PROPER BIT DESIGN IMPROVES PENETRATION RATE IN ABRASIVE HORIZONTAL WELLS

Richard J. Gentges
ANR Pipeline Co.
Detroit

Overall drilling penetration rates nearly tripled, and drill bit life nearly doubled compared to conventional bits when specially designed natural diamond and polycrystalline diamond compact (PDC) bits were used during a seven-well horizontal drilling program.

The improvement in drilling performance from better-designed bits lowered drilling costs at ANR Pipeline Co.'s Reed City gas storage field in Michigan.

Laboratory tests with scaled down bits used on abrasive cores helped determine the optimum design for drilling the gas storage wells. The laboratory test results and actual field data were used to develop a matrix-body natural diamond bit, which was later modified to become a matrix-body, blade-type polycrystalline diamond compact bit. This bit had excellent penetration rates and abrasion resistance.

BACKGROUND

Horizontal drilling technology has only recently been applied to natural gas storage reservoirs. According to a September 1991 survey funded by the Gas Research Institute, only about 76 horizontal gas wells had been drilled worldwide by that time. Since then, approximately 15 horizontal gas storage wells have been drilled in the U.S. alone, 7 of which have been drilled by ANR Pipeline Co.

ANR Pipeline began evaluating the suitability of horizontal wells in its storage operations in June 1990 as part of an internal study on costs associated with various deliverability restoration techniques. the study suggested that ANR's gas storage fields are well suited for horizontal wells.

Drill bit selection was found to be a critical factor in the success of a horizontal drilling program. Drilling penetration rate and drill bit life have an obvious and significant effect on overall drilling costs.

Drill bit selection typically proceeds by trial and error, often at great expense. To control costs and eliminate some of the guesswork associated with selecting the most appropriate bit, ANR contracted Slimdril International Inc. in Houston to assist with bit selection. Slimdril also provided the directional drilling services for ANR's first horizontal gas storage wells.

ANR's first two horizontal wells, RC-124 and RC-123, were drilled in 1991 in the Reed City storage field in Osceola County, Mich. Gas is stored in the Michigan Stray sandstone formation in this field. The Stray sand is a medium-fine grained Mississippian-age sandstone consisting predominantly of quartz with relatively minor amounts of calcite and clay minerals. The formation is considered only moderately hard but is relatively abrasive. Formation porosity averages 15%, permeability ranges about 100-400 md, and conate water saturation is approximately 20-25%. The formation thickness averages about 20 ft.

The planned well configuration consisted of 8 5/8-in. surface casing set and cemented to a depth of 800 ft. The radius section, with a 400-ft radius of curvature, was planned to be drilled with a 71/2-in. bit with 5/.-in. production casing set and cemented into the storage zone at about 1,200 ft true vertical depth. The plan included drilling a horizontal section of approximately 1,000 ft with a 4/2-in. bit; the completion would be open hole. The drilling window, or true vertical depth tolerance within the Stray, was 10 ft.

Because of the slim well bore and the lateral stresses imposed on a bit during horizontal drilling, Slimdril suggested using fixed cutter drag-type bits. Drag bits can provide superior side-tracking capability compared to conventional roller cone bits in this type of application, and drag bits eliminate the risk of a lost cone because of bearing failures and other problems.

BIT SELECTION

The selection process for a specific type of drag bit began with a review of historical bit runs from wells drilled in the Stray. Most Stray wells in the Reed City field were drilled with cable tools and completed open hole; however, a few wells were drilled with tricone insert bits, such as the F2 and F3 series manufactured by Smith International Inc.

Based on the historical bit records, Slimdril recommended a testing program using 4-in. core samples from the Stray to establish the drilling performance of natural diamond and thermally stable man-made diamond (TSD) bits. Polycrystaline diamond compact (PDC) bits were not investigated initially because the preliminary review suggested the Stray was too abrasive for this type of bit.

NATURAL DIAMOND BITS

ANR packaged and shipped 18 ft of core from prior drilling programs in the Stray to Slimdril in Houston. There, the core was cataloged and reassembled so that lithologic variations within the reassembled core were consistent with actual reservoir conditions. The core was then cemented into 3-ft long sections of PVC tubing.

