DC179729 DOI: 10.

2118/179729-PA Date: 17-February-16 Stage: Page: 71 Total Pages: 11

Hydraulics Calculations and Field

Application of Radial Jet Drilling
Wang Bin, Li Gensheng, Huang Zhongwei, Li Jingbin, Zheng Dongbo, and Li Hao,
State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum, Beijing

Summary jet bit to drill laterals (Fig. 2). The self-driving jet bit (Fig. 2) dif-
Multilateral-horizontal-well drilling is an efficient approach for fers significantly from a mechanical drill bit. It features forward
stimulating shallow, low-permeability, marginal, and coalbed- and backward nozzles to generate a high-pressure jet; the forward
methane (CBM) reservoirs. Radial-jet-drilling (RJD) technology, jet breaks the rock to create the borehole (30–50 mm), whereas
which uses a high-pressure water jet, aims to drill tens of laterals the backward jet drives the bit forward and expands the lateral di-
from a vertical wellbore. Hydraulics design is essential for suc- ameter behind the jet bit.
cessful field-drilling operations. However, detailed hydraulics cal- In addition, nearly 50–70% of jet-drilling-cuttings diameters
culations and design methods have not yet been published. are smaller than 1 mm (Chang 2006). Because the pump flow rate
In this study, the pressure loss, ejector force, and extending re- of RJD is lower than that of conventional drilling operations, most
sistance of an RJD system were measured at different flow rates cuttings flow back to the ground, whereas larger cuttings are dis-
at a full-scale RJD experimental facility. A set of pressure-loss posed of in the bottomhole of the vertical well (Fig. 1). According
models and a lateral-extending-force model of an RJD system to our field experience, the bottomhole space of the vertical well is
were then established, and a method to optimize the hydraulics of sufficient to contain the cuttings of an RJD operation.
RJD was proposed. Finally, a field case study with six radial later- In conventional drilling, hydraulics design aims to maximize
als in four different layers was conducted by use of our models the ROP under the limitations of economy, well structure, the
and optimization method. operating characteristics of the equipment and pipe, borehole sta-
The hydraulics calculations corresponded well with the field bility, and cutting-removal efficiency (Swanson et al. 1994). In
data. The model error was within 8%. The pressure loss of the RJD, the penetration process relies on the backward self-driving
high-pressure hose and jet bit represents a large proportion of the effect and forward high-pressure-jet rock-breaking effect of the
RJD-system pressure loss (41.2 and 55.8%, respectively). Accord- jet bit. The hydraulics design should satisfy the requirements of
ing to the operation profile, the calculated pump pressure will equipment-safety working pressure and lateral-extending capacity
help the field engineer to estimate the working status of downhole (Zhongwei et al. 2013).
tools. The results show that the pump flow rate should be opti- Previous works have focused on the equipment, procedure,
mized for different well configurations. The optimum flow-rate and stimulation effect of RJD by use of case studies from the field
range was determined by the minimum lateral-extending force, conducted in a variety of countries. However, the literature does
minimum rock-breaking jet-bit-pressure drop, and minimum not contain detailed operating hydraulics, such as the pump-flow
equipment-safety working pressure. To maximize the rate of pen- rate and pump pressure, or calculation and optimization methods
etration (ROP), the largest flow rate within that interval was have not been published. Ma et al. (2012) established the pres-
selected as the optimum flow rate. A flow rate of 57.24 L/min was sure-loss model of the circulation system of an RJD system. How-
optimal for the case well. ever, the pressure loss of the jet bit was not provided. Buset et al.
The present work provides a viable and detailed hydraulics- (2001) tested the rock-breaking ability and ejector force of a jet
calculation model and design method for RJD, and may be used bit. However, their test flow rate and ejector force were not suffi-
for both short-term troubleshooting and long-term operation plan- ciently high to be applied for field operation. Guo et al. (2009)
ning. It can serve as a guide for the development of safer and established a jet-bit-ejector-force analytical model for a jet bit,
more-effective RJD practices. but this model was not experimentally validated. Gang et al.
(2013) conducted RJD experiments and field tests. The results of
Introduction these experiments can provide a theoretical basis for hydraulics
calculations. Li et al. (2015) developed an analytical model to cal-
RJD, which uses a high-pressure water jet, aims to drill numerous
culate the jet-bit-ejector force, which was validated by use of both
radial laterals in one layer or multiple layers from the main wellbore
experiments and computational-fluid-dynamics simulations. How-
(Dickinson and Dickinson 1985; Dickinson et al. 1989). RJD is an
ever, some model coefficients can only be determined experimen-
efficient method to exploit low-permeability, shallow, marginal,
tally, which is inconvenient for practical application.
and CBM reservoirs at low cost (Li et al. 2000; Kamel et al. 2013;
The present work aims to provide a set of hydraulics-calcula-
Liang et al. 2011). Over the past 30 years, RJD services have been
tion models and an optimization method. Models that correlate
used in many countries, such as the US, Canada, China, and Argen-
the pressure-loss models and lateral-extending force were estab-
tina, yielding significant improvements in productivity (Dickinson
lished on the basis of RJD experiments. A method to optimize the
et al. 1993; Bruni et al. 2007; Cirigliano and Talavera Blacutt 2007;
design of hydraulics was then established on the basis of the RJD
Liang et al. 2011; Salem Ragab and Kamel 2013).
characteristics. Finally, an RJD field case study with six radial lat-
The RJD system (Fig. 1) differs significantly from conven-
erals in four different layers in Well A was conducted by use of
tional drilling. RJD uses a high-pressure water jet to drill into the
our models and optimization method.
rock formation by means of a jet bit. The RJD operation first
anchors the deflector at the target formation by use of coiled tub-
ing (CT). Then, a special casing cutter runs into the hole to mill Experimental Studies
the casing and cement. After the casing and cement are pene-
A full-scale RJD experimental system was established to test the
trated, the high-pressure hose and jet bit run into the hole until
RJD equipment for Well A. This system can measure the hydraul-
they reach the formation, and drilling fluid is pumped through the
ics, such as the pressure loss, ejector force, and extending resist-
ance of the RJD system. The RJD tools and fluids prepared for the
Copyright V
C 2016 Society of Petroleum Engineers experiment were the same as those used for Well A. A regression
Original SPE manuscript received for review 18 January 2015. Revised manuscript received
analysis of these hydraulic measurements yielded several correla-
for review 5 November 2015. Paper (SPE 179729) peer approved 11 December 2015. tion coefficients for the hydraulics-calculations models.

