Orkustofnun, Grensasvegur 9, Reports 2014

IS-108 Reykjavik, Iceland Number 11

DRILLING FLUID DESIGN FOR GEOTHERMAL WELLS

George Moshe Dayan

Kenya Electricity Generating Company Ltd. – KenGen
P.O. Box 785-20117
Naivasha
KENYA
[email protected], [email protected]

ABSTRACT

Drilling is a key process in the development of geothermal resources, whether at

the exploration stage or in a fully developed field. It constitutes up to 35-50% of
project costs. Successful completion of a well at minimal cost is, therefore,
imperative. One area that highly affects both successful completion of a well and
its cost is the drilling fluid program. Drilling fluids have advanced over the years,
more so in petroleum drilling; however, some of the technologies in the oil sector
may not be relevant in geothermal drilling and could just escalate the well costs. It
is imperative to design a simple drilling fluid program that will not only ensure a
quality well but also minimise the cost of the well. This paper seeks to address the
selection of a drilling fluid best fit for geothermal wells.

1. INTRODUCTION

Drilling is a key process in the development of geothermal resources, whether at the exploration stage
or in a fully developed field. It also constitutes one of the highest costs in geothermal projects, i.e. up
to 35-50% (Kipsang, 2013). It is therefore imperative not only to successfully complete a well but
also to ensure it is done at the minimum cost that is possible. One area that highly affects both the
successful completion of a well and its costs is the drilling fluid program. A drilling fluid program
entails details on the drilling fluid to be used in the various well sections and possible actions to be
undertaken in case of special hole conditions. The drilling program is one of the critical components
in successfully completing a drilling project. Selection of drilling fluids also involves appreciating the
costs involved with each system. These can be either direct or indirect costs. Direct costs are those
incurred in the purchase, handling and utilization of the drilling fluid. They include personnel costs
and technology costs, for instance hiring an air drilling package, among others. These normally
constitute up to 15% of the well costs (Kipsang, 2013). Indirect costs, on the other hand, are costs
which can be traced to the effects of the drilling fluid technology used. This is because different fluid
systems have varying effects on the well which could either enhance or impede the drilling process;
these costs could be in terms of several extra drilling days or fishing operations among others. The
reservoir quality is also affected by the drilling fluid as some fluids lead to more formation damage.
Drilling fluids have advanced over the years, more so in petroleum drilling; however, some of the
technologies in the oil sector may not be relevant in geothermal drilling and could escalate well costs.
It is imperative to design a simple drilling fluid program that will not only ensure a quality well but
also minimise the cost of the well.

123
Dayan 124 Report 11

2. DESCRIPTION OF DRILLING FLUIDS

The drilling process mainly involves cutting the rock and bringing the cuttings to the surface. A
drilling rig is equipped with several tools and equipment to aid in the cutting and transportation of
these cuttings. One of the key systems needed is the circulation system which includes pumps,
compressors, tanks, a water pond and, most important, the drilling fluid.

In its simplest form, a drilling fluid is any fluid circulated in a well in order to bring out the cuttings
from the wellbore. This can be as simple as plain water or as complex as a fluid mixture with several
chemical additives. For a long time drilling fluids were mainly designed to bring cuttings to the
surface; however, with advancement in the drilling sector, research has proven that drilling fluids
affect drilling performance and, eventually, well performance (Baker Hughes, 1995). Today drilling
fluids are designed to take care of more than just cuttings. There are basically three types of drilling
fluid systems: mud, air and aerated systems (Ava, 2004).

For drilling fluid to function,

a circulation system is
needed; this can be either
direct circulation or reversed
circulation; in this study, we
shall focus on direct
circulation. Figure 1 shows
the key equipment in a
drilling fluid direct
circulation system, for air/gas
or aerated drilling systems;
compressors and boosters are
incorporated into the system.

The drilling fluid is pumped

from the mud tanks up the
Kelly hose through the drill
pipe to the drill bit. At the
drill bit, the fluid jets out at
high pressure, lifting with it
the freshly cut rock beneath
the bit up the annulus, while
cooling the bit and the
formation at the same time.
The fluid containing the rock
must possess enough kinetic
energy to move all the way to
the surface with the cuttings.
Once the fluid reaches the FIGURE 1: Circulation system (Ava, 2004)
surface at the diverter, it
flows to the shale shaker through the flow line where the cuttings are separated and the fluid is re-
circulated. The quality of the fluid in the tank is continuously monitored for correct PH, viscosity and
other parameters (Skalle, 2011).

