PDC bit hydraulics design, profile are key to reducing balling

P.R. Hariharan
The University of California
Berkeley

J.J. Azar
The University of Tulsa
Tulsa

PDC bits with ribbed or open-face hydraulic designs and those with flat or rounded profiles tended to ball more often in the bit balling experiments conducted. Experimental work also indicates that PDC hydraulic design seems to have a greater influence on bit balling tendency compared to bit profile design.

The nondimensional ratio, Rd, which is the ratio of the rotary torque to the product of the WOB and the bit diameter, is a good indicator of bit balling and its degree. Correlation of specific energy with the ratio Rd has indicated that this ratio could be useful in detecting and describing the degree of bit balling.

The PDC cutter was introduced in 1973, with the concept of cutting rock by shear action, paving the way for the development of the PDC drill bit. With advantages of higher rates of penetration (ROP) and longer life, PDC bits have gained prominent use for drilling soft and medium-soft formations.

A large amount of drilling occurs in shale and other clay-bearing rocks, which readily react with water and swell. Although the shales and clay are soft enough to be drilled by PDC bits, swollen clays can become plastic and stick to the cutters and the body of the drill bit, causing bit balling. PDC bits are especially susceptible to bit balling because of their shear cutting action and the mechanism of chip generation.

The problems associated with bit balling include the following:

• ROP drops because of the buildup of a mass of clay/shale cuttings on the drill bit, preventing the cutters from contacting the formation.

• The built up mass at the bit causes it to behave similar to a piston in a cylinder, leading to surge and swab pressures while tripping.

There are five main factors that affect bit balling: formation type, drilling fluid, drilling hydraulics, bit design, and confining pressures.

Substantial research has been undertaken to identify the particular types of formations that ball readily. Balling mainly depends on the cation exchange capacity (CEC), which further relates to the presence of illites and montmorillonites. These clays are described as swelling or reactive because of their property to adsorb water and swell and stick to the bit body, stabilizers, etc.

Elaborate studies have been conducted to prevent bit balling using different chemical additives in the drilling fluid. Most of the chemicals work by encapsulating the drilled cuttings to prevent their direct interaction with the metal bit body. Most of the success in this direction has been specific to the type of shale drilled. Furthermore, most of the chemicals are somewhat toxic, severely limiting their use, especially offshore.

Another approach has been to increase the drilling hydraulics to provide better cleaning at the bit. Although studies indicated that increasing bit hydraulic horse- power led to better bottom hole cleaning, there was an upper limit to the level of hydraulics beyond which no further improvement was observed. This occurred because of the presence of many stagnant and vortex zones under the bit where cuttings were trapped and stuck.¹

Another important factor influencing bit balling has been bit design. PDC bit design incorporates numerous complicated combinations of bit profile, bit hydraulic design, number of cutters, back-rake and side-rake angles, bit cutter standoff distance, cutter shape, nozzle location, and nozzle direction. Relatively little research has been focused on bit design.

Background

An equation for specific energy showed that it could be used to describe the efficiency of the drilling process by examining the amount of energy spent in drilling a unit volume of rock.² This concept of specific energy has been used herein to correlate with the parameter Rd, a parameter to quantify the degree of balling.

One of the first PDC single-cutter studies showed that shale becomes more plastic and ductile with increasing confining pressure, making it more difficult to drill.³ Other single cutter studies under simulated pressurized conditions in Mancos shale concluded that the balling tendency decreased with increasing back-rake angles.^4-6 Imparting a side-rake angle to the cutter aided mechanically in directing the generated cuttings away from the bit body towards the periphery of the bit. In another study, bit balling under confined pressure conditions for both Pierre and Mancos shale increased.⁷ In Mancos shale, the dilatancy characteristics of the rock (which caused the rock pore pressure to decrease at the point of shearing by the PDC cutter causing the cuttings to stick to the face of the cutter) gave rise to bit balling.