March 2016 SPE Drilling & Completion 71

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 72 Total Pages: 11

CT unit

High-pressure hose Forward water jet

Water
Mud pit tank
Pump

Working string
CT
Casing
High-pressure
hose Deflector

Anchor
Jetting bit
Backward water jet
Cuttings

Fig. 1—Schematic of the RJD system.

steel pipe. We fixed a button-type force sensor to the test plat-

form, used a steel-hose gripper to grip the high-pressure hose, and
placed the gripper onto the force sensor to measure the ejector
force. Thereafter, fluid was pumped at different flow rates and
ejector-force data were recorded. In addition, the test was
repeated with the jet bit removed under the same conditions to
calculate the pressure loss of the jet bit and high-pressure hose.
Steady flow was maintained for 3 minutes in each test run.

Extending-Resistance-Test Procedure. As shown in Fig. 3, we

fixed the deflector and force sensor to the test platform. We con-
nected one side of the hose to the pump, and the other side was
passed through the deflector and connected to the force sensor.
Thereafter, fluid was pumped at different flow rates. When the
flow stabilized, the moving platform began to move at a speed of
Fig. 2—Jet-bit drilling and jet-erosion holes. 0.0055 m/s, during which time the pressures and resistance data
were recorded. Steady flow was maintained for 3 minutes in each
The RJD experimental system (Fig. 3) consisted of an ejector- test run. The peak frictional resistance was recorded as the deflec-
force-test module, an extending-resistance-test module, fluid-mix- tor resistance in each test run.
ing and -pumping equipment, and a data-acquisition system. The
ejector-force-test module included a high-pressure hose, jet bit, Hydraulics-Calculation Model and
and button-type force sensor. The extending-resistance-test module Optimization-Design Method
included a high-pressure hose, deflector, horizontal moving plat- Pressure Loss Models. The drilling fluid for RJD was water, the
form, and force sensor. The fluids were prepared and stored in a flow rate was low, and the CT and high-pressure-hose dimensions
water tank. The pumping equipment consisted of a triplex pump. A were relatively small compared with those used for conventional
number of pressure sensors and valves were used to measure and drilling. The pressure loss models are used in engineering applica-
control the flowline and pressure of the test modules. The force tions, we simplified the models based on the following assumptions
sensors were used to measure the ejector force and extending re- (a detailed validation of these assumptions is given in Appendix A):
sistance. The parameters of the jet bit, high-pressure hose, and de- • Assumption 1: On the basis of our calculations, the effect of
flector (Fig. 4) are shown in Table 1. The tested fluid was water. water compressibility on pressure-loss calculation was mini-
mal. To simplify the calculation, we assumed that the dril-
Ejector-Force-Test Procedure. Li et al. (2015) proposed an ling fluid was incompressible, the fluid density was 1000 kg/
ejector-force-measurement experiment scheme. We followed this m3, and the fluid viscosity was 1.005 mPas.
procedure and duplicated their experiment. As shown in Fig. 3, • Assumption 2: For RJD, we often use 11=2- or 1-in. CT to
the high-pressure hose was connected to a jet bit and placed in a satisfy the high-pressure condition in the field run. On the

Valve Valve
Water
inlet
Water
tank High-pressure
Pressure hose Steel-hose gripper
Valve
sensor Button-type
F force sensor
High-pressure Force
Deflector
hose sensor
Pump F Horizontal moving
Steel Pipe
platform

Jet bit

Fig. 3—Schematic of the full-scale RJD experimental system.

72 March 2016 SPE Drilling & Completion

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 73 Total Pages: 11

DpcT ¼
0:218q1:05 LcT
(  Q2:05 ;
D 3:9625ld
d4:95 l0:05 D0:1 log 
3:7065d qQ
 1:1098  0:8981 !#)2
1 D ld
log þ 4:7096    ð3Þ
2:8570 d qQ

DpsT ¼
0:2026qLsT
(  Q2 ;
D 3:9625ld
Fig. 4—Jet bit, deflector, and high-pressure hose in Well A. d5 log 
3:7065d qQ
 1:1098  0:8981 !#)2
basis of our calculations, the fluid flow of the RJD system is 1 D ld
log þ 4:7096     ð4Þ
always turbulent in the field run. To simplify the calculation, 2:8570 d qQ
we assumed that the fluid flow in the hose and CT was
always turbulent flow.
• Assumption 3: For RJD, the pump-flow rate is lower than where LST is the length of the CT straight section (m); LCT is the
that used for conventional drilling. On the basis of our calcu- length of the CT spiral section (m); d is the CT inner diameter
lations for Well A, the pressure loss in the annulus is only (m); D is the diameter of CT reel drums (m); D is the CT wall
0.087 MPa. To simplify the calculation, we assumed that the roughness (m); l is the drilling-fluid dynamic viscosity (Pas); Q
annular-pressure loss was negligible. is the pump-flow rate (m3/s); and Lh is the length of the high-pres-
The pump pressure for RJD can be calculated as follows: sure hose (m).
Pressure Loss in the High-Pressure Hose. The high-pressure
hose is made from a composite material and consisted of polyur-
DpL ¼ DpcT þ DpsT þ Dph þ Dpb ; . . . . . . . . . . . . . . . ð1Þ ethane outer layer, braided-steel-wire middle layer, and polyamide
inner layer. Polyamide is a plastic used in engineering applications.
where DpL is the pump pressure (MPa), DpCT is the pressure loss The absolute roughness of plastic is much lower than that of com-
in spiral section of CT (Pa), DpST is the pressure loss in the mercial steel (Eugene et al. 2006). The Blasius (1913) friction fac-
straight section of the coiled tubing (Pa), Dph is the pressure loss tor is a simple, commonly used equation for smooth pipes. We
in the high-pressure hose (Pa), and Dpb is the pressure drop at the assumed that a high-pressure hose is a smooth pipe. To simplify the
jet bit (MPa). calculation, we derived our model from the Blasius correlation
Pressure Loss in CT. According to a steady-state mechanical- (1913) by introducing a frictional-correction factor a.
energy balance, the general CT pressure-loss formula can be writ- Blasius (1913) developed the Fanning friction factor in turbu-
ten as follows: lent smooth-pipe flow:
0:079
2f qLv2 f ¼ ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð5Þ
.
Dpi ¼ ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . ð2Þ NRe 0:25
di
where NRe is the Reynolds number of the fluid, defined as
where f is the Fanning friction factor (dimensionless), q is the
fluid density (kg/m3), v is the fluid mean velocity (m/s), di is the qvd 4qQ
NRe ¼ ¼ : . . . . . . . . . . . . . . . . . . . . . . . . . ð6Þ
pipe inner diameter (m), and L is the pipe length (m). l pld
On the basis of experimental data, Zhou and Shah (2004) and
Shah and Ahmed Kamel (2010) proposed the Fanning friction fac- Substituting Eqs. 5 and 6 into Eq. 2 yields the following pres-
tor of Newtonian turbulent flow in CT. Substituting the coeffi- sure-loss model in the smooth pipe (Dpss):
cients fCT and fST into Eq. 2, the CT pressure loss of coiled DpCT L
and straight DpST sections can be expressed as follows: Dpss ¼ 0:2399q0:75 l0:25 Q1:75 : . . . . . . . . . . . . . . ð7Þ
d4:75

