International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52

Contents lists available at ScienceDirect

International Journal of
Rock Mechanics and Mining Sciences
journal homepage: www.elsevier.com/locate/ijrmms

Estimating rock strength parameters using drilling data T

b,⁎ a b
Sajjad Kalantari , Hamid Hashemolhosseini , Alireza Baghbanan
a
Department of Civil Engineering, Isfahan University Of Technology, Iran
b
Department of Mining Engineering, Isfahan University Of Technology, Iran

A R T I C L E I N F O A B S T R A C T

Keywords: Estimating rock strength parameters using operational drilling data can be a fast and reliable method. In this
Estimating rock strength parameters case, several researchers have proposed different analytical models based on force or energy equilibrium
Operational drilling data methods. Most of them propose methods to estimate uniaxial compressive strength through the investigation of
Analytical model interaction between the bit and rock in drilling process. Although in the proposed models, operational drilling
system, rock strength parameters, bit geometry and contact friction were considered, some of the important
factors such as crushed zone and its mechanical properties, contact frictions between the bit and rock and
friction between the rock and crushed zone need to be explicitly considered. In this research work, a theoretical
model is developed based on limit equilibrium of forces and considering contact frictions, crushed zone and bit
geometry in the rotary drilling process by a T-shaped drag bit. Based on the model, a method is used to estimate
rock strength parameters form operational drilling data. The operational drilling parameters such as thrust force,
torque, rate of penetration and speed of rotation were obtained by a developed portable drilling machine. The
portable drilling machine is able to drill the rocks with different strength range coincident with measure and
record the parameters. A set of drilling experiments were conducted on three different rocks ranged from weak,
medium and hard strength. Obtained results based on proposed model for uniaxial compressive strength, co-
hesion and internal friction angle of rock are well fitted to the results of the conventional standard tests.

1. Introduction
Unconfined Compressive Strength (UCS) of intact rock has been
considered as the main parameter in most rock mass classification
systems and also as a high limit of the rock mass strength. Determining
UCS using suggested standard methods is costly and time consuming
mainly due to difficult sample preparation and destructive procedure.
Therefore, many indirect methods such as scratch test, point load,
Schmidt hammer and punch tests have been proposed for this purpose.
Estimating the UCS of rock using drilling parameters is also an alter-
native method. In this method, continuous measurement of drilling
parameters and evaluating of rock strength during the drilling is pos-
sible. Beside the advantage of continues measuring, it can be used as a
quasi-nondestructive method in field. In large scale, this method can be
easily facilitated due to simply movement and no need for sampling. In
this case, the drilling methods as well as the bit are most important.
Regarding extensive use of rotary drilling systems, it has attempted
more attention compared with the other systems and most of the re-
presented methods are accommodated in this drilling system. The
drilling process is conducted through two stages of feeding and cutting.
At the first stage, the bit indents into the intact rock by the thrust force.
As a result of the indentation, a part of rock is crushed and pressed. The
compressed crushed zone is formed in front of bit due to the rake angle,
cutting face friction and spiral motion of the bit. The next stage of ro-
tary drilling occurs due to the rotational motion of the bit. At this stage,
the horizontal force is transferred to the rock along rotating bit but
unlike the first stage, the force is transferred to the rock through an
intermediary. The intermediary is the compressed crushed zone, which
damps a part of the force as frictional one. The crushed zone is more
important in investigating the interaction between the bit and rock as
well as estimating the strength parameters of rock. This issue, mostly
have not been considered in previous proposed models.
According to the literature, previous proposed models which esti-
mate rock strength using drilling technique or propose methods for this
aim through the investigation of interaction between the bit and rock is
categorized in two forces equilibrium-based and energy equilibrium-
based models. The former category has been introduced by Evans1 as
“Indentation Model” which estimates the necessary force to coal
plowing. It was also further developed by Nishimatsu2 as “Shear Model”
for estimating the necessary force to shear failure of rock which is based
on Merchant's metal cutting Model.3 There are other models in this
category which are developed based on different types of bits such as
Detournay et al.4 and Gerbaud et al.5, Chiaia,6 Wojtanowicz et al.7 or
Nakajima et al.8 models for drag and Polly crystalline Diamond

⁎
Corresponding author.

