The Experts below are selected from a list of 60 Experts worldwide ranked by ideXlab platform

Mark William Mahoney - One of the best experts on this subject based on the ideXlab platform.

  • Pitfalls in Performance-Data Tracking of Sucker-Rod Pumped wells
    All Days, 2006
    Co-Authors: Mark William Mahoney
    Abstract:

    Abstract Mean time between failure (MTBF) has been the standard for tracking field performance of sucker rod pumped wells but it can lead to improper conclusions and applications. A new method using cycle time and loading can greatly improve the understanding of the true performance of sucker rod pumped systems. Using proper data tracking of system performance and wear based on cycles and magnitude can greatly increase the understanding of a well system and be a proper guide to optimization. Introduction Mean Time Between Failure (MTBF) has been used for so long that it is accepted without question and sometimes can cause data to be skewed and misapplied. The primary downfall of MTBF is that most of the fields operating today have a wide variety of operating conditions from well to well and also go though optimization over time. For example at 10 SPM a sucker rod pumping system will complete 14,400 reversals per day multiply that by 365 and you have 5,256,000 cycles a year. This means that the rods have stretched and contracted, the pump barrel and tubing have had the Fluid load shifted on and off of them and almost all of the other pump parts have been subjected to stress reversals and wear over five million times in a year. Strokes per minute, different stroke lengths, timers and well controllers as well as a variety of different pumps to meet operation needs, make it difficult to apply days in operation as the primary decision making criteria. Cycle life should be the primary baseline for data and magnitude of reversals should be tied in. All of the components in a sucker rod pumping system have a lifetime but it is measured in cycles (reversals) and the magnitude of those reversals. Reversals and the magnitude of the reversals dictate the normal life of the sucker rods, sucker rod pumps, tubing, polished rods, etc. Understanding sucker rod pump dynamics Sucker rod pumps operate by displacement. On the upstroke the traveling valve closes and lifts a column of Fluid equal to the cross-sectional area of the plunger. At this time the Fluid load is on the plunger and sucker rods. During the downstroke the traveling valve opens when pressure in the pump chamber is greater than the hydrostatic load on the top of the valve and the Fluid load is then transferred to the tubing string, pump barrel and standing valve. This constant cycling of the Fluid load and the magnitude of the change is the main constant factor in the failure of sucker rod pumping systems. Other factors such as abrasion, corrosion, gas Pound and Fluid Pound contribute and accelerate failures but the one constant is cyclic loading. This load is based on the gross calculated Fluid load lifted by the pump (F) minus the pump intake pressure (PIP) to give the net load or Fo load. This Fo load can be incorporated into the cyclic loading to give an idea of the normal cycle life of sucker rod pump parts. Cyclic loading verses MTFB Most records will be in days of operation and this data should be converted to cycle to failure (CTF) if possible to keep from skewing data. For example lets look at group of 10 wells with 5 of them operating at 16 strokes per minute and the other five operating at 8 strokes a minute. When added up and averaged in days for run time, they will look something like the chart in table 1. If we look at the run time average the conclusion would be that group 1–5 is not performing anywhere near as well as the group 6–10 (183.8 days vs. 314.8) and if they were averaged together the MTBF would drop to 249.3 days. If we looked at these two groups in CTF the reality would look different. Group 1–5 averaged 4,234,752 CTF and group 6–10 averaged 3,626,496 CTF so a completely different picture from the MTBF model. So using CTF alone can be a big improvement over tracking the real wear and tear on equipment. If we also look at time clocking, pump-off controller cycles and well manger units that are tied to variable slippage drives, then the MTBF model is even worse. Simply stated would you buy a used car based on the statement that "it is only 2 years old" or would you want to see the mileage?

  • Laboratory-Instrumented Sucker-Rod Pump
    Spe Production & Facilities, 2003
    Co-Authors: A.l. Podio, Jaime Gomez, Benny J. Williams, Mark William Mahoney
    Abstract:

    Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (≥100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions.

A.l. Podio - One of the best experts on this subject based on the ideXlab platform.

  • Laboratory-Instrumented Sucker-Rod Pump
    Spe Production & Facilities, 2003
    Co-Authors: A.l. Podio, Jaime Gomez, Benny J. Williams, Mark William Mahoney
    Abstract:

    Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (≥100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions.

