SPE

SPE 14221

Optimization of PDC Bit Hydraulics by Fluid Simulation

by R. Crouse and R. Chia, * Sii Smith Tool
●SPE Member

Copyright 19S5, Society of Petroleum Engineers

This paper was prepared for presentation at the SOthAnnual Technical Conference and Exhibitionof the Society of Petroleum Engineers imid in Las
Vegas, NV September 22-25, 1985.

This paper was selected for presentation by an SPE Program Committee following review of informationcontained in an abstract aubmined by the
author(a). Contents of the paper, as presented, have not been reviewed by the S@etY of petroleum Engineem and are aubiect to com=tion by the
author(a).The material, aa presented, doea not necessarily reflect any positionof the S@ety of Petroleum Engineers, ita o~lcers, or members. Papers
presented at SPE meetings are subject to publiiatksn review by Editorial Committees of the Society of Petroleum Engineers. Permission to copy ia
restrictedto an abatract of not more than 300 words. Illustrationsmay not be copied. The abstract shouldcontainconspicuousacknowledgmentof where
end by whom the paper ia presented. Write Pubficafiims Manager, SPE, P.O. Sox S33836, Richardson, TX 75063-3S36. Telex, 730989 SPEDAL.

ABSTRACT diamond drill bit as a function of flowrate.

Fluid flow acroas complex PDC bit geometry ie Measurements of flow field characteristics
difficult to optimize by drilling tests in the acroea the face of a stud-mounted PDC bit have
laboratory. Insight into this flow field haa been been obtained by Glowka (2,3) in a test facility
developed within a computer program which of Sandia Laboratories, which simulatee downhole
simulates the fluid flow at the bit. The flow fields. Flow tracer tracking and cutter heat
development aspects of the computer code which transfer coefficient measurements were reported in
allow for the three-dimensional flow visualization comparing flow field characteristics of three bit
are reviewed. Graphical representation of the c6nfigurationa.
.Je~9ciCYVeCtOrS and pressure fields for SiX bit
geometries are given to ahow the effects of nozzle With the development of h@P6p@?d digitsl
location, orientation, bafflea and flow channela. computers in the 1970-s, several sophisticated
This in turn serves as data for a bottom hole numerical fluid mechanics codes have evolved (ace
cleanup, cooling rate and body erosion analysis. reference 4,5,6,7 for example) that can be applied
Dull bit observations are discussed and compared to problems of drill bit hydrodynamics. These
to the computer-derived hydraulic conditions. codes have already been applied to the computation
of turbulent, recirculating flows in three
INTRODUCTION dimensions, and can be extended to non-Newtonian
single and two-phase (solid/liquid) flows.
The continuing need to discover and produce
~ew reserves of oil and gas economically has The adaptation of these powerful computer
.
encouraged the development of improved drill bit coaes to artii bit L..A-A...,.
IlyUL&UY.udiAv.
<,..Y
nFFaFe $J,Jntque
v..-..

designs with the potential for reducing drilling opportunity for improving the hydraulic design and
coats. Much effort has been expended on operation of drilling bits at a small fraction of
determining the optimum set of operating the time and cost associated with the manufacture,
conditions, or bit “hydraulics”, that will testing, and evaluation of experimental bit
maxfraizethe rate of penetration. However, this designs in the laboratory.
type of operational information may not be helpful
.to the drill bit design engineer who wanta to At this time, both finite element and finite
improve the bit cleaning, bit cooling and difference computer codes are being developed for
formation erosion effects of the drilling fluid fluid mechanics studies. The finite element
through the design of the bit nozzle, and fluid “ methods have one distinct advantage in that
course arrangements. automatic mesh generation techniques for fitting
triangular and quadrilateral elements to complex
Simple analytical methods have been used to geometries have been developed. However, the
model certain aspects of drill bit hydrodynamics. proper choice of basic functions for representing
Myers and Funk (1) derived a aet of equations for convection dominated fluid flows is still a matter
predicting the pressure drop across the face of a for on-going research.