https://doi.org/10.1016/j.ijrmms.2018.02.013
Received 10 January 2017; Received in revised form 1 December 2017; Accepted 5 February 2018
1365-1609/ © 2018 Elsevier Ltd. All rights reserved.
S. Kalantari et al.
Compact (PDC) bits, Franca9 and Roxborough et al.10 and Hareland11
models for roller-cone bits. They proposed relationships between the
rock strength parameters and operational drilling parameters based on
drilling tools geometry. However except Nakajima et al.8 and Gerbaud
et al. model5, the effect of crushed zone has not been considered. The
Gerbaud et al. model has been aimed to investigate the interaction
between the bit and rock in order to increase the efficiency of PDC bits.
While compressed crushed zone is the main factor of cutting force
transmission from bit to the rock. The energy equilibrium-based
methods are adopted from the drilling specific energy, which is initially
introduced by Teale12 and represents the minimum required energy for
drilling a unit volume of rock as an index for any type of rock. This
amount of drilling specific energy is approximated as UCS of rock.
Several researchers13–22 used different methods to estimate rock
strength and bit wearing in rotary drilling by roller-cone, drag and PDC
bits based on the drilling specific energy of Teale.12 The energy equi-
librium-based analytical models resolve simply the problem of friction
and the crushed zone in drilling process. However they do not propose
an explicit relationship between the UCS of rock and drilling para-
meters. In this case, there is not any specific method for measuring the
minimum drilling specific energy.
Detournay and Defourny4 proposed a method to estimate the
minimum drilling specific energy (from cutting point), which is used to
estimated UCS of rock in scratch test. This method also can be used as a
method to estimate UCS in drilling process. In this research work, a
model is developed based on the limited equilibrium of forces in dril-
ling process by a drag bit. In the model, parameters of cohesion and
internal friction angle of rock are estimated from operational drilling
data and based on these parameters, the UCS of rock is calculated. At
the same time, the UCS of rock is estimated from the cutting point. In
addition the effects of different parameters on estimated strength
properties are also evaluated. A sophisticated drilling machine is de-
veloped and operational drilling parameters of three types of rocks with
different range of strength from weak in travertine to medium in onyx
and high in rhyolite were examined. The machine is able to drill the
rocks with different strength range coincident with measure and record
the parameters of torque, thrust force, indentation rate, rotation speed
and drilling depth via two independent motors. The obtained results
from the drilling experiments for strength parameters of rocks have
been compared with those results from the conventional standard tests.
2. Developed analytical model
Generally, the rotary drilling process of rock is conducted through
two stages. In the first stage, the indentation performs along with axial
motion of the bit, which is included the indentation, crushing and
smashing. The second stage is the cutting force, which is perpendicular
to the axial motion and only cuts the rock. Both stages require to per-
form the drilling process, since the axial motion of bit needs to keep the
cutting depth and feeding and the rotary motion perpendicular to the
indentation motion needs to cut the rock. The two stages occur in a
continuous and fast process and can be considered simultaneously. The
resultant of the two forces have important role. One of the resultant
forces functions along the bit axis and the other one is the resultant of
forces which are parallel to the bottom hole. The amount and ratio of
the two forces in frontal face of the bit mostly depend on the bit geo-
metry especially rake angle and the contact friction. Hereon, the fric-
tion force between the compressed crushed particles and rock as well as
the bit-end (head) friction affect the forces. Considering the condition, a
model is proposed based on the force limit equilibrium in a T-shaped
drag bit with negative rake angle, α according to Fig. 1(a) under the
atmospheric pressure of drilling process. For a blunt bit, its cross section
is rectangular and the area of wearing flat is l×D, where D is bit dia-
meter and l is the width of wearing flat. When the bit is ideally sharp,
the width of wearing flat is zero. By helicoidal motion of the bit over-
comes the rock strength/resistance and indents into the rock with thrust
46
International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52
on the bit. Concurrent with the indentation, the crushed materials are
confined and compressed between front face of the bit and rock.
Several factors control the formation of the crushed zone and its
compacting in front of the bit. These factors include the rake angle of
bit, continuous downward movement of the bit, cutting face friction
angle, and the presence of an enclosed space inside the drill hole, which
does not allow to move upward the cuttings material from bit in front.
In fact, the compressed crushed zone performs as the initial indenter
into the intact rock. The compressed crushed zone also transfers a part
of the applied forces of bit to the intact rock. In other words as a result
of the applied force (Fc) by the bit. The crushed zone transfers some part
of the force to the intact rock along the vertical cross-sectional area,
which equals the depth of cut multiple bit diameter. The other part of
the force transfers to the bottom of hole by compressed crushed zone.
Compressed crushed zone is in contact with different strengths of the
intact rock and bit. Due to the formation of a compressed crushed zone
in front of bit, bit cutting face and rock are subjected to a uniform
(hydrostatic) stress (σ0 ). The compressive stress in the crushed zone and
rock is enough to shearing and chip propagation along linear line.
When a sufficient large horizontal force transferred by the compressed
crushed zone into the rock, the rock fails with an angle of ψ and chip is
formed (Fig. 1(a)). The retention angle (β ) of bit is considered as zero to
decrease the contact of the bit perimeter (Fig. 1(b)). If bit moves ro-
tationally along cutting direction, the side rake angle (γ ) will be zero.
Since the hydrostatic condition is dominant in the compressed crushed
zone,5 the amount of the stress would be the same at different direc-
tions. The contact friction angle of bit cutting face and bit end wearing
face are the same and is considered as θf , since both of them have the
same material (tungsten carbide), and are in contact with the boundary
layer of crashed rock. In this condition, three different media of bit,
compressed crushed zone and failed intact rock are in contact.
In n-t coordinate system, the force, Ft is acting force in the direction
of bit rotation results from bit torque and the force Fn, along the normal
axis results from thrust force. Each of the forces has two components;
one of which is the cutting force and another one is the frictional force
as follows:
Fn = Fnc + Fnw (1)
Ft = Ftc + Ftw (2)
where Fnc , Ftc ,
and Fnw Ftw
are normal and tangential components of the
cutting force, normal and tangential component of the friction force
respectively. These parameters are linked with the following relation-
ship,
Fnc = Ftc tan (α + θf ) (3)
Ftw = Fnw tan θf (4)
τ0 = σ0tanφ′ (5)
A = dD (6)
where α, θf, φ′, τ0 and σ0 are the rake angle, contact friction angle in the
bit end wearing face and contact friction angle in bit cutting face,
frictional angle between compressed crushed zone and intact rock,
shear strength between compressed crushed zone and intact rock, hy-
drostatic pressure in compressed crushed zone respectively. A, d and D
indicate effective vertical cross-sectional area of cut, cutting depth and
the bit diameter respectively. It should be noted that for a helicoidal
motion of a drag bit, when the drilling is in stationary condition, Eq. (6)
is satisfied.6 Now, considering the compressed crushed zone, we have,
Ftc = Aσ0 + Aτ0tanα (7)
Fnc = Aσ0tanα + Aτ0 (8)
Also in the potentially fail face of wedge, shear, τ and normal, σ
tractions are as follow,
S. Kalantari et al. International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52

