By Petroleum Technology Transfer Council

Hydraulic jet pumps prove well suited for remote Canadian field

Low downtime and minimal maintenance improve remote oil field economics.

John Anderson, Nexen Canada; Roger Freeman, Penn West Energy Trust; Toby Pugh, Weatherford International

With 150 million bbl of recoverable reserves, the Nexen Hay River Bluesky oil pool is among Western Canada’s largest oil finds in the past 20 years. Initial pilot production testing led to a finding that, to economically produce the field, several things would be needed:

• Triple, lateral, horizontal producing wells.
• •n-field horizontal water injection/ water disposal wells for pressure maintenance.
• Artificial lift from the outset.

Further complicating output at Hay Pool, in remote, northeastern British Columbia, road access would be limited to four out of 12 months. This is because roads must be "frozen in" prior to use. So, it was extremely important, economically, that the artificial lift method chosen should be very reliable and require minimal maintenance. The method should be safe and capable of lifting large fluid volumes with significant pressure drawdown.

BACKGROUND

Hay field was found in 1984. Its development exploits a medium-to-heavy oil (24° API) reservoir about 40 mi northwest of Rainbow Lake in northeastern British Columbia near the Alberta border, Fig. 1. The productive Lower Cretaceous Bluesky sand is at a 1,060-ft TVD. Porosity ranges from 18% to 27%. Permeability ranges from 60 to 250 md. The oil zone is 13 to 15 ft thick.

Fig. 1. Typical well cross-section.

Hay lies under very flat muskeg that is accessible only during winter. Drilling sites are prepared when the surface is frozen, and ice roads can be used from mid-December to mid-March. At other times, they are only accessible by helicopter or hovercraft.

Pilot production tests. After drilling several vertical wells, it was found that the thin reservoir and its proximity to bottom water made vertical well production uneconomical. A horizontal well drilled in 1995 evaluated the potential to produce at adequate oil rates with reduced water cuts. Although this well had numerous drilling difficulties and was not economical due to high costs, it did test oil with a reduced water cut of 20%.

A pilot project began in 1998 to further evaluate the potential to produce oil with horizontal wells. First, the sole existing horizontal well from 1995 was worked over. After cleanout of the lateral section with a jetting tool, and reinstallation of the progressive cavity pump (PCP) at a greater TVD, output grew noticeably. One lesson learned from this workover was that well control was a key concern, as it took three days to kill the well. Second, four single-leg horizontals were drilled to average lateral lengths of 3,825 ft, Fig. 1. These wells proved that fairly long laterals could be drilled in the thin sand, and proved that the formation could handle open-hole completions.

Once drilled, the wells were suspended with a bridge plug in the intermediate casing until a service rig could move in and complete the well. During completion, well control was again a major factor, with more time spent on killing the well than in running the completion. The wells were equipped with PCPs landed in tangent sections of the wellbores. A variety of rod accessories (scrapers, spin-through couplings, etc.) was employed to determine what configuration worked best in directional wells with aggressive build rates and stringent drawdown requirements. Production tests confirmed the potential for economic recovery, with an average fluid rate of 315 bpd and a 13% water cut.

PRODUCTION CRITERIA

It was clear that day-to-day output presented a different challenge – a system that could produce reliably, while essentially unattended, would be critical to Hay’s financial success. Since it was found that both artificial lift and water injection would be needed, the choice of an artificial lift mechanism required much research within the participating partners and the general industry.

Many variables had to be considered – fluid handling, surface facility dimensions, power generation, equipment reliability and workover frequency. Several different mechanisms were considered:

• Electric submersible pump (ESP). Although it is a highly reliable artificial lift mechanism, the ESP could not be landed far enough downhole to achieve maximum drawdown, due to dogleg severity (150/100 ft).
• PCP. This system is used widely in Canada and has an established performance track record. Among the advantages that it offers are low power requirements, small surface plant design requirements (lower fluid handling requirements) and the belief, based on field experience, that a three-year running life is achievable, thereby minimizing workover requirements.
• Gas lift. The simplicity of the downhole completion, requiring only gas lift valves is very attractive. However, gas lift could not obtain sufficient pressure drawdown.
• Jet pumps, which were in limited use in Canada at the time, use the transfer of energy from power fluid (such as water or produced crude oil) pumped into the tubing at high pressure to lift formation fluid up the annulus. Advantages include low maintenance costs, ease of pump change-out and a centralized surface equipment package. However, a major disadvantage is the PCP’s nearly double power requirement.

