Drilling Hydraulics

– Dynamic
Conditions

Energy Department
Drilling Hydraulics - Dynamic Conditions

Here is what you will be able to do when you complete each objective.
1. Describe fluid flow using rheological models and an equation of state for the fluid.
2. Calculate turbulent flow parameters of drilling fluid in drillpipe and annuli.

Introduction
Fluid mechanics is of great importance in the science and technology of drilling engineering. Very
large fluid pressures are created in the wellbore and in the tubular pipe strings by the flowing of the
drilling fluid. As such, the science of fluid mechanics and the calculation of pressure drops and pump
power required is important in the drilling process and in every drilling situation encountered.
OBJECTIVE 1: DYNAMIC CONDITIONS
DRILLING HYDRAULICS - Non-Static or Flowing Conditions
Pressure determination at various points in the wellbore can be complex when the drilling fluid is in
motion. Frictional forces in the well can be difficult to evaluate mathematically. However, in spite of
the difficulty, the effect of these frictional forces needs to be determined for the calculation of the
flowing wellbore pressure during drilling or cementing operations; the wellbore pressure during
tripping operations; the optimal pump pressure; flow rate and bit sizes during drilling operations; the
mud cuttings-carrying capacity and the surface and downhole pressures that occur in the drillstring
during well control operations.

The three basic physical laws used in calculating flow in the drillpipe and annulus are:

 Conservation of mass
 Conservation of energy
 Conservation of momentum

Equations in calculating fluid flow can be derived by applying these three physical laws with an
assumed rheological model and an equation of state. Rheological models correspond to the
Newtonian, Bingham plastic and power-law fluid models, and equations of state correspond to
incompressible fluid, slightly compressible fluid, ideal gas and real gas behaviour.

Considering the case of general flow in a system:

R e f e r e n c e p la n e

D 1

1 H eat
e n e rg y D
2
Pum p Q
e n e rg y
W

The general energy balance equation can be simplified for most drilling operations to:

2 v 2
1
V d p - g D 
2
 W - F PH – F - the mechanical energy balance equation

where F = friction loss term and includes the effect of heat flow in the
system
V = specific volume of the fluid
PH = pump input energy term
v = flow velocity
and each term in the equation has the units of ft2/s2
 1 2 p
For an incompressible fluid, V is constant (   ) , so 
1
Vdp 

and we have:

v 2
 p -   g  D       W -   F ........  2.15  *PH -F
2

J m3 kg J
We obtain the pump power from: PH W  q      3   Wat s hp or PH = ΔPp*Q/1714 hp
kg s m s

Example: Drillstring Pressure

Determine the pressure at the bottom of the drillstring if the frictional loss in the drillstring is 655 psi,
the flow rate is 400 gpm, the mud density is 12 ppg and the well depth is 6500’. The ID of the drill
collars is 2 1/2” and the pressure developed by the pump is 5500 psi.

Solution: Drillstring Pressure

The average velocity in the drill collars is:

VDC = 24.5*Q/DDC2 = 24.5*400/2.52 = 1568 ft/min = 26.1 ft/sec

PH is 5500*400/1714 = 1285
and F = 655*400/1714 = 153

Taking station 1 as the mud pits where v1  0 , p1 = 0 (gauge), D1 = 0,

 2 2
 p2 - p1  -   9.81   D2 - D1  
2
 
 v2 - v1    W -   F (PH – F) thus becomes:

p2 =  (PH – F + 0.052*D – v22/2) = 12(1285 – 153 + (.052*6500) – 26.12/2) = 13,548 psi

Rheological Models for the Fluids

The rheological models used by drilling engineers for drilling and completions are:

1. The Newtonian fluid model

2. The Bingham plastic model
3. The power-law model

1. Newtonian Fluid Model: For two plates, each of area A, immersed in a fluid and separated by
F v dv
a distance y = L, if one is fixed and a force per unit area of         is
A L dy
applied to the other, it will move through the fluid with a velocity v.  is called the applied shear
dv
stress and is the resulting shear rate. The constant of proportionality, , is the viscosity of
dy
the fluid. Note that this applies only to the case of laminar flow.
2. Non-Newtonian Fluid Model: In this case, we do not have a constant for , and the apparent
viscosity will vary with the applied shear stress or rate of shear. The two types of non-Newtonian
behaviour we are concerned with are:

The Bingham Plastic Model:

 dv
where  y = S h e a r s tre s s
  p  y for    y
dy
the yield y
......  2.16 
dv
point and p = S h e a r ra te
dy
the plastic dv
viscosity, and   p  y for     y
- y dy
again, this
only applies
for laminar dv
and for   y     y ,  0
flow dy
.
n n 1
   K ( dd vy ) dv dv
    K  ...........  2.17 
dy dy
dv dv
dy dy and if n < 1 - pseudo-plastic fluid
n < 1 n > 1 n = 1 - Newtonian fluid
n > 1 - dilatant fluid
dv n - 1
   K dy
(
dv
dy
) where K= consistency index
. P s e u d o -p la s tic P o w e r L a w D ila ta n t P o w e r L a w n = flow behaviour index or
power-law exponent

Viscometers
The standard API test for drilling fluids uses a rotational viscometer.

 2 The inner cylinder or bob is fixed on a torsion

spring, the outer rotates.
The torsion spring measures the torque or drag
 on the bob from the fluid.
r
1

r1 = 1.7245 cm r2 = 1.7589 cm L = 3.8 cm

r and the viscometer is used in field tests.
2
bob
In the Fann viscometer, the spring is chosen so
ro to r
that T  360.5   where  is the angle
reading on the dial in degrees.
The velocity at any radius r is given by v  r   where  = angular velocity in radians/s, or
dv d
r  
dr dr
If the bob is held still against the spring tension, 1 = 0 at r1. The torque at any radius is T =  ·
(2rh) · r, and with T = 360.5 ·  , we have:
 360.5  
 
2  h  r 2
dv
and, for a Newtonian model fluid,    .
dy
dv
We obtain the value of by considering that if the fluid were to rotate as a solid body (there would
dy
dv
be no slip in the fluid and hence no shear),  .
dr
dv d
The shear in the fluid is thus given by  r .
dr dr
d d 360.5  
Hence, we obtain:     r  so that 
dr dr 2  h  r 3  

2 360.5   r2 dr
Integrating this, with no slip at the walls:  0 d   
2  h   r1 r3
360.5    1 1 
to give:    2 - 2 .
4  h   2  r1 r2 
2 N
and with  2  , where N = RPM, r1 = 1.7245 cm, r2 = 1.7589 cm and
60
h = 3.8 cm.

N
  300  ..............  2.18 
N

where  = fluid viscosity in centipoise [cP]

If the viscometer is operated at 300 RPM:

   N cP

so the viscosity in cP is given directly by the angular deflection.

Non-Newtonian Fluid Model

The equations are derived in the same way, but two parameters are required. Measurements are
therefore required at 2 different rotational speeds, usually 300 and 600 RPM.

For a Bingham plastic: p = 600 - 300 y = 300 - p

 600 510   300

For a power-law fluid: n = 3.322 log10 K 
 511
n
 300
Example: Rotational Viscometer
A rotational viscometer containing a non-Newtonian fluid gives a dial reading of 12 at a rotor speed of
300 RPM and a dial reading of 20 at a rotor speed of 600 RPM. Calculate the consistency index and
flow behaviour index of the power-law model for this fluid.
Solution: Rotational Viscometer

20
n  3.32 log 10  0.737
12
510  12
K  61.8 equiv. cP
 511
0.737

Laminar Flow

In drilling, we deal primarily with the flow of drilling fluids down the circular drillstring and up the
annular space between the drillstring and the casing or open hole. If the pump rate is low and the flow
laminar, the Newtonian, Bingham plastic or power-law model can be used to develop the
mathematical relationship between flow rate and frictional pressure drop. However, assumptions must
be made in that:

1. The pipe is in the center of the hole so that the annulus is concentric.
2. The drillstring is not rotated.
3. Sections of open hole are circular.
4. The drilling fluid is incompressible.
5. The temperature is constant or, at least, an average temperature can be used.

In reality none of these assumptions is completely true. The resulting equations will not perfectly
model the laminar flow of drilling fluids in the well, but they are generally accurate enough for all
practical purposes.