Table 1—Parameters of the RJD tools for Well A. nf is the number of forward nozzles of the jet bit (pieces). nc is the number of forward-
central nozzles of the jet bit (pieces). nb is the number of backward nozzles of the jet bit (pieces). df is the forward-nozzle diameter of the jet
bit (m). dc is the forward-central-nozzle diameter of the jet bit (m). db is the backward-nozzle diameter of the jet bit (m). hf is the diffusion
angle of the forward nozzle (degrees), and hb is the diffusion angle of the backward nozzle (degrees).

March 2016 SPE Drilling & Completion 73

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 74 Total Pages: 11

High-Pressure-Hose-Pressure Loss (MPa)

8 32
Data point of water Data points
7 Fitting curve R 2 = 0.99 28

Jet-Bit-Pressure Drop (MPa)

Fitting curve R 2 = 0.99
Data point of water plus 0.07 vol% drag reducer 24
6
Fitting curve R 2 = 0.99
20
5
16
4
12
3
8
2
4
1
0
0 10 20 30 40 50 60
25 30 35 40 45 50 55 Flow Rate (L/min)
Flow Rate (L/min)
Fig. 6—Pressure drop of the jet bit.
Fig. 5—Pressure loss of the high-pressure hose.

Lateral-Extending-Force Model. Under actual drilling condi-

On the basis of Eq. 7, we can introduce a frictional-correction tions, the high-pressure hose is in contact with the wellbore and
coefficient, a, such that the pressure-loss model of the high-pres- deflector. According to the force analysis in Fig. 7, the lateral-
sure hose can be written as follows: extending force includes the jet-bit-ejector force Fbit, the wellbore
resistance Fw, and the deflector resistance Ff. Thus, the lateral-
Lh 1:75
Dph ¼ aq0:75 l0:25 4:75
Q ; .................. ð8Þ horizontal-extending force can be calculated as
dhi
F ¼ Fbit  Fr ¼ Fbit  Fw  Ff ; . . . . . . . . . . . . . . . ð10Þ
where dhi is the inner diameter of the high-pressure hose (m).
The length of the high-pressure hose (Table 1) used in the where Fr is the lateral-extending resistance (N). The lateral can
experiment was 20 m. For a specific type of hose and a known test extend when the lateral-extending force, F, is greater than zero.
fluid, the coefficient a is the only unknown term in Eq. 8. This Ejector Force of the Jet Bit. The ejector force of the jet bit is
coefficient can be determined by substituting the experimental mainly generated by the momentum difference between the for-
data for the pressure loss and pump-flow rate into Eq. 8. Thus, we ward and backward jets. Li et al. (2015) proposed a theoretical
can use the model to calculate the pressure loss of the hose. The ejector-force model for a multinozzle jet bit (Fig. 8). The model
experimental results for our high-pressure hose are shown in was correlated and validated by both experimental data and nu-
Fig. 5. In addition, the drag-reduction efficiency of the drag re- merical-simulation results. The ejector force can be calculated as
ducer in our experiment was 41.67%.
q
Jet-Bit-Pressure Loss. Chen et al. (2006) proposed the fol- Fbit ¼ mQ2 ; . . . . . . . . . . . . . . . . . . . . . . . . . . . ð11Þ
lowing pressure-loss model for a multinozzle jet bit: A0
where A0 is the cross-sectional area of the jet-bit inlet (mm2), Q is
0:5136qQ2
Dpb ¼ ; . . . . . . . . . . . . . . . . . . . . . . . . . . ð9Þ the pump-flow rate (L/s), and the coefficient m is defined as the
C2 A2 self-propelled-ability coefficient:
where C is the jet-bit-discharge coefficient (dimensionless), A is  2
0:081A20 S k
the total area of nozzles (mm2), and Q is the pump-flow rate (L/s). m¼ S þ 0:5ð1 þ SÞ  D4c S
C2 dne
4 ðnf þ 1Þðk þ 1Þ
For a specific type of jet bit and a known test fluid, the coefficient  2
C is the only unknown term in Eq. 9. This coefficient can be 1
determined by substituting the experimental data for the pressure þ nb D2b coshb :            ð12Þ
nb ðk þ 1Þ
loss and pump-flow rate into Eq. 9. Thus, we can use the model to
calculate the jet-bit-pressure drop. The experimental results for A detailed calculation of coefficient m is presented in Appen-
the jet bit of Well A (Table 1) are shown in Fig. 6. dix B.
D
L1

Hole
deflector
α

Extending force
F = Fbit – Fw – Ff
L2

90
High-pressure
Ff hose

Fw Fbit R

Deflector Wellbore Ejector

β

resistance resistance force

Fig. 7—Force analysis of RJD process.

74 March 2016 SPE Drilling & Completion

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 75 Total Pages: 11

vb db Backward orifice

Forward orifice

vf
df
d0
Center orifice
θf vc
V0 dc
A0

vf
df
θb

vb db

Fig. 8—Flow field of the self-driving multinozzle jet bit (Li et al. 2015).