The project team made a decision that the field would utilize PCPs, although concern remained that any rod pumping system would be hard-pressed to provide adequate running time in this remote area.

INITIAL JET PUMP PILOT TEST

Two previously drilled horizontals were put on an extended production test from January to March. One single-leg horizontal was equipped with a high-volume Progressive Cavity Pump (PCP). Another single-leg horizontal was equipped with a hydraulic jet pump. Workover operations to install the pumps were extremely costly, due to well control difficulties, where one well took seven days to kill. Although the wells would flow readily under initial reservoir conditions, artificial lift would be required to draw the wells down far enough to achieve desired output rates.

The pilot production test confirmed the potential for commercial output and also demonstrated that the jet pump could produce volumes equivalent to those of a PCP. To obtain real-time bottomhole producing pressure, the PCP well was equipped with a Promore downhole sensor. Also, to achieve the drawdown needed to maximize output, the pump was located in a tangent section at 68° hole inclination.

The rod string above the pump was equipped with spin-through couplings to minimize wear in the build section, which had dogleg severities of about 13°/100 ft. A tubing rotator swivel and no-turn tool were also run. The pump positioning and rod design were dual-purpose, to provide the required drawdown and give the PCP an intentional "torture test." It was estimated that field development with PCPs would require a three-year service life for these difficult conditions. The jet pump was landed at an equivalent depth, inclination and dogleg severity. A schematic of the pilot testing is shown in Fig. 2.

Fig. 2. Jet pump and PCP completion configuration.

Surface equipment (triplex pump, prime mover, separator, etc.) used for the jet pump installation comprised items that were readily available in the field, rather than equipment ideally suited for the trial application. Cavitation problems in the power fluid system, and separation of produced fluids, contributed to the step rate type of performance. Nozzle selection and system horsepower versus bottomhole pressure (BHP) drawdown is always a compromise. However, in this pilot test, the bottomhole producing pressures of the PCP well and jet-pumped well were nearly identical. Pilot production test results are shown in Fig. 3.

Fig. 3. Pilot test production results.

Despite the jet pump’s performance, there was considerable apprehension about using the system exclusively. So, it was decided to use PCPs on two pads and jet pumps on one pad for the program’s first year.

FINAL ARTIFICIAL LIFT SELECTION

Upon further review of project aspects, the choice of artificial lift medium was revisited, and primary criteria for project success were reconsidered. It was concluded that the Hay project’s economic risks were now strictly related to operations and production.

After evaluation of the pilot production test and simulation, a major change in depletion strategy was made. Rather than inject produced water peripherally to the oil pool boundary, the lack of aquifer support necessitated that pressure maintenance via in-field water injection would be required with the same tight spacing as the producing wells. This meant that water injection lines would have to run to the producing pads, where there would be four, triple-lateral horizontal producers and four injectors per pad, two dual-leg horizontal injectors and two single-leg horizontal injectors.

Since a power fluid source was coming to the pad anyway, the only additional cost of running the jet pump would be the incremental horsepower needed to boost the entire water system pressure and supply the pump, using larger lines as needed. This could facilitate the entire field’s production and maintenance of pressure by pumping fluid from the central facility to the producing pads with a minimal electrical distribution system.

Cost estimates for full-field development were based on production of 6,000 bopd and 120,000 bwpd. Despite the jet pump’s higher horsepower requirement, and when all costs were considered (especially those associated with workovers and lost production), it was decided to change the allocation of pump types and equip one pad with PCPs and two pads with jet pumps.

Operations proceeded accordingly with the following installations:

• One six-well pad was completed with PCPs, with pumps landed in the build section. To achieve the required drawdown of some 50%, rotational requirements were 300 rpm.
• Two pads, totaling eight wells, featured jet pumps, with pumps landed 10 – 20 m above the formation in the 70° build section.