Fig. 1. (a) Geometry and mechanism of the acting forces in the proposed analytical model, (b) Geometric parameters of bit.

σ = σ0sin2 ψ − τ0cosψsinψ (9) have,

τ = σ0sinψcosψ + τ0sin2 ψ (10) 2C cos (φ′ − φ)

Ft = Fntanθf + A (
1 + tanφ′tanφ
Considering Mohr- Coulomb failure criterion of rock,
1 + tanφ′tanα − (tanα + tanφ′) tanθf
τ = C + σtanφ (11) × )
cos (φ′ − φ) − sin2 (φ′ − φ) + sin (φ′ − φ) (20)
where C and φ are cohesion and internal friction angle of rock re- Since A (vertical cross-sectional area of cut) is a function of a bit
spectively. The values of τ and σ from the Eq. (9) and Eq. (10) are as- diameter and depth of cut, the effect of bit diameter size and the speed
signed in the Eq. (11) then we have, of bit rotation on the results is almost removed when both sides of the
(σ0sinψcosψ + τ0sin2 ψ) = C + (σ0sin2 ψ − τ0cosψsinψ) tanφ (12) Eq. (20) are divided on A. In this case, two new terms of ε = Ft / A as the
drilling specific energy and ζ = Fn/ A as the drilling strength are in-
If τ0 is replaced with σ0tanφ′ (from Eq. (5),) and after some algebraic troduced and Eq. (20) changes the form as follows,
manipulation we have,
2Ccos (φ′ − φ)
1 1 ε = ζtanθf + (
σ0 ⎡ sin2ψ + tanφ′sin2 ψ − sin2 ψtanφ + tanφ′tanφsin2ψ ⎤ = C 1 + tanφ′tanφ
⎣2 2 ⎦ (13)
1 + tanφ′tanα − (tanα + tanφ′) tanθf
In order to obtain the minimum value of σ0 Eq. (13) is derived in × )
cos (φ′ − φ) − sin2 (φ′ − φ) + sin (φ′ − φ) (21)
respect to ψ. Under this circumstances the value of ψ will be obtained as
follows, This equation indicates a linear relationship between drilling spe-
cific energy and drilling strength of rock for a blunt bit, in which the
∂σ0
= 0, slope is tanθf . The intercept of this linear relationship is the second term
∂ψ (14) of Eq. (21) in right side, which is considered as ‘b’. The obtained ex-
π φ′ − φ perimental results of the drilling experiments conducted by different
ψ= −( ) researchers also indicate the presence of a linear relation between the
4 2 (15)
vertical and horizontal acting forces in the rock cutting.4,12,23–25
Finally minimum value of σ0 is obtained as follows, Denoting Eqs. (1), (2), (3), (4) in different form, we have:
2C cos (φ′ − φ) 1 Fn = Ftc tan(α + θf ) + Fnw
σ0 = × (22)
1 + tanφ′tanφ cos (φ′ − φ) − sin2 (φ′ − φ) + sin (φ′ − φ)
(16) Ft = Ftc + Fnw tanθf (23)
Now, considering the acting forces in the bit as well as the Eqs. (1), By subtracting the Eq. (22) from Eq. (23), the following equation is
(2), (4), (5), (7), (8), we have: achieved,
Fn = (σ0 Atanα + σ0tanφ′A) + Fnw (17) Ft − Fn = Ftc [1−tan(α + θf )] + Fnw (tanθf −1) (24)
Ft = (σ0 A + σ0tanφ′Atanα ) + Fnw tan θf (18) If Ftc
from the Eq. (2) is replaced with Ft −Ftw , we will have the
following relationship using Eq. (4):
The following equation is obtained by multiplying the value of
(−tanθf ) in the both sides of the Eq. (17) just addition the result to Eq. 1 1
(18), Ft = Fn + ⎜⎛1− ⎞ w
⎟ Ft
tan (α + θf ) ⎝ tan θf tan ( α + θf ) ⎠ (25)
Ft = Fntanθf + Aσ0 [1+tanφ′tanα − (tanα + tanφ′) tanθf ] (19)
When a bit is ideally sharp, the tanθf value is not zero, but bit-end
If the value of σ0 from Eq. (16) is placed in the Eq. (19), then we trends toward zero. Therefore the value of Ftw is negligible and the