2.1 Functions of drilling fluid

The initial key functions of drilling fluids were transporting cuttings to the surface, cooling the bit and
drill string and controlling the sub-surface pressures. Today, however, with the advancement in
Report 11 125 Dayan

drilling technologies, the drilling fluid has evolved to embrace several new functions. Nevertheless,
the preceding still remain the key functions that any drilling fluid program must meet. Moreover, it is
important that the drilling fluid is designed with the right properties for each function.

2.1.1 Hole cleaning

This involves carrying the cuttings from beneath the bit, transporting them to the surface and releasing
them. The aim is to transport all the cut material as fast as possible to avoid any accumulation failure
which could lead to several drilling challenges such as:
1. High torque which could lead to the drill stem snapping;
2. Stuck pipe, probably leading to the loss of the drill stem;
3. Hole pack off;
4. Damage to formation;
5. Excessive over pull during trips, hence reducing the life of the drill string; and
6. Slow rate of penetration.

Hole cleaning is the main action for which drilling fluid design is done. A fluid must have adequate
viscosity, density and flow, at the right rate to carry the cuttings to the surface.

2.1.2 Well control

When preparing a drilling fluid, one must always remember that it is actually the key well control
system. The pressures in a well increase with depth. When the formation fluid has higher pressures
than those of the fluid in the well, a kick or blow out may occur. On the other hand, an overbalanced
drilling fluid may cause formation damage by exerting excessive pressure on the formation wall. The
key property of the fluid to be monitored in this case is the density.

2.1.3 Maintain borehole stability

A single well profile has formations with varying properties: some fractured, erodible or swelling.
These could result in very problematic formations during the drilling process. The drilling fluid is the
main option in addressing these and ensuring the borehole is kept stable to ensure the drill bit and stem
runs through successfully to the total depth.

2.1.4 Protecting formation from damage

This has become an increasingly important function of drilling fluids. In geothermal drilling, aerated
fluids are used in the production zone with the aim of minimizing formation damage. Formation
damage may occur mainly due to plugging of the formation’s natural porosity, either by solids or
plugging associated with fluid filtration.

Other functions of drilling fluid include:

1. Cooling and lubricating the bit and drill stem;
2. Improving the rate of penetration by cleaning the surface to be drilled;
3. Contributing to drill string buoyancy; and
4. Retrieval of formation data.

2.2 Types of drilling fluids

A drilling fluid in its simplest form can be plain water; however, many times the properties of water
must be improved to achieve the various tasks required of a drilling fluid. The basic composition of a
drilling fluid is a base fluid, with a continuous fluid phase and can be water, oil or air. Various
Dayan 126 Report 11

additives are introduced to these base fluids to achieve a given property, for instance viscosifiers are
added to improve the viscosity of the fluid (Darley and Gray, 1988). The drilling fluids are, therefore,
classified depending on the base fluid used: water, oil or air. Nevertheless, all drilling fluids have
essentially the same properties; only the magnitude varies. These properties include density, viscosity,
gel strength, filter cake, water loss, and electrical resistance (Baker Hughes, 1995). It is important to
note that drilling fluids in geothermal drilling are rather simple since the formation that is drilled
through many times is under-pressurised.

2.2.1 Water based drilling fluids

These have water as the continuous base and can be either fresh or saline water. This is the main type
of fluid used in geothermal drilling, more so for the upper cased well sections. Active and inert solids
are normally added to change the mud properties. Common active additives include bentonite and
polymers, mainly used for improving on viscosity, which is important for the cuttings transport
capacity of the mud. Inert solids include particles added to the mud, such as formation particles; a
common substance normally added to increase mud density is barite (Darley and Gray, 1988). This,
however, is rarely used in geothermal drilling.

Bentonite mud made by mixing bentonite into water has the advantage of gelling and also forming a
filter cake around the wall of the well. The gel assists in suspending the cuttings in case the
circulation is stopped, reducing the chance of the cuttings dropping down on the string. The filter cake
is important in reducing loss of circulation since it forms an impermeable layer around the wall, and it
also helps in increasing well stability. The filter cake, however, can be problematic if it is too thick
and can also hinder good cementing jobs if not cleaned off. Polymers like bentonite are added with
the aim of improving the viscosity of the mud; however, they have poor gelling properties and do not
form a filter cake (Baker Hughes, 1995).