Limited studies address the influence of bit design on balling.^{1 8} Bladed bits have a lesser tendency to ball compared to other designs.⁸ Shallow cone profile bits have a greater tendency to ball. Although the study stressed the importance of bit design on balling, very few bits were used to draw any consistent conclusion. Warren, et al., also showed the dependence of bit balling on PDC bit profile on the basis of drilling experiments conducted in shale, limestone, and sandstone.¹ They concluded that flat faced/flat profiled PDC bits are likely to ball sooner than bladed bits. No details of torque variation during the onset of balling were discussed. Also, the range of bits used for the experiments was limited.¹

Cheatham, et al., indicated that the degree of balling could be identified by examining the ratio of the bit torque to the weight on bit (WOB).⁹ The study was restricted, however, to roller-cone bits alone. An "egg-beater" bit was designed specially for drilling in soft formations.⁷ The design concentrated on the size of cutters and the higher standoff distance and the peeling action imparted by the directed nozzles. Although the bit did not ball in the laboratory experiments, the bit has not been widely deployed as an antiballing bit.

The efficiency of the drilling process was examined by the nondimensional parameter, m, which is the bit specific coefficient of sliding friction.¹⁰ This parameter is the ratio of the rotary torque to the product of WOB and bit diameter. Although dimensionally is similar to the coefficient of mechanical sliding friction, the measured torque used in the computation consists not only of the frictional force between the cutter and the rock surface, but also some resolved components of the forces acting on the face of the cutter. These experiments were restricted to roller cone bits alone.

A similar nondimensional parameter, Rd, which is indicative of the drilling process but not equal to the coefficient of friction, has been used in this study to describe the degree of bit balling on PDC bits.

Bit selection

The two main aspects of PDC bit design, bit profile and bit hydraulics, affect bit balling the most. Hence, to have a range of bits being mainly used in the field, the PDC bits were selected based on the International Association of Drilling Contractors (IADC) classification of 1987 such that they spanned a wide spectrum of bits from those available.¹⁶ Nine different bits were selected covering the three profiles defined as flat, rounded, and parabolic. In each of the profiles the three different bit hydraulic designs were selected: open faced, ribbed, and bladed. For the flat and parabolic profiles, no bits with open-faced bit hydraulics design were available.

• IADC M315 is a matrix body parabolic profile bit. It has 55 cutters, of which 11 are gauge cutters and the remaining are set on the face of the bit. The bit has six interchangeable nozzles. The nozzle sizes used for the study were five 9/32 in. and one 12/32 in. with a total flow area (TFA) of 0.421 sq in.

• IADC M345 is a matrix body bit with a parabolic profile and a ribbed hydraulic design. It has 64 cutters of which 19 are on the face, 36 on the scribe, and 9 on the gauge area. The nozzle sizes used for the study were three 13/32 in. giving a TFA of 0.388 sq in.

• IADC M615 is a rounded profile matrix body bit with five interchangeable nozzles and 1/2-in. diameter cutters. The cutter density is medium. The bit hydraulic design is ribbed, which provides a large junk slot area facilitating the removal of generated cuttings more easily.

• IADC M646 is a matrix body, rounded profile bit with a ribbed hydraulic design. The nozzles used for the experiments were three 13/32 in. giving a TFA of 0.388 sq in.

• IADC S675 is a steel body rounded profile bit. The bit hydraulic design is open faced with no apparent channels and blades for directing the cuttings outward. The nozzles used were three 13/32 in. giving a TFA of 0.388 sq in.

• IADC S914 is a steel studded flat profile bladed bit with erosion-resistant hard surfacing. Of the 28 cutters, 8 are gauge cutters, and the remaining are distributed evenly among the four blades. There were five 10/32 in. nozzles used, giving a TFA of 0.383 sq in. The cutters were 1/2 in. in diameter.

• IADC M946 is a flat profile matrix body bit. The bit hydraulic design is ribbed. The bit has 55 cutters, of which 17 are on the gauge, 24 are on the face, and 14 on the scribe. The cutter size was 1/2 in. in diameter

Experimental setup

The University of Tulsa Drilling Research Projects rig was upgraded for these experiments. Additional experiments were later conducted at Amoco Production Co.'s Tulsa facility (Fig. 1 [12756 bytes]). The circulating system consists of a 1,600-hp mud pump capable of developing 3,000 psi, a high-pressure vessel rated at 2,000-psi working pressure into which the rock-sample is enclosed and drilled, a screen to remove the drilled cuttings, an hydraulic choke arrangement for throttling the return flow and simulating confining pressure. The return line is taken from downstream of the choke at atmospheric pressure back to the suction tank.

The drilling machine consisted of an hydraulic arrangement for applying weight on bit up to 100,000 lb. The values of the WOB used for the tests ranged from 3,000 to 24,000 lb. The rotary system consisted of a variable speed hydraulic motor. The rotary speed used for all the experiments was 120 rpm. The rotary speed was maintained constant for varying WOBs through an automatic feedback arrangement.

