Copyright 2007, Society of Petroleum Engineers This paper was prepared for presentation at the 2007 SPE Annual Technical Conference and Exhibition held in Anaheim, California, U.S.A., 11–14 November 2007. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, Texas 75083-3836 U.S.A., fax 01-972-952-9435.

Abstract Efforts to develop electrical submersible pumping (ESP) system technology for harsh environment applications has resulted in equipment designed and developed specifically for applications in higher temperature wells. This paper will discuss the development evolution of this equipment through Centrilift’s state-of-the-art high temperature test facility “hot loop” and continuing development by field operations in Canadian SAGD applications. Internal Development: The utilization of a state-of-the-art test facility that can be entered and exited at strictly planned / controlled intervals where temperature, system cycling and system monitoring can be controlled in a safe environment, has been instrumental in advancing the ESP equipment technology to a satisfactory operational level at 200° C – 220o C (392° F – 428o F) environments. This facility allows the equipment component design and materials evaluation to proceed quickly and under controlled procedures, with the end result being a fully tested and evaluated system for field applications. Several equipment modifications based on the test results will be discussed. Field Testing/ Development: The utilization of customer owned producing wellsites for initial application of the extreme temperature equipment is critical for the final ‘live’ well test of the design and tweaking of minor issues that would be difficult to detect under cleaner and controlled lab type environments. Further equipment design focus beyond 250°C will be dependent upon SAGD operators’ targets and requirements. INTRODUCTION: For years the gap between theory and practice has been the fodder for field testing and iterative field development carried out in producing wells throughout the oil and gas industry. It

is industry’s quest for a panacea that has made component level testing insufficient. Component supplier claims are easily evaluated under laboratory conditions, both in-house and through the use of outside laboratories. The testing of new system(s) designs is another issue altogether. In-house development and testing of all phases of Centrilift’s new extreme temperature ESP product offering would have been impossible without the construction and availability of the hot loop test facility located at headquarters in Claremore, Oklahoma. Field testing alone would have extended the development process and delayed product availability. OIL SANDS EXPLOITATION Alberta’s oil sands have now attained global recognition as an important reserve base (Fig. 1). Technology has allowed these reserves to be recognized as proven reserves and to be developed. With approximately 90% of the reserves too deep to mine, Steam Assisted Gravity Drainage (SAGD) is a proven method to is develop the deeper resources. EnCana, one of the largest landholders of this resource, is consistently optimizing its processes. One of the most significant well enhancements has been ESPs, which allow the optimizing the steam oil ratio (SOR) with greater well control and ease of operation. ESPs can run at lower pressures than steam lift or gas lift, which improves economics of the project by lowering the SOR. The high fluid temperatures of up to 218ºC (456ºF) are greater than the elastomers in traditional ESPs can handle. With extensive lab testing and innovative solutions, the high-temperature ESPs are now the preferred lifting method in the majority of SAGD projects. SAGD SAGD was first tested at Alberta Oil Sands Technology and Research Authority (AOSTRA) in the 1980’s (Fig. 2). Multiple pilots have been done since that time and in 2000 EnCana started the first commercial SAGD operation. SAGD consists of two parallel horizontal well pairs, approximately 400 to 1,000m in length, located at the bottom of the reservoir with approximately 5m of vertical separation. The upper well is used to inject steam into the formation, which comes into contact with the oil, condenses and lowers the viscosity of the oil. The resulting oil/water emulsion drains by gravity to the lower well, which produces the emulsion.

SPE 110701

Extreme Temperature ESP Development Steve Tetzlaff, Kelvin Wonitoy, Brad Ward, Lawrence Burleigh - Baker Hughes Centrilift; Adrian Dodds - Encana Oil & Gas Partnership

2 SPE 110701

Improving SAGD Since lower pressure can result in a lower SOR, development of an alternative to the gas lift or steam lift was required. As most of the reservoirs are under balanced, some sort of lifting mechanism is required. SAGD produces too large a fluid volume for traditional rod pumps, so gas lift was used. Gas lift is the injection of gas into the heel of the well to lower the hydrostatic gradient and allow the fluids to be produced to surface. Another method was to overpressure the reservoir so that the fluids would flow. But as field tests have shown, lower pressures result in lower SORs which improve the economics of SAGD, and at lower pressures, gas lift is not efficient. ESPs allow operations to control the fluid rates coming into the plant and optimize the well. The elimination of gas from the system reduces line and vessel sizes, which further reduce the capital requirements of the facility. Since each well can slow or increase its production depending on plant circumstances, there is increased plant control of the system. Starting of a well with an ESP eliminates the geysering that can occur with gas lift. EnCana continues to be the leader in SAGD with one of the lowest SORs in the industry and has installed numerous ESPs within its wells. It is the large scale of operation that has helped with some of the field testing that proves up laboratory results. Increased reliability is what is demanded from all mechanical devices and ESP’s are no exception. The downhole SAGD conditions are some of the most severe, yet with technology, these are being overcome as demonstrated by the field results. ESP TECHNOLOGY IN SAGD SAGD operations can employ a low or high pressure strategy to produce the field. A high downhole pressure strategy typically uses gas lift while a low downhole pressure strategy will use a downhole pump. The advantages of a low pressure operation include:

• Reduced SOR (steam to oil ratio) • More flexible production • Improved safety due to no gas handling on surface • Greater control of the steam chamber since pumps

offer greater flexibility • Gas over bitumen (GOB) production is not at risk • Reduced risk of surface steam blowout • Reduced risk of steam path forming from injector to

producer well • Reduced risk of steam path forming from injector to

gas pocket or water aquifer • Reduced surface equipment for gas handling

ESP systems are particularly well suited for a low pressure strategy (Fig. 3). Downhole pressure has a minimal influence on ESP performance. Standard ESP pumps can handle up to 30% free gas and even greater amounts with a Gas Avoider™, gas separator or gas handling pump stage. In the presence of free gas there are 3 strategies that can be employed:

• Separate the gas with a gas separator (not desirable in a horizontal orientation)

• Handle the gas via increasing pump intake pressure or a pump design that can handle greater free gas

• Avoid the gas In the horizontal orientation typical of SAGD wells, Centrilift’s Gas Avoider intake provides the most reliable solution. The Gas Avoider intake features a perforated tube containing two coiled titanium gas filled tubes. These tubes naturally float upward to partially block the perforations on the top side of the intake, biasing fluid flow in the bottom section of the casing where there is less free gas. The radial journal bearing material in the Gas Avoider has been upgraded to operate in varying fluid conditions. The challenges facing ESPs are:

• Effective gas handling due to gas breakout at lower pressures

• Optimizing execution to be more conventional • Continuously improving run time performance • Wellbore issues

Pump Type ESPs utilize multistage centrifugal pumps to produce fluid. Centrifugal pumps fall under the kinetic pump category as kinetic energy is imparted to the fluid by the rotating impeller. A stationary diffuser redirects the fluid towards the shaft, creating pressure or potential energy. A centrifugal pump will interact with a well to realize an operating point in equilibrium. The rate that a centrifugal pump will produce depends upon the total dynamic head (TDH) requirements of the production scenario. In an oil well this is dictated by the fluid level, the tubing friction, and the surface tubing pressure at the wellhead. A fourth component, velocity head, is negligible in the lower production rates in oil fields. The well productivity index (PI) determines the fluid level. In SAGD the rate of steam injection heavily influences the well PI. Piston, rod, or progressing cavity pumps are examples of positive displacement (PD) pumps. PD pumps produce a given rate at the given RPM or stroke rate. PD pumps use gears, screws, lobes or pistons to trap a fixed amount of fluid and force it to move. ESP systems are in general more suited for SAGD operations due to the distinctive production characteristics of SAGD wells. For example, if steam injection is interrupted or reduced the fluid level decreases and the productivity index shifts. A centrifugal pump will naturally shift to the left on its head flow curve, reducing the rate being produced and partially complimenting the reduced rate of steam injection. A positive displacement pump will continue to pump at the same rate – reducing the fluid level and the pressure and compromising the barrier between the injector and producer wells thus increasing steam breakthrough risk. ESP system development philosophy The ESP system comprises a variable speed drive, three phase power cable, a motor lead extension, a pothead, two pole

SPE 110701 3

squirrel cage induction motor, a seal section, a Gas Avoider, and a multistage centrifugal pump. The variable speed drive increases production flexibility while ensuring more reliable operation of the ESP system. The three phase power cable carries the power from the surface to the downhole system. The motor lead extension adapts the power cable to the pothead, which plugs into the motor providing the power. The motor rotates at 3500 rpm at 60 Hz power, providing the torque for the pump. The seal section has 5 functions: protect the motor via 2-6 labyrinth or barrier chambers, seal the well fluid via mechanical seals, allow for motor oil expansion and contraction, transmit the torque from the motor to the pump, and absorb the pump hydraulic thrust and shaft differential pressure force. The Gas Avoider enhances gas avoidance in a horizontal orientation and the pump produces the fluid. To upgrade the ESP design for extreme temperatures there were five key tactics applied.