47
S. Kalantari et al.
normal force of the bit will be transferred to the rock via horizontal
cross-sectional area of bit which is the result of the rake angle and
friction between the bit and compressed crushed zone. In other words,
the friction angle θf appears in this manner using a sharp bit. In this
case, the slope of the Ft-Fn relationship is 1/ tan (α+θ′f ) . This condition
applies to the ε−ζ relationship as well and the Eq. (21) is changed as
follows,
1 20
ε= ζ
tan (α + θf )
According to this equation, when a bit is sharp the rake angle and
friction between the bit and compressed crushed zone are deciding
factors. Meanwhile, the rake angle could be controlled by operator and
therefore optimizing of its value is necessary.
The effective range of the parameters must be investigated to esti-
mate the strength parameters of rock (internal friction angle, cohesion
and UCS of rock). According to the proposed model, the Eq. (21) and
Eq. (26) are the only output of the model. These linear equations obtain
from the trend lines of plotted drilling data (ɛ = Ft / A , ζ = Fn/ A ) in blunt
and sharp bits respectively. The slope and the intercept of linear Eqs.
(21) and (26) are two important factors, and both are presented in a
blunt bit, while only the slope of line exists in the sharp bits. In the both
conditions (sharpness and bluntness), the sloped curve is used to esti-
mate the internal friction angle of rock. In order to estimate the internal
friction angle of rock in a blunt bit, given rake angle (α ) and the value
of (θf) from the slope of ε−ζ linear curve, if the Eq. (8) is divided by Eq.
(7) and is placed equals to tan (α+θf ) , then the value of φ′ will be ob-
tained from the following equation,
tanα + tanφ′
tan (α + θf ) =
1 + tanαtanφ′
By estimating the friction angle between intact rock and compressed
crushed zone (φ′) from Eq. (27), the internal friction angle of rock (φ)
from the following equation that has been introduced by Kerisel26 and
Gerbaud,5 is obtained:
π
tanφ′ = tanφ
2 (28)
According to this model, variation of the slope of ε−ζ linear curve
with internal friction angle of rock (φ ) and the friction angle between
intact rock and compressed crushed zone (φ′) are shown in Fig. 2.
According to the Fig. 2, if the slope of ε−ζ linear curve changes in a
range of 0.45–0.75, the estimated internal friction angle of rock is also
ranged between 31 and 60° and the estimated friction angle between
compressed crushed zone and intact rock is also altered in a range of
43–69°. The effect of rake angle (α ) on both φ and φ′ is presented in
Fig. 3. As can be seen, when the rake angle is smaller than 20°, its effect
on estimated internal friction angle of rock and estimated friction angle
between compressed crushed zone and intact rock is negligible.
Another important parameter in this model is the intercept of the
90
φ φ'
80
70
60
φ , φ' (degree)
50
40
30
20
10
0
0 0.2 0.4 0.6 0.8
tanθf
Fig. 2. Variation of estimated internal friction angle of rock (φ ) and the friction angle
between the rock and compressed crushed zone (φ′) with slope of ε − ζ diagram.
International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52
80
φ φ'
70
60
50
φ, φ'(degree)
40
30
(26) 10
0
0 5 10 15 20 25 30 35
α(degree)
Fig. 3. Variation of estimated internal friction angle of rock (φ ) and friction angle be-
tween the compressed crushed zone and rock (φ′) with rake angle.
linear relation between ε and ζ of blunt bit, which is used to estimate
cohesion of rock. This parameter could be affected by different para-
meters such as φ , φ′, θf and the rake angle (α ) according to Eq. (21). All
of these parameters also control aspect ratio of ε−ζ linear curve, which
show a bilinear behavior where beyond tanθf = 0.5 the curve follows a
sharp trend (Fig. 4). The effect of rake angle according to the model on
estimated cohesion is shown in Fig. 5. According to the figure when the
rake angle is smaller than 25° its effect on the estimated cohesion is
negligible. The bits with a rake angle, smaller than 20° are preferable.
Now given the internal friction angle of rock and its cohesion, the un-
confined compressive strength of rock (qc) is calculated based on the
Mohr-Coulombs criterion as follows,
2Ccosφ
qc =
(27) 1−sinφ (29)
The UCS is also estimated using drilling data from ε−ζ diagrams for
the blunt and sharp bits. In this case, the vertical distance of intersec-
tion of the two linear diagrams from the drilling strength axis (ζ ) is
considered as UCS of a rock. This intersection point is the cutting point,
which has been defined by Detournay and Defourny.4
3. Drilling machine
A portable drilling machine (Fig. 6) was developed and used to
estimate required parameters for developed analytical model. In this
case different rock samples ranged from weak to medium and high
strength values are selected and examined. The machine is able to drill
in both field and laboratory scales. The machine measures and records
drilling data in a rock such as torque, thrust force, speed of rotation,
rate of penetration, and drilling depth. Two different servomotors are
used (Fig. 7). One of them provides thrust force up to 1200 N, which is
transferred to a rock by a ballscroll via a timing belt. The other ser-
vomotor called drilling servomotor, which is connected to the gearbox
with a maximum capacity of 6 N m and provides the necessary torque
for drilling. The latter servomotor is placed on the ballscroll and the bit
is connected to the gearbox through a drill chuck. The drilling depth is
1
Fig. 4. Variation of the slope of ε − ζ diagram (tan θf) and estimated cohesion of rock.
48
S. Kalantari et al. International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52