Solving the annular flow case allows us to consider flow in an open pipe as a special case where the
outside radius of the inner pipe  r1  0. It can be shown that at radial distance r from the center of
the pipe, the shear stress  is given by:

r d pf C
    1
2 dL r
d pf
where C1 = an integration constant = frictional pressure gradient in the direction of flow.
dL
dv
and the shear rate = - (smaller r, larger v, hence the negative sign).
dr
 dv
Combining these with the Newtonian fluid model,    -   , and integrating, results in:
 dr

 r 
d p  ln 2 
v 
1

4 d L 
f

  r2 - r 2 - r2 - r1  r
2 2 2

r    

ln 2
 r1 

1 d pf
In pipe flow, r1  0, so ln
r2
r1
  and v  
4 d L
 2
 r2  r 2 
 La mina r Flo w

Sliding motion.
Velocity at the wall tends to zero, therefore
erosion at the wellbore tends to zero.
Velocity is maximum at the centre.
Vmax = 2V where V = Average Particle Velocity

Laminar Flow

Flow with an
Ideal plug flow Parabolic flow
unsheared plug

V max

Fluid flows in parallel layers

Velocity at the wall = 0
Velocity is maximum at the center
Velocity in the unsheared plug is uniform
Vmax = 2 V for parabolic flow
» Where V = Average particle velocity

The total volumetric flow rate is given by q   v  2  r d r , so we obtain:

 
 
2
 d p  4 4 r2 2 - r12 
  r2 - r1 - 
f
q  
8 d L  r
ln 2 
 r1 

For annular flow:

2

q    r2  r1  v
2
 where v  mean flow velocity = Q/cross sectional area

Hence,
d pf 8 v
 ..............  2.19 
dL  
 2 2
r -r
2 
 r2  r12 - 2 1 
 r
 ln 2 
 r1 
Example: Bottomhole Pressure

A 9 ppg Newtonian fluid having a viscosity of 15 cP is being circulated in a 6500’ well containing a
7” ID casing and a 5” OD drillstring at a rate of 550 gpm. Calculate the static and circulating
bottomhole pressure, assuming a laminar flow pattern exists.

Solution: Bottomhole Pressure

Static bottom hole pressure =  · g · D = 9*.052*6500 = 3042 psi

Mean annular velocity: v = Q/(2.448*(d22 – d12) = 550/2.448(49 – 25) = 9.2 ft/sec

Frictional pressure gradient is:

dpf 80 .0150.4153
= =115.7Pa/m
dL 0.17 82 0.1272 0. 8 92-0. 6352 2 2 2 2
  +  - 0. 8 9 (μ*v)/(1500*(d2 + d1 –(d2 –d1 )/ln (d2/d1))) = 138/5100 = 0.027 psi/ft
 2   2  ln
0. 635
Total frictional p in the well = 6500*0.027  176 psi

Circulating bottom hole pressure is 3042 + 176 = 3218 psi

The Slot Flow Approximation

Flow in the annulus can also be approximated as flow in a slot where

A = w · h =  (r22 – r12) and h = r2 – r1

that is, the annular space is “unrolled” and the slot has the same area as the annular space.

d1
This approximation is OK as long as > 0.3 . In
d2
h
almost all cases in drilling, it can be shown that the slot
approximation is OK, making the equations simpler.
W

Hence, using the slot flow approximation, we obtain for laminar flow:

Annulus Pipe
d pf 12  v 8 v
For a Newtonian fluid  2
 r2 - r1  r2
2
dL
d pf 12  p v 3 y 8 p v 8 y
For a Bingham plastic   
dL  r2 - r1 
2
 r2 - r1  r2
2
3 r2
n n
 2 n  1
2 K 4   v  2 K  v  3 
n
For a power-law fluid d pf
  n  n
 r2 - r1 
n 1 n 1
dL r2
Summary

Equations in calculating fluid flow can be derived by applying the three physical laws using an
assumed rheological model and an equation of state for the fluid. Rheological models correspond to
the Newtonian, Bingham plastic and power-law fluid models and equations of state correspond
to incompressible fluid, slightly compressible fluid, ideal gas and real gas behaviour.