References and Illustrations at end of paper. On the other hand, finite difference codes in
2 OPTIMIZATION OF PDC BIT HYDRAULICS BY FLUID SIMULATION SPE 14221

Cartesian rectangular and cylindrical coordinate where there are no large gradients and the fluid
systems have been used extensively to compute is flowing out.
~e=vectie= g~~~~~~ed ~aminar and turbulent flows
with and without flow recirculation in two The fourth boundary condition is a periodic
dimensions (7). We would show in this paper how boundary. For the left and right pair, the model
the finite difference method could be used in width should be such that (IBAR-1) is equal to the
determining bit hydraulics at high Reynolds period length.
number. This three dimensional code, PDCFLOW, was
modified and extended by the authors from the two There is a subroutine which can contain any
dimensional code, SOLA (7). additional special boundary conditions. This
subroutine also would contain the description of
DESCRIPTION OF METHOD any internal obstaclea or cells which have no
F1,-.,.,
AA””. To define an QbsEacle, it is merely
Governing Equations necessary to define the velocity equal to zero for
all the cells within that obstacle.
The computer program used is based on the
solution of the three dimensional Navier-Stokes Geometrical Modeling
equation and the incompressibility equation. The
equations are solved using a centered finite The first step in the development of a drill
difference method with an upstream difference bit hydrodynamics computer code is to generate a
operator for stability. computing mesh that fills and subdivides the space
between the drill bit face and the walls of the
In order to integrate the Navier-Stokes borehole with a network of finite volume elements.
equations in t%rne,iiSir@a ~~zipitite~,
it iS The highly curved and three-dimensional shape of
necessary to convert the differential equations a drill bit face requires the use of special
into finite difference equations. This is done methods for mesh generation.
using a centered difference method.
To aid in the analysis of fluid flow problems
~ordfnate System using PDCFLOW, a set of programs have been written
to take the user through the procedure in a
The three dimensional Cartesian coordinate logical order. These programs are designed to
system used is shcmi iil Figiire 1. ‘HIethree exes ~..-= the user for any needed values.
nromn~
are x, y, and z with tndtces i, j, and k
respectively. The corresponding velocities are u, The steps to be followed are discussed
“v,ad ‘w. Tt.- ltienti~~~f the ve~Q~itieS for any
..,=L“--- briefly in this paper. First an input data deck is
cell is defined in Figure 1 with the velocity generated. TNs will contain information on the
being defined in the center of the face on the law of “alle
step aiz~, FNiTZ.=. ---., hQV hxg EQ run the
pQsitive side of the cell. The pressure is job, boundary conditions, and fluid properties.
defined as being at the center of the cell. The second step is to use PATRAN (8) to buiid
EedeIS ~f the bit body% padsP and the rock hole
The model is made up of IBAR by JBAR by KBAR outline. The output of these PATRAN models is
cells with a plane of fictitious or boundary cells translated into boundary conditions for the fluid
on all aiclea. The flctitioua cells are for the flow code. The next step is to define the location
application of the boundary conditions. of the cutters. From the cutter locations, a file
is created which will have the fluid code block
Initial Conditions out the areas where the cutters are. Fourth, the
nozzles are defined as in flow cells with fluid
At the begining of the calculation, the flowing in with a defined velocity and direction.
initial flow conditions must be defined. This can Now that all of the necessary files have been
be aa simple as defining all velocities and created, the fluid flow code, PDCFLOW, can be run.
~i~SSiii=S ~= Z~ZO CT S2Y ~C~er ~o~~ltion desired. This code can use as initial conditions, a no
flow case, or the final result from a previous
Boundary Conditions run. If a restart was used, then the time to
reach steady state is obtained much faster. Since
There are four built-in boundary conditions there is such a mass of information obtained from
for each of the aix surfaces. They can be invoked a run, it is easiest to understand the results
by setting the correct value in the input data by plotting the output. In plotting the results
cards. The first condition is a rigid free-slip the user has the option to view the bit from one
wall, that is the normal velocity must be zero and of three different directions, the top or either
the tangential velocity should have no gradient. side.