Other special water-based muds can be made to address specific well situations. These, however, are
rare in geothermal drilling, but are common in oil drilling, and include:
1. Emulsion muds-oil in water;
2. Inhibited muds-large amounts of dissolved salts added to the mud; and
3. Lime treated muds.

Caustic soda is normally added to improve the pH of the fluid.

Also used in the oil sector is oil-based mud where oil is used as the continuous fluid phase and
additives are put in to achieve various properties. This is rarely utilised in geothermal drilling, both
because of costs and environmental concerns.

2.2.2 Air and aerated drilling fluids

Air drilling involves the utilisation of compressed air as the drilling fluid. Normally, this air is
delivered through the string just like typical drilling mud, but with higher velocities and carries up
with it the cuttings. It is a very efficient method for drilling in dry or frozen formations. Once water
is encountered, then the cuttings transportation capacity is greatly hampered (Skalle, 2011). In
geothermal drilling, this is mainly applied for drilling the surface section of the well where many times
the rock is competent and very hard. Air hammer drilling is applied in Iceland and in the Kenya-
Menengai geothermal project for drilling the surface hole (Thórhallson, 2014).

There are various types of air drilling, all depending on the liquid volume fraction, LVP, which
indicates how much liquid is in the system and is a measure of the density of the fluid mixture: zero
LVP implies no liquid, while one implies 100% liquid (Hole, 2006). All air drilling systems have
LVP of less than one. Also of significance is the method of injection of the air stream into the system;
Report 11 127 Dayan

this can be either through the drill pipe or through the annulus (Lyons et al., 2009). In geothermal
projects, the most common method used is the drill pipe injection method.
1. Dust drilling: Compressed air is used as the sole drilling fluid; this is ideal in dry areas where
we do not expect to encounter liquids. It is used with air hammers. Normally, there is zero
LVP, implying the system is 100% air.
2. Mist drilling: Air drilling with the addition of liquids, normally water and soap. This is
introduced when a wellbore gets liquid influx during dust drilling.
3. Foam drilling: Foam is created by combining water, surfactants and air. Has better cuttings
carrying capacity.
4. Aerated drilling: In this type of drilling, air or nitrogen is added to the drilling mud. The mud
can be water-bentonite, water-polymer or water-foam. The water-foam drilling is also called
stable foam drilling and involves the use of water mixed with surfactants or air. This is the
most common form of aerated drilling used in the geothermal industry.

FIGURE 2: Down hole pressures (Hole, 2006)

The introduction of air drilling in geothermal systems is mainly due to the fact that most geothermal
formation pressures are significantly lower than the hydrostatic column of water at any depth within
the system, as illustrated in Figure 2 (Hole, 2006).

Figure 2 indicates the specific gravities of the various drilling fluid, indicating the importance of
aerated systems in keeping the drilling fluid under-pressurised in comparison with the formation
pressure.
Dayan 128 Report 11

2.3 Operating principle of drilling fluids

The working principle of any drilling fluid depends upon the function for which it is designed. Most
drilling fluids are, from the onset, designed to lift the cuttings to the surface and control the well. As
the drilling progresses and special conditions are encountered, the drilling fluid may be redesigned to
perform other functions. However, in all cases the drilling fluid parameters considered are viscosity,
velocity and density; other properties such as gel strength, filter cake, water loss, are dependent on the
former (Darley and Gray, 1988).

To discuss the operation of drilling fluids, we shall consider the main functions of drilling fluids as
identified earlier: hole cleaning, well control, protection of the formation and borehole stability.

2.3.1 Hole cleaning

Hole cleaning results from the continuous transportation of cuttings from beneath the bit to the surface
where they are released. This is the key function of any drilling fluid and, hence, is the main
controlling factor in fluid design. The key factors affecting cutting transportation are drilling fluid and
cuttings velocity. The drilling fluid velocity depends on the rheological parameters of the fluid,
density, the pump rate and the annular geometry of the well, whereas the key cuttings parameters are
density, diameter and shape (Mitchell and Miska, 2011).

