

6. R. L. Andsager and R. M. Knapp, “Acoustic Deter- Constituents,” National Institute of Standards and
mination of Liquid Levels in Gas Wells,” Journal Technology (NIST), 1993.
of Petroleum Technology, May 1967. 10. H. Tahani, “Determination of the Velocity of Sound
7. L. K. Thomas, et al., “Determination of Acoustic in Reservoir Fluids Using an Equation of State,”
Velocities for Natural Gas,” Journal of Petroleum PhD Dissertation, Heriot-Watt University, 2012.
Technology, July 1970. 11. J. J. Jakorsky, “Bottom Hole Measurements in
8. J. L. Savidge, et al., “Sound Speed of Natural Gas,” Pumping Wells,” Transactions of AIME, 1939.
SPE Gas Technology Symposium, June 1988. 12. K. Huddleston, et al., “Data Processing and Display
9. B. A. Younglove, et al., “Speed of Sound Data for Echo Sounding Data,” U.S. Patent 5,117,399,
and Related Models for Mixtures of Natural Gas May 1992.

Acoustic Fluid Level Equipment and Procedures 4-1
4
Acoustic Fluid Level Equipment and Procedures
In this chapter:
• Requirements for state-of-the-art equipment used to acquire and analyze acoustic records
• Background history of equipment and patents
• State-of-the-art equipment used in the field
• Recommended practices for using acoustic fluid level equipment
This chapter presents the specialized equipment and Low-frequency waves propagate with less at-
procedures necessary for acquiring acoustic records. The tenuation than high-frequency waves. Thus,
objective of this chapter is to outline the history, practi- the pulse should have a slow rise time and long
cal application, and complexities of generating a viable wavelength to obtain distinct echoes from deep
acoustic pulse, as well as building microphones that detect wells. The pulse should have a spectrum shifted
the pressure pulse and signal processing equipment that toward low frequencies (1 to 10 Hz).
records and displays acoustic signals. Also presented are • Clearly defined echoes from discontinuities of
best practices and recommended operating procedures cross-sectional area require a pulse of minimum
for installing the sound source, preparing the well, and duration in time with a short wavelength and fast
acquiring and recording an optimum, high-quality acoustic rise time. Thus, the pulse should have a spectrum
signal with minimal interference. with high-frequency content (20 to 80 Hz).
Thus, the designer is faced with the problem of

ACOUSTIC PULSE GENERATION AND creating a pulse generation system that satisfies both
SIGNAL ACQUISITION objectives, which is very difficult in practice. As a
The characteristics of the acoustic pulse used in echo- consequence, some systems emphasize low frequencies
metric surveys of oil and gas wells are described in to provide high-amplitude echoes from deep reflectors,
chapter 3. Acoustic pulses need sufficient amplitude and while other systems stress high frequencies to achieve
appropriate frequency content in order to generate clear better definition of echoes from shallow- and medium-
and distinct echoes from the fluid level and all other depth wellbore discontinuities.
cross-sectional area discontinuities in the wellbore over This problem is also addressed through signal
distances from a few hundred to several thousand feet. processing techniques (filtering and variable gain) ap-
Designing a pulse generation and recording system that plied to the received signal, either in real time or by
satisfies these requirements has to take into account the post-processing, to enhance the quality of the displayed
following two opposing characteristics of sound propa- record and thus facilitate the analysis.
gation and reflection that were discussed in chapter 3: The pressure of the gas in the well has a major impact
• Acoustic pulse attenuation increases as the on received signal quality since it affects the attenuation
square of the frequency content of the pulse. of the pulse, causing less attenuation in high-pressure

at Austin
wells, so that larger amplitude echoes from all discon- listing of the major developments related to oilfield ap-
tinuities and the fluid level are observed. plications, as recorded in the patent literature presented
Practical application of echometric systems in the oil- in the references at the end of this chapter. Readers can
field for acquiring and analyzing fluid level surveys also obtain the full copies of the patents at the United States
has to satisfy strict workflow and safety requirements: Patent and Trademark Office (USPTO) patent database.
• The instruments should be easily transported, • Lehr2, 1936
installed, operated, and uninstalled for a broad – Use of compressed gas and manually con-
range of pressure conditions from near vacuum trolled quick-opening valve to generate an
to several thousand psi. explosion pulse
• Tools have to function in contact with potentially – Valve to control the amplitude of pulses
hazardous and/or explosive gas mixtures. – Computation of acoustic velocity from gas
• The time required by the operator to perform properties
routine measurements should be as short as • Walker3, 1939
possible to allow frequent monitoring of a large – Measurement of the acoustic velocity by sam-
number of wells. pling gas in the wellbore and using a resonant
• The instruments should be rugged, require mini- tube
mum maintenance, and must operate reliably over – Determination of acoustic velocity from
long periods of time. downhole markers
– Enhancement of amplitude of collar echoes by
The development of echometric systems for oilfield
tuning the receiver tube to the collar echoes’
operations derives from systems for detecting obstruc- harmonics
tions in pneumatic mail (dispatch tube) systems, which – Oscillographic recording of transmitted pulse,
have been in use since the 1890s in major cities and received echoes, and timing marks
industrial complexes. Transmission of documents and
• Ritzman4, 1941
other small items over significant distances was ac-
– Generation of a sharp pulse with frequencies
complished by enclosing them in a cylindrical carrier above 40 Hz
cartridge that was pushed through a piping system by – Generation of pulses with a blasting cap com-
compressed air pulses to deliver them at the end of the mercially used for detonating explosives
line. Occasionally, the cartridges got stuck somewhere – Variable gain of recording to compensate for
in the piping and had to be retrieved. The location of a signal attenuation
stuck cartridge was determined by echo ranging, using a – Dynamic microphone
sound pulse generated by firing a blank cartridge pistol – High-pass filtering to enhance tubing collar
into the pipe at one end and recording the signal on a echoes
chart as a function of time1. The time elapsed between • Kremer5, 1946
the shot and the echo was used to calculate the distance – Dynamic microphone sealed from well gases
to the stuck cartridge, knowing that the sound propaga- – Static high-pressure microphone equalization
tion medium was the air in the pipe and that the acoustic system that allows detection of high-frequency
velocity in air approximates 1,100 ft/s. pulses
Oilfield applications that had been developed by • Wolf6, 1951
1930 required modification of this original system so – Generation of pulses by firing a blank cartridge
that it was more practical and worked with hydrocar- (10 gauge)
bon gases instead of air as the propagating medium. – Firing mechanism contained within a cylindri-
This spurred the development of numerous designs for cal breech block
pulse generation systems as well as signal recording – Dynamic microphone with static pressure
and analysis systems. The following is a summarized balancing

– Piezoelectric cylinder microphone – Explosion or implosion generation of the

– Calculation of wellbore pressure from the acoustic pulse
height of the fluid column, fluid density, and – Vibration noise-canceling, piezoelectric,
gas pressure pressure-compensated microphone
• Teichmann7, 1953 • Eicher17, 1985
– Broadband strain-gauge microphone – Piston-actuated valve operated by a second
– Pressure-balancing ports pilot valve
• Booth8, 1960 – Selectable explosion/implosion mode
– Generation of pulses by the vibration of a – Remotely operated gas gun
resilient diaphragm or piston • McCoy18, 1987
– Pulse frequency of less than 50 Hz – Poppet valve that isolates an integral gas
– Low-frequency response hot-wire microphone chamber
– Microphone located at acoustic anti-node of – Rotating control of the valve position
the chamber – Implosion pulse generation
– Variable gains switch for sensitivity control • Ahern19, 1988
during recording – Processor-controlled pulse generation and
• Towell9, 1961 analysis of received echoes
– Cartridge-firing device operating at high pres- – Charging of gas gun by means of a compressor
sure (1,500 psi or more) – Monitoring of well pressure during acoustic
– Safety catch to prevent premature firing measurement
• Clements10, 1963 – Digital filtering of acoustic signal
– Cartridge-firing device with a rim choke – Determination of acoustic velocity as a func-
– Choking of pulses to slow rise time and in- tion of depth
crease pulse length – Determination of acoustic velocity for each
• McCoy11, 1967 record
– Compact and quickly installed cartridge-firing • Barber20, 1989
acoustic pulse generation (10-gauge blank – Microprocessor-controlled pulse generation
cartridge) and acquisition of acoustic data
– Cylindrical piezoelectric microphone – Repetitive generation of acoustic pulses over
a period of time
• McCoy12, 1975
– Database of information from multiple wells
– Rapid-loading cartridge-firing (.45 caliber)
– Download of records to a powerful analysis
acoustic pulse generation
processor
– Rotatable sealing element with a firing chamber
• McCoy21, 1990
• Godbey13,14,15, 1982, 1983 – Measurement of variations of the fluid level
– Automatic repeated generation of acoustic over a given time period
pulses – Sharp filter for enhancing signals above 50 Hz
– Automatic timing between the generation of – Band-pass filters for enhancing signals in a
a pulse and the detection of an echo range of 15 to 25 Hz
– Generation of pulses by actuation of a solenoid – Muting of sonic echoes for a time slice to
valve to release chamber pressure isolate the liquid level echo
– Digital count that controls the pump to main- – Unattended acquisition of chart traces and
tain liquid level between desired levels collar counts
• McCoy16, 1983 • Nolen22, 1998
– Compact gas gun assembly with an integral – Quick venting of a small quantity of casing
chamber and a manually operated valve stem gas to the atmosphere

– Measurement of acoustic velocity in coiled – Correlation of received and transmitted signals

tubing filled with well gas to determine echo travel time
– Adjustment of the speed of sound for pressure – Filtering out of non-correlating signals
and temperature variation – Signal sweep of less than 5 to 1,000 Hz
– Calculation of the distance to the fluid level – Vibrating membrane with a permanent magnet
considering the variation of velocity with depth and an excitation coil
• Finnestad23, 1998 • Gibbs29, 2012
– Pilot-operated solenoid valve to permit rapid – “Green” method, by which fluid level in a well
and automatic actuation of the gas gun, despite space is determined without substantial venting
pressure differentials across the solenoid valve of gas to the atmosphere
– Automatic generation of explosion or implosion – Multiple valves to control pressure increase/
pulses decrease in chamber
– No moving O-rings – Actuation of valves to generate implosion/
• Burris24, 2000 explosion pulse
– Series of acoustic transmitters that converts the – Resonant frequency of coiled tubing from
signal from the output amplifier into a pressure Fourier Analysis (FFT) to determine acoustic
wave velocity
– Series of acoustic receivers that receives reflected • Hill30, 2012
acoustic signals from the acoustic transmitters – Placement of the gas emission port and a pres-
– Determination of distance directly or by har- sure wave receiving port within the tubing-
monics casing annulus
– Storing of signals for various liquid level condi- – Use of produced fluids from the well to generate
tions the acoustic signal
– Analysis of downhole conditions by compari- – Accelerometer as a pressure-wave measuring
son of recent data to stored data device
• Finnestad25, 2007 – Insertion of pulse generator and receiver into
– Digitization of acoustic signals at the sensor the wellbore
– Transmission of digital signals, wired or wire- • Burris31, 2013
lessly, to the analysis processor – Generation of a sonic event through the use of
• Guion26, 2008 a compressed fluid obtained from within the
– Automatic setting of the pressure in the acoustic well
gun chamber – Gas compression chamber with a double-acting
– Automatic implosion or explosion mode piston
– Chamber outlet designed to optimize the pulse – Explosion or implosion pulse
shape – Release of multiple pulses with each piston
– Shorter, sharper pulses compared to prior art cycle
• McCoy27, 2010 – Control of pulse shape, amplitude, and sequence
– Acoustic transmitter/receiver sensors in the by programmed actuation of valves
borehole to generate T (tube) waves
– Oil/water/gas interface detection This extensive list shows the great level of effort that
– Fluid properties computed from velocity has been devoted to developing an acoustic fluid level
– Wellbore pressures computed from surface system that provides accurate information in all types of
readings wells—from shallow to depths exceeding 10,000 feet—
• Zamow28, 2011 and for variable conditions of pressure, temperature, and
– Pulse with predetermined time-variable fre- fluid composition. The following section summarizes
quency the principal characteristics of state-of-the-art systems.

PULSE GENERATION nitrogen gas at a pressure that is greater than the pressure
in the wellbore in order to generate a positive pressure
The numerous methods and tools listed in the previous
pulse (increase). Alternately, the gun vents a small volume
section can be grouped into two main categories based
of the gas from the wellbore into the gas gun chamber in
on the primary application of the system:
order to generate a negative pressure pulse (decrease).
• Category 1: Tools for monitoring the well per-
The first method is defined as the explosion mode, and
formance at a specific point in time for routine
the second method is defined as the implosion mode.
surveillance of well operation
A typical manually operated gas gun includes a volume
• Category 2: Tools for monitoring the performance
chamber, a quick-opening valve, and a microphone,
of the well over an extended period of time, with
as shown in figure 4.1. The indicated annular pressure
the objective of monitoring well performance
amplitude change of ±5 psi was measured using a sepa-
changes
rate pressure transducer connected to the wellbore and
Category 1 tools are typically manually operated, located diametrically opposite to the pulse generator.
while category 2 tools are remotely actuated via solenoid The amplitude of the pressure pulse that is created
valves, which are in turn actuated by a computer proces- is directly proportional to the volume of the chamber and
sor according to a predetermined schedule. the pressure differential between the wellbore and the
gas gun chamber. The characteristics of the pulse and its
Manual Pulse Generation frequency content are a function of the rapidity of the
The majority of modern fluid level instruments use a gas pressure discharge and the geometry of the connection
gun with either a source of compressed carbon dioxide or between the gas gun and the wellbore.
MICROPHONE Pressure
100 psi increases 5 psi
during a short
time; then wave
GAS propagates in
GUN tube.
300 psi
QUICK
OPENING Pressure
VALVE MICROPHONE decreases 5 psi
300 psi during a short
EXPLOSION—DISCHARGE GAS INTO WELL time; then wave
propagates
GAS in tube.
GUN
100 psi
QUICK
OPENING
VALVE
IMPLOSION—REMOVE GAS FROM WELL
Figure 4.1 Pulse generation using a manually actuated gas gun with a quick-opening valve

Figure 4.2 is a picture of a typical manually operated

gas gun capable of generating acoustic pulses in either Gas Gun Cutaway
implosion or explosion modes.
Figure 4.3 is a simplified schematic of the internal 400
500
600 PRESSURE
components of a gas gun. They consist of a chamber GAUGE
300 700
200 800
(to volume
of fixed volume that can be isolated from the wellbore BNC
100 900
1000 chamber)
CONNECTOR
by means of a manually actuated plunger gas valve COCKING
ARM
that seals by means of an O-ring, the passageway that TRIGGER
communicates the chamber with the wellbore. The

gas-valve stem is seated manually by lifting the cock- CASING
PRESSURE
ing arm, which closes the chamber so that its pressure QUICK
CONNECTOR
can be increased either by filling it with gas (generally
nitrogen or carbon dioxide) or can be decreased to a CASING
BLEED
GAS VALVE
(shown open) VOLUME
VOLUME CHAMBER CHAMBER
PRESSURE GAUGE
LIFT TO
CLOSE
MICROPHONE GAS VALVE
CONNECTOR DISCHARGE MICROPHONE
Figure 4.3 Simplified schematic of the gas gun in

PULL TO figure 4.2
OPEN
GAS VALVE
QUICK pressure lower than the well pressure using the filler/
CONNECT
GUN FILLER OR PRESSURE bleed valve. The gas-valve stem is kept in the closed
AND BLEED TRANSDUCER position by a spring-loaded trigger-pin.
ADAPTER
CASING The outlet gas-valve will open rapidly when the
BLEED
trigger is pulled. This generates a pressure pulse. If the
pressure is greater in the volume chamber than in the
2-IN. THREADS
11½ V
casing annulus, a compression pulse (explosion) is gener-
ated. If the pressure is greater in the casing annulus than
in the volume chamber, a rarefaction pulse (implosion)
is created. A differential pressure must exist between
the volume chamber and the casing annulus to actuate
the valve and generate a pulse.
The microphone is a twin-disc pressure-sensitive
device that is vibration-canceling. It has a near-flat
MICROPHONE
AND MICROPHONE
frequency response from less than 1 to 100 Hz, so it is
PROTECTOR capable of detecting low-frequency echoes from deep
liquid levels and high-frequency echoes from tubing
Figure 4.2 Typical manually operated gas gun collars and other wellbore discontinuities.

Explosion versus Implosion Pulses causing environmental and health concerns, or must be
Whenever a significant pressure (60 to 100 psi for returned to the wellbore or injected into the flow line
wells in the 5,000- to 7,000-foot depth) is present at by recompressing it. Some of the latest technologies
the wellhead, using the wellbore pressure to generate a have tried to address this problem with limited success.
pulse by imploding the well gas into the gun chamber A question is often raised regarding the possible dif-
has the advantage of eliminating the need for an exter- ference in the quality or accuracy of the acoustic records
nal source of compressed gas, generally provided in obtained by each pulse generation method. Figure 4.4
the form of a cylinder of CO2 or N2. The disadvantage shows acoustic records obtained in quick succession
is that the well’s gas, present in the chamber after fir- with the same compact gas gun, using about the same
ing the pulse, either must be vented to the atmosphere, differential pressure in explosion and implosion modes.
Sec 0 1 2 3 4 5
316.2 mV
LIQUID LEVEL
Indicator @ 2.505 sec Apply Low Pass Filter Move Indicator < --- Left Right --->
Sec 0 1 2 3 4 5
316.2 mV
PERFORATIONS
Figure 4.4 Comparison of records from explosion and implosion pulses acquired in the same wellbore. Top:
Explosion pulse acquired with a 195-psi differential (400 psi in the chamber with 205 psi in the wellbore). Bottom:
Implosion pulse acquired with a 205-psi differential (0 psi in the chamber with 205 psi in the wellbore)

Since the output of the microphone is either positive

or negative depending whether the ambient pressure
increases or decreases, the scale of the vertical axis of the
implosion display is inverted so that the polarity of the
echoes are consistent with the long-established industry
convention: A down-kick represents an obstruction and
an up-kick represents an enlargement of the wellbore,
as discussed in chapter 3.
The round trip travel time to the large-amplitude echo
(indicated by the dashed vertical marker) is measured to
be essentially the same (2.505 and 2.503 seconds) for
each test. The sequence in time and the polarity of the
recorded echoes are identical for both records: a small
up-kick at the perforations followed by a large down-
kick at the liquid level and a smaller up-kick for the echo
of the perforations mirrored at about the liquid level.
This sequence of echoes between 2.3 and 2.8 seconds is
repeated at 4.6 seconds, corresponding to the repeated
pulse reflected from the wellhead at the surface. Since
the vertical scales in figure 4.4 are identical, the slightly
smaller amplitude of the echoes generated by implosion
indicates that the sensitivity of the microphone may
not be symmetric and yields a larger output for a given Figure 4.5 Wireless remote control gas gun
pressure increase than for a similar pressure reduction.
Other small differences include the random noise
observed near the perforation’s echo in the explosion wirelessly. The firing mechanism is typically actuated via
pulse in figure 4.4, which does not appear in the implo- a solenoid valve powered by a long-lasting battery that
sion pulse. It was probably a random event that did not functions as a pilot valve in order to accommodate well
repeat when the implosion record was acquired. pressures from atmospheric to several thousand psi. The
While there may be minor differences in the acoustic pressure discharge valve is designed to operate quickly
records, the main variable of interest—the round trip (in fractions of a second) to generate an acoustic pulse
travel time—is not affected by the type of pulse that with significant high- and low-frequency components,
is generated. as discussed in chapter 3.
As shown in the simplified schematic in figure 4.6,
Automatic Pulse Generation the pressure chamber is initially charged to a pressure
To monitor the performance of the well or the variation greater than the pressure in the well (250 psi compared
of the liquid level position over an extended period of to 50 psi in this example), which also sets the internal
time, remotely actuated gas guns are used with program- firing valve to the closed position.
mable software to perform data acquisition and process- When the solenoid of the pilot valve is actuated
ing automatically. Most automatic gas guns operate in electrically to open, the firing valve retracts quickly
explosion mode by charging the chamber through direct and allows the chamber pressure to discharge into the
connection to a source of regulated compressed gas and wellbore, generating the acoustic pulse that is detected
discharging the pulse at specified time intervals. Figure by the microphone. The acoustic signal is monitored and
4.5 shows a picture of an automatic gas gun that operates recorded for sufficient time to detect a possible liquid

VOLUME VOLUME
CHAMBER CHAMBER
250 psi 50 psi
PRESSURE 12 VOLT PRESSURE

RELIEF VENT RELIEF VENT
FILLER
CONNECTOR
FILLER
CASING CASING
BLEED BLEED
50 psi 50 psi
Figure 4.6 Remotely controlled gas gun operation. Left: Gas gun ready to fire. Right: Gas gun after firing by
actuation of the solenoid valve
level echo from the deepest point in the well. After this In particular, fluid level surveys in flowing gas wells
elapsed time, the firing valve is closed, and means are are performed using implosion methods with gas guns
provided to automatically recharge the pressure chamber, capable of working pressures in the range of 5,000 to
so that a new acoustic record can be acquired accord- 15,000 psi. These applications are discussed in detail
ing to the preselected schedule. Generally, a separate in chapter 9.
pressure sensor is included with the gas gun to monitor
and record the ambient pressure in the wellbore. This Other Pulse Generation Methods
second sensor has two purposes:
Several variations of the basic manual and automatic
• To ensure that the gas gun chamber is pressurized
gas gun systems have been developed, as illustrated in
to a level greater than the well pressure
the summary of relevant patents. Most efforts have been
• To record and monitor well pressure variations
focused on developing different pulse characteristics that
during the acquisition of the acoustic record with
facilitate identification of the liquid level echo as well
the objective of determining wellbore fluid flow
as of the echoes from tubing collars and other wellbore
and fluid distribution to be used in subsequent
discontinuities.
well performance calculations
Acoustic guns that use blank 10-gauge black powder
The recommended practice is to use a remotely fired shotgun shells or blank .45 caliber pistol shells to induce
gas gun, which provides maximum flexibility and safety a positive pressure wave have been in use for a long time.
since the operator can be located some distance from These devices are now considered obsolete because of
the well when the gas pulse is discharged. the hazards associated with using blank ammunition and
Manual gas guns are typically used for fluid level the potential for initiating an explosion by firing into
surveys in wells that exhibit high wellhead pressures. the wellbore when air has inadvertently been admitted

during completion or workover operations, creating an removal of the sensors, and streamlines the operator’s
explosive mixture with the hydrocarbon gases. Some workflow.
companies have banned the use of blank-cartridge Figure 4.5 shows a wireless sensor used for fluid
acoustic guns after the rupture of the surface casing of level and pressure recording, which communicates with
a well in Alberta, Canada resulted in a blowout. The a radio base station connected to a portable computer.
failure was attributed to acquiring fluid level data with Multiple wireless sensors can be used simultaneously,
a .45-caliber blank after a workover was performed. so that the user can select which combination is most
appropriate for the type of well being tested. The results
ACOUSTIC RECORD ACQUISITION are presented on the laptop screen in a graphical format,
as shown in figure 4.4. Simultaneous recording of the
AND RECORDING
fluid level and wellhead pressure allows the visualiza-
Most commercially available fluid level instruments tion of the fluid distribution in the wellbore and the
include acquisition and analysis software executable on subsequent calculation of pressure distribution.
laptops that communicate with a signal conditioning and The acoustic trace is generally acquired at a sam-
digitizing front end controlled by microprocessors run- pling rate between 500 and 1,000 samples per second
ning a firmware application. Figure 4.7 illustrates typical and digitized with a 16- to 24-bit resolution to provide
systems for fluid level data acquisition and processing. the maximum dynamic range for recording the micro-
Advances in the miniaturization of integrated cir- phone signal.
cuits and in wireless communications, as well as the
development of long-lasting batteries, have allowed
the development of very small and rugged sensors Microphones
that can be quickly and reliably deployed to monitor The microphone must be able to withstand the high am-
the operation of the artificial lift system and analyze bient pressure and the relatively large pressure change
the performance of the well. The absence of cables and caused by the firing of the acoustic gun. At the same
connectors improves the accuracy and reliability of the time, it has to be able to detect minute pressure oscilla-
data, reduces significantly the time for installation and tions that characterize the echo from a deep liquid level.
Figure 4.7 Typical laptop-based fluid level data acquisition, processing, and analysis systems

Most modern systems have microphones that are

specifically designed for acoustic fluid level applications
and are constructed using various types of piezoelectric
elements that either are directly sensitive to the varia-
tions in pressure or are coupled to mechanical elements
that produce axial, radial, or bending strains in response
to small dynamic variations of the ambient pressure.
Other acoustic fluid level survey systems, however,
are configured for using off-the-shelf pressure transduc-
ers that have adequate sensitivity to detect the small-
amplitude acoustic signal within the static high-level
background pressure. The frequency response of the
microphone and associated electronics must preferably
be flat from near DC to several hundred hertz, keeping
in mind that most of the acoustic energy contained in
the received echoes is in the subsonic range from 1 to
20 Hz. In order to achieve good resolution in the travel
time measurement, however, it is also important to detect
the higher-frequency components.
Signal Recording and Processing

The advent of portable laptop computers revolutionized
the instrumentation used in the field for monitoring the
performance of oil and gas wells. Analog recording and
filtering of the microphone output, as shown in figure 4.8,
gave way to recording by analog to digital conversion.
The ability to digitize in real time and store the
acoustic signal in the computer memory provides unlim-
ited tools with respect to enhancing the user’s ability to
interpret correctly the series of echoes that are generated
from the liquid level and the anomalies that are present Figure 4.8 Strip chart recording of an acoustic trace
in the wellbore. The portability of digital systems allows
for interpretation and analysis that can be performed in
the field, which results in a more efficient assessment of SAFETY CONSIDERATIONS
any conditions that may require scheduling equipment High pressure, high temperature, and combustible or
repairs or well workover. noxious gases are associated with oil and gas produc-
Post-processing of the digitized acoustic record tion operations. Extreme care must be exercised when
permits using different filtering schemes, spectral working in a hydrocarbon production area and especially
analysis, trace overlaying, and folding, as illustrated in when connecting or installing measurement equipment
the examples presented in chapter 2. It must be noted to wells. Even small concentrations of hydrogen sulfide
that a large number of operators are currently still using can be lethal, and personnel must be adequately protected
the latest generation of strip chart recorders for routine from this environment and properly trained to operate in
fluid level monitoring because of their simplicity and the vicinity of wells that produce this gas in conjunction
proven reliability. with hydrocarbons.

Most commonly, certain concentrations of combus- Hazardous locations are classified by division ac-
tible gases in air constitute explosive mixtures that can cording to the probability that an explosive concentration
be ignited by electrical sparks or the development of of hazardous material may be present.
high temperatures. Areas of the oilfield where there is • Division 1 defines locations where there is a
a high probability of occurrence of explosive mixtures high probability that an explosive concentration
are defined as “hazardous locations” and have been is present during normal operation. To be classi-
classified by the American Petroleum Institute (API). fied Division 1, there has to be a minimum of 100
hours per year or 1% probability that an explosive
Hazardous Locations in the Oilfield material is present. In Europe, Division 1 is sub-
The American Petroleum Institute provides guidelines divided into Zone 0 and Zone 1. Zone 0 has the
for classifying oilfield locations at petroleum facilities highest probability of an explosive concentration
for the selection and installation of electrical equipment being present (a probability of greater than 10%).
following basic definitions given by the National Fire • Division 2 defines locations where there is a low
Prevention Association in the National Electrical Code probability that an explosive mixture is present
(NEC), which is part of the 1996 edition of NFPA 70. during normal operation (a probability of 10
Electrical and electronic instruments that are used hours per year, or 0.1% of the time).
in these areas must meet strict guidelines if they are to The purpose of the API RP (Recommended Practice)
be operated without the additional use of combustible 500 is to provide guidelines for classifying locations as
32
gas detectors and the filing of documents detailing the Division 1 or Division 2 at petroleum facilities for the
procedures to be followed in order to obtain what is selection and installation of electrical equipment follow-
known as a “hot permit.” API publishes papers that ing basic definitions given in the 1996 edition of NFPA
describe the conditions found in oil and gas facilities 70, the National Electrical Code (NEC).
that would cause areas to be classified as hazardous Section 10 of RP 500 defines the area classification
locations. for locations surrounding oil and gas drilling and work-
over; for production facilities on land; and for facilities on
API Classification of Drilling and marine fixed (bottom-supported) platforms where flam-
Production Facilities mable petroleum gas and volatile liquids are produced,
Hazardous materials such as gas or dust are grouped in processed, stored, transferred, or otherwise handled prior
classes by their generic type; Class I includes flammable to entering the transportation facilities.
gases or vapors. In oil and gas operations, the most com- Production facilities considered for classification by
mon hazardous materials correspond to Class I and are Section 10 of RP 500 that impact measurements using
grouped according to their level of explosion hazard. acoustic fluid level equipment include all producing oil
The following groups are listed in order of the most and gas wells, whether flowing or artificially lifted. For
to the least easily ignited: these wells, the area within a radius of 10 feet from the
• Group A: Acetylene (a compound with a ten- center of the wellhead is classified as a Class I hazardous
dency to form copper acetylides, which are easily location and defined either as Division 1 or Division 2,
ignited by friction) depending on the exact location. Figure 4.9 illustrates
• Group B: Hydrogen, hydrogen mixtures the situation with regard to a pumping well with the
• Group C: Ethylene, most ethers, some aldehydes casinghead located in a cellar.
• Group D: Alkanes (butane, ethane, methane, Electronic instruments that are not properly designed
octane, and propane), hydrocarbon mixtures to be used within the boundaries of the hazardous area must
(diesel oil, kerosene, petroleum, and gasoline), be kept at a distance greater than the boundaries of the
alcohols, ketones, esters, amines, alkenes, and hazardous zone. This is one reason why the conventional
benzenoids acoustic fluid level instruments are provided with cables

STUFFING BOX
1.5 m (5')
.5 m (18")
.5 m (18")
3 m (10')
3 m (10')
GRADE
CELLAR
DIVISION 1 DIVISION 2
Figure 4.9 Hazardous areas classification and Figure 4.10 Certified intrinsically safe fluid level
extension for a rod-pumping wellhead recorder and gas gun
that are longer than 15 feet. When fluid level measure- be specially designed and certified to ensure that it meets
ments are performed offshore or in enclosed wellhead the intrinsic safety requirement33.
spaces, such as in Alaska’s North Slope, the operator The intrinsic safety method has the advantage of
must follow special procedures and is required to obtain being relatively simple to implement; it does not require
hot work permits. maintenance and each installation is relatively inexpen-
On the other hand, an instrument that is approved for sive. It requires that the instrument be designed to meet
operation in the hazardous area can be located in the vicin- the specifications and be tested by an authorized and
ity of the wellhead, and its use does not require filing for recognized agency before it can be certified as meeting the
special permits. In the United States, such instruments must safety regulations. Testing and certification is an expensive
be certified as intrinsically safe by an approved certifica- and time-consuming process, but once certification has
tion agency such as the Underwriters Laboratory (UL) been obtained, the manufacturer has only to build the
or Factory Mutual Research Corporation (FM, or FMRC). equipment as specified by the certifying agency with the
The official definition of intrinsic safety is as follows: appropriate markings affixed to identify its level of safety.
Intrinsically safe equipment and wiring shall not be ca-
pable of releasing sufficient electrical or thermal energy
under normal or abnormal conditions to cause ignition of RECOMMENDED OPERATING PROCEDURES
a specific atmospheric mixture in its most easily ignited Accurate interpretation of an acoustic fluid level record
concentration33. One such instrument shown in figure requires obtaining a record that has an optimum signal-
4.10 illustrates how the electronics, processor, and power to-noise ratio and is free of obscuring random events.
supply are self-contained in a specially designed sealed The following are some practical recommendations to
carrier. In addition, the gas gun for this system must also obtain such a record.

Installation of Sound Source mask signals of interest. Such is the case in the record
Ideally, the pulse should be generated within the well- in figure 4.12, which was obtained in a well where the
bore itself, as suggested by some of the latest patents. casinghead valve is at the bottom of a cellar while the
This may be practical when the sound generator is gas gun was located above ground level at a distance
permanently installed on the well for continuous or of about 12 feet.
long-term monitoring. When measurements are taken This resonance between time zero and 2.5 seconds
on a routine basis on multiple wells, it is necessary to does not interfere with the detection of the liquid level
install and remove the equipment with a minimum of echo (at 2.89 seconds) but obscures the echoes from the
effort and time. tubing collars, casting a doubt on the accuracy of the
In these cases, the ideal installation is one where acoustic velocity determination.
the pulse generator is connected as close to the wellbore
as possible with a fully open connection of uniform Well Preparation and Information
diameter, as shown in figure 4.11. Best results are obtained when the piping of the well
When this is not possible and a significant dis- is in good operating condition, free of corrosion and
tance (10 or more feet) exists between the sound pulse deposits that may prevent free communication between
generator and the wellbore, it is possible to create a the pulse generator and microphone to the interior of
resonant condition within the line between the well and the well. Connections and valves should be inspected
the gas gun that can persist in the acoustic record and and overhauled if necessary.
BEST SHOT DISTANCE FROM

MICROPHONE TO
CASING MUST BE
LESS THAN 5 FEET.
Figure 4.11 Connections to a well

Figure 4.12 Record showing decaying acoustic resonance in connection piping
A clear understanding of how the well is completed • Check the threads on the wellhead valve for
and the details of the wellbore are necessary to ana- corrosion, wear, or damage.
lyze correctly the acoustic record and obtain reliable • For 2” NPT (National Pipe Thread), make at least
calculated values. Good practice requires that the well 4H turns for safe connection to the wellhead.
information provided to the acquisition and analysis • Close off other connections from the wellbore
software be updated as needed. to the flow lines.
Acoustic fluid level measurements are especially dif-
• Adjust the desired pressure differential between
ficult in those wells where the wellhead pressure is at or
the gas gun volume chamber and the wellbore.
below atmospheric pressure because of the rapid decay
• Make sure the valve between the gas gun and the
of the amplitude of sound propagating in a rarified gas.
wellbore is open before a shot is fired.
In some cases, it may be necessary to inject nitrogen to
increase the pressure sufficiently to be able to clearly
The differential pressure between the gas gun cham-
detect the fluid level echo.
ber and the wellbore in conjunction with the volume of
the chamber controls the magnitude of the energy that
Acquisition and Recording
is transmitted downhole. Consequently, the amplitude
For safety and optimum quality of the acoustic record, of the echoes that are generated is directly related to the
the following recommendations should be followed: pressure difference used for pulse generation. Figure
• Know the pressure on the well at the surface 4.13 shows an example of acoustic records obtained in
before attaching the gas gun to the well. the same well with the same gas gun, first operated with
• Be alert that equipment ratings exceed well a differential pressure of 600 psi (top trace) and then
pressure. with a differential of 200 psi (bottom trace).

Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
316.2 mV
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
316.2 mV
Figure 4.13 Effect of chamber pressure on the amplitude of the liquid level echo. The top trace resulted from a 600-
psi differential; the bottom trace resulted from a 200-psi differential.
In this well, the liquid level is relatively high, so that, when acquiring acoustic records in wells with a
multiple repeats of the echo observed at about 3 sec- high liquid level that have the first echo at 1 second
onds are also recorded. Visual inspection shows that, or less, it is often convenient to reduce the differential
in this case, the peak-to-peak amplitude of the echoes pressure to a minimum. This reduces the probability of
is increased about three times by tripling the pressure overloading the microphone and electronic circuitry
differential. It should also be noted that the amplitude- and obtains a clearer definition of the early part of the
decaying high-frequency signal recorded between the acoustic record.
shot and the first echo also is magnified by increasing
the chamber pressure. This is an indication that such Random Acoustic Signals
a signal is being generated by the pulse and probably Although microphones can be built to minimize their
corresponds to a resonance of the piping. sensitivity to vibrations, the acoustic record may con-
The best level of differential pressure to be used in tain signals that are random or do not correspond to the
a particular well should be explored by trial and error transmission of the acoustic pulse. This type of signal
by experimenting with different values of chamber pres- will not appear at the same time if the acquisition of the
sure. The objective would be to determine the minimum pulse is repeated in succession. Figure 4.14 shows two
chamber pressure that results in a clear definition of the acoustic traces obtained a few minutes apart. The signals
liquid level echo as well as clear echoes from tubing col- highlighted at about 8.5 and 16.5 seconds in the upper
lars or other wellbore discontinuities. It must be noted trace are not present in the second trace. On the other

Sec 6 7 8 9 10 11 12 13 14 15 16 17
20
0
mV
-20
-40
SIGNALS NOT IN SIGNALS NOT IN
SECOND RECORD SECOND RECORD
-60
ft 0 4,000 5,000 6,000 7,000 8,000 9,000
Figure 4.14 Records showing random signals
hand, the echoes generated by diameter discontinuities of the acoustic pulse. Therefore, if a second record is
in the well (tubing anchor, top of liner, and liquid level) acquired, any signal generated by the pumping action
appear in both records and at the same times. will be shifted in time relative to the start of the shot.
Multiple acoustic records should be acquired every The amplitude and form of the noise will be fairly con-
time that signals are recorded and do not seem to cor- stant, since it is generated in the same manner every
respond to the known characteristics of the wellbore. pump stroke.
Only those echoes that correspond to actual features in In addition, when the record is sufficiently long
the well will be observed in all the records. in relation to the pumping speed, the time between two
pumping noise signals will approximate the pumping
Pumping-Related Noise period, as seen in figure 4.15 in a well where the pump-
Acoustic noise generated by rod pumping action is not ing speed was 6.38 strokes per minute, corresponding
random and is not synchronized with the generation to a pumping period of 9.404 seconds.
Figure 4.15 Noise generated by pumping action

Gas Flow Noise be observed, but all other echoes are masked by the
Acoustic noise that interferes with fluid level records high-frequency noise induced by the gas flow.
is typically related to gas flow through valves and fit- The lower trace shows that applying a proper
tings. This type of noise generally is characterized by low-pass digital filter makes it possible to observe the
higher-frequency content than that of the echoes from the echoes generated at the gas lift mandrels and the bottom
liquid level and other wellbore features. Consequently, packer. The acoustic record can thus be interpreted and
it should be possible to filter out the noise signal. Al- analyzed without having to close in the gas injection to
ternatively, the source of the noise should be identified the annulus of the gas lift well.
and eliminated, if possible.
Figure 4.16 shows a fluid level record that was ac- Calibration and Maintenance
quired in a gas lift well while gas was being injected at All instrumentation should be properly maintained and
the casinghead. The upper trace shows that the signal at checked periodically to ensure optimum performance.
time zero—corresponding to the firing of the shot—can The microphone is probably the most sensitive part of the
Sec 0 2 4 6 8 10 12 14 16 18
31.6 mV
Sec 0 2 4 6 8 10 12 14 16 18
31.6 mV
Figure 4.16 Acoustic record acquired in a well produced by gas lift

system with regard to mistreatment and impact loading. The pulse generator should be connected as close to
The gas gun should be properly handled. the wellbore as possible with a fully open connection
In wired systems, the cables and connectors of uniform diameter.
must be kept clean and dry at all times. The microphone The best level of differential pressure to create
cable in particular should be checked to ensure proper an acoustic pulse in a particular well should be explored
continuity and infinite resistance between the center by trial and error by experimenting with different values
conductor and the outer shield. of chamber pressure. Whenever unusual or unexplained
signals are observed, it is recommended that the record-
SUMMARY ing be repeated in order to differentiate random events
from actual echoes from wellbore discontinuities.
This chapter describes some of the equipment for acquir-
A clear understanding of how the well is com-
ing acoustic fluid level records that have been used in
pleted and the details of the wellbore are necessary to
the oil and gas industry since the early 1930s and have
analyze the acoustic record correctly and obtain reliable
evolved into state-of-the-art digital systems with tools
calculated values. Initially, a differential pressure of 100
for analyzing deep and geometrically complex wells.
to 200 psi should be used and then adjusted up or down
Generating an acoustic pulse of sufficient ampli-
based on the quality of the recorded signal.
tude and appropriate frequency content is the primary
requirement to obtain clear echoes from downhole in
deep wells. A pulse with a mix of low-frequency (1 to REFERENCES
10 Hz) and high-frequency (20 to 80 Hz) components 1. B. C., Batcheller, “Apparatus for Locating Obstruc-
would satisfy this condition, but is difficult to achieve in tions in Tubes,” U.S. Patent 602,422, April 1898.
practice. This has spurred the development of numerous
2. P. E. Lehr, “Method and Apparatus for Measuring
patents and alternative signal-processing techniques.
Well Depths,” U.S. Patent 2,047,974, July 1936.
The majority of the pulse generation equipment
uses an external gas source to create a rapid increase in 3. C. P., Walker, “Means For Measuring the Location
pressure that propagates within the wellbore, which is of Obstructions in Wells,” U.S. Patent 2,156,519,
known as an explosion pulse. In wells where the well- May 1939.
head pressure is relatively high (greater than 60 to 100 4. O. F. Ritzman, “Method of and Apparatus for Mea-
psi), the acoustic pulse can be generated by venting a suring Depths in Wells,” U.S. Patent 2,232,476,
small volume of gas from the well into a low pressure February 1941.
chamber, generating a rapid decrease in wellbore pressure 5. J. T. Kremer, “Well Sounding Microphone,” U.S.
known as an implosion pulse. Acoustic pulse generation Patent 2,403,535, July 1946.
can be performed manually or can be programmed for 6. A. Wolf, “Acoustical Well Sounder,” U.S. Patent
automatic acquisition according to a preset schedule. 2,560,911, July 1951.
Microphones for detection of the pressure pulse
7. C. F. Teichmann, “Microphone,” U.S. Patent
must operate in a high-pressure environment and be
2,648,828, August 1953.
insensitive to mechanical vibration while maintaining
8. S. E. Booth, “Measurement of Liquid Level in
very high acoustic sensitivity. Digital signal acquisition
Wells,” U.S. Patent 2,927,301, March 1960.
and recording provides the ability to reprocess the records
using various signal-processing techniques to optimize 9. B. H. Towell, “Well Sounding Gun,” U.S. Patent
the analysis, as discussed in chapter 5. 2,993,554, July 1961.
High pressure, high temperature, and combustible 10. R. J. Clements, “Means for Determining the Fluid
or noxious gases are associated with oil and gas wells. Level in a Well,” U.S. Patent 3,100,023, August 1963.
Extreme care must be exercised when connecting or 11. J. N. McCoy, “Echo Ranging Apparatus,” U.S.
installing measurement equipment to the wellhead. Patent 3,316,997, May 1967.

12. J. N. McCoy, “Wellhead Gun for Echo Ranging 25. D. Finnestad, “Apparatus, Method and System for
Apparatus,” U.S. Patent 3,915,256, October 1975. Digitally Transmitting Acoustic Pulse Gun Signals,”
13. J. K. Godbey, “Automatic Liquid Level Monitor,” U.S. Patent 7,277,358 B2, October 2007.
U.S. Patent 4,318,298, March 1982. 26. W. F. Guion, “Acoustic Generator for Distance
14. J. K. Godbey, “Automatic Liquid Level Controller,” Sounding with a New and Novel Portal Structure
U.S. Patent 4,318,674, March 1982. for the Efficient Firing of the Pressure Chamber,”
15. J. K. Godbey, “Automatic Liquid Level Monitor,” U.S. Patent 7,397,729 B2, July 2008.
U.S. Patent 4,391,135, July, 1983. 27. R. H. McCoy, “Using an Acoustic Ping and Sonic
16. J. N. McCoy, “Gas Gun Assembly,” U.S. Patent Velocity to Control an Artificial Lift Device,” U.S.
4,408,676, October 1983. Patent 7,784,538 B2, August 2010.
17. D. A. Eicher, “Gas Gun for Determining the Liquid 28. R. Zamow, “Method and Device for Recognizing
Level of a Well,” U.S. Patent 4,509,552, April 1985. the Depth Level in a Bore Hole,” EU Patent, EP
18. J. N. McCoy, “Echo Ranging Gun,” U.S. Patent 2169179 B1, November 2011.
4,637,463, January 1987. 29. S. G. Gibbs, “Systems and Methods for Measuring a
19. T. K. Ahern, “Method and Apparatus for Generating Fluid Level Within a Well,” U.S. Patent 8,261,819,
Data and Analyzing the Same to Determine Fluid September 2012.
Depth in a Well,” U.S. Patent 4,793,178, December 30. D. R. Hill, “Tool for Use in Well Monitoring,” U.S.
1988. Patent Application 2012/0325456 A1, December
20. D. D. Barber, “Automatic Liquid Level Recording 2012.
Device,” U.S. Patent 4,853,901, August 1989. 31. S. A. Burris, “Apparatus for Measuring of Fluid
21. J. N. McCoy, “Automatic Echo Meter,” U.S. Patent Levels and Pumping of the Same,” U.S. Patent
4,934,186, June 1990. 8,353,339, January 2013.
22. K. B. Nolen, “Determining Fluid Level in Wells 32. API Recommended Practice for Classification of
With Flow Induced Pressure Pulses,” U.S. Patent Locations for Electrical Installations at Petroleum
5,715,890, February 1998. Facilities Classified as Class I, Division 1 and Divi-
23. S. J. Finnestad, “Acoustic Pulse Gun Assembly,” sion 2, 2nd ed. (RP 500, 1997).
U.S. Patent 5,834,710, November 1998. 33. J. N. McCoy, W. Drake, and A. L. Podio, “Intrinsi-
24. S. A. Burris, “Well Pump Control Using Multiple cally Safe Well Analysis Instruments,” Proceedings
Sonic Detectors,” U.S. Patent 6,085,836, July 2000. of the SWPSC, 2007.

Methods of Determining Distance to the Liquid Level 5-1
5
Methods for Determining Distance to the Liquid Level
In this chapter:
• Best methods to obtain accurate results of fluid level depth for various well configurations
• Converting time record to distance
• Identifying collar echoes and echoes from wellbore discontinuities
• Determining acoustic velocity from gas properties
• Correlations and equations of state
• Calculating velocity from acoustic records in similar wells or past surveys
This chapter presents some recommendations for obtain- v = average acoustic velocity of gas between
ing the most accurate estimates of the distance to the the source and the reflector (ft/s)
liquid level from acoustic surveys. A number of different Δt = round trip travel time (seconds)
methods to obtain the distance using acoustic velocity As discussed in chapter 3, the acoustic velocity is a
are explained, from using collar echoes, tubing joints, or function of pressure, temperature, and composition of the
collar count to past acoustic surveys in the same region. gas. Consequently, it differs from well to well and also at
various points in a given well because of the increase of
CONVERTING ACOUSTIC PULSE TRAVEL pressure and temperature as a function of depth and the
TIME TO DISTANCE possible stratification of the gas column due to the differ-
ence in density of the various hydrocarbon components,
Echo signals are registered as the time required for the especially when gas is not being produced from the cas-
sound to travel from the pulse generator (gas gun) to the ing annulus.
wellbore cross-sectional area change (anomaly) and back For fluid level surveys in real wells, the following
to the microphone housed in the gas gun. This time is four methods are used to determine the average acoustic
known as the round trip travel time (RTTT) and, generally velocity:
speaking, is measured with an accuracy of ±1 millisecond. • Determination using identification and counting
The conversion of travel time to the actual distance of echoes from tubing or casing collars
from the microphone to the anomaly can be made using • Determination using the distance to a known
equation 5.1, if the average acoustic velocity for the gas anomaly in the wellbore
present in the wellbore between the gun and the anomaly • Calculation from gas gravity or composition
can be determined: • Estimation based on experience or previous
vΔt measurements
D = —– Eq. 5.1
2 All of these four methods involve varying degrees
where: of uncertainty, but generally, it is considered that the
D = distance between the sound source and the first two (the collar count and anomaly methods) yield
reflector (feet) the best estimates of the distance to the liquid level.

at Austin
Collar Count Method travel times, and collar echo identification becomes
When the well is completed with jointed pipe and when difficult and stops short of the liquid level echo. In this
externally upset couplings or internal gaps are present case, the distance to the liquid level has to be computed
at the joints, the acoustic record will include echoes that by extrapolation, using an average value of the acoustic
are generated at each tubing joint. These echoes will be velocity. This average velocity is calculated from the
spaced at a relatively constant frequency that will depend number of joints that are counted and the average joint
on both the average tubing joint length and the acoustic length for the particular tubing string. According to this
velocity in the gas. As the sound pulse propagates along method, the average acoustic velocity can be computed
the wellbore, some of the acoustic energy is reflected at using equation 5.2.
each tubing joint so that the amplitude of the propagating 2NcJL
v = ——– Eq. 5.2
pulse decreases as the distance traveled increases. The Δtc
echoes from the collars also decrease in amplitude and where:
eventually are indistinguishable from the acoustic noise
Nc = total number of pipe joints identified and
inherently present in the well. The attenuation of sound counted
is especially severe in low-pressure wells. In figure 5.1, JL = average joint length for pipe installed in
the vertical red line (labeled “C”) marks the point in the well (feet/joint)
the time record when this occurs (about 10.7 seconds). Δtc = time interval in acoustic record where
Ideally, if it were possible to identify the collar echoes echoes from pipe collars have been
all the way to the liquid level, the distance could be identified (seconds)
expressed as the number of tubing joints to the liquid v = average acoustic velocity in the section
of the wellbore where echoes from the
level, and the corresponding depth could be determined collars have been identified (ft/s)
very accurately from the tubing tally. Such determination
would not require knowledge of the average acoustic The average joint length should be computed from
velocity of the gas in the wellbore. the pipe tally of the specific well and should include
In general, as wells are drilled deeper and reservoir only full-size tubing joints (not pup joints). If the tally
pressures deplete, the wellbore pressure is reduced to low is not available, a representative average of 31.7 feet
values, the echo from the liquid level is recorded at longer per joint may be used for commonly used tubing joints.
Sec
0 5 10 15 20
1.5
0.5
mV
-0.5
1*
-1
-1.5
ft 0 2,000 4,000 6,000 8,000
Figure 5.1 Acoustic record showing decay of the amplitude of echoes from tubing collars. Echo (1) is caused by a
change in the casing’s internal diameter.

To maximize the number of tubing joints that can be Modern computer-based fluid level digital instru-
counted in the acoustic record, the amplitude of the echoes ments store in memory the sampled acoustic trace. This
from the pipe collars has to be optimized in relation to allows post-processing of the acoustic signal using a
the background noise. This can be achieved by increasing multitude of digital filters and other digital signal pro-
the amplitude and sharpness (high-frequency content) of cessing tools7,8. Spectrum analysis is generally used to
the transmitted pulse or by tuning the transmitted pulse determine a first estimate of the frequency band of the
frequency and the microphone or recording system to collar echoes. Then the acoustic trace is filtered using
the frequency of the echoes from the pipe joints1. The a band-pass filter centered at the prevailing frequency
development of electronic recording systems provided of the collar echoes9,10. This will improve the ratio of
the possibility of filtering the microphone signal to im- signal to noise and allow counting the collar echoes to
prove the signal-to-noise ratio of the received echoes. a certain depth. If the count is not satisfactory (close to
Systems were developed that included different filters the liquid level depth), the filtering process is repeated
that could be applied to the signal before displaying the using a slightly different filter bandwidth, and the col-
record2,3,4,5,6. Accordingly, a record could be acquired lar count is repeated and may yield a greater number of
with a low-pass filter to enhance the echo from the counted pipe joints.
liquid level, and a second record could be taken using a Figure 5.2 illustrates one method of collar analysis.
band-pass filter to highlight the echoes from the collars. The acoustic signal in the time interval from 0.5 to 1.5
Dual channel recording with dedicated filters simplified seconds is analyzed to detect the echoes from the pipe
and streamlined the workflow. collars (down-kicks highlighted by vertical dashed lines)
Figure 5.2 Digital processing of an acoustic record to detect echoes from tubing collars

and determine their prevailing frequency, which in this figure 5.3, where four sections of the acoustic signal
case corresponds to 14.86 Hz. This value is equivalent to from 0.5 to 1.5, 3 to 4, 7 to 8, and 10 to 11 seconds are
pipe joints per second within that time interval, so that a reproduced in detail.
direct relation can be established between a given round Several features should be noted:
trip travel time interval and the corresponding number • The peak-to-peak amplitude of the echoes de-
of pipe joints present in that section of the wellbore. creases significantly from the first interval (80
Care is required to identify the correct collar frequency, mV) to the last interval (0.3 mV).
especially when other downhole anomalies are present, • The number of collar echoes per second is slightly
such as cable splices or clamps used in ESP installations. different from section to section, as indicated on
These anomalies can generate additional echoes that the figure at the right margin of each interval.
can be mistaken for tubing collar echoes and result in • As time progresses, it becomes more difficult to
an erroneous collar frequency estimate, invalidating the distinguish the collar echoes from the ambient
calculation of acoustic velocity and liquid level depth. background noise.
As the average length of the pipe joints is known
By undertaking this process over the complete
(30.85 feet, in this example), the frequency of the collar
record, from time zero to the time of the echo from the
echoes can be converted to the average acoustic velocity
liquid level, it is possible to observe the variation of the
for that time interval using equation 5.2:
speed of sound (expressed in joints per second) in the
2 × 14.86 × 30.85 wellbore as a function of depth. The results are plotted
v = ———————– = 916.9
1 in figure 5.4, which shows a general decreasing trend
corresponding to a slowing down of the acoustic speed
This is the average acoustic velocity of the gas present
as a function of depth. The relatively low correlation
in the wellbore in the section between 230 and 670 feet.
coefficient of 0.78 is an indication of the variation of
As discussed earlier, because the temperature, pressure,
the actual length of individual pipe joints and of the
and composition of the gas are different throughout the
uncertainty introduced when the operator manually and
wellbore, this value is not necessarily representative of the
visually identifies the collar echoes in each second of the
velocity for a different section or of the average acoustic
record. The triangle marks the location of the average of
velocity for the total gas column between the wellhead
all the values (14.26 jts/s) and is plotted at the average
and the liquid level. The round trip travel time to the liquid
depth (3,587 feet). The red square represents the value
level is 17.409 seconds, so if one were to use this value to
of average acoustic velocity, using the marker analysis
compute the number of pipe joints to the liquid level, the
method discussed later in this chapter.
result would be 14.86 × 17.409 = 258.7 joints of pipe. In
Using the average collar echo frequency of 14.26
this well, the average length of a pipe joint is 30.85 feet,
jts/s and multiplying by the round trip time of 17.409
so the liquid level depth would be estimated at 7,980.8
seconds, the result is a total of 248.3 pipe joints to the
feet which would be inaccurate if the acoustic velocity in
liquid level or a depth of 7,660 feet. This value is 10.4
the lower sections of the well were significantly differ-
joints, or 320 feet, shorter than the value estimated ear-
ent from the estimated value. A more accurate estimate
lier using the collar echo frequency obtained from only
of liquid level depth would be preferred. The following
the upper part of the acoustic record and assuming this
sections discuss improved determination methods.
value applies to the whole acoustic record.
The process of manually identifying and counting
Stepwise Collar Echo Count collar echoes illustrated above has been automated
A similar analysis can be done on all sections of the and combined with adaptive digital filtering with the
acoustic record by scanning manually the signal in objective of improving the accuracy of determining the
uniform time intervals and determining the number of distance to the liquid level. This automatic algorithm is
collar echoes present in each section, as illustrated in described in the following section.

Figure 5.3 Identification of collar echoes in various segments of the acoustic record from figure 5.1, performed
visually and manually by the analyst

is selected as the best result and presented to the user

15 by placing a red marker on the record at the time where
14.8 the collar count is terminated.
COLLAR ECHOES PER SECOND
The result of such an iterative process is displayed

14.6
in more detail in figure 5.5, where the filtered acoustic
14.4 trace is segmented and displayed as 2-second intervals.
The collar echoes that are identified by the software
14.2
are indicated with vertical tick marks, and the average
14 number of echoes in the interval is displayed at the right
13.8 side of each row. The figure indicates that, for the sec-
tion in the first row (from time zero to 2 seconds), the
13.6
average number of echoes is 14.64 per second; for the
13.4 second row, the average number is 14.73; and so on, until
0 2,000 4,000 6,000 8,000
the tick marks stop at about the first third of the sixth
DEPTH, FEET
row. At 10.680 seconds in the filtered record, the collar
MANUAL COUNT y = -0.000164x + 14.850990
MARKER ANALYSIS R2 = 0.785087
echoes become indistinguishable from the background
AVERAGE noise, and thus the automatic count is stopped at 10.680
LINEAR (manual count)
seconds. Note that the tick marks are also missing at the
start of the acoustic record, from zero to 0.252 seconds.
Figure 5.4 Variation of collar echo frequency as a In this section, the acoustic signal that is generated
function of depth
by the pressure pulse discharged in the vicinity of the
microphone produces a high-level voltage that must be
Automatic Digital Filtering clipped to prevent damaging the digital circuits.
As the frequency of the collar echoes for the first section Thus, the record is not necessarily representative of
of the record is known (14.86 Hz), a narrow-band digital the actual acoustic signal, and the collar echoes cannot
filter can be designed and applied to the whole acoustic be identified until after this time. As indicated at the
record with the objective of recovering the majority of bottom of the figure, a total of 151 pipe joints have been
the collar echoes from the background noise. The data counted between 0.252 and 10.680 seconds of the record
processing logic6 includes comparing the amplitude of after applying to the acoustic trace a narrow-band filter
the collar echoes in the filtered trace to the amplitude of 4 Hz wide and centered at 14.8588 Hz.
the filtered background noise and determining the time The round trip travel time to the liquid level echo
in the record where the amplitude of the filtered collar is 17.409 seconds, so it is necessary to estimate the
echoes becomes less than the amplitude of the filtered number of joints that must be added to the 151 joints
background noise. This time in the record is considered to account for the additional time from the end of the
the limit where the count of tubing joints can be made automatic collar count to the liquid level echo. The time
with reasonable certainty. The number of collar echoes intervals where collar echoes have not been identified
that are identified is counted in this time interval. Then include 0.252 seconds at the beginning and 6.729 seconds
the process is repeated using a different narrow band (17.409 to 10.680) at the end—a total of 6.981 seconds.
filter, which is modified based on the frequency content Using the average number of joints per second of 14.48,
of an adjacent section of the record. The new time limit this time is equivalent to 101.1 pipe joints, which, when
where identification of echoes is possible is determined, added to the 151 joints counted, yields an estimate of
and the collar echoes are recounted. This adaptive filtering the liquid level depth at 252.57 joints.
operation is repeated for all sections of the record. The The conversion from the number of pipe joints to
process that yields the largest number of collar echoes the distance from the surface to the liquid level can be

Figure 5.5 Detailed collar count analysis of the record in figure 5.1
done either by referring to a detailed pipe tally with As a result of its simplicity and accuracy, this is by far
recorded lengths of individual pipe joints or by using the most common method for determining the distance
an average joint length for the pipe string in question. to the liquid level whenever echoes from pipe couplings
When a detailed pipe tally is not available, the average are present in the record. In addition to digital filtering
joint length can be estimated by dividing the depth to and spectrum analysis, other signal processing methods
the seating nipple (for a rod pump well), the depth to have been adapted from techniques used in seismic data
the wireline entry guide (in a gas well), or the depth to processing, such as cross-correlation, autocorrelation,
some other identifiable completion hardware, by the signal stacking, and wavelet transforms11,12,13. The objec-
number of full-length joints (not pup joints) that make tive of all these techniques is to enhance the signal-to-
up the string. When none of these values are known, it noise ratio of the acoustic record. For the most accurate
is customary to use a default value of 31.7 feet for the results, it is also important to optimize the pulse genera-
average tubing joint length. In this particular example, tion and acquisition system and procedure by selecting
the average pipe joint length was computed as 30.85 feet, the appropriate pulse generator and microphone and by
so the distance to the liquid level is computed with this verifying that they are properly installed on the well, as
value and is given as 7,792 feet in figure 5.5. discussed in chapter 4.

Distance to a Known Wellbore Anomaly the average velocity over a longer interval (5,913 feet)
(Downhole Marker Analysis) compared with the average velocity of 893 ft/s computed
The record shown in figures 5.1 and 5.6 was recorded from the collar count in figure 5.5, which covers a shorter
in the annulus of a well completed with two sections of wellbore section (about 4,800 feet, as shown by the “C”
casing with different internal diameters. (The casing in line in figure 5.1). The acoustic velocity calculated over
the upper section had a 7-inch outer diameter, a weight the greater length is probably more representative of
of 26 lbs./ft, and a 6.276-inch inner diameter; the casing the average over the total wellbore. Using this average
in the lower section had a 7-inch OD, a weight of 38 lbs./ velocity to compute the distance to the liquid level echo
ft, and a 5.29-inch ID.) The decrease in casing diameter gives a liquid level depth of 7,714 feet. That is a better
at the crossover point at 5,913 feet created an echo that estimate of the actual distance to the liquid level, since
is easily identified as a downward deflection (decrease this calculation does not use the average length of the
in area) at about 13.35 seconds. This measured time pipe joints, though it does rely on accurate information
and the known distance to the internal diameter change about the wellbore completion data. However, it should
can be used to calculate the average acoustic velocity be remembered that using a downhole anomaly for refer-
for the gas in the annulus between the surface and this ence yields a more accurate estimate of the liquid level
depth using equation 5.3. depth only when the anomaly is near the actual liquid
2La level. When echoes from multiple markers are present
v = —– Eq. 5.3
Δta in the acoustic record, one should use the marker closest
where: to the liquid level echo to estimate the distance.
La = measured distance from the surface to the In rod-pumped wells in which the tubing is anchored
known anomaly (feet) to the casing, it is often possible to see the echo from the
Δta = round trip travel time to the echo from the tubing anchor when the liquid level is below that depth.
known anomaly (seconds) Using the tubing anchor as the reference depth yields a
v = average acoustic velocity (ft/s) very good estimate of the average acoustic velocity and
In this instance, note that the computed average the distance to the liquid level14. The downhole marker
velocity of 885 ft/s (2 × 5,913 ÷ 13.35) corresponds to method also assumes that the distance to the anomaly is
Figure 5.6 Downhole marker analysis

entered correctly in the calculations. It is customary to using this method results in an estimate that places the
indicate the depths to specific hardware (packers, land- liquid level much deeper than it actually is, which may
ing nipples, valves, mandrels, and so on) that are run in prompt the operator to unnecessarily modify or adjust
the well based on physical measurement from the rotary the pumping system operation.
table (kelly bushing) of the completion or workover The most commonly used method of liquid level
rig. Because the time and distance corresponding to the calculation is to perform an automatic collar count after
echoes from these anomalies are determined relative to digitally processing the acoustic record to enhance the
the wellhead, it is important to know and use the kelly echoes from the pipe collars. In this instance, using this
bushing (KB) offset value to compute the correct acoustic method yields a liquid depth about two pipe joints, or
velocity and distance to the liquid level when the marker about 78 feet, deeper than the marker method. This would
depths are known relative to the rotary table. give an estimate of smaller pump submergence, but the
Table 5.1 summarizes the results obtained by these difference would probably not be significant enough to
three methods when applied to the record in figure 5.1. trigger a change in the mode of operation of the pumping
For all methods, the round trip travel time is 17.409 system. This liquid level depth and pump submergence
seconds and the average joint length is 30.85 feet. calculation also presumes that the operator has entered
Assuming that the downhole marker method yields a valid quantity for the average length of the pipe joints
the liquid level position with the minimum uncertainty, (30.85 feet per joint). If the default generic value of 31.7
the results show that the first method of using an esti- feet per joint were used instead, then the liquid level
mate of the acoustic velocity determined at the top of would be computed at a depth of 31.7 × 252.57 = 8,006
the wellbore gives the largest error for distance to the feet. In this case, the difference from the downhole marker
liquid level. This result must also be considered when method amounts to 292 feet, which is a larger deviation
using acoustic fluid level systems that rely on sampling than the deviation produced by the first method. It must
the gas at the wellhead to either directly measure the be stressed, therefore, that for every well, calculating or
acoustic velocity1,15,16 or measure the gas’s specific grav- determining the specific average joint length from an
ity and composition to compute an acoustic velocity, as accurate pipe tally must be given high priority to ensure
discussed in the next section. In this specific example, reliability of the computed fluid level depth.
Table 5.1
Comparison of Distance to Liquid Level Computed By Different Methods
Acoustic Number Distance Difference

Joints/ Velocity, of Joints to Liquid, from Marker
Method second ft/second to Liquid feet Method, feet
Collar frequency at top of well 14.86 916.9 258.7 7,980.8 266.8

Manual count of collar echoes — — 247.6 7,638 -76
Average collar frequency from
manual count 14.26 879.8 248.3 7,660 -54
Automatic collar echo filtering
and count 14.48 893 252.57 7,792 78
Downhole marker (casing weight
change) — 885 250.06 7,714 —

Calculating Acoustic Velocity from sampled gas for the effect of pressure and temperature
Gas Gravity or Composition by means of an equation of state model.
Often it may be necessary to undertake acoustic measure- Figure 5.7 is an example of a graph that can be used
ments in wells that are completed with coiled tubing, to estimate the average acoustic velocity if the average
non-upset tubing joints, or measurements inside casing pressure and temperature in the gas-filled portion of the
without a tubing string. In these cases, tubing collar well can be determined. This is an iterative approach
echoes will not be available for estimating the acoustic because the average pressure cannot be determined prior
velocity. In addition, when performing acoustic mea- to determining the liquid level. A reasonable estimate for
surements inside the tubing in low-pressure gas wells pressure at the bottom of a gas column can be determined19
or inside casing without a tubing string, it may not be using the equations shown in chapter 6. The problem in
possible to see echoes from the internal gaps in the col- this case is how to determine a representative average
lars between tubing or casing joints. It may be possible temperature of the wellbore gas. The points plotted in
to record echoes from the internal gaps of the collars in figure 5.8 are measured data during a wireline pressure
wells where the tubing or casing pressure is sufficiently survey in a pumping well that was performed shortly
high, so that sound attenuation is not too large. When after pulling the sucker rods. The temperature profile
analyzing records acquired inside casing, the average shows two distinct regions: one corresponding to the
joint length must be adjusted to match the range length gas section and one for the bottom liquid section. For
of the specific string of the pipe. acoustic velocity calculations, it is customary to compute
When collar echoes are not distinguishable in the re- the average temperature using an average between the
cord, the acoustic velocity of gases and gas mixtures can ground level temperature and the bottomhole temperature
be estimated from the thermodynamic, pressure, volume, (generally obtained from the header of wireline logs). In
and temperature properties of the fluid. Various methods this example, the star point on the blue line represents
have been developed that include sophisticated equation the computed average at about 117°F. Looking at the
of state models17,18. These methods require knowledge temperature distribution in the gas section (red line),
of the composition of the gas or, at a minimum, a value the average temperature in the gas is about 96°F, as
for the gas gravity plus the pressure and temperature indicated by the red star. At the average pressure of 450
of the fluid. The preferred method is to obtain a sample psi, the corresponding acoustic velocities obtained from
of the gas at the wellhead under flowing conditions to figure 5.7 are 1,360 ft/s at 96°F and 1,405 ft/s at 117°F.
ensure that it is representative of the wellbore gas and to In the well where the wireline temperature survey
determine its composition by chromatographic analysis was performed, the acoustic record shows the liquid level
of the sample. Separator or sale-line gas samples should echo at a round trip travel time (RTTT) of about 11.6
not be used, since by the time it reaches those points, the seconds. Using the overall average temperature or the
gas has experienced significant pressure and tempera- average temperature of the gas column, the difference
ture changes and its composition may be very different in computed liquid level depth would be 11.6 × (1,405
from that of the wellbore gas. In the case of coal bed – 1,394) × 1/2 = 261 feet, which is quite significant.
methane wells that principally produce methane without Unfortunately, a temperature survey on a pumping well
significant volumes of heavier hydrocarbons it is possible is very rarely conducted, though these surveys may be
to use samples taken downstream of the wellhead. As more common in flowing oil and gas wells or in gas lift
discussed in the previous section, the gas sample taken wells. Detailed temperature distribution information
at the wellhead is at temperature and pressure condi- should be used whenever it is available.
tions that are not representative of the average values A more precise calculation of the average acoustic
inside the wellbore. Therefore, the calculation of a more velocity and fluid level should start with a sonic ve-
representative acoustic velocity must include a method locity computed at surface pressure and temperature
for correcting the measured acoustic velocity of the conditions to estimate an initial value of the liquid level

PRESSURE (kg/cm2)
0 20 40 60 80 100 120 140 160 180 200

2,000
600
ACOUSTIC VELOCITY
1,900 GAS GRAVITY = 0.6 580
(BASED ON AIR = 1.0)
560
1,800
TEMPERATURE
540
FAHRENHEIT CELSIUS
200ºC
392ºF
ACOUSTIC VELOCITY (meters/sec)

1,700 520
ACOUSTIC VELOCITY (ft/sec)
161ºC
321ºF
500
1,600
250ºF 121ºC
480
215ºF 101ºC
1,500 460
144ºF 62ºC 440

117ºF VELOCITY
1,400 = 1,405 ft/sec
108ºF
96ºF VELOCITY 420
42ºC = 1,360 ft/sec
73ºF
23ºC
400
55ºF
1,300 13ºC
37ºF
3ºC 380
19ºF
-7ºC
1,200
360
VELOCITY = 1,400 ft/sec
340
1,100
0 500 1,000 1,500 2,000 2,500 3,000
PRESSURE (psia)
Figure 5.7 Effect of temperature on the average acoustic velocity of gas (with a 0.6 gravity) in a pumping well

Temperature Survey in Pumping Well
TEMPERATURE, DEGREES F
0 20 40 60 80 100 120 140 160 180 200

0
2,000
GAS COLUMN
4,000 AVERAGE = 96º F
LINEAR
AVERAGE = 117º F
DEPTH, ft
6,000
8,000
LIQUID
COLUMN
10,000
12,000
Figure 5.8 Temperature survey in a pumping well
depth. With the initial liquid level estimate, a calculation

of pressure and temperature at the liquid level is made
and a second sonic velocity is determined using the new
Courtesy of AMETEK, Inc.
average values. The process is repeated until the liquid

level depth converges.
Figure 5.9 shows two portable instruments that can be
used to analyze the sampled gas at the well site and obtain
accurate values of gas gravity and the percentage of each
hydrocarbon and other components in the gas mixture.
In many older wells, it may not be possible to access Figure 5.9 Portable instruments for gas gravity (left)
wireline log information to determine the bottomhole or gas analysis (right)

temperature. In these cases, one can use knowledge of Estimating Acoustic Velocity from Similar
the regional temperature gradient to calculate a fairly Wells or Past Acoustic Surveys
accurate number. Figure 5.10 shows a map of temperature In many oil and gas fields, it is customary to produce
gradient distribution over a portion of the Southwest of the wells to centralized production facilities by group-
the United States. ing wells into production/test satellites that operate
For the general location of the well in figure 5.8, at fairly constant pressure conditions. Therefore, it is
the average ground temperature commonly used is 75°F common to have the casinghead pressure of most wells
and the temperature gradient from the map is 0.9°F/100 at about the same pressure level. In these cases, it has
feet. Using these values, the computed temperature been observed that the acoustic velocity of the annular
at 10,420 feet is 168.8°F, which is very close to the gas in wells producing from the same reservoirs is fairly
measured temperature of 168.5°F. Similar maps can be similar from well to well.
obtained from other state agencies regulating oil and gas Table 5.2 shows an example of acoustic fluid level
production or from the United States Geological Survey data obtained in five wells producing in the same field.
(USGS, whose website can be found at www.usgs.gov). Note that the casing pressure levels are similar, so the
Courtesy of the Society of Petroleum Engineers (SPE)
Figure 5.10 Isothermal gradient map. Contour values are °F per 100 feet of depth.