For a rigid no-slip wall, the normal velocity TECHNIQUES FOR OBTAINING FLOW CHARACTERISTICS
must be zero and the tangential velocity must also
be zero. Cooling of the PDC Cutter

Outflow boundary conditions pose the most Very essential to the life of a PDC cutter is
difficult problem. They must let the flow paas the temperature it reaches while drilling. Heat
freely out while not affecting the upstream flow. is generated at the interface between the cutter
Care must be taken to locate the outflow “boundary SW4.+4-* against the rOCk:
and the zoc’k&tieto the ..~~ti.”..

—
 ..
SPE 14221 R. Crouse & R. Chia 3

This heat is removed from the cutter by convective plastic formation was identified as bit balling,
cooling of the mud flowing past the cutter. in which ahale cuttings clogged in front of the
PDC cutters. The shale cuttings recovered from
To increase ROP, the weight on the bit and/or the shale shaker contained many curved strips or
~~~ aee~=~t~, Fhich glosely resembled that of metal
the rotational apeed must ‘beincreased.
increasea the amount of heat produced at each cuttings. A shale chip has also the tendency to
cutter. If the temperature of the cutter gets too stick to the cutter face.
high, then thermal deterioration and wholegrain
pullout of the diamond layer will occur resulting From the chip generation point of view, the
in wear of the cutters. Because of this, one of chip clearance should be ample enough to allow
the limiting factors in ROP is due to the cooling smooth chip generation without cuttings compacted
of the PDC cutters. between the bit body and rock formation. On the
other hand, the large clearance for the chip means
To keep the temperature of the cutter as low large fluid areas which in turn induces low fluid
as possible, but still drill as fast as possible, velocity. The low fluid velocity may not have
the cooling of each of the cutters must be enough momentum to sweep away the plastic chip,
maximized. The cooling rate is dependent on the and reduces the cooling effect on the PDC cutter.
mud properties and the velocity and direction of Therefore, it is desirable to use bafflea to
-c-l
me ma flow- ~ioiiid tb,e c.u~~~re ‘lW hit dCXi@ler reduce the flow area in order to increaae fluid
does not have control over the mud properties but velocity and guide the mud flowing through
he can shape the bit to optimize the mud flow cutters.
around each cutter. The heat transfer coefficient
is proportional to the velocity of the mud to the The critical.velocity to remove the cutting
0.8 power (3). Therefore to increase the cooling, at any cross section can be estimated by an
the velocity around each cutter should be emperical equation, (11)
increased.
UC = .5*d**.4444*SQRT(Ds/Df-1),
Erosive Wear
where d in millimeters, is determined by the chip
The erosive process is a complex phenomenon dimensions, Ds and Df are the density of chip and
depending on many variablea. It is defined as the mud respectively.
proceaa of material removal from a surface
subjected to impingement attack by solid or liquid Weight-on-Bit Decrease due to Pump-off
media, being particulate in nature for the former.
The erosive process is caused by the directional Due to the restricted area the mud must flow
impact of solid particles, random impact and across the surface of a PDC bit, there will be a
sliding bed of solids. Experimental correlations pressure gradient over the bit face. This high
--- c-------IV
aLc LLqLsG..t&J
.~.c.A
..--..
to e~ti~te erosion wear. For pressure mud under the bit will reduce the
instance, the amount of material removed by a jet effective weight on the bit. i?roceduresfor
impinging on a plate is related to the solid measuring the decrease in bit weight are given in
particle velocity, concentration, particle size reference (12).
and angle of impingement (9,10).