  • Laboratory-Instrumented Sucker-Rod Pump
    SPE Production & Facilities, 2003
    Co-Authors: A.l. Podio, A.j. Mansure, Jaime Gomez, Benny Williams, Mark Mahoney
    Abstract:

    Summary Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions. Introduction The oil and gas industry continues to rely on sucker-rod pumping systems as the principal method of artificial lift for fields in which the reservoir pressure has been depleted and development of economic flow rates requires drawing the bottomhole pressure to the lowest possible level. Lifting costs represent one of the major operating expenses in these fields, and, thus, producers should maintain every pumping system at the maximum efficiency with a minimum of downtime and a long time between failures. Analysis of the pumping system's performance relies on surface measurements of load, position, and acceleration of the polished rod, prime mover power, pressures at the casing and tubing heads, and Fluid level in the annulus. These measurements are used to characterize operation of the downhole pump, the rod string, and the surface equipment based on a simplified theoretical description of the dynamics of the Fluid and the mechanical system. Early attempts by Gilbert 1 to verify the validity of this approach resulted in the development of the pump dynagraph, which allowed recording of the pump pressure during a pump stroke by installing a specially designed instrument above the pump plunger. Results of this work were vital for the development of accurate design and analysis procedures, such as the API RP 11L method,2 and the later development of numerical simulators of pump performance3 for design and analysis. To validate modern approaches to modeling the pumping system, the industry undertook a study to verify the calculated results by downhole measurement of rod loads. This resulted in a database of measurements for various wells and conditions that can be used by software developers to verify the formulation and solution algorithms applied in their programs.4,5 This study resulted in significant improvement in understanding mechanical rod forces. However, there is still little understanding of pump performance under downhole conditions, particularly the relationship between the pressure of the Fluid flowing through the pump and the mechanical loads developed during the pump stroke. This is the principal objective of the current study, which involves a two-fold approach - instrumentation of a clear sucker-rod pump in the laboratory followed by the development of an instrumented downhole pump for field testing. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at various field conditions. Laboratory Pumping System The experiments were conducted at the artificial lift facilities of the Petroleum and Geosystems Engineering Dept. at The U. of Texas at Austin. Fig. 1 shows a schematic of the experimental well. The pumping unit and wellhead are standard oilfield components. The pumping unit is a beam-balanced unit, API 16-53-30, driven by a 1-hp electric motor that operates through a mechanical, variable-speed drive. Clear acrylic pipe was used for the casing-tubing vertical wellbore. The inside diameter (ID) of the casing is 5 in., the length is 65 ft, the outside diameter (OD) of the tubing is 3 in., and the length is 50 ft. A plastic container with a 200-gal capacity is used as storage tank for the oil. The oil viscosity used was 3.42 centistokes at 25°C, and the oil specific gravity was 0.81. The distance from the stuffing box to the seating nipple is 49.9 ft. The arrangement of the rod string from the bottom to the top is delineated as follows.The sucker-rod pump plunger = 1.79 ft × 1.765 in.One coupling=0.4 ft × 1.765 in.One 2 ft × 0.625-in. pony rod.One 8-ft × 1.125-in. sinker bar (polished rod).One 4-ft × 0.625-in. pony rod.Four 6-ft × 0.625-in. pony rods.One 8-ft × 1.125-in. polished rod. The sucker-rod pump is a 1:1 replica of a standard API tubing pump. The working barrel, with a length of 4 ft, was constructed from plexiglass to allow the inner regions to be seen. The standing valve was attached to the bottom of the tubing. The steel plunger has a smooth sealing surface with a diameter of 1.765 in., a length of 21.5 in., and a plunger/barrel clearance of 0.003 in.

Mark Mahoney - One of the best experts on this subject based on the ideXlab platform.