Table 5.2 acoustic velocities are all within the range of 1,140 to
Acoustic Velocity of Annular Gas in Wells 1,168 ft/s with an average of 1,150 ft/s.
Producing From Same Reservoir Therefore, in those instances when acoustic measure-
ments have to be made in a well where it is not possible
Acoustic to observe echoes from the pipe collars or downhole
Casing Velocity,
Well ID Pressure, psi ft/second markers, one has the option of using the average acoustic
velocity that has been determined in similar wells in the
same field or lease, as long as the pressure levels are
1014H 660 1,140
within the same range. When pressure levels are signifi-
1074H 611 1,155 cantly different, this approximation should not be used.
It should also be noted that, as long as the casinghead
1114H 702 1,144
pressure in a given well does not change significantly,
1164H 762 1,168 the value of the acoustic velocity of the annular gas will
remain in a fairly narrow range. Figure 5.11 is a graph
1234H 651 1,145
that illustrates the variation of acoustic velocity over
a period of nearly three years in a well producing in a
Well 125T Variation of Acoustic Velocity from 3/2007 to 10/2010
1,100
1,050
DECREASING TREND
ACOUSTIC VELOCITY, ft/second
1,000
950
900
850
800
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83
ACOUSTIC RECORD NUMBER
Figure 5.11 Variation of the acoustic velocity of the annular gas in a given well over three years

mature field. The casinghead pressure was maintained Stopping the pump will cause the fluid level to
in the range of 12 to 15 psi during this period with fluid shift, so the pump should not be stopped for
level between 3,500 and 3,800 feet. extended periods of time.
The acoustic velocity was determined automatically • In pumping wells producing in mature fields with
using the collar count and has not been checked for low BHP, do not leave the casing-to-flow line
quality. The bar graph shows that the acoustic veloc- valve closed for a long time, because added pres-
ity exhibits a minor generally decreasing trend, but on sure may force the liquid level below the pump.
average it is in the 1,030 ft/s range.
The graph also shows that some acoustic records
yield velocities that vary from the normal velocity SUMMARY
(890 to 965 ft/s). This is a valuable indication that the Transmitted pulses and the recorded acoustic signal must
automatic collar echo analysis could be wrong and that be of sufficient quality for good results. Many expensive
the operator needs to verify the quality of the acoustic errors have been made by submitting an erroneous fluid
record, since the presence of excessive noise or insuf- level depth that was calculated using an echo that was
ficient pulse amplitude would yield unreliable acoustic actually caused by a casing liner or some other anomaly
velocity estimates. Keeping track of the values of in the casing ID or tubing OD. An accurate description
acoustic velocity from record to record over time is a of the wellbore geometry and well completion is neces-
recommended procedure for quality control of the data sary for reliable estimation of fluid level.
and the results from the liquid level analysis. A clear indication of the fluid level echo that can be
identified among multiple echoes should be achieved by
RECOMMENDATIONS FOR adjusting the amplitude of the transmitted pulse in relation
BACKGROUND NOISE to the background noise in the well. Knowledge of the
wellbore geometry is necessary for correct identification
If excessive noise exists:
of each of the echoes observed in the acoustic record. A
• Use a larger pressure differential between the detailed wellbore and completion diagram that includes
wellbore and the gas gun chamber. directional survey information should be available, and
• Close the valve between the microphone and the the data should be entered in the analysis software. The
well to determine if the source of the noise is round trip travel time to the echoes should be measured
downhole. If it is, then the noise level will decrease. preferably with an accuracy of ±0.001 seconds. This
Otherwise, the noise source is from the surface ensures a resolution in distance of about ±0.5 feet for
connections or is caused by wellhead vibrations. an acoustic velocity of 1,000 ft/s.
• Causing an increase in well pressure will improve A representative value of the average pipe joint
the signal-to-noise ratio and will facilitate identifi- length should be used when the estimated number of
cation of echoes in wells operating at low surface pipe joints to the fluid level is converted to distance.
pressures (near atmospheric). Using a generic average value (such as 31.7 feet per
• For best results, always inspect cables, connectors, joint) does not provide accurate results.
electronics, the microphone, and so on, and per- Always inspect the amplitude of the background
form recommended troubleshooting procedures. noise before acquisition. Measurements should be re-
• In pumping wells operated with noisy surface peated whenever excessive background noise is present
units (such as hydraulic linear drive and rotary or whenever unexplained signals are observed. Random
drive units), gas lift wells, or injection wells, use signals will not be observed in subsequent records or will
available digital filtering tools to eliminate noise be observed at different travel times. Signals caused by
from the record before considering whether it real features actually present in the well will be observed
will be necessary to stop the pumping system. at about the same travel time and will exhibit similar

features. Similar signals that occur periodically in the 8. DADiSP, Digital Signal Processing Software: http://
acoustic record should be checked for correlation with www.dadisp.com/filters.htm.
the period of pumping unit operation. 9. J. N. McCoy, K. N. Huddleston, and A. L. Podio,
Verify that the acoustic velocity being used to cal- “Data Processing and Display for Echo Sounding
culate depth to liquid is reasonable. A comparison of Data,” U.S. Patent 5,117,319, May 1992.
this calculated number to what is normal for the wells 10. J. N. McCoy, K. N. Huddleston, “Detection of Fluid
being measured in that area can prevent errors in the fluid Reflection for Echo Sounding Operation,” U.S.
level measurement. In rod-pumped wells, as the annular Patent 5,285,388, February 1994.
liquid level is lowered, pump load increases. So, take
11. R. Zamow, “Method and Device for Recognizing
advantage of the correlation between the annular liquid
the Depth Level in a Bore Hole,” EU Patent, EP
level and the pump dynamometer loads to validate the
2169179 B1, November 2011.
acoustic fluid level.
12. S. A. Burris, “Well Pump Control Using Multiple
Sonic Detectors,” U.S. Patent 6,085,836, July 2000.
REFERENCES 13. R. S. Grande, “Well Collar Identification Method,”
1. C. P. Walker, “Determination of Fluid Level in Oil U.S. Patent 5,748,471, May 1998.
Wells by the Pressure-wave Echo Method,” Trans- 14. C. P. Walker, “Pressure Wave Velocity Measuring
actions of AIME, 1937. System,” U.S. Patent 2,190,141, February 1940.
2. P. E. Lehr, “Method and Apparatus for Measuring 15. K. B. Nolen, “Determining Fluid Level in Wells
Well Depths,” U.S. Patent 2,047,974, July 1936. With Flow Induced Pressure Pulses,” U.S. Patent
3. J. N. McCoy, “Automatic Echo Meter,” U.S. Patent 5,715,890, February 1998.
4,934,186, June 1990. 16. S. G. Gibbs, “Systems and Methods for Measuring
4. A. Wolf, “Acoustical Well Sounder,” U.S. Patent a Fluid Level within a Well,” U.S. Patent 8,261,819,
2,560,911, July 1951. September 2012.
5. D. D. Barber, “Automatic Liquid Level Recording 17. L. K. Thomas, R. W. Hankinson, and K. A. Phillips,
Device,” U.S. Patent 4,853,901, August 1989. “Determination of Acoustic Velocity for Natural
6. T. K. Ahern, “Method and Apparatus for Generating Gas,” Journal of Petroleum Technology, July 1970.
Data and Analyzing the Same to Determine Fluid 18. National Institute of Standards and Technology
Depth in a Well,” U.S. Patent 4,793,178, December (NIST), Thermo-Physical Properties of Hydrocar-
1988. bon Mixtures Database: Version 3.2.
7. A. B. Williams and F. W. Taylor, Electronic Filter 19. J. J. Jakorsky, “Bottom Hole Measurements in
Design Handbook (McGraw-Hill, 2006). Pumping Wells,” Transactions of AIME, 1939.

Calculating Wellbore Pressure Distribution from Acoustic Fluid Level Surveys 6-1
6
Calculating Wellbore Pressure Distribution
from Acoustic Fluid Level Surveys
In this chapter:
• Pressure distribution in pumping wells
• Classification of wells by wellbore and producing conditions
• Gaseous liquid column gradient
• Liquid level depression test
• Gas-free liquid pump submergence
• Correlations and mechanistic models
• PBHP and SBHP calculations
Acoustic determination of the depth to the liquid in the properties as a function of pressure and temperature and
wellbore was introduced in the 1930s by C. P. Walker1, the lengthy iterative computations.
who also outlined graphical methods for obtaining the The advent of portable digital data acquisition and
pressure distribution in the well. At that time, the main processing provided the tools needed to routinely convert
objective was determining the depth of the gas/liquid fluid level measurements into estimates of downhole
interface in relation to the depth of the pump intake in pressure distribution in the wellbore at the well site3.
order to estimate the pump submergence. Downhole Today, surface pressure and pump submergence are eas-
pump submergence is defined as the amount (height) ily converted to pressure at both the depth of the pump
of liquid that exists above the pump intake. Since the intake and the depth of the producing formation and then
early days of rod pumping applications in the oilfield, reported with an analysis of acoustic fluid level records.
the submergence of the pump has been the parameter
most commonly used for monitoring and troubleshooting
well operation. Often abbreviated FAP for “fluid above WELL PERFORMANCE AND
pump,” it was (and still is today) periodically monitored POTENTIAL ANALYSIS
and recorded. Based on its value, the operation of the As discussed in chapter 1, the producing efficiency of a
pumping system can be adjusted to maintain an adequate well can be determined at a given time using an inflow
submergence, which has been defined as about 100 feet of performance relation (IPR) that expresses the effect on
fluid, to provide sufficient pump intake pressure to force pressure drawdown of the rate of production from the
the fluid into the pump at the operating pumping rate. formation. These relations require knowing the producing
The importance of knowing the pressure distribution bottomhole pressure (PBHP) and the static bottomhole
in the wellbore for detailed analysis of well performance pressure (SBHP) corresponding to a steady production
was recognized early in the 1930s2. However, for many flow rate. The simplest relation, applicable to wells
years it remained a research tool because of the dif- producing primarily liquids, is given by the productivity
ficulties involved in obtaining accurate values of fluid index (PI), defined as “barrels per day of gross liquid

at Austin
produced per pound per square inch pressure drop at a Other inflow performance relations, such as the Vo-
specified subsurface datum”4. gel IPR relation (which is discussed in chapter 1), have
The PI obtained in pumping wells from steady state been developed to describe the more complex behavior
fluid level measurements at two or more flow rates was of reservoirs where flow at the perforations includes
compared to productivity values calculated with pres- both liquids and gas. The wellbore pressures computed
sures obtained from bottomhole sensors. The comparison from fluid level measurements are also commonly used
showed that the results were similar: 1.41 bbl/day-psi for with these functions in the design and analysis of the
the acoustic method and 1.48 bbl/day-psi for a pressure artificial lift system for a specific well.
sensor5. This study also points out that the major source
of error in the value computed from the fluid level is the Fluid and Pressure Distribution
determination of the in situ density of the fluids. These in Pumping Wells
errors were minimized and negligible for all practical Acoustic fluid level measurements can be obtained in
considerations by establishing rates where an appreciable all wells, regardless of the type of artificial lift system
drawdown of pressure was obtained. Since that time, being used (sucker rod, electric submersible pump,
major improvements have been made in the accuracy progressing cavity pump, hydraulic, plunger lift, or gas
of fluid properties calculations for determining wellbore lift). In addition, the complete pressure distribution in
pressure distribution in producing and static wells, as the wellbore, as shown in figure 6.1, can be determined
discussed in more detail in the following sections. (provided a downhole packer is not present) without
Pcf = 10 psi Ptf = 40 psi FLOW RATE Q

Ptf
CONSTANT: Pcf
Flow Rate, Tubing Diameter,
Depth, GLR, Ptf, Pcf
ANNULUS TUBING
DEPTH
PRODUCING
FLUID LEVEL
PIP = 50 psi PDP = 2,540 psi

PUMP
5,000 ft
400 psi PBHP 2,600 psi SBHP

DRAWDOWN
PRESSURE
PBHP
Figure 6.1 Pressure distribution in a pumping well

the use of wireline-conveyed or permanently installed the pump intake by correcting the height of the gaseous
downhole pressure recorders. liquid column for the volume of free gas present. This is
Figure 6.1 shows schematically the wellbore defined as the equivalent gas-free liquid height.
configuration and pressure distribution that exist in a The PBHP is the pressure in the wellbore opposite
well producing at a stabilized flow rate by means of the perforations or producing zone when the well is
a downhole pump, regardless of the type of pumping operating at stabilized conditions. In horizontal wells,
system employed: rod pump, progressing cavity pump, the pressure at the depth of the heel is generally con-
electrical submersible pump, jet pump, and so on. The sidered to be representative of the producing pressure
measured depth to the fluid level is determined by ac- in the lateral wellbore.
quiring and processing the RTTT of an acoustic signal The SBHP is the pressure in the wellbore opposite
in the casing-tubing annulus. The casinghead pressure is the perforations or producing zone when the flow of
measured and its trend monitored for a short time with fluids from the formation has ceased. This assumes that
an accurate and very sensitive pressure transducer. The the pumping operation has been stopped, and all flow
surface pressure is then used to compute the pressure from the well’s surface has ceased for a sufficiently long
at the bottom of the annular gas column using the gas’s time that the BHP has equilibrated the reservoir pressure.
PVT properties and the geothermal temperature profile. Acoustic fluid level measurements are routinely
The casing pressure trend over time is used to determine used to compute downhole pressures in wellbores that
whether free gas is entering the well from the formation contain mixtures of gas and liquids. The acoustic method
and at what rate. The gas flow rate is used to estimate the of wellbore pressure calculation was first sanctioned by
fraction of gas and liquid present in the fluid between the Energy Resources Conservation Board (ERCB) of
the depth of the producing fluid level and the depth of Canada in 19786 and has since been adopted throughout
the formation. Then, the pressure at any depth in the the world as the most practical means of obtaining BHPs
wellbore is computed using the gradient of the fluid in rod pump wells without the need to pull the rods or tub-
below the gas/liquid interface. ing to install a downhole pressure gauge. It is extremely
The principal objectives of such analysis are to important to have representative fluid compositions and
determine the following: flow rates. The wellbore schematic and tubing tally need
• Pump intake pressure (PIP) or pump submergence to be up to date, and the casing pressures and fluid levels
• PBHP should be measured by experienced field personnel using
accurate and well-maintained equipment.
• SBHP
Further improvements to the standard ERCB
• Distribution and composition of fluids in the
method were proposed and applied7, including the use
wellbore
of an equation of state and advanced correlations to
The PIP determined acoustically is the pressure in determine fluid properties. After-flow liquid composi-
the casing-tubing annulus at the depth specified as the tion was computed by applying the productivity index
pump landing nipple or at the depth of the intake valve to the water phase and applying the Vogel correlation to
(standing valve). Depending on the configuration of the the oil rate to account for the varying water cut after the
piping installed below the pump intake (strainer, gas pump is shut down. The results of the study indicated
anchor, mud joint, and so on), the PIP computed in the that the average absolute difference between the SBHP
annulus may or may not be equal to the pressure at the obtained from refined analysis of the acoustic surveys
actual pump intake. In general, however, this value is and the SBHP measured with BHP recorders was of
considered to be equal to the pressure below the stand- the order of 0.8%. If the standard ERCB 5 method had
ing valve (for a rod pump) or at the depth where fluid been used, the difference would have been as high as
enters the pump. Pump submergence is expressed as the 10.4%, which is probably adequate for well potential
height of the liquid that is present in the annulus above calculations, as discussed in chapter 1.

CALCULATING PRODUCING PRESSURE pump intake set at the bottom of the perforated interval,
DISTRIBUTION can be established not only in wells that have pumps
operating continuously, but also in those that are on
Calculating the PBHP from fluid level surveys assumes
timers or pump-off controllers. This assumes that the
that, when the fluid level is determined, a pumping well
on-off periods are not excessively long (within 30 to
is operating at a stabilized condition. Therefore, the
45 minutes), so that an average value of the fluid level
standard volumes of fluids produced at the surface would
can be established and the interfaces—gas/fluid and oil/
be equal to the standard volumes of fluids entering the
brine—simply oscillate over relatively short distances.
well from the formation. In this case, it is considered
For a more accurate determination of pressure, it may
that there is no accumulation of material in the well-
be advisable to obtain fluid level records at the begin-
bore and that the wellbore is functioning similarly to
ning and the end of the pumping period to establish
a three-phase surface production separator with steady
the high and low bounds of the computed pressures.
interfaces between oil, brine, and gas. The condition of
In rod-pumping wells, whenever possible, fluid level
stabilized pumping operation is easily verified when the
measurements should be repeated at short intervals while
following conditions are met:
a dynamometer record is acquired continuously during
• The fluid level does not change as a function
the pump-down phase. This allows establishing the
of time.
range of pressures during the on-off periods and permits
• The casinghead pressure is constant under normal
verification that the POC cycle timing is set correctly.
operation.
For a stabilized pumping well, the degree to which
• The produced water/oil ratio is constant.
oil/water separation takes place depends on the following:
These conditions, illustrated in figure 6.2 for a well
• The time during which equilibrium of inflow-
completed with a single-size casing and tubing with the
outflow has existed: The longer the pumping
system is continuously operated, the more likely
that full separation of the liquid phases in the
REQUIRES A CONSTANT Pt annulus has been reached.
PRODUCTION RATE
GOR AND WOR • The rate at which annular gas is flowing: The
Pc
greater the velocity of the in situ gas in the an-
nulus is, the more turbulent the flow pattern and
REQUIRES A STABLE greater oil/water mixing will be maintained in
CASING PRESSURE
the gaseous liquid column.
GAS
• The stability of the oil/water mixture in the
particular well: The tendency to form emulsions
creates a greater uncertainty as to the composition
REQUIRES A STABLE and density of the liquid phase.
FLUID LEVEL FLUID LEVEL
OIL AND
In wells where the oil and water are readily sepa-
PERFORATIONS GAS rated by gravity, the water will accumulate below the
PBHP
PUMP pump intake and will be produced through the pump.
BRINE
GRADIENT Under these conditions, the liquid in the gaseous liquid
column will be primarily oil, and its density (adjusted
FLUID LEAVING THE WELLHEAD = FLUID
ENTERING FROM PERFORATIONS for pressure, temperature, and solution gas) in situ is
used in calculating the fluid gradient5.
Figure 6.2 Fluid distribution in a stabilized pumping The fluid flow pattern in the annular gaseous liquid
well column is a special case of multi-phase flow defined as

zero net liquid flow, because only the gas phase is flow- at or below the perforations. These wells are common in
ing upwards to the top of the wellbore and out of the depleted reservoirs and also in other reservoirs whenever
casinghead. At any point in the gaseous liquid column, pumping with the casing valve closed causes gas to ac-
the liquid simply recirculates in place as the gas flows cumulate in the annulus and forces all produced free gas
through. The in situ flow pattern depends on the gas flow through the pump. The wellbore pressure magnitude and
rate (in situ gas velocity) and can range from bubbly distribution is primarily controlled by the casinghead
flow to slug flow to churn flow, but in all cases there is pressure and the density of the gas, since very little or
no net displacement of the liquid. Therefore, the liquids no liquid is present in the wellbore above the formation.
present in the annulus will become saturated with gas Wells designated as type B wells have only saturated
at the pressure and temperature existing at the specific liquids in the wellbore. This implies that only liquid is
point, and their densities should be calculated along flowing through the perforations into the well and that the
the saturation curve of the gas solubility function. This pressure at the wellhead is sufficient to maintain all gas in
implies that the pressure distribution in the wellbore will solution in the liquid phase below the gas/liquid interface.
have to be calculated iteratively in small pressure incre- Wells designated as type C wells have gas and liquid
ments, as discussed in more detail later in this chapter. flowing simultaneously from the perforations into the
wellbore. The pump intake is commonly set some dis-
Wellbore and Completion Classification tance above the perforations or above the heel in a well
Most pumping wells can be classified into three groups, drilled horizontally. The tubing/casing annulus exhibits
depending on both the position of the pump intake rela- a gaseous oil column from the depth of the pump intake
tive to the producing formation and the position of the to the depth of the fluid level. The wellbore below the
fluid level relative to the pump intake. Figure 6.3 shows pump intake contains a mixture of gas and primarily
that wells designated as type A wells operate with the brine even when the produced water/oil ratio is very
fluid level at or near the pump intake and the pump set low. Type A and type C wells comprise the bulk of wells
using pumping systems, including horizontal wells and
wells with multilateral extensions.
An accurate surface measurement of the casinghead
pressure as a function of time can be used to determine
whether a well falls into one of the three categories. When
the annulus is connected to the flow line, if gas normally
flows out of the casing and if the casing pressure increases
after the casing-to-flow-line valve is closed, the well is
either type A or type C. The rate of casing pressure increase
is directly related to the flow rate of gas entering through
the perforations. On the other hand, when the fluid level
is above the pump and the casinghead pressure does not
increase with the casing-to-flow line valve closed, or
when the casing is maintained as shut-in during normal
pump operation, the well is type B.
TYPE A TYPE B TYPE C
LIQUID LEVEL LIQUID LEVEL LIQUID LEVEL When the pressure remains constant after shutting in
(at perforations) (above formation (above formation the casing valve to the flow line, there cannot be any free
and not producing and producing
annular gas) annular gas) gas entering the wellbore through the perforations, and
the annulus contains only liquids overlain by gas. This is
Figure 6.3 Classification of pumping wells by tubing characteristic of wells that are producing at a BHP greater
depth and fluid distribution than the bubble point pressure of the hydrocarbons in the

reservoir and is often found in pumping wells producing

from reservoirs undergoing water flooding or pressure Pc GAS
OIL
maintenance. Zero annular gas inflow is also typical of BRINE
wells that have been shut down for an extended period of
time (either voluntarily or because of pumping equipment
SECTION 1
failure), so that the BHP has equilibrated with the reser-
voir pressure and fluid inflow has ceased. The fluid level
survey can thus be used to obtain a good approximation Z SECTION 2
of the SBHP, as discussed later in this chapter.
The most commonly encountered tubing-casing an- SECTION 3
nulus configuration is illustrated in figure 6.2, in which SECTION 4

the pump intake is set at or above the perforations or the
heel of the horizontal extension. A mixture of free gas and SECTION 5
liquids is flowing into the wellbore from the formation.
Pz
Some of the gas will enter the pump, and the remaining
volume of gas will flow to the surface and percolate
through the annular liquid, which will generate a multi- Figure 6.4 Generalized wellbore configuration for
tapered tubing and casing
phase annular gaseous liquid column exhibiting zero net
liquid flow. In low-volume wells, the fluid level is often
near the pump inlet. pc = surface pressure at the casinghead (psi)
Pressure Calculation pfi = pressure increase due to the fluid column
in section i (psi)
In all three types of wells, the pressure at a certain depth pfn = pressure increase due to the fluid column
in the wellbore is computed as the sum of the measured in partial section n containing depth z
surface pressure plus the pressure resulting from the
The pressure increase resulting from a column of
columns of fluids that exist above the point of interest.
fluid is expressed as the product of its height (TVD) and
Considering the case where the wellbore may con-
the average gradient of the fluid contained in that sec-
sist of several diameters and the tubing string may be
tion. The fluid may be single-phase (gas, oil, or brine)
tapered, the well can be divided into various sections,
or may consist of a mixture of gas and liquid. Since
each characterized by a unique combination of casing or
fluid gradients should be computed at the pressure and
tubing diameters, the type of fluid inside, and the presence
temperature conditions that exist in the wellbore (in
of perforations. Figure 6.4 illustrates a common multi-
situ pressure and temperature), an iterative calculation
diameter well completion with tapered tubing and a casing
is undertaken, beginning with the known pressure and
liner. Assuming that the pressure calculation is performed
temperature at the surface, in order to calculate initial
down the annulus, the wellbore is divided into n distinct
values of the fluid densities and then step down to the
sections; in this example, n = 5. The fluid in sections 1,
point of interest in small increments of pressure and
2, and 3 consists mainly of gas. The fluid in sections 4
temperature, recalculating the fluid properties at each
and 5 generally consists of mixtures of gas and liquid.
step until convergence of pressure values is achieved.
The pressure at an arbitrary true vertical depth
Assuming that the well flow is stabilized, as was
(TVD) z can be computed as:
discussed earlier in relation to the simpler configuration
pz = pc + ∑ n–1
1 pfi + pfn Eq. 6.1 of figure 6.2, once the depth to the gas/fluid interface has
where: been determined from the acoustic record, the remaining
pz = pressure in the wellbore at true vertical issues are determining the:
depth z (psi) • Average gravity of the casing gas

• Fraction of liquid in the sections where a gas/ The liquid in the annular fluid column above the pump
liquid mixture is present intake is un-replaced (as a result of zero net flow) and
• Composition of the liquid in each section cannot provide any additional gas coming out of solution.
• Density of each phase If any gas is flowing out at the top of the well, it must
travel through the annulus liquid and, in the process,
These values are required to compute the corre-
keep the liquid saturated with gas. In the annular gas/
sponding fluid gradients and the pressure increase, pfi,
liquid mixture, most of the gas is distributed through-
due to each fluid column:
out; it originates from the formation and flows into the
pfi = Hi × γi Eq. 6.2 wellbore as free gas through the perforations10.
where: A foam layer on top of a solid liquid column can
γi = average gradient of fluid in section i (psi/ft) only exist in the annulus of a pumping well when both
Hi = vertical height of section i (feet) of the following conditions exist:
• Gas is not normally produced from the casing.
Gas Gravity • The pressure at the wellhead does not increase
significantly when the casing valve is closed while
The average annular gas gravity can be computed us- performing the acoustic fluid level measurement.
ing an equation of state from the measured acoustic
velocity, average temperature, and average pressure8,9. Occasionally, a foam layer may be observed in shut-in
Alternately, a sample of the gas may be obtained at wells that have not produced for a long time but have
the casing or tubing head, in which case the gravity of a casing valve that has not been fully closed, allowing
the gas is measured directly, as discussed in chapter 5. some gas to percolate to the surface at a very slow
Except for coal bed methane wells that produce mainly rate (less than 0.5 Mscf/day). The presence of a light
methane and water, the gas gravity should not be deter- foam at the top of a liquid column may attenuate the
mined from a gas sample taken downstream from the acoustic signal to the extent that a fluid level echo is not
well at a production separator or at the sales line. The observed in the acoustic record unless a large pressure
gas composition at those points is different from the pulse is generated. In some of these wells, to obtain an
composition of the gas inside the wellbore. interpretable fluid level record, it may be necessary to
inject brine in the annulus to destroy the foam layer.
Calculating the pressure distribution in the wellbore
FRACTION OF LIQUID IN of a pumping well therefore requires determining the
A GASEOUS COLUMN gradient of the gas/liquid mixture that is present from the
Whenever a well is not pumped off—there is significant depth of the fluid level to the depth of the formation. In
pump submergence—it is often wrongly assumed that the case of the well represented by figure 6.4, this implies
there is, in the annulus above the pump, a homogeneous determining the fraction of liquid and gas that exists in
liquid column topped by a layer of foam caused by gas sections 4 and 5 of the wellbore that are characterized
coming out of solution. This concept is invalid for any by different casing-tubing diameter combinations.
steady state pumping condition (for type A and type C
wells) where there is free gas flowing from the perfora- Liquid Level Depression Test
tions. As discussed earlier, free gas flow from the forma- The fraction of liquid present in the gaseous liquid
tion is easily detected during an acoustic fluid level survey column in a pumping well may be determined directly
that shows an increase in wellhead pressure during the using the Walker liquid level depression test, which was
acquisition of the echoes when the casing flow valve is originally patented in 193911. The objective of the test was
closed. The gas flow rate is estimated from the slope of to determine the fluid pressure at any given point below
the short (2 to 4 minutes) casing (or tubing) pressure the surface of the gaseous column, calculate its density,
versus time recording, as explained later in this chapter. and estimate the productivity index of the formation

without interrupting or interfering with the production of inflow from the formation will return to the conditions
of the well. Based on the results obtained from the test, that existed prior to increasing the wellhead pressure.
the most desirable size and location of the pump, the The Walker test is described in its patent as follows:
operating characteristics of the pumping equipment, From an exhaustive series of actual tests, it has
and the resulting flow rate could all be determined for a been discovered that the relation between the
given well in order to increase or maximize production. pressure on the liquid surface and the distance
The Walker test is based on the observation that, in through which the fluid will recede in any given
well having a uniform cross-sectional area is
a stabilized pumping well producing at a given flow rate, constant. By dividing the difference in fluid
the height of the gaseous liquid column above the forma- pressure by the difference in surface level, the
tion is dependent on the magnitude of the casinghead density of the liquid mixture of oil and gas in
pressure. An increase of the casing pressure causes the the well (that is, the pressure exerted by this
fluid level to be depressed and vice versa. If the casing mixture per unit of height of the column of
pressure is allowed to increase and stabilize at a certain liquid mixture) may be found. 11
value, then the fluid level will also stabilize at a deeper Figure 6.5 illustrates the pressure conditions for
point. In addition, the operation of the pump and the rate one stage of the liquid level depression test. When the
While Pumping at Constant Rate:

INCREASE CASING PRESSURE DEPRESS LIQUID LEVEL
BEFORE AFTER
Pc = 30 psig Pc = 250 psig
PRESSURE INCREASE = 220 psi
Pg = 10 psig
Pg = 20 psig
FLUID LEVEL
DROP 2,300 ft
2,500 ft Pgc = 310 psig
Pgc = 60 psig
200 ft
GRADIENT = 220 psi/2,300 ft = 0.095 psl/ft
PERFORATIONS
BOTTOMHOLE PRESSURES
PUMP PUMP
ARE EQUAL AT EACH
INTAKE INTAKE
STABILIZED CONDITION
Pwf = 350 psig Pwf = 350 psig
Pc = Casing Pressure
Pg = Gas Column Pressure
Pgc = Gaseous Liquid Column Pressure
Pwf = Pc + Pg + Pgc
Figure 6.5 Calculating the gaseous column gradient from a liquid level depression test. The pumping rate stabilized
at a 70-bbl/day liquid rate.

pump is displacing an average of 70 bbl/day through words, 2,300 feet of gaseous column are equivalent to
the 2M-inch tubing, the PBHP is stabilized at 350 psig, 220 psi, or the gradient of the gaseous liquid column
corresponding to the well’s IPR. The PBHP is balanced is equal to 220 psi divided by 2,300 feet, resulting in
by a casing pressure of 30 psig and a gaseous liquid an average of 0.095 psi/ft.
column of 2,500 feet. The pressure at the depth of the This gradient corresponds to the percentage of
gas/liquid interface is about 40 psi. Gas is flowing from gas and liquid that existed in the annulus above the
the 5H-inch casing to the flow line at a steady state rate formation prior to the depression test. Assuming that
of 51 Mscf/day. A back-pressure regulator is installed the well produces 40° API oil, its specific gravity is
on the casinghead, as illustrated in figure 6.6, and the 0.83 so the average fraction of liquid in the gaseous
casing pressure is allowed to increase and stabilize to column is computed as 0.095 ÷ (0.83 × 0.433) = 0.26,
a value of 250 psi by regulating the gas flow from the or 26% liquid concentration.
wellhead to the flow line. Correspondingly, the height Extensive studies of multi-phase flow in pipes and
of the gaseous column is depressed to a total of 200 annuli have shown that the fraction of liquid (liquid hold-
feet above the formation and the PBHP returns to a up) present at a given point in the wellbore is inversely
stabilized value of 350 psig for the stabilized flow rate related to the superficial velocity of the flowing gas (in
of 70 bbl/day. It can be seen that the casing pressure situ gas flow rate divided by cross sectional area). For a
increase of 220 psi has resulted in a reduction of 2,300 given cross-sectional area of the pipe or annulus, as the
feet of the height of the gaseous liquid column. In other gas flow increases, the liquid percentage decreases12–16.
• Attach BACK-PRESSURE VALVE and GAS

GUN to the TEE attached to CASING VALVE. Back-Pressure Test Setup
• Connect the BACK-PRESSURE VALVE outlet to
the FLOW TEE with the HIGH PRESSURE HOSE.
• Adjust BACK-PRESSURE VALVE to desired
pressure. POLISHED ROD
• Check frequently during depression to avoid
pushing gas into the pump.
• To ensure a stable condition, the casing pressure
and the liquid level must be unchanging.
GAS OIL AND GAS
VENT GAS THROUGH TUBING

HIGH PRESSURE CASING VALVE AND
HOSE BACK-PRESSURE
VALVE TO TUBING
GAS
GUN
CLOSE VALVE
TO FLOW LINE
“T”
ADJUST SPRING
TO SET CASING
PRESSURE BACK-PRESSURE VALVE
Figure 6.6 Wellhead arrangement for performing a liquid level depression test

Procedure for Walker Test Acquisition The time required for stabilization depends on the
and Analysis well’s production rate and particularly the annular gas
A properly performed Walker test yields the most ac- flow rate. The larger the gas flow, the shorter the stabi-
curate estimate of the gaseous liquid column gradient, lization time. The record in figure 6.7 corresponds to a
but requires installation of a back-pressure regulating depression test performed over four days. Each plotted
valve at the casinghead and usually takes several days point in the upper graph corresponds to a reading of
to complete. Measurements should be made at the casing pressure, and points in the lower graph show
highest fluid level first, followed by the progressively fluid level depth plotted as a function of time in hours.
lower fluid levels5. Measurements were made at three stabilized values of
Figure 6.6 illustrates a possible modification of casing pressure: 80, 180, and 250 psi, as shown in the top
the wellhead flow path for performing the depression panel. The corresponding fluid level behavior is shown
test while simultaneously acquiring fluid level and in the lower panel, where fluid level drops from 2,150
casing pressure data. The valve from the casing to the to 3,200 feet and then to about 4,000 feet. For this well,
flow line that is usually used to vent the gas from the the fluid level was depressed to a near-constant depth,
casing is closed. The gas flow is now directed through and the casing pressure was stabilized relatively quickly
the opposite casing valve connected to a tee fitting. (from point 2 to 3 and from 4 to 5), since the annular
Connected to the tee are two branches: gas flow rate was fairly high at about 150 Mscf/day17.
The figure also indicates that, once pressure stabi-
• An acoustic fluid level instrument is installed
lization is achieved, the fluid level and casing pressure
on one branch. The instrument is preferably one
remained fairly constant, as seen in data sections from
that can be programmed to automatically acquire
1 to 2, from 3 to 4, and from 5 to 6. The data in these
acoustic records according to a preset schedule.
stabilized sections are grouped very tightly when the
• A back-pressure regulator is connected to the height of the gaseous column (relative to the pump in-
other branch. The outlet of the regulator is take) is plotted as a function of the pressure at the gas/
connected to the sampling valve on the pump- liquid interface in figure 6.8. The dots between the groups
ing tee. Thus, the annular gas is mixed with correspond to the transient periods when the liquid level
the tubing flow stream and produced into the is moving down and the casing pressure is increasing.
production line. The slope of the line joining the groups of points is the
At the start of the test, it should be verified that inverse of the gradient of the gas/liquid mixture in the
the pumping system has been operating at a constant gaseous column, which is estimated as 0.0962 psi/ft.
displacement rate so that the fluid level and casing pres- If this line were extrapolated to zero height, it would
sures are stabilized. The following is a recommended intercept the horizontal axis at the pressure correspond-
procedure: ing to the pump intake depth. Alternatively, knowing
1. Maintain the well at normal pumping conditions. that for the last data group the annular pressure at the
top of the column at point 6 is 278 psi and the gaseous
2. Measure casing pressure and liquid level depth.
column height is 2,170 feet, the pump intake pressure
3. Increase casing pressure with back-pressure can be computed as:
regulator.
4. Wait a sufficient time for the fluid level and PIP = 278 + (2,170 × 0.0962) = 486.8 psi
casing pressure to stabilize.
If the pump intake were located at or near the depth
5. Verify the stabilization and record the new liquid of the producing formation, this pressure would be
level and casing pressure. equivalent to the PBHP. On the other hand, if the pump
6. Repeat steps 3 to 5 to obtain at least three pairs were set high above the producing zone (as illustrated
of casing pressure and fluid level values. in figure 6.4), calculating the PBHP would necessitate

275.0
Typical Test Data Points

CASING PRESSURE, psig
5 6
225.0
SPE 14254
175.0
3 4
125.0
1 2
75.0 Casing Pressure and Liquid
0 31 63 94 125 Level During Liquid Level
ELAPSED TIME, hours Depression Test
2,000
2 The pumping rate is kept constant
1 during the test.
2,563
LIQUID LEVEL, ft
3,125
4
3
3,680
5
6
4,250
0 31 63 94 125
ELAPSED TIME, hours
Figure 6.7 Typical fluid level and casing pressure recorded versus time during the Walker test
estimating the gradient of the gas/liquid mixture in the However, two questions are often raised with regard
wellbore section below the pump intake. This is because, to the liquid level depression test:
in this section, the cross-sectional area of the wellbore • Does the downhole pressure remain constant
is larger than the annular area between the casing and during the test while the casinghead pressure
tubing. Thus, the liquid hold-up (liquid fraction) would changes?
be higher and the gradient would also be larger. A multi- • Is the gradient of the gaseous liquid column
phase flow correlation would be required to obtain a constant independently of its height and cas-
good estimate of the PBHP12. inghead pressure?
The liquid level depression test is therefore a very
powerful tool for determining the effective gradient of The following two data sets attempt to provide
the annular gas/liquid mixture and for calculating the some evidence that these values remain constant dur-
pump intake and PBHPs. It can be performed without ing the test.
having to introduce downhole instruments in the well- Table 6.1 summarizes the results of a Walker test
bore or disturb production in other ways. undertaken in a pumping well where a BHP sensor was

Gaseous Column Height vs. Casing Pressure

for 150 MCF/D in 5" and 2⅞"
4,250
1–2
3,688 GRADIENT = 0.0962 PSI/FT
HEIGHT, ft
3,125
3–4
2,563
5–6
2,000
75 131 188 244 300
PRESSURE, psig
Figure 6.8 Height of the gaseous liquid column versus pressure at the gas/liquid interface during a liquid
depression test
Table 6.1
Results of Pumping Well Test with Differing Casinghead Pressures
Gaseous Column
Casing Liquid Level Pressure Sensor Height Above
Date Pressure (psi) Depth (ft) Reading (psi) Sensor (ft)
November 18 0 2,833 145.2 435
November 23 29 2,924 155 344
November 24 17 2,886 153.7 382
November 27 26 2,914 155 354
November 29 24 2,908 154.7 360
December 6 110 3,174 157.4 94
December 9 108 3,168 158.4 100

used to measure the pump intake pressure for different axis intercept of the best fit line for zero height (at the
casinghead pressures. depth of the pressure sensor) is equal to 157 psi, very
The recorded data points are plotted in figure close to the pressure value from the sensor readout of
6.9, where the blue curve represents the height of the 157.4 psi. This result indicates that, as long as the test
gaseous liquid column and the red curve represents is performed correctly and the flow conditions in the
the simultaneous readings of the downhole pressure wells are stabilized, the BHP remains constant.
sensor plotted as a function of the pressure at the gas/ The second data set corresponds to a liquid level
liquid interface depth. Note that, as the casing pressure depression test performed by taking a larger number
increased from 29 psi to the maximum of 110 psi, the of fluid level measurements as the casing pressure was
downhole pressure measured near the pump intake increased from 150 to 550 psi. The acoustic records in
only changed by 3.4 psi. During the test, the electrical figure 6.11 demonstrate the downward progression of
submersible pump was constantly operated at 47 Hz the fluid level echo as the casing pressure increases.
and 30 amps. For each of the 45 acoustic records, the liquid level
After re-graphing the annular pressure and gaseous depth is converted to gaseous column height above the
column height data and fitting a linear correlation pump intake and plotted as individual points, with the
as shown in figure 6.10, the gradient of the gaseous corresponding pressures at the casinghead and at the
column is shown to be 0.3578 psi/ft. The pressure gas/liquid interface in figure 6.12. Minor variations of
Measured BHP and Height of Gaseous Column

as Casing Pressure Increases
530
BHP (psi) OR HEIGHT OF GASEOUS LIQUID, ft
450
430
350 G/L INTERFACE BHP SENSOR
330
250
230
153.7 157.4
150
130
50
0
0 20 40 60 80 100 120 140
GAS-LIQUID INTERFACE PRESSURE, psi
Figure 6.9 Measured downhole pressure and gaseous liquid column height for varying annular pressures

PBHP and Gradient From Walker Test
157 psi
140
120
y = 0.3578x + 157
PRESSURE AT G/L/INTERFACE, psi
R2 = 1
100
80 PBHP FROM WIRELINE

RECORDER = 157.4 psi
GRADIENT = 0.3578 psi/ft
60 PBHP = 157 psi
40
20
PRESSURE AT G/L LINEAR PRESSURE AT G/L
0
0 50 100 150 200 250 300 350 400 450 500
HEIGHT OF GASEOUS COLUMN, ft
Figure 6.10 Estimate of the gaseous column gradient from a liquid level depression test
the fluid level are noticed, especially at the beginning and can be used to estimate representative values of
of the test, and are attributed to a lack of full stabiliza- pump intake pressure and PBHP. Nevertheless, this test
tion of the flow. A linear fit of the data points yields a is seldom performed because of the logistical require-
gaseous column gradient of 0.242 psi/ft with a correla- ments and the time that has to be devoted to periodic
tion coefficient (R2) of 0.9929, which indicates that the monitoring of the progress of the test. To overcome
assumption of a constant pressure gradient throughout these difficulties, a number of correlations have been
the gaseous liquid column is valid within the bounds of developed to estimate the gaseous column gradient from
the fluid level measurement uncertainties. This data set measurements of surface pressure and gas flow rate
is representative of the fluid level and pressure behavior that are taken in conjunction with routine acoustic fluid
observed during extensive series of liquid level depres- level records. In addition, some mechanistic models of
sion tests performed in 11 wells18. annular flow have been developed, but are not widely
Based on these field results and the experience of used because of their complexity and the difficulty in
numerous operators, it is now accepted that the Walker obtaining the physical and chemical properties of the
liquid level depression test yields representative values produced fluids that are required for input to the math-
of the effective gradient of the gaseous liquid column ematical models16,19.

Sec 0 1 2 3 4 5 6 7 8 9 10 11 12
LL = 607 ft
Sequential Acoustic
CP = 142 psi
Data From LL
Depression Test
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12
LL = 1,210 ft
1 DAY LATER
CP = 256 psi
Sec 0 1 3 4 5 6 7 8 9 10 11 12
LL = 1,628 ft 2 DAYS FROM

CP = 350 psi BEGINNING OF TEST
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12
LL = 1,981 ft
3 DAYS FROM
CP = 442 psi BEGINNING OF TEST
Sec 0 1 2 4 5 6 7 8 9 10 11 12
7 DAYS FROM
BEGINNING OF TEST—
LL = 2,173 ft
FLUID LEVEL 164 FEET
CP = 495 psi ABOVE PUMP
GAS/LIQUID INTERFACE DISPLACED FROM 607 TO 2,173 FT (1,566 FT IN 7 DAYS)
Figure 6.11 Partial sequence of acoustic fluid level records for tests analyzed in figure 6.12

CASINGHEAD PRESSURE STABILIZED PRESSURES AT

LINEAR (STABILIZED PRESSURES GAS/LIQUID INTERFACE
AT GAS/LIQUID INTERFACE) ANNULAR GAS COLUMN GRADIENT
2,500
GAS COLUMN
2,000
HEIGHT OF GASEOUS LIQUID, ft
1,500
1,000
CONSTANT DURING
TEST PUMP INTAKE
PRESSURE = 575 psi
500
GRADIENT BEST FIT LINE OF MEASURED PRESSURES
AND LEVELS: PRESSURE (psig) = 574.8 -0.242
(HEIGHT OF GASEOUS LIQUID)
0
0 100 200 300 400 500 600 700
PRESSURE AT GAS/LIQUID INTERFACE, Psig
Figure 6.12 Analysis of the liquid level depression test
Correlations for Determining the Gaseous obtained at representative field conditions, was neces-
Liquid Column Gradient sary to improve the calculation of BHP.
An empirical correlation for determining the gradient To this end, an extensive project was undertaken
of the gaseous liquid column was presented at the 1972 that involved performing accurate liquid level depres-
Southwest Petroleum Short Course (SWPSC)20 and sion tests in numerous pumping wells for a broad set of
attributed to W.E. Gilbert with permission from Shell operating parameters17 so that an empirical field-based
Oil Company. The purpose of the correlation is to cal- correlation could be developed. The wells tested included
culate a gradient correction factor that, multiplied by casing sizes from 4.5 to 7 inches and oil gravities between
the gradient of the liquid present in the mixture, yields 32 and 43° API. Long gaseous columns of more than
the gradient of the gas/liquid mixture. Applying this 5,000 feet were studied in wells up to 9,000 feet deep.
correlation to field cases indicated that the predicted Annular gas flow rates ranged from 13 to 150 Mscf/day,
gradient appeared to be larger than the actual gradient, which resulted in oil fractions in the gaseous column
as observed from pumping well performance analysis ranging from 20% to 77%. The data set includes 17
of dynamometer records. Alternate correlations were wells primarily located in the Midwestern United States
introduced14,21 to try to overcome this deficiency, but whose earth temperature gradient varies between 1.2
because all the published correlations had been devel- and 1.7° per 100 feet.
oped primarily from laboratory tests of gas/liquid flow The correlating parameter (Ycor) that was chosen
in relatively short columns (30 to 100 feet), it became for practical reasons was the gas flow rate per unit an-
apparent that additional gaseous column gradient data, nular cross-sectional area (Q ÷ A in SCF/day per square

inch). This parameter was chosen among other possible (such as the one represented in figure 6.4), the correlat-
parameters because of its simplicity and the fact that the ing parameter defined as Q ÷ A should be used for each
quantities can be easily determined in most wells from section of the wellbore to obtain the corresponding
field measurement and well completion information. In gradient correction factor. The field-based correlation
wells with a uniform cross-sectional area (with uniform is presented in figure 6.13 with double vertical axes that
casing and tubing sizes), this parameter is equivalent to account for both parameters.
the change in wellhead pressure per unit time multiplied For the liquid depression test shown in figures 6.7
by the length of the gas column from the surface to the and 6.8 performed in a well completed with 5-inch cas-
liquid level, or Da × dp/dt, where Da is the distance to ing and 2M-inch tubing with annular gas flowing at a
the liquid level and dp/dt is the slope (psi/min) of the rate of 150 Mscf/day, the annular cross sectional area
wellhead pressure buildup line obtained during the ac- is equal to 12.31 square inches. Therefore, the correlat-
quisition of the acoustic fluid level record, as described ing parameter Ycor is equal to 12,185 feet3/day/in2 and is
later in this chapter. In wells with variable cross sections indicated as the point circled in figure 6.13. The effective
fg, FRACTION OF GAS IN

GASEOUS LIQUID COLUMN
1.0 .9 .8 .7 .6 .5 .4 .3 .2 .1 0
100,000
50,000
40,000
50,000
30,000
40,000
30,000 20,000
20,000
10,000
10,000
5,000
4,000
dp Q
Da x 5,000
3,000
dt 4,000 A
3,000 2,000
psi-ft 2,000 Scf / D

min in.2
1,000
500
400
500
300
400
300 200
200
100
100
0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1.0
f0, EFFECTIVE OIL FRACTION
Figure 6.13 Effective oil fraction correlation from liquid level depression tests

fraction of oil in the gaseous liquid column is computed Applying the correlation in figure 6.13 to wells
from the ratio of the mixture gradient = 0.0962 psi/ft producing heavy oil (10 to 25° API) and outfitted with
and the oil gradient (density) = 0.437 psi/ft at wellbore BHP sensors indicated that the computed oil fraction
conditions, which yields an effective oil fraction of 0.22 was too low, which resulted in computing a pump intake
as plotted in the figure. pressure that was lower than the measured value. The
The measured gradient of the gaseous column in- difference was primarily attributed to the high viscosity
cludes the hydrostatic component (mixture density) and of the heavy crude. This prompted undertaking additional
the dynamic components of friction and kinetic energy field measurements during liquid level depression tests.
change. The friction component is primarily depen- Figure 6.14 summarizes the results obtained from 17
dent on the gas rate because the flow in the annulus is wells in western Venezuela, showing the effect of the
characterized as zero net liquid flow, while the kinetic viscosity of the oil on the effective liquid fraction18.
energy losses are dependent on the countercurrent mo- A similar trend has been reported from more ex-
tion and turbulence of the liquid. These two terms are tensive tests recently undertaken in the Boscan field in
not independently quantifiable, so it is not possible to wells producing oil from 9.2 to 13.8° API; the study has
clearly identify only the hydrostatic portion of the total yet to be published25. Tests were undertaken in wells
gradient with the purpose of defining an in situ liquid with progressing cavity, electrical submersible, and rod
fraction. However, based on published literature, these pump systems. Surface readouts of downhole pressure
two terms are not considered significant for the annular sensor data were used in a large number of the tests to
flow pattern conditions23, 24 found in most pumping wells. obtain the liquid fractions that are represented in the
The “effective” oil fraction fo is defined with the expanded correlation.
primary purpose of establishing a correlation between the
measured mixture gradient and the annular gas flow rate. Annular Gas Flow Rate Determination
gmt The correlating parameter for the “S” curves and other
fo = — Eq. 6.3
go commonly used correlations involves the annular gas
where: flow rate. Flow meters were used during the liquid level
depression tests experimental fieldwork, but in routine
fo = effective liquid fraction in gaseous oil
acoustic liquid level measurements, this flow rate is
column
computed using a short-term pressure monitoring test in
gmt = measured total gradient (psi/ft)
parallel with performing the acoustic fluid level test after
go = gradient of gas-free oil (psi/ft)
closing the casing valve. The flow of free gas from the
This term is primarily determined by the quantity of perforations does not stop when the casing to flow line
gas percolating through the annular liquid and thus can valve is closed and the flow of gas from the casing at the
be correlated with the annular gas flow by introducing surface is interrupted. Gas accumulates in the annular
a parameter related to the gas velocity, such as the gas volume, causing an increase in casing pressure. The rate
flow rate per unit cross-sectional area of the conduit, the of casing (or tubing) pressure increase (psi/min) is directly
in situ superficial gas velocity, or some combination of proportional to gas inflow rate and inversely proportional
variables such as Q ÷ AP0.4, as was proposed by Gilbert10. to the volume occupied by gas. Figure 6.15 illustrates the
The use of a correlation to perform pressure esti- initial and final pressure conditions and includes a plot of
mates implies that the flow conditions in a particular the casing pressure increase as a function of time.
well are similar to those in the wells used to establish the At the beginning of the test (time t1 = 0), the surface
correlation. When the well’s characteristics are outside pressure is 46.2 psi, and after 4 minutes of shut-in, at
the range of parameters represented by this correlation, time t2, it increases to 49.4 psi.
then the error in the computed effective liquid fraction Equation 6.4 can be used to compute the accumu-
becomes larger. lation of gas. It applies the real gas law and considers

“S” Curve for Heavy Oil Compared with the Echometer “West Texas Wells S Curve”
100,000
10,000
Q/A SCFD/in.2
HEAVY OIL
WEST TEXAS OIL

1,000
100
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
EFFECTIVE OIL FRACTION
Figure 6.14 Comparison of the effective oil fraction determined from liquid level depression tests in Venezuelan
heavy oil (10 to 11° API) wells and western Texas (32 to 43° API) wells
that the volume occupied by gas (which is the wellbore The resulting number of lb.-moles is converted to
volume less the volume occupied by liquid) does not gas volume at standard conditions by recalling that there
change and that the temperature and the compress- are approximately 0.791 scf (22.4 litres) of volume per
ibility factor (z) remain constant during the short time mole at standard conditions and there are 454 moles
of the test. per lb-mole. The gas flow rate is obtained by dividing
Vg the accumulated volume by the time duration of the test
n2 – n1 = (P2 – P1) —– Eq. 6.4
zRT (t2 - t1), yielding the following relation:
(p2 – p1) Vg
where: qg = 0.00068 ————– Eq. 6.5
(t2 – t1)
n2 = number of lb.-moles of gas present in the
wellbore at the end of the test (t2) where:
n1 = number of lb.-moles of gas present in the qg = gas flow rate (Mscf/day)
wellbore at the start of the test (t1) t1, t2 = time (minutes)
Vg = volume of the wellbore occupied by gas (feet3) p1, p2 = pressure (psi)
R = universal gas constant (10.73 feet3-psi/oR Vg = wellbore volume occupied by gas (ft3)
lb-mole)
z = gas deviation factor at the average pressure Because the pressure and time step changes can
and temperature of the wellbore gas be measured very accurately, the computed gas flow


Casing Valve Closed During Test

TIME = 0 TIME = 4 min.
GAS FLOW OUT = 45 MCF/D
46.2 Annular
49.4
psi Casing Pressure Buildup psi
Gas Flow
3.2 psi
45 Mscf/D
4.00 min
% Liquid
37
4.0
DELTA PRESSURE (psi)
49.6
CASING PRESSURE (psig)

3.2
48.8
2.4
48.0
1.6
47.2
0.8
46.4
0
45.6
-0.8
0 4.0
DELTA TIME (min)
GAS FLOW INTO WELL = 45 MCF/D
PBHP = 572.8 psi
Figure 6.15 Annular pressure increase during the acoustic fluid level acquisition used to estimate the gas inflow rate
rate can be very accurate and fully represent the actual each section must be computed considering the specific
instantaneous gas flow in the annulus. This was veri- internal and external diameters. Then the liquid fraction
fied during the numerous liquid level depression field in each section must be estimated by using a correlation
tests by means of gas flow meters attached directly to that requires knowing the gas flow rate, which is yet to
the gas exhaust line. However, the accuracy of the flow be estimated. Thus, multiple sets of iterations must be
rate calculation also depends on the estimation of the performed until convergence is achieved in all sections
volume occupied by the gas phase. This implies that it of the well.
is necessary to fully describe the wellbore geometry The problem is complicated further in wells with
and the piping used for the artificial lift installation. The long horizontal or lateral wellbores, because it is not
wellbore geometry was known accurately for the wells possible to estimate the fraction of liquid and gas present
tested for the purpose of developing the “S” curve cor- in the lateral section with a degree of certainty. At the
relation. Unfortunately, wellbore geometry information present time, there is insufficient field data acquired in
is incomplete or inaccurate for many wells. these complex-geometry wells to provide guidance or
The uncertainty inherent in the wellbore volume recommendations for estimating the uncertainty in the
calculation becomes more significant as the wellbore calculated wellbore pressure distribution. The following
geometry becomes more complicated. In a multi-section section refers primarily to conditions existing in vertical
well, such as the one shown in figure 6.4, the volume of and slightly deviated wellbores.

QUALITY CONTROL OF PRESSURE fluid level is steady and the oil/water interface in the
CALCULATIONS annulus is at the pump intake. When this condition is
not satisfied, the computed PIP underestimates the actual
In numerous vertical wells, it has been observed that
value. The uncertainty in PIP and PBHP increases as
the BHP computed from accurate acoustic fluid level
the difference in density between the produced oil and
records is within a few percent (3% to 5%) of the values
produced water increases. Table 6.2 illustrates the effect
obtained from downhole pressure sensors. However, in
of assuming a stabilized condition, while in reality the
some cases, significant differences have been observed
oil/water interface is halfway up the annular column
and are generally attributed to inaccurate data acquisi-
above the pump intake, as a function of the oil API
tion procedures or incomplete information about the
gravity. With regard to the information in the table, the
wellbore conditions. The accuracy of the computed
following parameters are constant:
pressures depends upon many factors, each affecting
• Gas-free liquid height = 587 feet
the final result with differing magnitude. This section
• Casinghead pressure = 46.1 psi
addresses some of the main factors that influence the
• Specific gravity of water = 1.05
accuracy of the computed pressure.
As the gravity of the oil decreases from 40 to
Production Stabilization 20° API, the difference between the PIP computed by
The algorithm used to compute the pressures below the assuming the well is stabilized and the actual PIP for
gas/liquid interface assumes that flow stabilization, as the correct position of the oil/water interface decreases
described in figure 6.2, has been achieved so that the from -11.4% to -5.6%.
Table 6.2
Effects of Unstabilized Flow on Computed PIP
Difference
API Gravity PIP @ (computed minus % Difference
Condition of Oil 5,226 feet (psi) actual [psi]) From Actual
Computed PIP, Assuming

Stabilized 42 252.7 — —
Actual PIP, with Oil/Water
Interface at ½ of Annular
Liquid Height 42 285.1 –32.4 –11.4

Liquid Height 30 294.0 –25.5 –8.67

Liquid Height 20 302.6 –17.2 –5.68

Percentage of Liquid in the Annular Gaseous casinghead pressure is zero. If the gaseous column height
Column were only 100 feet, the error would be only 2.5%. The
The gas/liquid fraction that exists in the annulus of percent error in PIP due to the uncertainty of the gaseous
vertical wells can be determined accurately using column gradient decreases as the casinghead pressure
the Walker test11. However, because of logistical and increases and the fluid level is depressed, as discussed
time constraints, the Walker test is seldom performed. in more detail in chapter 7.
Therefore, the liquid fraction must be calculated using These considerations show that the uncertainty in
a correlation that relates the well conditions (pressure, BHP computed from a fluid level measurement decreases
temperature, gas flow, fluid properties, and so on) to the significantly when the gaseous liquid column is short
gas/liquid fraction, as described in the previous section. and the casinghead pressure is high in relation to the
All the existing published correlations have been de- gaseous column pressure.
termined from tests in vertical wellbores. The effect of In deviated wellbores, it is important to consider
wellbore inclination on the distribution and concentration the correct well deviation survey data, since all pressure
of gas and liquid in pumping wells has only been studied calculations are based on TVD, while the acoustic survey
at laboratory conditions15 and has not been incorporated yields the measured depth of the fluid level. In addition,
in the existing correlations. Additional measurements of the accuracy of the gaseous liquid gradient correlation
liquid level depression tests in inclined and horizontal becomes more uncertain, as discussed previously.
wells should be undertaken in the field. In wells with horizontal or lateral extensions, the
inaccuracy of the bottomhole pressure calculation has
Fluid Level Depth not been quantified at this time.
The distance to the gas/liquid interface generally can
be determined acoustically within 15 feet (H pipe joint) Calculating SBHP in Pumping Wells
or better. Assuming that the oil is 30° API and that Whenever the pumping system fails as a result of me-
there is 50% liquid in the gaseous fluid, this distance chanical problems or other reasons and stays inoperable
is equivalent to 0.38 psi/ft × 15 feet × 0.5, or ± 2.8 psi. for an extended period of time, the fluids produced from
The relevance of this quantity must be considered in the formation accumulate in the wellbore and eventually
relation to the existing casinghead pressure. The greater exert sufficient back-pressure to nearly equilibrate the
the casinghead pressure, the lower the percent error information pressure that exists at the drainage radius of
troduced in the pressure calculations by the uncertainty the particular well. When equilibrium has been reached
in fluid level depth. and all inflow from the formation has ceased, the SBHP
is equivalent to the stabilized reservoir pressure26. When
Height of the Gaseous Liquid Column full equilibrium has not been reached, the longer the
When the height of the gaseous liquid column is consid- shutdown time is, the closer the SBHP will be to the
erable, for the purpose of computing PIP and PBHP, the reservoir pressure.
effect of the uncertainty on the percentage of liquid in Figure 6.16 shows the distribution of fluids in a static
the gaseous column (discussed in the previous section) well in which a series of fluid level surveys, each taken
is amplified. As the height of the gaseous column in- about once every 24 hours, have determined that the
creases, the uncertainty in computed pressures increases. position of the gas/liquid interface and the casinghead
In the previous example, assuming that the correct liquid pressure are no longer changing in time (or are close to
percentage is 40% (instead of 50%), the difference in becoming constant).
fluid gradient would be 0.38 × 0.1, or 0.038 psi/ft. For a The same balance of pressures performed using
gaseous column 1,000 feet in height, the computed PIP figure 6.1 is used to calculate the pressure distribution
could be too high by 38 psi compared with the actual in the wellbore, and the SBHP in particular. In a shut-in
PIP of 152 psi, yielding an error of 25%, assuming the well, the oil/brine interface cannot be assumed to be at

Calculation of SBHP Pt WELL SHUT-IN
Pc
STATIC BHP =
CASING PRESSURE +
GAS COLUMN PRESSURE +
OIL COLUMN PRESSURE + GAS
BRINE COLUMN PRESSURE.

STATIC FLUID LEVEL
OIL
NOTE:
All flow from perforations has stopped
fluids segregated by gravity. Position of OIL/BRINE INTERFACE
oil/brine interface above pump intake
must be computed.
BRINE
SBHP
Figure 6.16 Pressure balance and fluid distribution in a static well
the depth of the pump intake, as it is when the well is Prior to shut-in, there was 500 feet of gas-free oil
producing at a stabilized rate. The depth of the static above the tubing intake, as determined from the last
oil/water interface in the annulus has to be calculated recorded fluid level survey. Assuming that, during the
from the produced water/oil ratio and the gas-free liquid shut-in period, the volume of liquid equivalent to the
present above the pump before the pump was shut in or 1,000 feet of gas-free fluid has entered the wellbore at
the lift system failed. a water/oil fraction of 50%, the fraction of oil present
The fluids produced during the after-flow period ac- in the annulus is computed as 900 ÷ 1,500, or 0.6, and
cumulate in the wellbore and segregate based on the dif- the fraction of water is 0.4. With these numbers, the
ferences in their densities. In figure 6.17, the total height pressure due to each column of liquid can be calculated
of the liquid in the annulus of the well at equilibrium using the corresponding gradients of the fluids.
is 1,500 feet, and the casing pressure has increased to This method of computing the SBHP from acoustic
15 psi, as measured by the fluid level survey. Since at this measurements has been determined by most operators
point the casinghead pressure is constant—specifically, and several regulatory agencies to be sufficiently ac-
the value of dp/dt remains nearly zero during the usual curate for the purpose of determining well potentials
2 to 4 minutes of recording—the flow of free gas from and meeting report-filing requirements6. A typical static
the formation has ceased and only liquid is present in pressure survey report is shown in figure 2.23.
the wellbore. The problem is to correctly account for the The assumption that the after-flow water cut remains
composition of the liquid column in terms of fractions constant is a simplification, since it has been observed in
of oil and water to compute an accurate BHP. field tests that over time the water inflow decreases faster

However, in most pumping well cases, pressure

Producing Well Static Well transient tests are seldom performed. Thus, approxi-
40% OIL mate static pressures may be determined only when the
60% WATER pumping system fails. Then only one static fluid level
survey is performed at the end of several days, before
the well workover procedure is initiated. Therefore, the
CASING CASING assumption of a constant after-flow water/oil ratio has to
PRESSURE PRESSURE
IS 0 psi IS 15 psi be used, as illustrated in the SBHP example in chapter 2.
(assumed) (assumed)
SUMMARY
GAS COLUMN To obtain a best estimate of PIP or PBHP from fluid
PRESSURE =
2 psi (approx.) level measurements, the following guidelines should be
considered:
500' OF 500' OF
OIL OIL 1. Check that the well production is stabilized by
900' OF OIL
= 300 psi monitoring casing pressure and fluid level over an
(approx.)
400' OF extended period of time until their values remain
OIL
1,000' fairly constant.
OF
FILL-UP 2. Verify that the wellbore directional information is
600' OF WATER =
500' OF
OIL
300 psi (approx.) correct, so that calculating the pressure uses the
proper TVD values.
WATER OIL 3. Verify that the correct values of oil API gravity,
BHP •
617 psi
WATER
brine specific gravity, bottomhole temperature, and
surface temperature are used in the calculations.
Figure 6.17 Oil/water distribution for static well 4. Verify that the correct values for the inside diameter
conditions of the casing and the outside diameter of the tub-
ing are entered in the well file in order to compute
than the oil flow. This suggests that an improvement in accurate values of annular gas flow rate.
the calculations could be made by using a productivity 5. Verify that the correct average joint length is used
index relation for the water and a Vogel-type inflow in calculating the fluid level depth.
performance relation for the oil7. This type of mixed flow 6. Be aware that the value of the annular liquid percent-
calculation can be applied as long as several fluid level age is obtained from a generalized correlation that is
surveys are taken during the pressure buildup period and applicable mainly for a certain range of parameters.
intermediate PBHPs are computed. Then for each time 7. If a high gaseous column (more than one thousand
interval, an after-flow rate and water cut can be computed feet) is present when gas is flowing up the annulus,
and each liquid volume accounted for properly. This is minimize the height of the gaseous fluid column by
the analysis procedure adopted when performing an closing the casing valve to the flow line, allowing
automatic acoustic pressure buildup test27. In this test, the casinghead pressure to increase. Then depress
numerous fluid level records are acquired using a pre- the fluid level to a few hundred feet above the
established schedule of fluid level shots, as discussed pump intake. Use a back-pressure regulating valve
in detail in chapter 7. Liquid and gas after-flow are then to increase casinghead pressure, stabilize the fluid
calculated for each time interval between subsequent level and casinghead pressure, and maintain the well
fluid level records and for incremental pressure change production including the annular gas flow rate at a
determined as a function of time. constant value.