Exercise 1
1. A rotational viscometer contains a fluid that gives a dial reading of 20 at a rotor speed of 300
RPM and a dial reading of 40 at a rotor speed of 600 RPM.

a) Is the fluid in the Fann viscometer a Newtonian fluid? Why?

b) What is the apparent viscosity at 300 RPM in Poise, centipoise?

2. A rotational viscometer contains a Bingham plastic fluid that gives a dial reading of 22 at a
rotor speed of 300 RPM and a dial reading of 39 at a rotor speed of 600 RPM. Calculate the plastic
viscosity and yield point of the fluid.

3. A rotational viscometer contains a power-law fluid that gives a dial reading of 22 at a rotor
speed of 300 RPM and a dial reading of 39 at a rotor speed of 600 RPM. Calculate the consistency
index and flow behaviour index of the fluid.
OBJECTIVE 2:TURBULENT FLOW
Turbulent Flow
Consider first the case of Newtonian fluids: as the flow rate of the fluid in a pipe is increased from
low values, a point is reached where there is a transition from laminar to turbulent flow. This onset of
vD
turbulence is related to the value of the dimensionless Reynolds number N Re  . This

transition takes place at an average value of about NRe = 2 100, but the region 2 000 < NRe < 4 000 is
usually taken as a transition zone where the flow may be either laminar or turbulent.

Turbule nt Flow

Swirling motion
Average particle velocity is uniform throughout
the pipe
The swirling motion can erode the well bore.

Example: Flow Type

A 9 ppg brine having a viscosity of 1 cP is being circulated in a well at a rate of 2 m3/min. Determine
whether the fluid in the drillpipe is in laminar or turbulent flow. The ID of the drillpipe is 10.86 cm.

Solution: Flow Type

4q 42
v    3.6 m/s
d 2
60    0.10862

9 lbm/US gal  1 078 kg/m3

1 078  3.6  0.1086

N Re   421 455 and the flow is therefore turbulent as this is > 2 100.
0.001

The frictional p for turbulent flow is based on empirical correlations. The Fanning friction factor is:

d d pf
f  .............  2.21
2  v2 d L
 
and is a function of NRe and the relative roughness of the pipe where  is the average height of the
d
wall irregularities above the mean pipe diameter. Care should be taken not to confuse the Fanning
f
friction factor f with the Moody friction factor fM , since f  M .
4

Colebrook’s correlation for f can also be used:

1   1.255 
  4 log10  0.269   ............  2.22 
f  d N Re f 

 is typically: 0.15 · 10-8 m for drawn tubing

5 · 10-8 m for commercial steel
26 · 10-8 m for cast iron
30 to 300 · 10-8 m for concrete

For most drilling operations, we can assume smooth pipe (  = 0), and for 2 100 < NRe < 100 000 and
  0 , we can use the Blasius expression:
0.0791
f  0.25
............  2.23
N Re

d pf 2 v 2 f
For turbulent flow, we then have: 
dL d
16
Note that the Fanning equation can also be used for laminar flow if we take f  .
N Re

For smooth pipe and 2 100 < NRe < 100 000, we can combine the Blasius equation with the above
equation:
d pf 2   v 2  0.0791   0.25 0.1582   0.75  v 1.75   0.25
 
dL d  d 0.25   0.25  v 0.25 d 1.25

0.2414   0.75  q 1.75   0.25

 ..........  2.24 
d 4.25
and the large effect of d should be noted!

Flow in Annuli
For flow in annuli, the pipe flow equations are used with d replaced by the equivalent circular
diameter in terms of the hydraulic radius term:

rH 
cross  sec tional area


 r2  r1
2 2

r r d d
 2 1  2 1 .....  2.25 
wetted perimeter 2   r1  r2  2 4

The equivalent circular diameter is d e  4  rH  d 2  d1 .............  2.26 

and for d1 = 0 , de = d2
Turbulent Flow for the Bingham Plastic Model Fluid
It is found empirically that the frictional pressure loss in turbulent flow of a Bingham plastic fluid can
be predicted with the Newtonian fluid equations if the Newtonian viscosity term is replaced by the
plastic viscosity. Predicting the value of the Reynolds number at which turbulent flow starts is
difficult, and usually we use both the laminar and the turbulent equations and take the higher value of
NRe as the transition value.