For a specific type of jet bit and a known test fluid, the coeffi- where a and b are the correlation coefficients. The coefficients are
cient m is the only unknown term in Eq. 11. Thus, we can calcu- related to deflector-track structure, hose materials, and hose size.
late the jet-bit-ejector force by use of the model. We measured the Under our experimental conditions (Table 1), a and b are equal to
flow ratio (k) and the jet-bit-discharge coefficient (C) on the basis 0.88 and 6.06, respectively. Thus, we can calculate the deflector
of our experimental conditions (Table 1). The values of k and C resistance by use of the model.
are 0.43 and 0.56, respectively. Substituting k and C into Eq. 12 The submerged weight of the hose (Table 1) in water, qbh, had
yields a value for m of 6.46. The experiment results are shown in an experimental value of 0.96 N/m. The sliding-friction coeffi-
Fig. 9. To improve this model, we proposed an equation (Appen- cient, ls, between the high-pressure hose and borehole is related
dix B) to estimate the coefficient k instead of the use of an experi- to the rock-surface roughness and cleaning condition. In this pa-
mentally determined correlation. per, ls was assumed to be the sliding-friction coefficient between
According to our experiment results, ejector force is mainly polyurethane and concrete, the experimental value of which is
determined by pump-flow rate and jet-bit structure. It should be 0.35 (Ohnson and Storey 2000).
stressed that the jet bit has some surrounding effects in the actual
drilling environment, such as the borehole diameter, underpres- Hydraulics-Optimization-Design Method. According to the
sure-force effect (Buset et al. 2001), and standoff distance (Gang characteristics of RJD, the jet-drilling process is accomplished by
et al. 2013). According to our computational-fluid-dynamics sim- the backward jet self-driving effect and the forward high-pres-
ulation, the surrounding flow field of jet bit is complex, and it is sure-jet rock-breaking effect (Fig. 2).
difficult to develop a theoretical model to describe those sur- To drive the jet bit to persistently move forward, the minimum
rounding factors. Therefore, we simplified the model by neglect- flow rate should satisfy the two effects simultaneously: the ejector
ing those surrounding effects. The influence of these surrounding force, Fbit, should be greater than the lateral-extending resistance,
effects on ejector force needs further research. Fr, and the pressure drop of the jet bit, Dpb, should be greater than
Extending Resistance. The extending resistance is the sum of the threshold rock-breaking pressure, Pmin. For the RJD operation,
the wellbore resistance and deflector resistance. We estimated we conservatively estimated that the rock-breaking jet-bit-pres-
the deflector resistance by means of experimental regression sure drop should be greater than 15 MPa (Appendix D). The max-
(Fig. 10). A detailed correlation of the deflector resistance Ff is imum flow rate should ensure that the equipment-working
presented in Appendix C. pressure is lower than its yield pressure. Thus, we could determine
The extending-resistance model can be expressed as follows: the optimum flow-rate range. The largest flow rate in that range
was selected as the optimum flow rate to maximize the ROP. In
summary, the hydraulics-optimization-design method was pro-
Fr ¼ Fw þ Ff ¼ lqbh Lh þ aðDph þ Dpb Þ þ b; . . . . . . ð13Þ posed as shown in Fig. 11.

80 50
Model estimation Experimental data
70 Experimental value Fitting curve R 2 = 0.99
Drilling Resistance (N)

40
Ejector Force (N)

60
30
50
20
40

30 10

20 0
30 35 40 45 50 55 0 5 10 15 20 25 30 35 40
Flow Rate (L/min) Inner Pressure of Hose (MPa)

Fig. 9—Ejector force of the jet bit. Fig. 10—Deflector resistance with a pressurized hose.

March 2016 SPE Drilling & Completion 75

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 76 Total Pages: 11

Hydraulics models Optimization criterion

Circulation-pressure Maximum flow rate

loss
Equipment pressure
Jet-bit- ≤ Limit pressure? Maximize
Initial pressure drop ROP Optimum
flow rate Ejector force ≥ Resistance? flow rate
Jet-bit-
Jet bit pressure drop ≥ rock-
ejector force
breaking pressure?

Extending resistance
Minimum flow rate

Fig. 11—Hydraulics-design flow chart.

Hydraulics-Design Procedure and to 20, 40, and 100 m. The results are shown in Fig. 14. To drive
Field Application the jet bit to persistently move forward, the minimum flow rate
The reservoir and well parameters of Well A are shown in should ensure that the ejector force, Fbit, is greater than the lat-
Table 2. The preliminary well design consisted of six 100-m-long eral-extending resistance, Fr, and the pressure drop of the jet bit,
radial laterals drilled in four layers. The bottomhole assembly Dpb, should exceed 15 MPa. According to Fig. 12, the minimum
included a casing-milling assembly and jet-drilling assembly. The rock-breaking flow rate is 46.14 L/min (dashed pink line).
Well A tools are specified in Tables 1 and 3. According to Fig. 14, for a 100-m-long lateral, the minimum lat-
eral-extending-flow rate is 54.74 L/min (farthest-right black dot
line). To satisfy both conditions simultaneously, 54.74 L/min was
Hydraulics Optimization. Taking the No. 3 lateral of Well A as selected as the minimum flow rate at the No. 3 radial wellbore in
an example, the optimum flow rate was obtained and analyzed by Well A.
use of the proposed hydraulics-calculation models (Eqs. 1 through Optimum Flow Rate. As shown in Fig. 12, the pressure loss
13) and the design method (Fig. 11). of the CT, 1.16 MPa, comprises only 3% of the pressure loss in
According to the well design, the depth of the No. 3 lateral is the system, which was 41.01 MPa in Well A. The pressure-loss
2304.600 m. The designed lateral length is 100 m. Because the gradient of CT in Well A was 0.31 MPa/1000 m. The maximum
total length of the CT is known to be 4000 m, we were able to depth interval of the laterals was 18.30 m; thus, the depth of the
determine the lengths of the spiral-section tubing, the straight-sec- wellbore minimally influences the optimal flow rate in this case.
tion tubing, and high-pressure hose, which are 1695.40, 2304.60, In summary, in Well A, the optimum flow-rate range is
and 100 m, respectively. 54.74–57.24 L/min in Well A, and the optimal flow rate to maxi-
Maximum Flow Rate. The pressure loss and inner pressure of mize the ROP was determined to be 57.24 L/min.
the RJD circulation system were calculated by use of Eqs. 3 and 4
and Eqs. 8 and 9, respectively. The results are shown in Figs. 12
and 13. According to the results, the pressure loss of the high- Field Results. Six laterals were drilled in Well A by use of the
pressure hose and jet bit comprise the majority of the pressure hydraulics-calculation results and optimal flow rate presented in
loss of the entire system (41.2 and 55.8%, respectively). As shown this study. The specific details of the field results are shown in
in Fig. 13, the CT inner yield pressure was 83 MPa (horizontal Table 4. The RJD operating profile in the No. 4 lateral is shown
red dot line), and the high-pressure-hose inner yield pressure was in Fig. 15. The hydraulics-calculation results closely approxi-
40 MPa (horizontal blue dot line). To avoid CT or hose failure, mated the field data, with a model error of 8%.
the inner pressure should always be lower than the inner yield In addition, the downhole working status of the tools can be
pressure. On the basis of Fig. 13, we could easily select 57.24 L/ determined on the basis of the operation profile. Specifically, a
min (vertical blue dot line) as the maximum flow rate. Thus, the dramatic pressure drop would be observed if the high-pressure
shaded area of Fig. 13 is the safety-operation zone of the lateral. hose suddenly cracked under the harsh bottomhole environment.
Minimum Flow Rate. The ejector force and lateral-extending The RJD ROP is manually controlled by the run-in speed of the
resistance were calculated by use of Eqs. 11 and 12. The extend- CT. The jet bit should first penetrate the front formation rock and
ing resistance was estimated when the radial wellbore was drilled then slowly run the CT into the hole. Because the formation is

Table 2—Characteristics of Well A.

