By Petroleum Technology Transfer Council

Underbalanced drilling increases deliverability from gas storage wells

Stimulation improves well performance.

Randall Cade, Weatherford, Russ Frame, Kinder Morgan, Inc., Houston, and Bob Wallace, Kinder Morgan, Inc., Chicago

In 2003 Kinder Morgan, Inc. implemented a drilling program in the North Lansing gas storage field as part of an expansion project. Seventeen wells were directionally drilled, nine of which were drilled conventionally with a cased and perforated completion. The eight remaining wells were drilled underbalanced and left as openhole completions. The well program was designed specifically to compare the deliverability performance of underbalanced-drilled (UB) wells to cased and tubing-conveyed (TC) perforated. After approximately 18 months of operation, the eight UB wells are out performing the nine TC wells.

The North Lansing gas field was discovered in the late 1940s and produced from the Young carbonate member of the Rodessa formation over the next three decades until a large portion of the gas reserves were produced. In 1975 the operator decided to convert the field to gas storage. The Young, at about 6,850 ft, has four producing zones of roughly 70-ft gross thickness; the two lower zones have the better productivity. Permeability ranges from 10 to 340 md, with an average porosity of 14.3%.

The goal for this drilling program was to convert existing cushion gas to working gas, and to increase overall field injectivity and productivity. A reservoir simulation model with a robust reservoir description was created from a 100 well dataset. After achieving an acceptable history match, predictions for individual well deliverabilities for both UB wells and TC wells were made prior to drilling. These predictions provided a basis for measuring whether the planned wells performed as expected.

TUBING-CONVEYED WELLS

Each of the nine conventional TC perforated wells was drilled in a similar manner. After setting surface casing, a 83D4-in. hole was drilled from the surface casing shoe through the reservoir, using a Bentonite mud with low fluid loss and low solids content, creating an overbalance of several hundred psi. Next, 7-in. casing was run through the reservoir and cemented in place. After cleaning out cement, the wells were circulated with clean, filtered 2% KCL water in preparation for extreme underbalanced, tubing-conveyed perforating.

The completion string comprised 5.5-in. tubing, a production packer, and perforating guns loaded with 12 shots/ft. After firing the guns with about a 2,500-psi underbalance, the wells were flowed immediately for cleanup. The guns were then released and dropped into the rathole.

UNDERBALANCED WELLS

For the eight underbalanced wells, each well was drilled to just above the top of the reservoir, and 7-in. casing was set and cemented. The 7-in. shoe joint was drilled out with water using a 6-1/8-in. bit. An underbalanced condition was initially achieved by injecting nitrogen with the drilling fluid, water, until sufficient influx of storage gas production was achieved to maintain an underbalanced condition. No foaming agent was required for removal of cuttings with the resultant annular velocity achieved in this application. Drilling continued through the reservoir section, roughly 70 ft, with an additional 20 ft drilled for rathole.

Upon reaching TD, the drillstring was retrieved using “pipe light” methods to achieve minimal fluid pressure on the formation. Unfortunately, the wells unintentionally became overbalanced during logging operations that required a partially fluid-filled wellbore. A production packer was then set on electric line with a plug set in the tailpipe, to protect the formation from damage, followed by running of 51D2-in. large diameter tubing (LDT). As soon as possible after running the LDT, the well was unloaded with nitrogen and cleaned up.

DELIVERABILITY

Following connection to the surface system, each well’s actual performance was measured by its deliverability back pressure curve for comparison to its predicted performance. All of the conventional wells and four of eight UB wells were found to be underperforming compared to the prediction. The combined TC wells were 70 MMcfd less than expected, which was much greater than the 15 MMcfd below expectation of the UB wells, Table 1.

TABLE 1. Post-drilling and post-stimulation productivity compared to the expected results.

Normalizing for the actual footage of net pay encountered, the conventional wells averaged about 0.35 MMcfd/ft, equating to only 47% of the 0.75 MMcfd/ft expected. By contrast the UB wells delivered 0.54 MMcfd/ft, about 50% better than the conventional wells, but still 19% under expected deliverability of 0.66 MMcfd/ft.

