Bit Profile and Gauges Affect Well Trajectory

Embed Size (px)

DESCRIPTION

Bit Profile and Gauges Affect Well Trajectory

Citation preview

  • How Bit Profile and Gauges AffectWell Trajectory

    S. Menand, SPE, and H. Sellami, SPE, Armines/Ecole des Mines de Paris; C. Simon, SPE, DrillScan; A. Besson, SPE,TotalFinaElf; and N. Da Silva, SPE, Security DBS

    SummaryThe importance of wellbore deviation is well recognized by thedrilling industry. An analysis of a drilling systems directionalbehavior must include the directional characteristics of the drill bit.This paper presents a comprehensive analysis of the directionalbehavior of polycrystalline diamond compact (PDC) bits, includ-ing the effect of bit profile, gauge cutters, and gauge length. Nu-merical simulations as well as laboratory tests have been carriedout to better understand the mechanisms of PDC bit deviation andto evaluate the most important parameters affecting the directionalbehavior of PDC bits.

    The analysis presented in this paper shows that each part of thePDC bit (profile and active and passive gauges) plays a major rolein its walking tendency and steerability. A quantitative evaluationof how these factors contribute to well trajectory (inclination andazimuth) is given.

    For the first time, a full-scale directional-drilling bench wasbuilt to measure the walking tendency and steerability of PDC bits.The results obtained demonstrate that the bit profile, gauge cutters,and gauge length have a significant effect. A 3D theoretical rock-bit interaction model was developed to reproduce the drillingtest results.

    IntroductionThe oil and gas industry relies greatly on directional drilling todevelop petroleum reserves in environmentally sensitive areas orin restricted surface areas through an increasing number of multi-lateral, horizontal, and extended reach wells. Many directionalsystems can be used to drill and control the deviation of these morecomplex wells. Depending on well characteristics, one can select arotary bottomhole assembly (BHA), a steerable mud motor, or,more recently, a rotary steerable system (RSS). Whatever the sys-tem used, the drill bit has an influence on the systems directionalbehavior. This paper defines the contribution of the different PDCbit parts on its directional behavior (steerability and walking ten-dency) and their impact on well trajectory.BackgroundTheory. The directional behavior of a PDC bit is generally char-acterized by its walk tendency and steerability. The walk tendency,or bit turn, is a concept well known by drillers and a naturalphenomenon existing in any rotating cutting drilling heads. Fromthis tendency, Ho1 introduced the walk angle for PDC bits, theangle measured in a plane perpendicular to the bit axis, betweenthe direction of the side force applied to the bit and in the directionof the lateral displacement of the bit (Fig. 1). The walk anglequantifies the intrinsic azimuthal behavior of the PDC bit. Whenbit lateral displacement is on the left of the side force, the bit hasa left tendency. If the displacement is on the right of the side force,the bit has a right tendency. A neutral bit means that the lateraldisplacement is in the same direction as the side force. Accordingto the surface position (and considering the previous definition), ifthe bit goes to the left when we are in a building phase, then itstendency is left; if it goes to the right in the same phase, then its

    tendency is right. However, if the bit is going to the left whiledropping, its tendency is right; if it goes to the right, then it has aleft tendency. It is worth noting that an intrinsic neutral bit does notnecessarily give a zero turn rate, because this depends not only onbit characteristics behavior but also on the BHA behavior and theformation characteristics, mainly anisotropy.

    The bit steerability (BS) corresponds to the ability of a bit toinitiate a lateral deviation when submitted to lateral and axialforces. The bit steerability can be defined as the ratio of lateral toaxial drillability.

    BS =DlatDax

    . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (1)

    The lateral drillability (Dlat) is defined as the lateral displacementper bit revolution over the side force. The axial drillability (Dax) isthe axial penetration per bit revolution over the weight on bit(WOB). The BS (equivalent to the bit anisotropic index1,2) is gen-erally in the range of 0.001 to 0.1 for most PDC bits, depending onthe cutting profile, gauge cutters, and gauge-pad characteristics, asevaluated here. A bit with a high steerability means a strong pro-pensity for lateral deviation, enabling us to obtain a maximumpotential for the build or drop rates. Assuming that the BHA ap-plies a nonzero side force to the bit without bit tilt angle, the bitsteerability can be linked to the build or drop rate of well trajec-tories in the field.