The displacement was measured using a linearly variable differential transformer, and from this the rate of penetration was calculated. The rotary torque was measured by a torque meter on the drillstring. The rock used was Catoosa shale, mainly because of its local availability and its low cost compared to other shales like Pierre and Mancos.

Because the drilling fluid was not a variable in this study, it was kept relatively simple by using a basic bentonite water-based fluid of 9.1 ppg and about 13 cp viscosity.

Procedure

The Catoosa shale cores used for the tests were 15-in. in diameter and 36-in. long encased in Plexiglas coating. Just before the test, the top Plexiglas was cut, and the rock specimen was enclosed in a metal core barrel. The entire cylinder carrying the rock was then lowered into the pressure vessel.

The PDC bit was made up with the string, the top flange of the pressure cell was bolted, and the mud pumps started. As soon as the flow rate stabilized and the required confining pressure was reached, data acquisition began. The WOB was varied for a typical drilling test as follows: 3,000 lb, 6,000 lb, 3,000 lb, 9,000 lb, 6,000 lb, 12,000 lb, 9,000 lb, 15,000 lb, and 12,000 lb.

After the core was completely drilled through (except for the remaining last few inches), the bit was lifted off bottom, the rotary stopped, and the pump shut down. The pressure cell was then opened and the bit removed. The bit was lightly cleaned, examined for visual evidence of balling, and then photographed. The drilling parameters measured were WOB, rotary speed, rotary torque, displacement, time, flow rate, and bore hole pressure. The computed parameter was ROP.

Bit balling detection mainly centered around the value of the rotary torque during drilling. The literature showed that the relation between rotary torque and WOB is fairly linear until a limit from inadequate hydraulics makes the curve droop down.11-14 To have a quantitative indication, it was felt that the ratio of rotary torque to the product of the WOB and the bit diameter could be used to describe the effectiveness of the drilling process as well as the degree of bit balling.

The bit balling was detected in the following manner:

• Set WOB at W1, and record the corresponding torque T1.

• Raise WOB to W2, and record corresponding torque T2 (all other parameters are kept constant).

• Lower WOB back to W1, and record torque T3.

• Compare T1 and T3.

Based on the values, the following conclusion was reached: If T1 = T3 (within the ranges of experimental error), then the bit has not balled up; if T3 < t1, then the bit has balled up.

If the bit had not balled up, the WOB was then raised to the next higher value and Steps 1-4 repeated. The test was terminated when the core was almost completely drilled so that maximum number of data points could be obtained. At the conclusion of the test, the pressure cell was opened, the bit lightly cleaned and examined, and the bit was photographed.

Sixteen drilling tests were conducted: eight were conducted at 1,000-psi confining pressure and eight at 2,000 psi. Table 1 [43418 bytes] shows the complete test matrix and all the bits identified by their IADC codes. Table 2 [73467 bytes] shows the conditions of drilling hydraulics, and Table 3 [50365 bytes]shows the brief test results.

No bit balling

The results for bit M315 at 1,000-psi confining pressure and a flow rate of 350 gpm are discussed. Fig. 2 [13360 bytes] shows the variation of the WOB with the drilled depth. WOB was varied from 6,000 lb to 9,000 lb and then back to 6,000 lb. The torque corresponded by varying from 1,500 ft-lb to 1,800 ft-lb and then back to 1,500 ft-lb. This ability of the torque to recover back to the original value indicated that there was perfect cleaning at the bit, and balling did not occur. Furthermore, when the WOB was increased to 12,000 lb, the torque increased to 3,000 ft-lb. At this point, the core was totally drilled through, and no further data could be obtained. After the experiment, the bit was found clean and without any balling whatsoever. Fig. 3 [11794 bytes] shows the variation of the ROP and rotary torque vs. WOB. Both indicate a linear variation signifying that the drilling process proceeded without any bit balling.

The results for bit M315 at 2,000-psi confining pressure and 350 gpm were similar. ROP decreased at the higher confining pressure because of an increase in the rock strength. Again, both ROP and rotary torque increased nearly linearly with WOB.

The bit did not ball in this test. Fig. 4 [13715 bytes] shows the variation of Rd with depth for the drilling test at 1,000 and 2,000 psi. Rd remained constant through both tests. The average value of Rd was about 0.3.

