Fiberglass liners with polished bore seal assemblies are being successfully used to salvage old injection wells as less than half the cost of drilling the wells again. Components of conventional waterflood fiberglass liner completion equipment used in the past to improve well conditions and injection profile control were modified or changed.
Tertiary oil recovery techniques, such as carbon dioxide (CO2) flooding, require a closer economic evaluation of problems previously accepted during waterflooding. The Seminole San Andres Unit (SSAU) in Gaines County, Texas operated by Amerada Hess Corporation, is a field with some problems that are now being addressed due to the increased cost of flooding a reservoir with the water-alternating-gas (WAG) CO2 flooding technique. The cost of injecting one reservoir barrel of water in the SSAU is approximately six cents whereas the cost for injecting one reservoir barrel of CO2 is approximately two dollars and forty cents, a significant difference.
SSAU injection wells are an area that previously accepted problems are now being confronted to optimize the success of the CO2 flood. This paper reviews the history of problems with these wells and the development of a fiberglass liner with a polished bore seal assembly for solving these problems.
The dolomite San Andres formation of Permian age located near Seminole, Texas (Figure 1) was first discovered in 1936. The formation is produced from a total depth of 5300 feet (1.615 E + 03 m) with an average pay thickness of 126 feet (3.84 E + 01 m). The cap rock of t he formation is a dense dolomite with an anticline structure for trapping oil. Original oil in place was 971 million barrels (1.54 E + 08 m3). Primary depletion by solution gas drive recovered an estimated 268 million barrels (4.26 E + 07 m3) of oil. The field was unitized in 1969 for the purpose of waterflooding. Secondary recovery is expected to recover an additional 204 million barrels (3.24 E + 07 m3) of oil. In 1983 CO2 flooding the reservoir with the WAG method was begun as a tertiary recovery technique for recovering an estimated 138 million additional barrels (2.19 E + 07 m3) of oil. The field is being developed into an 80 acre (3.34 E + 05 m2) inverted nine spot pattern with currently 427 producing wells and 161 injection wells of which 107 are WAG injectors.
Over 90% of the WAG injection wells in the field were drilled in the 1940's. These wells were completed with production casing set into the cap rock of the formation followed by drilling the pay zone. They were oil producers until converted to injectors for the water and the CO2 floods.
The San Andres reservoir has a good waterflooding history of matching theoretical zonal injection (porosity-feet profile) and logged zonal injection. The injection profile logs combined with an efficient oil recovery have verified the reservoir's ability to waterflood property with the open hole completion technique.
As shown in Figure 2 the water injection well completion utilized an injection packer installed within 200 feet (6.10 E + 01m) of the casing shoe. Injection well work during the waterflood was minimal due to an equipment failure rate of less than five percent a year.
No changes were chosen from the open hole completion technique when these wells were placed on WAG injection due to their successful history during waterflooding. Although casing inspection logs ran prior to CO2 injection indicated some deterioration below the packer, the corrosion was not significant enough to have caused numerous failures in the past. Now five years since the commencement of CO2 injection in the field, casing problems are slowly occurring and no field wide history of poor injection profile control has been observed.
Prior to CO2 flooding a small number of injection wells experienced significant water loss due to a hole in the casing below the packer or an open hole thief zone. They were repaired for one or more of the following reasons:
Injection into the main pay zone was decreased to the point that withdrawals were affected.
The cost of water lost justified working on the well.
Life of a well was jeopardized due to the possibility of casing collapse.