76 March 2016 SPE Drilling & Completion

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 77 Total Pages: 11

Table 3—Bottomhole assembly in Well A.

90 150
CT 140 High-pressure hose Pyield = 40 MPa
80 130
Straight tubing 120 CT Pyield = 83 MPa
Pressure Loss (MPa)

70
110

Inner Pressure (MPa)

High-pressure hose
60 100
Jet bit 90
50 80
70
40 60
50
30 40
20 30
20
10 10
0
0 –10 Safety-operation zone
–20
0 10 20 30 40 50 60 70 80 90 100 110 0 10 20 30 40 50 60 70 80 90 100 110
Flow Rate (L/min) Flow Rate (L/min)

Fig. 12—Pressure loss/flow rate of tools in No. 3 lateral. Fig. 13—Hose and CT inner pressure/flow rate in No. 3 lateral.

field operator should immediately run out the CT a small distance

200 Ejector force and decrease the pump-flow rate. Thereafter, the field operator
Drilling resistance Lwellbore = 100 m
175 should repeatedly increase the pump-flow rate and try to run in
Drilling resistance Lwellbore = 60 m the CT. As shown in Fig. 15, four sharp pressure increases
150 Drilling resistance Lwellbore = 20 m occurred. In some cases, the threshold rock-breaking pressure of
125 hard rock was higher than the jet-bit-pressure drop; thus, the for-
Force (N)

mation rock was not sufficiently soft for penetration. At the 54-m
100
point of the No. 4 lateral, the high-pressure hose burst after
75 two attempts.
50
Conclusions
25
1. A set of RJD hydraulics-calculation models were established
0 on the basis of full-scale RJD experiments. An experimental
–25 correlation model was established for predicting the high-
10 20 30 40 50 60 70 80 90 pressure-hose pressure loss. The ejector-force model was
improved by proposing an estimation equation for an experi-
Flow Rate (L/min)
mentally determined coefficient k. The deflector-resistance
Fig. 14—Extending resistance and ejector force/flow rate in No.
model was established by use of a regression method of the
3 lateral. linear relation between the deflector resistance and high-pres-
sure-hose inner pressure.
2. According to the characteristics of RJD, an RJD hydraulics-
heterogeneous, it contains areas of hard rock. When the threshold optimization-design method was proposed. Specifically, the
rock-breaking pressure of the hard rock is lower than the jet-bit- optimum flow-rate range was determined on the basis of mini-
pressure drop, the hard rock prevents the jet bit from rapidly pene- mum lateral-extending force, minimum rock-breaking pressure
trating the rock while the CT is run in. In this situation, the high- drop, and minimum equipment-safety working pressure. The
pressure hose bends considerably, resulting in a sudden increase largest flow rate in that range was then selected as the optimum
in pressure. To prevent the high-pressure hose from bursting, the flow rate to maximize the ROP.

Table 4—Operation parameters of Well A.

March 2016 SPE Drilling & Completion 77

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 78 Total Pages: 11

80 70 D ¼ wall roughness of CT, m

70
Dpb ¼ pressure drop at jet bit, MPa
60 DpcT ¼ pressure loss in spiral section of CT, Pa

Pump-Flow Rate (L/min)