Bit balling

The results for bit M946 at 1,000 psi confining pressure and 200 gpm flow rate are discussed in terms of bit balling. Fig. 5 [16087 bytes] shows the variation of WOB and rotary torque with depth.

The torque immediately decreased with increased WOB, signifying that the bit was balled up. The maximum WOB that could be applied was 25,000 lb. Numerous data points could be obtained because of the slow drilling rate, which resulted from the bit balling.

Fig. 6 [12613 bytes] shows that the ROP and rotary torque decreased with increasing WOB, signifying that the bit was totally balled up and was not drilling ahead. The results were similar when the confining pressure was increased to 2,000 psi. The maximum WOB applied was 25,000 lb, as in the earlier experiment. ROP and rotary torque decreased when the bit balled up. The same trend with rotary torque was also noticed. Rd decreased as drilling proceeded, indicating bit balling in both tests (Fig. 7 [14086 bytes]). Furthermore, the bit was more severely balled up after drilling at 2,000 psi confining pressure (lower Rd).

Rd

Bit balling became more severe at higher confining pressures. The parameter Rd seemed to be an appropriate description of the drilling process and especially as an indicator of bit balling.

For bit M346, the bit balled at both 1,000 psi and at 2,000 psi. The balling was more severe at 2,000 psi, with Rd reaching values as low as 0.025.

For bit M615, Rd was constant at about 0.3 for both 1,000 psi and 2,000 psi tests. The bit did not ball during both tests.

For the first set of experiments at 1,000 psi for bit M645, Rd remained constant for WOB ranging from 3,000 to 12,000 lb. The bit was clean at the end of the test. In a second test at 1,000-psi confining pressure and WOB ranging from 12,000 to 18,000 lb, the bit balled quite severely. At a higher confining pressure, Rd decreased, indicating severe bit balling.

For bit M672, Rd remained constant at 0.3 for the test at 1,000 psi but decreased to 0.1 for the test at 2,000 psi.

For the two tests conducted with bit S914 at 1,000 psi, Rd remained at a high value of 0.25. At 2,000-psi confining pressure, the value decreased to about 0.175. Correspondingly, the amount of balling was extremely small.

R

esults

An interesting observation based on the above set of experiments was that the values of the ratio Rd were always between 0 and 0.4. For a nonballing bit, Rd always tended to be nearer 0.4. These ranges could be a function of the particular type of rock used for the set of experiments.

The variation of Rd from a nonballed condition could be used as an indicator of the efficiency of the drilling process. Moreover, the value of Rd at perfect cleaning conditions could be used to describe some characteristics of the rock itself.

The results of the experiments confirm that the tendency of PDC bits to ball is indeed dependent on bit design. It was also observed that bit balling is frequently irreversible.

Results also indicated that the balling tendency of the PDC bit is more dependent on bit hydraulic design rather than profile. Another important observation was that PDC bits are more likely to ball at higher confining pressures. In other words, these bits are likely to ball at lower WOBs while drilling at higher confining pressures.

For the stated conditions of drilling hydraulics and other drilling parameters, the bit designs that are likely to ball the most are those with a flat profile and with an open-faced bit hydraulic design (IADC M946). Similarly, the bits that are likely to ball the least are those with a parabolic profile and with a bladed hydraulic design (IADC M315).

The tests also indicate that bit hydraulic design is a more dominating factor than bit profile, as far as the effect of bit design is concerned on bit balling. Bladed bits (M315, M615, and S914) seem to be the ideal choice to avoid a bit-balling situation.

The reasons for these observations could be explained as follows. In the case of a flat profiled bit with an open-faced hydraulic design, the cutters are all in the same plane, and the drilled cuttings accumulate at the same place making prompt flushing very difficult. This effect is overcome in those designs with larger and fewer cutters. They help by providing a larger standoff between the cutter and the bit body, facilitating a better bottom hole cleaning and thus preventing balling. The effect of this parameter was out of the scope of the present study, but it would be worthwhile to examine the beneficial effects of a higher standoff distance in future bit balling studies.

Similarly, the reason for the absence of bit balling in parabolic profile bits may be attributed to the fact that cuttings are generated in these bits at different planes, and there is no interference between the cuttings at the different planes, thus making it easier for them to be flushed away promptly. For bladed bits, the bit balling problem is overcome to a certain extent by the junk slots between the blades where the cuttings are directed and allowed to accumulate before being flushed away.

