The Krishna-Godavari (KG) basin is a proven, petroliferous deltaic basin formed by the discharge of two large rivers—Krishna and Godavari—flowing on the east coast of India. In an exploratory well in the KG basin, a field with two separate reservoir sands was encountered, Fig. 1. Both of these reservoirs, named G and R, are tight formations, which require hydraulic fracturing for commercial productivity.
The G formation is composed primarily of siltstone approximately 200-m-thick, with permeability and porosity of about 0.1 mD and 10%, respectively. The approximately 35-m-thick R formation is composed of sandstone and clay, with better reservoir quality having permeability and porosity of 0.1 mD and 15%, respectively. Based on the estimated reserves in place, the G formation is the primary target. Table 1 summarizes key reservoir properties for both of these formations.
In the discovery well, there were oil shows in both formations. However, only the R formation flowed oil, at a rate of about 30 bpd, to the surface during the drillstem test (DST.) To prove commercial viability, a new appraisal well was drilled, fraced and tested—the results were not encouraging. The discovery well was then utilized as the second appraisal well. Re-entry was made to meet the expected fracturing loads.
The discovery well was completed with a 7-in. production liner in an 8 3/8-in. hole. Since the production liner was not rated for the expected fracturing pressures, it was decided to sidetrack the well from the 9 5/8-in. casing. The final well design consisted of a 7-in. intermediate liner, landed above the pay section at about 3,900 m, MD, followed by a 4 1/2-in. production liner, which was landed across the pay section, Fig. 2. The completion design was constrained by requirements for hydraulic fracturing and requirements for post-frac production testing.
Phase I (hydraulic fracturing). The optimum fracturing treatment design required an injection rate of 25 bpm with a maximum bottomhole pressure (BHP) of 12,200 psi. A 4 1/2-in. monobore completion was required to achieve this rate and handle the pressure. The monobore completion consisted of a 4 1/2-in. frac string strung into a 4 1/2-in. production liner hanger. The considerations for the hydraulic fracturing phase completion design included:
A 4 1/2-in. monobore design would allow pumping the frac under designed surface pressure limits at 25 bpm, with an estimated fracture gradient of ~0.92 psi/ft. Figure 3 shows completion testing for the pressure expected during the fracturing operation. The part of the completion above the 41/2-in. liner hanger and seal bore could be tested by applying 11,500 psi in the tubing and 5,000 psi in the annulus. But the same pressure was not sufficient to test the production liner, as the frac load was much higher. A two-stage testing sequence was adopted to pressure test the well, as per expected pressure loads during fracturing.
- This design efficiently catered to a standard plug-and-perf isolation technique while performing successive fracturing stages.
- A monobore design allowed for installation of the lower completion in an underbalanced condition.
Phase II (production testing). Some of the production requirement considerations, which made this completion unique, are listed below:
- Zone selectivity was required, to test each formation separately, to know the techno-commercial viability to appraise these sands individually.
- System analysis showed that 2 7/8-in. production tubing was best suited for long-term production testing in the first appraisal well. A workover was required to pull out the 4 1/2-in. frac string and then replace it with 2 7/8-in. production tubing.
- Post-frac well killing, during the workover, was suspected as one of the causes of poor productivity. Core flow testing confirmed the damage potential of the kill brine. To eliminate the need for post-frac well killing, the lower completion assembly was to be landed in underbalanced conditions.
- A slimhole selective completion assembly, consisting of two packers with a sliding sleeve in-between and a nipple profile below the lower packer, was selected. The 24.4-m length of the coiled tubing PCE lubricator complicated the completion assembly.
- While performing the workover on the 2 7/8-in. string, the lower completion assembly had to be utilized as a barrier. This was accomplished by installing a pre-set plug in the nipple profile, in the lower assembly. Suitable design changes were made in the setting tool to achieve this.
- The 2 7/8-in. lower completion assembly was conveyed to depth inside the 4 1/2-in. production liner by coiled tubing. This requirement was a key factor in the sidetrack design.
The primary objective of fracturing in this campaign was to test the flow potential of each reservoir. Fracture modeling indicated that the thinner R formation could be completed effectively with a single fracturing treatment. Since the lower G formation was about six times thicker, it required two fracture treatments.
KEY LEARNINGSThe high bottomhole static temperature of 325°F is at the limit of a standard borate, crosslinked frac fluid system. To accomplish this extensive frac fluid testing, which included fracture conductivity testing, a carboxymethyl hydroxypropyl guar (CMHPG) polymer crosslinked system, incorporating borate and zirconate crosslinkers, was selected.
