ROUTINE MAINTENANCE PROLONGS ESP TIME BETWEEN FAILURES

Sept. 21, 1992
Terry Hurst Chevron U.S.A. Production Co. Snyder, Tex. Robert W. Lannom, David L. Divine E.S.P. Inc. Midland, Tex. Routine maintenance of electric submersible motors (ESPS) significantly lengthened the mean time between motor failures (MTBF), decreased operating costs, and extended motor run life in the Sacroc Unit of the Kelly-Snyder field in West Texas (Fig. 1).
Terry Hurst
Chevron U.S.A. Production Co.
Snyder, Tex.
Robert W. Lannom, David L. Divine
E.S.P. Inc.
Midland, Tex.

Routine maintenance of electric submersible motors (ESPS) significantly lengthened the mean time between motor failures (MTBF), decreased operating costs, and extended motor run life in the Sacroc Unit of the Kelly-Snyder field in West Texas (Fig. 1).

After the oil price boom of the early 1980s, rapidly eroding profit margins from producing properties caused a much stronger focus on reducing operating costs. In Sacroc, ESP operating life and repair costs became a major target of cost reduction efforts.

The routine ESP maintenance program has been in place for over 3 years.

SACROC UNIT

The 49,900 acre Sacroc (Snyder Area Canyon Reef Operating Committee) Unit, produces from the Canyon Reef (Pennsylvanian) limestone reservoir at approxit ately 6,700 ft.1-3 The field was discovered in November 1948 and most of the wells were completed with 5 1/2 and 7-in. casing.

After the Sacroc Unit was formed in 1954, waterflooding started.

Initially, only 68 of the 1,000 plus producers were on artificial lift. These rodpumped wells were located in areas with low producing rates. Until the early 1970s, most wells in the unit flowed. These strong, high oil cut flowing wells let Sacroc easily reach the Texas regulated maximum allowable monthly production.

ESPS IN SACROC

In Sacroc, the first submersible pump was installed in 1970. In 1984, the number of ESP operating systems peaked at 688. The installation of ESPs coincided with waterflood and CO2 flood expansion projects that increased water production from 38,000 b/d to more than 1 million b/d.

The ESPs operate in a corrosive environment. Waterflooding increased hydrogen sulfide (H2S), compounding corrosion problems. In 1972, introduction of carbon dioxide (CO2) injection added more corrosion problems.4 5

The CO2 flood has made it difficult to properly design and size ESPs because reservoir performance conditions change rapidly as water and CO2 injection are constantly and frequently alternated.

Also, CO2 breakthrough at the producing wells severely hampers ESP performance.

Another operating problem is that as the wells have aged, casing problems have increased. The old casing impedes pulling and running ESPs and contributes to other equipment failures.

ESP failures also occur, because in Sacroc interruptible power was installed to reduce electrical costs. Although power costs were reduced, well downtime and equipment failure frequency increased after power outages. These failures are in addition to failures seen during thunderstorm season.

Another cause of ESP failures is that every year Sacroc undergoes a very active producer and injection well workover program. The workovers change reservoir inflow performance and alter producing rates. These changes cause the release of solid materials that wear and plug equipment.

With oil price deregulation in 1980, a very active program began that sized ESPS to ensure maximum oil production rates from every well.

Because of rapidly increasing water cuts and much lower oil prices, the number of operating ESPs began to decline after 1984. The reduction was accentuated by the collapse of oil prices in 1986. Today, Sacroc operates 380 submersible pump systems.

Sacroc's submersible pumps range in size from 300 to 6,200 b/d and are powered by 35-400 hp motors. Pumps are landed at an average depth of 6,500 ft, 55 ft above the top perforations. Approximately half of the wells have 5 1/2-in. casing.

Because of high producing rates, many pump systems consist of tandem pumps and motors ranging from two to five sections per pump and/or motor.

Submersible pumps from all major pump manufacturers have been, or are currently, in use. Sacroc is also heavily involved in mixing equipment from various vendors to reduce inventory costs by decreasing redundant pump sizes and motors from different vendors.

In 1985, a routine maintenance procedure for the electric submersible motor was developed and a trial study started.6 By late 1987, this study indicated that routine maintenance could prolong motor life more than the traditional dryout repair done by the manufacturer.

In January 1988, Sacroc began the routine maintenance on the majority of motors pulled.

ROUTINE MAINTENANCE

After pulling the pump, the submersible motor is sent to a test shop for routine checking and maintenance.

First, the motor is cleaned externally before an ohm meter reading verifies the phase balance. Unbalanced readings indicate a burnedout motor and the procedure ends.

If the readings are in balance, the oil in the motor is drained, and the dielectric strength of the oil is checked and recorded.