The tendency of the shale to ball more readily at the higher confining pressure could be due to the fact that plastic formations become more ductile at higher confining pressures from the normal brittle behavior at lower confining pressures.^{3 15} This ductile behavior at higher confining pressures causes longer and more plastic/sticky cuttings to be generated. Because of the larger cutting force needed to generate these cuttings, the water from the now ductile and plastic rock is squeezed out, causing the chip to stick differentially to the bit body and cause bit balling. However, simultaneous chemical reasons, such as the affinity of the surface negative charges on the newly generated chip surfaces to the positively charged surface of the metallic cation on the bit body, are also responsible for the shale chip cuttings sticking to the bit.

Hence, balling is a complicated process consisting of a chemical as well as a mechanical effect.

Acknowledgment

The authors appreciate the support provided by the member companies of the University of Tulsa Drilling Research Projects for this research.

References

1. Warren, T.M., and Armagost, W.K., "Laboratory Drilling Performance of PDC Bits," Paper 15617, SPE Annual Meeting, October 1986.

2. Teale, R., "The Concept of Specific Energy in Rock Drilling," International Journal of Mechanical and Mining Science, Pergamon Press, Vol. 2, pp. 57-73, 1965.

3. Cheatham, J.B., and Dalienls, W.H., "A Study of Factors Influencing the Drillability of Shales: Single Cutter Experiments with STRATAPAX Drill Blanks," Journal of Energy Resources Technology, September 1979, pp. 189-95.

4. Melaugh, J.F., and Salzer, J.A., "Development of a Predictive Model for Drilling Pressurized Shale with STRATAPAX blank bits," 1981.

5. Glowka, D., "Optimization of Bit Hydraulic Configurations", Journal of the Society of Petroleum Engineers, February 1983, pp. 21-32.

6. Huang, H.O., and Iversen, R.jpg., "The Positive Effects of Side Rake in Oilfield Bits Using Polycrystalline Diamond Compact Cutters," Paper 10152, SPE Annual Meeting, October 1981.

7. Zijsling, D.H., and Illerhaus, R., "Eggbeater PDC Drillbit Design Concept Eliminates Balling in Water Base Drilling Fluids," Paper 21933, SPE/IADC Annual Drilling Conference, Amsterdam, March 1991.

8. Koskie, E.T., and Appen, H.jpg., "A PDC Solution to Drilling Sticky Formations With Noninhibited Water Base Drilling Fluid: Experience in the Provincia Field in Colombia," Paper 14430, SPE Annual Technical Conference and Exhibition, Las Vegas, September 1985.

9. Cheatham, C.A., and Nahm, J.J., "Bit Balling in Water Reactive Shales During Full Scale Drilling," Paper 19926, Society of Petroleum Engineers/International Association of Drilling Contractors Annual Drilling Conference, February 1990.

10. Pessier, R.C., and Fear, M.J., "Quantifying Common Drilling Problems with Mechanical Specific Energy and a Bit Specific Coefficient of Sliding Friction," Paper 24584, SPE Annual Technical Conference and Exhibition, Washington, D.C., October 1992.

11. Kerr, C.J., "PDC Drill Bit Design and Field Applications Evolution," Journal of Petroleum Technology, March 1988, pp. 327-32.

12. Snedden, M.V., and Hall, D.R., "Polycrystalline Diamond: Manufacture, Wear Mechanisms, and Implications for Bit Design," Journal of Petroleum Technology, December 1988, pp. 1593-1601.

13. Steiger, R.P., and Leung, P.K., "Quantitative Determination of the Mechanical properties of Shakes," Paper 18024, SPE Annual Technical Conference and Exhibition, October 1988.

14. Feenstra, R., "Status of Polycrystalline Diamond Compact Bits: Part 2-Applications," Journal of Petroleum Technology, July 1988, pp. 817-21.

15. Zijsling, D.H., "Single Cutter Testing-A Key to PDC Bit Development," Paper 16529, Offshore Europe, September 1987.

16. Winters, W.J., and Doiron, "The 1987 IADC Fixed Bit Classification System," SPE/IADC Annual Drilling Conference, March 1987.

17. Hariharan, P.R., "Effect of PDC Bit Design and Confining Pressure on Bit Balling Tendencies while Drilling Shales Using Water Base Muds," MS thesis, University of Tulsa, 1993.

The Authors