TECHNOLOGY Multilateral well enhances gas storage deliverability

Matt C. Rowan, Michael J. Whims
ANR Storage Co.
Detroit

Multiple lateral drilling from an existing well bore can limit environmental exposure and overall drilling risk while optimizing gas storage field performance.

In 1994, an existing high-angle directional well was reentered, and five nearly horizontal fingers were created.

The Excelsior 6 gas storage field in Michigan comprises two Niagaran pinnacle reefs which are pressure connected through an interreef facies of low permeability. Each reef, known as the Excelsior 6 gas field (EX6) and the East Kalkaska 1 gas field (EK1), independently produced native gas through pressure-depletion drive.

In 1980, seven wells, six of which are in the EX6 field and one in the EK1, were drilled to convert both fields to gas storage service. Two additional wells were added to the EK1 in 1990.

Both reservoirs underlie an environmentally sensitive wetland area. The reservoirs were developed with wells drilled directionally from a central high land area overlying the eastern-most edge of the EX6 reservoir, resulting in relatively long (up to 4,000 ft) horizontal displacements to reach the EK1 reservoir to the west.

Accurate seismic delineation of the reefs had been hampered by shallow glacial deposits and difficulty in crossing the wetlands.

Geology

The reservoir gas trapping mechanism is provided by nearly impermeable dolomite of the Gray Niagaran formation at the reef base, dense carbonates, and evaporites around the reef flanks, and an evaporite cap rock of the A-2 evaporite formation which averages 67-ft thick. The base of the reef complex is at a depth of 5,800 ft and covers 142 acres for the EX6 and 87 acres for the EK1. Gross thickness of the EX6 and EK1 reefs are 369 ft and 398 ft, respectively.

Volumetric distribution between the EK1 and EX6 is 31% and 69% of the total gas capacity, respectively. Although total capacity of the two reefs is slightly in excess of requirement, because of performance variations between the two reefs, the EK1 could not be completely filled or withdrawn during the storage cycle. Gas movement between the two reefs is very slow and is considered negligible as long as differential pressures between the reefs are held at minimum levels.

Reservoir modeling indicated that the EK1 cumulative withdrawal volume could be increased significantly if the performance variation between the two reefs could be eliminated. A decision was made to increase the deliverability potential of the EK1 reef through maximum reef penetration at minimum cost.

The geologic review of the available seismic data and of the open hole logs indicated the southwestern portion of the EK1 to be salt plugged; the EK1 to extend to the northwest; the gas/water contact to be at 5,800 ft; and the best porosity development to lie above 5,700 ft.

As a result of this analysis, a candidate well should be designed to exploit the northwestern portion of the reef while maintaining a subsea depth above 5,700 ft. The EK1-2 well was selected because it provided the most flexibility for the drilling operations to exploit the northwestern part of the EK1 reef successfully.

Well design

Additional new well placement at the drill site was difficult because of limited area. Economics and seismic uncertainties suggested using an existing open-hole well bore to drill a number of exploratory laterals into the EK1 reservoir for the desired improvement.

The original EK1-2 well had 1338-in. casing set in a 1712-in. hole at 714 ft true vertical depth (TVD). A 1214-in. hole had been drilled to 4,522 ft TVD and 958-in. casing set. An 834-in. hole had been drilled to finish the build section to the desired tangent angle, which was held until the top of the cap rock was reached.

Once the cap rock had been identified by open hole logs, the 7-in. casing was run and set. Then, the 10-ppg salt-saturated mud was displaced with freshwater. A 618-in. hole was drilled to total depth. The well was acidized and completed open hole in 1980.

A schematic of the lateral extensions is shown in Fig. 1 (26365 bytes). The first extension was designed to be the main conduit for the subsequent extensions and to improve drainage in that area of the reef.

To maintain drilling operation flexibility in exploiting the northwestern part of the reef, the main conduit was designed as the shallowest and southern-most extension. The three subsequent extensions were designed to be evenly spaced along the initial extension to limit well bore interference.

The intersection of the initial extension and 5,600 ft plus the anticipated location of the northeasterly most extension were used as the end points for spacing the subsequent extensions.