Cuttings generated in a well experience a downward movement due to gravity, resulting in negative
velocity relative to the fluid’s velocity. This negative velocity is termed cutting slip velocity, Vsl, and
is key in determining cuttings transport. The movement of the cuttings up the annulus is the result of a
net upward velocity, called cutting transport velocity, Vt; this is the difference between the fluid
annular velocity Va and the cuttings slip velocity Vsl. This implies that in designing a fluid for
efficient hole cleaning, we must appreciate how the cuttings behave and calculate the expected slip
velocity (Rehm et al., 2012).

The slip velocity can be determined depending on the type of fluid and flow regime; for laminar flows,
there are different relationships for Newtonian and non-Newtonian fluids due to the effect of the
fluid’s rheological properties. In turbulent flow, the rheological properties have no effect (Darley and
Gray, 1988).

Hole cleaning, however, is not just about carrying the cuttings out but also ensuring minimal cuttings
concentration in the well during the drilling operation. For trouble free drilling, the cuttings
concentrations in the well, Ca, should be below 5%, and the fluid parameters must be controlled to
achieve this (Azar and Samwel, 2007). An additional way of designing a fluid is to consider its
cuttings capacity index (CCI); this is a relationship based on various fluid parameters that are key for
ensuring hole cleaning. This was arrived at from various field studies which concluded that for
efficient hole cleaning in vertical and near vertical wells, CCI ≥1. The drill bit also plays a key role in
effective hole cleaning; therefore, bit hydraulics must be considered when designing the drilling fluid
(Rehm et al., 2012).

The following equations relate the various key hole cleaning parameters and how to arrive at them:

Critical annular velocity

This is a key parameter and is the minimal annulus velocity at which the cuttings concentration in the
annulus reaches its threshold value and can be obtained by (Rahimov, 2009):
1667
(1)
60

where Vac = Critical annular velocity (ft/s);

Report 11 129 Dayan

Vsl = Slip velocity of cuttings (ft/s);

ROP = Rate of penetration (ft/hr);
Ca = Cuttings concentration in the annulus (%);
Db = Drill bit diameter (in); and
Dp = Drill pipe outside diameter (in).

To maintain a specific cuttings concentration (Ca) in the annulus, the annular velocity must not go
below this.

Cuttings slip velocity

Several correlations have been developed for slip velocity, all of them based on Stokes law with the
exception of those developed by Walker and Mayes (Mitchell and Miska, 2011).

Cuttings slip velocity for Newtonian fluid, Stokes law.

From Stokes law, the slip velocity (Vsl) of an object falling through a viscous fluid is given by:

18 (2)

2 3
(3)
3

4
(4)
3

where ds = Particle diameter (m);

ρS = Cuttings density (kg/m3);
ρf = Fluid density (kg/m3);
  Dynamic viscosity (kg/ms);
g = Acceleration due to gravity (m/s2);
CD = Drag Coefficient; and
f = Particle friction factor – obtained from graph in Appendix I.

Equations 2 and 3 apply for Newtonian fluids (e.g. water) in laminar and turbulent flow, respectively.
Equation 4 gives the drag coefficient corresponding to the slip velocity calculated (Mitchell and
Miska, 2011).

To determine the flow regime, we calculate the Reynolds number of the particles (Rep):

928.2 (5)

where ρf = Fluid density (kg/m3);

ds = Particle diameter (m);
Vs = Slip velocity (m/s); and
µ = Newtonian viscosity of the fluid (Pa s).

When 1.0, this defines a laminar flow regime; 1.0 2000 defines a transition flow
regime, and 2000 defines turbulent flow.

Calculation of slip velocity for Non-Newtonian drilling fluid

Stokes law does not give an accurate approach for determining the slip velocity in non-Newtonian
fluids, due to the effect of rheological parameters on the fluid (Guo and Liu, 2011). Several
correlations have been developed to determine slip velocity in non-Newtonian fluids, described below:
Dayan 130 Report 11

Moore correlation
Moore suggested the use of an apparent viscosity (µa) instead of the Newtonian viscosity of the fluid
as used in Stokes law (Rahimov, 2009). Apparent viscosity is based on a pseudo plastic fluid model
and is given by:

2
(6)
144 0.0208

And the slip velocity for laminar flow, where 1.0, is given by:

82.86 (7)

Slip velocity for transitional flow, 1.0 2000:

.

2.90 . .
(8)

Slip velocity for turbulent flow, Re 2000:

1.54 (9)

where K = Consistency index (lbf secn/100 ft2);

n = Flow behaviour index (dimensionless);
ρf = Weight of drilling fluid (ppg);
ρp = Weight of rock particles (ppg);
Va = Fluid velocity in annulus (ft/s);
µa = Apparent viscosity (cP);
Dh = Hydraulic diameter for annulus (in);
Dp = Pipe diameter (in); and
dp = Diameter of cuttings (in).