For different designs of bits this lift-off
The total erosion rate on a bit is assumed to force can be calculated if the pressure under the
bit is known. Since we know the pressure field
erosion rate = C*Ds*Cv*Q*V**m*PHI, under the bit from the flow simulation calculation
all that is needed to determine the force is to
where Da is the mud weight, in pounda per gallon, intergrate the pressure over the bit face.
Cv is the solid concentration,
Q is the flow rate, in gpm, BIT CONFIGURATIONS
V is the solid impinging velocity, in ftisec
PHI is the attack angle, in degrees. Baae Case: 6 #12 Nozzles. 3 Baffles. 300 GPM

Based on the measurements of weight removal The bit modelled ia an 8 1/2” diameter PDC
from Finnie et al (9) on erosion of tool steel, Tri-Cross-Flow Bit. A bit face view of the PDC
the constant C, UIand PHI can be found to be: bit is shown in Figure 1. The X and Z directions
of the grid used in the computer calculations are
C = 3.OE-09 ahown in relation to the bit. The Y direction
comes straight up out of this figure from the pin
m = 2.35 end of the bit toward the rock. I, J and K are
counters used in the computer code and they
PHI = ANG/37*EXP(-(ANG/37)**2) correspond to the X, Y and Z directions
+0.2*EXP(-((90-ANG)/37)**2) respectfully. There are six 12/32 inch nozzles in
the bit as seen in Figure 1. Three of the nozzles
h erosion curve is plotted on Figure 2. are on the bottom of the bit and the other three
are on the sloped sides to add to the radial flow
Cieaning as the diameter increases. The three nozzles on
the bottom of the bit are directed outward 5
The main problem of PDC bits drilling soft or degrees from the vertical. The three nozzles on
4 OPTIMIZATION OF PDC BIT HYDRAULICS BY FLUID SIMULATION SPE 14221

from the vertical. The flow rate for the bit is CONCLUSIONS
300 gpm.
C.leanin~
This bit incorporates three flow baffles
spaced evenly around the bit. These baffles The chip removal of a PDC bit is dependent on
restrict the area which the mud can flow in the cross flow of the mud along the rock surface.
thereby increasing its velocity and directing the Figure 3 shows the cross flow parallel to the rock
flow toward the cutters. for the model described in the base case. The
arrows indicate the direction of the mud flow and
The bit is assumed to be unworn so that the the length is proportional to the velocity of the
distance between the bit body and the rock surface mud flow. The six nozzles can be picked out by
is at its maximum value. As the bit is used, the the high velocity mud flowing out along the rock
cutters will wear down so that the three pads surface away from the nozzlea. The PDC cutters
touch the rock but in this model no wear has yet are seen as areas where there is no flow. An
occurred so there is still a small clearance overall view of the mud flow around the cutters
~tween the pads and the rock. Aa the wear for cooling and cleaning of the cutters is seen
occurs, the flow pattern may change a small here.
amount.
To understand the complexities of the mud
Case 2: 6 #12 Nozzles, 3 Baffles, 500 GFM flow under the bit, a side view of the bit is
shown in Figure 4. This view is along the 90
This case uses the same bit as the base case degrees - 270 degrees line shown in Figure 1, as
in the base case. The difference is that the flow seen by looking from the side of the bit marked
rate has been increased to 500 gpm. The flow is 180 degrees. The bit is shown upside down
directed out of the 6 #12 nozzles. (drilling up) to correspond to the convention used
in the Engineering drawings. At the top left side
Case 3: 6 #12 Nozzles, No Baffles. 300 GFM of Figure 4, the cutout for the nozzle can be
seen. Flow is being drawn into the jet from the
For this case the flow is 300 gpm flowing nozzle near the bit surface and then flows outward