Past experience has shown that, by following these 2. D. G. Hawthorn, “Subsurface Pressures in Oil Wells
recommendations, the acoustic fluid level survey in ver- and Their Field of Application,” API Ponca City
tical wells can yield values of pressure that are within Meeting, 1932.
5% of those obtained with downhole pressure recorders. 3. K. Huddleston, J. N. McCoy, A. L. Podio, and S.
To obtain a best estimate of the SBHP in pumping Weeks, “Analyzing Well Performance from Acous-
wells from acoustic surveys, the following guidelines tic Surveys,” Petroleum Society of Canada Annual
should be considered: Technical Meeting, 1986.
1. After the pumping system is stopped (or suffers 4. B. P. Kantzer and E. G. Trostel, “Oil Well Perfor-
mechanical failure), close in the well completely mance, Discussion and Proposed Terminology,”
by closing all surface flow paths. American Petroleum Institute, 1937.
2. Check that the inflow of fluids from the reservoir
5. J. J. Jakosky, “Bottom-hole Measurement in Pump-
has ceased by monitoring casing pressure and fluid
ing Wells,” Transactions of AIME, 1939.
level over an extended period of time until the pres-
sure change and fluid level change per unit time are 6. Directive 005: Calculating Subsurface Pressure via
near zero. Fluid-Level Recorders, Alberta Energy Regulator,
3. Verify that the wellbore directional information 1978.
is correct so that calculating the pressure uses the 7. A. D. Laird and R. F. Dupuis, “A New Method to
proper TVD values. Improve the Accuracy of Acoustic Pressure Calcula-
4. Verify that correct values of oil API gravity, brine tions,” Journal of Canadian Petroleum Technology,
specific gravity, bottomhole temperature, and surface 1990.
temperature are used in the calculations. 8. L. K. Thomas, R. W. Hankinson, and K. A. Phillips,
5. Verify that the correct values for the inside diameter “Determination of Acoustic Velocity for Natural
of the casing and the outside diameter of the tubing Gas,” Journal of Petroleum Technology, July 1970.
are entered in the well file in order to compute ac-
9. National Institute of Standards and Technology,
curate values of annular gas flow rate.
Thermophysical Properties of Hydrocarbon Mix-
6. Verify that the correct average joint length is used
tures Database: Version 3.2.
in calculating the fluid level depth.
7. When the wellbore pressures have stabilized, acquire 10. W. E. Gilbert, “Notes on Gradients Resulting from
a final fluid level record and annotate it as a “Static Gas Rising through Non-Flowing Liquid Columns,”
Acoustic Pressure Survey.” Laboratory Study, 1955.
8. Verify that the latest water/oil well test information 11. C. P. Walker, “Method of Determining Fluid Density,
is used in the calculations. Fluid Pressure and the Production Capacity of Oil
9. Include in the final BHP calculations the effect of Wells,” U.S. Patent 2,161,733, June 1939.
the fluid present in the wellbore during stabilized 12. G. W. Govier and K. Aziz, The Flow of Complex
production. Mixtures in Pipes (R. E. Drieger Publishing Co.,
Following these recommendations should result 1977).
in estimates of the SBHP pressure that are within 2% 13. Z. Schmidt, J. P. Brill, and H. D. Beggs, “Prediction
to 3% of the values obtained from wireline-conveyed of Annulus Pressure Gradients in Pumping Wells,”
pressure recorders. Proceedings of the American Society of Mechanical
Engineers Energy-Sources Technology Conference
REFERENCES and Exhibition, 1980.
1. C. P. Walker, “Determination of Fluid Level in Oil 14. J. K. Godbey and C. A. Dimon, “The Automatic
Wells by the Pressure-wave Echo Method,” Trans- Liquid Level Monitor for Pumping Wells,” Journal
actions of AIME, 1937. of Petroleum Technology, August 1977.

15. A. R. Hasan and C. S. Kabir, “Void Fraction in 21. M. J. Tarrillion, “An Empirical Investigation of
Two-Phase Flow in Vertical and Inclined Annuli,” Gradient Correction Factor Correlations for Liquid
International Journal of Multiphase Flow, August Columns Containing Gas Bubbles,” MS Thesis,
1990. University of Texas at Austin, 1978.
16. C. S. Kabir and A. R. Hasan, “Predicting Liquid 22. J. N. McCoy, A. L. Podio, and K. Huddleston,
Gradients in a Pumping Well Annulus,” SPE Pro- “Acoustic Determination of Producing Bottom Hole
duction Engineering, February 1998. Pressure,” SPE Formation Evaluation, September
17. J. N. McCoy, A. L. Podio, and K. Huddleston, 1988.
“Acoustic Determination of Producing Bottom Hole 23. D. Barnea, “Effect of Bubble Shape on Pressure Drop
Pressure,” SPE Formation Evaluation, September Calculations in Vertical Slug Flow,” International
1988. Journal of Multiphase Flow, January 1990.
18. J. N. McCoy, S. Bruning, et al., “Acoustic Determi- 24. A. Ansari, “Comprehensive Mechanistic Model for
nation of Pump Intake Pressure in Wells Producing Upward Two-Phase Flow,” MS Thesis, University
Low-API-Gravity Crude,” SPE Annual Technical of Tulsa, 1988.
Conference and Exhibition, 2007. 25. L. Marchan, “Study of Pressure Gradients in Gaseous
19. D. A. Papadimitriu, “A Mechanistic Model for Liquid Columns,” PetroBoscan, 2013.
Predicting Annulus Bottomhole Pressures for Zero 26. J. N. McCoy, et al., “Acoustic Static Bottom Hole
Net Liquid Flow in Pumping Wells,” MS Thesis, Pressures,” SPE Production Operations Symposium,
University of Tulsa, 1990. 1985.
20. F. W. Gipson and H. W. Swaim, “Designed Beam 27. E. R. Brownscombe, “Afterflows and Buildup Inter-
Pumping,” Proceedings of the 19th Annual South- pretation on Pumping Wells,” Journal of Petroleum
west Petroleum Short Course (SWPSC), 1972. Technology, 1982.

Applied Well Testing (Pressure Transient) 7-1
7
Applied Well Testing for Pressure Transient Data Acquisition
In this chapter:
• Automatic acoustic determination of formation pressure
• Factors that influence pressure transient testing
• Programmed fluid level surveys
• Data acquisition and processing background
• Recommended procedures for optimum quality of recorded data
• Examples of well surveys from the field
Proper reservoir management and production optimi- are produced through artificial lift pumping systems.
zation require up-to-date information about forma- Therefore, it is necessary to monitor and analyze the
tion pressure, permeability, and wellbore skin factor. performance of these systems. The principal tools that
Pressure transient tests using wireline-conveyed or are used in the field to determine indicators that influence
permanently installed surface readout pressure gauges production rates—such as reservoir pressure, formation
are commonly run into flowing wells. However, the permeability, productivity index, pump efficiency, and
presence of artificial lift equipment complicates and skin factor—include:
often precludes the use of wireline-conveyed devices. • Flowing BHP surveys
Thus, conventional pressure transient tests are seldom • Pressure buildup tests
performed in these wells. The result is poor reservoir
• Pressure drawdown tests
and production management.
• Inflow performance analyses
Since the 1980s3,6, the oil and gas industry has relied
on programmable equipment to calculate bottomhole These techniques are widely used in flowing wells and
pressure (BHP) from surface pressure and acoustically some gas lift wells, where the pressure information is
measured liquid level data in pumping wells. Over the easily obtained from wireline-conveyed BHP recorders.
years, advances in electronics, computer software, and In rod-pumped wells, the presence of sucker rods
transducer technology have vastly improved the data essentially precludes the practical, routine, and direct
quality and usability of this equipment. In fact, because measurement of BHP, thus eliminating the single most
the equipment provides real-time data with the quality important parameter for well analysis. Permanent in-
that is necessary for pressure transient analysis, this stallation of surface-indicating BHP gauges is not yet
method is considered to be a reliable11 and cost-effective cost-effective for wells with low production rates. As
way to determine BHP. discussed in chapter 6, one solution to this problem
From a management perspective, the industry’s is to calculate the BHP from the casinghead pressure
primary objective is to achieve maximum production measurement and then determine the annular fluid head
efficiency with minimum engineering and technical from echometric surveys that yield the depth of the gas/
labor. The majority of onshore oilwells in the U.S. liquid interface and the gradient of the annular fluids1,2.

at Austin
PROGRAMMED FLUID LEVEL SURVEYS The potential loss in production is probably the
main cost of performing an acoustic pressure transient
Acoustic fluid level surveys acquired at specific time
test; therefore, it is important to assure the successful
intervals can be used to monitor changes of wellbore
completion of the test by following strict procedural
pressure that occur when the pumping rate is varied. For
guidelines and by verifying data quality and system
example, in wells where the pump is driven by a vari-
reliability while the test is in progress, as described in
able speed drive, the pump rate can be set to a constant
the following sections.
slow speed, after which the production rate is allowed
to stabilize. The acoustic fluid level that is taken at a
specific time is analyzed to yield the producing BHP Automatic Acoustic Fluid Level Survey
for a specific flow rate. Afterward, the pumping speed The advent of portable, battery-powered, digital con-
is increased, fluid level and casing pressure are allowed trollers provided the necessary tools to automate the
to stabilize, and a new acoustic survey is acquired and acquisition of fluid level records over an extended period
analyzed to yield the new stabilized producing BHP of time3. In this instrument, the automatic firing of the
for the new flow rate. The pair of producing pressures acoustic pulse and recording of the time to the liquid
and flow rates can then be used to determine the well’s level echo and calculation of its depth do not require
productivity or inflow performance, as discussed in an operator to be present. Thus, it provides a method of
chapter 1. Using an IPR model and the two data points measuring fluid levels under transient flow conditions
also makes it possible to estimate the static reservoir when taken over long periods of time. This ability to
pressure, which is the pressure computed for a zero record liquid depth while the apparatus is unattended is
flow rate. The validity of this estimated static pressure particularly important in obtaining the buildup character-
is dependent on the applicability of the IPR model (for istics of a marginally producing well in which pressure
example, a PI model or a Vogel model) to the specific buildup must be measured over a long period of time.
well. The casing pressure is also recorded to facilitate the
A more reliable estimate of the static pressure can be determination of BHP in a producing well. This system
obtained when the operator is willing to stop the pumping was updated and expanded for monitoring the position
system and close in the well for an extended period of of the liquid level with the objective of controlling the
time to allow the wellbore pressure to increase until it operation of the pumping system to maintain a desired
balances the reservoir pressure and stops all fluid inflow pump submergence4.
from the formation. Fluid level records can be acquired A microcomputer-based system to automatically
periodically until changes in surface pressure and fluid acquire pressure transient data in pumping wells was
level become insignificant, which is an indication that developed as a hybrid system using analog filtering and
the equilibrium point has been reached. recording of the acoustic signal and digital calculation of
Unfortunately, the time required for stabilization BHP5. Such systems still depended in some measure on
may be several days or weeks, depending on the forma- the operator’s interpretation of the acoustic chart record-
tion and well completion characteristics. The potential ings to verify the computer’s automatic identification of
of a significant loss in production precludes performing the liquid level echo and determination of the average
pressure transient tests on a routine basis in multiple acoustic velocity in the annular gas and the depth to
wells in a given reservoir. From this fact stems the rec- the liquid level. Detailed acoustic records registered
ommendation discussed in chapter 6: Acoustic “static” on a strip chart were saved at periodic intervals. At the
fluid level tests should be performed whenever a well completion of the test, the operator could then analyze
that is produced by artificial lift is shut in because of the records and verify the results provided by the soft-
mechanical failure, so that these tests can yield an ap- ware. The reliability of the fluid level determination by
proximate value of the reservoir pressure and its areal automatic means was significantly improved by applying
distribution within the reservoir. fully digital instrumentation and a microprocessor that

Applied Well Testing for Pressure Transient Data Acquisition 7-3
substantially eliminated the effects of human error and pressure transient data in pumping wells by automati-
adjusted calculations for changing downhole conditions, cally calculating the BHP from echometric surveys of
such as continuous variation of the acoustic velocity, the annular fluid level. This portable system integrated
during the test6. specially designed, high-resolution A/D conversion and
Digital spectral analysis of the acoustic trace was conditioning with advanced signal processing and digital
the principal tool used to differentiate echoes from the filtering techniques8.
liquid surface and other reflections. By processing all This technology makes it possible to determine the
of the recorded reflections utilizing known values, such depth of the fluid level very accurately in most wells,
as the spacing of the tubing collars in the well, the aver- even in the presence of background noise caused by
age pressure pulse velocity of the gas in the well was gaseous liquid columns. Automatic signal generation
determined and used with the pulse travel time of the and recording is undertaken by the software at prede-
liquid surface reflection to calculate its actual location termined, operator-selected frequencies to maximize the
in the well. All information was stored in memory for quality of the pressure transient data. Surface pressure
further processing to accurately provide the location and temperature measurements are used in conjunction
of the liquid surface. The process was automated, and with gas gravity and gas acoustic velocity to determine
acquisition was performed according to a programmed the pressure at the gas/liquid interface and the pressure
schedule containing the following steps: at the sand face. Graphic displays allow the operator
to monitor the progress of the transient test by plotting
1. The present date and time was recorded, and the
in real time the current status of the well, the acoustic
well surface pressure was measured and stored.
signals, and the calculated pressures. Data that has been
2. A pressure pulse was initiated in the well annulus acquired at that point can be plotted as standard pressure
using remotely controlled valves and an external transient analysis graphs, including Horner, MDH, and
pressure source (nitrogen bottle). log-log plots.
3. All acoustic signals were detected and recorded This system has an overwhelming advantage because
for sufficient time to record the echo from the fluid it provides instantaneous information in the field regard-
level. ing the progress of the pressure transient test. Thus, the
4. The acoustic velocity in the wellbore was determined operator is able to decide on the best course of action to
at about 1,000-foot intervals. ensure that the test will yield accurate and complete data.
5. The round trip travel time to the liquid level echo Preliminary analysis of the data completed at the well
was determined. site can be followed up with detailed transient analysis
by exporting the data to other analysis software.
6. All information was stored in memory.
7. The process was then repeated according to a preset
schedule. SPECIAL REQUIREMENTS OF
PROGRAMMED ACOUSTIC DATA
The system operated unattended and did not require
ACQUISITION AND PROCESSING
a skilled, experienced operator. Throughout the test, the
recorded information was transferred to a microcomputer Typical transient tests may last for several days, during
via a serial port, where further processing of the data which time the pressure in the wellbore can change orders
was undertaken. of magnitude (from 1 to several hundred psi). Conse-
The development of powerful laptop computers quently, significant changes in fluid properties and the
and large-scale integrated electronic circuits allowed character of the fluid level echoes will occur during the
field measurement and processing of acoustic signals, test. Special processing and software tools may be needed
as well as real-time conversion of liquid level depth to to correctly identify the liquid level echo, estimate the
BHP7. This technology was applied to the acquisition of acoustic velocity, and calculate an accurate BHP. Figure

7.1 illustrates the changes in acoustic velocity observed pressure can be corrected by utilizing a simultaneous
during a pressure buildup test lasting about seven days. measurement of the temperature and a correlation function
Average acoustic velocity changed from 1,265 ft/s at of the temperature sensitivity of the transducer’s output.
the start to a low of 1,221 ft/s at the end of the test as The variation of casing pressure and temperature
the casinghead pressure increased from 75 to 320 psi. for this test is presented in figure 7.2. Daily tempera-
This implies that calculation of the depth to ture fluctuations experienced by the wellhead and the
the liquid level for each acoustic record (represented attached sensors are recorded via a sensitive thermistor.
by each plotted point in the figure) has to be performed This data must be used to compensate the output of the
using the acoustic velocity at that specific time. Because pressure sensor in order to record a pressure function
of the large number of data points, the process has to that represents the actual variation in pressure without
be automated. Then, the quality of the analysis has to the interfering noise created by temperature oscilla-
be verified by observing the continuity of the velocity tions. Without such a pressure correction function, the
function because there cannot be sharp discontinuities as recorded pressure oscillations would be transmitted to
long as the wellhead pressure variation is also a continu- the computed BHP and would severely affect the analysis
ous function of time, as shown in figure 7.2. of the well’s pressure response.
Maximum accuracy would require a pressure trans-
Surface Pressure Measurements ducer with a full-scale range as close as possible to the
At a certain pressure, the output signal of most pressure maximum wellhead pressure expected at the end of the
transducers varies with temperature. The calculated buildup test. In very harsh environments (for example,
1,270.00
1,260.00
ACCOUSTIC VELOCITY, ft/sec
1,250.00
1,240.00
1,230.00
1,220.00
0 1,562.50 3,125.00 4,687.50 6,250.00 7,812.50 9,375.00 10,937.50
TIME, min.
Figure 7.1 Acoustic velocity variation during a seven-day pressure transient test

350.00 90.00
300.00
PRES. TRANSDUCER TEMPERATURE, deg F

80.00
250.00
70.00
200.00
150.00
60.00
100.00
50.00
50.00
0 40.00
0 1,562.50 3,125.00 4,687.50 6,250.00 7,812.50 9,375.00 10,937.50
TIME, min.
Figure 7.2 Variation of casing pressure (circles) and wellhead temperature (triangles) during the transient test of
figure 7.1
Canada, Siberia, Patagonia, or the tropics), the recording column gradient is calculated as a function of pressure,
instrumentation and external battery should be placed with- temperature, and gas gravity or composition. The liquid
in an insulating weatherproof enclosure to protect them column pressure is a function of the composition of the
from extreme temperatures or humidity. The wellhead liquids and the in situ water/oil ratio. Gaseous liquid
sensors, cables, and connectors (if present) should also mixtures must be characterized by liquid percentage as
be protected from the environment. Wireless instruments a function of the gas rate. Flowing conditions and well
are easier to isolate from the environment, because they geometry determine the fluid distributions. For example,
can be placed inside plastic enclosures without affecting for steady state pumping conditions, the liquid above
radio signal transmission and reception. the pump intake is oil because of gravity segregation
occurring in the annulus. When the well is shut-in for
Wellbore Fluid Composition and Distribution a buildup, the inflow water cut remains essentially
Several papers have been presented on the correct meth- constant during the after-flow period and a moving oil/
ods for calculation of BHP from acoustic determination water interface develops in the wellbore during the test.
of annular liquid levels and are discussed in some detail These factors must be taken into consideration by the
in chapter 6. The BHP is always computed as the sum program when calculating BHP. In situ oil and water
of the wellhead pressure and the hydrostatic column densities must be calculated as a function of pressure
pressures due to the wellbore gas and liquids. The gas and temperature using conventional correlations11,12.

When the producing BHP is below the bubble point, during the process of liquid level depression until sta-
free gas is produced from the reservoir and is generally bilization. This is essentially similar to performing a
vented from the annulus at the surface. This annular Walker test, as described in chapter 6, which in addition
gas production reduces the liquid column gradient and will yield a good estimate of the annular fluid gradient
thus has to be taken into consideration when calculat- corresponding to the flowing condition. The measured
ing BHP. Experience indicates that a gaseous liquid gradient, when compared to the gradient computed from
column can exist in the wellbore above the formation the correlation, allows verification of the validity of the
during a period of time after the well is shut in because correlation and/or will provide an additional correction
gas flow from the reservoir may persist for a significant factor that may be used for the rest of the tests.
time even though liquid inflow may subside to almost By minimizing the amount of liquid present in the
nothing. The correlation presented in chapter 6, which wellbore at the beginning of the test, the height of the
was derived from a multitude of field measurements of liquid column at the end of the liquid after-flow period
gaseous liquid column gradients, may be used to account will be reduced. In this case, a major portion of the BHP
for this effect and to calculate the gas/liquid ratio of the will be provided by the surface casing pressure (that is
fluids in each section of the wellbore. measured very accurately) plus the pressure due to the
gas column, which can be computed accurately from
Pressure Distribution Calculation the gas properties using a good equation of state and
the gas composition.
Since all correlations yield approximate results, espe-
It must be noted that if the casing to flow line valve
cially when they are applied over a range of variables
is simply closed to depress the annular liquid level
that exceeds those used in the definition of the correlating
instead of installing a back-pressure valve to control
function, it is important to minimize the contribution to
the wellhead pressure, then the produced rate from the
the value of the BHP that depends on the gaseous liquid
reservoir will be reduced temporarily unless the pump
section(s). This can be accomplished by depressing the
is able to displace the additional annular liquid volume
gaseous liquid column to a depth just above the pump
that is pushed down towards the pump intake. Other-
intake prior to initiating the buildup test. In this situation,
wise, depressing the liquid without control may result
most of the pressure at the pump intake is due to the
in a temporary increase in the producing BHP that will
surface pressure and the gas column pressure, which can
affect the early time response of the reservoir.
be determined with accuracy. This can be accomplished
with a minimum change in producing BHP by allowing
a brief stabilization period. Another uncertainty of the RECOMMENDED TEST PROCEDURES
computed BHP stems from the generally unknown be- AND PRACTICAL IMPLICATIONS
havior of the water/oil ratio during the after-flow period. Running a pressure buildup test involves a major com-
It is generally assumed that the produced oil/water ratio mitment of time and labor as well as temporary loss
will be maintained after the pump is stopped, but this of income while the well is shut-in. Therefore, every
fact cannot be verified routinely. effort should be made to guarantee that the final data
To obtain the most accurate buildup test results when is of sufficiently good quality to yield an accurate
a long annular gaseous liquid column (greater than 1,000 representation of the formation permeability, skin, and
feet) is present in a pumping well, it is recommended that static reservoir pressure. The following recommended
before the initiation of the test, the liquid level be depressed procedures provide guidelines to help reach that objec-
to a few pipe joints above the pump by increasing the tive9. Although this reference is related to using wired
casinghead back-pressure while maintaining a steady or wireless programmable fluid level instruments, such
pumping rate. This is easily achieved by installing an as those shown in figure 7.3, the following steps are also
adjustable back-pressure regulator on the casinghead applicable to tests performed with other acoustic fluid
valve, which will maintain a constant casing pressure level instruments.

Figure 7.3 (A) Pressure transient fluid level acquisition setup using wired instruments. (B) Pressure transient fluid
level acquisition setup using wireless standalone instrumentation.

Beam-Pumped Wells are displayed in figure 7.4. The time required to

When performing pressure buildup tests on beam- depress the top of the gaseous column to the pump
pumped wells, apply the following procedure to help intake can be estimated by dividing the pressure
ensure the most accurate test results: exerted by the gaseous liquid column (psi) by the
1. Obtain all necessary data for acquisition and pressure casing pressure buildup rate (psi/minute). These
transient analysis. Review and update the base well values can be easily obtained using modern fluid
file. Obtain or draw a wellbore diagram to identify level instruments that measure the fluid level depth
all changes in annular cross-section that could be and the casing pressure buildup rate to calculate
used as downhole markers or that could interfere downhole pressures.
with automatic liquid level selection (liners, tubing 4. Inspect all well connections to the flow lines, cas-
crossovers, and so on). inghead, tubing head, and stuffing box. Check the
2. Prior to the date of the well test, perform acoustic valves for leaks and other irregularities. Report any
measurements to determine normal producing concerns to the operator so that all issues can be
conditions, acoustic velocity, casing pressure, and resolved before the date of the well test. It is impor-
existence of a gaseous liquid column. Perform a tant that the standing valve is holding; otherwise,
dynamometer test to determine pump fillage and there will be excessive backflow of liquid from the
effective pump displacement. tubing during the early stages of the buildup, and
3. If the height of gaseous liquid column is significant this will show up as additional liquid after-flow.
(more than 1,000 feet), perform a short-duration 5. Approximately 24 to 48 hours prior to the date of
(0.5 to 1 hour, uncontrolled pressure) liquid level the transient test, determine the average 24-hour
depression test by closing the casing to flow line production rate, water cut, and GOR by performing
valve to estimate the time that would be required a production test.
to depress the liquid level to the pump intake. Ex- 6. Review and update all data, and prepare the well
ample acoustic records for the short-duration test test procedure and checklist.
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Initial Fluid Level

Record
100.0 mV
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Fluid Level Record

After Closing Casing
100.0 mV
for 30 Minutes
Figure 7.4 Short-duration liquid level depression test showing the fluid level drop by 756 feet in 30 minutes

7. If the gaseous liquid column must be depressed, verify that the pump fillage and operation matches
install a back-pressure regulator on the casing to the results that were established during step 2 and
flow line outlet (if possible), and start increasing agree with the latest well test information. If the
casinghead pressure while monitoring the liquid difference is more than 10%, continue monitoring
level. Use the pressure transient data acquisition the dynamometer for at least 30 minutes to detect
module to monitor the progress of the depression any abnormalities. Figure 7.5 shows an example of a
test. As described in step 3, this test may take dynamometer record where the pump displacement
several hours or days to complete. Testing should has been adjusted for gas interference and solubil-
continue until the fluid level is approximately 60 ity to yield the effective liquid produced by the
feet above the pump intake. When this fluid level reservoir. If the pump operation is erratic because
is reached, the casing pressure should be stabilized of mechanical or other malfunctions, as shown in
to a constant value (±5% of the measured value), figure 7.6, then postpone the test until the pumping
and normal pump operation should continue until problem is fixed. Otherwise, it will be difficult to
the production rate and casing pressure are fully determine a steady flow rate, which is needed to
stabilized. interpret pressure buildup. If possible, the operator
8. Make sure that all instrumentation batteries are should monitor the daily production rate and try to
charged before starting the well test. On the day of maintain a fairly stabilized flow rate.
the test, after the equipment has been set up, take 9. Verify that all connections between the gas bottle
fluid level records to verify estimations that were and the remote fire gun do not have any leaks. Check
made earlier. Take a dynamometer record, and all cables and electrical connectors for tightness,
Figure 7.5 Dynamometer record acquired prior to initiating the pressure buildup test, showing pump displacement
of 110 bbl/day

8 8
TV
7 7
LOAD, Klbs
6 6
CBE
5 5
SV
4 4
3 3
0 100.00 200.00 0 55.5
TIME, sec. POSITION, in.
Figure 7.6 Example of erratic pump operation caused by a damaged traveling valve
or verify the wireless radio signal levels. Protect 10. Start the automatic acquisition software, and go
all sensors from rain. Place thermal insulation on through the setup procedure to acquire the zero offset
the pressure transducers that are exposed to daily of the pressure transducer. Select the transient test
temperature fluctuations, as shown in figure 7.7. module, and complete the test setup procedure. Use
Check connections to external power supplies, and a logarithmic schedule unless there is a reason for
verify that power is fed to the fluid level acquisition selecting another option. Take a pre-shot, and verify
instruments. that the program is picking the fluid level correctly.
(Adjust the signal window if necessary.) Verify
that the acoustic velocity and fluid level depth are
computed correctly, as established earlier in step 8.
11. Start the buildup acquisition while the well is still
pumping. (START the acoustic transient test.) Note
that the first pressure value corresponds to the pro-
ducing condition PBHP. As soon as the program
completes the processing of the first shot, stop the
pump. Set the brake, and lock out the motor switch.
Close the tubing flow valve to prevent the well from
flowing as the wellhead pressure builds up during
the test.
12. Monitor the progress of the test for at least 30
minutes, and check that the software has correctly
Figure 7.7 Thermal insulator placed on a pressure picked the fluid level. Check to make sure that all
sensor attached to a gas gun of the data is consistent. (The fluid level may rise

or fall depending on well conditions.) The casing 15. When returning to the well, connect to the fluid level
pressure should show a consistent trend. Make any acquisition instrument, and check the test progress
adjustments to obtain an accurate round trip travel screen. Verify when the last shot was taken, when the
time to the liquid level, as described in the software next shot is due, the presence of warning messages,
manual. Use the advanced acoustic record analysis the casing pressure, time to liquid, and so on. Take a
tools to verify that the correct echo is selected when shot manually, and observe the liquid level pick and
there are multiple echoes. The additional echoes depth calculation. Check a plot of casing pressure
could be repeats of the liquid level, as shown in figure versus time, and observe if there are any anomalies
7.8, or echoes from downhole area discontinuities. (step changes of pressure or abrupt changes of slope)
13. Determine the rate of casing pressure increase (psi/ that may indicate the presence of leaks at the well-
hour) to estimate the likely casing pressure for the head or transducer problems, as seen in figures 7.9A
time period between now and your planned return and 7.9B. Figures 7.9A and 7.9B show anomalies;
to the well site. Set the gas gun volume chamber Figure 7.9C shows the normal behavior of wellhead
regulator pressure to 200 psi above the estimated pressure during shut-in. Normally, the casing pres-
future casing pressure to ensure that fluid level shots sure versus time function should be a continuously
will be taken until that time. smooth curve, as illustrated in figure 7.9C, without
14. Check that the external power indicator is lit, and step changes or oscillations due to the daily ambient
check all connections (wired or wireless) before temperature variations. Some wireless instruments
leaving the well. Wait until a shot is taken automati- may also provide a means of remotely monitoring
cally before leaving the site. the progress of the test via an internet connection
Shot Entries Time Plots Velocity Analysis Select Liquid Level Depth Determination Log-Log Plot
Shot # ET (min) SEQ Bat(V) Csg(psi) T(F) Times(s) Vel(ft/s) Depth(ft) Cor Fact BHP(psia)
000012-H 24.02 1 12.0 1.6 38.1 1.823 1111.5 1013.1 1.00 71.6
000013-H 26.20 1 12.0 1.6 38.0 1.805 1109.1 1001.0 1.00 76.9
000014-H 28.38 1 12.0 1.5 38.0 1.788 1109.1 991.5 1.00 80.9
000015-H 30.57 1 12.0 1.5 38.0 1.772 1112.5 985.6 1.00 83.5
000016-H 32.73 1 12.0 1.5 38.0 1.755 1110.0 974.0 1.00 88.5
000017-H 34.92 1 12.0 1.4 38.0 1.738 1110.0 964.6 1.00 92.6
000018-H 37.10 1 12.0 1.3 38.0 1.722 1110.0 955.7 1.00 96.3
000019-H 38.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000020-H 48.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000021-H 48.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000022-H 48.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000023-H 58.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000024-H 58.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000025-H 58.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000026-H 68.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000027-H 68.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000028-H 78.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000029-H 78.98 1 11.0 1.3 38.0 1.701 1107.3 941.7 1.00 102.3
000030-H 84.25 1 11.9 -0.9 38.5 1.426 1107.9 789.9 1.00 166.1
000031-M 86.88 1 11.9 0.1 38.5 1.413 1103.2 779.4 1.00 171.7
Figure 7.8 Example record indicating the correct selection of the liquid level echo in the presence of a repeat
echo. The table within the figure shows the analysis of the current record and the test progress from 24.02 to 86.88
minutes.

2.50
1.88
1.25
0.63
15.00
0
12.50
-0.63
10.00
-1.25
7.50
-1.88
0 125.00
437.50
5.00
375.00
2.50
A

312.50
0
250.00
-2.50
0 2,500.00 5,000.00
187.50
125.00
B
62.50
0
0 6,250.00 12,500.00 18,750.00 25,000.00 31,250.00 37,500.00
TIME, min.
Figure 7.9 Examples of wellhead pressure variations observed in three different wells
and downloading the accumulated data to an office select “End transient test,” and exit the Pressure
computer without needing to travel to the well site. Transient module.
16. Make necessary adjustments to obtain accurate fluid 21. Select the Acoustic Test module. Select “shut-in”
level and depth values in subsequent shots. to indicate the well status. Take an acoustic record
17. Determine casing pressure increase rate, and adjust to establish the present value of the static BHP in
the regulator pressure. Check the pressure in the the well base file.
nitrogen bottle and battery voltage, and replace them 22. Select Dynamometer Test. Connect the polished
as necessary. rod transducer (PRT) or wireless polished rod
18. Make a copy of all data recorded up to this point, and transducer (WPRT) to the polished rod. Open the
move a copy of this data to a USB drive. Transfer tubing valve to the flow line, release the brake, and
the data from the USB drive to an office computer start the pumping unit.
for further analysis. Decide if the test has run suf- 23. Take dynamometer measurements to determine that
ficiently for meaningful buildup interpretation or if the pump is operating normally, and compute the
the test should be continued. net liquid volume displacement into the tubing.
19. If the test needs to continue, go back to step 14. 24. If the fluid level increased during the test, monitor
20. If the test should be terminated, take a manual shot. the fluid level periodically until it drops to one or
After the computer finishes processing the data, two joints above the pump intake. This may take

several hours depending on after-flow liquid volume valve (when present) after removing any needle valve and
and pump displacement. Monitor the casing pres- replacing it with a fully opening ball valve. Injection of
sure and casing pressure buildup rate. Data that is the gas into the casing is stopped and the valve from the
acquired during this step can be used to perform a tubing to the flow line is closed. Acoustic single shots are
drawdown analysis and establish well productivity. taken manually at 3- to 5-minute intervals until a clear
Doing so involves calculating the producing BHP fluid level echo is observed. Then the pressure transient
and the net inflow from the reservoir by calculating module is used to set up automatic acquisition of the data
the material balance and applying this calculation from that point forward.
to the wellbore control volume.
25. Open the casing valve to the flow line slowly to Gas Wells
return the casing pressure to its normal operating Refer to chapter 9 for details regarding acoustic fluid
value. For additional information, refer to the next level measurements and interpretation in gas wells.
section about ESP wells. For wells that are producing gas through the tubing
26. After the casing pressure has stabilized, repeat dy- with no packer in the annulus, the pressure transient mea-
namometer measurements to verify that the pump surements should be done in the annulus because there
is operating normally. If it is not, this could indicate will be a minimum amount of liquid accumulated above
a problem. Be sure to notify the operator. the tubing intake. Prior to shutting in the flow, several
27. If everything is normal, stop the pumping unit. Then, fluid level measurements should be taken to establish the
disconnect the dynamometer and remote fire gun. position of the liquid level in the annulus and to observe
Transfer all data to an external storage device. any variations during normal flowing conditions.
For wells that are producing gas through the tubing
28. Start the unit, and verify that everything is normal
with a packed-off annulus, the procedure is similar to
before leaving the well site.
that described above for gas lift wells. For wells that
ESP and PCP Wells are producing gas through the annulus and dewatered
For wells produced with ESPs or PCPs, the steps related through the tubing, the procedure is similar to that
to dynamometer measurements are not relevant. For ESP recommended for pumping wells. For wells that are
wells, it is very important to reduce the casing pressure producing gas through the tubing and the annulus, the
very slowly because hydrocarbon gases will dissolve preferred method is to undertake the pressure transient
in the downhole cable’s insulation, as the pressure in measurements through the annulus. These wells will
the annulus increases during the buildup test. A rapid exhibit a longer after-flow and wellbore storage effect
reduction of casing pressure will cause the insulation to than those wells producing through tubing and have a
swell and possibly damage the cable. A slow decrease packed-off annulus10.
in casing pressure allows the dissolved gas to evolve
gradually without causing the insulation to swell. Wells with Multiple Producing Zones
It is difficult to interpret the acoustic record correctly and
Gas Lift Wells determine the position of the liquid level if the record
Refer to chapter 11 for details regarding acoustic fluid includes multiple signals that were generated at section
level measurements in gas lift wells. area changes or at perforations. To correlate the acoustic
During the buildup test, fluid level measurements are record with downhole features, it is essential to obtain a
made through the tubing. Annular fluid level is monitored wellbore diagram that is complete and accurate. Figure
before shut-in and then checked periodically to observe any 7.10 shows an acoustic record that was correlated with
changes in pressure or fluid level during the transient test. perforated intervals.
After completing the fluid level test, the remote fired gas Whenever multiple open producing intervals or
gun is installed on the tubing, preferably above the swab zones are present in a given well, the wellbore pressure

can be computed as a function of time for all depths that EXAMPLE FIELD TESTS
correspond to the producing zones. Therefore, multiple Applying programmed acoustic fluid level measurements
records of BHP versus time can be determined from a to the acquisition of pressure buildup and drawdown tests
single test by reprocessing the acoustic data acquired has been used for many years on a variety of wells and
during the buildup test for each zone’s datum depth. The environmental conditions. The following are a selection
software must keep track of the position of the liquid level of field data with the purpose to illustrate the variety and
relative to a specific zone and take into account the cor- quality of the data that has been obtained. Note that the
rect fluid distribution. The main uncertainty is the correct pressure transient analysis plots that are presented here
composition of the liquid phase in terms of in situ water/ with the data were generated in the field to provide a
oil ratio. Because liquid flow has practically stopped, it is “quick look” presentation. This allows the user to decide
assumed that an oil/water interface will be established due whether the test had been run long enough to eventually
to gravity separation. Water and oil produced during the generate a complete detailed interpretation, which is cre-
after-flow will thus accumulate into a water zone overlain ated by transferring the acquired pressure data to special-
by an oil zone. The position of the interface is computed ized pressure transient analysis software. The published
by material balance and by accounting for the available literature contains more information regarding detailed
wellbore volumes based on the wellbore’s geometry. information about pressure transient analyses13-16.
The problem that needs to be considered when ana-
lyzing the buildup data is that in general it is not possible Well A
to determine or describe the cross-flow that may exist Figures 7.11 through 7.15 illustrate the data obtained
between zones because of their different reservoir pres- during a seven-day pressure buildup test in a 4,900-foot
sures and formation properties. beam pumping well. Production prior to shut-in was
Sec 13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18
5
10/18
-5
-10
mV
-15
-20
10/15
ft 8,000 8,500 9,000 9,500 10,000 10,500
Figure 7.10 Example of multiple sequential acoustic records with wellbore perforations. The dashed line shows the
progress of the depression of the liquid level echo during three days.

350.00 90.00
300.00

80.00
250.00
70.00
200.00
150.00
60.00
100.00
50.00
50.00
0 40.00
0 1,562.50 3,125.00 4,687.50 6,250.00 7,812.50 9,375.00 10,937.50
TIME, min.
Figure 7.11 Long-term pressure buildup test showing casing pressure (circles) and casinghead pressure transducer
temperature (triangles) as a function of elapsed time
350.00 1,000.00
300.00
2,000.00
250.00
DEPTH TO LIQUID, ft
3,000.00
200.00
150.00
4,000.00
100.00
5,000.00
50.00
0 6,000.00
0 1,562.50 3,125.00 4,687.50 6,250.00 7,812.50 9,375.00 10,937.50
TIME, min.
Figure 7.12 Depth to liquid level (triangles) and casinghead pressure (circles) versus time

350.00 1,875.00
300.00
1,562.50
250.00
1,250.00
200.00
BHP, psia
937.50
150.00
625.00
100.00
312.50
50.00
0 0
0 1,562.50 3,125.00 4,687.50 6,250.00 7,812.50 9,375.00 10,937.50
TIME, min.
Figure 7.13 Computed BHP (triangles) and measured casinghead pressure (circles) versus time
10,000.00
1,000.00
DEL PRESSURE, psia
100.00
10.00
0.10 1.00 10.00 100.00 1,00.00
DEL TIME
Figure 7.14 Resulting log-log plot of delta pressure with derivative versus delta time in hours

1,750.00
1,500.00
1,250.00
1,000.00
BHP, psia
750.00
500.00
250.00
0
1,000.00 1,000.00 100.00 10.00 1.00
(Tp + del t) / del t
Figure 7.15 Horner plot yields skin of 0.8 and P* of 2,018 psi.
226 bbl/day of water, 36 bbl/day of oil, and 24 Mscf/ and the derivative indicate that the wellbore storage ef-
day of gas. A total of 197 fluid level measurements were fect ends at about 20 hours. The corresponding Horner
recorded. The data was appended one time. Figure 7.11 plot shown in figure 7.15 shows a small skin of 0.8 and
shows the measured casinghead pressure increasing a P* of 2,018 psi.
throughout the test caused by a continued influx of gas.
The pressure transducer temperature exhibits daily varia- Well B
tions as large as 40°F, but temperature compensation and Figures 7.16 through 7.19 illustrate the results of a 4H-day
calibration minimize any noise induced by temperature buildup test in an oilwell producing with a progressing
on the pressure data. cavity pump at an average rate of 80 bbl/day from a depth
Figure 7.12 shows the rise in liquid level of about of 3,150 feet. The data was appended six times because of
3,000 feet experienced during the liquid after-flow pe- poor external battery quality, and a total of 136 fluid level
riod, which seems to end at about 5,000 minutes when shots were taken. Figure 7.16 shows that the liquid level
the liquid level stabilizes at about 1,900 feet. From that increased by about 2,600 feet and casinghead pressure
point on, the increase in the computed BHP shown in increased by 380 psi during the test. The x points in the
figure 7.13 is primarily due to the inflow of gas that figure correspond to bad data due to misfiring of the remote
continues while the liquid inflow has stopped. fire gas gun or other problems. They are not included in
The computed BHP is used to generate the log-log the calculations but are a part of the permanent record and
plot shown in figure 7.14, where both the unit slope line are shown for data quality control. The computed BHP

500.00 1,00.00
400.00 1,500.00
300.00 2,000.00
DEPTH TO LIQUID, ft
200.00 2,500.00
100.00 3,000.00
0 3,500.00
-100.00 4,000.00
0 1,250.00 2,500.00 3,750.00 5,000.00 6,250.00
TIME, min.
Figure 7.16 Liquid level (triangles) rises 2,600 feet and casing pressure (circles) increases by 378 psi during the 4½-
day test. The x points indicate bad data.
1,500.00 312.50
1,250.00 250.00
LIQUID AFTERFLOW, bbl/d

1,000.00 187.50
BHP, psia
750.00 125.00
500.00 62.50
250.00 0
0 -62.50
0 1,250.00 2,500.00 3,750.00 5,000.00 6,250.00
TIME, min.
Figure 7.17 BHP (circles) levels off and liquid after-flow (triangles) tends to cease after about 48 hours.

10,000.00
1,000.00
DEL PRESSURE, psia
100.00
10.00
1.00
1.00 10.00 100.00 1,000.00
DEL TIME
Figure 7.18 Resulting log-log plot of delta pressure with derivative (triangles) versus delta time in hours
1,500.00
1,250.00
1,000.00
BHP, psia
750.00
500.00
250.00
0
10,000.00 1,000.00 100.00 10.00 1.00
(Tp +del t) / del t
Figure 7.19 Horner plot shows skin of 8.9 and P* of 1,306 psi.

and liquid after-flow are plotted in figure 7.17, which from 4,998 feet at the start of the shut-in period to 1,916
shows the leveling-off of the liquid after-flow at about feet at the end of the test. The BHP increases from 168
48 hours and the corresponding near-stabilization of the to 1,607 psi during the same time period. Figure 7.22
BHP. The log-log plot in figure 7.18 shows by the end shows the log-log plot of the BHP data and indicates
of the test that the radial flow period is just beginning that the test was terminated prematurely because the
and that the test most likely ended prematurely. The derivative is changing, and the radial flow period has
Horner plot in figure 7.19 is characteristic of a severely not been reached despite the lengthy test.
damaged well with a skin near 9 and a P* of 1,306 psi.
Well D
Well C Figures 7.23 through 7.25 show the results of a buildup
Figures 7.20 through 7.22 show the results of a buildup lasting 3 days and 13 hours in a well producing 4 bbl
test extending 25 days and 17 hours. A total of 831 shots of oil and 4 Mcf of gas per day from a 56-foot thick
were taken, and the data set was appended 61 times. formation. A total of 235 shots were taken, and the data
The pumping well is completed with 7-inch casing and was appended four times.
2M-inch tubing and was producing 16 bbl/day of oil Wellbore storage effects are overcome and the
from a 12-foot pay zone at a formation depth of 5,379 radial flow period is reached after about eight hours of
feet. Figure 7.21 shows a very smooth decrease of the shut-in, but then there is an increase of the derivative,
round trip travel time as the depth to the liquid level rises as seen in figure 7.24, which may indicate the presence
437.50 110.00
375.00

100.00
312.50
90.00
250.00
187.50
80.00
125.00
70.00
62.50
0 60.00
0 6,250.00 12,500.0 18,750.00 25,000.00 31,250.00 37,500.00
TIME, min.
Figure 7.20 Variation of casing pressure and transducer temperature (triangles) during a 25-day and 17-hour buildup
test

1,750.00 10.00
1,500.00
8.75
1,250.00
7.50
TIME TO LIQUID, sec

BHP, psia
1,000.00
6.25
750.00
5.00
500.00
3.75
250.00
0 2.50
0 6,250.00 12,500.00 18,750.00 25,000.00 31,250.00 37,500.00
TIME, min.
Figure 7.21 Computed BHP (circles) increases as RTTT time to liquid decreases when liquid rises
10,000.00
1,000.00
DEL PRESSURE, psia
100.00
10.00
1.00
0 1.00 10.00 100.00 1,000.00
DEL TIME
Figure 7.22 Wellbore storage just beginning to be overcome and radial flow period starting in the well. The test was
terminated prematurely because the derivative (triangles) did not stabilize.

70.00 120.00
60.00
100.00
50.00
80.00
BHP, psia
40.00
60.00
30.00
40.00
20.00
10.00 20.00
0 1,000.00 2,000.00 3,000.00 4,000.00 5,000.00 6,000.00
TIME, min.
Figure 7.23 Casing pressure (circles) and BHP (triangles) during 3 days and 13 hours
100.00
10.00
DEL PRESSURE, psia
1.00
0.10
0.00 0.10 1.00 10.00 100.00
DEL TIME
Figure 7.24 Log-log plot shows boundary effect after radial flow period, as shown by increasing derivative (triangles).

112.00
100.00
87.50
75.00
BHP, psia
62.50
50.00
37.50
25.00
100,000.00 10,000.00 1,000.00 100.00 10.00 1.00
(Tp +del t) / del t
Figure 7.25 Horner plot yields a skin of 1.7 and P* of 102 psia.
of a boundary effect. Figure 7.25 shows that, using the The log-log plot in figure 7.28 shows that the radial
radial flow section of the data, the skin is estimated at flow period may be just beginning. As such, the test
1.7 and P* yields 102 psia. should have continued.
The Horner plot shown in figure 7.29 indicates that
Well E there may be significant skin. However, it is not pos-
The buildup in this well is an example where the liquid sible to quantify this prospect with the limited reservoir
level dropped to a deeper depth while the casing pressure data available.
increased during the test. The well was producing 1 bbl/
day oil, 18 bbl/day water, and 50 Mcf/day gas from a SUMMARY
depth of 5,020 feet. A rod pump was set at the bottom The acoustic buildup test examples illustrate the feasi-
of the producing interval at 5,770 feet, and initially the bility of performing these tests in pumping wells with
top of the gaseous liquid column is observed at 4,800 minimal disruption of the pumping system by carefully
feet, corresponding to the relatively high flow of gas setting up the well and instrumentation to minimize the
from the annulus. uncertainties inherent in acoustic BHP calculation.
The collapse of the fluid column as the casing pres- The acoustic liquid level instrument determines
sure increases is shown in figure 7.26. The correspond- the position of the liquid level in the annulus or tub-
ing BHP computed at 5,020 feet is seen to mirror the ing at specified time intervals while measuring the
increase of casing pressure in figure 7.27. wellhead pressure. The analysis software calculates

350.00 4,750.00
300.00
5,000.00
250.00
DEPTH TO LIQUID, ft
5,250.00
200.00
150.00
5,500.00
100.00
5,750.00
50.00
0 6,000.00
0 1,250.00 2,500.00 3,750.00 5,000.00 6,000.00 7,500.00 8,750.00
TIME, min.
Figure 7.26 Liquid level (triangles) drops as casing pressure (circles) increases.
350.00 437.50
300.00
312.50
250.00
250.00
200.00
BHP, psia
150.00
187.50
100.00
125.00
50.00
0 62.50
0 1,250.00 2,500.00 3,750.00 5,000.00 6,000.00 7,500.00 8,750.00
TIME, min.
Figure 7.27 Casing pressure (circles) and BHP (triangles) versus time

10,000.00
1,000.00
DEL PRESSURE, psia
100.00
10.00
1.00
0 1.00 10.00 100.00 1,000.00
DEL TIME
Figure 7.28 Log-log plot shows beginning of radial flow period, indicated by stabilization of derivative (triangles).
437.50
375.00
375.00
312.50
BHP, psia
250.00
187.50
125.00
62.50
1,000.00 1,000.00 100.00 10.00 1.00
(Tp + del t) / del t
Figure 7.29 Horner plot seems to indicate the presence of significant skin.

the fluid distribution and the position of the oil/water 4. J. K. Godbey and G. B. Ballard, “Automatic Liquid
interface and determines formation pressure versus Level Monitor,” U.S. Patent 4,318,298, April 1980.
time at a specified datum depth. The data acquisition 5. A. L. Podio, J. N. McCoy, and K. L. Huddleston,
system is designed for unattended long-term operation “Automatic Pressure Buildup Data Acquisition
where a dedicated computer controls progress of the and Interpretation Using a Microcomputer-based
test according to a predefined schedule. Electronics and Acoustic Liquid Level Instrument,” SPE Production
pressure sensors must be carefully selected to provide a Operations Symposium, 1987.
very stable system that operates reliably over extended 6. D. D. Barber, “Automatic Liquid Level Recording
periods of time. Device,” U.S. Patent 4,853,901, August 1989.
Special attention should be given when tests are
7. A. L. Podio and J. N. McCoy, “Computerized Well
performed in wellbores of complicated geometry to
Analysis,” SPE Latin American Petroleum Engineer-
verify that the echoes selected as the liquid level are
ing Conference, 1990.
not masked or confused with echoes from downhole
8. J. N. McCoy, A. L. Podio, and D. Becker, “Pressure
wellbore discontinuities.
Transient Digital Data Acquisition and Analysis
Periodic monitoring of the progress of the pressure
from Acoustic Echometric Surveys in Pumping
transient test is necessary to verify that the instrumenta-
Wells,” Permian Basin Oil and Gas Recovery Con-
tion is yielding accurate and reliable data and to decide
ference, 1992.
when the test can be terminated and still provide sufficient
data to perform detailed transient analysis. 9. D. Becker, J. N. McCoy, and A. L. Podio, “Best
Special care should also be given to maintain the Practices for Pressure Transient Tests Using Surface
equipment so that battery power and external gas sup- Based Measurements,” Southwestern Petroleum
ply are sufficient for the expected duration of the test. Short Course, 2007.
Otherwise, there will be gaps in the data during the 10. C. Fair, et al, “Gas/Condensate and Oil Well Testing-
periods when the equipment was not functional. Ap- From the Surface,” Proceedings of the SPE Annual
pending sections of data from tests that are interrupted Technical Conference and Exhibition, 2002.
by equipment malfunction should not be necessary with 11. Directive 040: Pressure and Deliverability Testing
proper planning and test monitoring. The most accurate Oil and Gas Wells, February 8, 2013, Alberta Energy
test results will be obtained by following the procedures Regulator.
detailed in this chapter. 12. A. R. Hasan and C. S. Kabir, “Determining Bottom-
hole Pressures in Pumping Wells,” SPE Production
REFERENCES Operations Symposium, 1983.
1. J. N. McCoy, A. L. Podio, K. L. Huddleston, and 13. R. N. Horne, Modern Well Test Analysis, 2nd ed.
B. Drake, “Acoustic Static Bottomhole Pressures,” (Petroway Inc., 1995).
SPE Production Operations Symposium, 1985. 14. R. Raghavan, Well Test Analysis (Prentice Hall,
2. J. N. McCoy, A. L. Podio, and K. L. Huddleston, 1993).
“Acoustic Producing Bottomhole Pressures,” SPE 15. C. S. Matthews and D. G. Russell, Pressure Buildup
Annual Technical Conference and Exhibition, 1985. and Flow Tests in Wells (Society of Petroleum En-
3. J. K. Godbey and C. A. Dimon, “The Automatic gineers, 1967).
Liquid Level Monitor for Pumping Wells,” SPE 16. R. C. Earlougher, Jr., Advances in Well Test Analysis
Annual Technical Conference and Exhibition, 1976. (Society of Petroleum Engineers, 1977).

Applications of Fluid Level Measurement to Pumping Wells 8-1
8
Applications of Fluid Level Measurements to Pumping Wells
In this chapter:
• Total monitoring of pumping system operation and wellbore fluid and pressure distribution
• Rod-pumped wells
• Well pressure survey
• Correlation of fluid level with dynamometer measurements
• ESP and PCP wells
• Recommended procedures and special considerations for quality control and analysis
Throughout the world, the most commonly used method and cannot displace liquid into the bottom of
to artificially produce oilwells is by sucker rod lift and the tubing at the rate that the formation could
has been since the early times of the industry. Efficient deliver it to the wellbore.
application of all types of well pumping systems re- • The pumping system’s theoretical displacement
quires knowledge of the position of the liquid in relation capacity equals or exceeds the formation produc-
to the intake of the pump. This quantity is defined as tivity, but the pump is operating inefficiently at
the pump submergence, and its determination was the a lower effective displacement rate, which in
primary reason for the early development1 of acoustic turn limits the liquid inflow from the reservoir.
fluid level instruments, as discussed in detail in chapter
4. The refinement of this technology and the advent of Experience has shown that the majority of pump-
portable computers have expanded the application of ing wells experience the second situation listed above,
fluid level measurements for optimization of the total where the low pump volumetric efficiency is the con-
pumping system through detailed analysis of the pres- trolling factor.
sure and fluid distribution in the well. The “First Law of Pumping” may be stated as: In a
Most operators want wells to produce at or near well that is artificially lifted by pumping, the reservoir
their capacity. When a well is producing at a maximum cannot produce more liquid into the wellbore than the
rate (defined as its potential), the producing bottomhole pump can displace from the wellbore into the tubing.
pressure (PBHP) will be very low compared to the static Fluid production (oil, water, and gas) from the forma-
bottomhole pressure (SBHP), which is equivalent to tion is controlled by the pump displacement, which means
the static reservoir pressure. If the PBHP is larger than that at stabilized conditions, the formation produces fluid
15% of the static reservoir pressure, then the current at the rate that fluid is removed from the wellbore by the
production may be significantly lower than what the pumping system. Depending on formation productivity,
formation is able to provide, indicating the reservoir is the PBHP will stabilize at a specific level and remain
not being produced efficiently. constant as long as the pump liquid displacement rate
Inefficient reservoir production by pumping may remains constant. In the annulus of the wellbore, the
be caused by one of two reasons: vertical distribution of produced fluids is controlled by
• The pumping system is operating efficiently at gravity, with gas overlaying a column of fluid generally
its maximum capacity, but is under-designed consisting of a mixture of gas and liquid. For a given

at Austin
PBHP, the height of the gaseous liquid column above measuring the casing pressure, and calculating the pump
the formation will depend on the pressure exerted by intake and bottomhole pressures (BHPs). Many expen-
the gas at the wellhead. sive errors have been made by submitting an erroneous
Changing the casinghead pressure will alter the fluid level depth that was actually an echo caused by
height of the gaseous liquid column in the annulus but a casing liner or some other anomaly in the casing ID
will not affect the PBHP when the system is producing or tubing OD. An accurate description of the wellbore
at stabilized conditions and when at least some gaseous geometry and well completion is necessary to reliably
liquid mixture exists above the perforations. This is the estimate the fluid level.
basis of the liquid level depression test, as discussed in Tools for determining the volumetric efficiency of
chapter 6. the pump depend on the type of pump that is installed
When the liquid flow rate of the pump is reduced, in the well. Generally, the actual pump performance
either by design or because of pump displacement inef- determined from direct measurement of production at
ficiency, the PBHP increases and stabilizes at a higher the surface is compared to the performance of the pump
value that corresponds to the reduced inflow rate. Cor- computed from a mathematical model that uses a detailed
respondingly, the height of the gaseous liquid column description of the pumping mechanism.
will increase, but there will not be a significant change in
the surface casinghead pressure because in most instal- PRODUCTION EFFICIENCY
lations, this pressure is controlled by the back-pressure IN ROD-PUMPED WELLS
from the surface production flow line or separator.
Therefore, the principal characteristic of wells where The operator is concerned with the overall efficiency
the pump inefficiency is limiting the liquid production is of the pumping system that includes the characteristics
the existence of a high fluid level (pump submergence) of the pump, the rods, and the surface unit. Therefore,
the analysis should not focus only on the liquid fillage
that results from the accumulation of liquid in the well-
of the pump but should quantify an overall volumetric
bore. Determining the depth of the liquid by acoustic
efficiency of the pumping system.
survey is the only practical method for identifying this
The pump effective displacement corresponds to
condition from surface measurements. However, in some
the volume of liquid actually transferred from the pump
wells, the annular fluid distribution may be inverted
intake to the pump discharge and into the bottom of the
(gaseous liquid overlaying a gas section), in which case
tubing. The factors that control the pump effective dis-
the presence of a high fluid level may not be an indica-
placement include various combinations of the following:
tion of pump inefficiency. One such instance related to
• Rod string stretch
the installation of tubing anchors above long perforated
• Tubing stretch
intervals is discussed later in this chapter.
• Plunger-barrel slippage
The task at hand is identifying clearly which of
• Valve leakage
the two cases listed above is limiting the production
• Incomplete pump liquid fillage
from the well. In both cases, there is significant pump
– Pump-off
submergence, and the pump intake pressure and PBHP
– Gas interference
are relatively large. In Case 1, the volumetric efficiency
– Choked pump
of the pump is high, while in Case 2, the volumetric
• Pump malfunction
efficiency of the pump is low.
An appropriate first step is to determine the pres- To determine the operating condition at the pump,
sure and fluid distribution in the wellbore. As discussed the results from the producing fluid level survey analysis
in chapter 6, most PBHPs, SBHPs, and analyses are should be used in conjunction with dynamometer records
obtained from surface measurements by acoustically that are processed to yield a detailed pump dynamometer
measuring the distance to the liquid level in the well, card analysis indicating pump liquid fillage and pump

effective displacement. If the following is known, the inflow performance of their wells because SBHP surveys
operator can determine whether or not the low produc- are seldom undertaken. The reason is that few operat-
ing efficiency is caused by incomplete pump fillage or ing companies are willing to shut in wells for extended
from over-pumping the well: periods of time. Every effort should be made to obtain
• The height of fluid column is above the pump representative values of the PBHP and the SBHP that
intake depth. can be used in determining the productivity index or
• The pump is incompletely filled with liquid. inflow performance relations. It thus should be clear
• Free gas is flowing up the casing annulus. that, in order to be able to make a judgment about how
• The pump intake is above the formation. efficiently a well is being produced, it is necessary to
• The PBHP is low in relation to the SBHP. have the SBHP and the PBHP as well as an accurate,
up-to-date well production test.
More efficient operations and lower costs will be If any parameter is missing, inaccurate, or guessed,
obtained when rod-pumped wells are operated with the the wrong conclusion may be reached regarding the po-
pump filled with liquid by eliminating gas interference tential of the well. If the producing efficiency is low and
using efficient downhole gas separation methods or by the incomplete pump fillage is due to over-pumping the
controlling pump run time. This is because the pump well, then the fluid level should be at or near the pump
displacement will match the inflow of liquid from the intake depth, as shown in figure 8.1 by the wellbore
reservoir into the wellbore and the pumping system will diagram on the right. The producing efficiency will
operate at maximum efficiency. be low because of incomplete pump liquid fillage, and
the upstroke load of the pump dynamometer will be at
Pump-Off: Excessive Pump Capacity the maximum level (FoMax) since the pressure inside
Accurate knowledge of well productivity is lacking in the pump barrel is minimal, as illustrated by the pump
most oilfields. Very few operators have determined the intake pressure of 31.3 psi from the fluid level survey.
Figure 8.1 Example of an over-pumped well

Note that the dynamometer analysis also gives a value pump intake depth, as shown at the right side of figure
of the PIP that is relatively low but differs by about 8.2. This condition is especially common when:
108 psi. The reasons for this mismatch are discussed in • The pump intake is set above the perforated
detail in reference 7, which concludes that the acoustic interval.
determination of PIP is generally more accurate. • Significant free gas is entering the wellbore and
To overcome this mode of operation, it is necessary percolating up the casing annulus to be produced
to match the pump displacement capacity to the produc- at the surface.
tion potential of the well. Simple changes in pumping • An inefficient downhole gas separator (or no down-
speed and/or stroke length can be made to reduce the hole separator) is installed before the pump intake.
pump displacement. Alternately, controlling the pump
run time with a pump-off controller or a percentage The high annular gaseous column corresponds to the
timer can adjust the number of strokes per day, so that high pump intake pressure (730 psi) so that the mixture
the cumulative daily pump displacement will equal the of gas and liquid filling the pump barrel at the top of
volume of liquid that flows into the wellbore. In all the plunger stroke is at a similar pressure. This causes
instances, operating the rod-pumping system with a the upstroke load of the pump dynamometer to be sig-
pump barrel full of liquid will result in more efficient nificantly below the maximum value (FoMax), as seen
operations and lower power usage. at the bottom left of figure 8.2. As soon as the plunger
moves down, the pressure in the pump barrel immediately
Gas Interference increases as the fluid is compressed, causing the load
If the producing efficiency is low and the incomplete to decrease along the typical gas compression versus
pump fillage is due to gas interference, then there will plunger position path until the traveling valve opens (at
be a significant volume of fluid in the annulus above the the point indicated by the vertical dashed line) and gas
Figure 8.2 Example of gas interference during pump operation

with liquid are transferred to the tubing. This type of valve is insufficient to fill the pump during the duration
incomplete pump fillage is defined as gas interference. of the upstroke, which may be as short as 3 to 4 seconds
When gas interference is the principal cause of for common pumping speeds. By the time the plunger
incomplete liquid pump fillage, the condition is typi- reaches the top of the stroke, the pump barrel is partially
cally caused by an inefficient downhole gas separation filled with liquid and gas at a pressure that can be as
system. The most effective downhole gas separation low as the vapor pressure of the liquid at the downhole
method is to locate the pump intake below the lowest temperature. The pump dynamometer card will essen-
formation gas entry point. An insignificant amount of tially have the same shape as that for the pumped-off
gas is dragged down to the pump intake when the liquid well condition. This condition is illustrated in figure 8.3.
annular velocity is 6 inches/second or less. When it is However, the corresponding fluid level record indicates
not possible to use this natural gas separator and the a very high liquid column and over 4,300 feet of pump
pump is set above or within the perforated interval, then submergence. The annular pressure at the pump intake
a properly designed downhole gas separator with the is of the order of 1,419 psi, which contradicts the pump
optimum size flow areas should be installed ahead of load being equal to FoMax during the upstroke that
the pump intake. When the separator’s liquid capacity normally indicates the pump intake pressure should be
is greater than or equal to the effective pump displace- low and slightly in excess of the casinghead pressure,
ment, the pump liquid fillage should be nearly 100%. which in this well is about 50 psi.
Detailed information about downhole gas separators is The high fluid level from the acoustic record normally
presented in chapter 10 of the Beam Lift Handbook8. would indicate that the pump should be either almost full
of liquid (if an efficient downhole gas separator is installed
POTENTIALLY MISLEADING ACOUSTIC below the pump) or show gas interference similar to the
FLUID LEVEL SURVEYS well in figure 8.2. If the operator had access to only the
In some rod-pumped wells, it is possible to observe a fluid level record, these would be the most logical conclu-
high fluid level and low pump volumetric efficiency. sions. Attempting to remedy the situation, the user may
This would lead the operator to conclude that the incom- increase pumping speed, replace the pump with a larger
plete pump fillage is the result of gas interference and/ plunger, or redesign the downhole gas separator. On the
or inefficient downhole gas separation, when in reality other hand, if the operator had access only to the dyna-
the problem has a different origin. Two situations have mometer record, the conclusion would be that the well is
been observed where the high fluid level is not directly pumped off and should be operated intermittently using
related to pump inefficiency: a pump-off controller or timer. None of these changes
• Restriction of liquid inflow to the pump (choked would solve the real problem of a choked pump intake.
pump intake) This example clearly illustrates the importance of
• Inversion of annular fluid distribution due to gas having both fluid level and dynamometer data, acquired
flow restriction at the tubing anchor or other point simultaneously or within a short time span, to correctly
in the annulus above pump intake visualize the performance of the well and the pumping
system. In figure 8.3, note that the pump dynamometer
Choked Pump figure includes a reference line labeled “Fo Fluid Shot”
This condition is typically associated with undersized that corresponds to the load level on the upstroke that
pump intake assemblies in relation to the instantaneous the measured pump card should exhibit in order to re-
pump displacement rate and fluid viscosity. Alterna- flect correctly the fluid level measured with the acoustic
tively, it may be due to obstruction of the pump intake record. This reference line would alert the operator that
by solids, paraffin, scale, sand screens, and so on. In there is a mismatch between the dynamometer and the
these cases, for existing pump submergence, the rate at fluid level records and a more thorough analysis needs
which liquid enters the pump barrel from the standing to be undertaken8.