Slimdril manufactured 2-in. diameter prototype drill bits for the tests. The TSD bit was an HT-1 design containing 48 TSD cutters (Fig. 1). The natural diamond bit was an ND-1 design containing 22 carats of natural diamond stones and nine TSD cutters in the core area of the bit (Fig. 2). Both the HT-1 and ND-1 were matrix-body bits with natural diamonds set into the gauge pads to reduce the tendency of drilling an undergauge hole as the bit wears.

The objective of the test was to determine which cutting structure would provide the maximum rate of penetration (ROP) and maximum resistance to the abrasive formation. The data recorded during the tests included flow rate (gpm), input pressure (psi), displacement (in.), weight on bit (lb), and ROP (ft/hr and instantaneous).

The purpose of gathering these data was to establish the drilling performance characteristics of each bit design. Both test bits drilled a total of 9 ft of core.

The 2-in. TSD bit achieved penetration rates of 50-200 ft/hr with initial bit weights of 1,000 lb. However, the TSD bit quickly developed large wear flats and required increased weight to maintain ROP. As wear progressed, vibrations increased and led to more frequent motor stalling. At the conclusion of the test, the TSD cutters had become rounded and dull, which resulted in the need for increased weight to maintain ROP.

Based on the test results and the significant wear, the TSD bit was judged unsuitable for drilling long lateral sections of the Stray. The TSD bit was believed suitable, however, for the radius section, which largely consisted of shale, anhydrite, and dolomite. This zone was believed to be less abrasive than the Stray.

The natural diamond bit drilled with penetration rates of 100-200 ft/hr during the test with bit weights of approximately 2,000 lb. Although greater weight was required to achieve this ROP, the ND-1 had little vibration, low overall wear, and less tendency to stall. Wear flats were not observed, and there was no evidence of polishing. A direct correlation was found between bit weight and ROP. The high ROP and durability of the ND-1 made it well suited for drilling the Stray.

Scaled up 7 1/2-in. versions of the TSD bit and 4 1/2-in. versions of the ND-1 bit were subsequently manufactured to drill ANR's first horizontal gas storage well.

FIELD RESULTS

Drilling the radius section of ANR's first horizontal well, RC-124, revealed the TSD bit was not well suited for the shale sections above the storage zone. Two TSD bits were required to drill 721 ft. Penetration rates in the radius section averaged approximately 6.5 ft/hr overall but were only about 3-4 ft/hr in the share sections. PDC bits, initially judged unsuitable for drilling the abrasive Stray, were considered an option for drilling the shale sections up hole.

A 7 1/2-in., eight-bladed, matrix-PDC bit was designed and manufactured to drill the radius section of ANR's second horizontal well, RC-123. The PDC bit drilled 521 ft of radius section in two runs. Average ROP during the two runs was 14.7 ft/hr - double the ROP achieved with the 7 1/2-in. TSD bit. The PDC bit was graded reusable after two runs.

Five 4 1/2-in. ND-1 bits were used to drill the lateral sections of the RC-124 and RC-123 wells. Average footage drilled per bit was 355 ft in 19 hr for an average ROP of 18.7 ft/hr. Dull grade inspection of four of the bits revealed that the overall cutting structure exhibited minimum diamond wear and polishing. Matrix washing was extensive enough, however, to cause some diamond loss in all four bits particularly in the nose region. The fifth bit was still drilling and was graded reusable when it was pulled because the well reached total depth. The gauge area on the bits showed evidence of matrix wiping, but no gauge diamonds were lost. All five bits had evidence of back-reaming damage.

The laboratory and field testing in 1991 confirmed the ability of natural diamond bits to drill the Stray. Because of the overall success of the program, ANR planned to drill an additional five horizontal wells in 1992. Based on the initial data and dull grade inspection of the field-tested bits, further laboratory testing was conducted in early 1992 to improve drilling performance and to evaluate the feasibility of drilling the Stray with PDC bits.

NEW BIT DESIGNS

In early 1992, additional drill bit tests were performed on core samples similar to those drilled during the previous tests. The 45 ft of core reassembled for testing consisted of 25 ft of core from the Stray and 20 ft of composite core. The composite core consisted of dolomite, shaley sandstone, and anhydrite from the radius section. The test assemblies were constructed identical to those used during the 1991 laboratory tests.