Pump Pressure (MPa)
60 DpL ¼ pump pressure, MPa
50
50 DpsT ¼ pressure loss in straight section of CT, Pa
40 Dpss ¼ pressure loss in smooth pipe, Pa
40 Dph ¼ pressure loss in high-pressure hose, Pa
30 hf ¼ diffusion angle of the forward nozzle, rad
30
20 hb ¼ diffusion angle of the backward nozzle, rad
20 Pump pressure l ¼ dynamic viscosity of fluid, Pas
10 Calculated pump pressure 10 ls ¼ sliding friction coefficient, dimensionless
Pump-flow rate q ¼ density of fluid, kg/m3
0 0 v ¼ backward flowrate factor, dimensionless
0 30 60 90 120 150 180 210 240 270
Operation Time (minutes)
Acknowledgments
Fig. 15—Comparison of the field operation profile and model The authors express their appreciation to the Major International
prediction of the No. 4 lateral. (Regional) Joint Research Program of National Natural Science
Foundation of China (No. 51210006), Major National Science
and Technology Program of China (No. 2011ZX05009-005), and
3. An RJD case in Well A consisting of six radial laterals in four Science Foundation of China University of Petroleum, Beijing,
different layers was drilled by use of the hydraulics-calculation (No. 2462013BJRC002) for their support of this work. In addi-
models and optimization-design method. The hydraulics-calcu- tion, the authors would like to thank Mao Sheng, Xianzhi Song,
lation model corresponded well with the field data, with a Huanpeng Chi, and Xiaojiang Li (China University of Petroleum)
model error of less than 8%. The optimal pump-flow rate was for revising the article and providing helpful suggestions.
57.24 L/min in Well A.
4. This work may be used for the diagnosis of short-term down- References
hole problems, long-term well design, and operating planning. Blasius, H. 1913. Das Ahnlichkeitsgesets Bei Reibungsvorganen in Flus-
The results may serve as a guide for the development of safer sigkeiten. Forsch Arb Ing-Wes.
and more-effective RJD practices. Bruni, M. A., Biasotti, J. H. and Salomone, G. D. 2007. Radial Drilling in
Argentina. Presented at the Latin American & Caribbean Petroleum
Engineering Conference, Buenos Aires, 15–18 April. SPE-107382-
Nomenclature MS. http://dx.doi.org/10.2118/107382-MS.
a ¼ correlation coefficient of deflector resistance, dimensionless Buset, P., Riiber, M. and Eek, A. 2001. Jet Drilling Tool: Cost-Effective
A0 ¼ cross-sectional area of the jet bit inlet, mm2 Lateral Drilling Technology for Enhanced Oil Recovery. Presented at
b ¼ correlation coefficient of deflector resistance, dimensionless the SPE/ICoTA Coiled Tubing Roundtable, Houston, 7–8 March.
C ¼ jet bit discharge coefficient, dimensionless SPE-68504-MS. http://dx.doi.org/10.2118/68504-MS.
d ¼ CT inner diameter, m Chang, Z. X. 2006. The Theory of Non-Isotropic Breaking Rock by Water
db ¼ backward nozzle diameter of the jet bit, mm Jets and Its Application (in Chinese). PhD dissertation, Taiyuan Uni-
dc ¼ forward central nozzle diameter of the jet bit, mm versity of Technology, Taiyuan, China.
df ¼ forward nozzle diameter of the jet bit, mm Chen, T. G. Guan, Z. C. 2006. Theory and Technology of Drilling Engineer-
dhi ¼ inner diameter of high-pressure hose, m ing (in Chinese). Dongying, China: China University of Petroleum Press.
di ¼ pipe inner diameter, m Chi, H. P., Li, G. S., Liao, H. L., et al. 2013. Experimental Research on Opti-
dne ¼ equivalent nozzle diameter, mm mum of Jet bit in Radial Drilling by Hydra-Jet. Fluid Mach. 41 (2): 1–6.
d0 ¼ jet bit inlet diameter, mm Cinelli, S. D. and Kamel, A. H. 2013. Novel Technique to Drill Horizontal
D ¼ drum diameter of CT, m Laterals Revitalizes Aging Field. Presented at the SPE/IADC Drilling
Dc ¼ diameter ratio of jet bit and central nozzle, m Conference and Exhibition, Amsterdam, 5–7 March. SPE-163405-MS.
Db ¼ diameter ratio of jet bit and backward nozzle, m http://dx.doi.org/10.2118/163405-MS.
f ¼ fanning friction factor, dimensionless Cirigliano, R. A. and Talavera Blacutt, J. F. 2007. First Experience in the
F ¼ lateral extending force, N Application of Radial Perforation Technology in Deep Wells. Pre-
Fb ¼ ejector force of jet bit, N sented at the Latin American & Caribbean Petroleum Engineering
Ff ¼ deflector resistance, N Conference, Buenos Aires, 15–18 April. SPE-107182-MS. http://
Fr ¼ lateral extending resistance, N dx.doi.org/10.2118/107182-MS.
Fw ¼ wellbore resistance, N Dickinson, W. and Dickinson, R. W. 1985. Horizontal Radial Drilling Sys-
k ¼ flow rate ratio between the forward and backward noz- tem. Presented at the SPE California Regional Meeting, Bakersfield,
zles, dimensionless California, 27–29 March. SPE-13949-MS. http://dx.doi.org/10.2118/
L ¼ pipe length, m 13949-MS.
LcT ¼ length of the CT spiral section, m Dickinson, W., Anderson, R. R. and Dickinson, R. W. 1989. The Ultra-
Lh ¼ length of the high-pressure hose, m short-Radius Radial System. SPE Drill Eng 4 (3): 247–254. SPE-
LsT ¼ length of the CT straight section, m 14804-PA. http://dx.doi.org/10.2118/14804-PA.
m ¼ self-propelled coefficient of the nozzle, dimensionless Dickinson, W., Dystra, H., Nordlund, R. et al. 1993. Coiled-Tubing Radials
nb ¼ the number of backward nozzles of the jet bit, pieces Placed by Water-Jet Drilling: Field Results, Theory, and Practice. Pre-
nc ¼ the number of forward central nozzles of the jet bit, sented at SPE Annual Technical Conference and Exhibition, Houston,
pieces 3–6 October. SPE-26348-MS. http://dx.doi.org/10.2118/26348-MS.
nf ¼ the number of forward nozzles of the jet bit, pieces Eugene, A., Theodore, B. and Sadegh, A. 2006. Marks’ Standard Hand-
NRe ¼ Reynolds number, dimensionless book for Mechanical Engineers, eleventh edition. New York City:
qbh ¼ submerged weight of the high-pressure hose in water per McGraw-Hill.
meter, N/m Gang, B., Gensheng, L., Zhonghou, S. et al. 2013. Experimental Research
Q ¼ pump flow rate, m3/s on the Technology of Hydra-Jet Sidetracking of Radial Micro-bore-
S ¼ area coefficient of nozzle front face, dimensionless hole. J. Eng. Sci. Tech. Rev. 6 (5): 137–142.
v ¼ fluid mean velocity, m/s Guo, R. C., Li, G., Huang Z. et al. 2009. Theoretical and Experimental
a ¼ frictional correction coefficient, dimensionless Study of the Pulling Force of Jet Bits in Radial Drilling Technology.

78 March 2016 SPE Drilling & Completion

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 79 Total Pages: 11

Table A-1—Pump pressure when considering water compressibility.