Example Well—Blocked Intake
22.5
Casing Pressure Well State:
20.0
50.3 psi (g) Producing Wrf + Fo Max
17.5
Casing Pressure Buildup Annular
0.037 psi Gas Flow 15.0 Wrf
S = 120 inch
min 57 scf/D
4.00 12.5
% Liquid
Sp = 107.1 inch
LOAD, klbf
Gas/Liquid Interface Pres.
100 10.0 SPM = 6.4
58.9 psi (g)
Liquid Level Depth
7.5 1¼-inch Plunger
Liquid Below Tubing
MD 4024.44 ft 5.0 Fo Max
Oil 0 %
2.5
Pump Intake Depth Water 100 %
0
MD 8330.00 ft % Liquid Below Tubing 106.8
100 % -2.5
TVD 8330.00 0 120.0
Formation Depth Liquid Below Tubing... POSITION, in.
MD 9390.00 ft
Pump Intake Pressure 5
Fo Max
1419.2 psi (g) 4 FoUp
3
] 4306 PBHP Fo Fluid Shot
ft
1880.0 psi (g) 2
] 4306 ft 1
Reservoir Pressure (SBHP) FoDn
0
2146.8 psi (g)
-1
-2 0 107.1
Strainer nipple plugged by fines and scale from tubing.
Figure 8.3 Example of incomplete liquid fillage due to choked pump intake
Annular Fluid Gradient Inversion rate than normal, so that the fluid in the lower part of
In wells that exhibit significant annular gas flow, con- wellbore has a higher than normal concentration of gas.
ditions within the annulus can cause a high concentra- Determining wellbore pressure distribution from
tion of liquid to overlay a section of the wellbore that acoustic fluid level records is discussed in detail in chap-
contains mainly gas. This condition is defined as a fluid ter 6, where the analysis assumes that the conditions in
gradient inversion, since normally the gradients of the the wellbore are such that there is a continuous gaseous
fluids or fluid mixtures in the wellbore usually increase liquid column from the depth of the echo identified as the
with depth, as illustrated in figures 6.1, 6.2, and 8.4. liquid level to the depth of the producing zone. This gas/
The normal increase of the gradient with depth is due liquid mixture is considered to overlay a denser column
to the tendency of the liquid to flow downwards under of mainly liquid that extends from the producing zone
the action of gravity (increasing liquid holdup) in con- to the bottom of the well. This distribution of fluids in
junction with the increase of the in situ density of the the wellbore is present in the majority of wells that have
gas as pressure increases with depth. When the gradient been surveyed acoustically. However, in certain types
inversion occurs, the liquid is either not able to fall to of wells and wellbore geometries, it is possible to find
the lower part of the well or is falling at much lower that the gaseous liquid column overlays a section of the

Pt
Pc Pc PRESSURE
GAS
FLUID LEVEL
DRAW DOWN
OIL + GAS
BRINE
PBHP GRADIENT
PBHP SBHP
DEPTH
Figure 8.4 Distribution of fluids and pressure versus depth in a stabilized pumping well
wellbore that contains mainly gas, giving rise to what tion of the wellbore, such as above perforated zones or
is defined as a gradient inversion. above a liner. Figure 8.5 illustrates schematically this
The gradient inversion effect is especially likely type of arrangement, which is often used in deep wells
to be present in wells that have multiple zone comple- producing from multiple zones.
tions that produce fluids at different gas/liquid ratios In these wells, the pump intake is often set just below
from reservoirs with different pressures. It is also likely the bottom perforations to minimize gas interference by
to be found in wells where the annular wellbore area means of natural separation, and the tubing anchor is
is reduced at some depth because of the presence of set at a significant distance above the pump in a section
partial liners, casing patches, tubing anchors, and so of the casing that is free of perforations. The presence
on. Gradient inversion is most likely to occur in small of the TAC causes an annular area discontinuity, so the
diameter completions because it is related to the increase liquid concentration of the annular fluid above the tubing
of the in situ velocity of the gas caused by a reduction anchor is different from the concentration of the fluids
of flow area. below the tubing anchor.
In rod-pumped wells, the most likely cause is the This gradient inversion phenomenon was first
presence of a tubing anchor-catcher (TAC) located some observed during a series of field tests that had the
distance above the pump intake in an intermediate sec- objective of quantifying the performance of various

PRESSURE
CASING
GAS
TUBING
DETECTED
“LIQUID LEVEL”
GASEOUS COLUMN
WITH 135 MCFD GAS + LIQUID
AND 23% LIQUID
TAC TAC
MOSTLY GAS
ENTERING ABOVE GAS
WITH LIQUID FROM
BELOW FALLING INVERSION OF
DOWN THE LOW GRADIENT
LIQUID
SIDE OF HOLE
LIQUID
Figure 8.5 Schematic representation of a well completion exhibiting fluid gradient inversion due to the TAC
types of downhole gas separators2,3. Some of the wells from the casing pressure buildup observed during the
simultaneously exhibited high fluid level with inef- fluid level survey.
ficient pump performance conditions that could not be The pumping system was operated with a pump-off
explained by conventional thinking, considering that the controller that was cycling frequently, and the daily liquid
pump intake was set below the bottom of the perforated production was about Gth of the pump displacement.
intervals. Figure 8.6 presents dynamometer data from The low production and pump fillage in the presence
one such well, completed with the pump intake set 135 of a high gaseous column with the pump intake set in
feet below the bottommost fluid entry zone at a depth the sump indicated a possible pump intake blockage, as
of about 10,400 feet. The tubing anchor is set at a depth discussed in the previous section. To verify this diag-
of 7,908 feet. nosis, the pump was stopped for an extended period of
The fluid level analysis showed a high fluid level time and then restarted “on-hand” while acquiring the
due to an annular gaseous liquid column extending 3,200 dynamometer data, as shown at the top left of figure 8.6.
feet above the pump, with 28% liquid concentration cor- Considering that the first 10 pump dynamometer
responding to a gas flow rate of 60 Mscf/day estimated cards in this series of strokes show almost 100% pump

7.5
5.0 25.00
2.5
21.88
LOAD, klbf
Wrf + Fo Max
0
18.75
-2.5
15.63 Wrf
-5.0
FULL PARTIAL FILLAGE

12.50
LOAD, klbf
-7.5
0 00.00 200.00 300.00 400.00
TIME, sec
9.38
25.0
22.5 6.25
Fo Max
TV
20.0 Fo From Fluid Level
3.13
LOAD, klbf
17.5
0
99.1
15.0
-3.13
12.5 0 120.0
POSITION, in.
10.0
0 120.0
POSITION, in.
Figure 8.6 Dynamometer records in a well exhibiting annular gradient inversion
liquid fillage, the possibility of having a blocked pump The tubing anchor geometry commonly used in these
intake must be rejected. Continued pumping shows wells provided a small 2.9-square-inch open flow area
the rapid onset of partial liquid fillage that stabilizes at compared to the large annular 14.4-square-inch flow
about 25% pump displacement. The surface and pump area of the wellbore.
dynamometer cards corresponding to the steady state Further analysis of the data, considering the com-
operation of the pump (shown at the right of the figure) mon factor of a tubing anchor installed above the pump,
are characteristic of a well that is pumped-off because could result in the following scenario:
there is insufficient liquid in the annulus to fill the pump The tubing anchor and the overlaying gaseous fluid
barrel. This is a clear contradiction, given the high column combine to create a choking mechanism that
gaseous fluid level and significant pump submergence regulates gas flowing up the casing annulus. Gas velocity
indicated by the fluid level record. through the passageways between the anchor slips and
Fluid level and dynamometer surveys that were the casing wall is large enough to prevent establishing
performed in similar wells completed with multiple a downward flow of liquid from the annulus above the
perforated intervals spanning several thousand feet, with anchor to the annulus below the anchor. When the pump
the tubing anchored either above the topmost perforations and annular flow stabilize, the height of the gaseous liquid
or between perforated intervals, yielded comparable column remains constant or may oscillate about an aver-
dynamometer records and fluid level observations. age depth in wells that exhibit a slugging tendency. The

gas and liquid entering the wellbore from the perforations • A sharp decrease of the fluid gradient as the
below the tubing anchor split into two streams: some of fluid level was depressed below the depth of the
the gas flows mainly upwards through the tubing anchor tubing anchor
and most of the liquid flows towards the bottom of the
The liquid level depression tests were performed
well at a reduced rate and occupies the low side of the
as follows:
wellbore, with gas filling the majority of the annulus and
1. The fluid level and dynamometer measurements
flowing towards the pump intake. The high pressure in
were verified, and it was confirmed that the pump
the free gas column between the tubing anchor and the
was not filling completely by acquiring dynamom-
pump inhibits the inflow of additional gas and liquid from
eter records simultaneously.
the perforated intervals present in the lower part of the
wellbore. These effects combine to create a “pumped off” 2. The well was shut down for 10 minutes. Upon restart-
diagnostic pump card shape. Depending on the relative ing, if the pump was full for only a few strokes and
magnitude of the pressures in the various zones, it is also then changed to incomplete pump fillage while the
possible that some liquid and gas cross-flow between annulus maintained a fluid level above the tubing
perforated zones is created, which also reduces the liquid anchor, then the liquid level depression test needed
flow towards the pump at the bottom of the well. to be performed on the well.
3. The casing valve was closed to build casing pressure
Liquid Level Depression Tests Confirm and depress the liquid level. The well continued to
Gradient Inversion be pumped.
Proof of the existence of the accumulated free gas col- 4. Fluid level measurements were obtained every 15
umn below the tubing anchor was needed to confirm minutes as the casinghead pressure increased and
the existence of the gradient inversion created by the the liquid level was depressed.
tubing anchor location above the pump. Liquid level 5. Five additional fluid level records were acquired
depression tests are widely used to determine the an- after the liquid level was depressed past the tubing
nular fluid gradient and PBHPs in pumping wells, as anchor.
discussed in chapter 6. Fluid level measurements as a 6. Additional fluid level tests were run to verify that the
function of casinghead pressure are obtained while the liquid level stabilized near the pump intake depth.
well is pumping at a constant rate and the valve from The casing pressure and fluid level data were then
the casing to the flow line is closed, thus causing an in- plotted simultaneously as a function of time. If the gas-
crease of the casinghead pressure through accumulation eous liquid column were continuous to the pump intake,
in the wellbore of the produced gas. The magnitude of the variation of fluid level depth versus time would be a
the fluid level depression is proportional to the casing continuous curve. If the gradient decreases significantly
pressure increase. The annular fluid gradient can then at the depth of the tubing anchor, the fluid level would
be estimated by equating the change in fluid level depth drop rapidly after being depressed below the tubing anchor
to the change in casinghead pressure. A back-pressure showing a discontinuity of the annular fluid gradient.
valve should be used to control the casing pressure that Figure 8.7 shows the resulting plot of the casinghead
will allow the producing conditions to stabilize. pressure and the depth to the fluid level versus time, sum-
To confirm the existence of the accumulated gas marizing the liquid level depression test run at the well
column between the tubing anchor and the pump, the characterized by the data in figure 8.6. At time 6:33:40
gradient of the fluid in the lower section of the wellbore PM, the liquid level was at 7,659.76 feet, approaching
needed to be estimated. The liquid level depression test the tubing anchor located at 7,908 feet. Thirty minutes
was conducted to determine the existence of either: later, the liquid level dropped again by about 250 feet
• A continuous gaseous liquid column extending and was just above the tubing anchor. Ten minutes after
all the way to the pump intake that, the liquid level was depressed past the tubing anchor

65 7,000
LIQUID LEVEL = 7,659.76 ft LIQUID LEVEL = ABOVE
TUBING ANCHOR 7,908 ft TUBING ANCHOR 7,908 ft
7/25/13 6:33:40 PM 7/25/13 7:01:24 PM
SURFACE CASING PRESSURE, psig
60 8,000
DEPTH TO LIQUID LEVEL, ft

55 9,000
50 10,000
LIQUID LEVEL BELOW TUBING ANCHOR
AT PUMP = 105,99 ft 7/25/13 7:10:34 PM
45 11,000
18:14:24 18:43:12 19:12:00 19:40:48 20:09:36 20:38:24
ELAPSED TIME (HH:MM:SS)
Figure 8.7 Casing pressure and liquid level depth as a function of time, showing a rapid drop in the fluid level past
the tubing anchor
and dropped almost 2,700 feet to the pump intake depth the tubing anchor. Then the fluid level dropped 1,041 feet
of 10,599 feet, where it stabilized for the remainder of in less than 15 minutes. This verifies that the wellbore
the depression test. This sharp discontinuity indicates below the tubing anchor was filled with free gas and
that the annular section of the wellbore below the tubing very little liquid, as shown schematically in figure 8.9.
anchor was filled primarily with gas and very little liquid. Gradient inversion is caused by the tubing anchor’s
Therefore, the pressure distribution in the well at normal annular area restriction. Liquid accumulates in the up-
operation and before initiating the liquid level depression per part of the wellbore in the gaseous fluid column.
test was not as shown in figure 8.4, where the gaseous The large velocity of the upwards gas flow through the
liquid column extends to the pump intake. Instead, the small clearance between tubing anchor and the casing
gaseous column terminated at the TA as shown in figure prevents the liquid from falling back to the lower part
8.9, confirming the hypothesis of the existence of the of the wellbore and also causes some gas to accumulate
fluid gradient inversion phenomenon. below the tubing anchor and flow all the way down to
To verify that this was not an isolated case, addi- the pump. The pressure created from the accumulated
tional tests were performed in similar wells in the same gas below the tubing anchor may restrict liquid inflow
field. Figure 8.8 shows the results of one of those tests. from the producing formations. Gravity causes the
Before the depression test was initiated, the gaseous liquid to flow at a reduced rate along the low side of
liquid column extended about 1,500 feet above the tubing the wellbore to the bottom of the well, resulting in a
anchor. After the casing-to-flow line valve was closed, it pumped-off condition. This condition occurs because
took about 12.4 hours to depress the gaseous column to the pump capacity exceeds the formation liquid inflow.

7,000 350
DEPTH TO LIQUID, ft
7,500 CASING PRESSURE, psig 300
8,000 250

14:19 6/9/2013
DEPTH TO LIQUID, ft
8,500 200
TAC AT 9,140
9,000 150
2:21 AM
9,500 100
2:36 AM
10,000 50
PUMP AT 10,181
10,500 0
0 100 200 300 400 500 600 700 800
TIME, mins.
Figure 8.8 Results of a liquid level depression test in a well with the tubing anchor set above the perforations
Pt
Pc
PRESSURE
Pc
GAS
FLUID LEVEL
GASEOUS
LIQUID
COLUMN
INCREASED
PRESSURE
BELOW TA
TA
ZONE 1
ZONE 2
ZONE 3
GAS
LIQUID
PBHP
BRINE
GRADIENT PBHP
DEPTH
Figure 8.9 Fluid and pressure distribution caused by the presence of the tubing anchor

Effect of Tubing Anchor on Well Performance freely enter through the separator ports and provide
Whether or not the fluid and pressure distribution due adequate pump liquid fillage. It should also be noted
to the presence of the tubing anchor and a high gaseous that specially designed tubing anchors with large flow
liquid column is causing a reduction in well production areas are commercially available and should be used in
depends on the well’s inflow performance characteris- wells producing annular gas at high flow rates.
tics. The liquid level depression test data at the point in
time when the fluid level drops below the tubing anchor TUBING DIAGNOSTIC ACOUSTIC SURVEYS
gives a very good estimate of the existing PBHP. In the One technique that is used to troubleshoot pumping wells
example shown in figure 8.8, another well exhibiting a that have stopped producing all together or at a much
high gaseous column above the tubing anchor set above lower rate than normal is to perform a tubing diagnostic
the perforated intervals at 9,140 feet, the casinghead acoustic survey. This type of survey involves shooting
pressure reaches a value of 275 psi and the liquid level a fluid level inside of the tubing in an effort to:
has dropped to the pump intake (2:36 AM). The PBHP • Identify the presence of holes in the tubing.
in the well was estimated from analysis of the fluid level • Determine if excessive gas flow has unloaded
survey to be at 325 psi. This pressure was also the pres- the tubing, that is, whether the well has kicked.
sure that existed at the bottom of the well before the test • Estimate the volume of gas flowing up the tub-
was initiated and agrees with the pump intake pressure ing and the percentage of liquid in the gaseous
(PIP) estimated from dynamometer records. If this 325- tubing column.
psi PBHP was high in comparison to the SBHP of the • Determine the effectiveness of the downhole
reservoirs (the ratio of PBHP to SBHP was greater than gas separation equipment. For example, if the
20%), then the well was not drawing down, and the gas percentage of liquid inside the tubing is low and
pressure was restricting liquid inflow. If, on the other the tubing fluid column is gassy, then the bot-
hand, the PBHP/SBHP ratio was less than 15% to 20%, tomhole assembly is ineffective at gas separation.
then the liquid inflow was not restricted significantly. • Evaluate inefficient pump displacement by moni-
As stated earlier, it is necessary to gather information toring tubing pump-up rate. A leaky pump can
about the PBHP and the SBHP in order to evaluate the display a pump card filled with liquid, but the fluid
performance of a well properly. level records acquired in the tubing can show the
It should be noted that, although the distance be- liquid level traveling up the tubing very slowly.
tween the tubing anchor and the pump intake was several • Indicate if the tubing is partially plugged with
thousand feet in the example wells described here, such paraffin. The presence of a lot of paraffin will
a long distance should not be a requirement for the tub- make a tubing acoustic record almost impossible
ing anchor in order to create additional back-pressure to interpret.
on the producing formations or to force gas to flow to
the pump intake. A similar problem may exist whenever Detection of Tubing Holes
the tubing anchor is located only a few pipe joints above Holes in the tubing are usually found deep in the well
the entry ports of the downhole gas separator that is in the vast majority of sucker rod installations. Rod-
attached to the pump intake. This is the configuration on-tubing friction often creates excessive wear that can
used by most operators when installing a downhole gas become a hole in the lower portion of the tubing string.
separator. The presence of the flow restriction caused by The presence of a deep hole in the tubing usually allows
the tubing anchor could create a high concentration of the fluid levels to equalize, causing the tubing fluid level
gas at the separator intake and reduce pump liquid fil- and the casing fluid level to be within a few hundred feet
lage. On the other hand, if the tubing anchor is installed of each other. The surface tubing and casing pressures are
below the downhole gas separator, then the liquid in the often within a few psi of the same value. To further verify
annular gaseous column above the tubing anchor can the possibility of a hole, several surface dynamometer

cards should be acquired to check for pump action and The automatic collar count method of determining
liquid fillage. Deep tubing leaks always produce flat distance may yield unreliable values due to the mixing
dynamometer cards, while a well with a shallow tubing of echoes generated at the rod couplings with echoes
leak will often produce full pump cards, but fluid will that may be generated at the internal gaps of the tubing
not reach the surface. In these cases, multiple fluid level joints. When echoes from the rod couplings are clearly
records taken in the tubing at the start and at the end identifiable, the conversion from couplings per second to
of a 30-minute pumping period will indicate the pres- feet/second should be completed using the average rod
ence of a tubing leak when the two fluid levels show length instead of the average tubing joint length. Figure
minimal change. 8.10 shows an example record where echoes from rod
Following the results of checking for holes in the couplings spaced at 25-foot intervals are identifiable in
tubing, the presence of the tubing hole should be verified the upper part of the tubing and yield an average acoustic
by running a tubing integrity test, which involves closing velocity of 973 feet/second. For fiber-reinforced rods,
the valve to the flow line and pressurizing the tubing, the average rod length is 37.5 feet.
either through pump action when some fluid is lifted to When echoes from the rod couplings are not visible,
the surface or through an external pressurizing system. the average acoustic velocity determined from a casing
When the location of the leak is relatively shallow, its acoustic record is used to calculate the distance to an
depth can be pinpointed from the up-kick echo that it echo observed in the tubing acoustic record. This option
generates in the acoustic record. is generally the easiest and most accurate method for
determining distance down the tubing. After all, in the
Acquisition of Fluid Level Records in the Tubing majority of the wells, the tubing head and casinghead
Shooting the fluid level inside the tubing usually re- pressures are about the same and thus the acoustic ve-
quires 50% less pressure differential between the gas locities are also similar.
gun chamber and the tubing than is normally used to When the pump is producing liquid up the tubing,
acquire acoustic records in the casing. A smaller pressure the presence of a hole can be masked or hidden because
difference is required for two reasons: of liquid filling the tubing at the depth of the hole. A
• The tubing shot is usually made through a one- recommended practice is to shut the pump down for a
inch bleeder valve. period of time to allow any liquid in the tubing above
• The space inside the tubing is more restricted the hole to drain out of the tubing so that the fluid level
by the sucker rod couplings than by the tubing is at or below the hole. This will allow the echo from
collars inside the casing annulus. the hole to be visible on the acoustic trace.
Figure 8.10 Frequency of rod couplings echoes in tubing (19.46 jts/s at 25.0 ft/jt)

Sec 0 2 4 6 8 10 12
C L
10.0 mV
Explosion
ft 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Figure 8.11 Example tubing record showing an echo from a deep hole
Determining Depth to Tubing Hole spreadsheet program or the application that is used for
Once the casing and the tubing fluid levels are properly analysis of the acoustic data. Figure 8.12 shows the overlay
analyzed, the tubing and casing acoustic traces can be of the tubing record (blue trace) and the casing acoustic
compared or overlaid and checked for an up-kick from the shot (black trace) for a well that exhibits a shallow tubing
tubing hole appearing on both traces at or near the same leak. Both echoes indicate an up-kick at about 0.55 sec-
depth. The liquid level marker is moved to the beginning onds that repeat (with inverted polarity) near 1.1 seconds.
of the up-kick from the hole on the tubing shot and to When the tubing leak is significant and the external
the beginning of the up-kick from the hole on the casing diameter of the tubing is large compared to the internal
shot. Figure 8.11 shows the acoustic record acquired in the diameter of the casing, then the liquid exiting the tubing
tubing that shows a distinct up-kick at about 5,030 feet. while the pump is operating can cause sufficient obstruc-
Whenever possible, it is useful to overlay the two tion in the casing/tubing annulus for the acoustic wave
acoustic traces. An overlay can be created using a to generate a visible echo in the casing acoustic record.
Sec -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3
150
100
50
mV
-50
-100
-150
ft 0 200 400 600
Figure 8.12 Comparison of a hole-in-tubing echo overlay of low-pass filtered casing shot to a raw tubing shot

Figure 8.13A is a plot of two acoustic traces acquired split in the 2K-inch tubing at 4,056 feet from the surface.
down the 4H-inch casing in a rod-pumped well with a The leak is such that no liquid is reaching the surface.
Liquid Leaking from Tubing at 4,056 Feet

A
PUMP ON
Sec 0 2 4 6 8 10 12 14
C LL
10.0 mV
Explosion
ft 0 1,000 2,000 3,000 4,000 5,000 6,000 7
PUMP OFF
Sec 0 2 4 8 10 12 14
C LL
10.0 mV
Explosion
ft 0 1,000 2,000 3,000 4,000 5,000 6,000
B
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
9.08 sec
Off
mV
Pumping
4,056 ft
-0.5
ft 0 2,000 4,000 6,000
Figure 8.13 (A) Acoustic traces acquired in the casing annulus while the pump is stopped and while the pump is
operating, showing an inversion of polarity of echo from the tubing leak. (B) Overlay of acoustic records acquired
when the pump was operating and when the pump was stopped.

Echoes from the hole on both acoustic traces are located wells that kick are just the opposite. When a well
at 9.08 seconds from the surface. While pumping, the has unloaded the tubing fluids, it is common to
echo from the hole is displayed as a down-kick in the observe a very high fluid level inside the casing
upper trace because the produced liquid discharging with a deep fluid level inside the tubing.
from the hole is partially filling the wellbore annulus, • The liquid column inside the tubing will almost
decreasing the cross-sectional area. After turning off the always be very gaseous, with only 15% to 20%
pump for some time and repeating the shot, the second of the column being liquid. Monitoring the
fluid level record (lower trace) shows the echo from change in tubing pressure versus time with the
the hole as an up-kick because the liquid is below the pump stopped and the flow line valve closed will
end of the split tubing joint, so that the wave sees an show increasing pressure versus time, indicating
enlargement of the cross-sectional area in the wellbore that a significant amount of gas is migrating up
at the failure point. the tubing.
Other than the inversion of the polarity of the echo • If a back-pressure valve is in use on the tubing
from the tubing rupture, when the two traces are overlaid and/or a choke on the casing, then the tubing and
as shown in Figure 8.13B, the character of all the other casing pressures will not be close together, as
recorded signals is practically identical. This shows that, when the pressures are nearly equal when there
in most wells, it is not necessary to stop the pumping is a tubing leak.
system in order to acquire a good quality acoustic record.
The combined analysis of the data from the tubing
TUBING GAS FLOW AND WELLS THAT KICK and casing fluid level shots and dynamometer records will
indicate that the downhole gas separation performance
In recent years, with the advent of completions in 2,000- to
is poor and that too much gas is being compressed by
3,000-foot-long multi-zone perforated intervals or with
the pump and produced up the tubing.
long horizontal wellbores, the industry has experienced
a great increase in the number of artificial lift systems
Recommended Troubleshooting Procedures
operating at relatively large liquid rates in gassy wells
with high PBHPs. These wells are characterized by flow After the well kicks and unloads the tubing, shooting
conditions with alternating gas and liquid slugs that can fluid levels down the casing and inside the tubing can
cause the introduction of large volumes of gas into the provide the operator with an inexpensive diagnosis of
tubing when downhole gas separation is inefficient. In what has happened in the well. The following procedures
addition, at a high PBHP, the oil that enters the pump are recommended for shooting fluid levels:
contains a large amount of dissolved gas that is released as 1. Acquire the acoustic fluid level record in the cas-
the fluid flows to the surface and the pressure decreases. ing. Analyze the data to determine the height of the
Excessive amounts of gas produced up the tubing can gaseous column, the flow rate of gas up the casing,
unload it of liquid, causing an effect that is similar to a and the percentage liquid inside the casing. Note
gas kick commonly experienced while drilling a well the acoustic velocity of the casing gas to be used
with under-balanced BHP. After the tubing is “unloaded,” for the analysis of the tubing acoustic record.
the lease operators discover that the well is not produc- 2. Turn off the pumping unit. Wait to close the tubing
ing liquid to the tank and erroneously may suspect the valve to the flow line until just before acquiring the
presence of a hole in the tubing or that foreign material first acoustic record in the tubing. When shooting
has stuck open a valve in the pump. down the tubing in a well suspected of kicking,
Wells that kick and unload the tubing fluids usually preserve the initial condition of the well to obtain
have the following traits: the highest quality data possible. Blowing down
• Unlike wells that have deep tubing leaks where the tubing pressure changes the conditions in the
the tubing/casing surface pressures equalize, tubing and should not be done.

3. Once the fluid level shot is acquired inside the tub- INEFFICIENT PUMP DISPLACEMENT
ing, the flow line valve is left closed to monitor the
When the pump liquid displacement is inefficient and
tubing pressure. Leaving the valve closed allows the
the pump submergence increases substantially, the
operator to determine the percent of liquid in the
conventional valve leakage test does not yield reliable
tubing and how much gas is flowing up the tubing,
information about slippage between barrel and plunger
as discussed in chapter 6.
and/or traveling valve leakage. In these cases, the pump
4. Repeat the acoustic fluid level measurement to inefficiency can be quantified by monitoring the rise of
observe the collapse of the gaseous column inside the fluid level in the tubing as a function of time, using
the tubing. The change in height and the increase in successive acoustic records taken in the tubing. The
pressure can be used to compute the gradient of the dynamometer card normally will indicate 100% liquid
gaseous mixture in the tubing. This value is useful fillage, but no liquid is reaching the tubing head, or the
for detailed analysis of the pump dynamometer card produced liquid rate is far below the pump displacement.
and calculation of the pump intake pressure from Based on the plunger travel and cross-sectional area,
the dynamometer. the maximum pump displacement can be calculated.
Knowing the tubing internal diameter and the rod string
Results from this test normally show that the well
size, the capacity of the tubing-rod annulus (bbl/feet) is
has a very high fluid level inside the casing annulus and
determined. These values are combined to determine the
a relatively high pump intake pressure, which encour-
fluid level rise per stroke that should be observed when
ages the well to flow up the tubing.
the pump is operating with a minimum of slippage or
valve leakage. Figure 8.14 illustrates such a diagram.
Operating Gassy Wells The top taper of the rod string is normally used because
Identifying that tubing fluids are being unloaded is an the fluid level in the tubing generally is found in the
important step in the process of properly operating a upper part of the well.
gassy well. Additional back-pressure on the tubing may At a specific pump displacement rate (plotted on the
be required to prevent unloading of the tubing due to x-axis), a point on a line that corresponds to the actual
excessive gas pumped into the tubing. Pump action can pumping speed indicates on the y-axis the distance the
usually be restarted by loading the tubing with water, liquid in the tubing should rise per stroke of the pump.
but the problem of tubing fluids being unloaded will
reoccur and require multiple interventions. Documenting Fluid Movement
The oil entering the pump at a high BHP contains Successive tubing fluid shots, displayed in figure 8.15,
a large amount of gas in solution. Even if full pump show the effect on the acoustic record of a 55-foot rise
liquid fillage is noted in the dynamometer analysis, the in tubing fluid level during 32 minutes of pumping. The
tubing will tend to unload because of the expansion of decrease in fluid level depth indicated an average fluid
the gas released from the oil when it reaches the much level movement corresponding to a little over 1.7 feet
lower tubing head pressure as it moves up the tubing. per stroke. The theoretical fluid level rise corresponding
Good downhole gas separation is critical to trouble- to the pump displacement of 170 bbl/day should have
free operation of these wells. Do not underestimate the averaged over 4 feet per stroke. The difference in fluid
value of downhole gas separation, and apply the available level rise per stroke indicates that the pump installed
technology to the design of an efficient downhole gas in the well was underperforming and was not pumping
separator. The operator should take action after analyz- sufficient liquid per stroke into the tubing.
ing the results of the fluid level shot down the tubing in The dynamometer pump card acquired simultane-
a very gassy well. When the data collected on the well ously shows good pump action, but liquid was not be-
is properly interpreted and acted upon, gassy wells can ing pumped into the tubing due to a leaky pump. The
be operated economically with minimum intervention5. fluid level shots down the tubing were used to make

Feet/Stroke Liquid Level Rise Between

1" Rod and 2.875" Tubing Annulus
16
DISTANCE WELL 1Y
TO LIQUID ELAPSED RISE
14 SHOT TIME RTTT (Sec) (ft) TIME (min) (ft/min) SPM 6.59
RISE OF LIQUID INSIDE ROD AND TUBING
CASING 01:14:56 PM 12.248 7300.0 0 FILLAGE 91%

12 TUBING 01:48:28 PM 1.221 727.7 33.533 PUMP DISP 127 bbl/d
TUBING 02:20:03 PM 1.129 672.9 31.583 1.736
ANNULUS, feet/stroke
10 TUBING 03:51:06 PM 0.973 579.9 91.050 1.021
6
CALCULATED
4 5.0 5.5 6.0
6.5 7.0 7.5

2
8.0 8.5 9.0
MEASURED RISE LOW
PUMP UNDER PERFORMING
9.5 10.0 SPM
0
100 200 300 400 500
LIQUID PUMPED INTO TUBING, BPD
Figure 8.14 Tubing fill-up rate as a function of pump displacement and SPM
the determination on whether the pump in this well is shot down the tubing showed clearly that the pump was
truly pumping. A visit to the pump shop on the next not lifting sufficient fluid.
day discovered that the operator had mistakenly run The previous example shows that—using an acous-
a pump with a PAP (grooved steel) plunger (fig. 8.16) tic instrument and shooting fluid levels down the tubing,
in this 10,955-foot deep well. With high differential even without a pump card or a valve test—an operator
pressure acting across the PAP plunger, lost pump can still “make a good call” and determine whether or
displacement due to slippage accounted for the very not his well needs to be pulled without incurring the
slow rise of liquid in the tubing. Initially, the well had additional expense of a pump truck.
a high fluid level with low differential pressure acting
across the PAP plunger, so the production from the well Paraffin Deposition
was good. As the well was produced, the lowered fluid One of the biggest problems in shooting and interpreting
level caused an increase of the pressure acting across tubing fluid level shots is the presence of one of the oil-
the plunger and pump slippage increased. No matter field’s worst enemies: paraffin. Shooting fluid levels down
what the dynamometer cards look like, the fluid level tubing packed with paraffin usually provides unreliable

Fluid Level Rise Inside Tubing
Sec -0.5 0 0.5 1 1.5 2 2.5 3
20
RTTT, sec
01/21/2014 01:48:28 pm
0
RTTT, sec
01/21/2014 01:57:08 pm
-20
mV
RTTT, sec
01/21/2014 02:12:56 pm
-40
RTTT, sec
01/21/2014 02:20:03 pm
-60
ft 0 500 1,000 2,000
Successive tubing fluid shots show PAP pump was under-performing by not pumping sufficient liquid per stroke.
Figure 8.15 Successive acoustic records taken during 32 minutes of pump operation
impossible. If the well has a known history of paraffin

buildup and is overdue for a treatment for paraffin, then
caution should be exercised when using a tubing fluid
level shot to diagnose and troubleshoot the problems
discussed in this chapter. Experience is beneficial when
interpreting the results from shooting down the tubing
on wells that may have a paraffin problem. If a well is
suspected of having paraffin in the tubing and if the tubing
acoustic trace exhibits unclear data and anomalies with
Figure 8.16 A PAP (grooved steel) plunger no identifiable fluid level, then it may be time to apply
hot oiling treatment to help remove the paraffin deposit.
results. The presence of moderate to severe amounts of Cleaning paraffin out of the tubing is one technique that
paraffin inside a well’s tubing will make shooting of can be used to make fluid level shots more valuable in
the fluid level inside the tubing untrustworthy, if not wells with paraffin.

FLUID LEVEL SURVEYS IN ESP WELLS In small casing diameter wells (4H inches or less), the
Electrical submersible pumps (ESP) must remain im- presence of the electrical cable and the metal bands that
mersed in the produced liquids so that the electrical are used to secure the cable to the tubing may generate
motor is cooled by the flowing fluids. Therefore, it is acoustical noise that interferes with the detection of the
vital to verify that the liquid level stabilizes a significant echoes from the tubing collars. Figure 8.17 shows typical
distance above the pump assembly at all times. Acoustic acoustic records with significant high-frequency noise
fluid level measurements provide the most common caused by the cable bands.
method to verify that adequate pump submergence is The analysis software may misinterpret the origin
maintained, even in wells that are outfitted with BHP of this high-frequency signal as coming from the tubing
sensors with surface readout. collars. This misinterpretation would in turn cause the
software to calculate an erroneous acoustic velocity and/
Acoustic Fluid Level Acquisition and Analysis or collar count, yielding an inaccurate pump submergence
Fluid level measurements can be performed and analyzed value. Analysts who analyze the software’s results should
successfully in most ESP installations where there is verify the indicated liquid level depth. As mentioned
adequate clearance between the tubing and the casing. earlier, the presence of paraffin accumulations in the
A
Sec -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13
1.5
0.5
mV
-0.5
-1
-1.5
ft 0 2,000 4,000 6,000 8,000
B
Sec 2 2.5 3 3.5 4
2
mV
-2
-4
ft 1,200 1,400 1,600 1,800 2,000 2,200 2,400 2,600
Figure 8.17 (A) Fluid level record acquired in an ESP. (B) Detail of an acoustic signal showing echoes from cable bands

annulus will also cause serious difficulties for acquisi- result in a low PBHP, indicating that production from
tion and analysis of fluid level records in these wells. the formation is near its potential. If the PBHP is low
(less than 10%) compared to the SBHP, the maximum
Presence of Gaseous Column production is being obtained from the well even though
Although ESP applications generally involve high water a high gaseous liquid column exists over the well during
cut and low gas/liquid ratios, an ESP-pumped well often normal operations. Figure 8.18 shows an example of
has a high gaseous liquid level over the pump. The well acoustic data that indicates that the maximum practical
produces a considerable amount of gas up the casing an- production rate is being obtained from the well, even
nulus, so a centrifugal gas separator is installed in series with more than a thousand feet of gasified fluid column
with the centrifugal pump. The separator discharges the over the pump. The existing producing efficiency cor-
gas into the annulus. The discharge aerates the liquid, responds to 98.1% of the potential. A larger-capacity
causing a gaseous liquid column to exist for a consider- pump or faster pump rotation speed would not increase
able height above the pump. The light column of gaseous production from this well sufficiently to justify a change
liquid over the pump and a low casing pressure may in operations.
Figure 8.18 Productivity analysis for an ESP well

COMPARING ESP DOWNHOLE PRESSURE and produced through a single tubing string. Depend-
SENSOR MEASUREMENT AND PIP FROM ing on the relative position of the pump intake and the
ACOUSTIC FLUID LEVEL primary fluid produced by each zone, the annular gas/
liquid concentration will be different from section to
An increasing number of recent ESP installations include
section of the wellbore. In these cases, there may be a
a downhole pressure sensor with surface readout for con-
significant difference between the measured downhole
tinuous monitoring of the pressure at the pump intake and
pressure and the pressure computed from the acoustic
verification of adequate operating pump submergence.
survey, depending on the location of the pressure sensor
In vertical wells that produce from a single formation,
relative to the producing zones. The following example
it has been observed that the measured pressure values
illustrates one such case.
generally match the pump intake pressure computed
from an accurate fluid level record. In many instances,
a discrepancy between the two values is an indication of Pressure Distribution and Annular Fluid Level
a malfunctioning downhole sensor or a sensor that has Gradient Discontinuity in Well with Multiple
experienced a significant zero shift over time. Producing Zones
The acoustic BHP calculation that was described The following example illustrates the need to consider
in chapter 6 assumes that the majority of the fluids are various multiphase flow regimes when describing the
produced from a single formation or from a multilayer wellbore pressure distribution in wells that produce flu-
zone distributed over a relatively short interval. This as- ids simultaneously from several formations distributed
sumption allows us to consider that the annular gas flow over a significant depth interval. The well in question is
is fairly uniform throughout the wellbore. The percent produced with an ESP; the intake is located below one
liquid and gas present at any point in the wellbore can of the producing zones and above a lower perforated
then be computed, taking into account the annular area zone. The flows from the various perforated intervals
changes, by means of the “S” correlation, as described are commingled and produced through the tubing by
in chapter 6. the pump.
The gas and liquid flow distribution along the Figure 8.19 shows a typical acoustic record cor-
wellbore can be different in those wells producing from responding to a stabilized flow condition with the well
multiple reservoirs where the flows are commingled producing at the well test rates of 90 bbl/day of oil, 513
Fluid Above Pump 2,702 ft TVD

Distance to Liquid 2,939 ft MD Equivalent
Gas Free Above Pump 771 ft TVD
RTTT (sec) 4.744 #JTS 92.81 AV 1,239 ft /sec JTS/sec 19.56
Sec 0 1 2 3 4 5 6 7 8 9 10
15
10
5
mV
-5
-10
-15
ft 0 100 200 300 400 500 600
Figure 8.19 Typical annular fluid level record for the well in figure 8.20

bbl/day of water, and 140 Mscf/day of gas. A clear echo a distance of 2,939 feet using an average tubing joint
from the fluid level is observed at 4.744 seconds, and the length of 31.67 feet/joint.
tubing collar count yields a total of 92.8 tubing joints Figure 8.20 is the corresponding fluid level summary
from the surface to the fluid level. This is converted to report generated based on the conventional analysis,
Fluid Above Pump 2,702 ft TVD

Liquid Level 2,939 ft Equivalent Gas Free Above Pump 771 ft TVD
Gun - WRFG 399

Current Potential
Oil 90 *-* BBL/D
Water 513 *-* BBL/D
Gas 140.0 *-* Mscf/D
IPR Method Vogel
PBHP/SBHP -0.00
Casing Pressure
Annular Gas Flow

Fluid Properties
Wellbore Pressures
PIP 315.6 psi (g) @ 5,641 ft
PBHP 384.4 psi (g) @ 5,994 ft
SBHP *-*
Depths
Collar Analysis (Automatic) Casing Pressure Buildup

47.5
47
46.5
46
45.5
45
44.5
0 0.5 1 1.5 2
Delta Time, Minutes
Acoustic Velocity 1,239 ft/s Buildup 1.0 psi (g)
Joints per Second 19.56 jts/sec Buildup Time 55 sec
Joints to Liquid 92.81 jts Gas Gravity 0.1024 Air = 1
Figure 8.20 Acoustic survey summary report displaying the computed annular pressure distribution

which presumes that all the fluids are produced from a The pressure sensor is believed to be reasonably ac-
single zone. The fluid level corresponds to a relatively curate, so the conclusion is that the PIP value, computed
high gaseous liquid column with an average annular from the acoustic fluid level record, is not representa-
liquid concentration of 28.5% extending from the liquid tive of the actual pressure and fluid distributions in the
level to the depth of the pump intake and a larger liquid wellbore. Such a discrepancy has not been observed in
concentration of 43.12% below the pump corresponding numerous comparisons of the computed and measured
to the greater wellbore flow area, as determined from pump intake pressures, so this occurrence deserves a
the “S” curve presented in figure 6.13. The pressure at detailed analysis and an explanation to alert the reader
the gas/liquid interface at 2,939 feet is computed as 49.3 that, in some cases, the acoustically determined downhole
psi, using the measured casinghead pressure of 44.6 psi pressure distribution may be in error.
and the annular gas specific gravity of 0.7296. Assum- The pressure versus depth plot shown in figure 8.21
ing a constant gaseous column liquid concentration, the illustrates and helps explain the discrepancy between
pump intake pressure at 5,641 feet is computed as 315.6 the two values by considering the possibility that, in
psi. This value is much lower than the pressure of 611 this well, the majority of the gas inflow occurs from
psi displayed by the downhole pressure sensor located the upper zone (4,636 to 4,813 feet) while the majority
at a depth of 5,667 feet, about two pipe joints deeper. of the liquid is produced by the bottom two zones. Any
Pressure Traverse from Fluid Level and from ESP PIP Sensor
0 100 200 300 400 500 600 700

0
1,000
TAM Pressure Sensor
2,000
GAS GRADIENT
0.016 psi/ft
DEPTH, ft
3,000
GAS/LIQUID INTERFACE
PRESSURE = 49.3 psi
4,000
GASEOUS COLUMN
GRADIENT 0.011 psi/ft FLUID GRADIENT
UPPER ZONE 0.424 psi/ft PRESSURE SENSOR
PBHP = 244.9 psi READING = 611 psi
5,000
GASEOUS COLUMN FLUID LEVEL

GRADIENT 0.093 psi/ft PIP = 315.6 psi
6,000
Figure 8.21 Pressure versus depth traverse showing the measured and computed pump intake pressure for the well
in figure 8.19

liquid that might be produced by the upper zone will This example also indicates that downhole pressure
tend to flow downwards to the pump intake, contributing calculations based on acoustic fluid level records, as
to the higher liquid concentration in the lower section described in chapter 6, must be evaluated critically when
of the wellbore. applied to wells where multiple zones are producing
The measured pressure at the sensor (611 psi) and fluids into the well or whenever any of the underlying
the computed pressure (244.9 psi) at the bottom of the assumptions may not be satisfied.
upper zone are used to compute the average gradient
of the fluid column between the bottom of the upper FLUID LEVEL SURVEYS IN PC WELLS
zone and the pressure sensor as follows: (611 – 244.9) Progressing cavity (PC) pumps can be easily damaged
÷ (5,667– 4,813) = 0.424 psi/ft. This also indicates a by incomplete liquid fillage during operation. The pump
high concentration of liquid (89%) based on the brine’s stator assembly is lubricated and cooled by the liquid
specific gravity of 1.10. passing through the pump. If only gas is being produced
From this analysis, one can develop the schematic by the pump, even for only a short period of time, the
diagram of the wellbore fluid distribution shown in figure pump will rapidly heat and cause failure in the polymer
8.22, including the type of flow that is expected in each coating of the pump assembly.
section. In the upper part of the annulus, gas is flowing When a PC pump is located at a depth of 4,000 feet
to the surface at a rate of 91.6 Mscf/day, with the top and the well is pumped down, the pressure increase
of the gaseous liquid column at 2,938 feet where the across the pump may be 1,300 psi or more. Any free
pressure is 49.3 psi. From this point down to the upper gas present at the pump intake will be compressed, and
perforated interval, the liquid in the gaseous column its temperature will increase 500°F or more during this
is recirculating and is characterized as zero-net liquid compression. If only gas is produced for a minute or two
flow with upwards gas percolation. The two-phase flow without liquid to cool the pump, the polymer coating
pattern is probably of the “churn” type, characterized will be destroyed.
by high gas velocity, low liquid velocity, and highly In general, a PC pump should be set below the
agitated flow in which gas bubbles interact with each formation for better downhole gas/liquid separation and
other and coalesce to form larger distorted bubbles with hence better lubrication and cooling of the pump. If the
unpredictable behavior. Below the bottom of the upper pump is set in or above the formation, a good downhole
zone, liquid flows downwards while some gas is slip- gas separator should be used so that the maximum liquid
ping upwards with an in situ liquid concentration of 80 inflow into the pump is obtained. An acoustic liquid level
to 90%. The liquid flows to the pump intake where it survey will accurately measure the distance to the liquid
mixes with liquid and some gas flowing upwards from level and calculate the percentage of liquid present in
the bottom perforated interval. the fluids surrounding the pump. This analysis aids in
Unless there is a detailed description of the reser- the efficient operation of PC-pumped wells6.
voir properties of each zone that would allow creating
a numerical simulation of the performance of each Acoustic Record Quality
reservoir, it is not possible to quantify and apportion PC pumps are generally driven by rapid rotation of the
precisely the produced gas and liquid flow rates for rod string that extends inside the tubing from the surface
each zone. However, the present example shows that to the pump. The flowing liquid provides some damping
having access to a pump intake pressure estimate that of the rod vibrations, and the frequency of the noise cre-
is independent of the acoustic analysis (such as from a ated by rotation is generally above the frequency band
downhole pressure sensor or PIP from a detailed analysis of the acoustic electronics. Therefore, there will not be
of dynamometer data) makes it possible to determine a significant level of background noise that might affect
the most likely distribution of fluid gradients and create detection of the echoes from the liquid level or the tub-
a qualitative map of the flow in the wellbore. ing joints, as seen in the example records in figure 8.23.

Mechanical Wellbore Properties
KB OFFSET 9.0 ft TUBING
AVG JT
BOTTOM WEIGHT
TOP (ft) OD (in) IO (in) LENGTH
(ft) (lb/ft)
(ft)
CASING
0 5,676 2.875 6.50 2.441 31.672
44.6 psi
PRODUCTION CASING
AVG JT
BOTTOM WEIGHT
TOP (ft) OD (in) IO (in) LENGTH
(ft) (lb/ft)
(ft)
0 6,044 7.000 23.00 6.366 40.000
WELL TEST: ANNULAR GAS FLOW = 91.6 MSCFD

OIL 90 BPD
WATER 513 BPD
GAS 140 MSCFD
LIQUID at 2,938 ft
PRESSURE 49.3 psi
GAS UP FLOW
ZERO NET LIQUID FLOW
LIQUID = 30.45%
LIQUID DOWNFLOW 4,636 ft–228.3 psi

4,814 ft–244.9 psi
STANDING VALVE
DEPTH 5,607.0 ft 5,186 ft

5,580 ft
PRESSURE SENSOR at
5,667' = 611 psi
5,680 ft
PLUGBACK
5,994 ft
COMPUTED PIP = 315.6 psi
DEPTH 60,44.0 ft
Figure 8.22 Analysis of pressure and flow distribution for a well producing from three perforated intervals

BN001
Sec -1 0 1 2 3 4 5 6 7 8 9 10 11 12
2
mV
-2
-4
-6
ft 0 2,000 4,000 6,000 8,000
BN0321A
Sec -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13
40
20
mV
-20
-40
ft 0 2,000 4,000 6,000 8,000
BN0372
Sec -1 0 1 2 3 4 5 6 7 8 9 10 11 12
150
100
50
mV
-50
-100
-150
ft 0 2,000 4,000 6,000 8,000
Figure 8.23 Examples of fluid level records acquired in wells producing with PC pumps

RECOMMENDED PROCEDURES the wellhead instruments and then to electronic

IN PUMPING WELLS hardware.
• Make sure that the wiring is not tangled, and is
In addition to general recommendations for proper
clear of walkways and work areas.
acquisition and analysis of acoustic records given in
• For wireless systems, verify adequate connectiv-
earlier chapters, the following focuses on procedures
ity and signal transmission to the base station.
related to measurements in pumping wells.
• Close off other connections to the casing annulus
Safety Considerations or flow line.
Personnel who perform acoustic fluid level measure- • Make sure the valve between the gas gun and
ments should read and understand the equipment’s the casing annulus or tubing is open before a
operation manuals and be familiar with this kind of shot is fired.
procedure. Furthermore, personnel should not attempt • Acquire several acoustic records to verify the
this procedure on their own until they have performed it consistency of the data.
under the direct supervision of an experienced operator. • Return the valves to their original position after
The minimum personal protection equipment rec- completing the test.
ommended include: hard hat, gloves, flame-resistant • Bleed the pressure between the gas gun and the
clothing, steel-toed boots, and safety glasses. Ensure wellhead.
that all tools are in proper working order and condition • For wired systems, disconnect the cables at the
prior to traveling to the well location. electronic hardware first and then at the wellhead
instruments.
Acquisition Workflow • Remove the gas gun from the wellhead.
• Return the well to the state it was in prior to testing.
The steps that follow indicate the recommended proce-
dure for acquiring acoustic records in a pumping well: Recommendations for Gas Gun
• Survey the well location for possible hazards or Connection to the Wellhead
unusual conditions.
The best results are obtained when the gas gun is con-
• Note the position of all valves, connections, at-
nected to the wellhead through a H-inch or larger fully
tachments, and so on, prior to initiating the test.
opening valve with the shortest possible distance (less
• Any alarm indicating the presence of H2S or
than 5 feet) between the gun and the casing or tubing,
hazardous gas will warrant stopping the job
as shown in figure 8.24.
immediately.
• Inspect the fluid level hardware before installing
it on the well. Stop the job immediately if you
notice any defects.
• Check the wellhead tubing and the casing pres-
sures.
• Stay alert if the equipment ratings exceed the
BEST SHOT DISTANCE FROM
well pressure. MICROPHONE TO
CASING SHOULD
• Check the threads on the wellhead valves for BE LESS THAN
corrosion, wear, or damage. 5 FEET.
• For 2-inch NPT, make 4H turns for a safe con-

nection to the wellhead.
• Use Teflon tape on the threads to prevent leaks
and protect threads from galling. Figure 8.24 Recommendations for connecting a gas
• For wired systems, first connect the cables to gun to a wellhead

The presence of pipe bends and/or elbows does not increase by leaving the flow line valve closed
affect signal transmission as long as the internal diam- because acoustic signal quality will improve with
eter of the piping is constant. Adaptors and pipe size increasing pressure, and the quality of echoes
reducers may be used if necessary, but they will result will improve.
in lower signal amplitude and quality. Shooting through • When all else fails, stop the surface pumping
a needle-type valve can obstruct or block the signal and equipment, and repeat the acoustic signal ac-
is not recommended. Shooting through a chemical pot quisition.
will cause a resonating signal and will give poor results.
Use the correct average tubing joint length because
Quality Control this value generally controls the accuracy of the cal-
culated distance to the fluid level. Enter the wellbore
In general, it is not necessary to stop the pumping opera-
deviation even if the pump is in the vertical section of
tion to acquire a good-quality acoustic record. Stopping
the well. It is required to compute the pressure distribu-
the pump will change the normal distance to the fluid
tion in the wellbore and at the pump intake.
level. In rod-pumped wells, acquire the fluid level data
Measure the surface pressure to calculate the well-
before stopping the pump to install dynamometer sensors.
bore and pump intake pressures. Measure the change
Pressure calculations require stabilized condi-
in surface pressure versus time to calculate the annular/
tions; verify that fluid level is steady. In wells operated
tubing gas flow rate.
intermittently (timers or POCs), the best practice is to
Enter accurate values for the tubing and casing
acquire fluid level records at the start and at the end of
diameters and weights; precise diameters and weight
the pumping period.
are required to calculate the gas flow rate and the liquid
The recommendations for accurately determining
fraction in the tubing or casing annulus. Enter correct
liquid level depth are:
values for oil and water densities. Apply the gas gravity
• Obtain a clear indication of the fluid level echo,
that was computed from acoustic velocity as determined
and obtain an accurate measurement of the round
through fluid level analysis using an equation of state
trip travel time. Inspect the well noise before
model or correlation. Such data is required to calculate
shooting. If excessive noise is present, close the
the pressure gradients of the various fluid mixtures.
valve between the gas gun and the wellbore, and
determine if the noise is coming from the surface
or downhole. SUMMARY
• Measurements should be repeated whenever Operating pumping wells at optimum efficiency with
excessive acoustic noise is present and fluid level minimal expense requires a clear understanding of the
echo is not clearly identifiable. distribution of fluids and pressures in the wellbore in
• Inspect the acoustic trace. If an insufficient liquid conjunction with knowledge of the inflow performance
level echo is obtained, the shot is not detected, or characteristics of the producing formation. Acoustic fluid
the data is otherwise unsatisfactory, then increase level surveys are the most practical and economical tool
the pressure differential between the gun’s volume available to obtain the necessary data to perform this
chamber and the wellbore and repeat acquisition. analysis for wells operated by the most common pump-
• For wired systems, inspect all connections and ing systems, such as rod pumps, ESPs, and PC pumps.
the acoustic cable for possible shorts. Successful analysis requires an accurate description
• For wireless systems, verify connectivity to the of the completion and pumping system in addition to
base station and signal level. accurate physical data. In rod-pumped wells, the most
• If the wellhead pressure is near atmospheric, accurate interpretation of the fluid level data is obtained
use the maximum available pressure in the gas when analyzed in conjunction with dynamometer and
gun chamber. Allow the wellhead pressure to pump performance information. Analyzing fluid level

data in the absence of dynamometer data or vice versa 2. J. N. McCoy, et al., “Evaluation and Performance of
often leads to the wrong conclusions. To validate the Packer-Type Downhole Gas Separators,” Proceed-
fluid level analysis, always take advantage of the cor- ings of the Production and Operations Symposium,
relation that exists between fluid level and the pump 2013.
dynamometer loads: a high fluid level (pump submer- 3. J. N. McCoy, et al., “Tubing Anchors Can Restrict
gence) yields a low value of the pump load (Fo) while Production Rates and Pump Fillage,” Unconven-
a low fluid level yields a high value of the fluid load in tional Resources Technology Conference, 2014.
the vicinity of FoMax.
4. J. Sparks and O. L. Rowlan, “Troubleshooting Rod
Whenever acoustic records are acquired in wells
Pumped Wells Using Tubing Level Shots,” Proceed-
with complex wellbore geometries, such as wells with
ings of the 54th Annual Southwest Petroleum Short
multiple tubular sizes or multiple producing zones, the
Course, 2014.
analyst must devote special care to verify that the results
obtained from conventional software are in agreement 5. J. Harris, et al., “Pumping Flumping Sucker Rod
with the personnel’s understanding of the flow condi- Wells,” Proceedings of the 49th Annual Southwest
tions likely to be present in the well. Special testing, Petroleum Short Course, 2010.
such as the liquid level depression test, may be needed 6. J. N. McCoy, “Analysis and Optimization of Pro-
to achieve an accurate analysis. gressing Cavity Pumping Systems by Total Well
Although the majority of fluid level records in pump- Management,” SPE Progressing Cavity Pump
ing wells are acquired in the annulus, it is also useful Workshop, 1996.
to determine the presence of a fluid level interface in 7. A. L. Podio, J. N. McCoy, and O. L. Rowlan, “Pump
the tubing and correlate it with the annular fluid level. Intake Pressure: Comparison of Values Computed
Doing so can yield valuable information about tubing from Acoustic Fluid Level, Pump Dynamometer
integrity. The tubing fill-up test is a very powerful tool to and Valve Check Measurements,” Proceedings of
determine pump volumetric efficiency. A large number the 48th Southwest Petroleum Short Course, 2009.
of pumping wells are also known to exhibit “flumping” 8. P. M. Bommer and A. L. Podio, The Beam Lift
characteristics—that is, they intermittently flow and Handbook (University of Texas Petroleum Exten-
pump. Special considerations and procedures can be sion Service, 2012).
used to operate these wells efficiently.
Example analyses presented in this chapter also il-
BIBLIOGRAPHY
lustrate some of the difficulties that must be overcome
when dealing with pumping wells that are completed and J. N. McCoy, et al., “Tubing Anchors Can Restrict Pro-
operated differently from the majority of wells that are duction Rates and Pump Fillage,” Proceedings of the
vertical, with uniform diameter tubulars and producing 54th Annual Southwest Petroleum Short Course, 2014.
from a single formation. J. N. McCoy, et al., “How to Maintain High Producing
Efficiency in Sucker Rod Lift Operations,” SPE Produc-
tion and Operations Symposium, 2003.
REFERENCES J. N. McCoy, et al., “Efficient and Accurate Monitoring
1. C. P. Walker, “Means For Measuring the Location of Rod Pump Well Performance Using Real Time Data
of Obstructions in Wells,” U.S. Patent 2,156,519, Processing and Visualization,” SPE Production and
May 1939. Operations Symposium, 2013.


Fluid Level Measurement Applications for Gas Wells 9-1
9
Fluid Level Measurement Applications for Gas Wells
In this chapter:
• Monitoring gas well performance with fluid level measurements in tubing and casing
• Determination and analysis of liquid loading
• Gas well troubleshooting
• Tubing and casing integrity testing
The principal objective when performing acoustic additional back-pressure acting on the formation when
measurements in a flowing gas well is to determine the there is liquid loading in the tubing. When gas velocity
quantity of liquid inside the tubing (or the annulus, when drops below the critical rate, production rates are reduced
the tubing is used for removing liquid from the wellbore by liquid accumulation in the tubing. Removing this
by means of a pump) and whether the produced liquid liquid requires applying a deliquifying technique, such
(1) is uniformly distributed over the length of the well as installing plungers or pumps, adding surfactants, or
as a mist or annular flow pattern or (2) has fallen back, redesigning the flow string to increase gas velocity1.
accumulating near the bottom of the well. The acoustic test in flowing gas wells is designed
In the first case, the gas flow rate is above a value to determine which flowing gradient conditions exist
defined as the critical rate, and the liquid is uniformly in a well. The test involves performing a series of fluid
distributed. The gas velocity is sufficient to continuously level and surface pressure measurements while the flow
carry liquid as a fine mist or small droplets to the surface, at the surface is stopped for a length of time sufficient
establishing a relatively low and fairly uniform flowing to identify the behavior and distribution of the fluids in
pressure gradient throughout the tubing. the tubing or tubing/casing annulus. The advantages of
In the second case, when the gas flow rate is below the acoustic test over wireline flowing pressure surveys
the critical rate, the gas velocity is not sufficient to carry include lower costs, because equipment is very portable,
all the produced liquid to the surface, and most of the and lower risks, because measurement tools are not
liquid accumulates and stays in the lower part of the introduced into a flowing well. An important byproduct
well. A flowing pressure traverse in the wellbore will of acoustic testing is determining the condition of the
show two different gradients: a light gas gradient above downhole hardware and integrity of the tubulars.
the gas/liquid interface and a heavier gradient in the
lower section of the well below the gas/liquid interface.
The gradient of the fluid below the gas/liquid interface DETERMINING LIQUID LOADING
reflects the liquid concentration, which is controlled OF A GAS WELL
by the gas flow rate. The liquid in this section of the The acoustic fluid level test is used to determine the tubing
wellbore recirculates in place, with zero net liquid flow, (or annular) pressure distribution in a flowing gas well
as the gas bubbles or slugs of gas percolate through the by momentarily shutting in the flow for the duration of
liquid, and only the gas flows to the surface. the test. An analysis of the acoustic fluid levels acquired
Knowledge of the flowing gradient and fluid dis- on gas wells can be used to determine the:
tribution in the well is important in determining the • Amount of liquid loading on the formation

at Austin
• Approximate gas inflow rate from the formation • Does the increase in tubing gas/liquid pressure
• Equivalent gradient of the gaseous liquid column displace liquid out of tubing?
in the tubing • What is the maximum production rate available
• Flowing BHP at the end of the tubing and at the from the well?
datum depth or the depth of the perforations2
The answers to these questions are important to
The fluid level test data should be analyzed while determine the type of flow behavior of the gas well and
it is collected at the well to answer the following well develop strategies to improve its performance.
performance questions: Flowing gas wells may be grouped into one of three
• Is gas flowing? If so, at what rate? different categories:
• Is there a gas/liquid interface in the wellbore? • Gas flow above the critical rate
• What is the depth to the top of the liquid in the • Gas flow below the critical rate
tubing and/or casing? • Static or shut-in
• What is the percentage of liquid in the fluid
column? The critical rate Qc is defined as the gas rate neces-
• How does the liquid level drop as the gas flow sary to keep the liquid in constant motion towards the
decreases? surface so that it is removed from the flowing conduit.
• How much liquid is in the tubing above the tub- Several empirical and theoretical relations3,4 have been
ing intake? developed and are used by the industry to determine
• What are the producing and static BHPs? the critical rate as a function of wellbore geometry,
• Is the reservoir inflow flow rate restricted because fluid properties, pressure, and temperature. Figure 9.1
of back-pressure from liquid loading? shows an example of a critical flow diagram that has
Turner Unloading Rate for Well Producing Water
3,000
4½ OD 3.958 ID
2,500 3½ 2.992
2⅞ 2.441
CRITICAL RATE, Mcf/Day
2⅜ 1.995
2,000 21/16 1.751
GAS VELOCITY
REMOVES LIQUID
1,500 WHEN Qg = Qc
1,000
500
LIQUID LOADING
PREDICTED
WHEN Qg < Qc
0
0 50 100 150 200 250 300 350 400 450 500
FLOWING PRESSURE, psi
Figure 9.1 Example of a critical rate diagram for different tubing sizes

been computed for various standard tubing sizes for the important to closely follow all recommended safety
flow of mixtures of water and natural gas at different procedures and perform a detailed risk assessment prior
pressures. For a given pipe size, the lines define the to performing the tests.
minimum flow rate required to keep the water in motion The most common wellhead attachment for flowing
at a given flowing pressure. gas well measurements includes an acoustic gas gun
For 2K-inch diameter tubing operating at a pressure with microphone and pressure transducer, connected
of 300 psi, the critical rate is about 550 Mscf/day. For rates to the tubing through a fully opening valve connection
less than this value, the tubing will be loaded with water. above the swab valve or at the point of connection of the
For rates above this value, the water will be uniformly tubing head pressure gauge. The ease of use, low cost,
distributed in the tubing as a mist, which will flow with the and reliability of gas guns have resulted in widespread
gas to the surface and will be produced continually. This use of this type of acoustic pulse generator.
value should be considered a theoretical quantity to be used In many cases, acoustic tests are also performed
as a guideline and design tool to help select an appropriate down the casing annulus, even when a packer is present,
tubing size for a given application. Flow conditions in to evaluate the integrity of the completion. The acous-
a given well will be affected by many parameters, such tic pulse travels through the gas in the wellbore and is
as actual liquid composition, dissolved solids, presence partially reflected by changes in cross-sectional area.
of surfactants, pipe roughness, and so on. Thus, the ap- The remaining energy is then reflected by the gas/liquid
plicable critical rate will vary from the predicted quantity. interface at the depth of the liquid level. These reflected
When the well is flowing at a gas rate less than the signals travel back to the surface of the well where they
critical rate, an acoustic fluid level test acquired down are detected by the microphone. If the liquid level echo
the tubing will usually show a liquid level echo. If the is not detected, then a larger pressure difference or larger
gas well is flowing gas at a rate greater than the critical volume chamber should be used to increase the energy
rate, then the gas/mist interface will be at or near the contained in the initial pressure pulse.
surface, and the initial fluid level shot will often not
show a distinct liquid level echo in the tubing. If the gas Implosion Method
well is shut-in for an extended period of time, the high- Because the wellhead pressure is generally high, the
pressure gas accumulates in the tubing, liquid droplets implosion method discussed in chapter 4 is preferred to
fall back and a gas/liquid interface is formed. In many generate the acoustic pressure wave. An exception should
cases, all the liquid is displaced out of the tubing bottom be made when the well gas contains hydrogen sulfide
and into the lowermost part of the wellbore. In a shut- at a sufficiently high concentration to create hazardous
in gas well, the acoustic record will often show echoes conditions. In these wells, the explosion method should
from the bottom of the tubing and the perforations as be used, as discussed below.
well as a liquid level echo very near the bottom of the
perforations. Using portable fluid level instrumentation Explosion Method
permits the operator to quickly conduct a simple, cost- At low wellhead pressures (generally less than 100 psi),
effective test and immediately identify underperforming an external pressure source will be required to charge
liquid loaded gas wells. In addition, the acoustic fluid the gas gun volume chamber to a pressure greater than
level data can be used to evaluate the integrity of the the well pressure in order to perform an explosion mode
completion, such as the presence of holes in the tubulars acoustic test. In most cases, compressed CO2 or N2 gas is
or other conditions that impair the well’s productivity. loaded into the gas gun volume chamber. In all cases, a
gas gun with a working pressure rating that exceeds the
Equipment Selection and Setup shut-in wellhead pressure should be used. Best results
Most gas wells operate at relatively high wellhead are obtained when shooting through a fully opening
pressures (from hundreds to thousands of psi), so it is swab valve, as shown in figure 9.2.

Figure 9.2 Examples of gas gun connections to gas wells
Depth Determination to the Liquid Level 3. Convert the RTTT to a distance by one of the meth-
In installations where a needle valve is present to isolate ods discussed in chapter 6: automatic collar count,
the pressure gauge normally connected to the tubing or downhole marker, or acoustic velocity.
casing, the needle valve should be replaced with a fully The acoustic record acquired inside the tubing may
opening H-inch high pressure ball valve. The very small not exhibit good quality echoes from the enlargements of
diameter of the needle valve seat distorts and reduces the inside of the tubing couplings. Thus, in some cases,
the sound transmission into the well, resulting in a very the automatic depth determination technique does not
poor quality acoustic record. yield a reliable liquid level depth. This depends on the
As discussed in detail in chapter 5, determining the well’s pressure level, the tubing type, and the gas proper-
distance to a liquid level reflection on an acoustic trace ties. Having access to a well completion diagram with
usually requires three steps: detailed description of all hardware present in the tubing
1. Identify correctly the echo on the acoustic trace that allows determining acoustic velocity from the depth of
reflected from the liquid level. known wellbore diameter changes and the correspond-
2. Determine the elapsed time from the firing of the ing measured RTTT. Usually, this is the best method for
shot to the reception of the echo from the liquid accurately determining the liquid level. Alternatively,
level. This time is defined as the round trip travel the acoustic velocity may be estimated from determining
time (RTTT). the specific gravity of a wellhead gas sample5.

ANALYZING TYPICAL GAS WELL concentration. This forms a fluid front with a higher
PERFORMANCE density that propagates downwards. When the density
contrast between the gas and the gas/liquid mixture
The following sections present some example data and
is sufficiently large, the transmitted acoustic pulse is
analysis for gas wells that are typical of each of the three
reflected at the interface and creates an echo that is
groups defined earlier. Additional examples of acoustic
clearly visible in the acoustic record. Figure 9.3 shows
measurements in gas wells are included in chapter 2.
the progressive fall of the “liquid” interface as recorded
Throughout the discussion, references to “tubing” also
during approximately 35 minutes after stopping the gas
include “annulus” unless otherwise indicated, since gas
flow at the tubing head.
wells can be produced through either.
Recommended Procedure
Fluid Level Records for Gas Flow
Above Critical Rate The following procedure assumes that acoustic measure-
ments are performed down the tubing with an instrument
When the gas flow rate is above the critical rate, any liquid
that has a working pressure greater than the maximum
being produced with the gas or condensing because of
shut-in wellhead pressure. It is also recommended that
temperature and pressure changes is usually uniformly
fluid level measurements be made in the casing/tubing
distributed in the tubing. The flowing gas velocity carries
annulus to verify completion integrity. The procedure
the liquid as a fine mist or small droplets to the surface,
for acquiring fluid level records in this type of gas well
and a relatively light and uniform flowing pressure
contains the following steps:
gradient is established in the tubing. At the stabilized
undisturbed flowing condition, liquid droplets are at the 1. Attach the acoustic instrument to the tubing head
surface, and a light mist exists in the tubing string from (swab valve).
the surface down to the bottom of the tubing6. 2. Acquire at least one acoustic record while the well
Acoustic surveys consisting of repeated fluid level is flowing.
records, acquired as the gas is flowing from the well, 3. Close the wing valve to the flow line.
have shown that a distinct echo from a “liquid” level 4. Acquire multiple acoustic records at short time
is usually not detected. This is because there is not a intervals (3 to 5 minutes).
significant contrast between the density of the gas and 5. Observe any increase in the tubing head pressure.
the density of the gas/liquid mist. In addition, a uniform 6. Observe echoes from falling gas/liquid interface.
mist or fog highly attenuates sound (especially higher
7. If possible, continue acquiring fluid level records
frequencies), causing the acoustic signal amplitude to
until the liquid level drops below the bottom of
deteriorate rapidly with distance. Therefore, detectable
tubing string.
echoes of the transmitted pulse are not created and ob-
served in the acoustic record, which primarily shows 8. Open the wing valve slowly, and return the well to
the large-amplitude signal of the shot firing, followed normal flow.
by the background acoustic noise created by the fluid 9. Acquire the acoustic record while gas is flowing at
as it flows through the wellhead piping and valves. the original tubing head pressure.
To analyze the fluid and pressure distribution in 10. Disconnect the acoustic fluid level instrument.
this type of gas well, the gas flow at the surface is shut 11. If possible, attach the instrument to the casing valve.
in, and a series of acoustic fluid level surveys are ac- 12. Acquire the acoustic record in the annulus
quired at constant time intervals of 3 to 5 minutes. As 13. Disconnect the acoustic fluid level instrument.
the gas velocity in the tubing decreases and the surface
pressure increases because of the interrupted flow, the Several measurements should be taken to ensure
liquid droplets start to fall back towards the bottom of the accurate detection of the gas/liquid interface and
the well, which creates a localized increase of the liquid establish the gaseous column gradient.

Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
16:48:04 RTT = 1.733 Seconds,

316.2 mV
Questionable Fluid Level Echo,

Explosion Shot
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
16:52:44 RTT = 2.779 Seconds,

316.2 mV
Explosion Shot
Liquid Level = 1,823 ft
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
16:56:59 RTT = 3.73 Seconds,

316.2 mV
Explosion Shot
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
17:02:53 RTT = 5.294 Seconds,

316.2 mV
Explosion Shot
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
17:08:27 RTT = 6.483 Seconds,

316.2 mV
Implosion Shot
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
17:11:17 RTT = 7.1 Seconds,

Implosion Shot
316.2 mV
Liquid Level Echo Clearly Detectable

Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
17:14:27 RTT = 8.027 Seconds,

Implosion Shot
316.2 mV
Fluid Level Below End of Tubing

at 5,256 ft
Figure 9.3 Acoustic records acquired during a gas well shut-in

Example Gas Well Flowing Above Critical Rate corresponding to the echo from the end of the tubing
The well is completed with 2K-inch tubing set in a was used as a downhole marker at a depth equal to the
packer at 5,596 feet. Three zones are perforated in the tubing string length. The distance to the liquid level
lower 4.5-inch casing at depths of 5,741 to 5,761 feet, on all nine shots was determined using the 1,312-ft/s
5,828 to 5,844 feet, and 5,914 to 5,936 feet. acoustic velocity, which was considered a representa-
At the time of the test, the well was estimated to be tive value, even though the average pressure in the
flowing at a rate of 750 Mscf/day up the tubing. With wellbore changed from 680 to 800 psi during the test.
a tubing head pressure of 644 psi, the 750-Mscf/day A plot of the depth of the gas liquid interface as
flow rate is above Coleman’s 4 critical rate for 2K-inch a function of time indicates that, on average, the gas/
tubing, and the gas/water mixture should be produced liquid interface front propagates to the bottom at a
at the surface in the mist flow regime. relatively constant speed since the points closely follow
The section of the well below the packer has a a straight line, as shown in figure 9.4. This behavior
diameter of 4.095 inches, resulting in a superficial gas has been observed in other wells with similar tubing
velocity of 1.97 ft/s. Based on various flow regime size and gas flow rates.
maps7,8, this section of the wellbore is significantly The analysis of these fluid level records can be
liquid-loaded, and the gas and water are flowing in the used to determine the average tubing fluid mixture
churn-slug regime. This seems to be validated by the gradient and estimate of the flowing BHP. A plot of the
production history that shows periods of constant gas computed gas/liquid interface pressures and height of
flow followed by periods of heading, with the gas rate the gaseous liquid column determined from the series
oscillating between 350 and 1,200 Mscf/day. of fluid level shots will usually also fall along a straight
line9, indicating that a constant pressure gradient ex-
Description of the Acoustic Tests ists below the gas/mist liquid level interface, as seen
Seven fluid level shots were acquired on this well using in figure 9.5.
the explosion technique (shots 1 to 5), and two records Since the relation is linear, the change in pres-
were acquired using the implosion technique (shots sure and height of the gas/liquid-mist interface from
6 to 7). As shown in figure 9.3, the acoustic velocity at least two fluid level measurements can be used to
could not be determined with certainty from the tubing calculate the gradient below the fluid level. Thus, it
collar’s recess reflections. At this high pressure, there is not necessary to continue acquiring data until the
should not have been a problem in seeing the echoes liquid falls all the way to the bottom of the tubing.
from the collar recesses. It is possible that the collar In this specific example, all the recorded points are
recess reflections are very weak due to: used to obtain a linear regression equation in order to
• Noise from the high gas flow rate being greater compute an average slope (-33.97 feet/psi), and the
than the amplitude of the reflected signals from corresponding mixture gradient is given by the recip-
the collars rocal of the slope (1/33.97 = 0.029 psi/ft). Knowing
that the liquid produced by the well is mainly water,
• Liquid droplets falling out of the mist, creating
the gradient value of 0.029 psi/ft is converted to an
an annular film, and covering the tubing collar
equivalent 6.8% of liquid present in the tubing at the
enlargements
time of the test (0.068 = 0.029/0.433). The gradient
Note that it is necessary to take several shots after of the dry gas above the gas/liquid-mist interface is
stopping the flow before the gas/liquid interface can about 0.018 psi/ft.
be detected with some confidence. Extrapolation of the pressure at the gas/liquid-
To analyze these series of tests, an average acoustic mist interface to a zero height of the gaseous liquid
velocity of 1,312 ft/s was determined using the last fluid will give a reasonably accurate estimate of the PBHP
level record, where the up-kick on the acoustic trace at 804 psi.

