|Publisher||Society of Petroleum Engineers||Language||English|
|Content Type||Conference Paper|
|Title||Bit Balling in Water-Reactive Shale During Full-Scale Drilling Rate Tests|
|Authors||Cheatham, C.A., Sperry-Sun Drilling Services; Nahm, J.J., Shell Development Co.|
SPE/IADC Drilling Conference, 27 February-2 March 1990, Houston, Texas
|Copyright||Copyright 1990, IADC/SPE Drilling Conference.|
This paper demonstrates a useful means of studying bit balling in laboratory drilling tests. Full-scale tests were conducted using a new 6-1/8 in. (156 am) tricone insert-type bit, realistic muds, and Pierre shale, which is similar to water-reactive shales which are apt to cause balling in the field. Balling ranged from slight to severe with an uninhibited water-base mud, and did not occur with an oil-base mud. The ratio of bit torque to weight-on-bit was a reliable indicator for detecting the degree of balling. The results show that proper modeling of bit balling in the laboratory must address the degree of overbalance which results from insufficient mud weight in the field. They also indicate that increasing mud weight in the field is an effective means to reduce or prevent bit balling if, and only if, balling is caused by mechanical (not chemical) wellbore instability. Although bit balling would become (irreversibly) severe within seconds, it could be prevented by significantly reducing weight-on-bit prevented by significantly reducing weight-on-bit quickly enough after incipient balling.
Slow rate of penetration (ROP) in shale intervals has been identified as a majo drilling problem. Overall drilling costs can be reduced in such slow drilling formations by increasing ROP. An important factor related to slow ROP is bit balling in water-reactive, or "sticky," shale as found, for example, in young sediments in the Gulf of Mexico. Balling occurs when rock-cuttings adhere to bit cutters, which results in a lower ROP. Based on a successful pilot study, we decided to use the full-scale pilot study, we decided to use the full-scale drilling simulator at th Drilling Research Laboratory (DRL) to develop a comprehensive methodology to reduce shale drilling costs. The first step in that effort used Mancos shale, medium-hard shale obtained from surface outcrops. Although some bit balling was observed in the Mancos shale tests, the degree of balling was insufficient to model the water-reactive shales which are apt to cause extensive balling in the field.
Large out crops of Pierre shale were located which, from a mineralogical standpoint, seemed to be a likely candidate for a model study of bit balling. As discussed in the next section, significant differences between Mancos and Pierre shales include the degree of water saturation and compressive strength.
Small plugs (2 inches (51 mm] long by 1 inch (25 mm] diameter) were taken from the large (36 inches (914 am] long by 15-1/2 inches (394 am] diameter) Pierre shale cores which were used for the drilling Pierre shale cores which were used for the drilling tests. Samples were submitted for X-ray diffraction analysis to determine the mineralogy. The results (Table 1) showed that Pierre shale contains 30 to 35 percent clay minerals by weight. The predominant percent clay minerals by weight. The predominant clay is montmorillonite, which exhibits relatively high swelling strain when exposed to water.
A petrophysical core analysis (Table 2) showed that Pierre shale has approximately 23 percent porosity and is 10 percen water-saturated. Attempts porosity and is 10 percen water-saturated. Attempts to measur liquid permeability revealed only that it is essentially impermeable.
Conventional triaxial compression tests were performed on the small plugs. At 2,000 psi (13.8 MPa) performed on the small plugs. At 2,000 psi (13.8 MPa) confining pressure, compressive strength was low (1,200 psi (8.3 Mpa]) and increased only slightly (1,300 psi (9.0 MPa]) at 8,000 psi (55.2 MPa) confining pressure (Table 3). As shown in Table 3, the compressive strengt of Pierre shale is significantly lower than that of Mancos.
|File Size||740 KB||10|