However, sometimes it is imperative that we ensure turbulent flow (for instance when pumping
cement slurry), so we need a more accurate method. Most commonly, an apparent viscosity  a is
calculated which can be used in the Reynolds number criterion: by combining the laminar pipe flow
equations for the Newtonian and Bingham plastic models, we obtain:

8 a  v 8 p  v 8 y
2
 2

r2 r2 3 r2
so that
 y  r2
a   p  .............  2.27 
3v
and for annulus flow:
 y   r2  r1 
a   p  ..........  2.28 
4v
and transition is taken as being at NRe = 2 100.

Example: Frictional Pressure Loss

A 10 ppg mud having a plastic viscosity of 40 cP and a yield point of y = 15 lbf/100 ft2 is being
circulated at a rate of 600 gpm. Estimate the frictional pressure loss in the annulus opposite the
drillcollars if the drillcollars are in a 6.5” hole, have a length of 1000’ and an OD of 4.5”. Check for
turbulence using the apparent viscosity concept. Use an equivalent diameter given by
de = 0.816 (d2 – d1) to represent the flow in the annulus and assume smooth pipe.

Solution: Frictional Pressure Loss

The average velocity is: v = Q/(2.448*(d22 – d12) = 600/53.8 = 11.14 ft/sec

and the apparent viscosity is:

6.5 4.5
7.18  0.3048
y D2 D1 12 12 2 2
a  p   40103     53. 10-3 Pa×s. + 5y (d2 – d1 )/ v = 40 + 5*15*2/11.14 = 53.5 cP
8v 83 .4
The equivalent diameter:
 6.5 4.5 
de  0.816  d2  d1   0.816     0.3048  0.04145 m *2 + 1.632 in
 12 12 
Dv  d v  0.04145  3.44 10 119.83
NRe   e   3206 9.28dev/μ = 9.28*10*11.14*1.632/40 = 4218
  53.3  10 3
and as this is > 2100, the flow is turbulent.
The Colebrook function gives f = 0.0098 for NRe = 4218:
Note: We could use the Blasius solution to estimate the starting value of:
0.0791
f  0.25
 0.0105
N Re

The pressure drop is:

d pf 2v2 f
 pf  L   L
dL de
2  10  119.83  3.442  0.01066  305
 Pa  2 225 kPa
0.04145

ΔPf = fv2/25.8de = 0.0098*10*11.142*1000/25.8*1.632 = 289 psi

An alternative method is to use the equation resulting from the Blasius equation:

Therefore, ΔPf = 0.75v-1.75μ0.25/1800d1.25 = 100.75*11.14-1.75*400.25/1800*1.6321.25 = 289 psi

Note that estimating f as in the first method is an approximation for the Bingham plastic. Alternative
methods (e.g., Hanks’ method) are available.

For comparison, the pressure drop may also be calculated using the expression for the laminar flow
case:

d p f 12  p v 3  y 12  40  10 3  3.44 3  7.18

   
d L  r2  r1   r2  r1   6.5 4.5 
2
 6.5 4.5 
 2 12  2 12   0.3048  2 12  2 12   0.3048
2

   
 3407 Pa/m
305
so that  pf  3407   1039 kPa 149 psi
1000

that is, much less than the pressure drop for the (correct) turbulent flow case.

Turbulent Flow for the Power-law Model Fluid

This is treated in the same way. Combining the laminar flow equations for flow in a pipe of the
Newtonian fluid and the power-law fluid:
n
n  1
2 K  v  3 
4 K v 
n n
4  8  a  v  n 2
 n 1
 6 
d2 d d n 1  n
 
2
K v
n 1 n
d 1 n  2
or, the apparent viscosity   a   6   .............  2.29 
8  n
and, in the Reynolds number expression:

8dn v
2n
dv  
N Re   ..........  2.30 
a  2
n

K 6  
 n

Values of f calculated with the Colebrook function are not accurate for power-law fluids, and the
Dodge and Metzner correlation is used:
 1 4  1 
n
0.395
 0.75 log10  N Re f 2   ........  2.31
f n   n 1.2

and, again, this applies for flow in a smooth pipe, an assumption which is acceptable for most drilling
applications. Fig. 4.34 shows values of f for this case.