through 6 #12 nozzles. This model is the same as as the jet impinges on the rock. This acts up a
in the base case except that the three flow baffles recirculating flow near the nozzles which helps
have been removed. This leaves a larger volume clean the cutters at the center of the bit.
between the bit body and the rock surface. The
distance between the bit body and the rock at the Figure 3 shows the overall flow around the
cutter faces remains the same since this is the cutters on the bit. To ensure proper cleaning of
necessary distance for chip removal. all the cutters, each one must be looked at more
closely. Each of the cutters on the bit is going
~
Cas 4: to be creating rock chips of different sizes
depending on the relative location of the other
Th. h{t
-..-“-- ~Q&lQ~ in this section is the same cutters both inward and outward of the cutter in
basic bit as modeled in the base case. The nozzle question and the penetration rate. To ensure
size for all six nozzles has been Increaaed to cleaning of any cutter, it is necessary that the
16/32 inch. The angle of the nozzles is the same mud will remove rock chips at least as big as the
as in the base case. The flow rate fs 300 gpm and chips being generated by that cutter. Figure 5
ihere are the three flow baffles. shows contours of chip size that will be removed
for cutter #3. This shows that only relatively
Case 5: 6 #12 Nozzles, New Nozzle Angle small chips will be cleaned from the cutting edge
of this PDC cutter. This can also be seen by
This model test the effects of the angle looking at the flow in front of cutter #3 in
which the nozzles are pointed. The bit is the Figure 3. Here it can be seen that the flow from
same as In the base case. The flow rate is 300 gpm the nozzle is stagnating at the cutting edge of
and there are three flow baffles. The three this cutter, then fiowing down the face & ths
nozzles on the top flat surface have been angled cutter toward the bit body to be recirculated back
out to 30 degrees from the vertical. The ‘three to the nozzle jet. This compares in Figure 5 with
nozzles on the sloped sides have angled out to 64 an increasing size of rock chip that will be
degrees from the vertical. carried away as you progress down the face of the
cutter toward the bit body.
Case 6: 3 #12 Nozzles, 3 Baffles, 300 CPM
Coolinq
The model for this section uses the same bit
as in the base case. The three nozzles on the The cooling of the cutters is dependent on
sloped sides have been blanked off. All of the the mud flowing past the cutters. It is necessary
mud flow now comes out of the three #12 nozzles on to place the cutters in relation to each other,
the top surface. The flow rate is 300 gpm and the nozzles and the flow baffles in such a way as
there are three flow baffles. to maximize the cooling of all the cutters. As
with the cleaning, the overall flow pattern as

—
SPE 14221 R. Crouse & R. Chia 5

seen in Figure 3 is a good indication of the can be seen in Figure 11 which shows practically
cooling of the cutters. Since this is only an no erosion. This indicates a large reduction in
overview, it is necessary to look at each cutter the turbulent flow in the center of the bit which
Individually. is responsible for the cooling and cleaning of the
center cutters. This shows that even though the
The rate of cooling of the cutters ia flow rate la the same, it is very important how
dependent on the heat transfer coefficient. the energy of the jets is used to clean all of the
Figure 6 shows contours of heat transfer cutters.
coefficient for cutter #9 of the model described
in the baae case. The location of this cutter in The angle of the nozzles can be changed to
relation to the nozzle can be seen In Figure 1. direct the flow at the cutters. In the model