Chien correlation
This is similar to Moore’s correlation in that it also uses the apparent viscosity in calculating Reynolds
number. Chien’s analysis is based on a Bingham fluid model. The following equations apply:

5 (10)

where µa = Apparent viscosity (cP);

µp = Plastic viscosity (cP);
τy = Yield stress (lbf/100 ft2); and
Va = Fluid velocity in annulus (ft/s).

36800 1
0.0075 1 (11)

Equation 10 gives the apparent viscosity to be used in polymer type drilling fluids. For bentonite,
Chien proposed the use of plastic viscosity (Rahimov, 2009). Equation 11 gives accurate slip velocity
Report 11 131 Dayan

μ
when the viscous properties of drilling fluids are abnormally high, i.e. when ρd 10; otherwise,
Chien proposed a simpler equation for normal fluids:

1.44 (12)

Walker and Mayes correlation

In this correlation, the drill cuttings are assumed to be shaped like a circular disc, unlike the above two
cases where the particles are assumed to be spherical. Hence, the particles will fall through the fluid
with the flat face lying horizontal. The shear rate, called the boundary shear rate ( ), at which a
particle’s movement switches from laminar to turbulent, is calculated by:
186
(13)

The shear stress developed by the particles as they fall through the drilling fluid is given by:

7.9 (14)

Once the stress is determined, the corresponding shear rate is determined by using annular power law
constants:
⁄
(15)

If or 100, then the slip velocity of a particle is in the laminar zone and is determined
by:
.
0.02 (16)

If or 100, then the slip velocity of a particle is in the turbulent zone and is determined
by:

0.28 (17)

where Tp = Particle thickness (in);

γb = Boundary shear rate (1/s);
γp = Shear rate corresponding to τp (1/s);
τp = Shear stress developed by particles (lbf/100 ft2);
Ka = Consistency index in annulus (lbf secn/100 ft2); and
na = Flow behaviour index in annulus (dimensionless).

Calculating minimum annular velocity for hole cleaning:

(18)

(19)
4 3600
where ROP = Rate of penetration (ft/hr)-Rd;
Ca = Cutting concentration factor in the wellbore = 0.04;
db = Bit diameter in inches; and
Dayan 132 Report 11

A = Cross-sectional area of flow path (in2).

But for best hole cleaning, the value of the rise velocity, Vt, should approach the annular velocity,
(Rahimov, 2009).

From this, the minimum flow rate Qmin required for this velocity, Va, is gotten by (Guo and Liu, 2011):

60 (20)

where Qm = Volumetric flow of drilling mud/water (m3/s); and

Va = Fluid velocity in annulus (m/s).

Another method based on a power law fluid, which closely represents the behaviour of drilling fluids,
involves the determination of the minimum velocity required to transport the fluid and cuttings
mixture Vmix; this is based on a cuttings concentration already defined earlier as Ca (Mitchell and
Miska, 2011). The below equations are used:
0.0475
0.05 (21)

(22)

(23)
4
where Vmix = Velocity of the cuttings and fluid mixture (m/s); and
Qc = The volumetric flow rate of the cuttings generated at the bit (m3/s).

By substituting Equation 23 into Equation 22, we solve for the value of Qm, which is the volumetric
flow of mud required to attain Vmix.

The efficiency of hole cleaning is checked by either monitoring the CCI, the annular volumetric
cuttings concentration, Ca or the ratio of the cuttings transport velocity (CTR) Rt to the cuttings
annular velocity Va, Rt. In vertical drilling, it is recommended that Rt should be a minimum of 0.5-
0.55 (Azar and Samuel, 2007).

400,000 (24)

1.27 (25)

1 (26)

where CCI = Cuttings carrying index;

CTR = Cuttings transport rise; and
ROP = Rate of penetration (m/hr)-Rd.

The K factor power law fluids is shown in Appendix II.

Hole cleaning in aerated drilling

When designing hole cleaning for aerated drilling, the practise is to design the incompressible drilling
fluid to have a minimum lifting capability for the planned open hole interval. This implies the fluid
Report 11 133 Dayan

should be capable on its own in maintaining a minimum concentration of rock cuttings in the largest
annulus section of the well (Lyons et al., 2009). The assumption in this approach is that the
incompressible fluid can carry the cuttings on its own, and that the injection of air into the fluid will
enhance this capacity.