The critical NRe for transition from laminar to turbulent flow is a function of the flow behaviour index
n: for any given value of n, the critical NRe is taken from Fig. 4.34 as the first deviation from the
straight line (with a minimum value of 2100).

For example, if n = 0.2, NRe,crit = 4200

The Dodge and Metzner correlation can also be applied to annular flow as above to get:
n
K  4
  8     v    d 2  d1  ........  2.32 
n 1 1 n
a 
12  n

and with this in the expression for NRe and with the equivalent diameter d e  0.816   d 2  d1  ,

n
 
9.792     v 
2n
 d  d  
N Re   2 1  ............  2.33
K  8  4  
  
n  

Example: Frictional Pressure Loss

A 15.6 ppg cement slurry having a consistency index of K =  335 equiv. cP and a flow behaviour
index of n = 0.67 is being pumped at a rate of 672 gpm between a 9.625” hole and a 7” casing.
Determine the frictional pressure loss per 100 ft of slurry. Use the expression for the equivalent
diameter d e  0.816   d 2  d1  .

Solution: Frictional Pressure Loss

The mean fluid velocity is v = Q/(2.448*(d22 – d12) = 672/106.8 = 6.29 ft/sec

NRe is 109000*15.6*6.29(2-0,67)/335 = 3612

For n = 0.67 is 2 100, so that the flow is turbulent.

For n = 0.67 and NRe = 3 612, f = 0.00815.

Therefore:

ΔPf = fv2/25.8de = .00815*15.6*6.292*100/25.8*2.142 = 9.1 psi

Particle Slip Velocity

Particles of higher density will tend to settle in a lower density fluid. In a Newtonian fluid, for a
particle settling at its terminal velocity vs, the downwards force due to its mass and gravity exactly
balances the sum of the buoyancy force and the viscous drag force:

 s  Vs  g   f  Vs  g  3   d s    vs

 ds
3
and with the solid particle volume Vs  , we get:
6

2
1 ds
vs      s   f   g .........  2.34 
18 

where this is known as Stokes’ Law and applies for very slow, or creeping, flow where there are no
turbulent eddies behind the particle: this will generally be the case for
 v d
N Re  f s s  0.1


If NRe > 0.1, empirical correlations must be used which involve friction factors defined by
F
f 
A  EK
where F = force on particle from viscous drag = 3  d s  vs
 ds
2
A = characteristic area of the particle =
4
1
 f vs
2
EK = kinetic energy/unit volume =
2
and, hence,
4 d   f
f   g  s2  s .............  2.35 
3 vs f

Hence, we get
4 d   f
vs  g s  s ............  2.36 
3 f f

Now, f is a function of NRe and the particle sphericity  (the Greek psi), defined as the surface area of
a sphere containing the same volume as the particle, divided by the surface area of the particle. The
correlation for f as a function of NRe for various  is given in Fig. 4.45.

24
Equation (2.36) can also be used for NRe < 0.1 if we define, for low NRe, f  , which results in
N Re
equation (2.34) again.

This expression may be used to calculate how fast a dilute suspended solids system will settle in a
stationary Newtonian liquid e.g., when circulation is stopped.
In a non-Newtonian fluid, the particle will settle if the net downward force is enough to overcome the
fluid gel strength. For a sphere, the surface area is  d s and the force to break the gel strength is
2

 d s   g . For no movement of the particle in the fluid, then,

2

 ds
3

 d s   g    s   f   Vs  g    s   f   g 
2

so that the gel strength  g needed to suspend a particle of diameter ds is:

ds
g    s   f   g
6
or the particle diameter must exceed

ds = g/10.4(s - f)………. 2.37

to settle through a fluid with a gel strength of  g

Example: Sand Particle Suspension

Calculate the maximum diameter sand particle having a relative density of 2.6 that can be suspended
by a mud having a density of 9 ppg and a gel strength of 5 lb/100 ft2.