As seen in Figure 6, the heat transfer coefficient described in case 5, this was done. The
is very high near the cutting edge of the PDC nozzles were angled outward 25 degrees. Figure 12
cutter. This will ensure good cooling at the shows the erosion pattern resulting from this
point where the heat is being generated. Also, case. The erosion is leas because the flow is
there is good cooling along the aide of cutter and directed In the direction it must travel between
stud body. This ia due to the proximity of the the rock and the bit. The cutters in the center
nozzles cauaing high velocity flow to waah of the bit suffer from this approach. Now there
alongside this cutter. ia less mud striking the rock and flowing to the
center of the bit to clean and cool the cutters
Erosion there.

The nature of the mud flowing in the confined m.

~u ~nhazce the c~oltng and cleaning of the
region between the bit body and the rock surface center cutters on the top flat section of the bit,
is such that it can cause severe body erosion. the three nozzlea on the side of the bit were
Designing the bit to achieve maximum veiocity to klm-lcd off. This is described in case 6.
“--------
improve the cleaning and cooling of the cutters The erosion pattern for this case IS shown in
will cause the greatest body erosion. To reduce Figure 13. This figure shows more erosion on the
the erosion some bits have “~eiisp?fiyedk-ith top surface and some erosion on the top edges of
tungsten carbide to create an erosion resistant the flow bafflea. This is due to aii of the
surface. Since measuring what is happening to the flow coming out of the top three nozzles. The
mud while a bit is drilling in the field is very flow around the cutters on the sloped side of the
difficult, a good comparison of the flow pattern bit is reduced for this caae becauae the flow is
is the resulting wear pattern on the dull bits. no longer pushed away from the nozzles on the
aides toward the cutters. A design like this will
Figure 7 shows the expected erosion patterna 14 0 -f
have better cieardng aridcco..ne - the cutters on
on a dull bit for the caae described in the baae the top surface but poorer cleaning and cooling of
case. Th.
. ... .m~ters
_ .- and nozzles were not shown On the cutters on the sloped aidea. Also excessive
thia figure for clarity. The computer analyais erosioiiof the bit body wmdd occur.
showa that the bit will erode just above the
nozzle on the sloped surface. Figure 8 is a Pump-off
photograph of this bit after it came back from the
fieiti. it sho~s the bcdy eroded where the Because of the restrictive area the mud must
computer predicted the erosion would occur. The flow through “Mtweeiittiebit szd the rock, there
erosion at this location ia due to the is a large pressure drop accross the face of the
reei=eui~tin~ fi~~ between the nozzle on the top bit . This is going to vary with how many cutters
and the one on the sloped surface. tuneflow miiatgc sre”uzd$the restriction due to
the baffles and the flow rate. This higher
By increasing the flow rate from 300 gpm to pressure under the bit will reduce the weight on
500 gpm the amount of erosion increases. Also the bit. Table 1 lists the llft-off force for the
there are more areas where erosion will occur. six cases presented here. For the baae case the
Figure 9 shows the erosion for the model described weight on bit is reduced by 549 lbs. The second
in case 2 where the flow rate waa increaaed to case with an increased flow rate shows a large
500 gpm. jump in the wieght reduction. The third case
shows a reduction In the lift-off because of the
Removing the flow baffles will allow the mud increased flow area under the bit. The fourth
to spread out more and thus decrease its velocity. case alao shows a reduction because of the
This will lessen the cleaning and cooling of the reduction in turbulence allowing the mud to just