30 (δ=24 mm/min & 420rpm & D=10mm & Sharp Bit)

800 Test-2-1 Test-2-2 Test-2-3 Test-2-4
25
700

20 600
Cohesion (MPa)

Thrust Force (N)

500
15 400
300
10
200
5 100
0
0 0 5 10 15 20 25 30 35 40 45
0 5 10 15 20 25 30 35 Depth (mm)

α(degree) Fig. 9. Measured thrust force versus drilling depth in a drilling test on onyx with different
repetitions.
Fig. 5. Variation of the estimated cohesion of rock with the rake angle.

controlled and recorded by two displacement-meter sensors on the

platform.
The machine is operated with different rate of penetration, rotary
speed, drilling depth, maximum value of the thrust force and the
maximum rate of torque with a computer or touch screen. The system
starts to drill as the self-control and measures also torque parameters
and thrust force for drilling in different depth and then saves the data in
an Excel file. The diagram of the thrust force and torque in different
drilling depth for a green onyx which are presented in Figs. 9 and 10
respectively. The rate of thrust force and its relative torque are observed
in the diagrams for a repeated test. As can be seen that after initial
violation in the surface, the values approaches to a constant rate (sta-
tionary drilling condition). The recorded thrust force and torque are
changed when different indentation rates as well as different rotation
Fig. 6. An overview of the developed drilling machine.
speed are proposed in machine. The recorded thrust force and torque
are changed when different indentation rates as well as different rota-
tion speed are proposed in machine. T-shaped drag bit (Fig. 8) with
tungsten carbide is used as cutting tool in this machine. Since drilling
machine is portable it can easily adopted for in situ drilling of outcrops
and also dimension stone with different sizes.

4. Experimental results

Different drilling tests were conducted on rocks with different UCS

Fig. 7. A schematic view of the developed drilling machine.
in lemon travertine as a weak rock, green onyx as medium rock and
rhyolite with as hard rock. Lemon travertine and green onyx consist
only one mineral composition while rhyolite with a porphyry fine grain
texture contains different minerals such as quartz, sanidine, plagioclase,
phenocrystals and biotit.
The drilling tests are conducted using 8–9 mm T-shaped drag bits
with tips of tungsten carbide (Fig. 8). The rake angles were 15–20 de-
grees according to new model (Figs. 3 and 5) and of other re-
searchers.25,27–29 Two blunt and sharp bits, were facilitated in drilling
tests. Both bit types have a rectangular cross section while end-bit
thickness of sharp and blunt bits are < 0.2 mm and > 1.2 mm re-
spectively as also used in conducted drilling test by Detournay and

(δ=24 mm/min & 420rpm & D=10mm & Sharp Bit)

1.8 Test-2-1 Test-2-2 Test-2-3 Test-2-4
1.6
1.4
Torque (N.m)

1.2
1
0.8
0.6
0.4
0.2
0
Fig. 8. T-shaped drag bits with tips of tungsten carbide used in laboratory. 0 5 10 15 20 25 30 35 40 45
Depth (mm)

Fig. 10. Measured torque versus drilling depth in a drilling test on onyx with different
repetitions.

49
S. Kalantari et al. International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52

1200
SHARP BLUNT all the three rock types were measured at laboratory following the
suggested standard methods by International Society of Rock Me-
Drilling Especific Energy (MPa)

1000
chanics (ISRM) and tabulated in Table 1. Figs. 11, 12 and 13 show the
= 0.57 ζ - 3.41
800 R² = 0.99 plots of drilling specific energy (ε = Ft / A ) versus drilling strength
600
(ζ = Fn/ A ) data. The presented data have been obtained from the results
of drilling tests on the three types of rocks with blunt and sharp bits
= 0.41 ζ + 13.26
400 R² = 0.99 according to the required parameters in the developed analytical
model. Cutting force (Ft) is calculated from the measured torque and
200
normal force (Fn) is obtained from the thrust force. Each data point on
0 these figures represents as a pair of drilling specific energy and drilling
0 500 1000 1500 2000
strength, which were obtained from the different recorded operational
Drilling Strength (MPa)
drilling parameters with sharp and blunt bits.
Fig. 11. Drilling specific energy versus drilling strength on travertine with a blunt bit Correlation between the drilling specific energy (ε ) and drilling
(D= 8 mm, α = 15°) and a sharp bit (D= 8 mm, α = 15°). strength (ζ ) data is a linear curve in to the sharp and blunt bits using a
least-squares regression method as shown in Figs. 11–13. Their corre-
700 lation coefficients are almost larger than 98%. This linear correlation
BLUNT SHARP
satisfies the analytic relationships (21) and (26). Therefore, in these
Drilling Especific Energy (MPa)