• Minimize elastomers – Extreme temperatures, temperature cycling, and free gas combine to create an environment that is not conducive to elastomers available today. The changing environment of the downhole well frequently requires elstomers to flex, further complicating the application.

• Allow for additional oil expansion – As temperatures increase the motor oil will expand proportionately. The design must also allow for surface temperatures nearing arctic conditions prior to installation.

• Allow for additional thermal expansion of mechanical components – Mechanical components grow radially and axially as temperatures increase. All clearances were reviewed and optimized.

• Allow for a wide range of bearing lubricity scenarios – Thrust and journal bearing interfaces in the motor, seal section, Gas Avoider, and multistage centrifugal pump had to be designed to allow for operation at temperatures over 400°F and for start-up temperatures well below that threshold.

• Increase electrical insulation – A common rule of thumb states that for every 10°C (18°F) the operating temperature is increased the insulation life is halved. The insulation system had to be increased accordingly.

LABORATORY VS. FIELD DEVELOPMENT: It was determined that in-house system testing would dramatically reduce the development cycle(s) necessary to produce a reliable system for these extreme temperature applications. The relationship between operating temperature and insulation life indicates that a 240°C (465°F) test for 23 days is equivalent to a 2 year test at 190°C (375°F) for the electrical insulation system. In-house testing allows for hands on operational control and monitoring of the system test characteristics as well as real time evaluation of design conditions in the event of component or system failure. These advantages help to reduce the development cycle time over what can be lengthy field tests. Time delays in design evaluation in the event of system failure and expenses relating to system installation and pulling due to rig availability along

with the costs associated with interrupted production needs to be factored into the decision to develop product in the field as was confirmed during development of geothermal operations. Test Facility: Centrilift’s hot loop development facility consists of a vertical 16” cased hole to a depth of 155 ft. with a 12” inner liner. The facility is fully enclosed and capable of providing bottomhole temperatures (BHT) of over 464o F (240o C) with a 1,750 gallon fluid capacity. The 4” closed flow loop system is production rated up to 4,360 m3/D (27,500 BPD) at discharge pressures up to 3,000 PSI (Fig.4). The loop’s electrical capability was upgraded in 2006 to 1,500 kVA. The downhole assembly is controlled by a standard 520 kVA variable speed drive / transformer set dedicated to the hot loop well. The climate controlled control room houses the emergency shutdown system as well as all equipment used to monitor and collect data while the downhole system is in operation. The data stream for these test runs include 16 points that are collected and stored in 5 minute increments. (Fig. 5) The test durations can vary up to 40 days, depending on the system elements under evaluation. Test datum may include periodic thermal cycling, which can be achieved by either driving the produced fluid temperature down utilizing the heat exchanger or by simply shutting the system down. System shutdowns and restarts are hard on the ESP system but do reflect operational conditions that the system can endure in the field. These data streams are often converted to data graphs for easier evaluation and publication (Fig. 6). Other tests may run the system components for a “Time at Temperature” evaluation. The hot loop temperature can ramp up from a fully cooled condition of 30°C (85° F) BHT to the desired operational BHT for a test in as little as four hours. Cooling the produced water down enough to pull and handle the ESP system can take a day or more, depending on the temperature. All existing phases of high temperature to extreme temperature development of ESP pumps, motors, seal / equalization assemblies and cable products have been developed and tested as a system in this facility. Testing has been initiated on Centrilift’s next generation ultra high temperature product development in this facility. Final design evaluation will be completed following a planned upgrade of the hot loop. The enhancement will allow certification of an ultra temperature ESP system to new projected SAGD operating temperatures. FIELD PERFORMANCE When considering SAGD field performance of ESPs, it is important to include bottomhole temperatures as a part of the discussion. Some SAGD producer wells are as cool as 150°C (300°F). The average SAGD producer well is nearly twice as hot as the average oilfield ESP well. The traditional oilfield well also produces fluid at a relatively steady temperature. Since Centrilift has qualified the system design in an R&D environment, the initial field trials immediately were exposed to average SAGD temperatures over 185°C (365°F). The operator also contributes to successful run times by providing high quality power with minimal THD (total harmonic distortion), a steady fluid temperature, and minimal shutdowns.