To fulfil this, the average velocity of the incompressible fluid Vf in the largest annulus section must be
equal or greater than the sum of the critical velocity Vc and the terminal velocity Vt of the average size
rock cutting particle in the drilling fluid (Lyons et al., 2009). This is the same as in normal drilling
while using mud, hence the equations described in the preceding section for minimum annular velocity
and flow rates will apply in determining .

However, the annular velocity of the air and the incompressible fluid mixture can be gotten by:

(27)

And the volumetric flow of air Qair at any depth in the well is given by:

(28)

where  Pg = Atmospheric pressure at the compressor inlet (N/cm2 abs);

P = Pressure entering the pipeline (N/cm2 abs);
Tg = Temperature of the air entering the compressor (°C);
Tav = Average temperature of the air over a depth interval;
Qg = Volumetric flow rate of the compressible gas (ft3/s); and
Qm = Volumetric flow rate of the incompressible fluid (ft3/sec).

Qm is the minimum mud flow rate required to attain the minimum annular flow velocity, Vf for
efficient hole cleaning. Equation 27 shows that it is possible to increase the annular velocity without
increasing the flow of mud by adjusting the flow rate of the air Qg; this means better hole cleaning
with minimal increase in fluid density, important for under-pressurised formations.

Aerated drilling using stable foam

Aerated drilling with stable foam introduces a new phase that must be monitored. Normally the
surfactants, water and air, flow down the drill pipe as a normal aerated fluid. However, as it goes
through the bit nozzles, the high shearing action leads to the generation of foam. This foam flows up
the annulus as stable foam. To maintain the stability of this foam and avoid disintegration into a new
phase, it is important to monitor a parameter known as the foam quality (Lyons et al., 2009). The
foam quality at the entrance is defined by:

Γ (29)

where Γ = Foam quality.

Usually, the foam quality should be maintained at 0.98 at the exit point and at least 0.6 at the bottom
hole. This is done by monitoring the pressure at which the mixture exits in the horizontal flow line at
the surface. A gauge is normally installed at the surface at the back pressure valve in the horizontal
flow line to check the exit pressure. The back pressure valve reading is calibrated to correspond to
various foam quality values and can then be adjusted to achieve the required foam quality (Rehm et
al., 2012). At the exit flow line, the volumetric flow rate, Qbp for the compressed air exiting is used to
calculate the foam quality and is gotten by:
Dayan 134 Report 11

(30)

where Pbp = Back pressure.

The back pressure valve is adjusted to get the pressure necessary for the desired foam quality of 0.98.

Air drilling
The use of air, alone, as the drilling fluid is sometimes considered; this is especially beneficial if the
water table is low or in areas with a scarce water resource. In geothermal drilling, an air hammer has
been used in drilling the upper well sections which have hard rock and may take longer to drill using
conventional rotary drilling. This has been practised with success in geothermal fields in Menengai,
Kenya and in Iceland.

In designing the air system, the main issue is the minimum gas volume requirement for good hole
cleaning. This can be determined by using the minimum velocity criterion or the minimum kinetic
energy criterion. We shall consider the minimum velocity criterion. This criterion uses the same hole
cleaning concept as in normal drilling, where the aim is to exceed the cutting terminal velocity; hence,
we first determine this velocity in the air. Once this velocity is determined, then the fluid flow rate is
designed to exceed it (Guo and Liu, 2011).

The minimum required gas velocity is given by (Guo and Liu, 2011):

(31)

4
(32)
3 1

The volumetric flow rate can then be obtained from:

60 (33)
144

(34)

where Qg = Volumetric flow rate of gas (cfm);

Qgo = Volumetric flow rate of gas in the standard condition (scfm);
Vg = Minimum required gas velocity (ft/s);
Ds = Solid particle (ft);
ρS = Cuttings density (lbm/ft3);
Vsl = Cutting slip velocity in air (ft/s);
CD = Drag coefficient accounting for the effect of particle shape;
Ψ = Particle sphericity factor, obtained from graph in Appendix I (-); and
DH = Hydraulic diameter of flow path (ft).