Solution: Sand Particle Suspension

Maximum diameter particle:

ds = g/10.4(s - f) = 5/10.4((2.6(8.33)-9) = 0.038 in

Carrying Capacity of a Drilling Fluid

The flow velocity in the drillpipe and the annulus vary due to the velocity profiles and it is thus
difficult to predict the carrying capacity of a mud. The particles tend to settle in the flowing liquid (or
gas), so the net rate of transport is usually obtained from correlations, fro example, the Moore
correlation, Chen correlation, Walker and Hayes correlation.

The transport velocity of the particle is vT  va  vs , where va is the mean annular velocity. The
transport ratio is defined by

vT vT v
or FT   1- s ........  2.38 
va va va

If the transport ratio is:

 positive, cuttings are moving towards the surface
 = 1, the particle slip velocity is zero and vT  va
 decreasing below 1, the slip velocity increases and the concentration of cuttings in the annulus
increases
Sample and Bourgoyne found that for va   0.61 m / s , the slip velocity vs was independent of va .
1
Hence, a plot of FT versus should give a straight line of slope  vs : an experimental measurement
va
of vs can be made on a recently-stirred sample of cuttings in mud, and so FT can be found for any va .

Formation Pore Pressure and Fracture Resistance

Formation pressure is the pressure in the pore spaces and is normally increasing at a rate of
approximately 9.9 kPa/m.

The geostatic load, or overburden stress ob, is resisted by the sum of the grain-to-grain contact stress
z and the pore fluid pressure p:  ob   z  p .

In addition, the horizontal stresses x and y may be greater or less than z, depending on the type of
formation.

As the pore fluid pressure p is increased, as fracturing pressure is reached, z  0 and ob = p if the
vertical stress is the minimum principal stress: the fracture will then be horizontal. If the minimum
principal stress is in the horizontal plane (x or y), the fracture will be vertical.

Summary
In drilling operations, the drilling fluid pump rate is often too high for the flow to be laminar and the
fluid flow will become turbulent.

Exercise 2
1. A cement slurry has a flow behaviour index of 0.3 and a consistency index of 9 400 equivalent
cP. Is being pumped in an annulus of 8.097" ID and 4.5" OD at a rate of 200 gpm. Assuming the
flow rate is laminar, calculate the frictional pressure loss per 1000 ft. Also estimate the shear rate
at the pipe wall.

2. Calculate the settling velocity of sand having a relative density of 2.6, a mean diameter of
0.018” and a sphericity of 0. 81 through water having a density of 8.33 ppg and a viscosity of 1 cP.
Exercise 1 Answers
1. Solution:
a)
N 20
At 300 RPM:   300   300   20 cP
N 300
40
at 600 RPM:   300   20 cP
600
 fluid is Newtonian as we get the same value.
b) Apparent viscosity at 300 RPM = 0.2 poise or 20 cP
2. Solution:
 p   600   300
 39  22
 17 cP
 0.017 Pa  s,
and  y   300   p
 22  17
 5 lb f /100 ft 2
 2.394 Pa
 p   600   300
 39  22
 17 cP
 0.017 Pa  s,
and  y   300   p
 22  17
 5 lb f /100 ft 2
 2.394 Pa

3. Solution:
 600
n  3.322 log10
 300
 39 
 3.322 log10  
 22 
 0.826
510  300
and K
 511
n

510  22

 511
0.826

 65 equiv. cP
Exercise 2 Answers
1. Solution:

v = 200/2.448(8.0972-4.52) = 1.803 ft/sec

From the power law model

ΔPf = [9400*1.8030.3/(144000(8.097-4.5)1+0.3)]*[(2 + 1/0.3)/0.208]0.3 = 0.078 psi/ft = 78 psi/1000 ft
w = [48*1.803/(8.097-4.5)]*(2+1/0.3) = 128 sec-1

2. Solution:
vsl = [138(s - f)ds2]/μ = 138*(2.6*8.33 – 8.33)0.0182/1 = 0.126 ft/sec

Module Self-Test
1. Calculate the settling velocity of sand having a relative density of 2.6, a mean diameter of
0.025” and a sphericity of 0. 81 through water having a density of 8.7 ppg and a viscosity of 3 cP.