4 SPE 110701

The effort placed in advanced R&D development has resulted in no failures attributed to the core Centrilift design. Failures have been attributed to execution – manufacturing, assembly, and installation. Execution is more critical in an environment with fluid temperatures nearly twice that of the typical ESP application. No failures due to abrasive handling or the mechanical components in the motor, seal section, gas avoider, and pump have been observed. One Centrilift ESP system is operating for over 500 days at an average fluid temperature nearing 190°C (375°F). For the electrical system this is equivalent to a 1,000 day run at 180°C (355°F). Another system is operating in a severely difficult environment with a peak fluid temperature nearing 210°C (410°F) and temperature fluctuations of +/- 45°C (80° F) and is still operating trouble-free. CONCLUSION Bitumen production is vital to satisfy the world’s increasing demand for oil as conventional oil supplies decrease. The success of ESP systems at extreme temperatures has contributed to low pressure thermal exploitation of in situ reserves being the preferred strategy for the numerous projects that will begin production over the next five years. Most of the technological developments continue to take place in the Alberta oil sands. Rigorous development in an R&D environment prior to field trials will accelerate the achievement of goals set by both operators and equipment / service providers. Two and a half years of intense design, material testing and system lab testing of pump system prototypes has yielded extreme temperature ESP systems capable of operating under actual field conditions with little to no modification. This provides the end user a reliable production system with minimal impact on operation expense or production losses that typically are experienced during straight “field test” type development. The population of Centrilift systems is growing and run times well over the one year mark have already been achieved. Work is ongoing into the development of even higher temperature pumps and it is sophisticated laboratory capabilities and innovative thinking that drives the relentless pursuit of increased run life to move ESPs to the next temperature level. References 1. Edmunds, N., and Chhina, H., “Economic Optimum Operating Pressure for SAGD Projects in Alberta”, JCPT, December 2001, V 40 No. 12, pages 13-17. 2. S.C. Solanki, R. Bowman, and D. Rowatt, “Steam Assisted Gravity Drainage with Electric Submersible Pumping Systems”, 2005 Gulf Coast Electric Submersible Pump Workshop held in Houston, Texas, April 27- 29, 2005.

SPE 110701 5

Fig. 1: Canadian oil sands

6 SPE 110701

Fig. 2: Typical SAGD well pair

SPE 110701 7

Fig. 3: SAGD ESP system

8 SPE 110701

Hot Loop

Test Well

Storage Tank

Heat Exchanger

Fluid Heater

System Control Valves

Fluid Expansion Chamber

System Fill

PumpInjection Pump

Liquid Level Probes

Fig.4: Hot Loop schematic drawing Date/Time Temp Out Temp In Temp Mot PSI Out PSI In DIS PSI Flow KW KVAR Current A Current B Current C Voltage A Voltage B Voltage C

6/5/2007 9:00 421.95 428.84 92.81 482.55 475.7 1182.0 132 80 69.34 26.03 26.56 27.00 2239.99 2261.96 2248.546/5/2007 9:05 428.44 428.84 92.81 486.53 479.6 1187.83 132 80 69.09 26.12 25.49 26.95 2246.09 2249.76 2268.076/5/2007 9:10 426.28 429.23 93.44 489.77 482.5 1190.76 133 77 68.85 27.64 24.90 26.51 2237.55 2243.65 2249.766/5/2007 9:15 425.49 429.04 92.81 493.25 486.4 1198.09 131 80 69.82 25.24 26.95 27.05 2237.55 2255.86 2258.306/5/2007 9:20 425.30 429.04 94.06 494.75 487.3 1193.70 133 80 69.82 27.39 26.61 27.15 2232.67 2250.98 2259.526/5/2007 9:25 425.30 428.84 94.06 495.24 488.3 1195.16 133 81 70.31 28.66 25.73 27.20 2237.55 2252.20 2260.746/5/2007 9:30 424.90 428.64 93.44 494.75 488.3 1198.09 132 81 70.31 28.21 26.71 26.17 2237.55 2250.98 2257.086/5/2007 9:35 424.71 428.64 94.69 493.00 486.4 1196.63 133 79 68.85 26.22 26.27 27.20 2230.22 2254.64 2250.986/5/2007 9:40 424.51 428.05 97.20 491.26 485.4 1192.23 132 76 68.60 25.73 25.20 26.81 2247.31 2250.98 2254.64

Fig. 5

SPE 110701 9

Cycle #

50

100

150

200

250

300

350

400

450

500

day

Temp[ºF] 1 2 3 4 5

440ºF

343ºF

411ºF

340ºF

84ºF

8/26 8/27 8/28 8/29 8/30 8/31 9/02 9/03 9/09 9/10 9/11 9/12 9/138/25 9/01 9/14 9/15

447ºF 443ºF444ºF

372ºF376ºF

110ºF

143ºF

453ºF

173ºF

372ºF

6

401ºF

Temperature ProfileHot Loop Test #4 - Aug/Sept 04

7 8 109443ºF

Fig. 6: Hot Loop Temperature Profile, Wellbore Fluid Temperature with Thermal Cycling