600
figures, there are two trend lines, one based on the data obtained from
500 = 0.34 ζ + 0.88
R² = 0.98 the drilling with a sharp bit and the other based on the data obtained
400 from the drilling with blunt bit. The sharp bit trend lines pass almost
300 from the origin of the coordinates. However, these trend lines, due to
= 0.24 ζ + 33.29 the different drilling efficiency in rocks, are deviated very small from
200 R² = 0.99
the origin of the coordinates and can be neglected. The obtained trend
100 lines from blunt bit have intercepts with relatively low slopes. Al-
0 though, these trend lines were obtained from operational drilling with
0 500 1000 1500 2000 blunt bit with different end wearing area, in all the three rocks corre-
Drilling Strength (MPa) lation coefficient for linear diagrams exceed 98%. This indicates that
Fig. 12. Drilling specific energy versus drilling strength on onyx with a blunt bit the slope of the friction line (trend line of blunt bit) depends only on the
(D= 9 mm, α = 20°) and a sharp bit (D= 9 mm, α = 20°). friction between the bit end wearing face and rock. This phenomenon
also can be seen in Eq. (21). Based on the estimated slopes of linear
1200 BLUNT SHARP curves of ε−ζ , (trend lines) of sharp bit and the estimated slope and
intercept of linear curves of ε−ζ (trend lines) of blunt bit in Figs. 11, 12
Drilling Especific Energy (MPa)

1000
and 13, the rock strength parameters are calculated according to the
= 0.34 ζ + 2.4
800 R² = 0.98 new method. The summary of the calculated strength parameters of the
rocks including internal friction angle, cohesion, and UCS are given in
600
Table 2. As can be seen in Figs. 11, 12 and 13, the slope of trend lines
= 0.26 ζ + 38.2
400 R² = 0.99 for sharp bits in travertine, onyx and rhyolite are 0.57, 0.34 and 0.34
respectively. Based on Eq. (26) the value of internal friction angle for
200 travertine, onyx and rhyolite are 45.3, 51.2 and 51.2 respectively. From
0
these figures, the slope of trend lines for blunt bits in travertine, onyx
0 500 1000 1500 2000 2500 3000 3500 and rhyolite are 0.41, 0.24 and 0.26 respectively. Based on Eq. (21) the
Drilling Strength (MPa) value of internal friction angle for travertine, onyx and rhyolite are
Fig. 13. Drilling specific energy versus drilling strength on rhyolite with a blunt bit 22.3, 13.5 and 14.6 respectively. Therefore, the obtained internal
(D= 9 mm, α = 20°) and a sharp bit (D= 9 mm, α = 20°). friction angles for all types of rocks are well fitted to those obtained
from the triaxial compressive tests when sharp bit is facilitated while
Table 1
they are far in blunt bit particularly for medium and hard rocks. De-
Measured mechanical properties of rocks from standard method. tournay et al.4 and Detournay et al.29 as well as Adachi et al.24 have
considered tanφ = tanθf ; in which the results do not fit with experi-
Rock type φ (degree) C (MPa) σc (MPa) mental results. The experimental results (Figs. 11–13) indicate that the
slope of linear curve of ε−ζ is smaller than one. In this case, the internal
travertine 47 8.5 48
Onyx 55 19 100 friction angle of rock is approximated smaller than 45° while the in-
Rhyolite 56 27.3 152 ternal friction angle of rocks generally is ranged between 35° and 60°.
Based on the new method in estimation of internal friction angle with
blunt bits, if the slope of ε−ζ curve changes in 0.45–0.75 ranges, in-
Richard.29 The penetration rates (drilling speed) and rotation speed for ternal friction angle of rock obtains 31–60° (Fig. 2). According to the
travertine, onyx and rhyolite were set as 12–36 mm/min and
300–360 rpm, 12–30 mm/min and 300–420 rpm and 6–24 mm/min
Table 2
and 360–420 rpm respectively. For each data point in Figs. 11, 12 and Obtained results from drilling experiments.
13, a drilling experiment with different rate of penetration and speed of
rotation was conducted. The number of tests in travertine, onyx and Rock type
rhyolite with sharp bit were 10, 12 and 9 respectively and with blunt bit
Parameters travertine Onyx Rhyolite
were 7, 9 and 8 respectively. Each test setup was repeated 4 times, as
shown in Figs. 9 and 10. When a drilling operation was in stationary φ(new model) (degree) 45.3 51.2 51.2
state, the parameters of thrust force and torque were recorded. The C(new model) (MPa) 12.1 21.1 25.2
average of the results of recorded parameters from different test re- qc(new model) (MPa) 58.3 119.8 143.12
qc(cutting point) (MPa) 56 111 154.5
petitions is calculated and used in derivations. Mechanical properties of