    Field and Laboratory Observations. Steerability. Many studieshave been carried out in the laboratory or in-situ to estimate theeffect of PDC bits on the build and drop rate of well trajectories.In analyzing the data from Gulf of Thailand wells, Perry3 reportedthat the profile and gauge length of PDC bits could affect the buildand drop tendencies of BHAs. Pastusek et al.4 conducted somedirectional tests in the laboratory to study the behavior of antiwhirlPDC bits and noticed that these bits had a lower side-cutting abilitythan conventional PDC designs. Pastusek et al.4 attributed thisdifference to the smooth gauge pads used for the antiwhirl bits andconcluded that the build rate of antiwhirl bits on steerable systemswas lower than on conventional PDC designs.

    OBryan and Huston5 studied the effects of gauge length on thebuild and drop tendencies of PDC bits. In testing two differentlengths (88.9 and 152.4 mm), the authors reported that the highestbuild/drop rate was obtained with the longer gauge. This phenom-enon was explained by a higher WOB on the PDC bit with thelonger gauge, generating a higher side force on the bit.5 Morerecently, Norris et al.6 carried out a study in the laboratory andin-situ to evaluate the bit side-cutting ability. One roller-cone bitand two PDC bits, with various gauge aggressiveness, were testedon Carthage marble in the laboratory. With varying WOB and sideforce applied on the bit, the authors observed a BS in the range of0.04 to 0.4. The lateral drillability of the PDC bit with an aggres-sive gauge was almost 10 times higher than the one with an un-aggressive gauge. However, some irregularities and ledges on theborehole were observed with the most aggressive PDC bit. Fur-thermore, the roller-cone bit showed a lower side-cutting abilitythan the two PDC bits. In analyzing field data, the authors noticeda good correlation between the PDC bit side-cutting ability evalu-ated in the laboratory and the build/drop rate measured in the field.

    Walking Tendency. Based on field observations, it is generallyaccepted that roller-cone bits almost always have a right tendency

    Copyright 2003 Society of Petroleum Engineers

    This paper (SPE 82412) was revised for publication from paper SPE 74459, first presentedat the 2002 IADC/SPE Drilling Conference, Dallas, 2628 February. Original manuscriptreceived for review 22 March 2002. Revised manuscript received 25 November 2002.Paper peer approved 3 December 2002.

    34 March 2003 SPE Drilling & Completion

  • and most PDC bits have a left tendency. Kerr7 noticed that PDCbits generally have a left tendency but emphasized that the azi-muthal behavior of the drilling system is influenced by formationcharacteristics, bit profile and size, formation dip, WOB, BHA,and other factors. In analyzing some well trajectories in the Gulf ofThailand, Perry3 concluded that the bit profile could affect theazimuthal behavior of the BHA. Indeed, a BHA with a flat PDC bitprofile showed a right tendency. Perry3 also supposed that thegauge cutters and length did not influence the turn rate. In studyingBHA azimuthal behavior in the Alwyn North field, Bannerman8confirmed the observations made by Perry3: the right turn mea-sured in the field is supposedly attributed to the flat PDC bitprofile, although the parabolic profiles exhibited a left tendency.

    Synthesis. These laboratory or in-situ studies show that a com-prehensive analysis of PDC bit directional behavior has never beenconducted to quantify the intrinsic azimuthal behavior of the PDCbit. Moreover, the directional behavior of a whole drilling systemcannot be explained solely by that of the bit. A bit with a highside-cutting ability does not necessarily produce a high inclinationrate on the well trajectory. This rate depends on the side force andweight applied on the bit, the bit tilt angle, and the rock formation.Likewise, the azimuthal behavior of a drilling system must not beattributed only to the walk tendency of the bit. Some frictionphenomenon along the BHA (mainly at stabilizers levels) cangreatly influence the azimuthal tendency of the drilling system.The formation effect (rock anisotropy) may be decisive in both thebuild/drop and azimuth rates for the trajectory.9

    Rock-Bit Interaction ModelDuring the past 30 years, Ecole des Mines de Paris has developeda methodology for designing and selecting cutting and drillingsystems. Drilling efficiency,10 wear reduction, vibration control,and efficient cleaning have been studied carefully. A 3D rock-bitinteraction model9,11 has been developed to calculate the direc-tional behavior of PDC bits in isotropic and heterogeneous forma-tions. The bit model takes into account the three bit parts thatinteract with the formationthe cutting structure, the active gauge(trimmers or gauge cutters), and the passive gauge (gauge pad), asshown in Fig. 2.