The motor insulation then is checked with a 1,000 y megohm meter (megger). A motor with a reading less than 2,000 megohms is considered wet and will be sent to the dryout area.

For the dryout, a dc current is passed through the motor's windings to raise the temperature to 240 F. After 72 hr, any moisture will be baked from the motor. If a vacuum is applied to the motor at the same time as the dc current, dryout is reduced to 24 hr.

One of this procedure's positive features, as compared to a manufacturer's dryout, is that the motor does not have to be disassembled and then remanufactured. This saves money and time and does not require passing the unit through the infant mortality section of the product reliability curve.6

After passing the megger test, clean, dry oil is circulat ed through the motor until the dielectric strength of the oil reaches 25 kv or greater. The motor is then ready to have the insulation's polarization index (PI) measured.7 8

The PI is determined by holding a fixed dc voltage on the windings and measuring the leakage current-vs.-time. The 1-min leakage current divided by the 10-min leakage current defines the PI. The PI should equal at least 2.0. If less, the motor is considered wet and returned to the dryout area.

The next test measures the motor's insulation resistance.6 7 While the leakage current is monitored, a dc test voltage is slowly raised to the calculated maximum.7 The minimum calculated insulation resistance should be in the 5,000-10,000 megohm range. If less, the motor is considered wet and returned for a dryout.

After passing the dc electrical tests, the motor is tested mechanically by having power applied to the motor. The motor is allowed to run until it reaches a stable temperature. The power is then removed, and the motor coasts to a stop. The time and total revolutions are counted and compared to the data base for good motors.6

Next, a transient impulse generator (TIG) tests the motor's insulation system from turn to turn.

Finally, the motor is vacuum purged and pressure filled with oil for storage or returned to the field.

Unless a dryout is required, the whole process normally takes no more than 2-4 hr.

RESULTS

The maintenance program started in January 1988. Since then, more than 2,200 motors have been through maintenance. Some motors have been through the procedure several times.

Records of motor run times until failure have been carefully retained and documented by motor serial number. The trend chart (Fig. 2) shows the average days of run time by month for motors that failed.

In January 1988, a total of 24 motors were pulled from the Sacroc field. Twelve of these motors were burnedout. The average run time of these 12 motors was 813 days before failure.

In a similar analysis during September 1991, 57 motors were tested and 7 were determined to be burnedout. The average run time of these 7 motors was 1,404 days. As shown in Fig. 2, the run time trend continues to increase due to the continued use of the routine maintenance program.

A similar method of analysis monitors the number of wet motors. Since the program's start, approximately 390 of the tested motors (17.5%) required dryout (Fig. 3). Of these, 96% or 374 were successfully dried out, tested, and returned to service.

Cost savings of the routine maintenance program is substantial.

The average 1991 price book lists the cost of repairing a 100 hp, 450 series (5 1/2-in. casing size) ESP motor, such as those used in Sacroc operations, as approximately $5,630 for a dryout and $9,940 for a rewind.

By comparison, routine maintenance including the dryout process is approximately $800.

ACKNOWLEDGMENTS

The authors thank Chevron U.S.A. Inc. and E.S.P. Inc. for their permission to publish this article.

REFERENCES

  1. Vest, E.L. Jr., "Oil Fields of the Pennsylvanian-Permian Horseshoe Atoll, West Texas," Geology of Giant Petroleum Fields, Memoir No. 14, AAPG, AAPG Meeting, Oklahoma City, Apr. 24, 1968.

  2. Perryman, T.L., "Workovers: Key to Sacroc Reservoir Control," Petroleum Engineer, February 1972.

  3. Dicharry, R.M., Perryman, T.L., and Ronquille, J.D., "Evaluation and Design of a CO2 Miscible Flood Project Sacroc Unit, KellySnyder Field," JPT, November 1973.

  4. Newton, L.E. Jr., and McClay, R.A., "Corrosion And Operational Problems, CO2 Project, Sacroc Unit," Paper No. SPE 6391.

  5. Patterson, K.W., "Downhole Corrosion Encountered in the CO2 Flood at the Sacroc Unit," Southwestern Petroleum Short Course, Lubbock, Tex., April 1979.

  6. Johnson, R.A., and Divine, D.L., "Reducing Costs Through the Routine Maintenance and Testing of Submersible Motors and Pumps," Gulf Coast Section of SPE ESP Workshop, Houston, Apr. 26-28, 1989.

  7. Recommended Practice for Insulation Testing of Large AC Rotating Machinery with High Direct Voltage, ANSI/IEEE Std. 95-1977.

  8. Recommended Practice for Testing Insulation Resistance of Rotating Machinery, ANSI/IEEE Std. 43-1974.

Copyright 1992 Oil & Gas Journal. All Rights Reserved.