50
S. Kalantari et al.
obtained results in very weak rock, the slope of ε−ζ curve is relatively
high. Therefore, estimating the internal friction angle from blunt bits
based on Eqs. (27) and (28) is preferred. However approximated in-
ternal friction angle from slope of ε−ζ linear curve of blunt bit in
moderate and hard rocks is relatively low. Therefore the results from
sharp bit is appropriated based on Eq. (26) and φ = tan−1 (1/ a) − α . The
parameter of ɑ indicates the slope of ε−ζ linear curve of sharp bit. Al-
though the internal friction angle of rock is linked with the rake angle
of a bit, if the rake angle is smaller than 20°, its effect on estimated
internal friction angle of rock is negligible as shown in Fig. 3. Therefore,
a rake angle ranged between 10° and 20° in sharp bit is preferable.
Using proposed formula, φ = tan−1 (1/ a) − α in a sharp bit is preferred,
since to contact face of the bit (cutting face) with rock is of tungsten
carbide and due to high toughness of the tungsten carbide and the
existence of a boundary layer of failed rock in cutting face. Therefore, it
can be made similar condition in the tip end of bit. In this case, friction
between the bit and compressed crushed zone can be a sign of the in-
ternal friction angle of rock and the experimental results confirm that.
Intercepts of the ε−ζ linear curves of a blunt bits use to obtain co-
hesions of rocks. In this regard, to obtain the value of internal friction
angle of rock and friction angle between the rock and compressed
crushed zone are used Eq. (27) and Eq. (28). The slope of ε−ζ linear
curve of blunt bit is used to estimate θf. Estimated cohesion from the
intercept of ε−ζ linear curve of blunt bits for travertine, onyx and
rhyolite are 12.1 MPa, 21.1 MPa and 25.2 MPa respectively. The
amounts of cohesion obtained from triaxial compressive test are
8.5 MPa, 19 MPa and 27.3 MPa for travertine, onyx and rhyolite, re-
spectively. The estimated cohesion for travertine is slightly high com-
pared with onyx and rhyolite. This highness can be resulted of more
plastic deformity of the weak rocks, because the plastic deformities are
high in drilling process especially in soft rocks.21 The effect of slope of
ε−ζ linear curve on the rock cohesion which is revealed in the intercept
of linear Eq. (21), is negligible in the presence of tanθf and in the case of
Fig. 4 slope < 0.5. However its increase more than this value affects
the cohesion. The case also exists in rocks with very low strength.
Therefore, it is better to estimate cohesion conservatively about rocks
with weak strength. The rocks with moderate and heavy strength the
effect of tanθf on cohesion is negligible and estimated cohesion is closer
to the standard result test. The issue is observed about onyx and
rhyolite. Based on proposed model in present study, the effect of rake
angle on rock cohesion will be negligible in all cases if the rake angle
is < 20° (Fig. 5).
Now, the UCS is calculated using equation
qc = (2C . cosφ)/(1 − sinφ) by estimating the values of cohesion and in-
ternal friction angle of rock. According to Table 2, the estimated values
of UCS by this method for travertine, onyx and rhyolite are 58.3, 119.8,
and 143.1 MPa respectively. The values for the three rocks were 48, 100
and 152 MPa from UCS standard tests. Therefore, it has been over-
estimated 21% and 20% for travertine and onyx respectively and it has
been underestimate 6% for rhyolite. This suggests that the mentioned
method is a suitable one for rocks with high strength although it slightly
overestimates UCS of rocks with moderate and weak. Another method
to estimate the UCS of rock is use of ε−ζ curve of blunt and sharp bit. In
this method, a point is obtained from the intersection of linear curves of
blunt and sharp bits which is called Cutting Point.4 The vertical dis-
tance of the point from drilling strength axis (ζ ) in ε−ζ diagram gives
the minimum value of drilling specific energy or intrinsic specific en-
ergy (ϵ ) which is the UCS of rock. Based on this method the values of
UCS are 56, 111 and 154.5 MPa for travertine, onyx and rhyolite which
are different from the values obtained from UCS standard test for tra-
vertine, onyx and rhyolite as < 16, < 11 and < 2%. The cutting point
method also has a close conformity with the results of the proposed
model. The proximity of estimated values for UCS from both of the
methods (cutting point and new method) and standard experimental
results show that the effects of different parameters such as bit geo-
metry, friction and crushed zone has been well-defined in the new
51
International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52
model, and provide the validity of DD model. Therefore, the cutting
point method also can be used to estimate the UCS of rock.
5. Conclusions
Based on the limit equilibrium of forces model, a method used to
estimate cohesion, internal friction angle and UCS of the three types of
rock (weak, medium, and high strength range) from operational drilling
data. The effects of parameters such as crushed zone, friction and bit
geometry considered. The obtained results of experiments suggest that
the estimated strength parameters for the three rocks using the drilling
data have a good conformity with the results of the conventional
standard experiments. However, points must be considered in order to
obtain desired results: In order to estimate the internal friction angle of
rock from drilling data particularly for medium and heavy strength
rocks it is better to use a sharp bit (Eq. (26)), however the blunt bits can
be suitable for very weak rocks. To estimate the cohesion parameter of