2.3.2 Well control

Well control is all about controlling kicks and eliminating kicks, hence limiting the chance of a
blowout. Kicks and blowouts occur when the fluid pressure in the formation exceeds the pressure of
the drilling fluid in the well. When a kick is not contained and stopped, it leads to a blowout which is
a more difficult situation. In geothermal drilling, the fluid acts both to cool the hot fluid and also keep
the formation fluid pressure under check. The key parameter checked is the mud weight. The
hydrostatic pressure due to the drilling fluid is obtained from the formula:
Report 11 135 Dayan

∗ ∗ 0.052 (35)
Another key parameter is the equivalent circulating density (ECD); this is the density of the mud due
to the effect of pump pressure applied on it for circulation. This is normally higher than the fluid’s
specific density, and is given by:

(36)
0.052 ∗
The ECD results in a new pressure called the bottom hole circulating pressure (BHCP) which is the
actual pressure the drilling fluid has during circulation (Azar and Samuel, 2007). It is important to
consider this pressure during the design process in order to ensure that it is kept low; it is given by:
∗ 0.052 ∗ (37)
where P =The hydrostatic pressure (bars);
MW =Mud density (kg/m3);
Pa =Annulus frictional pressure drop at a given circulation rate (Pa); and
D =Depth (m).

In geothermal drilling the production zone is often drilled using aerated water and foam which, many
times, is a form of underbalanced drilling; the aim is to allow formation fluid to flow into the well,
hence reducing damage to the formation. This, however, exposes the operation to higher chances of a
kick and/or a blowout.

2.3.3 Protection of formation

There are two main parts of borehole drilling, the upper section which is normally totally cased and
cemented, and the pay zone, which in geothermal wells is usually cased using slotted liners. The pay
zone, or the main hole, in many geothermal hot temperature wells ranges from 700 m up to 3000 m or
beyond. Prevention of formation damage is especially important when drilling the pay zone.
Permeability is one of the most important properties of a geothermal reservoir. During drilling, the
cuttings generated can clog the rocks, leading to loss of permeability. The inert components of drilling
fluids can also flow into the formation, blocking aquifers or other permeable zones.

In geothermal drilling, aerated drilling is often introduced in the production zones. This is because it
is possible to achieve better hole cleaning with compressed air, hence reducing the tendency of
cuttings to clog the formation. The aerated
drilling fluid has a lesser density than normal
mud, hence causing less formation damage. In
some cases water is used to drill with high
viscous polymer sweeps, used when making
connections. The aim in both cases is to reduce
the formation damage by having as few cuttings
and other solids getting into the formation as
possible (Hole, 2006).

Designing a drilling fluid to help reduce

formation damage involves appreciation of the
well’s drilling window, which is the margin
between the fracture and pore pressure, shown in
Figure 3. As a drilling practise, wellbore
pressure should always be greater than pore
pressure, both in static and dynamic conditions.
However, it should not surpass the fracture FIGURE 3: Drilling window
gradient; otherwise, costly drilling challenges (Rahimove, 2009)
Dayan 136 Report 11

such as loss of circulation and formation damage occur. Geothermal reservoirs are characterised with
subnormal pore pressures. Hence, the drilling window is wider. The tighter the drilling window, the
tougher the drilling will be (Vollmar et al., 2013). However, when drilling the main hole, pressures
below the pore pressure are desired in order to reduce formation damage. Hence, aerated drilling is
used to achieve this.

2.3.4 Maintain borehole stability

The aim of drilling is to create a pathway in the ground to access a given resource; in a geothermal
system, the resource is hot fluid from the ground. This may require drilling to depths ranging from
500 m to 3000 m or beyond. It is important to keep this path, borehole, intact as we drill on. This is
done by casing the various well sections and cementing them. However, prior to casing and
cementing, there is still need for well stability. The drilling fluid provides this stability.

In an undisturbed formation,
the rock matrix and the pore
pressure are able to withstand
the overburden pressure.
When a borehole is made in
this system, an imbalance of
forces is created. There will
be a net tangential force
which tends to cause the
formation to move to fill the
new cavity formed. This
increases with depth due to
the increase in overburden
pressure. The diagrams in
Figure 4 illustrate the FIGURE 4: Stress conditions in a borehole (Ava, 2004)
scenarios before, during, and
after drilling (Ava, 2004).

The drilling fluid provides a radial pressure which balances against the instability created in the
formation due to the borehole. The density of the drilling fluid is a key factor of this radial pressure.
In some cases, barite or other inert solids are added to the drilling fluid in case of an over pressurised
formation (Ava, 2004).