Once the production liner was landed and cemented, the 4 1/2-in. frac string was run in and stung into the liner hanger PBR (~ 6 m in length). Post perforation, pre-frac flowrates for each zone were measured as a baseline. This was followed by the successful placement of three hydraulic fracture treatments. The maximum proppant placed in a single frac was 387,000 lb, which is the largest in terms of the amount of proppant placed in India. Figure 4 shows the pressure-matched frac geometries with openhole log, and Table 2 lists key fracturing parameters. Based on these results, two frac stages in the G formation covered 95% of the net pay height, and one contained frac was placed in the R formation.
- A lower completion with bottom packer, perforated pup, nipple profile (with a pre-installed plug in the nipple profile) and mule shoe;
- The middle completion with upper packer, 2 7/8-in. sliding sleeve, selective nipple profile with seal bore ratch latch mechanism;
- The upper completion with 2 7/8-in. tubing, sliding sleeve, dynamic seals with a latching anchoring assembly.
The packer setting BHA for the lower and middle completion, which was run via coiled tubing, made tool strings of 13.1 m and 23.5 m, respectively. To rig up and run in the 23.5-m tool string, on coiled tubing, a swivel joint was incorporated into the tool string.
Identifying a suitable running tool to set the packer assembly was a challenge. The only tool available was a conventional coiled tubing-deployable packer setting tool. It works on the principle of a mandrel and a piston, in which the ball sits on the ball-seat and generated hydraulic pressure breaks the shear pins at a predetermined pressure. This, in turn, initiates the piston movement to set the packer.
The challenge here was to run this assembly in with a pre-set plug in the nipple profile below the packer, so as to provide isolation once the packer is set. However, as this lower assembly was to be set hydraulically with a ball drop mechanism, the pre-set plug would block the flow path for pumping the ball. To allow fluid flow while pumping the ball, the packer setting tool was modified to have ports below the ball seat in the tool.
Installation of the lower completion packer went according to plan, but, while setting the middle packer assembly, the packer setting tool failed to set the packer and release the tool string. The rise in pressure required to set the packer was not observed. The middle part of the completion had to be installed with seals and latch at the bottom, to anchor the middle completion into the seal bore of the lower packer. Releasing the setting tool from the packer, after it has latched, was not possible without shearing the ball seat.
Coiled tubing was then disconnected from the hydraulic disconnect in CT BHA. This left the lower portion of the BHA with a GS internal fish neck profile as its top. The remaining portion of the BHA and the setting tool had to be retrieved, to set an additional third packer to isolate the R formation. Fishing in 4 1/2-in. liner was challenging and costly, because of small clearances with conventional rig-less fishing tools, and a rig-based intervention had to be conducted.
TOOL FAILURE ANALYSIS
A thorough investigation was conducted to analyze the root cause of the packer setting tool failure. The packer setting tool had eight O-rings to contain the hydraulic pressure and initiate the setting mechanism. The O-rings ensure that the tool operates only when there is a positive pressure inside the running string. Failure of any of these O-rings opens a flow path to the running-string/casing annulus, rendering the tool useless.
To simulate the downhole hydraulic condition, a test was specially designed, which replicated the ball dropping, ball seating and pressure build-up mechanism inside the packer setting tool, Fig. 5. Tests were also conducted on the recovered packer setting tool, which confirmed that the O-rings were leaking heavily.
Further investigation of the tool design showed that if any one of the eight O-rings fail, the tool fails due to the lack of a back-up O-ring. Though the tool is being used extensively in the industry, no published literature was found to bench test the tool on site before running in. The service provider did not have any setup or engineered bench-test to replicate the downhole setting condition, which could confirm if the setting tool is good to work.
An extended two-month production test was conducted for the G and R formations. This testing validated the fracture performance and confirmed the commerciality of the prospect. Table 3 lists a summary of the production performance parameters. The G formation was tested at an average rate of 877 bopd on a 7/16-in. fixed choke. This was seven times more than the previous appraisal well. Table 4 shows a comparison of the production results with the previous appraisal well.
The successful conclusion of the project led the operator to draw a number of conclusions and recommendations for future work. These are:
- Systematic planning and the engineered well design were the key reasons for achieving the production gains.
- Successful placement of three hydraulic fracturing jobs helped to achieve 95% of gross height coverage in the G formation and complete coverage in the R formation.
- An underbalanced completion deployed on coiled tubing was a successful method adopted to avoid well killing in the G-formation, which was a major reason for the approximate seven-fold increase in production, as compared to the first appraisal well.
- Effective modifications in the lower completion assembly were made in the planning stage to fully accomplish a reservoir-friendly workover operation, without killing the well.
- Root cause analysis of the packer setting tool failure highlighted the requirement to improve the industry’s standard hydraulic packer setting tool design.
- Test setup built for root cause analysis could be utilized successfully in the future for bench testing the condition of packer setting tools.
- It is recommended to conduct pre-job function testing of any hydraulic setting tool and make it a standard practice.
- Extended production testing was an effective way of establishing the productivity of these reservoirs and evaluating the fracture performance.