Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Optimizing Electric Submersible Pump Replacement Costs N.W. Kupsch

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

OPTIMIZING ELECTRIC SUBMERSIBLE PUMP

REPLACEMENT COSTS

N.W. KUPSCH

this article begins on the next page F


JCPT90-05-04 OPERATIONS Optimizing electric submersible pump replacement costs NATHAN W. KUPSCH Amoco Canada Petroleum Co. Ltd. ABSTRACT Electric submersible pumps (ESPS) have made a significant con- tribution to the petroleum industry as an economic means of ar- tificial lift. As current depressed oil prices linger, it has become especially critical that costs associated with operating ESPs be kept to a minimum. One method, which has resulted in a great deal of success for Amoco Canada, has been qualifying used motors and pumps for re-use through cleaning and testing procedures. Among
Amoco Canada's operations are two large fields,- the South Swan Hill Unit and the Nipisi Gilwood Unit No. 1. Together, a total of 235 electric submersible pumps are i . n serv- ice. As ESPsfail downhole, equipment is examinedforpolential testing application. To the end of June 1989, 457pumps have been cleaned and tested while 221 motors have also been put through the test program. A high requalijication rate and in@nsive testing cost have combinedfor substantial savings over traditional repair procedures. The success of the program has been further enhanced by encouraging ESP performance
from equipment rerun subse- quent to being cleaned and tested. Introduction Amoco Canada Petroleum Company Ltd. is the U@t Operator of two major miscible flood projects. The locations of these fields are shown in Figure 1. The South Swan Hills Unit is located 225 km northwest of Edmonton, Alberta, and comprises 188 producing wells with 70 injection wells. The 124 electric submersible pump locations account for almost 9707o of the 30 000 M3 per day aver- age fluid volume. Production is from the Swan Hills formation of the Beaverhill Lake Group, a limestone reef complex. Bottom- hole
temperature averages 1120C at a depth of 2 500 m KB. The Nipisi Gilwood Unit No. I is located approximately 350 km north of Edmonton. The average Unit fluid production is 23 500 M3 per day. Included in the 230 producing wells are I I I ESPs which are responsible for 9307o of this volume. An addi- tional 74 wells are used for injection. The producing reservoir is the Gilwood (sandstone) formation at a depth of 1 700 m and BHT of 600C. During the past few years, an average of 125 ESP replacements have taken place annually in the above locations. Included in the $55,000 average
replacement cost were ESP equipment expenses of $28,000 each. This resulted in an annual expenditure of ap- proximately $3,500,000 for ESP repair and/or purchase. In April 1987, a used pump and motor testing program was ir-dtiated, s@ar to that described by D.L. Divine(l). In the 26 months since that time, ESP replacement expenditures have been reduced by almost $2 milhon. Keywords: Electric submersible pump, ESP, Pump test, Motor test, Cost reduction, ESP run life, Equipment requalification. Paper reviewed and accepted for publication by the Editorial September-October 1990, Volume 29,
No. 5 ESP Operations Amoco Canada replaces ESPs through the most economic means available. This has historically been with new equipment purchase and/or flat rate repairs. Research into the failures experienced showed that while one component failed, companion equipment was often in good condition and could be re-used. Standby in- ventories have been maintained separately for each Unit in order to minimize costs and eliminate deferred production due to the unavailability of equipment. This inventory provided an ideal sit- uation for testing motors and pumps, because replacement pieces
could be taken from stock with no production loss incurred while waiting for test results. The first ESPs tested by Amoco Canada for potential re-use were done iii the U.S. during mid-1986. Made up primarily of equipment with short r-un lives, the group was tested with very successful results. With the significant cost savings r@ on these pieces, fur-tber applications were sought. Amoco owned a large amount of unrepaired ESP equipment as a result of purchasing new pumps when economics were favourable over repair and/or exchange. Most of this equipment had simply been pulled and
stored, some for up to three years. A total of 108 pump housings and 46 motors were tested from this stock after a very limited selection process. Although experiencing a lower success rate in testing, cost savings still proved very attractive. With industry ac- ceptance of ihe testing concepts, major ESP vendors then decided to offer similar services locally. At this time, Amoco Canada in- itiated its routine pump and motor testing program. Submersible Pump Testing Selection of Pumps for Testing The initial st@ep in the pump testing program is to identify those pumps which are suitable
candidates for the testing procedure. Various factors may influence the success ratio of the program. Among these may be: run life of the equipment; presence of abra- sives (sand); foreign material plugging (scale, asphaltene); fluid production rates relative to pump size; mechanical condition when pulled; and, time and method of storage. In order to qualify a pump for testing, the above points should be evaluated to ensure a reasonable degree of confidence in achiev- ing a successful test. In the early stages of a testing program, all pumps may be tested in order to set selection
criteria. Although a lower percentage may meet specifications, the low testing cost should still make the program economic. The most important information used to select pumps for test- ing is data obtained when the ESP is pulled. Bent housidgs, se- vere corrosion, broken shafts, twisted splines, bushing wear or excessive shaft play may immediately disqualify a pump from the test program. @le a seized pump may indicate potential damage, often the cle@ng procedure can free it and an acceptable test may Board of The Journal of Canadian Petroleum Technology. 63

You might also like