  • Laboratory-Instrumented Sucker-Rod Pump
    SPE Production & Facilities, 2003
    Co-Authors: A.l. Podio, A.j. Mansure, Jaime Gomez, Benny Williams, Mark Mahoney
    Abstract:

    Summary Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions. Introduction The oil and gas industry continues to rely on sucker-rod pumping systems as the principal method of artificial lift for fields in which the reservoir pressure has been depleted and development of economic flow rates requires drawing the bottomhole pressure to the lowest possible level. Lifting costs represent one of the major operating expenses in these fields, and, thus, producers should maintain every pumping system at the maximum efficiency with a minimum of downtime and a long time between failures. Analysis of the pumping system's performance relies on surface measurements of load, position, and acceleration of the polished rod, prime mover power, pressures at the casing and tubing heads, and Fluid level in the annulus. These measurements are used to characterize operation of the downhole pump, the rod string, and the surface equipment based on a simplified theoretical description of the dynamics of the Fluid and the mechanical system. Early attempts by Gilbert 1 to verify the validity of this approach resulted in the development of the pump dynagraph, which allowed recording of the pump pressure during a pump stroke by installing a specially designed instrument above the pump plunger. Results of this work were vital for the development of accurate design and analysis procedures, such as the API RP 11L method,2 and the later development of numerical simulators of pump performance3 for design and analysis. To validate modern approaches to modeling the pumping system, the industry undertook a study to verify the calculated results by downhole measurement of rod loads. This resulted in a database of measurements for various wells and conditions that can be used by software developers to verify the formulation and solution algorithms applied in their programs.4,5 This study resulted in significant improvement in understanding mechanical rod forces. However, there is still little understanding of pump performance under downhole conditions, particularly the relationship between the pressure of the Fluid flowing through the pump and the mechanical loads developed during the pump stroke. This is the principal objective of the current study, which involves a two-fold approach - instrumentation of a clear sucker-rod pump in the laboratory followed by the development of an instrumented downhole pump for field testing. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at various field conditions. Laboratory Pumping System The experiments were conducted at the artificial lift facilities of the Petroleum and Geosystems Engineering Dept. at The U. of Texas at Austin. Fig. 1 shows a schematic of the experimental well. The pumping unit and wellhead are standard oilfield components. The pumping unit is a beam-balanced unit, API 16-53-30, driven by a 1-hp electric motor that operates through a mechanical, variable-speed drive. Clear acrylic pipe was used for the casing-tubing vertical wellbore. The inside diameter (ID) of the casing is 5 in., the length is 65 ft, the outside diameter (OD) of the tubing is 3 in., and the length is 50 ft. A plastic container with a 200-gal capacity is used as storage tank for the oil. The oil viscosity used was 3.42 centistokes at 25°C, and the oil specific gravity was 0.81. The distance from the stuffing box to the seating nipple is 49.9 ft. The arrangement of the rod string from the bottom to the top is delineated as follows.The sucker-rod pump plunger = 1.79 ft × 1.765 in.One coupling=0.4 ft × 1.765 in.One 2 ft × 0.625-in. pony rod.One 8-ft × 1.125-in. sinker bar (polished rod).One 4-ft × 0.625-in. pony rod.Four 6-ft × 0.625-in. pony rods.One 8-ft × 1.125-in. polished rod. The sucker-rod pump is a 1:1 replica of a standard API tubing pump. The working barrel, with a length of 4 ft, was constructed from plexiglass to allow the inner regions to be seen. The standing valve was attached to the bottom of the tubing. The steel plunger has a smooth sealing surface with a diameter of 1.765 in., a length of 21.5 in., and a plunger/barrel clearance of 0.003 in.

Jaime Gomez - One of the best experts on this subject based on the ideXlab platform.

  • Laboratory-Instrumented Sucker-Rod Pump
    Spe Production & Facilities, 2003
    Co-Authors: A.l. Podio, Jaime Gomez, Benny J. Williams, Mark William Mahoney
    Abstract:

    Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (≥100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions.

  • Laboratory-Instrumented Sucker-Rod Pump
    SPE Production & Facilities, 2003
    Co-Authors: A.l. Podio, A.j. Mansure, Jaime Gomez, Benny Williams, Mark Mahoney
    Abstract:

    Summary Although sucker-rod pumps are installed in nearly 90% of all oil wells and many gas wells (for liquid unloading) in the United States and have been widely used for decades, there are many issues regarding their hydraulic performance that are not well understood. This is caused by the difficulty of obtaining downhole pump-performance data. Many persistent problems in sucker-rod pumping, including partial pump fillage, gas interference, gas locking, Fluid Pound, sticking valves, rod downstroke compression loading, equipment failure, reduced production, etc., are difficult to diagnose from the surface. Currently, verification of sucker-rod pump problems can only be inferred by removing the pump at great expense. Thus, root-cause analysis depends on guesswork and component analysis. Knowledge of pump characteristics downhole would allow problems to be predicted rather than simply diagnosed after they have persisted long enough to result in failure. To develop a knowledge base on sucker-rod pumps, a two-fold approach is being pursued: instrumentation of a clear sucker-rod pump in the laboratory, followed by the development of an instrumented downhole pump. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at field conditions. A key element to both the laboratory and downhole instrumented pumps is measuring the compression chamber pressure (pressure within the pump barrel). The instrumentation has been designed to collect high-speed (100 samples a second) data so that transient behavior (ball chatter, etc.) can be observed. Data is archived while the pump operates under various conditions, from full to pumped off. This paper presents results of tests with the laboratory pump that have resulted in new insights about pump friction and the techniques developed to measure dynamic and static pump friction. Analyses of the compression-chamber pressure are leading to a better understanding of what happens when both valves are closed and leading to the development of techniques to perform real-time diagnoses that determine fillage and gas locking. The laboratory data showed that compression-chamber data can be insightful in understanding pump conditions. Introduction The oil and gas industry continues to rely on sucker-rod pumping systems as the principal method of artificial lift for fields in which the reservoir pressure has been depleted and development of economic flow rates requires drawing the bottomhole pressure to the lowest possible level. Lifting costs represent one of the major operating expenses in these fields, and, thus, producers should maintain every pumping system at the maximum efficiency with a minimum of downtime and a long time between failures. Analysis of the pumping system's performance relies on surface measurements of load, position, and acceleration of the polished rod, prime mover power, pressures at the casing and tubing heads, and Fluid level in the annulus. These measurements are used to characterize operation of the downhole pump, the rod string, and the surface equipment based on a simplified theoretical description of the dynamics of the Fluid and the mechanical system. Early attempts by Gilbert 1 to verify the validity of this approach resulted in the development of the pump dynagraph, which allowed recording of the pump pressure during a pump stroke by installing a specially designed instrument above the pump plunger. Results of this work were vital for the development of accurate design and analysis procedures, such as the API RP 11L method,2 and the later development of numerical simulators of pump performance3 for design and analysis. To validate modern approaches to modeling the pumping system, the industry undertook a study to verify the calculated results by downhole measurement of rod loads. This resulted in a database of measurements for various wells and conditions that can be used by software developers to verify the formulation and solution algorithms applied in their programs.4,5 This study resulted in significant improvement in understanding mechanical rod forces. However, there is still little understanding of pump performance under downhole conditions, particularly the relationship between the pressure of the Fluid flowing through the pump and the mechanical loads developed during the pump stroke. This is the principal objective of the current study, which involves a two-fold approach - instrumentation of a clear sucker-rod pump in the laboratory followed by the development of an instrumented downhole pump for field testing. The laboratory pump allows the development of diagnostic techniques in which pump performance can be verified visually. The downhole pump will allow testing at various field conditions. Laboratory Pumping System The experiments were conducted at the artificial lift facilities of the Petroleum and Geosystems Engineering Dept. at The U. of Texas at Austin. Fig. 1 shows a schematic of the experimental well. The pumping unit and wellhead are standard oilfield components. The pumping unit is a beam-balanced unit, API 16-53-30, driven by a 1-hp electric motor that operates through a mechanical, variable-speed drive. Clear acrylic pipe was used for the casing-tubing vertical wellbore. The inside diameter (ID) of the casing is 5 in., the length is 65 ft, the outside diameter (OD) of the tubing is 3 in., and the length is 50 ft. A plastic container with a 200-gal capacity is used as storage tank for the oil. The oil viscosity used was 3.42 centistokes at 25°C, and the oil specific gravity was 0.81. The distance from the stuffing box to the seating nipple is 49.9 ft. The arrangement of the rod string from the bottom to the top is delineated as follows.The sucker-rod pump plunger = 1.79 ft × 1.765 in.One coupling=0.4 ft × 1.765 in.One 2 ft × 0.625-in. pony rod.One 8-ft × 1.125-in. sinker bar (polished rod).One 4-ft × 0.625-in. pony rod.Four 6-ft × 0.625-in. pony rods.One 8-ft × 1.125-in. polished rod. The sucker-rod pump is a 1:1 replica of a standard API tubing pump. The working barrel, with a length of 4 ft, was constructed from plexiglass to allow the inner regions to be seen. The standing valve was attached to the bottom of the tubing. The steel plunger has a smooth sealing surface with a diameter of 1.765 in., a length of 21.5 in., and a plunger/barrel clearance of 0.003 in.