Liquid Level vs. Time—Gas Well 34

TIME, min.
0:00 5:00 10:00 15:00 20:00 25:00 30:00 35:00
0
1,000
GAS /LIQUID INTERFACE DEPTH, ft
FALL RATE = 146 ft/min
2,000
3,000
FALL RATE = 192 ft/min

4,000
5,000
6,000
7,000
Figure 9.4 Position of the gas/liquid interface as a function of time
Gaseous Column Height vs. Pressure From FLUID Level Data

Gas Well 34
4,500
HEIGHT OF GASEOUS LIQUID COLUMN, ft
4,000
3,500 REGRESSION EQUATION

COLUMN HEIGHT = -33.968 × PRESSURE + 27293
3,000
GASEOUS COLUMN
2,500
GRADIENT = 0.029 psi/ft
2,000
1,500
PBHP = 804 psi
1,000
500
0
660.0 680.0 700.0 720.0 740.0 760.0 780.0 800.0 820.0
GAS/LIQUID INTERFACE PRESSURE, psi
Figure 9.5 Height of a gaseous liquid column as a function of tubing pressure

Example Gas Well Flowing Below Critical Rate liquid level changes so that liquids previously held up
Acoustic fluid level measurements acquired in a flow- by gas flow begin to fall. The accumulation of liquids
ing gas well producing below critical rate will usually increases the gradient in the lower part of the tubing,
show a distinct liquid level echo some distance below causing an increase in BHP that eventually reduces the
the surface of the well. The flowing pressure traverse rate of gas inflow from the formation. Figure 9.7 presents
will show two distinct slope values: a very light gas pressure traverse graphs based on acoustic fluid level
gradient above the gas/liquid interface and a heavier records acquired while simultaneously measuring the
gaseous liquid gradient below the gas/liquid interface. wellbore pressure at the bottom of the tubing, using a
Below the liquid level, the flow is characterized as zero quartz pressure gauge located at 7,150 feet.
net liquid flow with gas bubbles or slugs percolating Four records were initially acquired while the well
through the liquid. Upon exiting the gaseous liquid was flowing with a tubing head pressure of about 60 to 65
surface, the gas flows the remaining distance up the psi. Subsequently, the flow was stopped, and additional
tubing to the surface, as illustrated in figure 9.6. This fluid level records acquired in succession. Note that,
pressure distribution characterizes what is defined as a as the tubing head pressure increases and liquid falls
liquid loaded gas well. back, the measured BHP also increases in response to
When gas flow from a liquid loaded well is shut in the decrease in gas flow rate and the pressure transient
at the surface, the pressure increases, gas flow velocity buildup in the formation. A more detailed description
in the tubing decreases, and the flow regime below the of this test is presented in chapter 2.
Acoustic Fluid Level Survey in a

Type 2 Well Loaded with Liquid
DEPTH
GAS GRADIENT ABOVE FLUID LEVEL
FLUID LEVEL
GASEOUS LIQUID
GRADIENT BELOW
GAS BUBBLES OR SLUGS

MOVE UP THROUGH LIQUID COLUMN
PRESSURE
Figure 9.6 Pressure traverse in a liquid loaded gas well

BMT 35 Pressure-Depth Traverses Before and After Shut-In
PRESSURE, psi
0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 400.00 450.00
0.00
1,000.00 FLOWING 1
FLOWING 2
2,000.00 FLOWING 3
FLOWING 4
3,000.00 SHUT-IN 1
DEPTH, ft
SHUT-IN 2
SHUT-IN 3
4,000.00
SHUT-IN 4
SHUT-IN 5
5,000.00
WIRELINE GAUGE SHUT-IN 6
MEASUREMENTS SHUT-IN 7
6,000.00 SHUT-IN 8
SHUT-IN 9
7,000.00 SHUT-IN 14
8,000.00
Figure 9.7 Pressure versus depth traverses
The four lines (flowing 1 to 4) correspond to the The first shut-in acoustic record corresponds closely
tubing pressure distribution when gas was flowing at to the condition that existed in the well when gas was
the surface and the wireline tool was being lowered into flowing. It may be considered that the pressure distribution
the well. Consequently, there is no data for the pressure corresponds to the average flowing condition. Subsequent
at 7,150 feet. graphs show how the pressure at the tubing head is in-
After hanging the BHP tool and shutting in the well, creasing and the gas/liquid interface is moving down as
the gradient of the gaseous liquid column is obtained by well as the gradual increase of the gradient of the gaseous
joining the pressure computed at the gas/liquid interface liquid column. The gradient increase corresponds to the
from the acoustic record with the pressure measured at liquid falling back to the bottom of the tubing as the gas
the bottomhole sensor. Note in the figure how the slope flow rate decreases. The last plot (Shut-in 14) was taken
of the lines (feet/psi) are fairly constant for the first two prior to retrieving the wireline tools and shows that a
records (Shut-in 1 and 2) but then decrease significantly 700-foot column of mostly liquid (87% liquid, consider-
as the tubing head pressure increases, corresponding to ing condensate and water with a gradient of 0.421 psi/
the increase in liquid concentration in the lower part of ft) has accumulated at the bottom of the tubing and the
the tubing (Shut-in 3 through 9). pressure at 7,150 feet has stabilized at about 404 psi (table
The gradient of the gas/liquid mixture calculated 9.1). Figure 9.9 shows the sequence of acoustic records
using the measured BHP and the pressure at the gas/ acquired during this test.
liquid interface is plotted in figure 9.8 as a function of It is important to note that an echo from the top of
shut-in time (pink curve). The gradient increases from a the wireline tools is clearly observed when the tools were
minimum of 0.05 psi/ft at shut-in to a maximum of 0.405 located at 2,500 feet in depth. The echo’s round trip travel
psi/ft after 90 minutes of shut-in. time is used to calculate an accurate average acoustic

Gaseous Column Gradient

0.45
0.40
0.35
GRADIENT, psi/ft
0.30
0.25
0.20
0.16
0.10
0.06
0
-150.000 -100.000 -50.000 0.000 50.000 100.000 150.000 200.000
TIME, MINUTES
COMPUTED FROM ANNULAR S CURVE COMPUTED FROM MEASURED PRESSURES
Figure 9.8 Gaseous column gradient changes in time
velocity for the gas in the tubing in order to calculate acoustic fluid level measurements acquired after stopping
the depth of the gas/liquid interface. After the gas flow the tubing flow should yield a fairly accurate estimate of
is stopped at the surface, the gas/liquid interface drops the gaseous liquid column gradient and the flowing BHP.
rapidly and stabilizes near the bottom of the tubing after
about 28 minutes. The following table summarizes the Estimating BHP from Fluid Level
results of the acoustic measurements during the test. Measurement in Tubing
The majority of the liquid accumulated at the bot- The percent liquid in the annular gaseous column is
tom of the tubing corresponds to the volume of liquid normally computed from the generalized correlation
that was suspended in the gaseous liquid column when (the “S” curve discussed in chapter 6, fig. 6.13) that was
the gas was flowing normally. Assuming the capacity of developed from field data of annular gaseous column
the tubing (bbl/feet) is fairly uniform from top to bottom depression tests10. The correlation is thus applicable to
(same weight/feet of tubing pipe), the ratio of the height annular flow with some confidence, but it is suspected
of liquid to the length of the tubing [(698 × 0.87)/7,150 to be less accurate when applied to tubular flow. Having
= 0.085] corresponds to the fraction of liquid that was the pressure data from the quartz gauge gives invaluable
present in the gaseous liquid column when the gas was information regarding the applicability of the “S” curve
flowing out of the tubing. to gas flow in the tubing.
The results from this field test indicate that, in a liquid Figure 9.8 compares the measured pressure at
loaded well producing gas below critical rate, the first few 7,150 feet with the pressure computed from each fluid

Table 9.1
Test Summary Table
Elapsed Elapsed Computed

Time, Time, Measured Tubing Gas/Liquid Height of
from from Tubing Pressure Buildup Interface Gas/Liquid Gaseous
Start of Flow Pressure, Buildup, Time, Pressure, RTT, Interface Liquid
Test Shut-In psi psi minutes psi seconds Depth, ft Column, ft
0:00:00 61.7 7.5 1.50 68.1 5.566 3,626 3,607
0:37:33 57.7 15.8 2.50 63.2 5.043 3,332 3,901
0:36:13 59.0 8.4 1.25 64.7 4.752 3,165 4,068
0:19:25 60.2 10.4 1.00 64.5 3.819 2,530 4,703
0:19:06 63.4 12.9 1.25 67.5 3.435 2,276 4,957
2:08:13 0:00:00 63.0 17.5 2.00 67.6 3.611 2,392 4,841
2:12:44 0:04:31 95.6 6.6 2.00 103.5 4.400 2,926 4,307
2:17:36 0:09:23 109.4 6.1 4.00 120.4 5.463 3,632 3,601
2:22:56 0:14:43 116.9 2.1 2.00 129.9 6.359 4,233 3,000
2:27:54 0:19:41 121.7 2.6 4.20 138.4 7.508 4,967 2,266
2:32:40 0:24:27 124.4 0.3 4.00 143.8 8.975 6,000 1,233
2:38:54 0:30:41 124.9 0.239 4.00 144.9 9.204 6,148 1,085
2:43:32 0:35:19 125.2 0.208 4.00 145.4 9.269 6,192 1,041
2:49:37 0:41:24 125.5 0.161 4.00 145.9 9.356 6,257 976
2:54:38 0:46:25 125.8 0.153 4.50 146.4 9.420 6,317 916
3:02:38 0:54:25 126.1 0.081 4.00 147.8 9.518 6,382 851
3:07:43 0:59:30 126.2 0.154 8.25 148.1 9.568 6,415 818
3:17:48 1:09:35 126.4 0.200 13.00 148.5 9.662 6,478 755
3:27:47 1:19:34 126.4 -0.104 6.00 148.7 9.730 6,523 710
3:37:35 1:29:22 126.2 -0.062 3.50 148.4 9.736 6,528 705
3:51:05 1:42:52 126.0 -0.57 2.00 148.2 9.740 6,529 704
4:10:33 2:02:20 125.6 -0.093 4.25 147.0 9.762 6,546 687
4:24:35 2:16:22 125.4 -0.122 5.00 147.6 9.765 6,548 685

9:16:55 No tool in well—just shut-in for 3 minutes. Tool at 7,150 ft. Closed in well—0 minutes.
RTT = 5.566 sec. RTT = 3.611 sec.
Tool at 2,500 ft. Well flowing—just shut-in to shoot. Tool at 7,150 ft. Closed in well for 5 minutes.
RTT = 5.04 sec. RTT = 4.4 sec.
Tool at 5,000 ft. Well flowing—shut-in to shoot. Tool at 7,150 ft. Closed in well for 10 minutes.
RTT = 4.752 sec. RTT = 5.463 sec.
Tool at 6,000 ft. Well flowing—just shut-in to shoot. Tool at 7,150 ft. Closed in well for 15 minutes.
RTT = 3.819 sec. RTT = 6.395 sec.
Tool at 7,000 ft. Well flowing—after 10 minutes stop. Tool at 7,150 ft. Closed in well for 20 minutes.
RTT = 3.435 sec. RTT = 7.508 sec.
Tool at 7,150 ft. Closed in well for 28 minutes.

RTT = 8.975 sec.
Figure 9.9 Sequence of acoustic records

level record, using the annular “S” curve correlation to effective gradient in the tubing, which occurs at high
determine the effective gradient of the gaseous liquid gas flow rates, is also lower (12%) compared to the
column. There is a significant difference at the beginning minimum gradient for annular flow (19%).
of the shut-in period between the measured pressure (314 The conclusion is that the annular gaseous column
psi) and the computed pressure (464 psi), indicating that “S” curve generally overestimates the percentage liquid
the “S” curve is estimating a larger percentage of liquid present in the gaseous column when it is applied to tubular
in the gaseous column than is actually present. As the flow in liquid loaded gas wells. The analysis of acoustic
gaseous liquid column collapses and accumulates near fluid level surveys acquired in the tubing while the gas/
the bottom of the tubing, the difference between the liquid interface is falling should consider this effect, and
two values decreases. Towards the end of the test when a correction ranging from 10% to 20% reduction of the
the gas flow is minimal, the computed (407.4 psi) and liquid fraction, given by the conventional “S” curve,
measured (402.84 psi) values differ by 1.1%. should be applied before calculating the BHP.
The computed gradient in the tubing converted to A more reliable estimate of the percentage of liq-
fraction of liquid is presented in figure 9.10, overlaid uid in the tubing below the gas/liquid interface can be
on the annular “S” curve. obtained by performing a liquid level depression test
The tubular “S” curve shown in figure 9.10 indi- with two fluid level measurements undertaken shortly
cates that, for flow in the tubing, the percent liquid cor- after stopping the flow at the surface. These measure-
responding to a certain gas flow rate, given in terms of ments should be used to relate the depression of the gas/
a tubing pressure buildup rate of dp/dt, corresponds to liquid interface to the increase in wellhead pressure in
a lower percent liquid when compared to the same gas order to establish the gaseous column gradient used to
flow conditions in an annulus. The predicted minimum compute the PBHP.
Annular and Tubular “S” Curves
100,000
10,000
(dp/dt) × L
1,000
100
10
0 20 40 60 80 100 120
EFFECTIVE LIQUID %
ANNULAR “S” CURVE TUBULAR “S” CURVE FROM WELL 35
Figure 9.10 Comparison of annular and tubular “S” curves

Determining Static Bottomhole Pressure of the echo generated by the change in internal diameter
The use of acoustic surveys to determine the static (when shooting down the tubing) or external diameter
shut-in pressure in gas wells is an accepted and ac- when shooting down the casing. The amplitude of the
curate practice. It provides operational and economic repeated echoes decays exponentially, as seen in figure
advantages over the use of downhole gauges in that the 9.11, so it is generally possible to view the echoes
measurement does not require introducing tools in the generated at greater depth from other features of the
well. In a shut-in gas well, liquid is often pushed out of flow string and the liquid level. If the liquid level were
the wellbore, and the liquid level is usually identified in the upper part of the tubing, it is possible to confuse
at or near the bottom of the perforations. The downhole its echo with the repeats from the safety valve. In this
marker method is the most appropriate technique to case, the identification of the liquid level echo can be
determine distances in a shut-in gas well. It is important accomplished by making the liquid level move through
to record the measured depth to all known downhole changes in the pressure conditions over a period of time.
reflectors, such as the depth to: Three wells in the North Sea had wireline-conveyed
pressure bomb surveys and static fluid level surveys
• Gas lift valves/mandrels
performed at the same time. Initially, when the static
• All zones of perforations
pressures determined from the two methods were com-
• The subsurface safety valve, crossovers, and pared, differences in excess of 100 psi were observed.
liners The error in the acoustic BHP calculations was due to
The most accurate depth to the static liquid level is using incomplete wellbore information and applying
determined by identifying and using the reflection from average default values of acoustic velocity and fluid
the downhole anomaly that is closest to the liquid level. properties for the calculation of liquid level depth and
If the wellbore is deviated, then the deviation survey pressure. After inputting the correct representative well
must be used in the calculations to determine pressures information and reprocessing the acoustic fluid level
based on true vertical depths. To calculate the fluid data, the differences between the wireline-conveyed
gradients of the liquids, both the oil and water produc- pressure bomb’s static pressures and the acoustic fluid
tion rates and their corresponding gravities should be level’s static values were reduced to only a few psig.
known. In deep wells, the bottomhole temperature has Figure 9.12 shows a section of the detailed description
a significant impact on the acoustic velocity and the of the wellbore that indicates a significant enlargement
pressure gradient. Errors will result if default values for of the internal diameter at a depth of 13,204 feet due to
temperature are used in the calculation. crossover from 3H- to 4H-inch threads.
Figure 9.11 displays the wellbore schematic for an The acoustic record acquired inside tubing in this
offshore gas well and the corresponding acoustic trace well (fig. 9.13) shows the echoes from the downhole
obtained down the tubing. An accurate wellbore sche- safety valve between 1 and 3 seconds and the echoes
matic is essential for interpreting the acoustic record from the crossover and the mill-out extension at about
shot inside the tubing since all components that exhibit 18 seconds, followed by the echo from the liquid level
changes of internal diameter will generate echoes of the (LL) at 19.161 seconds.
acoustic pulse. Knowing the depth to each item allows a The depth of the liquid level is computed using
very accurate estimation of the acoustic velocity in each the average acoustic velocity of 1,499 ft/s, which was
section of the wellbore, and the user can pinpoint the calculated from the round trip travel time corresponding
depth to the liquid level. The figure shows the relation- to the echo (upward deflection due to area enlargement)
ship between the echoes observed in the acoustic record from the crossover at 13,204 feet in depth, as shown by
and the tubing description from the wellbore diagram. the dashed line in figure 9.14.
When a subsurface safety valve is present in the The conversion from fluid level depth to SBHP
upper part of the tubing, there will be numerous repeats should consider the in situ density of the produced fluids

316.2 mV
Sec
0
0
MULTIPLE REPEAT
ECHOES FROM
SUBSURFACE
SAFETY VALVE
2
500
4
1,000
6
1,500
8
10
2,000
PACKER BORE
AND EXTENSION
12
2,500
14
PERFORATIONS
2,500
16
3,000
18
LIQUID LEVEL
Figure 9.11 Offshore gas wellbore schematic and corresponding acoustic trace acquired down the tubing after long
shut-in time

Wellbore Diagram
INTERNAL
DIAMETER
680.62 PUP JOINT 3½ in. FOX 7.7 PPF 3.008 3.897
12,123 7 in. LINER PBR TOP TIW
687.77 TUBING (312) FOX 2.943 3.878

3½ in. 7.7 PPF
13,133.96 PUP JOINT 3½ in. FOX 7.7 PPF 3.068 3.995
13,141.10 FLOW COUPLING CAMCO 2.969 3.910
13,146.83 SEATING NIPPLE CAMCO 2.812 3.905

2.812 D
13,156.27 TUBING JOINT (1) 2.943 3.878
13,203.32 CROSSOVER 3½ in. FOX FOX 3.000 4.500

BOX × 4½ in. FOX PIN
13,204.55 ANCHOR BAKER 3.877 5.472

81 - 47 KC22
13,207.09 PERMANENT PACKER BAKER 3.875 5.865

85 - 47x38 SABL-3
13,211.69 MILLOUT EXTENSION BAKER 4.420 5.020

80 - 47
13,216.90 CROSSOVER 3½ in. FOX BAKER 3.015 5.460

BOX × 4½ in. FOX PIN
13,225.01 TUBING (312) FOX 2.943 3.878

3½ in. 7.7 PPF
Figure 9.12 Detail of the wellbore completion

Sec 0 2 4 6 8 10 12 14 16 18 20
10.0 mV
0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
19.161 sec 453.087 Jts 14,362.86 ft
Liquid level extrapolated from marker selection

of a downhole marker.
Crossover 3½ to 4½ above packer
Figure 9.13 Acoustic record corresponding to the wellbore in figure 9.12
(in this case, mainly gas and condensate, which are a Verifying the actual position of the downhole valve can
function of pressure and temperature). An appropriate be done efficiently by acquiring acoustic fluid records
equation of state should be used to calculate the gradi- down the tubing when the valve is in the open position
ent of all the fluids, as described in chapter 6. Figure and then repeating the test after closing the valve, as
9.15 shows the result of the SBHP (reservoir pressure) shown in figure 9.16.
calculation. The upper acoustic record shows echoes between 17
The computed SBHP of 1,692 psi compares very and 20 seconds that correspond to the downhole tools
favorably with the pressure of 1,697 psi measured by detailed in figure 9.12. This indicates that the down-
the wireline pressure bomb. hole safety valve is at least partially open. The lower
acoustic record shows the signal recorded after moving
Testing of Downhole Safety Valve Operation the safety valve to the closed position. Note that only
Safety regulations require periodic testing of proper repeated echoes from the closed valve are recorded,
operation of surface controlled subsurface safety valves and the first occurs at 0.966 seconds, corresponding
(SCSSV). The procedures to be followed for the test vary to the safety valve depth of 725 feet. This verifies that
depending on the actual location of the well but always the valve moved from the open to the closed position.
involve switching the valve from the open to the closed It does not verify or test whether the valve is able to
position through actuation of the controls at the surface. sustain a differential pressure and/or would not leak.

Determine Liquid Level From Marker Depth
17.25 17.50 17.75 18.00

3.2 mV
Downhole Marker Indicator

Indicator @ 17.615 sec
< --- Left Right --->
Enter Joints to Downhole Marker Show Full Shot Trace
416.53 Ave. Joint Length 31.7 ft Show One Second Interval of Shot Trace
Enter Depth to Downhole Marker Scale Up Scale Down
13,204 ft
Calculated Results
Description of Downhole Marker:
Liquid Level Depth 14,362.86 ft
Crossover 3½ to 4½ above packer Accoustic Velocity 1,499.18 ft/s
Calculate Done
Figure 9.14 Detail of the echo from the 3½- to 4½-inch crossover
Pth = 1,250 psi (The high-frequency noise at about 10 and 23 seconds

BHP Pgl = 1,501 psi are random signals due to structural vibrations caused
Ps = 1,692 psi
Casing Pressure BHP Collars by offshore platform activities.)
Deviated Wellbore Well State:
In situations where the downhole safety valve is not
Casing Pressure
1250.0 psi (g) Static
operating properly (because it is either stuck open or
Casing Pressure Buildup stuck closed), the acoustic records obtained in the open
0.0 psi and closed positions would not be significantly different,
Oil Column Height
4.00 min MD as shown in figure 9.17.
Gas/Liquid Interface Pres. 615 ft Note in both records that the echo at about one sec-
1501.5 psi (g) TVD
ond, due to the presence of the SCSSV, is followed by
Liquid Level 612
echoes from deeper downhole, indicating that the valve
MD 14362.86 ft Oil Column Height
TVD 9633.93
did not change position in response to the control signals
MD
Formation Depth 615 ft from the surface. Also, note the random high-frequency
MD 14978 ft TVD noise due to platform activity.
612
TVD 10246
Reservoir Pressure (SBHP)

1692.13 psi (g)
Figure 9.15 Fluid and pressure distribution in the shut-
in well

SCSSV Operation

OPEN
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
10.0 mV
Select Liquid Level Depth Determination Casing Pressure BHP Collars CLOSED
Sec 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
100.0 mV
Indicator @ 13.173 sec Apply Low Pass Filter Move Indicator <— Left Right —>
Figure 9.16 Acoustic records acquired in the tubing in a shut-in gas well with a properly operating SCSSV
APPLICATIONS TO TROUBLESHOOTING As discussed in detail in chapter 3, the polarity of

GAS WELLS the recorded echoes indicates either enlargements or
In addition to determining the flowing characteristics of reductions in the cross-sectional area of the casing-
gas wells, acoustic surveys provide a tool to visualize tubing annulus or in the internal diameter of the pipe
the condition of the tubing, casing, perforations, and if shooting inside the tubing. This characteristic of the
other downhole features without needing to introduce acoustic signals provides a very useful tool to identify the
wireline sensors into the tubing11,12,13. presence of known or unknown dimensional anomalies

Faulty Safety Valve

OPEN
Sec 0 2 4 6 8 10 12 14 16 18
31.6 mV
Indicator @ 15.999 sec Apply Low Pass Filter Move Indicator <— Left
Select Liquid Level Depth Determination Casing Pressure BHP Collars CLOSED
Sec 0 2 4 6 8 10 12 14 16 18
31.6 mV
Indicator @ 16.647 sec Apply Low Pass Filter Move Indicator <— Left
Figure 9.17 Acoustic records acquired in the tubing of a shut-in gas well with a SCSSV stuck open

present in the wellbore. Increases in area commonly by features inside the well, not just signals caused by
include perforations, open hole, sliding sleeves, mill-out external random events.
extensions, parted casing, parted tubing, tapered tub- The upper trace shows a significant echo with a
ing, and the end of the tubing string. Decreases in area downward deflection (at about 7.2 seconds) followed by
commonly include liner top, tubing anchors, packers, a large up-kick (at about 10.2 seconds) that corresponds
landing nipples, paraffin/scale deposits, subsurface safety to the depth of the open-ended tubing and the enlarge-
valves, liquid level, cement top, and so on. ment of the wellbore at the end of the annulus. Based
Figure 9.18 displays acoustic records acquired by on the available completion record, the down-kick at
shooting down the tubing of a gas well. Several records about 4,750 feet is not due to a known tubing feature, so
were acquired to verify that the observed echoes are it could be caused by scale or paraffin deposits creating
consistently present in the record and thus generated a tight spot in the pipe. The lower trace, acquired in a
Using Acoustic Surveys to Identify What is Downhole

At 4,750 ft, TIGHT SPOT IN TUBING
Sec 0 2 4 6 8 10
END OF
TUBING
10.0 mV
TIGHT
SPOT
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000
JTS/sec 20.8768 sec Acoustic Vel. 1,377.87 sec RTTT 10.199 sec 205.397 Jts 6,778.09 sec
At 5,050 ft ⅛ in, HOLE IN TUBING
Sec 0 2 4 6 8 10 12
⅛ in HOLE
31.6 mV
LIQUID
LEVEL
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
JTS/sec 19.8413 sec Acoustic Vel. 1,285.71 sec RTTT 11.664 sec 233.654 Jts 7,570.38 sec
Figure 9.18 Examples of identification of downhole features from the polarity of the acoustic echoes

different well, shows a significant echo with upwards

deflection near 5,000 feet followed by the echo from
the liquid near the lower part of the tubing. A pressure
integrity test verified the presence of a hole, which upon
inspection of the tubing string was about 1/8 inches in
diameter, as seen in figure 9.19.
When multiple echoes are observed in an acoustic
record, the correct interpretation and analysis becomes 1.995 inch
complicated and may require constructing a characteris-
tic diagram showing the position and polarity of the wave
front as a function of time, as discussed in chapter 3.
Figure 9.20 shows an acoustic record where five distinct Figure 9.19 Hole in a gas well’s tubing string caused
echoes are observed with at least two echoes exhibiting by corrosion
Multiple Echoes
Which is Liquid Level?
1–Upkick 2–Down 3–Down 4–Upkick 5–Upkick
Select Liquid Level Depth Determination Casing Pressure BHP Collars De-Liquification
Sec 0 2 4 6 8 10 12 14 16 18 20 22
4
31.6 mV
3 5
1 2
Go to Automatically Selected Kicks
<< Prev Kick Next Kick >> Reset
Scale Shot Trace
Scale Up Scale Down Reset
[Alt-1] Gun Parameters

Pulse Type: Explosion Implosion
Shut Down: Casing Tubing Advanced Options ? < Pg Up Pg Dwn >
Figure 9.20 Acoustic record acquired inside the tubing of a gas well with a shallow corrosion hole

the downward deflection that could indicate the echo the acoustic wave is totally reflected at the gas/liquid
from the liquid level (echoes 2 and 3). To correctly interface, and changes in the cross-section submerged
identify the echo from the liquid level, the first step is in the liquid cannot generate measurable echoes.
to review the mechanical description of the tubing to The initial analysis must be validated by construct-
identify all changes in cross-sectional area that could ing the characteristic diagram of the wave presented in
generate echoes. In this case, the internal diameter of figure 9.21. The diagram shows how the polarity of the
the tubing is uniform, with only a landing nipple and repeat echo (3) from the hole in the tubing is inverted
wireline entry guide near the bottom. Thus, the initial to a downwards deflection when the wave encounters
analysis of the record indicates that the echo at about the hole the second time after the initial echo (1) is
6.25 seconds was caused by a possible hole, and the reflected back down at the tubing head and back up at
echo from the liquid level is most probably the signal the hole. The additional up-kicks (4 and 5) are gener-
at 10.7 seconds. The problem with this initial analysis ated by multiple internal reflections of the wave as it
is the presence of a second downwards deflection past travels between the depth of the hole and the depth of
the echo from the supposed liquid level. In general, the liquid level.
31.6 mV Assume Hole in the Tubing–Wave Path Analysis

Sec 0
1 2 3 4 5
2
RTTT = 6.256
4
1
6
HOLE IN TUBING? 10.714

8
RTTT = 4.458
2
10
LIQUID LEVEL?
12
3
14
RTT1 = 6.256 UPKICK

4
16
RTT2 = 10.714 DOWNKICK

5
RTT3 = 2 x 6.256 = 12.512 DOWNKICK
18
RTT4 = 6.256 + 2 x 4.458 = 15.172 UPKICK

20
RTT5 = 2 x 6.256 + 4.458 = 16.97 UPKICK

22
Dashed Arrows Indicate Polarity Inversion
Figure 9.21 Characteristic wave path diagram corresponding to the presence of a hole in the tubing above the liquid
level

The detailed characteristic diagram therefore veri- decline. Efforts to return production rate to its previous
fies that the initial analysis of the acoustic record was level often fail when various methods to deliquify the
correct and that the tubing string should be pulled and well are applied. The rate reduction occurs gradually and
the faulty joint replaced. increases as the hole enlarges, so reservoir depletion is
also considered as a possible cause of the problem.
HOLES IN GAS WELL TUBING Figure 9.22 shows the production behavior of a
Based on extensive field data, including acoustic records, gas well where the growth of a corrosion hole caused
it can be concluded that the presence of holes in the tub- a significant drop in gas flow rate that was originally
ing of gas wells is a fairly common problem that often diagnosed as a liquid loading problem.
seems to be unexpected and overlooked by the operator. The presence of the hole was verified with fluid level
One complication is that the performance of a gas well measurements down the tubing and down the casing. After
with a hole in the tubing is often misdiagnosed as liquid replacing the faulty tubing, the production rate returned
loading since the hole also causes the production rate to to the level exhibited before the hole was generated.
64 2,250
DROP IN PRODUCTION
DUE TO LIQUID LOADING? 2,000
56
SMALL HOLE
BEGAN TO CAUSE
PROBLEM 1,750
48
1,500
FOUND HOLE IN TUBING
40 WITH FLUID LEVEL SHOT
1,250
Mscfd
E3m3
32
1,000
DRAMATIC DROP IN RATE REPLACED

24
TUBING AND
820 Mscfd 250 Mscfd 750
GAS FLOW
RETURNED TO
16 2006 RATE
500
TURNER CRITICAL
320 Mscfd
8
250
0 0
2005 2006 2007 2008
TIME, years
Figure 9.22 Production history of a gas well with a hole in the tubing

ACOUSTIC SURVEY IN PACKER-LESS the tubing and in the annulus at the tubing intake depth
GAS WELL are equal. Note that this assumes that a check valve is
not installed at the bottom of the tubing string.
In a gas well operating at stabilized conditions but at a
The difference in pressure at the surface between
rate less than the critical rate (so that there is a gas/liquid
tubing and casing can be attributed primarily to the dif-
interface inside the tubing), the fluid level in the casing ference in density between the fluid inside the tubing
should be at the depth of the tubing intake, as shown and the gas in the casing. This assumes that the frictional
in figure 9.23. The annular volume is completely filled pressure loss due to upwards flow in the tubing is neg-
with gas, and all fluid produced from the formation is ligible compared to the hydrostatic pressure loss due to
flowing up the tubing. the liquid concentration.
For these conditions, the volume of liquid accu- The following relation gives an estimate of the
mulated in the tubing can be estimated from the differ- volume of liquid accumulated inside the tubing:
ence between the flowing tubing head pressure and the Vliquid = 0.00224176 × (Pcasing - Ptubing)
casinghead pressure, since the pressure at the bottom of × (Tbg ID_^2) ÷ SpGr of Liquid Eq. 9.1

CASING
Sec 0
Casing Fluid Level Should be at EOT

EXPLOSION
(ft)
0
9,000
2
1,500
ACOUSTIC
4
VELOCITY
1,273.8 ft/s
3,000
9,500
SG = 0.72
6
MARKER:
4,500
TOP PERF
8
10,364 ft
10,000
16.281 sec
6,000
10
10,500
EOT
12
7,500
14
LIQUID LEVEL
11,000
9,000
10,536 ft
16.548 sec
DM
16
10,500
LL
11,500
18
Figure 9.23 Annular fluid level in a liquid loaded gas well operating at stabilized conditions

where Pcasing and Ptubing are in psi, tubing ID is in inches, Using the corresponding pressure and height values
and Vliquid is in bbls. shown in figure 9.24 in the equation above, the result is:
Having determined the height of the gaseous column
Gaseous column gradient = (512 – 83) ÷
in the tubing, an estimated gradient is computed from
(10,536 – 8469) = 0.212 psi/ft
the fluid level record as:
Knowing that the liquid is a mixture of brine and
Gaseous column gradient = (Pcasing – Ptubing)
condensate that has a specific gravity of 0.81, the liquid
÷ (Gaseous column height) Eq. 9.2
fraction is computed as:
where the gradient is in psi/ft and the gaseous column
Estimated liquid fraction = 0.212 ÷ (0.81 × 0.433)
height is in feet.
= 0.604, or about 60% liquid
Knowing the specific gravity of the liquid (water
+ condensate), the estimated gradient of the gaseous Because the pressure values are easily measured
liquid column is converted to a fraction of liquid by the at the surface, this estimated gaseous column gradient
following relation: assumes for the sake of simplicity that the gas column
contribution to the pressures at the bottom of the tub-
Liquid fraction = Gaseous column gradient ÷
ing and annulus are about equivalent. A more accurate
(Specific gravity of liquid × 0.433) Eq. 9.3
estimate is obtained by using the pressure at the gas/
Pressure at EOT Should Be Equal
Tubing Pressure Tubing Shot Figure 9.24Casing

Pressure
Shot distributionCasing
in Pressure
a gas well without a packer
Sec
Sec
(ft) 0
82.5 psig 511.8 psig

Explosion
Explosion
(ft) 0
0
0
Casing Pressure Buildup Casing Pressure Buildup
0.3 psi 0.1 psi
min min
1,000
1.50 2.25
2
2
1,500
Gas/Liquid Interface Pres. Gas/Liquid Interface Pres.

2,000
104.9 psig 666.4 psi (g)

4
Liquid Level Depth Liquid Level Depth

3,000
3,000
MD 8468.55 ft MD 10535.96 ft
6
6
4,000
Tubing Intake Depth Tubing Intake Depth

4,500
MD 10536.00 ft MD 10536.00 ft
8
8
5,000
TVD 10536.00 TVD 10536.00

6,000
6,000
Formation Depth Formation Depth

10
10
MD 10644.00 ft MD 10644.00 ft
7,000
12
12
7,500
Tubing Intake Tubing Intake

658.1 psig 666.4 psig
8,000
PBHP PBHP
14
14
9,000
702.0 psig 710.3 psig

16
10,500
658 psig
18
Figure 9.24 Pressure distribution in a gas well without a packer

liquid interfaces if it is possible to compute them using over wireline flowing pressure surveys include lower
the gas properties. In this example, the pressure values costs, since equipment is very portable, and lower risks,
are 104.9 psi in the tubing and 658 psi in the casing. since measurement tools are not introduced into the well.
Using these values, the gaseous column gradient is 0.224 The acoustic survey also allows visualizing downhole
psi/ft and the liquid fraction is 0.64, which are of the features to facilitate developing a better understanding
same order of magnitude of the values computed with of conditions present in the wellbore. In particular, it
the surface pressures. allows detecting holes in the tubulars that cause poor
There are several possible reasons that cause the production performance. The reliability of the analysis
annular liquid level to not be at the depth of the tubing of the acoustic survey is enhanced by devoting sufficient
intake in some wells. The most common include: time to the acquisition of multiple records and using
• Erroneous identification and analysis of the accurate wellbore descriptions.
acoustic echo attributed to the liquid level. This
may be due to incorrect acquisition procedures
or the presence of multiple echoes from perfora- REFERENCES
tions, liners, tapers, and so on 1. J. Lea, H. Nickens, and M. Wells, Gas Well Del-
• The presence of a hole in the tubing that allows iquification (Gulf Professional Publishing, 2003).
gas from the casing to enter and mix with the 2. O. L. Rowlan, J. N. McCoy, and A. L. Podio,
tubing flow “Acoustic Liquid-Level Determination of Liquid
• Surface flow of gas from the casing due to a Loading in Gas Wells,” Western Regional/AAPG
leaky valve Pacific Section/GSA Cordilleran Section Joint
• Cross-flow between perforated intervals in reser- Meeting, 2006.
voirs with different pressures 3. R. G. Turner, M. G. Hubbard, and A. E. Dukler,
“Analysis and Prediction of Minimum Flow Rate
Identification of a hole in the tubing is often time
for the Continuous Removal of Liquids from Gas
consuming when the hole is below the gas/liquid inter-
Wells,” Journal of Petroleum Technology, 1969.
face in the annulus or is so small that its echo is buried
in the background noise. In these cases, the well has to 4. S. B. Coleman, et al., “A New Look at Predicting
be shut in for extended periods of time. Shutting in the Gas Well Liquid Load-Up,” Journal of Petroleum
well allows the casing and tubing pressure to increase Technology,1991.
and displace the liquid out of the tubing and annulus 5. J. N. McCoy, “Acoustic Velocity of Natural Gas,”
until the hole is uncovered and the noise level in the Echometer Company, 1996.
well is reduced. Figure 9.25 shows the progression of 6. W. E. Gilbert, “Flowing and Gas Lift Well Perfor-
acoustic records used to detect a small hole in the tubing mance,” Drilling and Production Practice, 1954.
at a depth of 4,325 feet from the surface.
7. G. W. Govier and K. Aziz, The Flow of Complex
Mixtures in Pipes (Van Nostrand Reinhold Co.,
SUMMARY 1972).
Acoustic fluid level surveys offer powerful tools for 8. Y. Taitel and A. E. Dukler, “A Model for Predicting
analyzing the performance of gas wells by determining Flow Regime Transitions in Horizontal and Near
the distribution of fluids and pressures in the tubing Horizontal Gas-Liquid Flow,” AIChE Journal,
and/or casing without having to introduce tools or sen- January 1976
sors in the well. The survey determines the presence 9. C. P. Walker, “Method of Determining Fluid Density,
and extent of liquid loading that may be the cause of Fluid Pressure and the Producing Capacity of Oil
reduced productivity. The advantages of acoustic tests Wells,” U.S. Patent No. 2,161,733, June 1939

Sec 0 2 4 6 8 10
Hole Not Visible

100.0 mV
Time 12:12:27 Csg 125.9 psi

Shot Casing/Flowing Up Tubing
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000
Sec 0 2 4 6 8 10 12
100.0 mV
Time 12:24:15 Csg 135.5 psi

Shot Casing/Tubing Flow Shut-In
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Sec 0 2 4 6 8 10 12
Time 12:29:14 Csg 138.8 psi

31.6 mV
Shot Casing/Tubing Flow Shut-In
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Sec 0 2 4 6 8 10 12
Time 12:35:58 Csg 143.1 psi

100.0 mV
Shot Tubing/Flow Shut-In
Explosion
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000
Sec 0 2 4 6 8 10 12
Time 13:09:38 Csg 150.8 psi

100.0 mV
Shot Tubing/Flow Shut-In
Explosion
5
(ft) 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 /16 in Hole at 4,325 ft
Figure 9.25 Sequence of acoustic records in a gas well with a hole in the tubing

10. J. N. McCoy, A. L. Podio, and K. L. Huddleston, 12. J. N. McCoy, O. L. Rowlan, F. Collier, R. S. Lestz,
“Acoustic Determination of Producing Bottomhole and A.L. Podio, “Applications of Acoustic Liquid
Pressure,” SPE Formation Evaluation, 1988. Level Measurements in Gas Wells,” Proceedings of
11. J. N. McCoy, O. L. Rowlan, D. Becker, and A. L. the 48th Southwest Petroleum Short Course, 2008.
Podio, “Acoustic Liquid Level Measurements in Gas 13. J. N. McCoy, O. L. Rowlan, and A.L. Podio, “Acoustic
Wells,” Proceedings of the SPE Annual Technical Liquid Level Testing of Gas Wells,” Proceedings of the
Conference and Exhibition, 2006. SPE Production and Operations Symposium, 2009.

Fluid Level Measurement Applications for Gas Lift Wells 10-1
10
Fluid Level Measurement Applications for Gas Lift Wells
In this chapter:
• Unloading status and operating valve identification
• Determining static and producing BHP
• Pressure distribution at steady flowing conditions
• Pressure distribution at shut-in conditions
• Recommended equipment and procedures
• Example acoustic records
Optimizing the design and operation of wells produced by the upper valves (known as the unloading valves) are
continuous or intermittent gas lift, requires determining closed. Details about gas lift systems and operations are
the SBHP, the PBHP, the well inflow performance, and discussed briefly at the end of this chapter.
quantifying the overall gas lift system efficiency. The
normal gas lift well is assumed to be a continuous flow
well in which a packer is placed immediately above the
formation at the bottom of the tubing. The inside of the
GAS COMPRESSOR OIL AND GAS
tubing is open from the bottom to the top of the well, PRODUCED
but in some cases, it may have a standing valve. This GAS IN
prevents backflow from the tubing to the lower part of
the wellbore when gas injection is stopped.
The packer is used to stabilize the fluid level in the
casing annulus and prevent injection gas from blowing TUBING VALVE
around the lower end of the tubing in wells with a low
flowing BHP. The packer is particularly important for
gas lift when the injection gas line pressure varies or the
injection gas supply is interrupted periodically. When
the installation does not include a packer, the liquid that
accumulates in the annulus must be displaced after each
shutdown. Any changes in the injection gas line pres-
sure causes the working fluid level to oscillate unless
a packer is set. This causes additional flow of liquid
through the lower valves and possibly more wear of the
RESERVOIR
valve seat and stem.
Figure 10.1 illustrates a typical continuous injection
gas lift well showing that gas is being injected from the
bottommost valve (known as the operating valve) while Figure 10.1 Typical continuous injection gas lift well
For distribution by Petroleum Extension-The University of Texas10-1

at Austin
BENEFITS OF ACOUSTIC MEASUREMENTS and through a fully opening valve of adequate diam-
IN GAS LIFT WELLS eter. Acquisition of acoustic records through a needle
valve causes excessive signal attenuation and should
Throughout the life of a gas lift well, the distance to the
be avoided. Figure 10.2 shows the preferred modes of
annular fluid level provides important information for
installation:
analyzing the operating performance of the well and
checking the adequacy of the design of the gas lift system. • Directly to the swab valve for tubing measure-
During the valve unloading process, the casing and ments at elevated pressures
tubing pressures are monitored and recorded while the • Connection to the casing valve opposite the gas
fluid level is monitored. This ensures that the process is injection line
following the desired liquid displacement rate and that,
In wells operating with injection pressures less than
as each deeper valve is uncovered, the upper valves close
1,500 psi, the full two-inch diameter connection should
as programmed. After the gas lift system has stabilized,
be used. At higher operating pressures, it is common to
it is necessary to monitor the annular liquid level so that
have H-inch NPT connections, but the signal amplitude
it remains constant at or below the operating valve. Sig-
reduction caused by the smaller diameter is compensated
nificant oscillations of the liquid level are an indication
for by the lower attenuation of the acoustic signal at
of possible valve operation problems or of gas injection
elevated pressures. If thread adapters or reducers are
rate variations. In a well with a packer, the deepest liq-
needed, they should possess the largest pipe size that fits
uid level is an indication of which mandrel (uncovered
the wellhead connection. The smallest valve between
when the unloading process was completed) is deepest.
the gas gun and the wellbore should be fully opening
Wellbore integrity problems, such as holes in the
with an internal diameter (ID) of at least K inches.
tubing or casing, can be detected by periodically obtain-
Figure 10.3 shows a visual comparison of the quality
ing acoustic records and comparing them with previous
of the acoustic record obtained with different gas guns
records acquired at the time of installation of the tubulars.
and connection methods. Note the increased amplitude
Periodic acquisitions of acoustic fluid levels are useful
of the background noise in the record that was acquired
in identifying downhole problems, such as stuck-open
through the needle valve. The small size of the valve
valves, a leaking packer, leaking check valves, and other
seat seems to behave as a high-pass filter, enhancing the
problems. During workovers with an open wellhead, the
magnitude of the background noise that originates from
stability of the kill fluid depth is determined through
the flow of gas through the injection line.
repeated fluid level monitoring. These and other ap-
plications are discussed in more detail in the following
Background Noise
sections of this chapter.
In a gas lift well, the majority of the background noise
originates at the injection line by gas flowing through
EQUIPMENT INSTALLATION AND DATA the control valve or choke. Since the frequency of the
ACQUISITION noise is relatively high in comparison to the frequency
Complete analysis of gas lift system operation gener- content of the echoes generated at the gas lift mandrels
ally requires performing fluid level measurements, or the liquid level, it is possible to improve the signal-to-
both down the annulus and down the tubing. Because noise ratio using properly designed digital filters. Figure
the level of pressure at the wellhead is usually several 10.4 shows the raw acoustic signal in the upper trace
hundred psi, the majority of the acoustic records are (A) acquired while gas injection was occurring. After
acquired by implosion of the wellbore gas into the gun’s processing with a low-pass filter, the signal is displayed
pressure chamber. in the lower trace where the transmitted pulse at time
As recommended in chapter 4, the connection of zero and the echo from the liquid level at 7.34 seconds
the acoustic pulse generator to the well should be short can be clearly seen. The wellbore diagram showing the

Figure 10.2 Acoustic fluid level instrument connection to the casing or tubing of a gas lift well.
location of the gas lift valves is overlaid on the acoustic displayed immediately below in figure 10.4, it is clearly
trace. Despite the signal improvement by filtering, it is seen that the echoes from the safety valve and the two
difficult to correlate with confidence the position of the upper gas lift valves correlate with their respective posi-
valves with their corresponding echoes. tions in the wellbore.
In Figure 10.4, trace (B) shows the acoustic record Other sources of background noise in gas lift
acquired in the same well after closing in the gas injection installations include compressors, reciprocating pumps,
valve. When the injection gas flow is stopped, the noise and wave-induced vibrations in offshore platforms. Some
level is reduced significantly to the point that, even in acoustic microphones are designed to be insensitive to
the raw signal, it is possible identify the shot and echo vibrations by using multiple receivers while others may
from the liquid level. Furthermore, in the filtered trace be more prone to picking up interfering vibrations.

Comparison
Sec 0 1 2 3 4 5
31.6 mV
5,000 psi Gun, No Valve
Implosion
(ft) 0 1,000 2,000 3,000
Sec 0 1 2 3 4 5 6
31.6 mV
Compact Gas Gun
Implosion
(ft) 0 1,000 2,000 3,000
Sec 0 1 2 3 4 5 6 7
31.6 mV
Through Needle Valve
Implosion
(ft) 0 1,000 2,000 3,000 4,000
Figure 10.3 Comparison of acoustic records for different connection methods and gas guns
Random Noise flagged liquid level echo. This is especially a problem

in offshore installations where multiple wellheads are
Noise caused by random events—such as inadvertently located in a restricted space, for example, on a single
hitting the acoustic sensor or the microphone cable deck of the platform. Distinguishing the real echoes from
or signals induced by operations underway in nearby wellbore discontinuities is the main reason for acquiring
wellbores—is one of the leading causes of a wrongly multiple records that can be compared by superposition.

A Implosion Shot Down Casing—600 psi

Gas-Lift Injection ON
Sec -1 0 1 2 3 4 5 6 7 8 9
40
20
-20
-40
ft 0 1,000 2,000 3,000 4,000 5,000
Reflections from mandrels can be difficult to identify even with wellbore overlay.
Sec -1 0 1 2 3 4 5 6 7 8 9
-2
-4
ft 0 1,000 2,000 3,000 4,000 5,000
B Implosion Shot Down Casing—600 psi

Gas-Lift Injection OFF
Sec -1 0 1 2 3 4 5 6 7 8 9
15
10
-5
-10
-15
ft 0 1,000 2,000 3,000 4,000 5,000
Reflections from mandrels are clearly identifiable.

Sec -1 0 1 2 3 4 5 6 7 8 9
-2
-4
ft 0 1,000 2,000 3,000 4,000 5,000
Figure 10.4 Acoustic record acquired while injection gas is flowing (top) and acoustic record acquired after stopping
the flow of injection gas (bottom)

In figure 10.5, the echoes from the three mandrels are ANALYZING GAS LIFT WELL FLUID
visible in both records, which were acquired within 8 LEVEL RECORDS
minutes of each other. The echoes from the mandrels
The analysis tools and procedures described in chapter
align correctly with the mandrels shown on the wellbore
5 are directly applicable to the interpretation of acoustic
diagram. The black trace shows an additional signal at
records acquired in gas lift wells, as discussed in the
about 10.4 seconds that could be interpreted as a liquid
next section.
level echo (large-amplitude down-kick). However, this
signal is not present in the repeat shot, so it must have
originated from some random event that occurred during Determining Acoustic Velocity
acquisition of the record. Since most gas lift installations operate with injected gas
If the black trace alone were available for analysis, that is compressed after processing the stream in a liquid
that signal could be interpreted incorrectly as being the recovery facility, the composition of the injected gas
liquid level. Since it is below the bottommost gas lift includes primarily low gravity hydrocarbons (methane
valve, it would be an indication that the packer located and ethane). Thus, the acoustic velocity in the annulus
near the end of the tubing is leaking and the annular fluid can be characterized by the graphs shown in figure
has been pushed out. This interpretation might trigger 10.6. The acoustic velocity for injection gas pressures
an unnecessary workover operation. The complexity of from 600 to 1,600 psi ranges from 1,250 to 1,500 ft/s,
analyzing gas lift acoustic records merits the acquisition depending primarily on the temperature.
of multiple records to increase the confidence in the Even though the composition of the injection gas
quality of the data and its interpretation. is fairly homogeneous, the variation of temperature in
Sec -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13
-5
mV
-10
-15
-20
Figure 10.5 Repeat acoustic record (blue) that does not show a random signal at about 10.4 seconds. (Note the LL
marker located arbitrarily at 11.0 seconds.)

Acoustic Velocity Profiles in Gas-Lift Wells
1,500
LIQUID LEVEL:
10,200.8 ft
1,400 250 DEGREE F
956.6 psig
1,300 SURFACE:
0.0 ft
70 DEGREE F
701 psig
1,200
TEMP 40 70 100
DEGREES F 130 160 190
220 250 280
310 WELLBORE
1,100
0 500 1,000 1,500 2,000 2,500 3,000
PRESSURE, psia
Figure 10.6 Chart depicting acoustic velocity as a function of pressure and temperature for 0.6 gravity hydrocarbon
gas
the wellbore—cool at the surface and hot at the bot- The consequence of this variation is that the rela-
tom—results in a significant variation of the acoustic tion between acoustic pulse round trip travel time and
velocity in the wellbore. The dashed line in figure 10.6 distance is not a linear function of time but varies as
shows what the variation of velocity would be in a well a function of depth. As discussed in chapter 5, most
where the liquid level is at 10,200 feet of depth. The of the existing acoustic fluid level analysis software
velocity is 1,270 ft/s at the surface where the temperature use a constant average value of the acoustic velocity
is 70°F and increases to 1,520 ft/s at the bottom of the to establish the relation between round trip travel time
gas column where the temperature is 250°F. and distance. This assumption results in a mismatch

between the calculated positions of the echoes from the level or a hole in the tubing or casing, it is possible to
gas lift mandrels and their actual position, as shown in use the known depth to the mandrel nearest to the echo
figure 10.7, which includes the wellbore overlay. The in question and the round trip travel time to the known
bottom horizontal scale (in feet) is based on the average mandrel to compute the acoustic velocity. Then the
acoustic velocity—computed from the round trip travel average acoustic velocity from the surface to the depth
time of the echo from the top of the perforations (at of the known mandrel is used to calculate the unknown
about 10,200 feet)—and thus is a fairly representative distance to the echo from the unidentified reflector with
value for the gas in this wellbore. Despite this finding, a minimum of extrapolation error. This procedure com-
the difference in depth between the valve positions and pensates in part for the variations of acoustic velocity
their corresponding echoes becomes progressively more commonly observed in most wellbores due to changes
significant towards the bottom of the well where the in temperature, pressure, and gas composition as a
temperature is highest. function of depth.
The presence of the gas lift mandrels located at The polarity of the reflection from side-pocket man-
accurately known depths provides an excellent tool for drels depends whether the acoustic record is acquired
refining the calculation of the distance to the liquid level through the annulus or through the tubing, as shown in
and any other anomaly that can cause the generation of figure 10.8. These recordings were acquired in a gas lift
an echo. When the distance to the perforations is used well that had been shut-in long enough for the liquid
to determine the average velocity of gas in the wellbore, levels in the casing and the tubing to stabilize and equalize
anomaly depths immediately above the perforations will because of liquid U-tubing through the screened orifice
be calculated more accurately than the anomaly depths located just above the packer.
near the midpoint of the tubing string. In this example, the round trip travel time of the echo
from the third gas lift mandrel was used to compute the
Using Mandrels as Markers average acoustic velocity and determine the depth scale
To accurately calculate the distance to an echo generated for both records. When acquiring the acoustic record down
at a variable depth downhole reflector, such as the liquid the tubing, the polarity of the echo from each mandrel
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13
15
10
5
mV
-5
-10
-15
0 2,000 4,000 6,000 8,000
Figure 10.7 Acoustic record acquired down the tubing of a shut-in well showing a mismatch of echoes and gas lift
mandrels due to varying acoustic velocity with increasing depth

Mandrel Design Determines

Direction of Kick
Sec 7 8 9 10
15 SHOT DOWN TUBING DISPLAYS UPKICK AT MANDREL
10
5
mV
-5
-10
INCREASE IN AREA = UPKICK
-15
ft 0 1,000 2,000 3,000 4,000 5,000 6,000
Sec 7 8 9 10
SHOT DOWN CASING DISPLAYS DOWNKICK AT MANDREL
4
2
mV
-2
-4
DECREASE IN AREA = DOWNKICK
ft 0 2,000 4,000 6,000
Figure 10.8 Mandrel design determines the polarity of the echoes.
shows an upward deflection corresponding to the increase signals are additive, so a single echo is observed on the
of the internal cross-sectional area. On the other hand, record but of greater amplitude than expected.
since the area of the annulus at the depth of the mandrel On the other hand, in numerous gas lift installations,
is reduced because of the increase in outer diameter of the spacing of the valves is not uniform. This generates
the mandrels, the polarity of the echo is a downward multiple additional echoes spaced between the primary
deflection for a record taken in the casing-tubing annulus. echoes from each valve. The multiple echoes complicate
Notice that, in the example records in figure 10.8, the analysis of the record as far as identifying which are
the gas lift valves installed below the top valve are ap- the primary echoes from the deeper valves. This fact is
proximately spaced at the same distance from each other illustrated in figure 10.9 where the acoustic record was
(about 950 feet). As the sound pulse is reflected back acquired in the annulus of a deep well that has a total of
from the second valve, an internal echo is generated when seven gas lift valves and a packer just below the bottom-
passing the top valve. This echo travels back down to the most valve. Note the complexity of the acoustic record
second valve where it is again reflected to the surface where at least 11 echoes can be identified.
and then received at the microphone at about the same The first two echoes correspond to the top two valves
time as the first echo from the third valve. These two followed by the repeat echo of the top valve. Then there

Sec 0 1 2 3 4 5 6 7 8 9 10
150
Repeat GL 1 GL 3 ?? Repeat GL 2
100
Repeat GL 1
50
mV
-50 GL 4 ??
GL 1
-100
GL 2 Mirroring GL2 by GL1
-150
0 1,000 2,000 3,000 4,000 5,000 6,000
Figure 10.9 Complex acoustic record in a deep gas lift well
is an echo that might be from the third valve and then reflections of the echoes will exhibit alternate polarity.
the internal reflection of the second gas lift valve at the This effect is illustrated in figure 10.10 where the record
first gas lift valve. The analysis is further complicated was acquired down the tubing in a static well with the
by the presence of an unidentified echo between valve 3 liquid level about 700 feet below the top valve.
and 4, which raises the question of whether the informa- A total of nine echoes are visible in the record,
tion about the layout of the gas lift system is wrong and but only one corresponds to the primary echo of the
the depth of some of the valves are different in reality liquid level. The first is an up-kick that corresponds to
from what is listed in the wellbore diagram. Since all the top gas lift valve. The second is a down-kick from
the unidentified echoes exhibit downwards deflections, the liquid level. The rest of the echoes are repeats and
they correspond to annular restrictions. Thus, they cannot internal reflections created between the liquid level
be attributed to the presence of holes in the tubing or and the top valve. A detailed analysis can be performed
casing since these would exhibit upwards deflections. by creating the wave characteristic plot, as described
This example illustrates the importance of having ac- in chapter 3, showing the position and polarity of the
curate well information with the correct depth of all pulse as a function of time and distance. Dashed lines
the mandrels and other downhole markers in order to indicate up-kicks and solid lines indicate down-kicks,
analyze the record correctly. with inversion of the echoes occurring each and every
When the record is taken down the tubing in an time the wave reaches the side-pocket mandrel.
installation with side pocket mandrels, the analysis is The two reflectors are the gas lift mandrel and the
even more complex. The polarity of the echo is inverted liquid level. Value A (4.334 seconds) is the round trip
whenever the wave arrives at a mandrel, so multiple travel time to the mandrel. Value B (1.068 seconds) is

Sec 0 1 2 3 4 5 6 7 8 9 10 11 12
4 7
3 5+6
2 2 4
mV
-2
1
-4
ft 0 2,000 4,000 6,000 8,000
2 4 5 6
3 8 (1) 4.334 up
7 (2) 5.402 LL
1
(3) 6.47 up
(4) 7.538 up
A 4.334
(5) 8.606 up
(6) 8.668 RM down

2,842 ft (7) 9.736 (7) 9.674
B 1.068 (8) 10.804 LL LL

LL: 3,511 ft
B 1.068 1B 2B 3B 4B 5B 6B 7B
A 4.334 5.402 6.47 7.538 8.606 9.674 10.742 11.81
2A 8.668 9.736 10.804 11.872 12.94 14.008 15.076
3A 13.002 14.07 15.138 16.206
Figure 10.10 Acoustic data showing a problem identifying multiple echoes because of internal reflections down the
tubing

the round trip travel time between the liquid level and to show an interval velocity point outside the normal
the mandrel. trend (such as the fictitious point highlighted in yellow),
Quality control of the automatically computed it would be a strong indication that the calculation is in
acoustic velocity can be undertaken by using the known error: either the time or the distance between reference
distance between successive gas lift valves to compute points is wrong.
the average acoustic velocity of the gas in that specific The normal increasing velocity trend also verifies
section of the wellbore. This velocity is defined as the that the gas gravity is uniform, and the value computed
interval velocity to differentiate it from the average from the acoustic velocity can be used with confidence
velocity of the gas between the surface and the depth in pressure calculations. The continuous nature of the
of a specific valve. velocity function allows it to be curve-fitted with respect
Accurate fluid level depth calculations in deep to the round trip travel time with a polynomial relation
wells must account for acoustic velocity variation in the that has a relatively high correlation coefficient, as shown
deepening wellbore. In gas lift wells, the injection gas in figure 10.12.
composition is uniform and corresponds to low gravity Once this correlation function is established, it can be
hydrocarbons. Therefore, the variation of acoustic veloc- used to directly compute a more accurate distance to any
ity due to increase in temperature is likely to be repre- specific echo, knowing its round trip travel time without
sented by a continuous function of depth, as displayed using the bulk average acoustic velocity that is applied
in figure 10.11. The continuous trend that is exhibited when generating the distance scale of the acoustic graph.
by the two series of points validates the consistency of For example, the echo highlighted in figure 10.13 at 7.447
the analysis. If the analysis of an acoustic record were seconds, which is caused by a gas lift valve located at a
Acoustic Velocity vs. Depth for Gas Lift Well
INTERVAL VELOCITY
1,000
AVERAGE FROM
SURFACE
2,000
DEPTH, ft
3,000
4,000
5,000
6,000
1,250 1,260 1,270 1,280 1,290 1,300
Figure 10.11 Depth profile of the acoustic velocity of injection gas from acoustic records in figure 10.8

Velocity vs. RTTT
1,300
1,295 y = 0.6498x2 - 0.8452x + 1257.4

R2 = 0.9905
1,290
SERIES 1
1,285 POLY. (Series 1)
1,280
1,275
1,270
1,265
1,260
1,255
0 2 4 6 8 10
RTTT, sec
Figure 10.12 Correlation function of acoustic velocity versus the RTTT
Sec 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
40
20
mV
-20
-40
ft 3,000 4,000 5,000 6,000
Figure 10.13 Acoustic record analyzed automatically to generate a depth scale

depth of 4,854 feet RKB, is indicated by the program to Open Valves

be at a depth of 4,931 feet, based on the distance scale The reverse check in a gas lift valve is especially impor-
automatically determined by the software using the tant if any valves are located below the working fluid
tubing joints count and the tubing average joint length. level. The check prevents backflow from the tubing into
These values yield a constant average acoustic velocity the casing, which is particularly important if the well
of 1,319 ft/s that is used to generate the depth scale of produces sand and has a packer. The acoustic record
the graph. The estimated valve depth corresponds to a examples presented in chapter 2 include several acoustic
difference of 77 feet in comparison to the actual depth. records in gas lift wells where some valves were stuck
On the other hand, applying the acoustic velocity open and the check valves were faulty.
correlation function determined in figure 10.12 with the
measured round trip travel time to the valve echo, the OPERATING PRESSURE DISTRIBUTION FOR
computed acoustic velocity from the surface to the third
VARIOUS GAS LIFT CONFIGURATIONS
valve is calculated as 1,293 ft/s. This yields an estimate
of the distance to the valve of 4,815 feet for a total depth The pressure distribution in the annulus or tubing can
of 4,836 feet, considering the KB correction of 21 feet. be estimated from the acoustic survey, as discussed in
The result yields a smaller error of 18 feet. detail in chapters 6 and 7.
Whenever echoes from multiple markers are visible Static Bottomhole Pressure
in the acoustic record, the marker that is closest to the The general concepts of SBHP calculation based on fluid
echo (for which a distance is being sought) should be level surveys discussed in the aforementioned chapters
used as a reference. are directly applicable to gas lift wells.
In a normal gas lift well, the SBHP can be accurately
TROUBLESHOOTING GAS LIFT SYSTEMS estimated using an acoustic liquid level instrument by
If an installation is properly designed, all gas lift valves closing in the tubing and gas injection line at the surface
above an operating valve should be closed, and all valves and allowing the well pressures to stabilize. Determining
below should be open. Inefficient multipoint gas injec- the liquid level in the tubing using the acoustic liquid level
tion can result, preventing unloading an installation to instrument and measuring the stable wellhead pressure
the maximum depth of lift for the available operating allows calculation of the BHP by adding the pressures
injection gas pressure when upper valves are left open. exerted by the columns of fluids that exist above the
Closing in the injection and the production lines, formation depth to the measured surface tubing pressure.
thus allowing the pressures and fluid levels to stabilize, The liquid above the formation consists of water
is the recommended procedure to investigate whether or and oil at a ratio dependent on the completion geometry
not the valves are stuck open or if there are holes in the and the water cut that is produced by the well. In a well
tubing where gas is leaking. These faults can be detected that has tubing extending to the formation, the average
by comparing acoustic records acquired down the tubing gradient of the liquid in the column is primarily based on
and down the casing, as shown in figure 10.14, where the produced water/oil ratio (WOR). When a significant
a repeatable but unexpected echo at about 1.6 seconds distance exists between the end of the tubing or depth
could indicate the presence of an anomaly in the upper of the packer and the perforations, the liquid gradient
part of the tubing, most likely a corrosion/erosion hole. below the end of the tubing is primarily controlled by
Frequent periodic monitoring of annular fluid level the density of the water because of the liquid hold-up
and pressure allows developing a historical database of effect. A detailed volumetric balance calculation must
variations as a function of casing pressure and injection be performed to distribute the fluids correctly in relation
gas rate. This can be used to check the efficiency of the to changes in the cross-sectional area when the wellbore
installation and detect changes in performance that may consists of several sections because of the presence of
be related to reservoir depletion. liners or tapered tubulars.