Petrol. Sci. 6 (4): 395–399. http://dx.doi.org/10.1007/s12182-009- Swanson, B. W., Thorogood, J. L. and Gardner, A. 1994. The Design and
0060-6. Field Implementation of a Drilling Hydraulics Application for Drilling
Harvey, A. H. 1998. Thermodynamic Properties of Water: Tabulation Optimization. Presented at European Petroleum Computer Conference,
from the IAPWS Formulation 1995 for the Thermodynamic Properties Aberdeen, 15–17 March. SPE-27548-MS. http://dx.doi.org/10.2118/
of Ordinary Water Substance for General and Scientific Use. Report, 27548-MS.
National Institute of Standards and Technology. Zhongwei, H., Gensheng, L., Zhijun, T. et al. 2013. Technology of Hydra-
Li, Q. 2008. Research on Drilling Long Hole in Soft Coal Seam by High Jet Sidetracking of Horizontal Micro-Radial Laterals. Petrol. Drill.
Pressure Water Jets (in Chinese). Master’s thesis, Chongqing Univer- Tech. 41 (4): 37–41. http://dx.doi.org/10.3969/j.issn.1001-0890.2013.
sity, Chongqingcity, China. 04.009.
Zhou, Y. and Shah, S. 2004. Fluid Flow in Coiled Tubing: A Literature
Li, J., Li, G., Huang, Z. et al. 2015. The Self-Propelled Force Model of a
Review and Experimental Investigation. J Can Pet Technol 43 (6):
Multi-Orifice Nozzle for Radial Jet Drilling. J. Nat. Gas Sci. Eng. 24
52–61. PETSOC-04-06-03. http://dx.doi.org/10.2118/04-06-03.
(May): 441–448. http://dx.doi.org/10.1016/j.jngse.2015.04.009.
Li, Y., Wang, C., Shi, L. et al. 2000. Application and Development of
Drilling and Completion of the Ultrashort-Radius Radial Well by High Appendix A—Validation of the Assumptions
Pressure Jet Flow Techniques. Presented at the International Oil and Assumption 1. The drilling fluid is an incompressible Newtonian
Gas Conference and Exhibition in China, Beijing, 7–10 November. fluid. For water compressibility, the variation of water density
SPE-64756-MS. http://dx.doi.org/10.2118/64756-MS. does not significantly affect the pressure-loss calculation.
Liang, Z., Ge, Y., Li, J. et al. 2011. Application of Hydraulic Injector Ra- We used Well A as an example. We evaluated the pump pres-
dial Horizontal Well Technology in CBM Exploration (in Chinese). sure for different field conditions and water densities on the basis
J. Liaoning Tech. Univ. (Nat. Sci.) 30 (3): 349–352. of water-properties data from Harvey (1998) and the models pre-
sented in this study. The results are shown in Table A-1. We eas-
Ma, D. J., Li, G. S., Huang, Z. et al. 2012. A Model of Calculating The
ily calculated the pressure-loss differences as less than 1 MPa
Circulating Pressure Loss in Coiled Tubing Ultra-short Radius Radial
when considering water compressibility. To simplify the calcula-
Jet Drilling. Petrol. Explor. Dev. 39 (4): 494–499. http://dx.doi.org/
tion, we assumed the drilling fluid to be incompressible.
10.1016/S1876-3804(12)60072-X.
Ohnson, A. A. J. and Storey, R. J. 2000. Frictional Behaviour of a Polyur- Assumption 2. The fluid flow in the hose and CT is turbulent.
ethane Rubber Sliding on Concrete: Analysis of a Slip and Fall Acci- For RJD, we often use 11=2- or 1-in. CT to satisfy the high working
dent. Technol. Law Insur. 5 (3–4): 111–116. http://dx.doi.org/10.1080/ pressure in the field run. On the basis of the dimension of the tools
13599370010060140. in Well A (Tables 1 and 4), we calculated the Reynolds number
Salem Ragab, A. M. and Kamel, A. M. 2013. Radial Drilling Technique of the fluid in the CT, high-pressure hose, and annulus. As shown
for Improving Well Productivity in Petrobel-Egypt. Presented at the at in Fig. A-1, when the flow rate exceeds 10 L/s, all the fluid flow
the North Africa Technical Conference and Exhibition, Cairo, 15–17 in the RJD system is turbulent. To simplify the calculation, we
April. SPE-164773-MS. http://dx.doi.org/10.2118/164773-MS. assume that the fluid flow in the hose and CT is always turbulent.
Shah, S. N. and Ahmed Kamel, A. H. 2010. Investigation of Flow Behav-
ior of Slickwater in Large Straight and Coiled Tubing. SPE Prod & Assumption 3. The annular-pressure loss is negligible. The flow
Oper 25 (1): 70–79. SPE-118949-PA. http://dx.doi.org/10.2118/ rate of RJD is lower than that used in conventional drilling. On
118949-PA. the basis of the dimensions of tools in Well A, the pressure loss in
the annulus was only 0.087 MPa. We neglected the annular-pres-
sure loss to simplify the calculation.
100,000 RJD includes a lateral annulus and a casing annulus. Accord-
CT d = 0.031 m
High-pressure hose d = 0.01 m ing to Fig. A-1, the lateral-annular flow and casing-annular flow
Casing annular de = 0.048 m were considered to be turbulent. The annular-pressure loss can be
80,000 Lateral annular de = 0.016 m calculated by use of Eq. 2 and the hydraulic diameter, de:
Re = 4,000 turbulent flow

60,000
de ¼ d2  d1 ; . . . . . . . . . . . . . . . . . . . . . . . . . . . ðA-1Þ

where d2 is the inner diameter of the casing/lateral and d1 is the

Re

40,000 outer diameter of the tubing/high-pressure hose. In this study, the

lateral-annulus roughness was assumed to be the concrete-pipe
roughness, which is approximately 0.0003 m. We used Well A as
20,000 an example. The calculation results are shown in Table A-2.

4000
0
0 10 20 30 40 50 Appendix B—Calculation of the
Flow Rate (L/min) Self-Propelled-Ability Coefficient m
On the basis of the Li et al. (2015) model, the self-propelled-abil-
Fig. A-1—Reynolds number of the RJD system. ity coefficient m is given as follows:

March 2016 SPE Drilling & Completion 79

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 80 Total Pages: 11

Table A-2—Pressure loss of the RJD annulus.

Table B-1—Flow ratio of the forward- and backward-nozzle (k) data.

 2 Appendix C—Calculation of the Deflector

0:081A20 S k
m¼ S þ 0:5ð1 þ SÞ  D4c S Resistance Ff
C2 dne
4 ðnf þ 1Þðk þ 1Þ
 2 As shown in Fig. 7, the force condition of the high-pressure hose is
1 complex when it passes through the deflector. It is difficult to de-
þ nb D2b coshb ;           ðB-1Þ
nb ðk þ 1Þ velop a theoretical model to estimate the deflector resistance, Fw.
According to our preliminary test and numerical-simulation study,
where Dc and Db are the diameter ratio of the jet bit and nozzle the deflector resistance is mainly influenced by the track structure
(dimensionless), which were defined as follows:
of the deflector (Fig. 7), as well as the inner pressure of the hose
d0 d0 and the mechanical properties of the hose. Specifically, the deflec-
Dc ¼ ; Db ¼ : ..................... ðB-2Þ
dc db tor resistance decreases with deflector-bend radius R, track diame-
ter D, and bend angle a, and increases with deflector-exiting angle
A0 is the cross-sectional area of the jet-bit inlet (mm2):
b, as well as the hose inner pressure, elastic modulus E, and rigid
p modulus G. Because of the simplification made for field applica-
A0 ¼ d02 : . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ðB-3Þ
4 tion, we only model the inner pressure of the hose to estimate the
dne is the equivalent nozzle diameter (mm), which can be deflector resistance under the given deflector, and the hose.
described as follows for a multinozzle jet bit: The results of our experiment (Fig. 9) indicated a linear rela-
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi tionship between the resistance and the inner pressure of the hose.
dne ¼ nf df2 þ nc dc2 þ nb db2 : . . . . . . . . . . . . . . . . . ðB-4Þ Therefore, we could easily identify a correlation between the hose
inner pressure and deflector resistance:
S is the area coefficient of the front face (dimensionless),
which can be calculated as follows: Ff ¼ apInner þ b: . . . . . . . . . . . . . . . . . . . . . . . . . ðC-1Þ
 