    Cutting Structure. The rock-bit model includes an elementaryPDC-interaction model that takes into account PDC geometry (cut-ter size and geometry, back rake angle, chamfer, wear, and fric-tion) and the rock characteristics (cohesion, angle of internal fric-

    tion, uniaxial compressive strength, pore pressure, and dip angle).From the cutting structure, a cutting profile is defined geometri-cally and can be divided into two parts according to IADC clas-sification12: the inner cone (height C) and the outer structure(height G). The cutting structure is defined through the rock-bitmodel by its cutting profile (geometric parameters) and its cutterposition and orientation.

    Active Gauge. The active gauge corresponds to PDCs that aretruncated (trimmers or gauge cutters) to the bit diameter corre-sponding to the transition between the cutting structure and thepassive gauge. From single-cutter laboratory experiments, a trim-mer-rock interaction model was developed and integrated into therock-bit model. The active gauge is then defined by its length LAG,trimmers number, trimmer back rake angle, and rock-friction sur-face, depending on the trimmer truncation.

    Passive Gauge. The passive gauge (or gauge pad), which plays agreat role in stabilizing the PDC bit, can have many design fea-tures. The main passive-gauge characteristics are the length, thecircumferential coverage (depending on the blade spiral angle),and the surface roughness (smooth gauge pads, such as the low-friction gauge pads13,14 or aggressive gauge pads, depending onthe carbide- or diamond-insert type for protection). According tothese features, the passive gauge is characterized in the rock-bitmodel by its length LPG and its blade characteristics (number,spiraled or straight, diamond- or carbide-insert type), which definea friction surface with the borehole.

    Kinematics. The bit is assumed to rotate continuously on its axisand is given a prescribed axial and lateral motion. The motion ofthe bit is described with five variablesthree for a transla-tion movement and two for a rotation movement. After prescrib-ing a bit motion, the rock-bit model then calculates the forces onall cutting elements and integrates the single forces on thebit surface to produce global forces and moments averaged for onebit revolution.

    Results. The 3D rock-bit model calculates WOB and lateral forceon the bit required for axial and lateral motion, imbalance force,efficiency index, and wear evolution. It also computes the steer-ability and the walk angle of each part (cutting structure and activeand passive gauges) of the bit. It is important to note that the bitsteerability, calculated from the rock-bit model, is mainly a func-tion of WOB, lateral force, and rock strength and anisotropy.

    Assuming all the PDC cutters have an identical back rake anglealong the bit profile, Menand11 found that the walk angle is thena function of the inner cone depth, C, and the outer structureheight, G, and can be simply approximated by

    = arctan2C G

    tanc + fC + G. . . . . . . . . . . . . . . . . . . . . . . . . (2)

    Fig. 1Definition of the walk angle according to Ho.1

    Fig. 2Description of the PDC bit structure.

    35March 2003 SPE Drilling & Completion

  • C and Gthe inner cone and outer structure heights, respectively,according to the IADC classification;12 cthe back rake angle;and fthe friction angle between the PDC and the rock.

    Directional Laboratory TestsDirectional Drilling Bench. To measure bit steerability and walk-ing tendency, the drilling bench at Ecole des Mines de Paris wasmodified, enabling us to test the drill bit under simulated downholeconditions (Fig. 3). The new system tests the side-cutting abilityand the walk tendency of a bit up to a 311.1-mm diameter. Thedirectional tests can be performed with a maximum 15-ton WOBand a lateral force up to 1500 daN. The directional test principle(Fig. 4) is as follows. During axial bit penetration, a lateral force,Fx, is applied to the rock sample, which is free to move in thedirection of the applied force, generating a lateral displacement.Two sets of strain gauges are mounted on the drilling shaft tomeasure the bending moments (magnitude and orientation). Thetotal resulting lateral force, Flat, at the bit is computed with the

    bending moments readings; the orientation difference between thelateral displacement and the resulting lateral force, Flat, at the bitgives the bit walk angle (Fig. 4).