2. A 40 cP oil is flowing through 9000’ of 3” ID tubing at a rate of 2500 bbl/day. Calculate the
frictional pressure loss in the tubing. Assume that the flow pattern is laminar.

3. A Bingham plastic fluid has a plastic viscosity of 50 cP and a yield point of 12 lb/100 ft2.
Assuming that the flow pattern is laminar, calculate the frictional pressure gradient resulting from:

a) a flow rate of 50 gpm through a drillstring having an ID of 3.826".

b) a flow rate of 90 gpm through an annulus of ID = 7", OD = 10".

4. What is the pressure at the bottom of the drill collars if frictional losses are 900 psi with a flow rate
of 350 gpm, density is 10 ppg and TD is 8000’? DC ID = 2 ¾” and the pump pressure is 2600 psi.
The viscometer readings are 75 at 600 rpm and 35 at 300 rpm.

5. What are the n & K values for a drilling fluid having a PV = 18 and YP = 12?
 Module Self-Test Answers
1. Solution:

vsl = [138(s - f)ds2]/μ = 138*(2.6*8.33 – 8.7)0.0252/3= 0.373 ft/sec

2. Solution:

Q = 2500*42 = 105000 gal/day = 73 gpm

Mean annular velocity: v = Q/(2.448*(d12) = 73/2.448(9) = 3.3 ft/sec
Frictional pressure gradient is:
dpf 80 .0150.4153
= =115.7Pa/m
dL 0.17 82 0.1272 0. 8 92-0. 6352 2
  +  - 0. 8 9 (μ*v)/(1500*( dp1 )) = 132/5100 = 0.01 psi/ft
 2   2  ln
0. 635
Total frictional p in the well = 9000*0.01  90 psi

3. Solution:

a) Mean velocity: v = Q/(2.448*(d12) = 50/2.448(14.6) = 1.4 ft/sec

Frictional pressure gradient is:

dpf 80 .0150.4153
= =115.7Pa/m
dL 0.17 82 0.1272 0. 8 92-0. 6352 2
  +  - 0. 8 9 (μ*v)/(1500*( dp1 )) = 70/5100 = 0.014 psi/ft
 2   2  ln
0. 635
b) Mean annular velocity: v = Q/(2.448*(d22 – d12) = 90/2.448(100 – 49) = 0.72 ft/sec

Frictional pressure gradient is:

dpf 80 .0150.4153
= =115.7Pa/m
dL 0.17 82 0.1272 0. 8 92-0. 6352 2 2 2 2
  +  - 0. 8 9 (μ*v)/(1500*(d2 + d1 –(d2 –d1 )/ln (d2/d1))) = 36/10500 = 0.0034 psi/ft
 2   2  ln
0. 635
4. Solution:
5.
Static bottom hole pressure =  · g · D = 10*.052*8000 = 4160 psi

Mean annular velocity: v = Q/(2.448*(d12) = 350/2.448(7.56) = 18.9ft/sec

μ = 75 – 35 = 40 cP
Frictional pressure gradient is:
dpf 80 .0150.4153
= =115.7Pa/m
dL 0.17 82 0.1272 0. 8 92-0. 6352 2
  +  - 0. 8 9 (μ*v)/(1500*( dp1 )) = 756/11344 = 0.07 psi/ft
 2   2  ln
0. 635
Total frictional p in the string = 8000*0.07  560 psi

Pressure is 4160 + 560 + 900 = 5620 psi

6. Solution:
7.
F300 = PV + YP = 18 + 12 = 30: F600 = PV + F300 = 18 + 30 = 48

n = 3.322 log (F600/F300) = 3.322 log (48/30) = 0.68

K = 5.11 F300/511n = 5.11*30/5110.68 = 2.21

Assignment
1. Rheological models to describe fluid behavior encountered in the drilling industry are:

1) The Newtonian model.

2) The Bingham plastic model.
3) The power-law model.
Describe each of these models.

2. Describe the conditions for laminar flow in pipes and annuli.

3. Describe the conditions for turbulent flow in pipes and annuli.

References
“Applied Drilling Engineering” by A.T. Bourgoyne, M.E. Chenevert, K.K. Millheim, F.S. Young,
SPE Textbook Series Vol 2, 1991.