The primary objective of these tests was to evaluate the performance of two additional bit designs under consideration for the next horizontal wells:

• A natural diamond bit on which the TSD center cutters and 50% of the natural diamond cutting structure were replaced with natural diamond cube stones

• A matrix-body blade-type PDC bit with 8-mm cylinder cutters.

Three bits were manufactured for the tests: one 2-in. diameter natural diamond cube stone bit (ND1C) and two 2,-/s-in. diameter matrix-body bladed PDC bits (MP-3).

Fig. 3 shows the ND1C bit design with three primary blades containing a total of 15 carats of Congo cube stones (three stones per carat) and 14 carats of West African natural diamond stones (7-8 stones per carat). The matrix-body PDC bits were three-bladed designs with three secondary gauge pads, and natural diamonds were set in the gauge pad area of each blade (Fig. 4). The cutting structures were 8-mm diameter PDC cylinders. Each bit was designed with three water ports.

The test procedure and data recorded were identical to those from the 1991 tests. Flow rate, input pressure, displacement, weight on bit, and ROP were recorded using transducers connected to a personal computer. The data acquisition rate was one data point per second. Photographs were taken of each bit before and after the tests to evaluate bit wear.

The ND1C bit drilled three core sections, each about 3 ft long. The first core section was drilled at a rate of 30 ft/hr with 1,000 lb weight on bit. During drilling of the second core, the ROP increased to 120 ft/hr primarily because of increased diamond exposure as the carbide covering the diamonds wore. The ROP for the third core section decreased to 100 ft/hr because of the development of ",ear flats on the diamond cubes. An inspection after the tests revealed the natural diamond cube stones had evidence of fracturing and premature development of wear flats. Based on these results, the ND1C was judged unsuitable for drilling the Stray.

The MP-3 bit drilled the first Stray core section with penetration rates of 200 ft/hr with 750 lb weight on bit. The bit developed a slight chamfer at its breakover point. This wear required an increase in weight on bit to 1,500 lb to maintain ROP while the second and third core sections were drilled. Bit wear overall was low and concentrated in the breakover region of the bit.

After having drilled three core sections, the MP-3 bit maintained penetration rates equivalent to the sharp ND1C bit and had ROPs 50% greater than those of the ND1C bit after it drilled three core sections.

The second MP-3 bit drilled with penetration rates of 280 ft/hr in the shaley sandstone, dolomite, and anhydrite core samples. The bit drilled 21 ft of composite core. Again, wear was predominantly limited to the breakover region of the bit, indicating that these core sections were somewhat abrasive.

Results from these laboratory drill tests demonstrated that PDC bits could deliver improvements in both drill bit life and ROP, thus refuting the initial hypothesis that the abrasiveness of the Stray sandstone would cause premature failure of PDC bits. Wear patterns in the breakover region of both PDC bits indicated that design modifications to enhance cutter density in this region of the bit would improve bit life.

The redesigned bit (MP-9) had nine blades (three primary and six gauge pad) and three water ports (Fig. 5). To increase the erosion resistance of the body and abrasion resistance of the primary cutters, the bits incorporated variable back rake, variable cutter density, increased cutter exposure, greater blade relief between cutters, and a new series of diamond cutters.

FIELD RESULTS

These redesigned heavyset PDC bits were used in the radius and horizontal sections in five horizontal wells drilled by ANR in 1992. The radius sections were drilled with 7 7/8-in. MP-9 heavy-set PDC bits, and the horizontal sections were drilled with 4 3/4-in. bits of the same design.

The overall rate of penetration for both the radius and lateral sections of the five horizontal wells drilled in 1992 was 19.1 ft/hr compared to 10.5 ft/hr for the two horizontal wells drilled in 1991. The 1992 ROP increased by 821/c over the previous year.

Bit life also increased significantly. The average footage drilled per bit in 1992 was 759 ft compared to 414 ft in the 1991 drilling, program, an increase of 84%. Fig. 6 illustrates the overall ROP in sequential order for each of the seven horizontal wells drilled to date.

ACKNOWLEDGMENT

The author thanks the management of ANR Pipeline Co. for permission to publish this article and the corporate communications staff for its review and helpful suggestions.

Copyright 1993 Oil & Gas Journal. All Rights Reserved.