    The lateral drillability, Dlat, of the bit is calculated from thelateral displacement of the rock sample measured by the linear-variable differential transformer (LVDT) sensor and the resultinglateral force Flat. The axial drillability, Dax, is measured from therate of penetration (ROP), rotation speed, and WOB.

    Test Procedure. All the tests were carried out in the Vosges sand-stone (homogeneous, porous, medium-hard sandstone, uniaxialcompressive strength40 MPa). A 1150-kg/m3 water-based mud(bentonitic) was used with the mud flow fixed to 600 L/min.During the tests, the rotation speed was held constant at 60 rpm,while the WOB and lateral force were varied to evaluate theireffect on steerability and walking tendency.

    Off-bottom tests were also performed to test the lateral drill-ability of active and passive gauges. During the off-bottom test, the

    Fig. 4Principle of the directional test.

    Fig. 3Drilling bench and directional drilling bench in Pau, France.

    36 March 2003 SPE Drilling & Completion

  • bit was maintained above the bottom of the hole and a lateral forcewas applied, enabling testing of the gauge interaction with only theborehole formation.

    Characteristics of the PDC Bits Selected. Three PDC bits havingdifferent profiles have been tested on the directional drillingbenchBits A, B, and C (see Fig. 5). The back rake distributionis identical along these three profiles, ranging from 15 inside thecone to 30 in the outer structure. The common characteristics ofthe bits are a 215.9-mm diameter, eight highly spiraled blades with13.3-mm PDC cutters, and four nozzles. The bits have differentactive gauge lengths, ranging from 15 mm for Bit A to 30 mm forBit C. The bits have passive gauges with different types of pro-tection inserts. To evaluate the effect of the three different parts ofthe bit (cutting structure and active and passive gauges), each bitwas tested with five different configurations, as shown in Fig. 6.First, each bit was tested with passive gauge lengths LPG101.6,50.8, and 25.4 mm. Then, the bits were tested with only their activegauge and cutting structure (no passive gauge). Last, each bit wastested with only the cutting structure (i.e., without any active orpassive gauges).

    ResultsSteerability. For the various bits tested, one can notice that the bitsteerability highly increases with the reduction of the passivegauge length (Fig. 7). All the tests plotted for this figure have beencarried out with the same WOB and lateral force. The highestmeasured steerability is for Bit A. These results are explained

    mainly by the different active gauge lengths and bit profiles andare confirmed by the 3D rock-bit model calculation (see Table 1).

    Tests carried out without a passive gauge (i.e., with only theactive gauge and cutting structure corresponding to Bit Configu-rations 4 and 5) have revealed that the highest steerability for BitConfiguration 4 was obtained with Bit A, and the lowest steer-ability was for Bit C (see Fig. 8). This result can be attributedmainly to the active gauge length because Bit C has the longestactive gauge and Bit A has the shortest one. Nevertheless, one canalso notice that the highest steerability for Bit Configuration 5 (testwith the cutting structure alone) was observed with Bit B, althoughBit C exhibited the lowest steerability (Fig. 8). This result can beanalyzed by examining the bit profiles. Indeed, the highest steer-ability is obtained for Bit B with a flat profile (IADC bit profilecode 9), although the lowest measured steerability corresponds toBit C with a medium taper and cone (IADC bit profile code 5).

    Some tests performed with various lateral forces demonstratedthat steerability of a PDC bit depends on the intensity of the sideforce. For example, Bit Cs steerability with a 50.8-mm passivegauge (Configuration 2) is increased by 30% with a 25% increasein the lateral force. The off-bottom tests confirmed that the lateraldrillability of the active and passive gauges depends on the lateralforce applied. Indeed, the off-bottom lateral drillability of Bit B inConfiguration 3 is multiplied by almost three as the lateral forceincreases from 268 to 710 daN (see Fig. 9). WOB seems to haveno effect on the lateral drillability of the three bits tested.

    Walk Tendency. For the various bits tested with an active orpassive gauge, one can clearly notice that the PDC bits have a lefttendency, whatever the passive gauge length is (see Fig. 10). Even

    Fig. 7Bit steerability measured on the directional drillingbench (whole bit).Fig. 6The three PDC bit profiles tested.

    Fig. 5Description of the five bit configurations tested.