rock it is necessary to use a blunt bit and the estimated value of co-
hesion using this method obtains from the intercept of linear diagram of
drilling specific energy - drilling strength of blunt bit. It is necessary to
investigate various rocks with weak strength ranges. The UCS can be
calculated based on the equation qc = (2Ccosφ)/(1 − sinφ) or using
cutting point method, due to conformity of the results. Cutting depth
has an important effect on the obtained results of strength parameters
of rocks. Using torque and thrust force resulted from lower indentation
rates (e.g. 0.1 mm/s) proposes confusing results. In other words, low
cutting depth affects the drilling specific energy. Therefore, it is better
to use higher indentation rates of drilling results to estimate strength
parameters of rock.
References
1. Evans I. A theory on the basic mechanics of coal ploughing. In: Proceedings of
International Symposium on Mining Research. London. vol. 2: 761–798; 1962.
2. Nishimatsu Y. The mechanics of rock cutting. Int J Rock Mech Min Sci.
1972;9:261–270.
3. Merchant ME. Basic mechanics of the metal cutting process. J Appl Phys.
1945;66:168–175.
4. Detournay E, Defourny P. A phenomenological model for the drilling action of drag
bits. Int J Rock Mech Min Sci. 1992;29:13–23.
5. Gerbaud L, Menand S, Sellami H. PDC Bits: All comes from the cutter rock interac-
tion. In: Proceedings of the IADC/SPE Drilling Conference held in Miami, Florida, USA;
21–23 February; 2006.
6. Chiaia B, Borri-Brunetto M, Carpinteri A. Mechanical modelling of the mechanics of
core drilling in geomaterials. Mach Sci Technol. 2013;17:1–25.
7. Wojtanowicz AK, Kuru E. Mathematical modeling of PDC bit drilling process based
on a single-cutter mechanics. J Energy Resour Technol. 1993;115:247–256.
8. Nakajima I, Kinoshita S. Theoretical studies on cutting force of rock fracture me-
chanism in rock cutting. J Min Metall Inst Jpn. 1979;95:49–55.
9. Franca LFP. Drilling action of roller-cone bits: Modeling and experimental validation.
J Energy Resour Technol. 2010;132:1–9.
10. Roxborough FF, Philips HR. Rock excavation by disc cutter. Int J Rock Mech Min Sci.
1975;12:361–366.
11. Hareland GA. drilling rate model for roller cone bits and its application. In:
Proceedings of the CPS/SPE International Oil & Gas Conference and Exhibition in China
Held in Beijing, China, 8–10 June; 2010.
12. Teale R. The concept of specific energy in rock drilling. Int J Rock Mech.
1965;2:57–73.
13. Karasawa H, Ohno T, Kosugi M, Rowley JC. Methods to estimate the rock strength
and tooth wear while drilling with roller-bits—part 2: insert bits. J Energy Resour
Technol. 2002;124:133–140.
14. Karasawa H, Ohno T, Kosugi M, Rowley JC. Methods to estimate the rock strength
and tooth wear while drilling with roller-bits—part 1: Milled-ToothTooth Bits. J
Energy Resour Technol. 2002;124:125–132.
15. Ohno T, Karasawa H, Kosugi M, Rowley JC. Proposed practical methods to estimate
rock strength and tooth wear while drilling with roller-cone bits. J Energy Resour
Technol. 2004;126:302–310.
16. Bingham MG. How rock properties are related to drilling. Oil Gas J. 1964;50:94–101.
17. Wolcott DS, Bordelon DR. Lithology determination using downhole bit mechanics
data. In: Proceedings of the 68th Annual Technical Conference and Exhibition of the SPE,
Houston, Texas; October 3–6; 1993; pp. 769–778.
18. Hoberock L, Bratcher GJ. A new approach for determining in-situ rock strength while
drilling. J Energy Resour Technol. 1996;118:249–255.
19. Burgess TM, Less WG. Measuring the wear of milled tooth bits using MWD torque and
weight-on-bit. In: Proceedings of the Drilling Conference SPE/IADC, New Orleans; 1985;
pp. 453–462.
S. Kalantari et al.
20. Pessier RC, Fear MJ. Quantifying common drilling problems with mechanical specific
energy and a bit-specific coefficient of sliding friction. In: Proceedings of the 67th
Annual Technical Conference and Exposition of the SPE. Washington, DC, October 4-7.
P. 373-387; 1992.
21. Warren TM. Penetration rate performance of roller cone bits. SPE Drill Eng.
1987;2:9–18.
22. Lia Z, Ken-ichiItakura. An analytical drilling model of drag bits for evaluation of
rocks strength. Soils Found. 2012;52(2):216–227.
23. Richad T, Detournay E, Drescher A, Nicodeme P, Fourmaintraux D. The scratch test
as a means to measure strength of sedimentary rocks. In: Proceedings of the SPE
/NSRM Eurock’ 98. Norway, 8-10July. P. 15-22; 1998.
24. Adachi JI, Detournay E, Drescher A. Determination of rock strength parameters from
cutting tests. In: Proceedings of the 2nd North American Rock Mechanics Symposium on
Rock Mechanics Tools and Techniques. Balkema, Rotterdam; 1996; p. 1517–1523.
52
International Journal of Rock Mechanics and Mining Sciences 104 (2018) 45–52
25. Jianyong P. Interpretation of single cutter tests for rock mechanical properties. In:
Proceedings of 46th U.S. Rock Mechanics/Geomechanics Symposium, Chicago; 2012.
26. Kerisel J. Cours de mécanique des sols. Ecole National des Ponts et Chaussées. Paris;
1975.
27. Rajabov V, Miska S, Mortimer L, Ozbayoglu E. The effects of back rake and side rake
angles on mechanical specific energy of single PDC cutters with selected rocks at
varying depth of cuts and confining pressures. In: Proceedings of the 2012 IADC/SPE
Drilling Conference and Exhibition, San Diego, California, 6–8 March; 2012.
28. Coudyzer C, Richard T. Influence of the back and side rake angles in rock cutting. In:
Proceedings of the AADE National Technical Conference and Exhibition, Houston; 5-7
April 2005; 2005.
29. Detournay E, Richard T, Shepherd M. Drilling response of drag bits: theory and ex-
periment. Int J Rock Mech Min Sci. 2008;45:1347–1360.