Drilling program

The drilling program included the following procedures:

• Set a cement kick-off plug in the original open-hole well bore.

• Orient the bottom hole assembly (BHA) to sidetrack out of the side of the original well bore.

• Time drill (no weight on bit) the kick off.

• Slide drill (drillstring not rotating) the build section.

• Change the BHA and directionally drill the main extension.

• Spot a viscous pill to the next extension kick-off point.

This procedure would then be repeated for three subsequent extensions spaced at progressively shallower depths along the initial extension.

The primary hole-cleaning parameter was flow rate. Hole cleaning models were used to predict minimum flow rates and optimal rheologies for hole cleaning. Flow rates of 200-300 gpm were used to drill the extensions.

Maintaining turbulent flow along the extensions with mud that exhibited excellent shear characteristics provided the best hole cleaning. High-viscosity pills were also used to assist hole cleaning.

Rotating the drillstring and backreaming were used sparingly to limit drilling shock and fatigue of the bent-housing motor. Before tripping, circulation continued until the return of cuttings had minimized.

Becoming sidetracked in the open hole, bit life, and rate of penetration were the parameters used for bit selection. A diamond sidetrack bit and polycrystalline diamond compact bit were determined to be uneconomical when compared with a custom mill bit and with an enhanced roller cone bit, to drill the open-hole sidetrack and the extensions, respectively. Bit hydraulic models were used to optimize bit nozzle size with hole cleaning requirements.

Low-speed, high-torque motors that could accommodate the required hole cleaning flow rate were used. To aid the flow rate requirement and maximum horsepower output, the stator of one motor was bored out. The length from the bend in the motor to the bit box was a secondary concern. The contractor preferred to limit this distance to realize better directional control.

A 312-in., 15.5 lb/ft drillstring was used to drill the extensions. Casing wear was of high concern but was minimized by the use of drill pipe rubbers. Hole cleaning requirements for the 618-in. hole made triplex pumps desirable. Triplex pumps offer higher working pressure and fewer fluctuations than duplex pumps. Rig power requirements were designed on 100,000-lb overpull and a 10% safety factor. Maximum power was needed if backreaming the end of the second extension would be needed. Rig maintenance and safety were also analyzed carefully.

A foam acid stimulation with 1,430 cu ft of N2 and 6,750 cu ft of HCl acid was performed following the drilling operation. Because of mechanical problems with setting an inflatable packer, however, individually acidizing each new extension was not possible. The treatment had to be bullheaded.

Results

About 22% of the total time that the drilling operation equipment was on location was spent removing original packers and completion equipment, and 78% of the time was spent on drilling operations. Therefore, two separate analyses are presented on the detailed time requirements for various aspects of the project.

Table 1 (28164 bytes) presents the time distribution for the total number of hours required to complete the entire project. Table 2 (28833 bytes) differentiates the total number of hours for completion equipment removal and drilling.

Fig. 2 (16035 bytes) shows the percentage distribution of the items in Table 2 (28833 bytes), to qualify better the time and costs associated with the drilling part of the project. Of the 556 hr spent on drilling operations, 50% of the time was spent drilling, 24% tripping, 12% preparing equipment, and 14% on miscellaneous events.

Extension A, the initial extension, was directionally drilled 1,116 ft. Extensions B, C, and D were open-hole sidetracked from Extension A and directionally drilled 560 ft, 724 ft, and 513 ft, respectively. The cement kick-off plug set in the original well bore was drilled out. The total footage of open hole completion is 3,250 ft, including the original well bore.

The current engineering interpretation indicates a significant increase in withdrawal capacity from the EK1 reef as a result of the multiple laterals.

The drilling of multiple laterals from a single well bore was successfully achieved. Preliminary analysis from this past winter withdrawal indicates that the multiple lateral project was successful in improving the cycling capacity of the EK1 reef.

Without further testing, it is not known if the well is operating at 100% efficiency. Based on the type of stimulation performed, it is possible that some well bore damage associated with the drilling operation still exists. If pressure buildup information verifies the existence of damage and a successful stimulation is performed, additional performance improvement can be anticipated.

The Authors