    37March 2003 SPE Drilling & Completion

  • the tests carried out with the cutting structure and the active gaugedemonstrated that the bits have a left tendency. When the cuttingstructures were tested alone, Bit A demonstrated a right tendency,Bit C a left tendency, and Bit B a neutral tendency. Measured onthe directional drilling bench, these walk tendencies correlatedwell with the values computed from the rock-bit model (Table 1).

    Bit B showed a tendency to spiral in the hole because thewalking tendency was successively neutral, left, right, neutral, etc.(see Fig. 11). Nevertheless, the mean walk angle measured wasclose to 0. These spiraling problems, observed only for Bit B, canbe generalized to bits with a flat profile.

    Bit-BHA Coupled Computer ModelIn coupling the 3D rock-bit model with a 3D BHA mechanicalmodel, Ecole des Mines de Paris developed software enablingprediction of the inclination and azimuth of well trajectories.Based on the finite-element method, the 3D mechanical modelallows the deformed shape of the structure, forces exerted on thesystem, and contact forces between any part of the drillstring andthe borehole wall to be known. In integrating the directional be-

    havior of both the BHA and bit, the software calculates the theo-retical 3D equilibrium curvature of the drilling system.

    Case StudyPDC Bit Characteristics. To evaluate the influence of walk ten-dency and steerability on the well trajectory, some PDC bits, withan assumed BS and walk angle, were selected for the analysis. Foreach bit (Bits X, Y, and Z), with various bit steerabilities (Table2), the walk angle was varied between 20 (bit intrinsic lefttendency) and +20 (bit intrinsic right tendency).

    Well and BHA Characteristics. To observe the influence of thebit directional behavior on well trajectory, two assemblies thatproduce a significant side force on the bit were selecteda drop-ping and a building assembly (Fig. 12). The data used come fromtwo wells in phase 250.8 mm drilled by TotalFinaElf with the samePDC bit (Bit W). The run on Well 1 was performed with thebuilding assembly from 1380 to 2534 m measured depth (MD),producing a measured build rate of 0.29/30 m and turn rate of0.11/30 m. The Well 2 run was performed with the droppingassembly from 2405 to 3881 m MD, producing a measured drop

    Fig. 8Bit steerability measured on the directional drillingbench (cutting structure and active gauge).

    Fig. 9Off-bottom lateral drillability vs. lateral force for Bit B(Configuration 3).

    Fig. 10Bit walk angle measured on the directional drillingbench. Fig. 11Spiraling tendency of Bit B (Configuration 5).

    38 March 2003 SPE Drilling & Completion

  • rate of 0.55/30 m and turn rate of 0.30/30 m. Table 3 gives theparameters used for BHA simulations. As discussed previously, bitsteerability depends on the side force applied. In the two casesstudied, two theoretical bit steerabilities have been calculated be-cause the side force generated by the dropping assembly is greaterthan the one generated by the building assembly. The theoreticalsteerability of Bit W is 0.03 for Well 1 (building assembly) and0.04 for Well 2 (dropping assembly). The intrinsic theoretical walkangle is 12 (left tendency).

    Results. The bit-BHA model was used to compute the build/dropand turn rates for the two wells. In the calculations performed, allstabilizers are full gauge, which prevents evaluating any walk ratecaused by BHA walking tendency. Concerning Well 1 (Fig. 13),one clearly notices that the bit steerability has an influence on thepredicted build/drop rate of the drilling system because it variesfrom 0.12/30 m with Bit Z to 0.34/30 m with Bit X. Thetheoretical bit steerability (BS0.03) calculated for Bit W, used todrill Well 1, produces a predicted build rate very close to themeasured one (0.29/30 m). For Well 2 (Fig. 14), the predictedbuild/drop rate varies from 0.42/30 m with Bit X to 0.38/30 mwith Bit Z. The theoretical bit steerability of Bit W for this well isnot high enough to give a predicted drop rate close to the measuredvalue (0.55/30 m), but qualitatively, the increase in bit steer-ability because of the higher side force is consistent with the higherdrop rate observed in the field. Moreover, as discussed previously,one must keep in mind that the build/drop rate is caused by notonly the side force applied on the bit but also by the bit tilt angle.It is interesting to note that in both cases, the bit steerability hassuch an influence that it can turn the drilling system from a build-ing to a dropping angle. This result is caused by the bit tilt andlateral force acting in opposite directions. These results confirmthe impact of bit steerability on the well trajectory and show astrong necessity to calculate an accurate bit steerability to predictthe inclination of well trajectories correctly. The simulations haveshown that the walk angle has no influence on the predicted build/drop rate, whatever the bit steerability.