cutters. Figure 10 shows this case as described flow out from under the bit. The fifth case show
in case 3. Here the erosion is less over the very little change due to the nozzle angle. The
bit, resulting from the reduced mud velocity. last caae has a small increase becauae due to all
of the flow going out the top three nozzles.
The eroafon seen on the body ia due to the
high velocity jet coming out of the nozzle, ACKNOWLEDGEMENTS
hitting the rock surface then circulating down and
hitting the bit body. By increasing the nozzle The authors thank Sii Smith Tool for
size to 16/32 inch as described in case .4, permiaslon to publish this paper.
the velocity of the jet ia greatly reduced. This
6 OPTIMIZATION OF PDC BIT HYDRAULICS BY FLUID SIMULATION SPE 14221

nl?vrmv
mctr.nn,Y PwC2
w. II LIFT-OFF FORCE FOR DIFFERENT DESIGNS

(I) Myers, H. M., and FurdK, J. ~., “Fliiid %r=tics

in a Diamond Drill Bit,” Society of Petroleum no. of flow rate no. of nozzie iiit-off
Engineers Paper SPE 1696, 1967. a~4Jli__ ~ _ force lbf

(2) Glowka, D., “Optimization of Bit Hydraulic

Configurations,” Society of Petroleum ~~ fj#!~ 300 3 5 549
Engineers Paper SPE 10240, 1981.
R1 6 #12 500 3 5 1545
(3) Ortega, A., and Glowka, D. A., “Studies of the
Frictional Heating of Polycrystalline Diamond R2 6$12 300 0 5 343
Compact Drag Tools During Rock Cutting,”
Sandia National Laboratories Report SAND 80-267 R3 6 #16 300 3 5 309
June, 1982.
R4 6 #12 300 3 30 583
(4) FETF’LU,The Pacifica Technology Finite Element
Thermal and Fluid Dynamics Computer Program R6 3 #12 300 3 5 642
Series, Originally Developed by il.K.
Gartling.

(5) Engieman, M. S.,’’FIMP-A Fluid I!pamies

Analysis Package”, Advances in Software
y ‘* pa.
:.
/ FLOU BAFFLES

Engineering, 1982.

(6) Welch, J. E., Harlow, F. H., Shannon, J. P.,

Daly, B. J., “ The FL4CMethod, A Computing
Technique for Solving Viscous, Incompressible,
Transient Fluid-Flow Problems Involving Free
Surfaces, Los Alomos Scientific Laboratory,
LA-3425, November, 1965.

y4 “?pl~ , B, D., Romero, N. C.>

(7) Hirt, C. %!.,.-..,-
“SOLA - A Numerical Solution Algorithm for
Tranaient Fluid Flows”, Los Alomoa Scientific
~=~~~tnrv.--.,, LA-5852, April, 1975.

(8) pATRAN, A Finite Element Modelling Program

from PDA Engineering.

(9) Finnie, I., Wolak, J. and Kabil, Y., “Erosion

of Metals by Solid Particle”, J. of
Materials, Vol. 2, No. 3, Sept. 1967, PP. im”
682-700.
Fig. 1—8.5-in. Smithpax bitdesign.
(10) Roco, M. C., Nair, P., Addie, G. R., Dennis,
J,, ‘%roaion of Concentrated Slurries in
Turbulent Flow”, Energy Sources Technology
Conference, New Orleans, February, ~984, pp.
69-77.

(11) Graf, W. H., lQHydraulics of Sediment

.

!
Transport”, McGraw-Hill, 1971, pp. 89.

(12) Winters, W. J., and Warren, T. M., “Determing

the True Weight-on-Bit for Diamond Bits”, .
Society of Petroleum Engineers Paper SPE 11950,
1983.

ATTACK aax

Flg.2—Normalized erosionvs. angle ofsttackformud flow.

 tNlimm3 ,/ ,/3s 15 613
,, ,/B5 15: 613
s# ,2 Inzzi”. a V*, z *!!!-
Iivleum T . .l!S.3E+ Cm-E= =
1.41 x = .3533475
6 ●12 M2ZkS. = W. 3 ‘i{ fl- ZY.=1-.W=.
T= .1 S-? CYCLE= ~
xzPLaI’a N-m-

Fig. 4-CI’OM section showing the mud flow.

F@ 34%38 flow ●long thO rock Wfw.

-.

-J
x

— ——

F19. ~cat.trmmbr rate for Cuner ~.

F~. ~ska of chip removalfor Cutter3.

SPE ]. 4221
 \
KSSTART 8. SPAX 320 U&F&: 12 ‘Z]es,
8.5 941TWAX w!TKUT

Fig. 10-Ero310n along the blf body for C@se 5.3.

Fig. 4-EroelOn along the bit body for Caee 5.2.

..

i
k fWSTWT ?s30 gpf 6 # 12 mzzkJ< -W WZ
REST#7T 6 tl16 ?07.21. S, = 9P. 3 bffla &53fil TWAX I
8.s 3411TWAX BIT

Fig. 12-Er03im ●long tha bit body for Case 5.5.

Fig. 1 l–~ along the blt body for Ca3e 5.4.

RSSTMT & gpf 3 u 12 nozzles, 3 haffl I

8.5 3H1TWriX 1

Fig. 12-ErtmlOn along the blt body for Cme 5.6.

sPE 14221