Compare Shots Down Tubing and Casing
Sec 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5
LL
40 Tubing
20
mV
-20
HOLE
-40
Sec 0 1 2 3 4 5 6 7 8 9 10 11 12 13
LL
40 Casing
20 GLV9
mV
-20
-40
ft 0 2,000 4,000 6,000 8,000
Figure 10.14 Comparison of records acquired in the tubing and casing, showing an echo from the hole in the tubing
When no packer is used in the installation, the static Figure 10.15 illustrates some of the most common
fluid level through the tubing and through the annulus configurations and operating characteristics of gas lift
should be acquired. This acquisition helps verify that wells.
the bottom of the tubing is open and that there is free • Case A: The well is fully unloaded, and gas is
communication between the two branches. Differences being injected at the bottom valve.
in the two fluid levels correspond to different water cuts • Case B: The well is in the process of unloading,
in the accumulated liquids when the surface tubing and or the injection gas volume rate is insufficient
casing pressures are equal. to continue unloading the well and reaching the
A representative value of the SBHP is required for bottom valve or screened orifice. Gas is being
selecting the depth to the top unloading valve. The inflow injected at the second valve. A screened orifice
performance characteristics of the formation computed is located at the bottom of the annulus.
from the SBHP and PBHP are useful in selecting the • Case C: The packer is set at a significant distance
depths for additional valves, determining gas injection above the perforations. The well is in the process
rates, and optimizing the gas injection system. of unloading, or injection gas is insufficient to

CASING
TUBING
GAS LIFT
VALVE
LIQUID
SCREENED
ORIFICE
PACKER
A B C D
Figure 10.15 Gas lift well configurations
continue unloading. Whenever possible, the tubing that the gas lift installation was properly designed and
intake should extend to the top of the perforated is operating efficiently. The annular fluid level survey is
interval to achieve the most efficient performance. used to estimate the pressure that exists in the annulus at
• Case D: The well is fully unloaded with the bot- the depth of the valves. In turn, this pressure distribution
tom valve set high above the packer. is compared to the design tubing pressure traverse that
was used to define the depth of the operating valve and
For each of these configurations, it is possible to
calculate the injection gas/liquid ratio.
compute the pressure distribution in the annulus and
The relation between casing and tubing pressure
to estimate the flowing pressure at some points in the
tubing and the PBHP. at the depth of the valves depends on the completion
configuration and the position of the annular liquid
Producing Bottomhole Pressure in relation to the lowest valve or orifice. As shown in
At any time during the operation of a gas lift well, it figure 10.15, cases B and C illustrate gas lift systems
is important to understand the dynamic pressure distri- which have annular liquid above the lowest valve. In
bution in the casing and in the tubing. This will verify general, the annular liquid consists mainly of the kill

fluid—that is, the liquid that was used to control the well increase has to be estimated using a multi-phase flow
pressure during the installation of the tubing string. This gradient chart or software.
is generally a brine treated with a corrosion inhibitor, or
Case D
it could be a mixture of produced brine and oil. In all
cases, the gradient (density) of the packer fluid is fairly The PBHP is the casing pressure plus the gas column
well known and is considered to be a constant. pressure, minus the pressure drop through the valve,
After the system is put in operation and the produc- plus the fluid column pressure between the tubing intake
tion and injection rates and the fluid level are permitted and the injection valve, plus the fluid column pressure
to stabilize, the pressure in the casing annulus opposite between the formation and the tubing intake. Multiphase
any of the valves submerged in liquid is equal to the flow gradient curves or software are required to compute
pressure in the tubing at the same depth because the these additional pressures.
valves are open, and there is no outflow of liquid across The uncertainty of the value of the computed PBHP
the valve orifices. is minimized when the gas lift completion includes a
The casing annulus pressure opposite the lowest valve or an orifice immediately above the top of the
uncovered valve in a well which has been operated packer and the packer is set at a depth just above the
for a long period of time indicates the lowest pressure producing formation that is for cases A and B.
which has existed in the tubing since last equalization
of tubing and casing annulus pressures. This assumes EXAMPLE FLUID LEVEL AND
no leaks at the check valves and packer. PRESSURE SURVEY
The liquid level in the casing annulus normally The usefulness of fluid level surveys for monitoring gas
is lower than the point of gas injection because of the lift installations was recognized early in the history of
difference in the gradient of the gas-free packer fluid the technology1 and has been greatly enhanced with the
in the casing annulus above the bottom valve and the development of digital and wireless hardware.
gradient of the flowing gaseous fluid mixture which One of the important applications is verifying the
exists in the tubing between the gas injection valve and gas lift design calculations by comparing the measured
the bottom valve. annular pressure distribution with the estimated tubing
Case A pressure gradient curves. The procedure involves per-
forming a survey down the casing at stabilized condi-
The BHP in the tubing is the pressure in the casing
tions and acquiring repeated fluid level measurements
annulus at the bottom valve minus the pressure drop
down the tubing after shutting in the flow at the surface.
across the valve. The pressure drop across the valve
The dynamic fluid level survey in the annulus in
can be estimated by the gas lift valve manufacturer. The
figure 10.16 shows the liquid level at about 17 feet
pressure in the casing annulus at the bottom valve is the
above the bottommost gas lift valve located at 5,619
surface casing pressure plus the gas column pressure.
feet. Additional fluid level records were taken in the
Case B annulus over a period of two hours, and they all showed
The PBHP in the tubing is the casing pressure plus the consistently that the liquid level stayed above the bot-
gas column pressure, plus the liquid column pressure tom valve. This is a strong indication that probably the
above the bottom valve or screened orifice. third valve was open and acting as the operating valve.
The corresponding casing pressure traverse is plotted
Case C in figure 10.17 as the blue line.
The PBHP is the casing pressure plus the gas column After acquiring the liquid level survey in the annulus,
pressure, plus the liquid column pressure above the the instrument was installed on the tubing and connected
bottom valve, plus the fluid column pressure between to the swab valve. The flow was stopped, closing the
the bottom valve and the formation. This last pressure wing valve to the flow line, and a fluid level record was

0 1 2 3 4 5 6 7 8 9 10
Depth
2,801.57
3,873.26
4,854.95
5,619.06
Save & Close Cancel
Figure 10.16 Casing survey: Liquid level at 5,602 feet above fourth gas lift valve
Pressure vs. Depth After Shut-In
PRESSURE, psi
0 50 100 150 200 250 300 350 400 450 500
0
*
TUBING FLOWING
1,000 GRADIENT CURVE
2⅜", 2600 GLR, 275 BPD
2,000
CASING
TUBING 9:53
3,000
DEPTH, ft
TUBING 9:56
TUBING 10:44
4,000
* TUBING 10:15
5,000
*
6,000 *
7,000
Figure 10.17 Pressure traverses for the casing and tubing during shut-in. Flowing gradient (dashed line) from
McMurry gradient charts.

acquired after a few minutes. This first acoustic record design pressure drop of 75 psi. This indicates that the
shown in figure 10.18 is labeled at 9:53 and shows the well appears to be operating in accordance with the
liquid level echo at a depth of 2,523 feet. Subsequent design parameters.
records show the liquid level in the tubing dropping This example illustrates the use of a complete fluid
below the gas lift mandrels and at 10:05 is recorded at level analysis to validate the gas lift design and verify
a depth of 4,543 feet. The last record at 10:44 shows the that the artificial lift system is operating efficiently.
liquid level just past the bottommost valve at a depth
of 5,619 feet.
The tubing pressure and liquid level data are plot- BACKGROUND INFORMATION ABOUT GAS
ted in figure 10.17, showing the four pressure traverses LIFT INSTALLATIONS
where the bottom point represents the pressure at the The primary purpose of the gas lift system is to lift large
tubing intake just below the packer. It is interesting to volumes of fluid in an efficient manner and maintain
note that in less than 50 minutes the flowing column the well production rate at the desired level when the
of gas and liquid collapsed, leaving the tubing com- reservoir pressure declines because of depletion or when
pletely filled with gas. The gradient of the gaseous the produced water/oil ratio increases. Increased water
liquid column appears to remain constant and is fairly production causes a decrease in the gas/liquid ratio of
similar to the multi-phase flow gradient given by the the fluid in the tubing, with a corresponding increase of
gas lift design charts for this tubing size, liquid rate, the flowing gradient and increase of the tubing intake
and gas/liquid ratio. At the depth of the injection valve, pressure required to lift the fluid to the surface. In these
the difference in pressure between the casing injection cases, flow rate maintenance or increase is accomplished
gas and the flowing tubing pressure agrees with the by injecting gas into the tubing to reduce the gradient
100
50
10:44
10:05
mV
-50
-100
-150
9:53
-200
ft 0 2,000 4,000 6,000
Figure 10.18 Acoustic records acquired down the tubing after shut-in

of the flowing fluid so that the available flowing BHP from the bottom to the surface at the existing tubing
is sufficient to lift the fluids to the tubing head and feed head pressure. Such a pressure distribution is illustrated
them to the surface facilities at the design flow rate. in figure 10.19. The pressure traverse below the point
At steady state flowing conditions, the pressure dis- of gas injection (blue line) includes liquid and only
tribution in the well is such that equilibrium is reached formation gas, whereas the traverse above the point of
between the existing PBHP and the tubing intake pressure gas injection (orange line) includes both the formation
required to flow the desired rate of oil, gas, and water and injection gases.
PRESSURE, psi
Ptf = 100 psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500
0
50% OIL–50% WATER

TUBING SIZE: 2.5 in. I.D.
1,000 PRODUCTION RATE: 500 bbl/day
AVERAGE FLOWING TEMP: 140°F
GLR: 0–1,000 (IN STEPS OF 100)
2,000
CASING GAS
GRADIENT
3,000
4,000
DEPTH, ft
PRODUCED
5,000 FLUID
GRADIENT
6,000
7,000
PI = 0.4
Pwf = 2,500 Q = 500 b/d
∆p = 2000 psi
DEPTH TO
8,000
INJECTION
VALVE
Psf = 4,500
9,000
400 GLR
10,000
Figure 10.19 Pressure distribution in the tubing and annulus of a gas lift well

In this example, the formation is producing at a point at the surface until the line intersects the annular
steady state liquid rate of 500 bbl/day, which—given the pressure line (blue) of the produced fluids. This is the
reservoir static pressure (Ps) of 4,500 psi and the formation deepest point where gas can be injected into the tubing
productivity index of 0.4 bbl/day-psi—results in a PBHP from the casing. In practice, a shallower depth is used
of 2,500 psi at the perforations’ depth of 10,000 feet. to allow for the pressure drop caused by gas flowing
Considering the gradient of the produced fluids, through the injection orifice or valve, generally of the
based on a 50% water cut and a gas/liquid ratio of 50 order of 50 to 100 psi.
scf/bbl when the PBHP is 2,500 psi, the produced liquids To obtain the maximum benefit from the injected
could only rise in the well from the depth of the perfo- gas, it must be injected as near the producing interval
rations to a depth of about 4,000 feet (blue line) where as possible. The injection gas pressure must be greater
all the pressure would be spent and liquids could not than the flowing producing pressure at the same depth.
reach the surface wellhead. Liquid would accumulate In addition, when a well is shut-in, the BHP increases
in the wellbore, causing an increase in BHP and reduc- to a value close to the reservoir pressure, which could be
ing the inflow from the formation even to the point of several thousands of psi in a deep reservoir. To restart
killing the well. flow of the well by means of gas lift would require a
Injecting sufficient gas into the tubing at any point casinghead pressure close to the SBHP to initiate gas
below the 4,000-foot depth reduces the density or gradient injection at the bottom of the tubing. This is neither prac-
of the gas/liquid mixture. Thus, at a certain gas injection tical nor economical. For this reason, additional valves
rate, the available pressure of 2,500 psi is sufficient to are distributed along the tubing to unload the liquid in
raise the liquids to the surface and establish flow out of the the well in stages, using the available casing injection
tubing to the flow line and into the processing facilities. pressure by injecting gas from the upper valve first and
The questions to be answered are how much gas to then through increasingly deeper valves. Meanwhile, the
inject and at which depth. The solution of this design tubing flowing gradient is reduced and the production
problem requires estimating the pressure drop in the rate and PBHP both stabilize.
tubing for gas/liquid mixtures of different compositions Gas lift valves placed in or on the tubing are either
and flow rates. This is normally done using published casing-pressure-operated or tubing-fluid-pressure-oper-
multi-phase flow gradient charts or commercially avail- ated, depending on whether their opening and closing
able gas lift design software. pressures are controlled by the pressure distribution in
In this example, using the flow gradient chart for the the casing or in the tubing. The daily production rate
2.5-inch tubing size, it is necessary to inject gas from a from a continuous-flow gas lift installation should be
depth of 6,500 feet at the rate sufficient to increase the controlled by the injection gas volumetric flow rate to
gas/liquid ratio from 50 to 400 scf/bbl above the point the well.
of injection. For the desired 500 bbl/day production, the As discussed in chapter 1, knowing accurate values
injection gas rate would be 350 × 500 = 175,000 scf/day. for the static formation pressure, producing pressure,
The deeper the point of injection, the more efficient and formation productivity or inflow performance is
the process is, since it will require a lower injection gas a necessity when designing or analyzing any artificial
rate to achieve the necessary reduction in flowing gradi- lift installation. This is especially true when designing
ent. The actual depth of the injection point is primarily or analyzing a gas lift installation. Acoustic fluid level
controlled by the available pressure of the injection measurements in combination with records of surface
gas and thus is a function of the capacity of the com- pressures and flow rates are the primary tools used for
pression facility associated with the gas lift system. In visualizing the pressure conditions and fluid distribution
this example, the operating gas injection compressor existing in the gas lift well.
develops 1,000 psi at the casinghead so that a casing Wells drilled from offshore platforms have vary-
pressure versus depth line (red) can be drawn from this ing degrees of deviation that must be accounted for in

predicting vertical multiphase flow. Most gas lift design 4. If a valve is exposed—that is, the liquid level is
programs can take this varying deviation into account. below the valve—then gas injection through it
has occurred at some point and may still be in-
Valve Operation jecting gas into the tubing. Normally, the annular
Gas lift valves in a casing-pressure-actuated system liquid level will not be deeper than the deepest
operate as follows: valve unless the packer is leaking (or there is no
1.
A valve opens when it is exposed to a casing an- packer).
nulus pressure in excess of a pre-set value based Gas lift valves in a tubing-pressure-actuated system
on charging the valve dome with nitrogen at operate as follows:
surface conditions. The valves are arranged in the 1.
The gas lift valves will open if the pressure in
tubing string so that the highest pressure opening the tubing exceeds a preset value. The lowest
valve is at the top, and the opening pressure of opening pressure valve is set near the top of the
each valve decreases with depth. Gas injection tubing, and the opening pressure settings increase
in the casing annulus into the tubing lightens with valve setting depth.
the gradient in the tubing above the valve that
2. When the well is shut in, the liquid level in the
is open and injecting gas. All valves below this
tubing will be high, and most of the tubing-
point are open and permit liquid or gas to pass
pressure operated valves will be open. When
into the tubing if the tubing pressure is less than
gas injection into the casing annulus starts, the
the casing annulus pressure. The liquid level in
annular fluid level is depressed. Annular liquid
the casing annulus will be depressed until the
will U-tube into and up the tubing, causing the
pressure in the casing annulus at the bottom
tubing pressure opposite each valve to increase.
valve is equal to the pressure in the tubing at the
Annular gas is injected into the valve nearest
bottom valve or until the bottom gas lift valve is
the top, which has sufficient fluid pressure in
exposed to gas. Check valves prohibit backflow
the tubing to open the valve. The bottom valves
from the tubing into the casing annulus.
will be open and permit liquid to pass from the
2. A gas lift valve above the injection point will casing annulus into the tubing.
close if the pressure in the casing annulus is 3. Gas lift valves will become exposed as the liquid
reduced below its preset operating value by ad- level drops, and gas injection will begin if the cas-
ditional flow of gas into the tubing through the ing annulus pressure exceeds the tubing pressure
lower valve when it is uncovered. This is known and sufficient tubing pressure exists to open the
as the closing pressure of the gas lift valve and gas lift valve. Upper gas lift valves usually will
is preset prior to running the valve into the well. close since injection at the lower depth decreases
The difference between opening and closing the tubing pressures causing the upper valve to
pressures is known as the valve spread. close. In a normal installation, if liquid exists in
3. The bottom valve should be always open unless the annulus above the bottom valve, the bottom
it is the injection valve and there is insufficient valve will be open, and the pressure in the tubing
gas to maintain the pressure in excess of its at that depth will be equal to the pressure in the
opening pressure; then it will throttle or intermit. casing annulus. If the liquid is below the bottom
In this case, the casing pressure might increase gas lift valve, gas injection has and may still exist
enough for the next higher valve to reopen and at the bottom valve.
inject gas at a shallower depth. When a valve is
throttling, the pressure drop across the valve is Valves and Mandrels
not known, but the pressure in the tubing is less The choice between tubing-conveyed valves and
than the pressure in the casing annulus. wireline-retrievable valves depends primarily on the

costs associated with pulling the tubing and whether mandrel because the valve is offset from the centerline
introducing a workover fluid may damage the deliver- of the tubing, as shown in figure 10.20. Most side-pocket
ability of a well. mandrels have a full-bore sectional area greater than the
In tubing-conveyed types, the mandrel that holds the tubing’s to permit wireline operations, such as pressure
gas lift valve and reverse check valve is part of the tubing surveys. In deviated wells, special devices are used to
string. Replacing or repairing the valves requires pulling correctly orient and locate the valve for insertion into the
the tubing. The wireline-retrievable-valve mandrel is also mandrel. Gas lift retrievable-valve mandrels with orienta-
attached to the tubing but is designed with a pocket receiver tion sleeves are designed to ensure that the valve enters
within the mandrel. A gas lift valve can be removed or the pocket regardless of pocket orientation relative to the
installed by wireline operations without pulling the tub- vertical. Operators report few problems in setting valves
ing using a special kick-over tool. This tool is used for in mandrels at deviations from 60° to 70°.
locating the mandrel pocket and selectively removing or Detailed descriptions of such equipment can be
installing the valve. The mandrel is called a side-pocket found in API Spec. 11V12.
VALVE MOUNTED INSIDE THE

MANDREL (WIRELINE RETRIEVABLE)
LATCH
LATCH RETAINING SHOULDER
PACKING (VALVE POCKET TO SEAL)
PORTS TO ANNULUS
VALVE
PACKING (VALVE TO POCKET SEAL)
SIDEPOCKET (VALVE RECEIVER)
PORT TO TUBING
A B C
Figure 10.20 Typical side-pocket mandrels

VALVE MOUNTED OUTSIDE THE

MANDREL (TUBING MUST BE PULLED
TO HAVE ACCESS TO THE VALVE)
CONVENTIONAL GAS LIFT VALVE
REVERSE-FLOW CHECK
THREAD FOR INSTALLING VALVE

AND CHECK TO MANDREL
A B C
Figure 10.21 Typical conventional mandrels with external valves
General Considerations for Gas Lift Design pressure. Based on estimated flowing gradient curves
A common design is based on a constant decrease in (or software), the minimum gas/liquid ratio that must
the operating injection gas pressure for each succeeding be injected at that depth is determined for the particular
lower valve. This design is known as the API gas lift tubing size and desired liquid rates. A good estimate
design technique in RP 11V63. Other designs may be of the formation productivity or inflow performance is
applied when high pressure injection gas is available. necessary to estimate the PBHP. Figure 10.22 shows
Single-element, unbalanced, gas lift valves with the intersection of the casing pressure (red line) and the
nitrogen-charged bellows are the most widely used in tubing pressure gradient (blue line) to be the injection
the industry and available through all major gas lift depth at 6,500 feet with a PBHP of 2,450 psi.
equipment manufacturers. The API design method is
based on installing the same port size in all valves and Well Unloading
applying a constant decrease in the operating injection The primary function of a string of gas lift valves is to
gas pressure for each lower gas lift valve. The gas lift unload a well with the available injection gas pressure
port size is selected to allow the injection gas throughput to a maximum depth of lift that fully uses the energy of
rate necessary for unloading and gas lifting the well. expansion of the injection gas for the available injection
The first step is to determine the depth of the operating gas pressure. The pressure decrease and valve spread
valve based on the available continuous injection gas should be based on the valve specifications to minimize

PRESSURE, psi
Ptf = 100 psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

1,000 psi TUBING SIZE: 2.5 in. I.D.
1,000 INJECTION PRODUCTION RATE: 500 bbl/day
PRESSURE
2,000
CASING GAS
GRADIENT
3,000
4,000
DEPTH, ft
PRODUCED
5,000 FLUID
GRADIENT
6,000
400 GLR
7,000
Pwf = 2,450
DEPTH TO
INJECTION
8,000 VALVE
Psf = 4,500
9,000
10,000
400 GLR x 500 bbl/d

INJECTION GAS = 200 MCF/d
@ 1,000 psi
Figure 10.22 Determination of the injection depth (6,500 feet) for the desired formation flow rate and operating gas/
liquid ratio. PBHP = 2,450 psi

the probability of upper valves remaining open while depress the liquid and U-tube it via the operating valve
lifting from a lower valve. into the tubing until it flows out at the surface.
Figure 10.23 shows that, to depress the liquid to the
Unloading Sequence for Casing Pressure depth of 6,500 feet and through the valve, the casinghead
Operated Systems pressure would have to be at least 2,800 psi to balance
Starting with a dead well with a static fluid level some- the column of liquid in the tubing. This is not practical,
where below the surface, gas is injected in the casing to so the unloading process is undertaken in steps with a
PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

1,000
PRODUCTION RATE: 500 bbl/day
2,000
3,000
KILL FLUID
4,000 GRADIENT
DEPTH, ft
5,000
6,000
OPERATING
VALVE DEPTH
7,000
8,000
9,000
10,000
Figure 10.23 Pressure in the tubing and casing when the kill fluid reaches the surface by U-tubing through the
operating valve

lower injection casing pressure—say, 1,000 psi—which start to be injected into the tubing. This point at 2,500
is more likely to be readily available. feet indicates the depth where the first gas lift valve
Figure 10.24 shows the intersection of the casing should be located.
pressure (red line) and the tubing pressure gradient (blue To facilitate the unloading process, it is custom-
line) that corresponds to the deepest point where annular ary to initiate the process with the well connected to
liquid could be depressed and gas at this pressure could a tank at atmospheric pressure. Then, as gas flow is
PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
1,000 psi 50% OIL–50% WATER

Ptf = 0 psi INJECTION
PRESSURE
2,000
UNLOADING
VALVE
CASING GAS
3,000 GRADIENT
4,000
DEPTH, ft
5,000
KILL FLUID
GRADIENT
OPERATING 6,000
VALVE
7,000
8,000
Ps = 4,500
9,000
10,000
Figure 10.24 Depth and pressure when the first valve was unloaded

established, it is directed to the production manifold the first valve decreases and stabilizes at the minimum
and separation train. gradient indicated by the orange line in figure 10.25,
Once gas is flowing from the unloading valve into assuming that sufficient injection gas flow rate is avail-
the tubing, the multiphase gradient in the tubing above able to increase the gas/liquid ratio.
PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

2,000 1ST UNLOADING

VALVE DEPTH
3,000
KILL FLUID
4,000 GRADIENT
DEPTH, ft
5,000
6,000
OPERATING
VALVE DEPTH
7,000
8,000
MINIMUM
FLOW
GRADIENT
9,000
10,000
Figure 10.25 Unloading of the liquid to the flow line after the first valve was uncovered. This assumes sufficient gas
volume to achieve the minimum gradient.

As the pressure in the tubing decreases, the liquid valve should be located. This process is continued as
level in the annulus drops further, until at 4,000 feet a shown in figures 10.25 through 10.28.
balance is reached again between the casing and tubing The actual installation depth of the valves is adjusted
pressures. This is the depth where the next unloading to account for the reduction in annular gas pressure that
PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

2,000
3,000
2ND UNLOADING
VALVE DEPTH
KILL FLUID
4,000 GRADIENT
DEPTH, ft
5,000
6,000
OPERATING
VALVE DEPTH
7,000
8,000
MINIMUM
FLOW
GRADIENT
9,000
10,000
Figure 10.26 Depth and pressure when the second valve was uncovered and gas was injected into the tubing

PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

2,000
3,000
3RD UNLOADING
VALVE DEPTH
KILL FLUID
4,000 GRADIENT
DEPTH, ft
5,000
6,000
OPERATING
VALVE DEPTH
7,000
8,000
MINIMUM
FLOW
GRADIENT
9,000
10,000
Figure 10.27 Depth and pressure when the third valve was uncovered
is necessary to allow the upper valves to close while ing gas into the tubing, it may not be possible to unload
continuing to inject gas through the lowest valve that the well all the way to the operating valve, especially if
has been uncovered. When multiple valves are inject- there is limited gas volume available.

PRESSURE, psi
0 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500

0
50% OIL–50% WATER

2,000
3,000
KILL FLUID
4,000 GRADIENT
DEPTH, ft
5,000
6,000 400 GLR

FLOW
GRADIENT OPERATING
VALVE DEPTH
7,000
OPERATING
VALVE
UNCOVERED
8,000
9,000
PBHP
10,000
Figure 10.28 Depicting the depth and pressure when the operating valve was uncovered and gas was injected at 400
GLR
Monitoring the Unloading Operation monitor the variation of casing and tubing pressure,
The process of unloading should be monitored closely to as shown in figure 10.29. Preferably, a programmable
verify that all is proceeding according to plan. Portable automatic fluid level instrument should also be used to
instrumentation can be connected to the well to at least record fluid level data frequently.

Figure 10.29 Portable pressure recorder monitoring the casing and tubing pressure during gas lift well unloading
The circular two-pen recording shown in figure sure occur as the annular liquid is depressed to
10.30 is probably the most common record that is used the top valve.
in practice and is easily annotated to highlight the main • 3:00 P.M.: The casing pressure drops when the
events. Note that the unloading sequence in this rela- second valve is uncovered. The tubing pressure
tively shallow well spans a period of about 12 hours. shows a corresponding surge when the additional
As described later, the unloading process should not gas reaches the surface.
be rushed to avoid damaging the valve seats and stems • 5:20 P.M.: The third valve is uncovered, and the
because of excessive liquid flow rate. casing gas pressure drops even further.
The following is an approximate time sequence • 6:00 PM: A pressure surge at the tubing indicates
of events: that additional gas flow reaches the surface.
• 10:00 A.M.: The static load fluid level was near
• 7:45 P.M.: The operating valve is uncovered.
the surface in the casing and tubing before initial
Casing pressure drops slightly, and tubing pres-
unloading began. The wellhead pressure remains
sure shows a small surge about 15 minutes later.
relatively constant during U-tubing operations
before injection gas enters the tubing for the first From this point forward, the pressures stabilize,
time through the top gas lift valve. indicating that a steady state condition has been reached.
• 12:00 P.M.: A surge in wellhead tubing pressure Monitoring the gas injection flow rate is generally done
and a decrease in the injection gas casing pres- at the injection manifold, as shown in figure 10.31.

10 A.M. 11 A
.M .
M.
9 A. 1000
NO
. ON
.M 900
8A
800
00
10 U-TUBING
1
.
.M
P.
0 OPERATIONS
M
A
90
.
7 80
0 500 TOP VALVE
10
00
90
00
2P
80
.
0
.M
400
70
.M
6A
0
0
60
.
60
50
0
300 40
0
0
0
30
50
200 0
20
3 P.M
0
.
40
5 A.M
WELLHEAD 100 10
0
TUBING 0
30 SECOND VALVE
.
PRESSURE
0
20
1000
0
10
200
300
100
900
400
500
800
600
4 P.M.
700
4 A.M.
700
600
94
800
500
84
400
100
300
900
200
79
1000
10
INJECTION-GAS
20
CASING PRESSURE
0
0
THIRD VALVE
30
5 P.M
10
.
3 A.M
0
0
40
20
.
0
0
30
50
0
0
40
60
0
50
0
0
60
.
6N
.M
70
70
0
0
2A
IG
80
00
00
HT
80
9
10
FOURTH (OPERATING) VALVE 0

90
Figure 10.30 Two-pen pressure-recorder chart illustrating continuous-flow gas lift unloading operations with choke
control of the injection gas
Figure 10.31 Gas lift injection manifold

Procedures for properly unloading a gas lift instal- the casing pressure has reached at least 300
lation must be followed to prevent possible damage to psig. Most companies use a standard choke size
gas lift valves and attain the design depth of lift. If a in the injection gas line for U-tubing and initial
permanent metering facility is not installed in the injec- unloading operations. A typical injection gas
tion gas line to the well, a portable flow meter should be choke size ranges from 6⁄64 to 8⁄64 inches for the
installed prior to unloading for adjustment of the injection U-tubing operation.
gas rate to the well. Ideally, the meter or flow computer 4. After the casing pressure has reached 300 to 500
should be located near the well’s injection gas control psig, the injection gas rate can be adjusted to
device so that the effect of changes in the adjustment of allow a 100-psi increase per 10-minute interval
the injection gas volume can be easily observed. until gas begins to circulate through the top gas
The initial U-tubing is the most critical operation lift valve. (The top valve is uncovered.) After
during the unloading procedure. There is no reason to the top gas lift valve is uncovered and gas has
hurry the U-tubing of the kill fluid to uncover the top gas been injected through this valve, a high pressure
lift valve. Because the tubing is full of liquid, there is no differential cannot occur across the lower gas lift
drawdown in flowing BHP. In fact, the BHP increases valves. Any time the casing injection gas pres-
until gas lifting from the top valve occurs. sure is increased above the opening pressure of
The kill fluid production rate is controlled by the the top valve, the valve will open and prevent a
rate of increase of the injection gas pressure, which in further increase in the injection gas pressure.
turn depends on the injection gas rate. Because most gas 5. If the gas lift installation does not unload to the
lift installations include a packer, the kill fluid enters the bottom valve or the design operating gas lift valve
tubing through the open gas lift valves. When the kill depth, adjustment of the injection gas rate to
fluid has not been properly treated and contains solids the well is required. An excessive or inadequate
and debris and a high line injection gas pressure is ap- injection gas rate can prevent unloading.
plied to the casing, the annular liquid will be displaced
at a high rate and erode the valve seats and stems. Proper adjustment of the injection gas volume to
The following procedure is recommended for moni- a well is not permanent for most installations. The in-
toring and controlling the unloading operations for all jection gas requirements change with well conditions.
gas lift installations to prevent damage to the gas lift Therefore, continuous monitoring of the injection gas rate
valves and surface facilities. and the wellhead and injection gas pressure and frequent
1. If the well has been shut in and the tubing pres- acquisition of acoustic fluid levels is recommended to
sure exceeds the separator pressure, bleed down maintain efficient gas lift operations.
the tubing through a small flow line choke. Do
not inject lift gas before or while the tubing is SUMMARY
being bled down. This chapter addresses the most common applications
2. Remove all wellhead and flow line restrictions, of fluid level measurements to monitor and optimize
including a fixed or adjustable choke, if the well the design and operation of wells produced by continu-
does not flow after all kill fluid has been produced. ous or intermittent gas lift. Fluid level measurements
If the gas lift installation is in a new well or a are used from the beginning of gas lift installation and
recompletion that could flow, a 24⁄64- to 32⁄64-inch commissioning to determine the progress of the liquid
flow line choke is recommended until the well unloading phase through the establishment of steady
has cleaned up and does not flow naturally. state flow and monitoring of daily operations. Acoustic
3. Inject lift gas into the casing at a rate that does records are obtained both in the annulus and the tub-
not allow more than a 50-psi increase in casing ing to establish the actual operating pressure and fluid
pressure per 10-minute interval. Continue until distribution, which can be compared with the values

from the system originally designed using multiphase REFERENCES

flow software or charts. Recommendations are given
1. G. Webber, “Fluid Level Indicator Useful in East
regarding proper installation of the acoustic hardware to
Texas,” Oil and Gas Journal, December 1938.
ensure that accurate records are obtained. In particular,
connections through small orifices or needle valves 2. API Specification for Gas Lift Equipment, 1st ed.
should be avoided. (Spec. 11V1, 1995).
For quality control, multiple records should be 3. API Recommended Practice for Design of Con-
acquired to identify random signals that are not gener- tinuous Flow Gas Lift Installations Using Injection
ated by wellbore features and could be interpreted as Pressure Operated Valves, 2nd ed. (RP 11V6, 1999).
acoustic echoes. In deep wells with numerous gas lift
valves, the records will exhibit a large number of primary BIBLIOGRAPHY
echoes and multiple reflections. A reasonably accurate
H. P. Arendt, C. Dines, and T. Heard, “Pumpdown
analysis can be achieved only by referring to an accu-
(TFL) Technology for Subsea Completions,” Journal
rate wellbore diagram detailing all downhole tools and
of Petroleum Technology, 1978.
their actual depths. Variations of acoustic velocity with
temperature must be considered when analyzing these J. R. Blann, R. Garcia, F. Guaramata, et al., “Advances
deep well records, as illustrated by several examples. in Heavy Oil Lifting in the Morichal Area of Venezuela,”
Acoustic surveys acquired at frequent intervals over the SPE Mid-Continent Operations Symposium, 1999.
life of the well are useful in troubleshooting and detecting J. R. Blann, and G. M. Laville, “Gas Lifting a Major Oil
malfunctions, such as valves stuck open, leaking pack- Field in Argentina with High CO2 Content Associated
ers, corrosion/erosion holes, and other failures. Even Gas,” SPE Production & Operations, 1997.
in wells where a packer isolates the annulus from the
R. J. Dickens, “High-Pressure Gas Lift for Deep, Sour
perforations, it is possible to estimate the PBHP from
Production,” SPE Production Engineering,1988.
the acoustic survey by adding the pressure drop due to
multiphase flow in the tubing to the annular pressure S. Noonan and K. Decker, “Subsea Gas Lift Design and
computed at the operating valve. Analysis for Chevron’s Subsea Developments,” ASME/
The SBHP can be estimated accurately from acoustic API Gas Lift Workshop, 2001.
surveys acquired down the tubing after shutting in the R. Stinson, “Equipment Development for Gas Lift in
flow and waiting for the pressure to stabilize. In addition Deepwater,” ASME/API Gas Lift Workshop, 2001.
to the conventional chart recording of the tubing and cas-
ing pressure, the liquid unloading process is easily and G. Takacs, Gas Lift Manual (PennWell Publishing Co.,
efficiently monitored by repeated fluid level measure- 2005).
ments in the annulus that verify when the operating valve C. Taylor, “Application of Acoustic Fluid Level Mea-
is uncovered and the system is operating as designed. surements in Gas Lift Wells,” Echometer Co., 2014.


Fluid Level Measurement Applications for Plunger Lift Wells 11-1
11
Fluid Level Measurement Applications for Plunger Lift Wells
In this chapter:
• Detailed analysis of plunger cycle performance
• Basics of plunger lift to produce liquid and gas
• Acoustic survey and monitoring
• Plunger fall characteristics
• Problem detection and analysis
• Benefits of plunger tracking
Plunger lift is a low-cost method for lifting liquids is repeated throughout the day to produce liquids and
(water, condensate, and/or oil) from gas and oil wells. gas from the well.
In general, the objective is to remove as much of the In plunger lift wells, acoustic fluid level instruments
liquid accumulating in the well as possible and to are used to monitor and analyze the progress of the cy-
increase gas production by minimizing back-pressure clical plunger operation in real time and determine the:
on the formation. The plunger lift system reduces the • Position and depth of the plunger as a function
cost of operating a well compared to other artificial lift of time
methods because the formation pressure supplies most • Fall velocity of the plunger
of the energy required to lift the liquids. • Rise velocity of the plunger
During plunger lift operations, a motor-controlled • Plunger travel time to the liquid and to the bot-
valve is opened and closed at specified intervals to cycle tom of the tubing
a gas-driven plunger from the bottom to the top of the • Tubing and casing pressure as a function of time
tubing and remove any liquid that accumulated at the
• Volumes of gas and liquid flowing into and out
bottom of the well. When the surface valve to the flow
of the well
line is closed, the produced gas and liquid accumulate
inside the well’s casing and tubing. When the flow is The objective is to visualize in detail the performance
stopped at the surface, the plunger falls down to the of the plunger lift system to determine the appropriate
bottom of the tubing. After a predetermined amount cycle time for optimum operation. Acquisition and
of time, the surface flow valve opens, and the tubing analysis of acoustic and pressure data is generally per-
head pressure drops to the flow-line pressure. The dif- formed automatically, using a portable computer with
ferential force across the plunger—due to the drop in appropriate software. Thus, the operator can quickly
pressure in the tubing above the liquid column and the and efficiently determine the adjustments necessary to
high well pressure below the plunger—lifts the plunger optimize the plunger lift operation1,2.
and a portion of the liquid above the plunger to the The following sections present a brief overview of
surface. Gas and some liquid continue to flow out to plunger lift operation and describe in detail the equip-
the flow line until the motor valve is closed. The open ment and procedure used to acquire and interpret the
and shut-in operational cycle of the plunger lift system acoustic data for plunger lift analysis.
For distribution by Petroleum Extension-The University of Texas11-1

at Austin
PLUNGER LIFT SYSTEM attenuate the impact of the plunger upon arrival
and provides access for plunger removal and
Although there are several variations, the most common
inspection.
plunger lift installation is represented schematically in
• The motorized valve: Provides open/close con-
figure 11.1 and includes the following components: trol of the flow from the tubing to the flow line.
• The lubricator/catcher: The uppermost stopping Generally, it is diaphragm-operated.
point for the plunger where it stays locked during • The controller: Generally, a microprocessor-
the flow period. It acts as a shock absorber to based system that controls the duration of the
LUBRICATOR ELECTRONIC
CONTROLLER
FLOW TEE
WITH O-RING
MOTORIZED
VALVE
BYPASS
VALVE
BLEED
VALVE
EXTERNAL
CABLE
CATCHER
WITH ARRIVAL
MASTER SENSOR
VALVE
PLUNGER
BUMPER SPRING
TUBING STOP
Figure 11.1 Schematic diagram of a typical plunger lift installation

plunger cycle by opening and closing the motor Figure 11.2 illustrates some of the surface compo-
valve. Detects the presence of the plunger in the nents of the plunger system.
lubricator and may monitor other parameters
such as pressures. Types of Plungers
• The arrival sensor: A transducer that senses the Currently, over thirty types of plungers are commercially
arrival and/or the presence of the plunger inside available in North America3 that are designed to meet
the lubricator. the requirements imposed by varying conditions and
• The bumper spring: A shock absorber that at- performance of the wells during their life cycle. Figure
tenuates the impact of the falling plunger when 11.3 shows examples of commonly used plungers with
it arrives at the bottom of the tubing. different types of shapes, sizes, and seals.
• The tubing stop and check valve: Allows seating
of the bumper spring and prevents liquid that ac- Types of Controllers
cumulates above the plunger from flowing back Controllers differ in terms of their complexity, flexibility,
from the tubing to the wellbore below. and programmability and correspondingly in terms of
• The plunger: Elongated cylindrical element man- their cost and need for an experienced technician for
ufactured in numerous configurations, which pro- setup and maintenance.
vides an interface between liquid above and gas
below during the unloading phase of the plunger Timers
cycle. It provides a partial seal that prevents The simplest design consists of manually entered on/off
excessive slippage of gas and liquid as it is dis- set points. They require constant monitoring and adjust-
placed from the bottom to the top of the tubing. ment as the well’s inflow and artificial lift performance
Figure 11.2 Surface components of the plunger installation

Figure 11.3 Various types of plungers

change. Smart timers adjust the on/off points based on reaches the top of the accumulated liquid at the
plunger ascent velocity. bottom of the tubing. The plunger then falls through
Pressure Differential or Pressure Set Point
the accumulated liquid at the bottom of the tubing
and rests on a plunger catcher or bumper spring.
These controllers sense the pressures at the tubing head During shut-in, most of the gas produced from the
and casing and open/close the motor valve based on their formation accumulates in the annulus, causing the
difference and/or their values. The operating limits are casing pressure to increase. A portion of the gas
input by the production technician. also enters the bottom of the tubing, commingled
Programmable Logic Controllers with liquid produced from the formation. The gas
These controllers use a combination of pressure values, percolates through the liquid and flows to the top
differential pressure, plunger travel time, flow rate, and of the well, causing the tubing head pressure to
so on to determine the best operating cycles by adapting increase. The liquid accumulates at the bottom
to the well conditions. of the tubing. The casing pressure should build
The logic used by these controllers varies in com- high enough to lift the accumulated fluids and the
plexity in relation to the microprocessor performance, plunger to the surface during the next step when
but in general, the control algorithms are based on the motor valve opens.
comparing the actual plunger velocity to some basic 2. The unloading period begins after a predetermined
value for the specific type of plunger, installation, and amount of time has elapsed from the start of the
pressure levels. shut-in period. Based on satisfying some required
If the plunger ascends too fast, it is assumed that the: operational criteria, the valve controller opens
• Casing pressure is too high the motor valve between the tubing and the flow
• Slug size is too small line. The tubing head pressure drops to near the
• Line pressure has fallen value of the line pressure, and correspondingly
the pressure at the top of the liquid column ac-
Then the controller may respond by increasing
cumulated above the plunger also decreases. The
the after-flow time and/or decreasing the length of the
pressure below the plunger is due to the annular
shut-in period.
BHP from the gas stored in the casing. The dif-
If the plunger ascends too slowly, it is assumed
ferential pressure across the plunger generates
that the:
the force that lifts the plunger and the liquid slug
• Casing pressure is too low
to the surface. As gas from the casing flows into
• Liquid slug size is too large
the tubing and below the plunger, the bottomhole
• Line pressure has increased
casing pressure decreases. Additional fluids flow
• Plunger seal is becoming less efficient
from the reservoir into the well and the bottom
Then the controller may respond by decreasing of the tubing.
the after-flow time and/or increasing the length of the 3. The after-flow period begins after the liquid and
shut-in period. the plunger arrive at the surface. The flow valves
are open; the liquid slug is discharged to the flow
Plunger Lift Operation Cycle
line. When the plunger arrives, it is held at the
The plunger lift cycle can be divided into three distinct surface inside the lubricator by differential pres-
phases, as shown schematically in figure 11.4. sure caused by the flow of gas up the tubing and
1.
The shut-in period begins when the flow-line mo- into the sales line. If the gas velocity in the tubing
tor valve closes, the flow of gas is stopped, and is high enough to lift liquid to the surface, some
the plunger drops out of the lubricator and into additional liquid will also be produced with the
the tubing. The plunger falls through gas until it gas.

Conventional Plunger Cycle
VALVE CLOSES, SHUT-IN CONTINUES, PLUNGER ARRIVES.

SHUT-IN BEGINS AND VALVE CLOSED VALVE OPENS, TUBING PRESSURE VALVE CLOSES,
PRESSURE STARTS WITH PRESSURE UNLOADING SPIKE MAXIMUM FLOW STOPS AND
INCREASING INCREASING BEGINS AFTER-FLOW BEGINS CYCLE REPEATS
SHUT-IN UNLOADING AFTER-FLOW
Figure 11.4 Three phases of plunger lift: shut-in, unloading, and after-flow
During the after-flow period, the gas rate decreases one valve opening—must be monitored. If the operator
as gas is transferred from the annulus to the tubing, where is only interested in determining the plunger fall velocity,
at some point liquids are no longer carried to the surface then data acquisition should start just before the shut-in
because the gas velocity becomes too low. The liquid in period begins and continue until after the beginning of
the tubing will fall back and accumulate at the bottom. the unloading period. To analyze a well that behaves
The motor valve is closed after a predetermined time erratically may require the acquisition of data over
elapses or when a specific control criterion is satisfied several cycles to correctly analyze the operation of the
starting the next shut-in period. plunger lift installation and identify potential problems.
Determining Plunger Position and Velocity

ACOUSTIC FLUID LEVEL MONITORING OF To visualize the performance of the plunger lift system
PLUNGER WELL OPERATION in detail and determine the appropriate cycle time for
A complete analysis of the plunger lift system requires optimum operation requires accurate knowledge of the
that at least one complete cycle—consisting of one valve position of the plunger at all times during the cycle,
opening plus one valve closing or one valve closing plus including when it is traveling up or down inside the

tubing. Acoustic fluid level instrumentation is ideally volume chamber and contaminate the gun mechanism.
suited for this purpose, since acoustic pulses can be used Other types of gas guns may require the gas gun chamber
to monitor the distance from the surface to the plunger. to be charged to a pressure greater than the maximum
This can be done either actively by acquiring acoustic expected well pressure in order to keep the internal gun
records (explosion or implosion mode) or passively by mechanisms closed throughout the test and to protect
monitoring acoustic pulses generated by the plunger as the internal mechanism from the well fluids.
it travels past each tubing coupling, as discussed in the Detailed discussion of the various methods that may
following sections. be used for monitoring the plunger position during the
The most common data acquisition hardware for shut-in periods is found in various other publications4,5.
plunger lift monitoring includes an acoustic gas gun The gun’s microphone is connected to electronic
with microphone and pressure transducer connected circuits to amplify and process the acoustic pulses and
to the tubing through a H-inch or larger, fully opening background noise generated throughout the cycle. A
valve connection on the lubricator and a second pressure second pressure transducer is connected to the casing-
transducer connected to the casinghead. Figure 11.5 il- head to monitor and record the variation of pressure.
lustrates typical installations Note that if a needle valve Depending on the type of plunger, the properties of
is present on the well at the point of gun connection, it the fluids, and the condition of the tubing, the speed of
should be replaced with a fully opening ball valve before the plunger falling or rising ranges from a few hundred
connecting the gas gun in order to record the best quality to a little over 1,000 feet per minute. In comparison, the
acoustic signal. The preferred gas gun used should be speed of acoustic pulses propagating in the gas is typi-
cocked manually so that well fluids will not enter the cally between 1,000 and 1,500 feet per second. Thus, the
Figure 11.5 Typical installations of acoustic hardware used for plunger lift monitoring

round trip travel time of an acoustic pulse generated at the operator. Typical acquisition frequency is one record
surface that reflects from the top of the plunger while it every 2 to 3 minutes until the plunger reaches the bot-
is traveling at a depth of 5,000 feet is between 6 and 10 tom of the tubing. Each record is analyzed to calculate
seconds. During this short time, the position of the plunger the position of the plunger at the time when the record
would change by at most 30 to 50 feet, which implies that was acquired, based on detecting the echo from the
it is possible to use the conventional acoustic fluid level plunger and processing the data to calculate its distance
equipment to determine the position of a moving plunger from the wellhead. Since it is customary to also acquire
with an accuracy of the order of one tubing joint. The fluid level data in the annulus, the acoustic velocity of
uncertainty of the computed depth increases as the well the gas in the tubing is obtained from the analysis of
depth increases. However, for most wells, this method the annular fluid level record. Usually, the difference
yields a position-versus-time relation that is sufficiently in pressure between the tubing and the casing is only a
accurate to determine the plunger fall velocity as it travels few hundred psi, so it is acceptable to use the acoustic
from the wellhead to the bottom of the tubing. velocity determined in the casing to calculate the depth
of the plunger in the tubing. The difference in plunger
Active Acoustic Monitoring of Plunger Position depth between two successive records and the differ-
Acquisition of the acoustic records can be performed ence in acquisition time are used to calculate an average
manually or by using an automated scheduler that is velocity of the plunger. These values are plotted as a
programmed to collect data without monitoring by an function of time, as shown in figure 11.6.
Worn Brush Plunger 477 fpm Fall
TIME OF DAY
10:04:48 10:07:41 10:10:34 10:13:26 10:16:19 10:19:12 10:22:05 10:24:58 10:27:50
0 800
1,000 700
ACOUSTIC SHORT PLUNGER DEPTH, ft
VALVES CLOSE, PLUNGER HITS FLUID VALVES OPEN,
PLUNGER FALL VELOCITY, ft/min

2,000 PLUNGER 260 ft ABOVE TUBING PLUNGER BEGINS 600
BEGINS BOTTOM AT 10:22:00 RISE TO SURFACE
FALL 10:07:00 10:25:00
3,000 500
PLUNGER AVG PLUNGER FALL
VELOCITY VELOCITY 477 ft/min
4,000 400
5,000 300
PLUNGER
DEPTH
6,000 200
7,000 DEPTH TO LIQUID IN TUBING 7,140 ft 100

STANDING VALVE DEPTH 7,400 ft
8,000 0
Figure 11.6 Chart depicting the depth of the plunger versus time obtained from manually acquired acoustic records

The red points show the position of the plunger at the coupling recess immediately below the bottom of
a specific time until the plunger enters the liquid ac- the plunger expands to the lower pressure that exists
cumulated at the bottom of the tubing; then they show above the plunger when the top of the plunger moves
the depth to the liquid at 7,140 feet. The blue points below the coupling recess. The coupling recess acts as a
represent the average velocity of the plunger (ft/min) “mini” gas gun chamber and “fires” a pulse that travels
for the interval between two depth values. through the gas to the surface, where it is detected by the
Note that the plunger velocity is not a constant but microphone and the tubing pressure transducer. These
tends to decrease as a function of depth from about 580 acoustic pulses are normally obtained when a plunger
to 400 feet per minute for this particular type of plunger. falls down the tubing in a well that produces a limited
The plot also shows that this particular plunger’s fall amount of liquid, so the tubing interior is relatively dry.
to the bottom lasted about 15 minutes. This type of test A 0.2-psi amplitude pressure wave (green curve) and an
should be repeated to establish the consistency of the acoustic signal (black curve) are generated as the plunger
plunger behavior and determine a reasonable average fall falls past the 111th and 112th tubing collar recess at a
time to be used for initial setup of the valve controller. depth of 3,584 feet, as shown in figure 11.7.
Acoustic records can also be acquired while the By monitoring and recording at the surface these
plunger is at the bottom of the tubing, with the objective pulses, which were generated at tubing coupling recesses,
of monitoring the rate of liquid inflow by determining the plunger travel can be followed on a continuous
the depth of the top of the liquid slug at two or three basis. Therefore, it is not required to periodically fire
different times during the shut-in period. Knowing the the gas gun to determine the position of the plunger
capacity (bbl/feet) of the tubing string, the change in by echo ranging. The method of acoustically record-
volume of accumulated liquid can be converted to a ing plunger-generated pressure waves from the tubing
producing flow rate from the formation. This value collar recess has been defined as “passive” monitoring
should agree with the volume measured from a recent of the plunger position during the fall. A schematic for
production well test. the instrumentation setup is shown in figure 11.8, with
pressure sensors connected to both the tubing and casing
Passive Acoustic Monitoring of Plunger Position and with the acoustic gas gun and microphone installed
When using a modern digital acoustic fluid level instru- on the tubing. For passive monitoring, a high rate (30
ment, the sensitivity of the system is such that the gas Hz or greater) for data acquisition is used to record the
gun microphone can be used to detect plunger location signals from both tubing and casing pressure sensors,
during the plunger fall by monitoring and digitally plus the acoustic signal from the microphone.
recording the low-amplitude acoustic noise inside the When the plunger enters the liquid, the tubing recess
tubing as a function of time. During the majority of the acoustic pulses are generally not transmitted through the
field tests recorded to date, it has been observed that an liquid and detected at the microphone, but in all cases,
acoustic pulse is generated when the plunger falls past a the acoustic noise amplitude level drops, indicating that
tubing collar recess. Because of the upward flow of gas the plunger is submerged in the liquid.
in the tubing and the downward velocity of the plunger, However, in some wells, it is possible to see acous-
a difference in pressure exists across the plunger as it tic pulses as the plunger falls through the liquid. The
falls. The pressure difference depends on the area of acoustic and tubing pressure data in figure 11.9 show an
the plunger and the clearance between plunger outside example where tubing collar recess pulses are detected
diameter (OD) and tubing inside diameter (ID). This at the microphone, both when the plunger falls through
difference in pressure across the plunger is generally the gas above the liquid and after the plunger enters the
of the order of 2 to 10 psi. liquid at the bottom of the tubing. The frequency of the
As the plunger moves through a tubing collar, a pres- pulses (pulses per unit time) depends on the speed of
sure wave is generated when the gas volume occupying the plunger. In this case, it drops from an average of

C111 C112—3,584 ft deep

424.80 0.03
424.70 0.2 psi 0.02
424.60 0.01
424.50 0
424.40 -0.01
424.30 ACOUSTIC SIGNAL SHOW PLUNGER FALL -0.02

PAST 111TH AND 112 TUBING COLLAR
424.20 -0.03
95.812 95.867 95.923 95.978 96.034 96.089
Figure 11.7 Record showing the expanded portion of an acoustic signal and the tubing pressure versus the time
recorded during plunger fall
LUBRICATOR
ACOUSTIC GAS
VALVE GUN AND TUBING
CONTROLLER PRESSURE SENSOR
FULL BORE
MASTER
VALVE
FLOW
MOTOR VALVE
CASING
PRESSURE
LIQUID SLUG SENSOR
PLUNGER
BUMPER SPRING
Figure 11.8 Acoustic recorder and pressure sensor installation schematic

Plunger Fall in Gas and Liquid

Raw Data Select Cycle Cycle Limits Plunger Fall Gas Property Cycle Analysis Plots
446.00 201 ft/min 39 ft/min 0.06

Gas Liquid
444.00
PRESSURE 0.04
442.00
PLUNGER HITS LIQUID
440.00
0.02
MICROPHONE SIGNAL
438.00
0
436.00
PLUNGER ON BOTTOM
434.00
-0.02
432.00
430.00 -0.04
104.331 107.903 111.276 114.749
TIME, min
X-Axis Range 13 mins Full Trace Options
Figure 11.9 Example acoustic record showing pulses generated by the plunger as it falls through the gas and
through the liquid at the bottom of the tubing
about six pulses per minute in the gas section to about one need to be used. Figure 11.10 shows an example record
pulse per minute in the liquid section, corresponding to for a plunger cycle from shut-in to shut-in that lasted
the slowing down of the plunger when it is falling inside about 93 minutes and consists of about 170,000 data
the liquid. When the plunger finally rests at the bottom points for each of the three traces, plotted versus time:
on top of the bumper spring, the noise level drops again • Casing pressure (blue)
and a small increase in tubing pressure is observed at the • Tubing pressure (green)
surface. These features make it possible to determine fairly • Acoustic amplitude (black)
accurately the time when the plunger reached bottom.
Adequate reliability of the computer and electronic
Data Acquisition and Recording for Passive hardware and excellent stability of the acquisition software
Monitoring are required to manage successfully the large volume of
recorded data and avoid having to repeat this lengthy
The acoustic signals from downhole and surface pressure
test because of unexpected failures or program crashes.
data are digitized and stored in a computer for display
in real time and subsequent analysis. Data points are ac- Identifying and Annotating Key Events
quired at a default rate of 30 per second. For high-pressure To aid in the analysis of the plunger record, it is impor-
wells or fast-falling plungers, faster sampling rates may tant that specific events taking place during the cycle be

320.0 0.5
300.0
280.0
260.0
240.0
220.0 0
200.0
180.0
160.0
140.0
120.0 -0.5
0.000 13.889 27.778 41.667 55.556 69.444 83.333
Figure 11.10 Acoustic and pressure record for a complete plunger cycle
identified by annotating the acquired data. For events peak amplitude of the acoustic pulses decreases
such as the time when the control valve opens or closes as time increases.
or the plunger starts to fall, reaches the liquid, stops at • The plunger reaches liquid: The signals generated
bottom, and so on, the specific points of their occurrence as the plunger passes through the tubing collars
in time are marked on the record and flagged with cor- disappear, followed by reduced acoustic noise.
responding notes. This process of annotating the record • The plunger reaches the bottom of tubing: Further
can be done while the data is acquired, when the opera- reduction in the acoustic noise amplitude.
tor is monitoring the process, or after completion of the • The start of unloading: The tubing and casing
acquisition when the data is reviewed and analyzed. pressure start to drop. The acoustic signal level
Events related to the opening and closing of the shows increase in noise.
control valve are identified by rapid changes in the value • The liquid reaches tubing head: There is a sharp
increase in acoustic signal and tubing head pres-
of the recorded pressure. Events related to the movement
sure. An inflection point often is observed in a
of the plunger and flow of fluids are characterized by
casing pressure versus time graph.
changes in amplitude of the acoustic signal and some-
• The plunger arrives in the lubricator: The tubing
times also by changes in pressure.
pressure reaches a peak and then starts to drop as
The following are some of the acoustic and pres-
flowing fluid changes from liquid to gas. If the
sure signal characteristics corresponding to key events
plunger arrives dry (no liquid reaches the surface),
occurring during the plunger cycle: then a sharp acoustic pulse is detected.
• The plunger release: The acoustic signal amplitude • The start of the after-flow: A decrease in tubing
increases, and the tubing pressure may drop slightly. pressure and inflection point on casing pressure
• The plunger falling in gas section: The peak-to- is observed.

Unexpected events—such as the plunger fall being of the plunger system in terms of gas and liquid produc-
interrupted during shut-in or the motor valve leaking or tion per cycle, to identify problems, and to help optimize
actuating at the wrong time—will cause anomalies in the operation of the system.
the pressure and acoustic record that must be identified
and annotated. In wells where the cycle times seem er- Instantaneous Plunger Fall Velocity
ratic, it is recommended that the operator monitor the An important result of acoustic plunger lift analysis is
progress of the test, take accurate notes, and record the the accurate determination of the position of the plunger
time when these occur. A complete analysis may require as a function of time so that an instantaneous plunger
the acquisition of data for multiple plunger cycles, espe- fall velocity can be computed in addition to its average
cially in wells that exhibit random fluid production tests. value. When pulses generated by the falling plunger are
Tubing and casing pressure are recorded simultane- detected in the acoustic record, this instantaneous veloc-
ously with the acoustic signals to undertake calculation ity value can be obtained from the difference in arrival
of the pressure distribution as a function of time in the time of the signals from two consecutive tubing joints.
tubing and wellbore. The variation of pressures versus Starting at the time when the plunger begins to
time during the plunger cycle is then used to calculate fall, the pulses are each identified and numbered on the
volumetric flow of gas from the reservoir into the well record (in an expanded time scale), as shown in figure
and from the well to the flow line. The objective of these 11.11. The depth to each joint is obtained from a detailed
calculations is to present an analysis of the performance tubing tally or estimated using the average tubing joint
Velocity: Plunger Fall Speed Between

Two Consecutive Collar Pulses
Plunger velocity @ joint 22 equals the change in depth divided by the difference in pulse arrival time.
Velocity = (D23 – D22) / (T23 – T22) = 230.9 ft/min
D22 = 676.2 D23 = 708.4
C21 C22 C23 C24 C25 C26 C27
T22 = 5.663 T23 = 5.802
|<-- <<< Color # C22 <<< |<--

5.802 mins <-- --> Add Delete Cursor Mode
(A) (1) # Time (mins) Velocity Depth

C19 5.359 -209.24 611.90
C20 5.519 -201.95 644.00
Looking at this Minute C21 5.663 -223.78 675.20
C22 5.802 -230.92 708.40
C23 5.961 -203.37 740.60
C24 6.109 -217.08 772.80
Falling through Gas C25 6.254 -221.22 805.00
C26 6.411 -205.53 837.20
C27 6.558 -218.72 869.40
C28 6.716 -204.81 901.60
Figure 11.11 One-minute section of an acoustic record of pulses generated at the tubing couplings, and the method
for determining the plunger fall velocity

length multiplied by the pulse sequential number (for 11.13. A total of 227 collars have been counted when
example, the depth to coupling no. 22 is equal to 22 × the plunger hits the liquid at a fall velocity of -135.42 ft/
32.2 feet/joint = 708.4 feet). min. (The negative sign indicates downwards velocity.)
The process of identifying and counting the pulses The blue curve represents the plunger depth as it
generated at sequential tubing couplings and calculat- falls through the gas in the tubing and reaches the top
ing the plunger velocity is generally performed by the of the liquid at the bottom. The last point represents the
acquisition software, either automatically or with some time when the plunger stops on top of the bumper spring.
user intervention. With access to the digitized acoustic In some cases, this time may not be determined with
record, the information can also be transferred to a ge- accuracy. The green graph represents the general trend
neric spreadsheet to create a detailed graph, as seen in of the computed plunger fall velocity corresponding to
figure 11.12, which allows identifying the pulses and each sequential tubing joint. The oscillations in value
determining the arrival time from each tubing joint. are partly due to uncertainty in the time determination
In either case, the calculated results are generally (the sampling rate of 30 points per second provides a
presented as a graph and/or table of plunger velocity and resolution of ±30 feet for an acoustic velocity of 1,150
position as a function of real time, as shown in figure ft/s) and partly due to variations of the actual length of
Recorded Collar Pulses During Plunger Fall

6.00E-02
4.00E-02
2.00E-02
AMPLITUDE, mV
0.00E+00
-2.00E02
-4.00E-02
-6.00E-02
4,500 6,500 8,500 10,500 12,500 14,500
ELAPSED TIME, sec
Figure 11.12 Graph of acoustic pulses recorded digitally during plunger fall

Normal Fall Velocity (During Shut-In)

Close Eject Print Invert Velocity Axis
Collar # C227
Depth to Plunger 7,309.40 ft Plunger Fall Velocity 134.48 ft/min Elapsed Time 47.452 mins
-100 Falling through Gas 0

Gradually Slows from
240 ft/min to 135 ft/min
-120
1,600
-140
SLOWER
DEPTH TO PLUNGER, ft
-160
3,200
PLUNGER VELOCITY
-180
PLUNGER DEPTH 4,800

-200
FASTER Normal Fall Falling

-220
Velocity Profile: through 6,400
• Tubing is okay Liquid
-240 • Liquid in bottom
Plunger Hits Liquid—7,313.58 ft
Bottom of Tubing—7,773.00 ft 8,000

-260
0.000 8.081 17.361 26.042 34.772 43.403 52.083 60.764
ELAPSED TIME, mins
Figure 11.13 Record showing the plunger depth and velocity as a function of time elapsed since the start of plunger fall
each tubing joint. Despite this uncertainty, it is possible plunger velocity trace, notice that the left vertical scale
to observe how the velocity of this particular plunger in is amplified and that the general trend of the velocity
this particular tubing string is not a constant but tends to is to consistently slow down as time (plunger depth)
decrease (from about 240 to 135 ft/min) as the plunger increases. When liquid accumulates in the bottom of
falls to the bottom. The actual time for the plunger to fall the tubing, it is normal for the plunger to initially fall
from the surface to the liquid at 7,314 feet is indicated quickly and gradually slow down as it gets deeper into
as 47.4 minutes, giving an average velocity of 154 ft/ the well. Significant deviations from this trend would be
min in the gas section. an indication that the identification of the collar signals
Figure 11.13 should be used for quality control of should be reviewed and verified or that there may be a
the analysis process because it represents the position problem with the operation of the well. An operational
of the plunger and its instantaneous velocity up to the problem may exist in the well if the plunger falls at a
depth of the collars counted. The normal trend is for constant speed or abruptly slows down or speeds up
the plunger depth to increase smoothly as a function of during its fall, as discussed in the following section on
time. Although there seems to be significant noise on the troubleshooting plunger wells.

Factors Affecting Plunger Fall Velocity annulus or the tubing. Therefore, the best estimate of the
Several studies have been published related to the gas gravity will be obtained from measurements within
variation of plunger velocities in relation to the type the well or using gas samples taken at the wellhead.
and design of the plunger3,4,6. The following are some The gas gravity may be determined in various ways:
of the major observations: • A value computed from the acoustic signals
recorded during the plunger fall
• Diameter of plunger: A larger diameter causes
• A measured value determined from testing a gas
slower plunger fall.
sample taken at the wellhead
• Effectiveness of seal between plunger and tubing:
• A computed value based on composition of a gas
A better seal means that the plunger falls slower.
sample taken at the wellhead
• Brush stiffness: If the bristles do not provide an
• A value previously determined from acoustic
effective seal, then the plunger falls faster.
measurements in the well
• Gas through the plunger: If gas can pass through
the plunger (specifically, bypass), then the plunger In general, it can be assumed that the average gravity
falls faster. of the gas in the tubing is fairly well-represented by the
• Tubing: If the tubing is sticky, the plunger falls average gravity of the gas in the annulus. Therefore, a
slower. fluid level record acquired connecting to the casinghead
• Gas flow rate into the tubing: Gas flow into tub- can be used to obtain an average acoustic velocity to be
ing reduces plunger fall velocity used in conjunction with an equation of state to compute
• Old age/increased wear: As the plunger wears the gas gravity, as discussed in chapter 6.
out, the worn plunger falls faster. Alternately, acquiring a liquid level record in the
• Wellbore deviation: More than 20° of deviation tubing during the shut-in period also provides a direct
impacts plunger fall velocity. estimate of acoustic velocity. However, very often it is
– Padded plungers fall faster because of loss of not necessary to shoot a fluid level, since acoustic ve-
seal. locity in the tubing gas can be computed by analyzing
– Solid plungers fall slower because of increased in more detail the signals generated by the plunger as
friction. it falls past the tubing collars. When the pulses gener-
• Gas pressure or density of gas ated at the tubing couplings are clearly identified in
– High pressure: Plunger fall is slow. the acoustic record, the acoustic velocity in the tubing
– Low pressure: Plunger fall is fast. gas can be computed by observing the round trip travel
• Liquid/gas density time of repeat echoes generated at the tubing head. The
– Increased density: Plunger falls slowly. acoustic pulse generated when the plunger goes past a
• Pressure: High pressure also causes the plunger collar propagates, via the gas in the tubing above the
to fall more slowly through liquid. plunger, to the wellhead.
• Surfactant: Surfactant lightens the gradient and There it is detected by the microphone and then is
the plunger falls faster. reflected back down through the gas until it reaches the
top of the plunger and is reflected back to the surface,
Determining Gas Properties where the microphone detects it as a first echo. Since the
The specific gravity of the produced gas must be deter- acoustic velocity in the gas is of the order of 900 to 1,400
mined to accurately calculate the pressure distribution ft/s and the plunger velocity is much slower, of the order
in the tubing and casing and to undertake calculations of of 200 to 1,000 ft/min (3.3 to 16 ft/s), the plunger has
volumetric rates of gas influx from the reservoir during not fallen a significant distance by the time the acoustic
the plunger cycle. The properties of gas sampled at the pulse catches up with the plunger and is reflected to the
sales line, or even at the separator, generally are not surface. The repeat reflection process can be observed
representative of the properties of the gas within the several times before the plunger reaches the next collar.

In figure 11.14, collar C56’s echo occurs at 9.837 DETECTING AND TROUBLESHOOTING
minutes and its repeat echo (marked with the vertical OPERATION PROBLEMS
line E56) occurs at 9.877 minutes. The depth to collar
In addition to determining the performance of plunger
C56 is 1,513 feet, and the repeat echo travels from the
wells, acoustic surveys provide tools to determine the
microphone to the top of the plunger and back to the condition of the tubing, casing, perforations, and other
microphone in 0.04 minutes. For this example, the downhole features without needing to introduce wireline
acoustic velocity is calculated to be 1,252 ft/s, and the sensors into the tubing. The measured tubing and cas-
corresponding gas specific gravity is calculated to be ing pressure trends when combined with the features
0.746. observed in the acoustic fluid level record, as discussed
In wells that exhibit low-amplitude collar echoes in detail in chapter 3, provide a very powerful tool to
in the casing annulus or poor collar pulses in the tubing identify the presence of known or unknown anomalies
because of excessive noise in the acoustic records, the present in the wellbore. Of particular interest are the
gas gravity should be obtained by direct measurement presence of holes and obstructions in the tubing that
on a gas sample taken at the wellhead. affect the normal cycling of the plunger.
Determine Gas SG and Acoustic Velocity

Raw Data Select Cycle Cycle Limits Plunger Fall Gas Properties Cycle Analysis Plots
Determine Gas Specific Gravity by Selecting Echo Repeats 1,256.26 ft/sec

|< --- <<<
C55 C56 E56 C57 C58 C59 C59
ft/sec <-
<-
Add Delete Cursor
E56 1,252.16
E59 1,260.36
Enter Gas Specific Gravity, Compositional Analysis, or Acoustic Velocity ft/sec
Gas Gravity Determination Dialog...
ft/sec
Retrieve Gas Gravity from Previous Acoustic Set
Retrieve Data Set...
Result
Gas Specific Gravity 0.746314 Air = 1
Reset Gas Gravity ? < Pg Up Pg Dwn >
Figure 11.14 Determining the acoustic velocity from repeat echoes of the collar pulses

Figure 11.15 shows a common problem in plunger falling plunger, and the increase in pressure below the
lift wells that produce small amounts of liquid and gas. plunger forced gas out the hole into the casing. When
The small difference between the tubing and casing the plunger fell past the hole, the higher gas pressure
pressure at the start of the plunger fall indicates that from the casing equalized back into the tubing, thereby
there is little or no liquid accumulating in the tubing. equalizing the pressure at the hole. For this type of tubing
In addition, as the tubing pressure increases during the pressure increase to occur after the plunger falls past the
shut-in period, the liquid in the tubing is pushed out so hole, the tubing must have some liquid in the bottom,
that the plunger arrives at the bottom of dry tubing. For trapping the gas between the plunger and the hole.
these low producing wells, a standing valve is needed Figure 11.17 shows the plunger suddenly stopping
to hold the liquid in the tubing. as it was falling to bottom during the shut-in period of
Figure 11.16 shows how the tubing pressure behaves the cycle. In this example, the plunger had fallen 1,866
when the plunger falls past a hole in the tubing, where feet from the surface, and the sharp 3-psi increase in the
the hole is above the liquid in the bottom of the tubing. tubing pressure at this depth indicates that the tubing
In this example, the plunger falls past the hole at a depth pressure is no longer reduced because of the pressure
of 1,800 feet from the surface; the increase in tubing drop across the moving plunger. The operator, who was
pressure easily identifies the depth to the hole. When monitoring the plunger fall in real time, quickly identi-
the plunger began the fall, the gas pressure in the tubing fied that the plunger was stuck and began corrective
above the plunger was depressed by the weight of the measures to free the plunger.
Raw Data Select Cycle Cycle Limits Plunger Fall Gas Property Cycle Analysis Plots
160.00 0.30
TUBING PRESSURE INCREASE
AND CASING PRESSURE DROPS
WHEN PLUNGER ON BOTTOM 0.20
150.00
0.10
0
140.00
-0.10
130.00
-0.20
-0.30
120.00
3 psi DROP WHEN
RELEASED FROM CATCHER -0.40
110.00 -0.50
0.000 6.944 13.889 20.833 27.778 34.722
X-Axis Range 39 mins Full Trace Options
Figure 11.15 Acoustic and pressure record in a plunger well with dry tubing

[B] [3]
240.5
240.0
239.5
3 psi DROP WHEN
239.0 RELEASED FROM CATCHER
238.5
238.0
237.5
237.0 HOLE IN TUBING
236.5
1.178 3.956 6.733 9.511 12.289
Figure 11.16 Tubing head pressure increase observed after the plunger drops below a hole in the tubing
Figure 11.18 shows how the plunger fall velocity is between the plunger and the liquid in the tubing result
nearly constant at 215 ft/min, as the plunger falls from in the plunger slowing down to a fall velocity of 200 ft/
the surface to the hole. Once the plunger falls past the min. Located in the 156th tubing joint from the surface,
hole at 5,054 feet from the surface, then the trapped gas the J-inch diameter hole in the tubing was causing the
420 0.2
410 0.0
400 -0.2
TUBING PRESSURE INCREASES
WHEN PLUNGER STICKS
390 -0.4
380 -0.6
TUBING PRESSURE DROPS

WHEN PLUNGER STARTS TO FALL
370 -0.8
SHUT-IN BEGINS
360 -1.0
0.716 1.271 1.827 2.382 2.938
Figure 11.17 Graph showing recorded pressure changes corresponding to the plunger falling or stopping

Plunger Fall Velocity Slows Past Hole
Collar # C156
Depth to Plunger 5,054.40 ft Plunger Fall Velocity -221.75 ft/min Elapsed Time 23.613 mins
-190 Plunger Slowed from 215 ft/min 0

After Falling Past Hole @ 5,050 ft
-195
1,600
-200
-205
3,200
-210
4,800
-215
-220
6,400
-225
-230 Bottom of Tubing—7,778.00 ft 8,000

0.000 6.944 13.889 20.833 27.778 34.722 41.667
ELAPSED TIME, mins
Figure 11.18 Record showing the effect of a hole in the tubing on the plunger fall velocity
plunger to not arrive at the surface during the unloading annotated with the unloading at point [A] through the
period of the cycle many times per day. end of the flow period at [B] and continuing to the end
The operator quickly identified the hole and elimi- of the shut-in period at [C]. Liquid arrives at the tubing
nated the need to drop a standing valve and to pressure- head 12 minutes after the start of the unloading followed
test the tubing before it was pulled. by the plunger 2 minutes later. This corresponds to an
average plunger rise velocity of 580 ft/min. The after-
Field Example flow period lasted 70 minutes, and the shut-in period
A complete plunger cycle was acquired starting a few lasted 68 minutes, during which the plunger fell to the
minutes before the controller opened the flow-line valve liquid level after 30 minutes.
until about the same time of the next shut-in phase. In this well, the plunger fall velocity is fairly con-
Figure 11.19 shows the recorded data for the cycle, stant and slightly less than 300 ft/min. Accordingly, the

Raw Data Select Cycle Cycle Limits Plunger Fall Gas Properties Cycle Analysis Plots
[A] [B] [C]

300.0
250.0
200.0
150.0
100.0
50.0
0.000 27.778 55.556 83.333 111.111 138.889
X-Axis Range 163.923 mins Full Trace Options...
Define Plunger Lift Cycle Using Following Markers

Marker [A]: Valve Opens [Unloading Begins] 6.091 mins
Marker [B]: Valve Closes [Shut In Begins] 94.884 mins
-––> Marker [C]: Valve Opens [Unloading Begins] 162.891 mins
Current Marker Values Annotations

Tubing Pressure 208.2 psig Show Annotations in Graph
Casing Pressure 264.3 psig
-x-
Acoustic -0.000303
Apply to Marker Options...
<-
<-
Reset Current Cycle ? < Pg Up Pg Dwn >
Figure 11.19 Complete recording of pressures and acoustic data during an entire plunger cycle
plunger position versus time exhibits an almost linear SUMMARY

relation, as shown in figure 11.20. At the bottom of the
Using conventional computer-based acoustic fluid level
tubing, there is a gaseous liquid column consisting of
instruments and software, monitoring the complete
61% liquid and extending over 1,117 feet.
Based on these results, the shut-in period was plunger cycle becomes a fairly simple task. By ac-
decreased in several steps from 68 to 33 minutes. This curately measuring the plunger fall velocity and depth
decrease resulted in an increase in gas production from to the liquid level, the minimum shut-in time for the
168 to 241 Mscf/day. However, it also caused an in- plunger lift installation can be determined. The plunger
crease in the plunger rise velocity to close to 1,000 ft/ fall measurements ensure that the plunger will reach the
min, which is excessive. bottom of the tubing by the end of the shut-in period.