    Concerning the azimuth predictions, one can clearly observethat the bit walk angle and steerability have an influence on thepredicted turn rate (Figs. 13 and 14). For Well 1, with an intrinsicleft-tendency bit, the simulations give a left turn up to 0.06/30 m,although with an intrinsic right-tendency bit, the predicted turn isto the right. This result is accentuated for Well 2 because thepredicted turn rate is in the range of 0.7/30 to 0.7/30 m, de-pending on the intrinsic bit walk angle. It is also interesting to notethat for a given bit walk angle, the predicted turn rate depends onbit steerability, and the influence grows more important assteerability increases. This tendency can be attributed to the bitside-cutting ability, making the bit walk on the borehole wall.Comparison between predicted and actual turn rates for Well 2shows that the theoretical bit steerability (BS0.04) and walkangle (12) produce a turn rate very close to the measuredvalue (0.3/30 m).

    Synthesis. Even though the directional behavior of a drilling sys-tem can not be attributed only to the bit directional behavior (for-mation effect, curvature of the borehole, hole enlargement, frictionphenomenon, etc.), these simulations have shown that bit steer-ability and walk angle have a strong influence on well trajectory.ConclusionsAnalysis of the directional behavior of PDC bits presented in thispaper leads to the following conclusions.1. The walk angle of a PDC bit depends not only on the bit profile

    but also on the active and passive gauges. Directional lab tests

    have demonstrated that the various bits tested with a passivegauge had a left tendency, despite their bit profiles and PDCsetup.

    2. The walk angle of a PDC cutting structure is calculated with asimple equation that links the inner cone and outer structureheights and the PDC back rake angle.

    3. The active and passive gauges dramatically affect the walkangle of PDC bits.

    4. The directional tests enable observation of spiraling problemsand define the minimum requirements for avoiding suchphenomena.

    5. The steerability of a PDC cutting structure depends greatly onthe bit profilethe flatter the profile is, the more steerablethe bit is.

    6. Bit steerability is a nonlinear function of the active gauge lengthand decreases as the active and passive gauge lengths increase.

    7. Bit steerability depends on the applied side force.The bit-BHA simulations and comparisons with field results haveshown the following.1. The bit walk angle has no influence on the build/drop rate of

    well trajectories.2. There is a strong correlation between bit steerability and build/

    drop rate.3. An accurate calculation of bit steerability is necessary to make

    a good trajectory prediction.4. The bit steerability and walk angle have an influence on the

    predicted turn rate.

    NomenclatureBS bit steerability, dimensionlessC inner cone depth, L, mm

    Dax axial drillability, L/m/rev (mm/Mg)/revDlat lateral drillability, L/m/rev (mm/Mg)/revFlat resulting lateral force, mL/t2, NFx lateral force applied by the jack, mL/t2, NFy lateral force in the y-axis direction, mL/t2, N

    Fig. 12Description of the BHAs.

    39March 2003 SPE Drilling & Completion

  • G outer structure height, L, mmLAG active gauge length, L, mmLPG passive gauge length, L, mm walk angle, rad, degreesf friction angle between PDC and rock, rad, degreesc back rake angle, rad, degrees

    Acknowledgments

    Part of this work was carried out within the EEC Thermie PAB-BIT project conducted by Ecole des Mines de Paris/Armines, To-talFinaElf, and Security DBS. The authors would like to thank theEuropean Commission for its financial support enabling us to carryout a part of the work presented in this paper. Thanks are also ad-dressed to DrillScan Co. for performing well trajectory calculations.

    References

    1. Ho, H.S.: Method and System of Trajectory Prediction and ControlUsing PDC Bits, U.S. Patent 5,456,141 (1995).

    2. Barton, S.: Development of Stable PDC Bits for Specific Use onRotary Steerable Systems, paper SPE 62779 presented at the 2000IADC/SPE Asia Pacific Drilling Technology, Kuala Lumpur, 1113September.

    3. Perry, C.J.: Directional Drilling With PDC Bits in the Gulf of Thai-land, paper SPE 15616 presented at the 1986 SPE Annual TechnicalConference and Exhibition, New Orleans, 58 October.