1 nf In the field case, the inner pressure of the high-pressure hose
S¼ þ . . . . . . . . . . . . . . . . . . . . . ðB-5Þ
D2c 2coshf D2c (DpInner) is related to the hose-pressure loss (Dph) and the jet-bit-
where k is the flow ratio of the forward and backward nozzles pressure drop (Dpb) because of the frictional pressure loss:
(dimensionless) and C is the jet-bit-discharge coefficient. k and C
are related to the nozzle structure and material roughness, which PInner ¼ Dph þ Dpb : . . . . . . . . . . . . . . . . . . . . . . . ðC-2Þ
can be determined by means of a jet-bit experiment (Li et al.
2015). According to the experimental results, the discharge coeffi- Substituting Eq. C-2 into Eq. C-1 yields
cient of the multinozzle jet bit, C, varies from 0.5 to 0.8.
The flow ratio of the forward and backward nozzles, k, is mainly Ff ¼ aðDph þ Dpb Þ þ b; . . . . . . . . . . . . . . . . . . . . ðC-3Þ
related to the nozzle-area ratio, which can be estimated as follows:
nf df2 þ dc2 where a and b are the correlation coefficients. The coefficients are
k¼ ; . . . . . . . . . . . . . . . . . . . . . . . . . . ðB-6Þ related to the deflector-track structure, hose materials, and hose
vnb db2
size. For a specific type of deflector and high-pressure hose, the
where v is the backward-flow-rate factor (dimensionless). On the coefficients a and b are the only unknown terms for Eq. C-3. On
basis of our experimental observation, the backward-nozzle-flow the basis of the experimental data, we could draw a linear-fitting
rate is lower than the forward-nozzle-flow rate when their nozzle line. Here, a and b are the slope and intercept of the line, respec-
total areas are equal. Therefore, we introduce v as the backward- tively. Thus, we can calculate the deflector-resistance force by use
flow-rate factor. As shown in Table B-1, the flow ratio k can be of the model. For our deflector and high-pressure hose (Table 1),
estimated accurately when the value of v is 0.9. the experimental results are shown in Fig. 10 and Table C-1. We

Table C-1—Deflector-resistance-experiment data.

80 March 2016 SPE Drilling & Completion

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001

 DC179729 DOI: 10.2118/179729-PA Date: 17-February-16 Stage: Page: 81 Total Pages: 11

Table D-1—Threshold rock-breaking-pressure experimental results. P(z) is the confining pressure

(MPa),rc is the compressible strength (MPa), E is the Young’s modulus (GPa), and v is the Poisson’s
ratio. In general, the confining-pressure gradient is 1 MPa/1000 m.

research interests are the theory and technology of drilling,

completion, and stimulation, such as high-pressure/high-tem-
perature drilling, hydraulic-pulsed-cavitating-jet drilling, ultra-
short-radius drilling, wellbore cleanout, multistage hydraulic
fracturing, and supercritical fluid flow. Gensheng has auth-
ored 75 technical papers and holds 19 patents. He was spon-
sored by China National Funds for Distinguished Young
Scientists in 2001. Gensheng received the 2012 National
Award for Technological Invention of China, the 2007 National
Award for Science and Technology Progress of China, and
the 2010 Science and Technology Progress Award of the
Department of Education of China. He is a member of SPE.
Fig. D-1—Soft sandstone and coal samples. Gensheng holds a master’s degree in oil/gas-development
engineering from East China Petroleum Institute and a PhD
could easily obtain the values of a and b, which were 0.88 and degree in oil/gas-well-drilling engineering from China Univer-
sity of Petroleum.
6.06, respectively.
Huang Zhongwei is a professor at the Department of Petroleum
Engineering at the China University of Petroleum, Beijing. His
Appendix D—Estimation of the Threshold research interests are hydrodynamics in drilling, completion,
Rock-Breaking Pressure Pmin and stimulation, such as multiphase flow, sand cleanout, cut-
tings transport and hole cleaning, high-pressure-water jets,
To penetrate the reservoir formation, the minimum pressure drop multistage hydraulic fracturing, and ultrashort-radius drilling.
of the jet bit should exceed the threshold rock-breaking pressure. Zhongwei received the 2012 National Award for Technological
RJD is mainly applied to drill shallow, soft, and CBM reservoirs. Invention of China and the 2010 Science and Technology Pro-
On the basis of jet-bit rock-breaking experiments (Chang 2006; Li gress Award of the Department of Education of China. He is a
2008; Chi et al. 2013), we conservatively estimated the threshold member of SPE. He holds a bachelor’s degree in petroleum en-
rock-breaking pressure Pmin to be 15 MPa. The experimental gineering and a PhD degree in oil/gas-well-drilling engineering,
results are shown in Table D-1 and Fig. D-1. both from China University of Petroleum, East China.
Li Jingbin is a PhD degree candidate at the Department of Pe-
troleum Engineering at the China University of Petroleum, Bei-
jing. His work is mostly focused on hydrodynamics in drilling,
Wang Bin is a master’s degree candidate at the Department completion, and stimulation, such as ultrashort-radius drilling,
of Petroleum Engineering at the China University of Petroleum, wellbore-fluid flow, and high-pressure-water jets.
Beijing. His research interests are hydrodynamics in drilling,
completion, and stimulation, such as ultrashort-radius drilling, Zheng Dongbo is a master’s degree candidate at the Depart-
wellbore-fluid flow, and high-pressure-water jets. Bin holds a ment of Petroleum Engineering at the China University of Pe-
master’s degree in petroleum engineering from Yangtze Uni- troleum, Beijing. His work is mostly focused on drilling
versity, China, and a master’s degree in oil-/gas-well-drilling engineering and high-pressure-water jets.
engineering from China University of Petroleum, Beijing.
Li Hao is a master’s degree candidate at the Department of
Li Gensheng is on the faculty of the Chinese Academy of Engi- Petroleum Engineering at the China University of Petroleum,
neering and is a professor at the Department of Petroleum En- Beijing. His work is mostly focused on drilling engineering and
gineering at the China University of Petroleum, Beijing. His high-pressure-water jets.

March 2016 SPE Drilling & Completion 81

ID: jaganm Time: 15:43 I Path: S:/DC##/Vol00000/160001/Comp/APPFile/SA-DC##160001