J.f. Lea - One of the best experts on this subject based on the ideXlab platform.

  • Application of Artificial Neural Network to Pump Card Diagnosis
    SPE Computer Applications, 1994
    Co-Authors: G.m. Nazi, J.f. Lea, Kaveh Ashenayi, Frank Kemp
    Abstract:

    Abstract Beam pumping is the most frequently used artificial lift technique. Down hole pump cards are used to evaluate performance of the pumping unit. Pump cards can be generated from surface dynamometer cards using one-dimensional wave equation with viscous damping, as suggested by Gibbs. Pump cards contain significant information describing the behavior of the pump. However, interpretation of these cards is very tedious and time consuming. Hence, an automated system capable of interpreting these cards is very useful. This work presents a DOS based computer program capable of correctly classifying pump cards. The program uses a hybrid artificial neural network (ANN) to identify significant features of the pump card. The hybrid ANN utilizes classical and sinusoidal perceptrons. The network is trained using error back propagation technique. The program has been able to correctly identify problems in over 100 different training and test pump cards. The ANN requires total of 80 data points as input. Sixty data points are collected from the pump card perimeter itself and the remaining 20 data points represent the slope at selected points on the pump card perimeter. The problem conditions are grouped into distinct classes such as malfunction anchor tubing, Fluid Pound, gas interference, and worn traveling valve. The network is capable of identifying one or more of these problem conditions for each pump card. Several examples will be presented and discussed. Introduction Sucker-rod pumping system is a very popular means of providing artificial lift for wells. It is estimated that 90 percent of artificially lifted wells use sucker-rod pumping systems. In order to assure proper operation of these systems it is necessary to monitor them on a regular basis. This is done by means of a dynamometer which is attached to the polished rod of the pumping unit. The plot of load versus position generated by the dynamometer is known as a surface card. Using the technique proposed by Gibbs this surface card can be used to obtain down hole "pump card". The pump card presents the load versus position at the pump down hole instead of the load versus position at the polished rod on the surface. The shape of the pump card can be used as a diagnosis tool. For example, when the pump is functioning properly and pumping to its full capacity the pump card will have a rectangular shape. When the pump capacity is not being fully utilized then we may have a condition called "Fluid Pound". The petroleum industry is in the process of automating the control of pumping wells. Computers are being used to continuously monitor the pump and transmit data to a central office for further analysis. The shear volume of data received makes it necessary to develop tools for automatic analysis of the pump cards. Various pattern recognition techniques have been used to analyze the down hole pump card. Traditional techniques require significant processing time. P. 151^

  • Modeling Forces on a Beam-Pump System During Pumping of Highly Viscous Crude
    SPE Production Engineering, 1991
    Co-Authors: J.f. Lea
    Abstract:

    Summary In previous dynamic predictive models of beam-pump performance, drag forceson the rod string commonly are modeled by an input empirical drag coefficientmultiplying the local rod velocity. This paper shows how forces on the rods andpump plunger can be modeled theoretically and calculated for viscous flow. Results of viscous-flow calculations to develop drag coefficients compare wellwith measured field data. Introduction Solutions to the wave equation to describe beam-pump performance usuallyinclude an empirical drag term to account performance usually include anempirical drag term to account for viscous drag on the rods and other forces. Bastian el al. discussed concerns about drag coefficients, primarily fordiagnostic programs. Doty and Schmidt discussed the use of experimentallydetermined drag coefficients. Most drag coefficients used for predictivecalculations are found from comparison of field data to calculated results. They cannot be extrapolated to extreme to extreme conditions. The governingequation of motion (wave equation for tapered rod strings) solved to modelbeam-pump performance with inertial, elastic, and drag terms is: ...............................................(1) where for the upstroke and 0. 15 for the downstroke for many applications. Normally, the Fluids pumped are complex mixtures of gas, oil, and water. Therefore, the practice has been to use the empirical drag coefficients tomodel beam-pump motion. If the produced Fluid is single-phase, however, and theviscosity and the speed of pumping are such that viscous flow is produced inthe pump-plunger/barrel and the rod/tubing annuli, then it produced in thepump-plunger/barrel and the rod/tubing annuli, then it becomes possible tocalculate velocity profiles, shear stress, and the resulting drag overFluid-exposed surfaces. This paper solves the laminar-or viscous-flow equationsof motion in the pump-plunger/barrel, rod/tubing, and coupling/tubing annuli, Once the equations are solved, the pressure and shear-stress distributions canbe found and the resultant forces on a beam-pumping system can be calculated. The governing equations are developed and entered into a dynamic wave-equationmodel for prediction of beam-pump operation. Also, this paper shows how dragcoefficients can be developed for input to existing wave-equation modelswithout program alterations, if the drag term programmed is as in Eq.1. Abeam-pump example of the production of an oil with a viscosity of severalcentipoise is studied. Calculated results are compared with actual dynamometerdata from the example well. Forces Acting on Rod String and Pump for a Beam-Pumped Well Fig. 1 shows a beam-pumping system (rods and downhole pump) during theupstroke and downstroke. Several forces, in addition to the static Fluid load, act on the rods and pump.The rod/Fluid interaction.The rod/tubingsliding friction.The pump/Fluid interaction (between the barrel andplunger).The pump plunger-to-barrel sliding friction.The upward forcecaused by the pressure drop of Fluid flowing through the traveling valve on theplunger downstroke.Fluid Pound caused by incomplete fillage (partly from Force 5).Inertial effects.Other forces, such as stuffing box frictionand rod In this study, Forces 1 and 3 are solved for viscous single-phase flow. Forces 2 and 4 are assumed to be no higher for a high-viscosity situation thanfor a low-viscosity situation, so their magnitudes can be inferred fromcommonly used, current empirical drag coefficients. For the cases analyzedhere, rod/tubing friction forces are considered to be much less thanviscous-Fluid drag forces as long as the assumption holds that high viscositydoes not greatly increase rod/tubing friction. Fluid Pound forces can be largein fast-pumping situations. but pumping speed is very slow for the highlyviscous situations studied here. Incomplete fillage is simulated when necessaryas an input. Inertial loading is neglected with slow pumping speeds. Byrd and Hale made a related study of the rod coupling/piston effect on the load of asucker-rod pumping system. They showed that during the upstroke and downstroke, the average Fluid velocity, past the couplings and tubing is (2a) (2b) for the upstroke, and (2c) and (2d) for the downstroke, where and Fluid velocities relative to couplings andtubing, respectively; rod velocity; = area of pump barrel; = inside-tubingcross-sectional area; and coupling cross-sectional area. The Fluid velocitypast the rods can be obtained by appropriate replacement of the area of thecouplings, Ac, by the area of the rods, in Eq. 2. Examination of theseequations shows that, when the pump plunger is small compared with the tubingand is rising, the rods rise faster than the average Fluid velocity. This givesa downward drag on the rods. When the pump size is near the tubing size, thereis little drag on the rods. When pump size is near the tubing size, there islittle drag on the rods. When the pump size is equal to or greater than thetubing area, then the rods experience a net Fluid drag lifting them. The rodare always in tension, however, because of the large gravity force of the Fluidload on the plunger and the pressure to offset friction at the plunger to raisethe plunger and the pressure to offset friction at the plunger to raise theFluid. The same comments apply to the couplings, except that the Fluid movesfaster past the couplings because of the reduced annulus area between thecouplings and the tubing. Note that, on the downstroke, the Fluid is forced up as the rods go downindependently of the pump size. This, along with pump-plunger/barrel viscousdrag, causes rod fall problems, especially with high-viscosity Fluids. Byrd and Hale derive a drag term for the couplings considering the area jump that occursat the couplings. This derivation is outlined in Appendix A and could be usedwith the model developed here. For the case modeled here, however, the velocityis low and Byrd and Hale's terms for coupling drag are not as important. Byrdand Hale's terms may be considered when large couplings have high-velocitychanges across them. SPEPE P. 420