    4. Pastusek, P.E. et al.: Directional and Stability Characteristics of An-tiwhirl Bits With Nonaxisymmetric Loading, paper SPE 24614 pre-

    sented at the 1992 SPE Annual Technical Conference and Exhibition,Washington, DC, 47 October.

    5. OBryan, P.L. and Huston C.W.: A Study of the Effects of Bit GaugeLength and Stabilizer Placement on the Build and Drop Tendencies ofPDC Bits, paper SPE 20411 presented at the 1990 Annual TechnicalConference and Exhibition, New Orleans, 2326 September.

    6. Norris, J.A. et al.: Development and Successful Application ofUnique Steerable PDC Bits, paper SPE 39308 presented at the 1998IADC/SPE Drilling Conference, Dallas, 36 March.

    7. Kerr, C.J.: PDC Drill Bit Design and Field Application Evolution,JPT (March 1988) 327.

    8. Bannerman, J.S.: Walk Rate Prediction on Alwyn North Field byMeans of Data Analysis and 3D Computer Model, paper SPE 20933presented at the 1990 SPE EUROPEC, The Hague, 2224 October.

    9. Simon, C.: Modelisation of PDC bit directional behavior in aniso-tropic formation, PhD dissertation, Ecole des Mines de Paris (1996).

    10. Gerbaud, L. et al.: New PDC bit design increased penetration rate inslim wells, paper presented at the 1997 Energy Week Annual Con-ference and Exhibition of ASME, Houston, 2830 January.

    11. Menand, S.: Analysis and validation of a PDC drilling bit directionalbehavior model, PhD dissertation (confidential), Ecole des Mines deParis (2001).

    12. Winters, W.J. and Doiron, H.H.: The 1987 IADC Fixed Cutter BitClassification System, paper SPE 16142 presented at the 1987 SPE/IADC Drilling Conference, New Orleans, 1518 March.

    13. Warren, T.M., Brett, J.F., and Sinor, L.A.: Development of a Whirl-Resistant Bit, SPEDE (December 1990) 267.

    14. Sinor, L.A. et al.: Field Testing of Low-Friction-Gauge PDC Bits,SPEDC (March 1993) 21.

    Fig. 13Effect of bit steerability and walk angle on the predicted build/drop and turn rates (Well 1).

    Fig. 14Effect of the bit steerability and walk angle on the predicted build/drop and turn rates (Well 2).

    40 March 2003 SPE Drilling & Completion

  • SI Metric Conversion Factorsft 3.048* E01 m

    in. 2.54* E+01 cmlbf 4.448 222 E+00 N

    lbm/gal 1.198 264 E+02 kg/m3psi 6.894 757 E+00 kPaton 9.071 847 E01 Mg

    *Conversion factor is exact.

    Stephane Menand is currently a research scientist in the Dept.of Mining and Underground Works Engineering at the ParisSchool of Mines (ENSMP) in Fontainebleau, France. e-mail:[email protected]. His main areas of interest are di-rectional drilling, drillstring mechanics, and rock-cutting tools.Menand holds a PhD degree in drilling engineering from

    ENSMP. Hdi Sellami heads the Rock Cutting and Drilling groupof the Dept. of Mining and Underground Works Engineering atthe Paris School of Mines (ENSMP). e-mail: [email protected]. He is an expert on rock fragmentation for mining,tunneling and oil drilling applications, and has provided sev-eral PhD theses, written numerous papers, and patented vari-ous techniques for cutting and drilling hard rocks. Sellami holdsa PhD degree from ENSMP. Christophe Simon, after 10 years ofresearch at the Paris School of Mines (ENSMP), has startedDrillScan to commercialize expertise and software in the fieldsof directional drilling, torque and drag, and bit performance.e-mail: [email protected]. Simon holds a PhDdegree in drilling engineering from ENSMP. Alain Besson is theformer head of Downhole Drilling Tool section of TotalFinaElf,Paris, France. e-mail: [email protected] Da Silva is Design Support Manager for Halliburton (Se-curity DBS) plant located in Brussels, Belgium. e-mail:[email protected]. His main areas of interest arecore heads and drilling bits, but more specifically directionaldrilling and hard/abrasive drilling.

    41March 2003 SPE Drilling & Completion