Velocity and Depth Graph
0
0
-100
2,000
-200 4,000
-300 6,000
-400 8,000
Bottom of Tubing—8,686.97 ft
-500 10,000
0.000 6.944 13.889 20.833 27.778 34.722 41.667
ELAPSED TIME, mins
Figure 11.20 Graph depicting the plunger fall velocity and position for the well in the field example
Maximum production from the plunger lift installation optimization of plunger lift production achievable with
will be obtained by having the shortest possible shut- a minimum of effort and avoids the usual waste of time
in time. due to trial-and-error procedures.
The operator will save time by quickly detecting
the presence of holes and eliminating the need to drop
REFERENCES
a standing valve and pressure-test the tubing before it
is pulled. Detailed monitoring of the plunger behavior 1. D. Phillips and S. Listiak, “How to Optimize Produc-
increases the safety of operations by allowing the operator tion from Plunger Lift Systems,” World Oil, April
to know where the plunger is in the tubing. If a plunger 1998.
is not going to the bottom and the well is pressured up, 2. D. J. Becker, O. L. Rowlan, J. N. McCoy, and A.
then the plunger will surface-dry at a very high veloc- L. Podio, “Plunger Lift Optimization by Monitor-
ity, which can cause equipment damage and exceed the ing and Analyzing Wellbore Acoustic Signals and
mechanical integrity limits of the lubricator. Having a Tubing and Casing Pressures,” Proceedings of the
detailed analysis of the operation of the well makes SPE Eastern Regional Meeting, 2006.

3. R. Nadkrynechny, O. L. Rowlan, C. Cepuch, J. F. 6. R. Nadkrynechny, C. Cepuch, and O. L. Row-

Lea, and J. N. McCoy, “Measured Plunger Fall Ve- lan, “Test Results from Plunger Well Simulator,”
locity Used to Calibrate New Fall Velocity Model,” ALRDC Gas Well Deliquification Workshop, 2008.
SPE Production and Operations Symposium, 2013
4. O. L. Rowlan, J. N. McCoy, and A. L. Podio, BIBLIOGRAPHY
“Determining How Different Plunger Manufacture
Features Affect Plunger Fall Velocity,” SPE Produc- O. L. Rowlan, J. F. Lea, and J. N. McCoy “Modified Foss
tion and Operations Symposium, 2003. and Gaul Model Accurately Predicts Plunger Rise Veloc-
ity,” SPE Production and Operations Symposium, 2009.
5. J. F. Lea, O. L. Rowlan, and J. N. McCoy, “Mea-
surement and Calculation of Key Events During O. L. Rowlan, J. N. McCoy, and A. L. Podio, “Analyzing
the Plunger Lift Cycle,” SPE Annual Technical and Troubleshooting Plunger-Lifted Wells,” SPE Annual
Conference and Exhibition, 2007. Technical Conference and Exhibition, 2006.


Appendix A-1
Figure Appendix: Figure Credits

Owner Web Site
Figure Owner Web Site
1.1 Typical oil lifting cost as a function of artificial The University of Texas at petex.utexas.edu
lift efficiency and produced water cut Austin, PETEX
1.2 Calculation of overall pumping efficiency from The University of Texas at petex.utexas.edu
the pressure distribution in the annulus and tubing Austin, PETEX
1.3 Detailed analysis of the pumping well The University of Texas at petex.utexas.edu
performance from the acoustic fluid level survey Austin, PETEX
1.4 Acoustic fluid level height during a liquid level The University of Texas at petex.utexas.edu
depression test in an ESP well outfitted with a Austin, PETEX
bottomhole pressure sensor
1.5 Data input and output form for the well The University of Texas at petex.utexas.edu
performance analysis Austin, PETEX
1.6 Schematic of the fluid and pressure distribution The University of Texas at petex.utexas.edu
in a pumping well Austin, PETEX
1.7 Inflow performance based on the productivity The University of Texas at petex.utexas.edu
index determined from a two-rate flow test Austin, PETEX
1.8 Schematic representation of gas saturation in The University of Texas at petex.utexas.edu
an oil reservoir at a pressure lower than the Austin, PETEX
bubble point pressure
1.9 Producing bottomhole pressure is 200 psi for The University of Texas at petex.utexas.edu
liquid and gas flow from perforations. Austin, PETEX
1.10 Inflow performance relations for single-phase The University of Texas at petex.utexas.edu
(blue) and two-phase (red) flow in reservoir Austin, PETEX
1.11 Graphical representation of a multi-rate flow The University of Texas at petex.utexas.edu
test for determining the well inflow performance Austin, PETEX
1.12 Vogel’s dimensionless curves for different stages The University of Texas at petex.utexas.edu
of reservoir depletion and best fit reference curve Austin, PETEX
1.13 Schematic comparison of IPR from a multi-rate The University of Texas at petex.utexas.edu
test and the corresponding Vogel relation based Austin, PETEX
on single well test flow rate q2
1.14 Comparison of IPR relations from Gallice and The University of Texas at petex.utexas.edu
Wiggins Austin, PETEX
1.15 Acoustic fluid level and casing pressure acquired The University of Texas at petex.utexas.edu
during a multi-rate test in a well pumped with a Austin, PETEX
variable-speed ESP
1.16 Variation of PBHP versus time during the The University of Texas at petex.utexas.edu
variable rate test shown in figure 1.15 Austin, PETEX
1.17 Inflow performance from a three-rate test The University of Texas at petex.utexas.edu
Austin, PETEX
2.1 Example acoustic record with unexplained The University of Texas at petex.utexas.edu
signals Austin, PETEX
2.2 Vertical well with a rod pump set at 5,115 feet. The University of Texas at petex.utexas.edu
Average production is 27 bbl/day of oil, 60 bbl/day Austin, PETEX
of water, and 40 Mscf/day of gas.
2.3 Acoustic record with typical echoes from the The University of Texas at petex.utexas.edu
liquid level and tubing collars Austin, PETEX
2.4 Detailed analysis of echoes from tubing collars The University of Texas at petex.utexas.edu
recorded between 0.5 and 1.5 seconds from pulse Austin, PETEX
generation

A-1
A-2 ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
2.5 Summary report including calculation of downhole The University of Texas at petex.utexas.edu
pressures and production potential Austin, PETEX
2.6 Wellbore trajectory and pump location The University of Texas at petex.utexas.edu
Austin, PETEX
2.7 Comparison of raw and filtered acoustic record The University of Texas at petex.utexas.edu
Austin, PETEX
2.8 Overlay of records acquired at the start and the The University of Texas at petex.utexas.edu
end of pump-down Austin, PETEX
2.9 Summary acoustic fluid level report with pressure The University of Texas at petex.utexas.edu
distribution and potential analysis Austin, PETEX
2.10 Acoustic record showing multiple echoes The University of Texas at petex.utexas.edu
Austin, PETEX
2.11 Detailed analysis of an acoustic record using the The University of Texas at petex.utexas.edu
depth to the downhole marker at the echo where Austin, PETEX
the tubing tapers from 4H to 3H inches
2.12 Summary acoustic fluid level report, including The University of Texas at petex.utexas.edu
the calculation of downhole pressures and Austin, PETEX
production potential
2.13 Well with multiple perforated intervals The University of Texas at petex.utexas.edu
Austin, PETEX
2.14 Acoustic record with echoes from the blast joint The University of Texas at petex.utexas.edu
and perforations Austin, PETEX
2.15 Detail of an acoustic record showing (A) the The University of Texas at petex.utexas.edu
repeat echo from the top perforations and absence Austin, PETEX
of an identifiable liquid level echo and (B) the
liquid level marker located at pump intake depth,
based on dynamometer determination of
pumped-off condition
2.16 Well completion schematic The University of Texas at petex.utexas.edu
Austin, PETEX
2.17 Wellbore trajectory, ESP location, and producing The University of Texas at petex.utexas.edu
fluid level Austin, PETEX
2.18 Noisy record due to a resonating cavity or cable The University of Texas at petex.utexas.edu
banding Austin, PETEX
2.19 Record in Figure 2.18 filtered with a low-pass The University of Texas at petex.utexas.edu
filter Austin, PETEX
2.20 Acoustic velocity determination using echoes The University of Texas at petex.utexas.edu
from tubing collars after filtering the raw data Austin, PETEX
2.21 Producing fluid level analysis and pressure The University of Texas at petex.utexas.edu
distribution summary Austin, PETEX
2.22 Acoustic fluid level record for a shut-in well The University of Texas at petex.utexas.edu
Austin, PETEX
2.23 Summary report with producing and static fluid The University of Texas at petex.utexas.edu
level analyses Austin, PETEX
2.24 Wellbore schematic and trajectory for Example 6 The University of Texas at petex.utexas.edu
well with hole in tubing. Austin, PETEX
2.25 First acoustic record acquired through the tubing The University of Texas at petex.utexas.edu
string Austin, PETEX

Appendix A-3
2.26 Overlay of three acoustic records acquired in The University of Texas at petex.utexas.edu
quick succession Austin, PETEX
2.27 Calculation of acoustic velocity using the known The University of Texas at petex.utexas.edu
distance to the echo from the tubing taper Austin, PETEX
2.28 Determining the depth to the hole in the tubing The University of Texas at petex.utexas.edu
Austin, PETEX
2.29 Identification of repeat echoes from multiple The University of Texas at petex.utexas.edu
reflectors inside the tubing Austin, PETEX
2.30 Wellbore diagram of a flowing gas well with a The University of Texas at petex.utexas.edu
downhole safety valve Austin, PETEX
2.31 Superposition of acoustic records acquired with The University of Texas at petex.utexas.edu
an open (black) and a closed (blue) properly Austin, PETEX
operating safety valve
2.32 Comparison of acoustic traces for a The University of Texas at petex.utexas.edu
malfunctioning subsurface safety valve Austin, PETEX
2.33 Wellbore diagram for a well undergoing a casing The University of Texas at petex.utexas.edu
integrity test Austin, PETEX
2.34 Acoustic record acquired in the intermediate casing The University of Texas at petex.utexas.edu
annulus and processed with a low-pass filter Austin, PETEX
2.35 Acoustic record acquired in the intermediate The University of Texas at petex.utexas.edu
casing annulus, showing echoes from the Austin, PETEX
couplings
2.36 Determining the acoustic velocity of nitrogen gas The University of Texas at petex.utexas.edu
in a wellbore using echoes from the casing collars Austin, PETEX
2.37 Summary report for a casing integrity test The University of Texas at petex.utexas.edu
Austin, PETEX
2.38 Well completion schematic and simplified The University of Texas at petex.utexas.edu
directional survey for Example 9 well with Austin, PETEX
stratified gas column
2.39 Acoustic record acquired in a deep rod-pumped The University of Texas at petex.utexas.edu
horizontal well Austin, PETEX
2.40 Dynamometer diagrams showing the range of The University of Texas at petex.utexas.edu
pump liquid fillage during 63 strokes Austin, PETEX
2.41 Detailed results of the tubing joints count The University of Texas at petex.utexas.edu
Austin, PETEX
2.42 Variation of acoustic velocity as a function of The University of Texas at petex.utexas.edu
the RTTT Austin, PETEX
2.43 Determining average acoustic velocity from the The University of Texas at petex.utexas.edu
known depth of the end of the tubing Austin, PETEX
2.44 Summary report for a horizontal well containing The University of Texas at petex.utexas.edu
a stratified gas column Austin, PETEX

3.1 Acoustic pulse amplitude versus time, recorded The University of Texas at petex.utexas.edu
at the gas gun where z = 0 Austin, PETEX
3.2 Propagation of an acoustic pulse, as observed at The University of Texas at petex.utexas.edu
different times by three recorders located at Austin, PETEX
different depths
3.3 Construction of the waveform observed at The University of Texas at petex.utexas.edu
position Z1, applying the characteristic diagram Austin, PETEX
to the waveform generated at Z = 0

3.4 Observed amplitude versus time record at The University of Texas at petex.utexas.edu
point Z1 Austin, PETEX
3.5 Reflection and transmission at a medium The University of Texas at petex.utexas.edu
discontinuity Austin, PETEX
3.6 Wellbore schematic, characteristic diagram, and The University of Texas at petex.utexas.edu
synthetic acoustic record for a well with a gas/ Austin, PETEX
liquid interface at 3,000 feet
3.7 Acoustic fluid level record with multiple echoes The University of Texas at petex.utexas.edu
and an acoustic velocity of 1,129 ft/s Austin, PETEX
3.8 Reflection and transmission at an area The University of Texas at petex.utexas.edu
discontinuity Austin, PETEX
3.9 Schematic wellbore diagram, characteristic The University of Texas at petex.utexas.edu
diagram, and synthetic acoustic record for the Austin, PETEX
liquid level below the casing liner top.
3.10 Echoes from tubing couplings, a tubing crossover, The University of Texas at petex.utexas.edu
and the fluid level Austin, PETEX
3.11 Schematic PVT diagram for hydrocarbon fluids The University of Texas at petex.utexas.edu
Austin, PETEX
3.12 Acoustic velocity in a hydrocarbon gas with a The University of Texas at petex.utexas.edu
specific gravity of 0.6 as a function of pressure Austin, PETEX
and temperature
3.13 Sonic velocity in a hydrocarbon gas with a The University of Texas at petex.utexas.edu
specific gravity of 1.2 Austin, PETEX
3.14 Pulse amplitude attenuation and spreading in The University of Texas at petex.utexas.edu
wellbore gas at 106 psi Austin, PETEX
3.15 Exponential amplitude decay of an acoustic pulse The University of Texas at petex.utexas.edu
propagating inside coiled tubing Austin, PETEX
4.1 Pulse generation using a manually actuated gas The University of Texas at petex.utexas.edu
gun with a quick-opening valve Austin, PETEX
4.2 Typical manually operated gas gun The University of Texas at petex.utexas.edu
Austin, PETEX
4.3 Simplified schematic of the gas gun in figure 4.2 The University of Texas at petex.utexas.edu
Austin, PETEX
4.4 Comparison of records from explosion and The University of Texas at petex.utexas.edu
implosion pulses acquired in the same wellbore. Austin, PETEX
4.5 Wireless remote control gas gun The University of Texas at petex.utexas.edu
Austin, PETEX
4.6 Remotely controlled gas gun operation. The University of Texas at petex.utexas.edu
Austin, PETEX
4.7 Typical laptop-based fluid level data acquisition, The University of Texas at petex.utexas.edu
processing, and analysis systems Austin, PETEX
4.8 Strip chart recording of an acoustic trace Dr. A. L. Podio
4.9 Hazardous areas classification and extension for a The University of Texas at petex.utexas.edu
rod-pumping wellhead Austin, PETEX
4.10 Certified intrinsically safe fluid level recorder and Dr. A. L. Podio
gas gun
4.11 Connections to a well The University of Texas at petex.utexas.edu
Austin, PETEX

Appendix A-5
4.12 Record showing decaying acoustic resonance The University of Texas at petex.utexas.edu
in connection piping Austin, PETEX
4.13 Effect of chamber pressure on the amplitude of The University of Texas at petex.utexas.edu
the liquid level echo. Austin, PETEX
4.14 Records showing random signals The University of Texas at petex.utexas.edu
Austin, PETEX
4.15 Noise generated by pumping action The University of Texas at petex.utexas.edu
Austin, PETEX
4.16 Acoustic record acquired in a well produced by The University of Texas at petex.utexas.edu
gas lift Austin, PETEX
5.1 Acoustic record showing decay of the amplitude The University of Texas at petex.utexas.edu
of echoes from tubing collars. Austin, PETEX
5.2 Digital processing of an acoustic record to detect The University of Texas at petex.utexas.edu
echoes from tubing collars Austin, PETEX
5.3 Identification of collar echoes in various segments The University of Texas at petex.utexas.edu
of the acoustic record from figure 5.1, performed Austin, PETEX
visually and manually by the analyst
5.4 Variation of collar echo frequency as a function The University of Texas at petex.utexas.edu
of depth Austin, PETEX
5.5 Detailed collar count analysis of the record in The University of Texas at petex.utexas.edu
figure 5.1 Austin, PETEX
5.6 Downhole marker analysis The University of Texas at petex.utexas.edu
Austin, PETEX
5.7 Effect of temperature on the average acoustic The University of Texas at petex.utexas.edu
velocity of gas (with a 0.6 gravity) in a pumping Austin, PETEX
well
5.8 Temperature survey in a pumping well The University of Texas at petex.utexas.edu
Austin, PETEX
5.9 Portable instruments for gas gravity (left) or gas AMETEK, Inc. www.ametek.com/
analysis (right)
5.10 Isothermal gradient map. Society of Petroleum Engineers www.spe.org/unitedstates/
5.11 Variation of the acoustic velocity of the annular The University of Texas at petex.utexas.edu
gas in a given well over three years Austin, PETEX
6.1 Pressure distribution in a pumping well The University of Texas at petex.utexas.edu

Austin, PETEX
6.2 Fluid distribution in a stabilized pumping well The University of Texas at petex.utexas.edu
Austin, PETEX
6.3 Classification of pumping wells by tubing depth The University of Texas at petex.utexas.edu
and fluid distribution Austin, PETEX
6.4 Generalized wellbore configuration for tapered The University of Texas at petex.utexas.edu
tubing and casing Austin, PETEX
6.5 Calculating the gaseous column gradient from a The University of Texas at petex.utexas.edu
liquid level depression test. Austin, PETEX
6.6 Wellhead arrangement for performing a liquid The University of Texas at petex.utexas.edu
level depression test Austin, PETEX
6.7 Typical fluid level and casing pressure recorded The University of Texas at petex.utexas.edu
versus time during the Walker test Austin, PETEX

6.8 Height of the gaseous liquid column versus The University of Texas at petex.utexas.edu
pressure at the gas/liquid interface during a Austin, PETEX
liquid depression test
6.9 Measured downhole pressure and gaseous liquid The University of Texas at petex.utexas.edu
column height for varying annular pressures Austin, PETEX
6.10 Estimate of the gaseous column gradient from The University of Texas at petex.utexas.edu
a liquid level depression test Austin, PETEX
6.11 Partial sequence of acoustic fluid level records The University of Texas at petex.utexas.edu
for tests analyzed in figure 6.12 Austin, PETEX
6.12 Analysis of the liquid level depression test The University of Texas at petex.utexas.edu
Austin, PETEX
6.13 Effective oil fraction correlation from liquid The University of Texas at petex.utexas.edu
level depression tests Austin, PETEX
6.14 Comparison of the effective oil fraction The University of Texas at petex.utexas.edu
determined from liquid level depression tests Austin, PETEX
in Venezuelan heavy oil (10 to 11° API) wells
and western Texas (32 to 43° API) wells
6.15 Annular pressure increase during the acoustic The University of Texas at petex.utexas.edu
fluid level acquisition used to estimate the gas Austin, PETEX
inflow rate
6.16 Pressure balance and fluid distribution in a The University of Texas at petex.utexas.edu
static well Austin, PETEX
6.17 Oil/water distribution for static well conditions The University of Texas at petex.utexas.edu
Austin, PETEX
7.1 Acoustic velocity variation during a seven-day The University of Texas at petex.utexas.edu
pressure transient test Austin, PETEX
7.2 Variation of casing pressure (circles) and The University of Texas at petex.utexas.edu
wellhead temperature (triangles) during the Austin, PETEX
transient test of figure 7.1
7.3 (A) Pressure transient fluid level acquisition setup Dr. A. L. Podio
using wired instruments. (B) Pressure transient
fluid level acquisition setup using wireless
standalone instrumentation.
7.4 Short-duration liquid level depression test The University of Texas at petex.utexas.edu
showing the fluid level drop by 756 feet in Austin, PETEX
30 minutes
7.5 Dynamometer record acquired prior to initiating The University of Texas at petex.utexas.edu
the pressure buildup test, showing pump Austin, PETEX
displacement of 110 bbl/day
7.6 Example of erratic pump operation caused by a The University of Texas at petex.utexas.edu
damaged traveling valve Austin, PETEX
7.7 Thermal insulator placed on a pressure sensor Dr. A. L. Podio
attached to a gas gun
7.8 Example record indicating the correct selection The University of Texas at petex.utexas.edu
of the liquid level echo in the presence of a repeat Austin, PETEX
echo.
7.9 Examples of wellhead pressure variations The University of Texas at petex.utexas.edu
observed in three different wells Austin, PETEX

Appendix A-7
7.10 Example of multiple sequential acoustic records The University of Texas at petex.utexas.edu
with wellbore perforations. Austin, PETEX
7.11 Long-term pressure buildup test showing casing The University of Texas at petex.utexas.edu
pressure (circles) and casinghead pressure Austin, PETEX
transducer temperature (triangles) as a function
of elapsed time
7.12 Depth to liquid level (triangles) and casinghead The University of Texas at petex.utexas.edu
pressure (circles) versus time Austin, PETEX
7.13 Computed BHP (triangles) and measured The University of Texas at petex.utexas.edu
casinghead pressure (circles) versus time Austin, PETEX
7.14 Resulting log-log plot of delta pressure with The University of Texas at petex.utexas.edu
derivative versus delta time in hours Austin, PETEX
7.15 Horner plot yields skin of 0.8 and P* of The University of Texas at petex.utexas.edu
2,018 psi. Austin, PETEX
7.16 Liquid level (triangles) rises 2,600 feet and The University of Texas at petex.utexas.edu
casing pressure (circles) increases by 378 psi Austin, PETEX
during the 4H-day test.
7.17 BHP (circles) levels off and liquid after-flow The University of Texas at petex.utexas.edu
(triangles) tends to cease after about 48 hours. Austin, PETEX
7.18 Resulting log-log plot of delta pressure with The University of Texas at petex.utexas.edu
derivative (triangles) versus delta time in hours Austin, PETEX
7.19 Horner plot shows skin of 8.9 and P* of The University of Texas at petex.utexas.edu
1,306 psi. Austin, PETEX
7.20 Variation of casing pressure and transducer The University of Texas at petex.utexas.edu
temperature (triangles) during a 25-day and Austin, PETEX
17-hour buildup test
7.21 Computed BHP (circles) increases as RTTT time The University of Texas at petex.utexas.edu
to liquid decreases when liquid rises Austin, PETEX
7.22 Wellbore storage just beginning to be overcome The University of Texas at petex.utexas.edu
and radial flow period starting in the well. Austin, PETEX
7.23 Casing pressure (circles) and BHP (triangles) The University of Texas at petex.utexas.edu
during 3 days and 13 hours Austin, PETEX
7.24 Log-log plot shows boundary effect after radial The University of Texas at petex.utexas.edu
flow period, as shown by increasing derivative Austin, PETEX
(triangles).
7.25 Horner plot yields a skin of 1.7 and P* of 102 The University of Texas at petex.utexas.edu
psia. Austin, PETEX
7.26 Liquid level (triangles) drops as casing pressure The University of Texas at petex.utexas.edu
(circles) increases. Austin, PETEX
7.27 Casing pressure (circles) and BHP (triangles) The University of Texas at petex.utexas.edu
versus time Austin, PETEX
7.28 Log-log plot shows beginning of radial flow The University of Texas at petex.utexas.edu
period, indicated by stabilization of derivative Austin, PETEX
(triangles).
7.29 Horner plot seems to indicate the presence of The University of Texas at petex.utexas.edu
significant skin. Austin, PETEX

8.1 Example of an over-pumped well The University of Texas at petex.utexas.edu
Austin, PETEX

8.2 Example of gas interference of pump operation The University of Texas at petex.utexas.edu
Austin, PETEX
8.3 Example of incomplete liquid fillage due to The University of Texas at petex.utexas.edu
choked pump intake Austin, PETEX
8.4 Distribution of fluids and pressure versus depth The University of Texas at petex.utexas.edu
in a stabilized pumping well Austin, PETEX
8.5 Schematic representation of a well completion The University of Texas at petex.utexas.edu
exhibiting fluid gradient inversion due to the TAC Austin, PETEX
8.6 Dynamometer records in well exhibiting annular The University of Texas at petex.utexas.edu
gradient inversion Austin, PETEX
8.7 Casing pressure and liquid level depth as a The University of Texas at petex.utexas.edu
function of time, showing rapid drop in the fluid Austin, PETEX
level past the tubing anchor
8.8 Results of a liquid level depression test in a well The University of Texas at petex.utexas.edu
with the tubing anchor set above the perforations Austin, PETEX
8.9 Fluid and pressure distribution caused by the The University of Texas at petex.utexas.edu
presence of the tubing anchor Austin, PETEX
8.10 Frequency of rod couplings echoes in tubing The University of Texas at petex.utexas.edu
(19.46 jts/s at 25.0 ft/jt) Austin, PETEX
8.11 Example tubing record showing an echo from a The University of Texas at petex.utexas.edu
deep hole Austin, PETEX
8.12 Comparison of a hole-in-tubing echo overlay of The University of Texas at petex.utexas.edu
low-pass filtered casing shot to a raw tubing shot Austin, PETEX
8.13 (A) Acoustic traces acquired in the casing annulus The University of Texas at petex.utexas.edu
while the pump is stopped and while the pump is Austin, PETEX
operating, showing an inversion of polarity of
echo from the tubing leak. (B) Overlay of acoustic
records acquired when the pump was operating
and when the pump was stopped.
8.14 Tubing fill-up rate as a function of pump The University of Texas at petex.utexas.edu
displacement and SPM Austin, PETEX
8.15 Successive acoustic records taken during The University of Texas at petex.utexas.edu
32 minutes of pump operation Austin, PETEX
8.16 A PAP (grooved steel) plunger Dr. A. L. Podio
8.17 (A) Fluid level record acquired in an ESP. The University of Texas at petex.utexas.edu
(B) Detail of an acoustic signal showing Austin, PETEX
echoes from cable bands
8.18 Productivity analysis for an ESP well The University of Texas at petex.utexas.edu
Austin, PETEX
8.19 Typical annular fluid level record for the well in The University of Texas at petex.utexas.edu
8.20 Acoustic survey summary report displaying the The University of Texas at petex.utexas.edu
computed annular pressure distribution Austin, PETEX
8.21 Pressure versus depth traverse showing the The University of Texas at petex.utexas.edu
measured and computed pump intake pressure Austin, PETEX
for the well in figure 8.19
8.22 Analysis of pressure and flow distribution for a The University of Texas at petex.utexas.edu
well producing from three perforated intervals Austin, PETEX
8.23 Examples of fluid level records acquired in wells The University of Texas at petex.utexas.edu
producing with PC pumps Austin, PETEX

Appendix A-9
8.24 Recommendations for connecting a gas gun to a The University of Texas at petex.utexas.edu
wellhead Austin, PETEX
9.1 Example of a critical rate diagram for different The University of Texas at petex.utexas.edu
tubing sizes Austin, PETEX
9.2 Examples of gas gun connections to gas wells The University of Texas at petex.utexas.edu
Austin, PETEX
9.3 Acoustic records acquired during a gas well The University of Texas at petex.utexas.edu
shut-in Austin, PETEX
9.4 Position of the gas/liquid interface as a function The University of Texas at petex.utexas.edu
of time Austin, PETEX
9.5 Height of a gaseous liquid column as a function The University of Texas at petex.utexas.edu
of tubing pressure Austin, PETEX
9.6 Pressure traverse in a liquid loaded gas well The University of Texas at petex.utexas.edu
Austin, PETEX
9.7 Pressure versus depth traverses The University of Texas at petex.utexas.edu
Austin, PETEX
9.8 Gaseous column gradient changes in time The University of Texas at petex.utexas.edu
Austin, PETEX
9.9 Sequence of acoustic records The University of Texas at petex.utexas.edu
Austin, PETEX
9.10 Comparison of annular and tubular “S” curves The University of Texas at petex.utexas.edu
Austin, PETEX
9.11 Offshore gas wellbore schematic and The University of Texas at petex.utexas.edu
corresponding acoustic trace acquired down Austin, PETEX
the tubing after long shut-in time
9.12 Detail of the wellbore completion The University of Texas at petex.utexas.edu
Austin, PETEX
9.13 Acoustic record corresponding to the wellbore in The University of Texas at petex.utexas.edu
9.14 Detail of the echo from the 3H- to 4H-inch The University of Texas at petex.utexas.edu
crossover Austin, PETEX
9.15 Fluid and pressure distribution in the shut-in well The University of Texas at petex.utexas.edu
Austin, PETEX
9.16 Acoustic records acquired in the tubing in a The University of Texas at petex.utexas.edu
shut-in gas well with a properly operating SCSSV Austin, PETEX
9.17 Acoustic records acquired in the tubing of a The University of Texas at petex.utexas.edu
shut-in gas well with a SCSSV stuck open Austin, PETEX
9.18 Examples of identification of downhole features The University of Texas at petex.utexas.edu
from the polarity of the acoustic echoes Austin, PETEX
9.19 Hole in a gas well’s tubing string caused by The University of Texas at petex.utexas.edu
corrosion Austin, PETEX
9.20 Acoustic record acquired inside the tubing of a The University of Texas at petex.utexas.edu
gas well with a shallow corrosion hole Austin, PETEX
9.21 Characteristic wave path diagram corresponding The University of Texas at petex.utexas.edu
to the presence of a hole in the tubing above the Austin, PETEX
liquid level
9.22 Production history of a gas well with a hole in the The University of Texas at petex.utexas.edu
tubing Austin, PETEX

9.23 Annular fluid level in a liquid loaded gas well The University of Texas at petex.utexas.edu
operating at stabilized conditions Austin, PETEX
9.24 Pressure distribution in a gas well without a The University of Texas at petex.utexas.edu
packer Austin, PETEX
9.25 Sequence of acoustic records in a gas well with The University of Texas at petex.utexas.edu
a hole in the tubing Austin, PETEX
10.1 Typical continuous injection gas lift well The University of Texas at petex.utexas.edu
Austin, PETEX
10.2 Acoustic fluid level instrument connection to The University of Texas at petex.utexas.edu
the casing or tubing of a gas lift well. Austin, PETEX
10.3 Comparison of acoustic records for different The University of Texas at petex.utexas.edu
connection methods and gas guns Austin, PETEX
10.4 Acoustic record acquired while injection gas is The University of Texas at petex.utexas.edu
flowing (top) and acoustic record acquired after Austin, PETEX
stopping the flow of injection gas (bottom)
10.5 Repeat acoustic record (blue) that does not show The University of Texas at petex.utexas.edu
a random signal at about 10.4 seconds. Austin, PETEX
10.6 Chart depicting acoustic velocity as a function The University of Texas at petex.utexas.edu
of pressure and temperature for 0.6 gravity Austin, PETEX
hydrocarbon gas
10.7 Acoustic record acquired down the tubing of a The University of Texas at petex.utexas.edu
shut-in well, showing a mismatch of echoes and Austin, PETEX
gas lift mandrels due to varying acoustic velocity
with increasing depth
10.8 Mandrel design determines the polarity of the The University of Texas at petex.utexas.edu
echoes. Austin, PETEX
10.9 Complex acoustic record in a deep gas lift well The University of Texas at petex.utexas.edu
Austin, PETEX
10.10 Acoustic data showing a problem identifying The University of Texas at petex.utexas.edu
multiple echoes because of internal reflections Austin, PETEX
down the tubing
10.11 Depth profile of the acoustic velocity of injection The University of Texas at petex.utexas.edu
gas from acoustic records in figure 10.8 Austin, PETEX
10.12 Correlation function of acoustic velocity versus The University of Texas at petex.utexas.edu
the RTTT Austin, PETEX
10.13 Acoustic record analyzed automatically to The University of Texas at petex.utexas.edu
generate a depth scale Austin, PETEX
10.14 Comparison of records acquired in the tubing The University of Texas at petex.utexas.edu
and casing, showing an echo from the hole in Austin, PETEX
the tubing
10.15 Gas lift well configurations The University of Texas at petex.utexas.edu
Austin, PETEX
10.16 Casing survey: Liquid level at 5,602 feet above The University of Texas at petex.utexas.edu
fourth gas lift valve Austin, PETEX
10.17 Pressure traverses for the casing and tubing The University of Texas at petex.utexas.edu
during shut-in. Austin, PETEX
10.18 Acoustic records acquired down the tubing after The University of Texas at petex.utexas.edu
shut-in Austin, PETEX

Appendix A-11
10.19 Pressure distribution in the tubing and annulus The University of Texas at petex.utexas.edu
of a gas lift well Austin, PETEX
10.20 Typical side-pocket mandrels The University of Texas at petex.utexas.edu
Austin, PETEX
10.21 Typical conventional mandrels with external The University of Texas at petex.utexas.edu
valves Austin, PETEX
10.22 Determination of the injection depth (6,500 feet) The University of Texas at petex.utexas.edu
for the desired formation flow rate and operating Austin, PETEX
gas/liquid ratio. PBHP = 2,450 psi
10.23 Pressure in the tubing and casing when the kill The University of Texas at petex.utexas.edu
fluid reaches the surface by U-tubing through Austin, PETEX
the operating valve
10.24 Depth and pressure when the first valve was The University of Texas at petex.utexas.edu
unloaded Austin, PETEX
10.25 Unloading of the liquid to the flow line after the The University of Texas at petex.utexas.edu
first valve was uncovered. Austin, PETEX
10.26 Depth and pressure when the second valve was The University of Texas at petex.utexas.edu
uncovered and gas was injected into the tubing Austin, PETEX
10.27 Depicting the depth and pressure when the third The University of Texas at petex.utexas.edu
valve was uncovered Austin, PETEX
10.28 Depicting the depth and pressure when the The University of Texas at petex.utexas.edu
operating valve was uncovered and gas was
injected at 400 GLR Austin, PETEX
10.29 Portable pressure recorder monitoring the casing Dr. A. L. Podio
and tubing pressure during gas lift well unloading
10.30 Two-pen pressure-recorder chart illustrating The University of Texas at petex.utexas.edu
continuous-flow gas lift unloading operations Austin, PETEX
with choke control of the injection gas
10.31 Gas lift injection manifold Dr. A. L. Podio
11.1 Schematic diagram of a typical plunger lift The University of Texas at petex.utexas.edu
installation Austin, PETEX
11.2 Surface components of the plunger installation The University of Texas at petex.utexas.edu
Austin, PETEX
11.3 Various types of plungers The University of Texas at petex.utexas.edu
Austin, PETEX
11.4 Three phases of plunger lift: shut-in, unloading, The University of Texas at petex.utexas.edu
and after-flow Austin, PETEX
11.5 Typical installations of acoustic hardware used The University of Texas at petex.utexas.edu
for plunger lift monitoring Austin, PETEX
11.6 Chart depicting the depth of the plunger versus The University of Texas at petex.utexas.edu
time obtained from manually acquired acoustic Austin, PETEX
records
11.7 Record showing the expanded portion of an The University of Texas at petex.utexas.edu
acoustic signal and the tubing pressure versus Austin, PETEX
the time recorded during plunger fall
11.8 Acoustic recorder and pressure sensor installation The University of Texas at petex.utexas.edu
schematic Austin, PETEX

11.9 Example acoustic record showing pulses The University of Texas at petex.utexas.edu
generated by the plunger as it falls through Austin, PETEX
the gas and through the liquid at the bottom
of the tubing
11.10 Acoustic and pressure record for a complete The University of Texas at petex.utexas.edu
plunger cycle Austin, PETEX
11.11 One-minute section of an acoustic record of The University of Texas at petex.utexas.edu
pulses generated at the tubing couplings, and Austin, PETEX
the method for determining the plunger fall
velocity
11.12 Graph of acoustic pulses recorded digitally The University of Texas at petex.utexas.edu
during plunger fall Austin, PETEX
11.13 Record showing the plunger depth and velocity The University of Texas at petex.utexas.edu
as a function of time elapsed since the start of Austin, PETEX
plunger fall
11.14 Determining the acoustic velocity from repeat The University of Texas at petex.utexas.edu
echoes of the collar pulses Austin, PETEX
11.15 Acoustic and pressure record in a plunger well The University of Texas at petex.utexas.edu
with dry tubing Austin, PETEX
11.16 Tubing head pressure increase observed after The University of Texas at petex.utexas.edu
the plunger drops below a hole in the tubing Austin, PETEX
11.17 Graph showing recorded pressure changes The University of Texas at petex.utexas.edu
corresponding to the plunger falling or stopping Austin, PETEX
11.18 Record showing the effect of a hole in the tubing The University of Texas at petex.utexas.edu
on the plunger fall velocity Austin, PETEX
11.19 Complete recording of pressures and acoustic The University of Texas at petex.utexas.edu
data during an entire plunger cycle Austin, PETEX
11.20 Graph depicting the plunger fall velocity and The University of Texas at petex.utexas.edu
position for the well in the field example Austin, PETEX

Glossary G-1
Glossary
absolute pressure n: total pressure measured from an to water (if its API gravity is greater than 10, it is lighter
absolute vacuum. It equals the sum of the gauge pressure and floats on water; if it is less than 10, it is heavier and
and the atmospheric pressure and is expressed in pounds sinks in water). It can thus be considered the inverse
per square inch absolute (psia). of a petroleum liquid’s specific gravity and is used to
absorption coefficient n: characterizes the loss in energy compare densities of different petroleum liquids. See
as the wave propagates through a layer of material, which specific gravity.
is constant for the material for a given wave length. amplitude n: the difference in height or magnitude
acoustic fluid level instruments n pl: specialized equip- between the maximum displacement of a periodic wave
ment necessary for obtaining, organizing, and analyzing and the point of no displacement.
data to assist in visualizing the performance of oil and gas annular area n: 1. the space that surrounds a cylindrical
wells, typically consisting of a sound source to create an object within a cylinder. 2. the space around a pipe in a
acoustic pulse, a microphone to detect the pressure pulse, wellbore, the outer wall of which may be the wall of either
and signal processing equipment and sensors to record the borehole or the casing. Sometimes termed the annulus.
and interpret the acoustic signal. annular gaseous liquid column n: a multi-phase liquid
acoustic fluid level survey n: a well-logging method column generated by a mixture of free gas and liquids
in which sound pulses are generated at the surface and flowing into the wellbore from the formation. Some of
transmitted into a well. The time it takes for the sound the gas will enter the pump, and the remaining volume
pulses to travel down the well to the fluid level and for the of gas will flow to the surface and percolate through the
echoes to return to the surface is measured and recorded. annular liquid.
Subsequent analysis and interpretation of the record (log) annulus n: see annular area.
provides a tool to determine the pressure distribution and
fluid level in the well. API abbr: American Petroleum Institute.
acoustic velocity n: the rate at which a sound wave travels area n: the extent of a surface enclosed within a bound-
through a medium, calculated by distance divided by travel ary; the extent of the surface of all or part of a solid. For
time. two-dimensional plane surfaces, area is usually stated in
after-flow n: The flow associated with wellbore storage square units. For example, a rectangle 2 feet long on one
following a surface shut-in. When a well is first shut in at the side and 3 feet long on the other side has an area of 2 ×
surface, flow from the formation into the bottom of the 3 feet, which equals 6 square feet (ft2). Thus, the area of
the rectangle is 6 ft2.
wellbore continues unabated until compression of the
fluids in the wellbore causes the downhole pressure to rise. artificial lift n: any method used to raise liquid to the
If the wellbore fluid is highly compressible and the well surface through a well after reservoir pressure has declined
rate is low, the after-flow period can be long. Conversely, to the point of being insufficient to cause liquids to cause
high-rate wells producing little gas have negligible after- the liquid to flow to the surface.
flow periods. attenuation n: the loss of energy or amplitude of waves
American Petroleum Institute (API) n: an oil trade as they pass through media.
organization founded in 1920 that leads others in setting
standards for oilfield drilling and producing equipment. background noise n: acoustic noise already present at
Headquartered in Washington, D.C., it publishes materials the location of the microphone plus any noise generated
concerning exploration and production, petroleum mea- by the electronic equipment used to process and record
surement, marine transportation, marketing, pipelining, the acoustic signal.
refining, safety and fire protection, storage tanks, valves, back-pressure n: 1. the pressure maintained on equipment
training, health and environment, policy, and economic or systems through which a fluid flows. 2. in reference
studies. Contact: 1220 L Street NW, Washington, D.C., to engines, a term used to describe the resistance to the
20005; 202-682-8000; www.api.org. flow of the exhaust gas through the exhaust pipe. 3. the
American Petroleum Institute (API) gravity n: a mea- operating pressure level measured downstream from a
sure of how heavy or light a petroleum liquid is compared measuring or control device.

G-1
G-2 ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
ball valve n: a valve using a spherical closure element given temperature, causes it to consist only of liquid,
(ball), which is rotated through 90° to open and close rather than liquid and gas.
the valve. bushing n: 1. a pipe fitting on which the external thread
barrel (bbl) n: 1. a measure of volume for petroleum is larger than the internal thread to allow two pipes of
products in the U.S. One barrel is the equivalent of 42 different sizes to be connected. 2. a removable lining or
U.S. gallons or 0.15899 cubic metres (9,702 cubic inches). sleeve inserted or screwed into an opening to limit its size,
One cubic metre equals 6.2897 barrels. 2. the cylindrical resist wear or corrosion, or serve as a guide.
part of a sucker rod pump in which the pistonlike plunger
moves up and down. Operating as a piston inside a cylin-
carbon dioxide (CO2) n: a colorless, odorless gaseous
der, the plunger and barrel create pressure energy to lift
compound of carbon and oxygen. A product of combus-
well fluids to the surface.
tion and a filler for fire extinguishers, this heavier-than-air
bbl abbr: barrel.
gas can collect in low-lying areas, where it may displace
BHP abbr: bottomhole pressure. oxygen and present the hazard of anoxia. In contact with
blast joint n: a tubing sub made of abrasion-resistant steel, it may cause corrosion.
material. It is used in a tubing string where high-velocity
casing n: 1. steel pipe placed in an oil or gas well to pre-
flow through perforations may cause external erosion.
vent the wall of the hole from caving in, to prevent move-
bottomhole pressure (BHP) n: 1. the pressure at the ment of fluids from one formation to another, and to
bottom of a borehole. It is caused by the hydrostatic pres- improve the efficiency of extracting petroleum if the
sure of the wellbore fluid and, sometimes, by any back- well is productive. Most casing joints are manufactured
pressure held at the surface, as when the well is shut in to specifications by API, although non-API specification
with blowout preventers. When mud is being circulated, casing is available for special situations. Casing manu-
bottomhole pressure is the hydrostatic pressure plus the factured to API specifications is available in three range
remaining circulating pressure required to move the mud lengths. A joint of range 1 casing is 16 to 25 feet (4.8 to
up the annulus. 2. the pressure in a well a point opposite 7.6 metres) long; a joint of range 2 casing is 25 to 34 feet
the producing formation, as recorded by a pressure-tight (7.6 to 10.3 metres) long; and a joint of range 3 casing
container mechanism referred to as a bottomhole pressure is 34 to 48 feet (10.3 to 14.6 metres) long. The outsider
gauge or bottomhole pressure bomb. diameter of a joint of API casing ranges from 4H to 20
bottomhole pressure gauge n: an instrument placed in inches (114.3 to 508.0 millimetres). Casing is made of
a well to measure bottomhole pressure. The instrument many types of steel alloy, which vary in strength, corro-
can be either permanently installed in the well or run sion resistance, and so on. 2. large pipe in which a carrier
and retrieved on wireline. Also called bottomhole pres- pipeline is contained. Casing is used when a pipeline
sure bomb. passes under railroad rights-of-way and some roads to
brine n: water that contains a large quantity of salt, shield the pipeline from the unusually high load stresses
especially sodium chloride, dissolved in it; salt water. of a particular location. State and local regulations identify
British thermal unit (Btu) n: a measure of heat energy specific locations where casing is mandatory.
equivalent to the amount of heat needed to raise 1 pound of casinghead n: a heavy, flanged steel fitting connected to
water 1°F at or near its point of maximum density (39.1°F). the first string of casing. It provides a housing for slips
Equivalent to 0.252 kilogram-calories or 1,055 joules. and packing assemblies, allows suspension of intermediate
Btu abbr: British thermal unit. and production strings of casing, and provides the means
bubble flow n: a type of gas-liquid flow pattern in which for the annulus to be sealed off. See annular area.
the gas is distributed as small, dispersed bubbles through casinghead pressure n: the pressure in a well that exists
the liquid. between the casing and the tubing or the casing and the
bubble point n: the pressure and temperature conditions at drill pipe.
which the first bubble of gas comes out of solution in oil. casing pressure n: see casinghead pressure.
bubble point line n: the outside edge of the two-phase casing-tubing annulus n: in a wellbore, the space between
region of a PVT diagram at a pressure level where, at a the inside of the casing and the outside of the tubing.
given temperature, part of a liquid begins to convert to gas. casing valve n: the valve located at the casinghead that
bubble point pressure n: the amount of pressure that, connects the production casing annulus to the production
when applied to a given mixture of hydrocarbons at a flow line.

Glossary G-3
check valve n: a valve that permits fluid to flow in one kilograms per cubic metre. The density of fresh water is
direction only. If the gas or liquid starts to reverse, the 1 g/ml, 8.33 lbm/gal, or 62.4 lbm/ft3.
valve automatically closes, preventing reverse movement. dew point n: the temperature and pressure at which a
Sometimes referred to as a one-way valve. liquid begins to condense out of a gas. For example, if a
churn flow n: a type of gas-liquid flow pattern where gas constant pressure is held on a certain volume of gas but
and liquid are flowing upwards in a pipe as a turbulent the temperature is reduced, a point is reached at which
mixture. droplets of liquid condense out of the gas. That point is
CO2 form: carbon dioxide. the dew point temperature of the gas at that pressure.
Similarly, if a constant temperature is maintained on a
coal bed methane n: a form of natural gas, consisting
volume of gas but the pressure is increased, the point at
principally of methane, extracted from the fractures and
which liquid begins to condense is the dew point pressure
matrix of coal beds.
at that temperature.
collar echoes n: echoes generated by tubing joints in downhole gas separator n: a special assembly of pipes
the wellbore. that are attached to the intake of a downhole pump with
composition n: the combination of ingredients or com- the objective of minimizing the volume of gas that can
ponents in a mixture or substance. enter the pump.
compressibility n: the change in volume per unit of down-kick n: a negative signal and a downward deflec-
volume of a liquid caused by a unit charge in pressure at tion of the acoustic signal to the bottom of the screen or
a constant temperature. chart caused by an increase in pressure as detected by the
compression pulse n: a pressure pulse generated by acoustic fluid level instruments.
a gas gun into a wellbore when the pressure is greater drawdown n: 1. the difference between static and flowing
in the volume chamber than in the casing annulus. See bottomhole pressure. 2. the distance between the static
explosion pulse. level and the pumping level of a fluid in the annulus of
correlation coefficient n: a calculated value ranging a pumping well.
from -1 and +1 that represents the linear correlation or dynamometer n: 1. an instrument or assembly of instru-
dependence between two variables or data values. ments used to measure torque and other force-related
corrosion n: the gradual loss of a material, usually a metal, properties of rotating or reciprocating machinery. 2. in
due to chemical or electrochemical reactions, which can sucker rod pumping, a device used to indicate a variation
eventually destroy a structure. in load on the polished rod as the rod string reciprocates.
A continuous record of the result of forces acting along
critical point n: the point at which, in terms of tempera- the axis of the polished rod is provided on a dynamometer
ture and pressure, a fluid cannot be distinguished as being card, from which the performance of the well pumping
either a gas or liquid, i.e., the point at which the physical equipment is interpreted and analyzed.
properties of a liquid and gas are identical.
critical rate n: a rate of gas flow at which the liquid is echo n: the reflection of a sound wave traveling in a gas
uniformly distributed and the gas velocity is sufficient to that is generated when the wavefront encounters a dis-
continuously carry liquid as a fine mist or small droplets continuity, such as a change in the density of the medium
to the surface, establishing a relatively low and fairly or a change in the geometry of the space where the wave
uniform flowing pressure gradient throughout the tubing. is propagating.
electrical submersible pump (ESP) n: an electric
ρ abbr: density. centrifugal downhole pump used in oil production that
Darcy’s Law n: a law stating that the rate of flow of a is designed with vane and fin configurations to provide
fluid through a rock varies directly with the amount of adequate lift over a broad range of flow rates.
interconnected pore space (permeability) and applied equation of state n: a mathematical expression that
pressure and varies inversely with the viscosity, or flow defines the physical state of a homogeneous substance
resistance, of the fluid. by relating volume to pressure and absolute temperature
density (ρ) n: the mass of a substance per unit volume. for a given mass of the substance.
For instance, the density of a drilling mud may be 10 equivalent gas-free liquid height n: the height of the
pounds per gallon, 74.8 pounds per cubic foot, or 1,198.2 gaseous liquid column for the volume of free gas present.

ESP abbr: electrical submersible pump. foam n: a two-phase system, similar to an emulsion, in
EUE abbr: external upset end. which the dispersed phase is a gas or air.
explosion mode n: a mode of creating an acoustic signal free gas n: a hydrocarbon that exists as a gaseous phase
using a gas gun with a source of compressed carbon di- at a given pressure and temperature rather than in solution
oxide or nitrogen gas at a pressure that is greater than the in the liquid phase.
pressure in the wellbore in order to generate an increasing free wave n: an unbounded wave whose propagating
pressure pulse. medium may be considered uniform and infinite in all
explosion pulse n: a rapid increase in pressure that directions.
propagates within the wellbore. frequency n: the number of cycles completed by a periodic
external upset end (EUE) n: on tubing, casing, or drill quantity in a unit of time—for example, the number of
pipe, the thickening at each end of the joint, such that the wavefronts passing a point in a certain amount of time,
internal diameter of the joint is not affected, but remains usually one second. Typically measured in hertz.
uniform throughout the joint’s length. Only the outside
diameter is enlarged at each end. Pipe is thickened, or
upset, at each end to increase its strength so that threads, gaseous liquid column n: the fluid column below the
couplings, or tool joints may be attached. gas/fluid interface, consisting of a gas-liquid mixture
with gas bubbling through the liquid all the way from the
producing perforations to the gas/fluid interface.
first break n: the point on a wave diagram correspond-
ing to the time of the beginning of amplitude deflection gas gun n: a device installed at the wellhead (tubing or
(positive or negative) from the background sound level. casing), consisting of a quick-opening gas valve and a
volume chamber. Generally utilized to generate a pressure
flowing pressure traverse n: a type of graphic curve pulse through the wellbore gas.
developed to display pressure existing in a wellbore as
a function of depth. gas interference n: a type of incomplete pump fillage
caused by free gas mixed with liquid being admitted
flowing well n: a well in which flow is maintained natu- through the pump intake and transferred to the tubing.
rally without the use of artificial lift equipment and the
flow rate is controlled by adjusting surface back pressure. gas lift n: the process of raising or lifting fluid from a
well by injecting gas down the well through the tubing or
flow line n: the surface pipe through which oil travels the tubing-casing annulus. Injected gas aerates the fluid
from a well to the processing equipment or to storage. to make it exert less pressure than the formation does;
flow rate (q) n: the mass or volume of material flowing the resulting higher formation pressure forces the fluid
per unit of time (lbm/day, bbl/day, ft3/sec, and so on). out of the wellbore. Gas may be injected continuously or
Also known as the rate of flow. intermittently, depending on the producing characteristics
fluid gradient inversion n: a condition in which a high of the well and the arrangement of the gas-lift equipment.
concentration of liquid overlays a section of the wellbore gas lift mandrel n: a device installed in the tubing string
that contains mainly gas. of a gas lift well, onto which or into which a gas lift valve
fluid level n: the distance from the earth’s surface to the is fitted. There are two common types of mandrel. In the
top of the liquid in the tubing or the casing in a well. The conventional gas lift mandrel, the gas lift valve is installed
static fluid level is taken when the well is not producing as the tubing is placed in the well. Thus, to replace or
and has stabilized. The dynamic, or pumping, level is repair the valve, the tubing string must be pulled. In the
the point to which the static level drops under producing side-pocket mandrel, the valve is installed and remove by
conditions. wireline while the mandrel is still in the well, eliminating
flumping n: the flow behavior in a pumping well when the need to pull the tubing to repair or replace the valve.
fluids are intermittently flowing through the pump even See side-pocket mandrel.
though the pump is stroking the plunger (flowing + pump- gas lift valve n: a device installed on a gas lift mandrel,
ing = flumping). which in turn is installed on the tubing string of a gas
flumping well n: a pumping well in which there is a high lift well. Tubing and casing pressures cause the valve to
gas rate, so that both liquid and gas can be produced out open and close, thus allowing gas to be injected into the
of the casing valve and there is not a definite gas/liquid fluid in the tubing to cause the fluid to rise to the surface.
interface in the annulus. See gas lift mandrel.

Glossary G-5
gas/oil ratio (GOR) n: the ratio of produced gas to implosion pulse n: a rapid decrease in pressure that
produced oil. propagates within the wellbore.
gas saturation n: the fraction of gas contained within the inflow performance relationship (IPR) n: the functional
porosity of a reservoir. relationship between the flow rate, flowing bottomhole
gauge n: 1. the diameter of a bit or the hole drilled by the pressure, and static bottomhole pressure, used to assess
bit. 2. a device, such as a pressure gauge, used to measure well performance.
some physical property. 3. in electrical circuits, a measure injectivity test n: a procedure conducted to establish the
of the diameter (thickness) of the wire or other device rate and pressure at which fluids can be pumped into the
that conducts electricity through a circuit or from one treatment target without fracturing the formation.
point to another. In general, gauges for wire conductors interval velocity n: the average acoustic velocity calcu-
range from 24 to 4/0 (000). Some common sizes are 6, lated by using the known distance between successive
12, and 16, with 6-gauge wire being larger in diameter gas lift valves, rather than the average velocity of the
than 16-gauge wire. gas between the surface and the depth of a specific valve.
GOR abbr: gas/oil ratio. intrinsically safe adj: descriptive of an instrument that
gradient n: the change in pressure per unit of depth, is approved for operation in the hazardous area and has
typically in units of psi/feet or kPa/m. been certified by an approved certification agency, such
gradient correction n: a factor that gas accounts for the as the Underwriters Laboratory (UL) or Factory Mutual
reduction of the gradient of a gas-liquid mixture due to Research Corporation (FM or FMRC).
the presence of free gas. intrinsic safety n: the condition of being incapable of
gradient inversion n: a condition in a well’s annulus in releasing sufficient electrical or thermal energy under
which gaseous liquid overlays a section of the wellbore normal or abnormal conditions to cause ignition of a
that contains mainly gas. specific atmospheric mixture in its most easily ignited
concentration.
H2S form: hydrogen sulfide. IPR abbr: inflow performance relationship.
head n: the point in an acoustic record corresponding to
the time of arrival of the wave. KB abbr: kelly bushing.
Horner plot n: graph of plotted numerical data that uses KB correction n: the difference in distance between
a logarithmic scale on one axis and a linear scale on the the wellhead and the rotary table of the completion or
other axis. workover rig.
hydrogen sulfide (H2S) n: a flammable, colorless gaseous kelly bushing (KB) n: an adapter that serves to connect
compound composed of hydrogen and sulfur, which in the rotary table to the kelly. The rotary motion from the
small amounts has the odor of rotten eggs. Sometimes rotary table is transmitted to the bushing and then the
found in petroleum, it causes the foul smell of petroleum kelly itself through the square or hexagonal flat surfaces
fractions. In dangerous concentrations, it is extremely cor- between the kelly and the kelly bushing. The kelly then
rosive and poisonous, causing damage to the skin, eyes, turns the entire drill string because it is screwed into the
breathing passages, and lungs and attacking and paralyzing top of the drill string itself.
the nervous system, particularly that part controlling the kelly bushing (KB) offset n: the distance from the rotary
lungs and heart. In large amounts, it deadens the sense table to the wellhead, necessary for computing marker
of smell. Also called hepatic gas or sulfureted hydrogen. depths.
kick-over tool n: a special tool used for locating the
ideal gas law n: a linear equation of state describing the mandrel pocket and selectively removing or installing a
behavior a hypothetical ideal gas, which is often written gas lift valve without pulling the tubing.
as PV = nRT and states that the pressure and volume of a kill fluid n: mud or other fluid in a wellbore whose weight,
gas are inversely proportional to each other and are directly or density, creates pressure great enough to equal or exceed
proportional to the amount and temperature of the gas. the pressure exerted by formation fluids.
implosion mode n: a mode of creating an acoustic signal
in which the gas gun admits a small volume of the gas landed adj: descriptive of the condition of a downhole
from the wellbore into the gas gun chamber in order to element in a well, such as a pump, of being fixed at a
generate a decreasing (negative) pressure pulse. specific depth.

linear motion pumping unit n: surface unit for rod pump- equipment in the United States. It is part of the National
ing where the polished rod is driven by a linear motion Fire Codes series published by the National Fire Protection
system. Could be mechanical, electrical or hydraulic. Association (NFPA). Also known as NFPA 70.
liquid fraction n: the amount of liquid present in a given National Fire Prevention Association (NFPA) n: a trade
section of the wellbore, casing, flowline, and so on, ex- organization founded in 1896 that creates and maintains
pressed as the ratio of the volume of liquid divided by the private, copyrighted standards and codes for usage and
volume of the specific section of wellbore or pipe. It is a adoption by local governments. Headquartered in Quincy,
dimensionless fraction between zero and one. Massachusetts, it publishes fire and building safety codes
liquid hold-up n: the fraction of liquid present at a given and standards, including the Fire Code and the National
point in the wellbore. Electrical Code. Contact: 1 Batterymarch Park, Quincy,
liquid level n: the depth at which the first liquid is found MA 02169; 800-344-3555; www.nfpa.org.
in a well. National Institute for Petroleum and Energy Research
liquid level depression test n: a procedure for calculat- (NIPER) n: a laboratory for petroleum, fossil fuel, and
ing pump intake pressure by installing a back-pressure natural gas research, now known as the National Energy
regulator in the casinghead flow line and causing a con- Technology Laboratory (NETL).
trolled increase in pressure until the fluid level stabilizes National Institute of Standards and Technology (NIST)
just above the pump intake. The pump intake pressure is n: a measurement standards laboratory and non-regulatory
then utilized with the producing bottomhole pressure to agency of the United States Department of Commerce,
evaluate the well performance. See Walker test. founded in 1901. Headquartered in Gaithersburg, Mary-
liquid loaded gas well n: a type of well producing gas land, its mission is to promote innovation and industrial
from the tubing or casing in which a very light gas gradient competitiveness. Contact: 100 Bureau Drive, Gaithers-
exists above the gas/liquid interface and a heavier gaseous burg, MD 20899; www.nist.org.
liquid gradient exists below the gas/liquid interface, with natural gas separator n: a method of downhole gas
gas bubbles or slugs percolating through the liquid. separation in which the pump intake is located below the
liquid slugging n: the accumulation of batches of water, lowest formation gas entry point, causing an insignificant
oil, or other liquid in a gas pipeline or wellbore. amount of gas to be dragged down to the pump intake at
log-log plot n: a graph of plotted numerical data that uses low liquid annular velocities.
logarithmic scales on both the horizontal and vertical axes. NEC abbr: National Electrical Code.
needle valve n: a type of small valve used for flow meter-
mandrel n: a cylindrical bar, spindle, or shaft around ing, having a tapered needlepoint plug or closure element
which other parts are arranged or attached or that fits and a seat having a small orifice.
inside a cylinder or tube. In gas lift, the mandrel is the NFPA abbr: National Fire Prevention Association.
section of pipe that receives a gas lift valve. NIPER abbr: National Institute for Petroleum and Energy
marker analysis n: a method of calculating acoustic ve- Research.
locity by using the distance to a known wellbore anomaly NIST abbr: National Institute of Standards and Technol-
and the measured time to an easily identified deflection ogy.
(up or down) in the acoustic record. nitrogen (N2) n: a colorless, odorless, inert gas that
mass n: the quantity of matter that a substance contains, forms about 78% of the atmosphere and can be found
independent of external conditions such as the buoyancy in hydrocarbons.
of the atmosphere or the acceleration caused by gravity.
mixture density n: the density of a mixture of hydrocar- operating valve n: the bottommost valve in a continuous
bon liquids, gases, and water, expressed as mass per unit injection gas lift well.
volume in either oilfield units such as lbs./ft3, lbs./gal, or
lbs./bbl or in metric units of gm/cm3. p abbr: pressure.
packer n: a piece of downhole equipment that consists
N2 form: nitrogen. of a sealing device, a holding or setting device, and an
National Electrical Code (NEC) n: a regionally adoptable inside passage for fluids. It is used to block the flow of
standard for the safe installation of electrical wiring and fluids through the annular area between the pipe and the

Glossary G-7
wall of the wellbore by sealing off the space between a piston in a reciprocating pump. 3. the device in a fuel-
them. In production, it is usually made up in the tubing injection unit that regulates the amount of fuel pumped
string some distance above the producing zone. A pack- on each stroke.
ing element expands to prevent fluid flow except through POC abbr: pump-off controller.
the packer and tubing. Packers are classified according potential n: the maximum flow rate of a well, not neces-
to configuration, use, and method of setting and whether sarily an achievable rate.
or not they are retrievable (that is, whether they can be
pressure (p) n: the force exerted uniformly by a substance
remove when necessary or whether they must be milled
per unit of area, measured in pounds per square inch (psi)
or drilled out and thus destroyed).
or kilopascals.
PAP adj: grooved steel pump plunger.
pressure buildup test n: the measurement and analysis
paraffin n: a saturated aliphatic hydrocarbon having the
of bottomhole pressure data acquired after a producing
formula CnH(2n + 2) (e.g., methane, CH4; ethane, C2H6).
well is shut in. Soon after a well is shut in, the fluid level
Heavier paraffin hydrocarbons (e.g., C18H38) form a
in the wellbore usually reaches a stable state in which
wax-like substance that is called paraffin. These heavier
bottomhole pressure rises smoothly and is easily mea-
paraffins often accumulate on the walls of tubing and
sured, allowing for interpretable test results to accurately
other production equipment, restricting or stopping the
estimate static formation pressure.
flow of the desirable lighter paraffins.
PBHP abbr: producing bottomhole pressure. pressure gradient n: 1. a scale of pressure differences in
which there is a uniform variation of pressure from point
PCP abbr: progressing cavity pump.
to point. For example, the pressure gradient of a column
PDP abbr: pump discharge pressure.
of water is about 0.433 pounds per square inch per foot
peak n: the point of maximum amplitude on a wave. (9.795 kilopascals per metre) of vertical elevation. The
perforation n: a hole made in the casing, cement, and normal pressure in a formation is equivalent to the pres-
formation through which formation fluids enter a wellbore. sure exerted at any given depth by a column of 10% salt
Usually, several perforations are made at a time. water extending from that depth to the surface (0.465
permeability n: the ability, or measurement of a rock’s pounds per square inch per foot, or 10.518 kilopascals
ability, to transmit fluids, typically measured in darcies per metre). 2. the change (along a horizontal distance)
or millidarcies. Formations that transmit fluids readily, in atmosphere pressure. Isobars drawn on weather map
such as sandstones, are described as permeable and tend display the pressure gradient.
to have many large, well-connected pores. Impermeable pressure traverse n: a graphical representation of the
formations, such as shales and siltstones, tend to be finer- pressure in the wellbore as a function of depth.
grained or of a mixed grain size, with smaller, fewer, or
pressure-volume-temperature (PVT) n: properties of
less interconnected pores.
matter, which in a fluid are related and can be used to
PI abbr: productivity index. compute the fluid’s density and gradient
PIP abbr: pump intake pressure.
producing bottomhole pressure (PBHP) n: The pres-
plane wave n: a wave of constant frequency and ampli- sure measured in a well at or near the depth of the prod-
tude with wavefronts that are an infinitely long straight ucing formation, obtained while the well is being produced
line, which travel in the direction perpendicular to the at normal conditions.
wavefronts.
producing formation n: the underground rock zone from
plunger n: 1. a basic component of the sucker rod pump which fluids are produced into the wellbore.
that consists of a hollow tube with the traveling valve and
producing pressure n: the pressure inside the well at the
some form of seal on the outside. The external seal may be
depth of the producing formation.
fiber or elastomer rings, or it may simply be a very close
tolerance fit with the barrel. The seal partially prevents production rate n: the flow rate of the fluids flowing
fluids from migrating back beneath the plunger around from the producing formation, generally expressed as
the outside, while allowing some slippage to lubricate stock tank barrels per day (STBO/day).
the surfaces. The traveling valve allows fluid to flow up productivity index (PI) n: a mathematical means of ex-
the hollow inside the plunger during the downstroke and pressing the ability of a reservoir to deliver fluids to the
then closes to prevent fluids from migrating back below wellbore, expressed as the flow rate or volume delivered
the plunger during the upstroke. 2. the rod that serves as per psi of drawdown (bbl/day per psi).

progressing cavity pump (PCP) n: a type of subsurface recommended practice n: a technical specification ad-
pump in which the rods are rotated instead of recipro- opted and recommended by the American Petroleum In-
cated. The pump consists of an auger-like assembly that stitute for safe and optimal handling of oilfield equipment.
rotates inside a cylinder. The reverse screw action forces reflection n: the return or rebound of particles, waves, or
fluid up the tubing to the subsurface. Especially useful energy from the interface between two media.
in heavy crude-high solids content lift applications. reservoir n: 1. a subsurface, porous, permeable rock
pump discharge pressure (PDP) n: the pressure at the body in which oil and/or gas has accumulated and been
bottom of the tubing and above the pump discharge. trapped. Most reservoir rocks are limestones, dolomites,
pump displacement n: the volume of fluid displaced sandstones, or a combination thereof. 2. a container or
by the pump during a complete pump stroke. vessel that stores fluid, such as a reservoir on an accu-
pump effective displacement n: the volume of liquid mulator that holds hydraulic operating fluid.
actually transferred from the pump intake to the pump reservoir pressure n: the average pressure within the
discharge and into the bottom of the tubing. reservoir at any given time. Determinining this value is
pumping well n: a well in which the flow rate is con- best made by bottomhole pressure measurements with
trolled by pump displacement. The producing BHP is the adequate shut-in time. If scheduling a shut-in period long
response of the formation to the rate demanded by the enough for the reservoir pressure to stabilize is impractical,
pump. Fluid level and casing pressure adjust themselves then various techniques of analysis by pressure buildup
to balance the PBHP (casinghead back-pressure affects or drawdown tests are available to help determine static
fluid level depth but does not influence the PBHP as long reservoir pressure.
as a gaseous fluid column exists above perforations). round trip travel time (RTTT) n: the total time required
pump intake pressure (PIP) n: 1. the pressure in the for a signal pulse to travel from a specific source to a
annulus at the depth of the pump intake. 2. the pressure specific destination and back again.
below the plunger when the standing valve opens during RP abbr: recommended practice.
the upstroke of a rod pump. RTTT abbr: round trip travel time.
pump off v: to keep the pressure at the bottom of the
well as low as possible, by not allowing a liquid column
safety valve n: a valve that automatically shuts off the flow
to form above the perforations and keeping the surface
of fluids from within the production string if the wellhead
pressure on the casing-tubing annulus as low as possible.
is damaged or an accident causes an uncontrolled outflow
pump-off controller (POC) n: a device used to place the of fluids at the surface.
pumping unit in intermittent operation without changing
SBHP abbr: static bottomhole pressure.
the speed and/or the stroke, in the unusual circumstance
that the unit cannot be slowed to match the pump rate side-pocket mandrel n: a mandrel offset from the cen-
to the reservoir rate. terline of the tubing.
pump submergence n: the difference in hydrostatic single-phase adj: in a mixture of hydrocarbons, having
head between the pump intake depth and the dynamic the same fluid properties. Can refer to solid, liquid, or gas.
fluid level above the pump. The pump submergence is sinusoidal adj: 1. of or relating to a sinusoid. 2. oscillating
continuously monitored to adjust the pump flow rate smoothly and repetitively, similar to the form of a sine
and avoid a pump-off condition. wave. It occurs often in pure and applied mathematics,
PVT abbr: pressure-volume-temperature. as well as physics, engineering, signal processing and
many other fields.
q abbr: flow rate. skin n: the area of the formation that is damaged because
of the invasion of foreign substances into the exposed
rarefaction pulse n: a pressure pulse generated by a gas section of the formation adjacent to the wellbore during
gun into a wellbore when the pressure is greater inside drilling and completion. In flow equations such as Darcy’s
the casing annulus than in the volume chamber. See Law, skin is expressed in dimensionless units; a positive
implosion pulse. value denotes formation damage, while a negative value
real gas law n: a non-linear equation describing the rela- indicates improvement. Also called skin effect.
tionship between the volume of a gas and its composition, slug n: 1. a quantity of fluid injected into a reservoir
pressure, and temperature. to accomplish a specific purpose, such as chemical

Glossary G-9
displacement of oil. 2. in multi-phase flow, a pocket of sucker rod string. Basic components are the barrel, plunger,
fluid of one type that is flowing interspersed with the rest valves, and hold-down. Two types of sucker rod pumps
of the flow. For example, pockets of gas may alternate are the tubing pump, in which the barrel is attached to
with slugs of liquid in flow in a pipe. the tubing, and the rod, or insert pump, which is run into
slug flow n: a type of gas-liquid flow pattern characterized the well as a complete unit.
by a series of liquid plugs (slugs) separated by relatively sucker rod pumping n: a method of artificial lift in
large gas pockets. which a subsurface pump located at or near the bottom
solenoid n: a cylindrical coil of wire that behaves like a of the well and connected to a string of sucker rods is
magnet when electrical current is conducted through it. used to lift the well fluid to the surface. The weight of
sonic velocity n: see acoustic velocity. the rod string and fluid is counterbalanced by weights
specific acoustic impedance n: the ratio of pressure at attached to a reciprocating beam or to the crank member
a point in a specific fluid to the fluid’s particle velocity. of a beam pumping unit or by air pressure in a cylinder
specific gravity n: the ratio of the density of a substance attached to the beam.
to the density of a standard substance, such as water or air. superficial gas velocity n: the velocity of gas moving
static bottomhole pressure (SBHP) n: The pressure through a pipe, defined as the in situ volumetric flow rate
measured in a well at or near the depth of the formation, of that fluid divided by the cross-sectional area of the pipe.
obtained when the well has been shut-in long enough that superficial velocity n: a hypothetical (artificial) flow
the surface and bottomhole pressures in the wellbore have velocity calculated as if the given phase or fluid were the
stabilized and there is no inflow of fluids. only one flowing or present in a given cross-sectional area.
static fluid level n: the level to which fluid rises in a well surfactant n: a chemical that preferentially adsorbs at
when the well is shut in long enough for the bottomhole an interface, lowering the surface tension or interfacial
pressure to equalize with the formation pressure. tension between fluids or between a fluid and a solid.
static pressure n: 1. the stationary or line pressure exist- swab valve n: the topmost valve on a wellhead that pro-
ing in a vessel or pipe. 2. the pressure exerted by a fluid vides vertical access to the wellbore.
on a surface that is at rest in relation to the fluid. 3. the
pressure exhibited at the surface or a point downhole
during the time that the well is shut-in. 4. the surface or TAC abbr: tubing anchor-catcher
bottomhole pressure after sufficient time has elapsed for tail n: the point corresponding to the end of the wave pulse.
the pressure to become stable. test point n: a given value of flow rate and flowing bot-
steady state n: a state or condition of a system or process tomhole pressure used as a control.
that does not change in time, which in a well means that transducer n: a device that converts variations in a
the producing flow rate is stabilized and the gas/oil ratio physical quantity, such as pressure or brightness, into an
and water/oil ratio are consistent with past performance. electrical signal and vice versa.
strip chart n: a chart containing one or more electrical transmission n: the propagation and resultant change in
or mechanical input trends printed on a long strip of energy of a wave moving from one medium to a medium
paper, used to analyze well performance and obtain the of different density.
distance to the liquid level in the casing annulus of a well
tubing n: relatively small-diameter pipe that is run into
by acoustic means.
a well to serve as a conduit for the passage of oil and gas
sucker rod n: a special steel pumping rod. Several rods to the surface.
screwed together make up the mechanical link from the
beam pumping unit on the surface to the sucker rod pump tubing anchor-catcher (TAC) n: a device used to an-
at the bottom of a well. Sucker rods are threaded on each chor the tubing string to the casing at a desired depth by
end and manufactured to dimension standards and metal maintaining tension in the tubing string while catching
specifications set by the petroleum industry. Lengths are and preventing any pipe from falling into the well.
25 or 30 feet (7.6 or 9.1 metres); diameter varies from H tubulars n pl: any kind of pipe. Oilfield tubulars include
to 1J inches (12 to 30 millimetres). There is a continuous tubing, casing, drill pipe, and line pipe. Also called tu-
sucker rod (trade name: CorodTM). bular goods.
sucker rod pump n: the downhole assembly used to two-phase adj: descriptive of a region of matter in which
lift fluid to the surface by the reciprocating action of the both liquid and gas coexist.