First-year output results proved the jet pump conclusively as the best choice. Once the drilling season was over, and summer conditions allowed very limited access, the screw pumps all failed, and downtime was considerable with commensurately high workover costs. Rod jobs were performed on wells by a helicopter-mobilized mast unit. However, tubing could not be pulled until a workover unit could regain full access. The wells were worked over the following winter, and pumps were placed in their vertical sections. Unfortunately, while the screw pumps’ performance did improve, most of them failed again.

By contrast, there was no downtime due to jet pump failure. Output was maintained from these eight wells throughout the year. A comparison of the downtime hours by failure type is shown in Fig. 4.

Fig. 4. PCP pump downtime by failure type.

Several issues that hindered operations had to be addressed, such as scale that inhibited the ability to circulate the pumps out of the well, and problems with emulsion at the facilities. The first full year of production proved conclusively that the jet pump approach was, by far, the most viable, with the resultant decision that all new wells would be completed, and all existing PCP installations worked and recompleted with jet pumps. Production data that compare the running lives of both pump types are shown in Fig. 5.

Fig. 5. Typical PCP and jet pump production performance.

THE JET PUMP

Fig. 6. Jet pump schematic.

A typical jet pump (Fig. 6) is activated by pumping high-pressure power fluid (in this case, produced water) down the tubing. The power fluid enters the pump at the nozzle, where a transfer of energy occurs to the produced fluid entering the pump throat from below. The combined fluid is discharged into the tubing/ casing annulus above the packer and then to the surface, where it is separated, and the water is used for both power fluid and water injection requirements.

In this application, the jet pump offers several major advantages:

• No moving parts, so it is very reliable with low maintenance.
• If wear occurs, change-out is easy, either by reverse circulation to the surface or wireline retrieval.
• It can pump high fluid volumes.
• It can be run in severe build sections and placed very close to the formation to maximize drawdown.
• It allows for localization of power sources into one facility.

The jet pump also has several disadvantages that can make it unsuitable for some operations. Among these are:

• Lower efficiencies when compared to other pumps, requiring much more horsepower.
• High fluid volumes required for activation mandate large surface facilities with higher capital costs.
• Very dependent on backpressure.

ADDITIONAL JET PUMP ADVANCES/ ADVANTAGES

Jet pump completions allowed for use of the same tubulars and packer as used in the injection well completions, thus simplifying operations. Because well control was a paramount concern at Hay, it was critical that the completion be run in the hole as soon as possible after drilling finished. The jet pump completions were simpler, cheaper, faster and safer than other alternatives.

Nexen had a jet pump recorder housing built that allowed the bottomhole producing pressure on any well to be determined by simply circulating the pump in the well, producing for a defined period, and circulating the pump and pressure recorder back out of the hole. Some wells had permanent sensors with surface read-out. However, the jet pump recorder carrier provided additional flexibility in determining BHP at a substantial cost savings.

Nexen designed a simple four-way valve system and wellhead, resulting in fail-safe pump operation and greatly reduced wellhead maintenance. For the water source wells, a jet pump was used in conjunction with a booster pump, to supply greater power fluid pressure. A second centrifugal pump was used to reduce casing pressure and deliver produced fluid to the central facility. These wells, equipped with 2-7/8-in. tubing, initially produced 5,600 bopd.

Jet pumps are being used at Hay and can produce up either casing or tubing. They can be configured to produce the well up the casing on original cleanup, and then produce up the tubing for long-term output in potentially corrosive environments. The project has also proved to be larger than originally anticipated, with output at 9,000 bopd and 230,000 bwpd.

CONCLUSIONS

Jet pumps may not be right for some applications but are ideal for Hay field, due to reliability, low downtime and minimal maintenance. All 77 producing wells were equipped with jet pumps, and plans were implemented to equip 17 additional wells. The jet pump should be considered for all field applications, where it can address major requirements, especially when high-volume lift is necessary. Detailed, up-front planning is a must, to ensure that facilities can handle fluids for both power fluid and injection. Utilizing existing injection line networks can help project economics greatly. 

ACKNOWLEDGMENTS

The authors thank the management of Nexen Canada Ltd. for permission to publish this article, and Weatherford Completion and Production Systems for its support.

REFERENCE

¹   Sander, M., R. Norgaard, K. Kern and T. M. Warren, "Project management and technology provide enhanced performance for shallow horizontal wells," Dallas, Texas, Feb. 26 – 28, 2002.