Underperformance of the UB wells is thought to be primarily caused by the unintended killing/ filling of the wells during the logging operations. The reason for the TC well underperformance is less certain, and part of an ongoing study.

STIMULATION

Because nearly all the wells initially underperformed, the decision was made to attempt to restore productivity by acid stimulation using a Pulsonics tool. The Pulsonics tool creates alternating bursts of fluid that help to break and mobilize many types of near-wellbore damage, by means of circumferential harmonic forces acting on the formation face or casing.

Initial results on two test wells showed good improvement. The procedure involved moving the tool through the completion interval on coiled tubing while injecting treatment volumes ranging from 2,400 gal to 5,800 gal of 15% HCL per well, depending on net pay. Following stimulation, data were again collected to create deliverability curves.

Stimulation improved deliverability dramatically. The combined post-stimulation deliverability of the TC wells increased from 70 MMcfd below expected deliverability, to only 2.2 MMcfd below. By contrast, the UB wells changed from 15 MMcfd below expectation to 105 MMcfd above expectation. When comparing results on the basis of deliverability per foot of net pay, similar results were seen, Table 1.

A normalized plot, Fig. 1, summarizes the change in deliverability for the entire process. On this plot, the cumulative total deliverability is normalized to 1.0; values above 1.0 indicate total deliverability is better than expected, and values below 1.0 are poorer than expected. The TC wells delivered only 47% of the expected flowrate after drilling, but that increased to 98% after stimulation. The story is dramatically better for the UB wells.

Fig. 1. The UB wells delivered 83% of the expected rates, but stimulation increased that to 214% of expected deliverability.

The question of why the UB wells performed much better than the TC wells and double the expectation is part of an ongoing analysis. The relatively large number of wells involved is considered sufficient to eliminate bias from back pressure curve interpretation, reservoir description anomalies, drilling practices, and perforating practices. All wells were interspersed among existing wells and none were subject to geological, reservoir pressure, or other bias.

Possible reasons for this difference include drilling and perforation damage, flow turbulence, interval of formation face open and exposure to kill fluids and acid, net pay estimation, and perforation effectiveness. Future well tests, buildup analysis, and additional well completions may help determine which of the factors above are most important; this study is in progress and should be completed in the next few months.

RESULTS

On average, the UB wells cost 7% less than the TC wells in this program. Mud costs, casing, and rate of penetration all reduced the UB costs compared to the TC wells. Further cost reductions may be possible by using coiled tubing rather than jointed pipe to drill the open hole, which could reduce one of the possible causes for formation damage.

UB wells outperformed TC wells in this application, and UB is now the preferred completion type for future wells. Analysis is ongoing to understand the reasons for the greater-than-expected UB well deliverability, when stimulated with the Pulsonics tool. Even greater improvements to deliverability and possible cost reductions are expected with changes to the drilling and completion program. A future planned drilling project will compare coiled tubing-drilled UB completions to the current UB completions.  

THE AUTHORS
	Randall Cade is US chief reservoir engineer for Weatherford’s SURE group, which focuses on reservoir applications of underbalanced drilling. Prior to joining Weatherford, he held a variety of positions over 19 years focused on reservoir engineering, horizontal and multilateral well technology, and operations in the US and internationally, with ARCO and other companies. Cade earned a BS in Engineering from the Colorado School of Mines in 1983.
	Russ Frame is a system engineer in the Storage Operations department of Kinder Morgan, Inc. Russ has over 18 years’ experience working in gas storage as a reservoir engineer and drilling engineer. Prior to Kinder Morgan, Frame was a production engineer for 10 years with Tenneco Oil Co. E&P and four years for Natural Gas Pipeline Co. Russ earned a BS in Engineering Geology from Kansas State University in 1970.
	Robert Wallace is director of Gas Storage Operations for Kinder Morgan, Inc. He has over 25 years’ experience in gas storage as a reservoir engineer and drilling engineer. Wallace has been a leader in developing and applying underbalanced completion technology to the gas storage industry. He earned a BS in Petroleum Engineering from University of Missouri at Rolla in 1980.