United States Geological Survey n: a scientific agency water/oil ratio (WOR) n: the ratio of produced water to
of the United States government and a bureau of the produced oil.
Department of the Interior, founded in 1879. Headquar- wave equation n: a differential equation expressing the
tered in Reston, Virginia, the agency’s scientists study properties of motion in waves.
the landscape of the United States, its natural resources, wellbore n: a borehole; the hole drilled by the bit. A well-
and its natural hazards. Contact: 12201 Sunrise Valley bore may have casing in it or may be open (uncased), or
Drive, Reston, VA 20192; 703-648-5953; www.usgs.gov. part of it may be cased and part of it may be open. Also
unloading valves n pl: the upper valves in a continuous called a borehole or hole.
injection gas lift well. wellbore pressure n: 1. bottomhole pressure. 2. casing
up-kick n: a decrease in pressure as detected by the pressure.
acoustic fluid level instruments, resulting in a positive wellbore storage effect n: a distortion in the reservoir
signal and an upwards polarity in the recorded signal. pressure transient response because of wellbore storage.
wellhead n: the equipment installed at the top of the
V abbr: volume. wellbore. A wellhead includes equipment such as the
valley n: the point of minimum amplitude on a wave. casinghead and tubing head. adj: pertaining to the well-
valve spread n: the difference between opening and head (for example, wellhead pressure).
closing pressures in a gas lift valve. well performance n: the relationship between the fluid
variable speed drives n pl: a type of adjustable-speed flow rate and the pressure drawdown between the well-
drive used in electro-mechanical drive systems to con- bore and the formation pressure.
trol AC motor speed and torque by varying motor input wireline n: a slender, rod-like or threadlike piece of metal,
frequency and/or voltage. usually small in diameter, which is used for lowering
viscosity n: a measure of the resistance of a fluid to flow, special tools and/or sensors (such as logging sondes,
also thought of as the thickness of a liquid. Viscosity has perforating guns, and so forth) into the well. Also called
units of centipoise (cp) or Pascal-sec (Pa-sec). slick line.
Vogel’s Equation n: a realistic inflow performance wireline-conveyed adj: lowered into a well and later
model for two-phase flow, developed through numeri- retrieved by means of a wireline rig.
cal simulation of well performance, which describes a WOR abbr: water/oil ratio.
dimensionless reference curve applicable to most wells work n: the acting of a force through distance or the
producing in reservoirs below bubble point pressure with overcoming of a resistance to motion, for example by
a minimum deviation. lifting a mass of material from a certain depth during a
volume (V) n: a confined three-dimensional space. Units given time. No work is done unless motion is produced.
correspond to the dimensions of the confined space, for Mathematically, work is force times distance.
example, liters (l), cubic centimetres (c3), gallons (gal),
cubic feet (ft3), and so on. zero net liquid flow n: the fluid multi-phase flow pat-
tern in the annular gaseous liquid column, in which the
Walker test n: a procedure for determining pressure dis- gas phase is flowing upwards to the top of the wellbore
tribution in oil wells by the pressure-wave echo method. and out of the casinghead while liquid only recirculates
See liquid level depression test. in place.

Index I-1
Index
Throughout this index, an f indicates a figure and a t indicates acquisition and recording of, 4-10 to 4-11
a table on that page. to consider real behavior of an acoustic wave, 3-2
converting time scale to depth, 3-14
absorption coefficient, 3-20 to 3-21, 3-22 correctly operating SCSSSV, 2-25
accelerometer, 4-4 examples of, 2-4t
acoustic echo. See echoes. graphical representation of, 3-13 to 3-14
acoustic fluid level measurements. See also gas lift wells; gas guidelines for, 2-1
importance of composition in, 3-16
wells; plunger lift wells; pumping wells.
incorrectly operating SCSSSV, 2-25
advantages of over wireline equipment, 9-28
multiple echoes, 2-12f, 3-10f
automatic, 1-9, 7-2 to 7-4, 7-23, 7-26
overlay of, 2-23f
diversity in records of, 2-35
pulse generation systems and accuracy of, 4-7 to 4-8, 4-7f
as an essential tool, 1-22 to 1-23, 3-1
rod-pumped wells, 2-32f
history of, 4-2 to 4-4, 6-3
tasks involved, 3-6
importance of accurate results of, 5-15
unexplained echoes in, 2-2
misleading, 8-5 to 8-13
in wells with multiple producing zones, 7-13, 7-14f
mixed flow calculations from, 6-24
in a well with blast joints, 2-15f
public domain software for, 1-10, 1-11f
quality control of data, 2-1 to 2-2 acoustic resonance, 4-14, 4-15f
recommendations for acquisition of signal for, 2-3 acoustic signal
recommended operating procedures, 4-2, 4-13 to 4-19 digitalization of, 4-11, 4-19
from SBHP, 6-23 history of developments of, 4-4
technological advances and, 4-10 operating procedures in acquiring, 4-15 to 4-16
testing in stabilized conditions, 1-1, 1-6 overview and history of, 4-1 to 4-4
typical analysis of, 1-5f processing techniques used with, 4-1
well performance and, 1-1, 1-4 to 1-5 random, 4-16 to 4-17, 4-17f
acoustic fluid level measurements, examples of recommendations for acquisition of, 2-3
acoustic records and analysis, 2-4f recording and processing, 4-11
corrosion survey of intermediate casing, 2-4t, 2-27 to 2-30 safety considerations while acquiring, 4-11 to 4-12
deviated wellbore, 2-4f, 2-8 to 2-10, 2-8f, 2-10f, 2-20, technological advances in, 4-10
2-22, 2-22f acoustic velocity
ESP well casing shots, 2-4t, 2-15 to 2-20, 2-17f, 2-19f average, 8-14
ESP well with hole in tubing, 2-4t, 2-20 to 2-24 calculating from gas composition, 3-17
simple wellbore with uniform casing and tubing calculating round trip travel time and, 3-23
diameters, 2-4t, 2-5 to 2-8, 2-5t, 2-6f, 2-7f calculation of using known distance, 2-23f
stratified annular gas column, 2-4t, 2-30 to 2-35, 2-32f, 2-37f collar count method and, 2-19f, 2-30f
surface controlled subsurface safety valves (SCSSSV), comparing for reasonability, 5-13 to 5-15, 5-14f, 5-16
2-4t, 2-25 to 2-27 correlation function of RTTT versus, 2-35, 2-35f, 10-12,
tapered tubing and casing liner, 2-4t, 2-10 to 2-11, 2-11f, 10-13f, 10-14, 10-27
2-12f, 2-13 to 2-14, 2-13f defined, 3-1
well with blast joint opposite upper perforations, 2-4t, from depth of tubing, 2-36f
2-14 to 2-16, 2-14f, 2-16f in an ESP well, 2-18
acoustic pulse. See also amplitude attenuation; round trip estimating, 5-13 to 5-15, 5-14t
travel time (RTTT). gas and, 3-2, 3-15 to 3-16, 3-20, 3-23
amplitude versus time, 3-3 to 3-5, 3-3f, 3-4f gas composition and, 2-34 to 2-35, 3-16 to 3-20
converting travel time to distance, 2-1, 5-1 gas lift wells and, 10-6 to 10-7, 10-12, 10-12f, 11-16
generated by plunger, 11-7, 11-9, 11-11, 11-11f history of computation of, 4-2
generation of, 4-5 to 4-10 pressure and temperature and, 3-18f
observed at different depths, 3-4f pressure buildup test and, 7-4, 7-4f
observed at different locations, 3-4 in a simple wellbore, 2-5, 2-7
overview and history of, 4-1 to 4-4 in tubing, 8-14
recommendations on generation of, 4-14, 4-14f, 4-19 variation of, 5-14 to 5-15, 5-14f, 7-4f
recording, 4-10 to 4-11 verifying range of, 2-2
acoustic record analysis. See also digital filters. adiabatic gas law, 3-15
acquiring multiple records for, 2-2, 2-3, 2-9 to 2-10, after-flow
2-10f, 4-17, 10-35 ERCB standards and, 6-3

I-1
I-2 ACOUSTIC FLUID LEVEL MEASUREMENTS IN OIL AND GAS WELLS HANDBOOK
gas, 1-9 blank cartridge pulse, 4-9 to 4-10

in a plunger lift well, 11-5 to 11-6 blast joints, 2-4f, 2-14, 2-15f
shut in wells and, 1-9, 6-23, 7-5 blowout, 4-10
water/oil ratio (WOR) behavior, 7-6 bottomhole pressure (BHP)
Alberta Canada blowout, 4-10 annular gas production and, 7-6
American Petroleum Institute (API), 4-12 calculating, 7-5, 7-6
amplitude. See also acoustic pulse. estimating, 9-11, 9-14
different pulses and, 2-3 example results, 7-16f
estimating, 3-23 importance of measurement of, 3-1
example of, 3-8 multiple producing zones and, 7-13
history of reading, 4-2 pressure transient analysis for, 7-1
inspection of, 5-15 programmed fluid level surveys and, 7-2 to 7-3
low, 2-2, 2-25, 2-27 when calculating fluid level depth, 6-22
negative, 3-12 bottomhole pressure sensor, 1-8f, 1-9
overview of, 3-23 bubble point, 1-12 to 1-17, 1-14f, 1-24, 6-5, 7-6
peak-to-peak, 3-2, 3-11, 3-22, 4-16, 5-4 bubble point line, 3-16, 3-16f
of pressure pulse, 4-5 bumper spring, 11-3
subsurface safety valves and, 9-15
versus time, 3-3 to 3-5, 3-3f, 3-4f
cable banding, causing noise, 8-21, 8-21f
amplitude attenuation
coiled tubing and, 3-22f calibration and maintenance of instrumentation, 4-18 to 4-19
developments in reducing, 4-2 casing
in low pressure wells, 5-2, 5-2f acoustic surveys to troubleshoot, 11-17 to 11-18, 11-18f
mist and, 9-5 change in diameter and echo reflection, 5-8, 5-8f
overview of, 2-10, 3-9 to 3-11, 3-20 to 3-22, 3-21f, 4-1 integrity test, 2-27 to 2-30, 2-28f, 2-29f, 2-31f
stepwise collar echo count, 5-4 small diameter, 8-21
wellhead pressure and, 4-15 changing, 8-2
analyzing well performance 2000, 1-10, 1-11f, 1-23 to determine well classification, 6-5
estimating acoustic velocity and, 5-14 to 5-15
annular fluid gradient
example results, 7-15f, 7-16f
discontinuity in, 8-23 to 8-26
fluid levels and, 6-12t, 6-13, 6-13f, 6-22, 8-10
inversion of, 8-2, 8-6 to 8-10, 8-7f, 8-8f, 8-9f
gradient inversion and, 8-10, 8-11f
overview of, 8-1 to 8-2
monitoring, 6-3
annular gaseous liquid column, 6-4 to 6-5, 6-7, 6-22 PBHP and, 1-7
annular gas flow rate, 6-10, 6-18 to 6-20, 6-20f shut-in wells and, 6-5
annulus, sound pressure wave propagation in, 3-2, 3-13 to stable, 1-7, 1-8, 6-4
3-15 casing pressure
anomalies automatic recording of, 7-2
downhole, identifying, 2-25, 5-4 buildup of with tapered tubing, 2-11f
paraffin, 8-20 in corrosion survey, 2-27
in plunger lift wells, 11-13, 11-17 example results, 7-15f
API gas light design technique in RP 11V6, 10-24 fluid level depression and, 8-9
API gravity of oil, 6-20, 6-20t fluid levels and, 6-8, 6-11f
arrival sensor, 11-3 in a multi-rate flow test, 1-21f
artificial lift systems in plunger lift wells, 11-13
costs of, 1-2, 1-2f policies in measuring, 1-23
designing for well’s potential, 1-20, 1-24 relation to tubing pressure, 10-10f, 10-16, 10-17, 10-18
production and, 8-1 in a simple wellbore, 2-7f
well efficiency and, 1-2, 1-23 to 1-24 temperature and, 7-4
wireline-conveyed equipment in, 7-1 variations of during pressure transient test, 7-5f
automatic acoustic fluid level survey, 1-9, 7-2 to 7-4, 7-23, casing-pressure-actuated gas lift system
7-26 gas lift valves in, 10-21, 10-22
automatic acoustic pressure buildup test, 6-24 monitoring unloading, 10-31 to 10-34, 10-32f
automatic pulse generation, 4-8 to 4-9 unloading sequence, 10-26 to 10-34, 10-26f, 10-27f
AWP 2000, 1-10, 1-24 unloading valves, 10-28f, 10-29f, 10-30f
casing valve, 1-7, 1-10
background noise, 5-6, 5-15, 8-26, 10-2 to 10-3. See also noise. catcher, plunger lift well, 11-2
back-pressure regulating valve, 1-8, 6-9f, 6-10, 7-6, 8-17 chamber pressure, 4-16f
beam-pumped wells, 7-8 to 7-13, 7-15f to 7-17f, 7-17 check valve, 11-3
Benedict-Webb-Rubin (BWR) equation of state, 3-17 chemical pot, 2-3, 8-30

Index I-3
choked pump, 8-5, 8-6f productivity of, 2-10, 2-11f

chromatographic analysis, 2-18f, 5-10 SBHP and, 9-15
churn type of liquid flow, 8-26 typical acoustic record of, 2-4f
“C” marker, 3-15, 5-2 dew point, 3-16, 3-16f
coal bed methane, 3-17, 5-10, 6-7 differential pressure, 4-6, 4-15 to 4-16, 4-16f, 4-19, 5-15
coiled tubing, 4-4, 5-10 digital filters. See also acoustic record analysis; noise.
collar count method. See also tubing joints. automatic for determining liquid level, 5-6 to 5-7, 5-7f,
for acquisition of fluid level records in tubing, 8-14 5-9t
automatic digital filtering for, 5-3, 5-3f, 5-6 to 5-7, 5-7f, in the collar count method, 5-3, 5-3f, 5-6 to 5-7, 5-7f, 5-9t
5-9t comparison of raw and filtered, 2-9f
compared to downhole marker method, 5-9, 5-9t in a gas lift well, 10-2 to 10-3
example results, 2-33 to 2-34, 2-34f gas noise and, 4-18, 4-18f
gas composition and, 2-34 to 2-35 high-pass, 4-2
history of, 4-2 history of, 4-3
indistinguishable echoes in, 5-10 low-pass, 2-8 to 2-9, 2-9f
overview of, 3-10, 5-2 to 5-4 objective of, 3-15
in plunger lift wells, 11-9, 11-13f, 11-14f, 11-16 to 11-17, overview of, 4-11
11-17f to reduce background noise, 5-15
stepwise collar echo count, 5-4, 5-5f, 5-6f, 5-9t technology and, 5-3
collar echoes, 3-14, 5-10 digital processing of acoustic trace, 5-3, 5-3f
controllers for plunger lift wells, 11-2, 11-3, 11-5 digital spectral analysis, 7-3
correlations. See also Vogel’s Equation. distance scale, 3-15
acoustic velocity and RTTT, 2-35, 2-35f, 10-12, 10-13f, downhole gauges, 9-15
10-14, 10-17 downhole marker analysis
fluid level and dynamometer, 8-31 compared to other methods, 5-9t
for gas properties, 3-19 to 3-20 example of, 2-12f
linear, 6-13 gas lift mandrels, 10-8 to 10-14
percentage of liquid in annular gaseous column, 6-14, history of, 4-2
6-16 to 6-18, 6-22 overview of, 5-8 to 5-9, 5-8f
pressure and temperature, 3-17 plot of, 5-4f, 5-6f
“S” curve, 6-18, 6-19f, 6-20, 9-11, 9-14, 9-14f for shut-in wells, 9-15
well fluid composition and distribution, 7-5 to 7-6 downhole pressure, 1-23, 2-2
corrosion in the tubing, 2-4f, 9-23f, 9-25 downhole pressure sensors, 3-1, 8-23 to 8-26
critical point, 3-16, 3-16f downhole pump submergence, 6-1
critical rate down-kick echo, 2-11 to 2-12, 3-3, 3-12, 3-13, 8-17. See also
fluid levels for gas flow above, 9-5 echoes.
fluid levels for gas flow below, 9-7, 9-9 to 9-11, 9-9f drainage radius, 1-12
overview of, 9-1 to 9-3, 9-2f drawdown, 1-11f, 1-12, 1-14, 1-15, 1-23
dry tubing, 11-18, 11-18f
Darcy’s Law, 1-12 dynamometer
database information, 4-3 compared to the PIP value, 8-4
deep wells, 9-15, 10-9, 10-10f, 10-12, 10-35 in conjunction with fluid level survey, 2-15, 2-16f, 8-2 to
delta pressure, 7-16f, 7-19f 8-3, 8-9, 8-10, 8-17, 8-18, 8-30 to 8-31
depleted reservoirs, 2-5, 6-2 example results, 2-32, 2-33f
depth high, choked pump and, 8-5
acoustic record generating profile of, 10-13f, 10-14 readings for a well exhibiting fluid gradient inversion, 8-8
acoustic velocity and, 3-20, 5-1, 10-12f to 8-9, 8-8f, 8-9f
gas composition and, 2-34 to 2-35 showing pump displacement, 7-8, 7-9f
to liquid level, field data and, 7-15f
loss of tubing joint echoes and, 5-2, 5-4, 5-6f echoes. See also down-kick echo; up-kick echo.
plunger lift wells and, 11-8 analysis of multiple, 2-24, 2-24f
pressure-depth traverse, 9-10f from the casing collars, 2-30f
pressure distribution and, 8-25, 8-25f, 9-10 clearly defined, 4-1
detailed pipe tally, 5-7 complex, caused by gas lift mandrels, 10-10 to 10-11,
deviated wellbore. See also wellbore. 10-10f, 10-11f
calculating fluid level depth and, 6-22 foam layer and, 6-7
example results of, 2-4t, 2-8 to 2-10, 2-8f, 2-10f, 2-20, graphical representation of, 3-14, 3-14f
2-22, 2-22f inverted polarity, 10-10
locating valve in a, 10-23 liquid levels and, 2-3, 2-18

mismatch in with gas lift mandrels, 10-8f, 10-9 flow meters, 6-18
multiple, 2-12f, 3-10f flow rate
polarity of, 3-23, 4-8, 8-17, 9-22f drawdown and, 1-14
pressure buildup test and, 7-11f gas lift wells, 10-19
primary requirements for clear, 4-19 overview of, 1-7
rod couplings and, 8-14, 8-14f pattern of, 6-5
tubing collars, 2-5, 2-6f, 2-7f pressure below bubble point and, 1-15, 1-15f
from tubing joints, 5-2 stabilized, 1-12, 1-15, 6-3
unexplainable, 2-2, 2-2f well performance and, 1-1
visualizing features creating, 2-1 flow stabilization, importance to accurate readings, 6-21
wells with blast joints, 2-15 fluid above pump (FAP), 6-1
in a well with blast joints, 2-15 to 2-16, 2-16f fluid column
Echometer, 1-10 calculating PBHP and, 1-7 to 1-8
effective oil fraction, 6-17f, 6-18, 6-19f choked pump and, 8-6f
electrical submersible pump (ESP) normal, 8-6
acoustic fluid level acquisition, 8-21f in a stabilized pumping well, 6-4, 6-4f
calculating producing bottomhole pressure in, 1-7 fluid distribution in a static well, 6-23f
comparing downhole pressure sensor measurement to fluid gradient inversion. See annular fluid gradient;
acoustic fluid level survey, 8-23 to 8-26 gaseous liquid column gradient.
example records from, 2-15 to 2-16 fluid level interface, 8-31
fluid level surveys acquisition, 8-21 to 8-22 fluid levels
gaseous column in, 8-22 annular, importance of periodic monitoring of, 10-14
with hole in tubing, 2-20 to 2-24 casinghead pressure and, 6-12t, 6-13, 6-13f, 6-22, 8-10
multi-rate flow test with, 1-21 to 1-22, 1-21f casing pressure and, 6-8, 6-11f
noise production from, 2-18 comparing to dynamometer surveys, 8-9, 8-10, 8-17,
pressure transient analysis in, 7-13 8-30 to 8-31
productivity analysis for, 8-22f depressing for acoustic pressure buildup test, 1-10
summary of, 2-21f high, 8-5, 8-8, 8-17
typical annular pressure distribution, multiple producing importance of accuracy in, 8-2
zones, 8-24f measuring, 6-4
typical fluid level record, multiple producing zones, 8-23 policies in measuring, 1-23
to 8-26, 8-23f producing, 2-18, 2-19f, 2-21
Walker test and, 1-8f in pumping wells, 6-2 to 6-3
Energy Resources Conservation Board (ERCB), 6-3 records for gas flow above critical rate, 9-5, 9-7
in a shut-in gas well, 9-19f
environmental concerns, 4-4, 4-7
static, 2-20, 2-20f, 2-21f
equation of state, 3-2, 3-15, 3-17, 3-19, 6-3, 6-7 surveying ESP wells, 8-21 to 8-22
equipment. See instruments. typical, 8-23f
equivalent gas-free liquid height, 6-3 fluid level survey. See acoustic fluid level measurements.
ESP. See electrical submersible pump (ESP). fluids
explosion pulse composition and distribution in wellbore, 7-5 to 7-6
compared to implosion pulse, 4-7 to 4-8, 4-7f documenting movement of, 8-18 to 8-19
in a gas well, 9-3 history of developments in determining properties of,
generation of, 4-8 4-4
history of, 4-2 particle velocity of, 3-2
history of development of, 4-4 produced, to generate an acoustic signal, 4-4
overview of, 4-5, 4-5f, 4-19 properties of and reflection, 3-7 to 3-11, 3-7f
external upset end (EUE) connections, 3-14 reflection at the discontinuity of, 3-7 to 3-11, 3-7f
wellbore pressure distribution and, 6-2
Factory Mutual Research Corporation (FM, FMRC), 4-13 flumping well, 1-7, 8-31
Fetkovich’s approximation, 1-20, 1-20f, 1-21 foam, 3-8, 6-7
first break, 3-3 formation pressure, 1-1, 1-15, 1-15f
flowing gas wells, 2-26f, 4-9 Fourier Analysis (FFT), 4-4
flowing pressure traverse, 9-1, 9-9, 9-9f free gas, 1-23, 6-7, 6-18, 8-10 to 8-11
flowing wells free gas flow, 1-12, 6-23, 7-6, 8-4
categories of, 9-2 free-gas phase, 1-14 to 1-15, 1-16
fluid surveys in, 9-1 free wave, 3-3
performance of, 1-4, 2-25 frequency content of a pulse, 5-3 to 5-4
permanent pressure readout gauges in, 7-1 fuel consumption, 1-4

Index I-5
gas composition gas interference, 8-3, 8-4 to 8-5, 8-4f

depth and, 2-34 to 2-35 gas lift injection manifold, 10-33f
effect on acoustic velocity, 2-34 to 2-35, 3-16 to 3-20 gas lift mandrels
portable instruments for analyzing, 5-12f on the acoustic record, 10-6
gas compression chamber, 4-4 causing complex acoustic record, 10-10f
gas emission port, 4-4 overview of, 10-22 to 10-23, 10-24f
gaseous liquid column using as markers, 10-8 to 10-14, 10-9f
determining fraction of liquid in, 6-7 using for liquid level calculations, 10-8
flow pattern of, 6-5 gas lift wells
fraction of gas in, 6-17f, 6-18 acquiring multiple records, 10-4, 10-6
height of, 6-11f, 6-13, 6-13f, 6-22, 8-2, 9-8f analyzing fluid level records, 10-6 to 10-14
overview of, 1-7 background information about, 10-19 to 10-21
presence of in an ESP well, 8-22 benefits of fluid level measurements for, 10-2
properties of, 6-4 casing pressure operated systems, 10-26 to 10-34
SBHP calculations and, 1-9 to 1-10 design of, 10-24
during shut in, 7-6 equipment installation and data acquisition, 10-2 to 10-6
Walker test and, 6-7 to 6-11 examples of fluid level and pressure surveys, 10-17 to
gaseous liquid column gradient 10-19
acoustic velocity and, 2-34 to 2-35 instrument connection to, 10-3f
calculating, 6-6, 6-8f, 7-5, 9-27 overview of fluid level measurement applications for, 10-1
changes in, 9-10, 9-11f, 9-14 pressure buildup test and, 7-13
correlations for determining, 6-16 to 6-18 troubleshooting, 10-14
increase in for shut in well, 9-10 typical, 10-1f
Walker test and, 6-14f valve operation, 10-22
gases valves and mandrels, 10-22 to 10-23
acoustic velocity in, 3-15 to 3-16, 3-23 well unloading, 10-24, 10-26 to 10-34
after-flow, 1-9 gas/liquid interface
calculating speed of sound in, 3-15 to 3-16 changes in, 9-11, 9-11f, 9-14
composition of and acoustic velocity, 2-34 to 2-35, 3-23 depth and, 9-7, 9-8f
considering phase behavior of, 3-2 example results, 3-8 to 3-9, 3-9f
excessive amounts of in a well, 8-17 finding distance to, 6-22
obtaining sample of, 5-10 flow rate and, 1-7
pressure of and signal quality, 4-1 to 4-2 formation of, 9-3
properties of, 3-2, 11-16 to 11-17 gradients and, 9-1
real behavior of, 3-2, 3-17 liquid level depression test and, 6-12f, 6-15f
solubility of, 5-10 to 5-13, 6-5 for wells flowing below critical rate, 9-9, 9-10
sonic velocity and, 3-1 well types and, 6-5
specific gravity of, 3-17, 3-18f, 3-23, 9-4 gas/liquid phase, 3-16 to 3-20
stratification, 3-20, 5-1, 5-9 gas/liquid ratio, gas lift wells, 10-19
gas flow, 4-18, 4-18f, 6-10 gas/mist interface, 9-3, 9-5
gas flow rate gas/oil ratio (GOR), 1-6, 1-9
acoustic velocity and, 3-20 gas properties, 3-15
calculating, 6-18 to 6-20, 6-20f, 7-5 Gas Research Institute, 3-19 to 3-20
calculating PBHP and, 1-7 to 1-8 gas saturation, 1-14 to 1-15, 1-14f
critical rate and, 9-1 gas separator
estimating, 6-7 centrifugal, 8-22
overview of, 6-3 effectiveness of, 8-13
in situ flow pattern depending on, 6-5 gas interference and, 8-4, 8-5
gas gravity, 5-9, 5-10 to 5-13, 5-12f, 6-7 gassy wells and, 8-18
gas gun tubing anchor installation and, 8-13
for flowing gas well, 9-3, 9-4f gassy wells, 8-17, 8-18
in a gas lift well, 10-4f gas velocity, 8-9, 8-11, 11-5
history of developments of, 4-3 to 4-4 gas wells
manually operated, 4-5, 4-6, 4-6f, 4-9 acoustic fluid level measurement survey overview, 9-1
overview of, 2-3 acoustic measurement test summary table, 9-12t
for plunger lift wells, 11-7, 11-7f acoustic survey to identify downhole features, 9-22f
with pressure sensor attached, 7-11f acoustic tests for, 9-8f
recommendations in pumping wells, 8-29 to 8-30, 8-29f analyzing typical performance, 9-5 to 9-19
wireless remote control, 4-8, 4-8f, 4-9f categories of, 9-2

description of acoustic tests for, 9-7 determining acoustic velocity in wells using, 10-6 to 10-
determination of liquid loading in, 9-1 to 9-4 7, 10-7f
determining static bottomhole pressure in, 9-15, 9-18 overview of, 10-5f, 10-21
flowing above critical rate, 9-5, 9-7 packers and, 10-1
flowing below critical rate, 9-9 to 9-11 injection line, background noise and, 10-2
holes in tubing, 9-25 injectivity tests, 1-9
overview of, 9-1 instruments
packer-less, 7-13, 9-26 to 9-28 calibration and maintenance of, 4-18 to 4-19
pressure buildup test and, 7-13 for controlling plunger lift system, 11-2, 11-3, 11-5
principal objective of, 9-1 to detect waves, 3-3
sequence of acoustic records for, 9-13f digital, 1-7
troubleshooting overview, 9-20, 9-22 to 9-25 for a gas lift well, 10-3f, 10-4f
typical acoustic record during shut in, 9-6f hazardous areas and, 4-12 to 4-13
geometric discontinuity, 3-11 to 3-13, 3-11f, 3-13f, 5-15 intrinsically safe, 4-13, 4-13f
Gilbert, W. E., 6-16 maintenance of, 7-26
GOR. See gas/oil ratio (GOR). plunger lift acoustic record, 11-7
gradient correction, 6-16 portable for analysis, 5-12f
gradient inversion portable pressure recording monitor, 10-31f
annular fluid, 8-2, 8-6 to 8-13, 8-7f, 8-8f, 8-9f for pressure transient analysis, 7-2 to 7-3, 7-7f
gravity and, 8-11 proper calibration of, 2-2
protecting, 7-5
hazardous locations classifications, 4-12, 4-13f storing acoustic trace, 5-3
head of the wave, 3-3 strip-chart, 1-7, 4-11, 4-11f
Heriot-Watt University, 3-20 intake valve, pressure at, 6-3
high-frequency waves, 4-1, 4-16, 5-3, 8-21 interference. See gas interference.
high-pass filters, 4-2 internal reflections, 2-24, 9-24, 10-10, 10-11f
horizontal wellbores, 1-21, 2-32f, 2-33, 6-20, 6-22 interval velocity, 10-12
Horner plot, 7-17, 7-17f, 7-19f intrinsically safe certification, 4-13, 4-13f
hot work permit, 4-12, 4-13 intrinsic safety, 4-13
hydrocarbon gases, 3-16, 3-18f, 3-19f IPR. See inflow performance relation (IPR).
isentropic gas law, 3-15
ideal gas, 3-2, 3-15, 3-19 to 3-20 isothermal gradient map, 5-13, 5-13f
ideal gas law, 3-15
implosion pulse Jones equation, 1-19, 1-20f, 1-21
compared to explosion pulse, 4-7 to 4-8, 4-7f
for flowing gas well, 4-9 kelly bushing correction (KB correction), 2-1, 5-9
in a gas lift well, 10-2, 10-5f kick, liquid level, 2-3
in a gas well, 9-3 kicked well, 8-13
history of development of, 4-3, 4-4
kick-over tool, 10-23
overview of, 4-5, 4-5f
kill fluid, 1-9, 10-2, 10-16 to 10-17, 10-26f, 10-34
inclined wellbore, 1-21
inflow performance relation (IPR). See also Vogel’s Equation.
to determine producing efficiency, 6-1 landed rod pump, 2-8
determining from multi-rate flow test, 1-17 to 1-18, 1-17f, laptops for data acquisition, 4-10, 4-10f, 4-11, 7-3, 7-7f
1-23f lateral wellbores, 6-20, 6-22
determining from one-rate well test, 1-18 to 1-20 linear motion pumping units, 1-17
free gas production and, 1-16 liquid column gradient, 7-6, 10-14, 10-21
other performance models, 1-20 to 1-21, 1-20f liquid column pressure, 7-5
pressure below bubble point, 1-14 to 1-17 liquid holdup, 6-9, 6-11
productivity and, 1-24 liquid level
programmed fluid level surveys and, 7-2 converting travel time to distance to, 5-1
sample plot, 1-13, 1-13f determination of depth to, 2-35, 9-4
in single and two-phase flow, 1-16f effect of chamber pressure on, 4-16f
well performance and, 1-1, 1-10 to 1-14 gas lift wells, 10-16
inflow water cut, 7-5 kick from, 2-3
injection gas not at tubing intake depth, 9-28
changing requirements for, 10-34 plunger lift wells and, 11-16
depth profile, 10-12, 10-12f recommendations for determining, 8-30
determination of depth for, 10-25f unclear echo and, 2-3

Index I-7
liquid level depression test. See also Walker test. reading acoustic record and, 3-6
beam-pumped wells, 7-8f recommendations for reducing, 2-3
to confirm annular gradient inversion, 8-10 to 8-13 non-upset tubing joints, 5-10
correlations for finding gradient and, 6-16 to 6-18 “N” type wave, 3-3
data points for, 6-10, 6-11f
to determine effective oil fraction, 6-17f, 6-19f obstructions, echoes generated by, 3-8
estimating gaseous column gradient and, 6-14f offshore platforms
for a gas well, 9-14 deviation in, 10-21 to 10-22
history of, 6-16 KB correction and, 2-1
during pressure buildup test, 7-6 random noises during acoustic tests, 10-4
pumping wells, 8-12f schematic and acoustic trace of, 9-15, 9-16f, 9-17f, 9-18f
liquid level echo, 2-3, 2-18, 2-27 subsurface safety valves in, 2-25
liquid level marker (LL), 2-28, 8-15 oil, gravity of, 6-20
liquid level survey oilfield facilities, hazards of, 4-12
automatic digital filtering for collar count and, 5-6 to 5-7, oil/water ratio, 6-24f
5-7f, 5-9t oil/water separation, 6-4
calculating acoustic velocity from gas, 5-10 to 5-13 one-dimensional wave equation, 3-15
collar count method, 5-2 to 5-4, 5-9t one-rate well test, 1-18 to 1-20, 1-20f
stepwise collar echo count, 5-4 operating conditions, changing for liquid level, 2-3
liquid loaded gas wells, 9-3, 9-25, 9-26f operating costs
liquid pump fillage, 8-5 acoustic fluid level measurements and, 3-1
liquid slugging, 2-33, 11-5 artificial lift efficiency and, 1-23
low-frequency waves, 4-1 energy supplied and, 1-2, 1-4
low-pass filter, 2-8 to 2-9, 2-18f, 2-29f, 5-3 gas lift wells, 11-1
low-pressure wells, 5-2, 5-10 overall efficiency and, 1-2
lubricator, plunger lift well, 11-2 operating procedures, acoustic fluid level survey, 4-13 to
4-19
mandrels. See gas lift mandrels. operating valve, 10-1
manual pulse generation, 4-5 to 4-6 oscillographic recordings, 4-2
microphones
history of, 4-2 to 4-3 packer in the gas lift well, 10-1
output of, 4-8 PAP plunger, 8-19, 8-20f
overview of, 4-10 to 4-11, 4-19 paraffin, 3-8, 8-13, 8-19 to 8-20, 8-21 to 8-22
sensitivity of, 4-18 to 4-19 particle velocity, 3-8, 3-12
in the typically operated gas gun, 4-6, 9-3 patents, oilfield applications, 4-2
mixture density, 6-18 PBHP. See producing bottomhole pressure (PBHP).
multi-phase flow, 6-9, 6-11 PC pump. See progressing cavity (PC) pump.
multi-rate flow test PDP. See pump discharge pressure (PDP).
inflow performance relation (IPR) and, 1-17 to 1-18, 1-17f, peak of the wave, 3-3
1-23f peak-to-peak amplitude, 3-2, 3-11, 3-22, 4-16, 5-4
monitored with acoustic fluid level records, 1-21 to 1-22, perforated zones, 8-7
1-21f, 1-22f perforations, 2-15, 2-15f, 2-16f, 6-2, 11-17
Phillips Natural Gas and Gasoline Department, 3-17
National Electrical Code (NEC), 4-12 PIP. See pump intake pressure (PIP).
National Fire Prevention Association, 4-12 pipe joint length, 5-15
National Institute for Petroleum and Energy Research plane wave, 3-3, 3-7 to 3-11, 3-7f, 3-23
(NIPER), 1-21 plunger lift wells
National Institute of Standards and Technology (NIST), 3-20 acoustic monitoring of, 11-6 to 11-7
natural gas separator, 8-5 anomalies in data, 11-13
needle-type valves, 2-3, 8-30, 9-4, 10-2, 10-4f data acquisition, 11-11 to 11-16, 11-12f
nitrogen gas, injecting to pressurize wellbore, 2-27, 4-15 determining appropriate cycle time, 11-1
noise. See also digital filters. determining plunger position and velocity, 11-6 to 11-8
background, 5-6, 5-15, 8-26, 10-2 to 10-3 field example, 11-20 to 11-21, 11-21f, 11-22f
gas flow, 4-18 gas properties, 11-16 to 11-17
high-frequency, 2-18, 2-25 monitoring of plunger position, 11-7f, 11-8 to 11-9, 11-8f,
obtaining second record because of, 2-2 11-10f, 11-11, 11-11f
pumping-related, 4-17, 4-17f operation cycle, 11-5 to 11-6, 11-6f
random, 4-16 to 4-17, 10-4, 10-6, 10-6f overview of, 11-1, 11-2f

plunger fall velocity, 11-6, 11-13 to 11-16, 11-13f, 11-14f, example field data, 7-14 to 7-23
11-15f gas lift wells, 7-13
systems for, 11-2 to 11-3, 11-3f gas wells, 7-13
troubleshooting, 11-17 to 11-20 instruments for, 7-7f
types of plungers, 11-3, 11-4f lack of, 6-24
unloading sequence, 11-5 overview of, 7-1
POC. See pump off controller (POC). programmed survey and, 7-2 to 7-3
portable pressure recording monitor, 10-31f recommended procedures and implications, 7-6
pressure special requirements for, 7-3 to 7-4
acoustic velocity and, 3-16 to 3-20, 5-1 technological advances and, 7-2 to 7-3
gas gun and, 4-6 in wells with multiple producing zones, 7-13 to 7-14, 7-14f
significant at the wellhead, 4-7 pressure transient analysis, field test data on
sonic velocity and, 3-2 Well A, 7-14, 7-15f to 7-17f, 7-17
at wellhead, below atmospheric pressure, 4-15 Well B, 7-17, 7-18f to 7-19f, 7-20
pressure, volume, and temperature (PVT) tests, 1-10, 3-16, Well C, 7-20, 7-20f to 7-21f
3-16f, 3-19, 3-20, 3-23 Well D, 7-20, 7-22f to 7-23f, 7-23
pressure-balancing ports, 4-3 Well E, 7-23, 7-24f to 7-25f
pressure buildup test pressure traverse, 10-18f
acoustic velocity and, 7-4, 7-4f pressure wave receiving port, 4-4
beam-pumped wells, 7-8 to 7-13, 7-11f producing bottomhole pressure (PBHP)
on ESP and PCP wells, 7-13 bubble point and, 1-15, 1-15f, 7-6
example field data, 7-14 to 7-23 calculating, 1-7 to 1-9, 6-4 to 6-7
gas lift wells, 7-13 comparing to SBHP, 8-22
gas wells, 7-13 determining optimum well performance and, 1-1, 1-4
recommended practices for, 7-6 gaseous liquid column and, 8-22
in wells with multiple producing zones, 7-13 to 7-14 in a gas lift well, 10-16 to 10-17
pressure differential for plunger lift wells, 11-5 high, in wells that kick, 8-17
pressure distribution importance of obtaining, 8-3
annular gas flow rate, 6-18 to 6-20, 6-20f multi-rate flow test for, 1-21 to 1-22, 1-22f
calculating, 6-4 to 6-7, 7-6 overview of, 1-6 to 1-7, 6-3
caused by presence of the tubing anchor, 8-12f plot of, 1-16, 1-16f
depth and, 8-25, 8-25f, 9-10 programmed fluid level surveys and, 7-2
of a ESP well, 2-19f, 2-21f pump inefficiency and, 8-2
gaseous liquid column and, 6-7 to 6-18 for reaching well potential, 8-1
in a gas lift well, 10-14 to 10-17, 10-20, 10-20t recommended practices, 6-24
in gas well without packer, 9-27f schematic of, 1-11f
history of calculating, 6-1 using to determine well performance, 1-23
liquid level depression test and, 6-11 to 6-16 in a Walker test, 6-10 to 6-11, 6-14f
obtaining data for, 1-10 producing wells, 2-15 to 2-16
overview of, 6-3 producing zones, wells with multiple, 7-13 to 7-14, 7-14f,
in a plunger lift well, 11-8 8-23 to 8-26, 8-24f
pressure traverse and, 9-10, 9-10f production. See well productivity.
pumping wells and, 6-2 to 6-3, 6-2f, 8-7f production flow rate, 1-4
quality control of data, 6-21 to 6-24 productivity index (PI)
in a shut-in gas well, 9-19f calculating, 1-13
in a static well, 6-23f changing pump displacement rate and, 1-14
summary of deviated well, 2-11f data for, 1-14
summary of simple well, 2-7f overview of, 6-1 to 6-2
tapered tubing, 2-13f well performance and, 1-1
typical, 8-24f programmable logic controllers for plunger lift wells, 11-5
wells with multiple producing zones and, 8-23 to 8-26 programmed fluid level surveys, 7-2 to 7-4
pressure drawdown, well performance and, 1-1 progressing cavity (PC) pump
pressure pulse, 4-5 calculating producing bottomhole pressure in, 1-7
pressure sensors, 4-9, 7-10f field data, pressure buildup tests, 7-17, 7-18f to 7-19f, 7-20
pressure set point for plunger lift wells, 11-5 fluid level surveys in, 8-26, 8-28f
pressure transducer, 4-11, 9-3, 11-7, 11-7f pressure transient analysis in, 7-13
pressure transient analysis pulse generation system
beam-pumped wells, 7-8 to 7-13 collar count method and, 5-3, 5-7
on ESP and PCP wells, 7-13 designing, 4-1

Index I-9
history of, 4-2 to 4-4 random noise, 10-4, 10-6, 10-6f

overview of, 4-5f random signals, 5-15 to 5-16
pump, varying speed of, 2-2 real gas law, 6-18 to 6-19
pump capacity, excessive, 8-3 to 8-4 recording instrumentation, protecting, 7-5
pump discharge pressure (PDP), 1-3 to 1-4 reflection
pump displacement capacity, 8-1, 8-4, 8-13, 8-19f at the discontinuity of fluid properties, 3-7 to 3-11, 3-7f
pump-down, 2-8, 2-10f from downhole anomaly, 9-15
pumped off well, 1-20, 8-10 gas lift valves and, 10-10
pump effective displacement, 8-2 at a geometric discontinuity, 3-11 to 3-13, 3-11f, 3-13f
pump fillage, 8-5 overview of, 3-7
pumping rate, 7-2 polarity of, 10-8
pumping wells. See also sucker rod lift. reflection coefficients, 3-8, 3-11, 3-12 to 3-13, 3-23
background noise and, 5-15 regulatory agencies, acceptance of acoustic survey results,
calculating producing bottomhole pressure in, 1-7 3-1, 6-23
choked pump, 8-5, 8-6f reservoir pressure, 1-13, 1-19, 1-24
classification of, 6-5 to 6-6, 6-5f reservoirs, depleted, 2-5, 6-5
comparing to downhole pressure sensor measurement, resonance, 4-14, 4-15f
8-23 to 8-26 resonant tube, 4-2
field data, pressure buildup tests, 7-20, 7-20f to 7-21f resonating cavity, causing a noisy record, 2-18f
fluid and pressure distribution in, 6-2 to 6-3, 6-2f rod couplings, 8-14, 8-14f
fluid level surveys in ESP wells, 8-21 to 8-22, 8-21f rod-pumped wells. See also pumping wells.
fluid level surveys in PC wells, 8-26, 8-28f accurate interpretation of fluid level data, 8-30
gradient inversion in, 8-6 to 8-13 background noise and, 5-15
inefficient pump displacement, 8-18 to 8-19 creating acoustic noise, 4-17, 4-17f
misleading acoustic level surveys in, 8-5 to 8-13 downhole marker analysis, 5-8 to 5-9
noise generated by, 4-17 gradient inversion and, 8-7
overall efficiency of, 1-2 PBHP in, 1-7
overview of, 8-1 to 8-2 production efficiency in, 8-2 to 8-3
PC pumps and, 8-26 round trip travel time (RTTT). See also acoustic pulse.
production efficiency in, 8-2 to 8-5 accuracy of, 5-15
recommended procedures and implications in, 8-29 to acoustic fluid level measurements analysis and, 2-1
8-30 acoustic velocity and, 3-23
tubing diagnostic acoustic surveys in, 8-13 to 8-18 correlation function of acoustic velocity versus, 2-35,
pump intake, 6-5, 8-4 2-35f, 10-12, 10-13f, 10-14, 10-27
pump intake pressure (PIP) function of in gas lift wells, 10-7 to 10-8
calculating, 1-4, 1-8, 6-3, 7-6 gas wells and, 9-4
compared to the dynamometer reading, 8-4 knowledge of the acoustic velocity of gas for calculating,
discrepancies in, 8-25, 8-25f 3-20
in ESP well, comparing to acoustic level survey, 8-23 to overview of, 5-1
8-26 pulse generation system and, 4-8
percent of error in, 6-22 in a simple wellbore, 2-5
recommended practices, 6-24 software calculation mismatch, 10-7 to 10-8, 10-8f
unstabilized flow and, 6-21, 6-21t software for, 1-10
in a Walker test, 6-10 to 6-11 in a typical pumping well, 6-3
pump liquid fillage, 8-3, 8-26
pump-off, 8-3 to 8-4, 8-3f safety considerations
pump off controller (POC), 6-4 acoustic fluid level measurements in gas wells, 9-3
pumps in acquisition and recording of acoustic records, 4-11 to
calculating usefulness of, 1-3 to 1-4, 1-3f 4-13, 4-15, 4-19
choked, 8-5, 8-6f hazardous locations classifications, 4-13f
efficiency of, 1-3, 1-3f history of, 4-2, 4-3
excessive capacity in, 8-3 to 8-4 in pumping wells, 8-29
inefficient displacement and, 8-18 to 8-19 remotely fired gas gun, 4-9
leaking, 8-13 safety valve
linear motion, 1-17 acoustic records and operation of, 2-25
variable speed drives in, 1-13, 1-17, 1-23 correctly operating SCSSSV, 2-26f
pump submergence, 3-1, 6-1, 6-3, 8-1, 8-2 incorrectly operating SCSSSV, 2-27f
PVT (pressure, volume, and temperature) tests. See pressure, subsurface downhole, 2-25, 2-26f
volume, and temperature (PVT) tests. “S” curve, 6-18, 6-19f, 6-20, 9-11, 9-14, 9-14f

sensors, technology and, 4-10 for reaching well potential, 8-1

sharp pulse, 4-2 recommended practices, 6-25
Shell Oil Company, 6-16 schematic of, 1-11f
shot using to determine well performance, 1-1, 1-4, 1-23
ESP well casing, 2-15 to 2-16 static fluid level, 1-6, 2-20
tubing, 8-14, 8-15, 8-15f static reservoir pressure, 8-1
undetected, 2-3 static wells, 6-22, 6-23f, 6-24
shut-in wells steady state of well performance, 1-6, 1-7
acoustic fluid level record in, 2-20, 2-20f stepwise collar echo count, 5-4, 5-5f, 5-6f, 5-9t
casinghead pressure and, 6-5 stratified annular gas column, 2-4f, 2-37f
foam layer, 6-7 strip-chart instruments, 1-7, 4-11, 4-11f
formation fluids and, 1-9 sucker rod lift, 8-1, 8-13. See also pumping wells.
gaseous liquid column and, 7-6 superficial gas velocity, 6-18
gas lift wells, 10-19f superficial velocity, 6-9
gas wells, 9-3, 9-6f, 9-20f surface casing pressure, 7-6
plunger lift wells during, 11-5 surface controlled subsurface safety valves (SCSSSV), 2-4f,
producing bottomhole pressure (PBHP) and, 1-16 2-25, 9-15, 9-18 to 9-19, 9-20f
SBHP estimation during, 1-9 surface pressure, 2-2, 6-3, 7-4
static fluid level tests during, 7-2
side-pocket mandrel, 10-23, 10-23f “T” (tube) waves, 4-4
signal acquisition. See acoustic signal. tail of the pulse, 3-5
signal-to-noise ratio, 2-3, 5-7, 5-15 TAM. See Total Asset Monitor (TAM) software.
single-phase liquid flow, 1-12, 1-16f, 6-6 tapered tubing, 2-4f, 2-10
single-rate well test. See one-rate well test. technology
software acoustic fluid level measurements and, 4-10, 8-1
for acoustic fluid level measurements, 1-10, 1-11f, 4-10 in automatic fluid level surveys, 7-2 to 7-3
mismatch in RTTT and distance, 10-7 to 10-8, 10-8f BHP calculations and, 7-1
for plunger lift wells, 11-11 digital filters, 5-3
sound speed and PVT properties, 3-20 for oilfield applications, developments in, 4-2 to 4-4
solid obstructions, echoes generated by, 3-8 pressure distribution and, 6-1
sonic echoes, 4-3 temperature
sonic velocity acoustic velocity and, 2-34 to 2-35
computation of, 5-10, 5-12 bottomhole, 5-13
defined, 3-1 calculating acoustic velocity and, 5-10, 5-11f, 5-12, 5-12f
in a hydrocarbon gas, 3-19f casing pressure and, 7-4
propagation of, 3-2 causing variations in acoustic velocity, 5-1, 10-12, 10-12f
sound pressure wave propagation, 3-13 to 3-15 effect on acoustic velocity, 3-16 to 3-20
sound pulse, generation of, 3-1 to 3-2. See also acoustic pulse. output signal of pressure transducers and, 7-4
sound source, installation of, 4-14, 4-14f. See also gas gun. PC pumps and, 8-26
Southwest Petroleum Short Course (SWPSC), 6-16 sonic velocity and, 3-2
specific acoustic impedance, 3-8 using regional temperature gradients, 5-13f
specific gravity of gas, 3-17, 3-18f, 3-23, 9-4 variations of during pressure transient test, 7-5f
spectrum analysis, 5-3 test point, 1-21
speed drives, variable, 1-13, 1-17 thermal insulator, 7-11f
speed of sound, 3-4, 3-15 to 3-16, 3-20, 5-4 thermistor, 7-4
stabilized pumping operation, 6-4 thermodynamic properties of variables, 3-2, 3-8, 3-15, 5-10
stabilized reservoir pressure, 6-22, 7-2 three-phase flow. See multi-rate flow test.
standing valve, 6-3, 10-1 timers, for plunger lift wells, 11-3, 11-5
static bottomhole pressure (SBHP) Total Asset Monitor (TAM) software, 1-10
acoustic fluid level measurements and, 6-23 transmission at a geometric discontinuity, 3-11 to 3-13, 3-11f,
calculating, 1-9 to 1-10, 6-22 to 6-24, 6-23f 3-13f
comparing to PBHP, 8-22 transmission coefficients, 3-8, 3-11, 3-23
determining for gas wells, 9-15, 9-19f true vertical depth (TVD), 2-20, 6-6, 6-22
example on obtaining, 2-20 tubing. See also tubing holes.
gas lift wells configurations and, 10-14 to 10-16, 10-16f acoustic surveys to troubleshoot, 11-17 to 11-18, 11-18f
importance of obtaining, 8-3 acquisition of fluid records in, 8-14
operators hesitation in surveying, 8-3 changes in diameter of, 3-10
overview of, 1-6, 6-3 diagnostic acoustic surveys overview, 8-13 to 8-18

Index I-11
different diameters and pressures, 6-6 damaged, 7-10f

gas flow and wells that kick, 8-17 to 8-18 exposed, 10-22
gassy wells and, 8-18 gas lift wells and, 10-14, 10-21f, 10-22, 10-24, 11-2
internal reflections down, 10-10 to 10-11, 10-10f, 10-11f history of developments of, 4-4
pressure, relation to casing pressure, 10-10f, 10-16, 10-.17 needle-type, 2-3, 8-30, 9-4, 10-2, 10-4f
to 10-18 pressure discharge, 4-8
sound pressure wave propagation in, 3-2, 3-13 to 3-15 quick-opening, 4-5, 4-5f
tally of, 6-3 spacing of, 10-9
unloading, 8-17 to 8-18 valve spread, 10-22
tubing anchor, as downhole marker, 5-8 vapor phase, 3-20
tubing anchor-catcher (TAC), 8-7, 8-9, 8-11, 8-12f, 8-13 variable speed pumps, 1-22
tubing collars, 2-5, 2-6f, 2-7f, 2-19f venting gas, 7-6
tubing-fluid-pressure-operated, 10-21 Vogel’s Equation. See also correlations; inflow performance
tubing holes relation (IPR).
acoustic records to detect, 8-13, 10-15f compared to multi-rate test, 1-19f
corrosion, 9-23f ERCB standards and, 6-3
detecting, 9-25, 9-28 overview of, 1-18 to 1-21, 1-18f, 1-24
determining depth to the hole, 8-15 to 8-17, 8-15f, 8-16f wellbore pressure distribution and, 6-2
determining in ESP well, 2-20 to 2-24 well performance and, 1-1
gas lift wells, 10-2, 10-14
masked by liquids, 8-14 Walker test. See also liquid level depression test.
misdiagnosed as liquid loading, 9-25 acoustic data sample from, 6-15f
plunger lift wells, 11-19, 11-19f, 11-20, 11-20f analysis of, 6-16f
pressure integrity test to verify, 9-23 calculating amount of liquid in the annular gaseous
typical acoustic record of, 2-4f, 9-29f column, 6-22
wave path diagram, 9-24f calculating gaseous column gradient with, 6-8f
tubing intake, liquid level not at depth of, 9-28 casing pressure and, 6-12t
tubing joints. See also collar count method. difficulties in performing, 6-14
average joint length, 5-2, 5-7 height of gaseous column and, 6-13
calculating distance from surface to liquid level with, 5-2, overview of, 1-8 to 1-9, 1-8f, 6-7 to 6-9
5-6 to 5-7 during pressure distribution calculations, 7-6
example of, 2-14, 2-34, 2-34f procedure for, 6-10 to 6-16
signals from, 3-10 wellhead arrangement for, 6-9f, 6-10
tubing pressure, 11-13 water cut, 1-2, 1-2f, 6-23 to 6-24
tubing-pressure-actuated system, 10-22 water holdup, 1-7
tubing shot, 8-14, 8-15 water/oil interface, 6-20, 6-21, 7-14
tubing stop, 11-3 water/oil ratio (WOR)
tubing taper, 2-12f, 2-23f behavior of during after-flow, 1-9, 7-6
TVD. See true vertical depth (TVD). gas lift wells, 10-14, 10-19
two-pen pressure-recorder, 10-31, 10-32f in a stabilized pumping well, 1-6, 6-4
two-phase liquid flow, 1-12, 1-16f, 1-18, 2-33, 8-26 in a static well, 6-24
wave, 3-1, 3-2, 3-3
Underwriters Laboratory (UL), 4-13 wave equation, 3-1, 3-2 to 3-6, 3-3f, 3-15
United States Geological Survey, 5-13 waveform, 3-1, 3-3, 3-5 to 3-6, 3-5f
United States Patent and Trademark Office (USPTO), 4-2 wave propagation, 3-1 to 3-2, 3-5 to 3-6, 3-5f, 3-10 to 3-11,
University of Oklahoma, 3-19 to 3-20 3-23
unloading, tubing, 8-17 to 8-18. See also well unloading. wellbore. See also deviated wellbore; wellbore geometry.
unloading valves, 10-1 calculating pressure of, 4-3
up-kick echo, 2-11, 3-3, 3-12, 3-13, 8-15, 8-17. See also distance to known anomaly, 5-8 to 5-9
flow into, 1-15f
echoes.
fluid composition and distribution in, 7-5 to 7-6
U-tubing, 10-8, 10-34 general configuration of, 6-5 to 6-6, 6-6f
horizontal, 1-21, 2-32f, 6-20, 6-22
vacuum, annulus at, 2-27 inclination of, 6-22
vacuum, sonic waves and, 3-1 integrity problems in, 10-2
valley of the wave, 3-3 mechanism of flow into, 1-11f, 1-12
valves pressure calculation of, 6-6 to 6-7
background noise and, 5-15 typical pressure and flow distribution, 8-26, 8-27f
configurations of, for gas lift, 10-16 to 10-17, 10-16f wellbore completion data

acoustic velocity and, 9-4 inflow performance relation (IPR) and, 1-10 to 1-22
determining distance to liquid level and, 5-15 necessary information for analysis of, 1-4 to 1-5
downhole marker analysis and, 5-8 to 5-9 requirements for visualization of, 1-23
of an ESP well, 2-21f stabilized conditions for determining, 1-23
gas wells, 9-17f well tests, 1-5 to 1-10
general configuration of, 6-5 to 6-6 well potential
importance of, 2-2, 4-19, 6-3, 8-2 analysis of, 6-1 to 6-2
offshore gas wells, 9-15, 9-16f designing the artificial lift and, 1-2
overview of, 4-15 ESP well, 2-20
schematic of, 2-17f overestimating, 1-17
stratified annular gas column, 2-32f producing near, 8-1
wellbore diagram pump displacement capacity and, 8-4
casing integrity test, 2-28f, 2-31f SBHP and, 1-6
flowing gas well, 2-26f well productivity
multiple producing zones, 7-13 analysis of ESP wells and, 8-22f
overlaying on acoustic record, 2-1 deviated wellbore and, 2-10, 2-11f
overview of, 1-4 to 1-5 efficiency in a rod-pumped well, 8-2 to 8-3
wellbore geometry. See also wellbore. hole in tubing and, 9-25, 9-25f
complex, 7-26, 8-31 identifying what is limiting production in, 8-2
critical rate as function of, 9-2 knowledge of lacking in oilfields, 8-3
determining distance to liquid level and, 5-15 lack of pressure transient tests and, 7-1
fluid distribution and, 7-5 liquid loading and, 9-1
importance of knowledge of, 1-4, 2-1, 6-20, 8-2 loss of during SBHP test, 1-9, 7-2
wellbore radius, 1-12 loss of during variable rate flow test, 1-17
wellbore schematic. See wellbore completion data. overestimating, 1-17, 1-24
wellbore storage effect, 1-9 rate of, 1-12
wellbore summary report reasons for inefficiencies in pumping wells, 8-1
annular pressure distribution, 8-24f simple well, 2-7, 2-7f
casing integrity test, 2-31f tapered tubing well, 2-13 to 2-14, 2-13f
general example of, 2-7f, 2-10, 2-10f, 2-13, 2-13f wells
pressure distribution, 2-19f accepted concepts in production, 1-7
producing and static fluid levels, 2-21f classification of, 6-5 to 6-6, 6-5f
stratified annular gas column, 2-37f gassy, 8-18
well configuration data. See wellbore completion data. keeping pumped off, 1-20
well drawdown, 1-6 with multiple producing zones, 8-7, 8-23 to 8-26
well efficiency, 1-2, 1-2f, 1-3, 1-3f operated intermittently, 8-30
wellhead over-pumped, 8-3, 8-3f
conditions at, 2-27 potential of, 1-13 to 1-14, 1-23, 8-3
connecting a gas gun to, 8-27f preparation of, 4-14 to 4-15, 4-14f
distance to gas gun, 2-3 that kick, 8-17 to 8-18
gas samples taken from, 3-7, 5-10 well spacing, 1-12
wellhead pressure well stimulation, 1-24
background noise and, 5-15 well unloading, 10-2, 10-24, 10-26 to 10-34, 11-5. See also
high, 4-9, 4-19 unloading, tubing.
low, 4-15 wireless communications, advances in and acoustic signal
obtaining acoustic signal and, 3-22 acquisition, 4-10
variations in, 7-12f
wireline-conveyed equipment, 1-9, 7-1
well performance
wireline flowing pressure survey, 9-1, 9-10
analyzing from acoustic fluid level measurements, 1-4 to
1-5 wireline log, 5-12, 6-3
defined, 1-1 wireline retrievable valve mandrel, 10-23
determining optimum and current, 1-1 to 1-4 WOR. See water/oil ratio (WOR).
effect of tubing anchors on, 8-13
efficiency and, 1-2 to 1-3, 1-3f zero net liquid flow, 6-5, 6-6, 6-18, 8-26, 9-1, 9-9
importance of SBHP and PBHP in, 1-4 zones, multiple producing, 8-27f

To obtain additional training materials, contact:
PETEX®
PETROLEUM EXTENSION
J.J. Pickle Research Campus
10100 Burnet Road, Bldg. 2
Austin, TX 78758
Telephone: 512-471-5940
or 800-687-4132
FAX: 512-471-9410
or 800-687-7839
Visit our website: petex.utexas.edu
To obtain information about training courses, contact:
PETEX®
HOUSTON TRAINING CENTER
4702 N. Sam Houston Parkway West, Suite 800
Houston, TX 77086
Telephone: 281-397-2440
or 800-687-7052
Visit our website: petex.utexas.edu/our_courses


Acoustic Fluid Level Measurements-1

Uploaded by

Copyright:

Available Formats

Acoustic Fluid Level Measurements-1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Acoustic Fluid Level Measurements-1

Uploaded by

Copyright:

Available Formats

What is the purpose of the document?

What is the purpose of the document?

What topics are covered in the document?

What topics are covered in the document?

For distribution by Petroleum Extension-The University of Texas at Austin

For distribution by Petroleum Extension-The University of Texas at Austin