An unstable formation is also caused by fluid seepage from the borehole into the formation. Drilling
fluids, such as bentonite, form an impermeable layer on the wall of the borehole which reduces the
seepage. This is of great significance in clay formations (Chemwotei, 2011).

2.4 Drilling fluid equipment

The drilling fluid equipment forms the circulation system of the rig and is mainly used for applying
pressure on the fluid and providing a channel for fluid to flow. The basic setup is made up of mud
pits, mud pumps, mud mixing equipment and contaminant removal equipment. The mud pumps
provide the pressure needed for the drilling fluid to flow through this system into the well and up with
the cuttings. The selection of the pumps will depend upon the hydrostatic pressures expected to be
handled during the drilling process.

In aerated drilling, an additional system is included, depending on the type of gas to be used. Most
geothermal drilling uses aerated water and foam with two primary compressors and one booster.
Report 11 137 Dayan

2.4.1 Criteria for selection of drilling fluid equipment

The key equipment in the drilling fluid system is the mud pump. This is responsible for supplying the
hydraulic pressure needed to move the drilling fluid from the pump through the entire drill string, and
back to the mud tanks. Figure 5 illustrates the sum of the pressures that constitute the pressure to be
supplied by the pump (Baker Hughes, 1995). Pumps are rated for hydraulic power, maximum
pressure and maximum flow rate. Any of these can be used for our design criteria in selecting our
pump. Once the drilling program is known, it is possible to determine the required flow rates and
pressure in drilling the various well sections. Of significance is that the pump must supply adequate
flow to achieve the required annular velocity for effective hole cleaning.

FIGURE 5: Pressure loss in a pump (Baker Hughes, 1995)

The required pump flow rate is calculated based on annular volumes and required up flow velocity
(Lapeyrouse, 2002).

∗ (38)
24.5
where Dh = Well or casing diameter (m);
Dp = Outside diameter of drill pipe (m); and
v = Desired annular velocity (m/s).

The ideal minimum velocity when using mud is 0.3 m/s and while using water is 0.7 m/s (Chemwotei,
2011).

The flow rate required is calculated for the various well sections and the maximum flow rate required
is then used to select the pump. Another significant parameter is the pump pressure rating; this should
normally be at least 1.5 times the total pressure losses, which are illustrated in Figure 5. The pressure
losses are comprised of:
Dayan 138 Report 11

1 2 3 4 5 (39)
where p1 = Losses through surface equipment;
p2 = Losses through drill pipe/drill collars;
p3 = Losses through the rock bit;
p4 = Losses between the outer diameter of the drill pipe and drill collar, and wall
= of the hole; and
p5 = Losses in the mud motor (when used, e.g. directional drilling).

FIGURE 6: Well profile (Hole, 2007)

The pressure losses have the form (Gabolde and Nguyen, 2014):
(40)
Calculation for N in the drill string is given by:
.
(41)
901.63 .

Calculation for N in the annulus is given by:

Report 11 139 Dayan

.
(42)
709.96
where p = Pressure loss (kPa);
N = Pressure losses for pure water;
B = Coefficient corresponding to circulating mud;
Do = Annulus outside diameter (in);
Di = Annulus inside diameter (in);
Q = Drilling fluid flow rate (l/min); and
L = Length of drill string (for N in drill pipe is the length of the drill pipe for the drill
= collar and surface equipment) (m).

The coefficient B corresponds to circulating mud and is given by:

. .
(43)
Pressure loss in bit nozzles is given by:

(44)
2959.41 0.95
Hence, we get:
(45)
where d = Specific gravity (kg/l);
µp = Plastic viscosity (cP);
A = Area of nozzles (in2);
N1 = Pressure loss coefficient in the surface equipment;
N2 = Pressure loss coefficient in the drill pipe;
N3 = Pressure loss coefficient in the drill collar;
N4 = Pressure loss coefficient in the hole/drill collar annulus; and
N5 = Pressure loss coefficient in the hole/drill pipe annulus.

Once these are calculated, the pump power required is calculated from:

(46)
600
where Ppower = Pumping power (kW);
p = Pump discharge pressure (bar) =1.5*Ptotal;
Q = Fluid flow rate (l/min);
ηm = Pump mechanical efficiency, assumed to be 0.85; and
ηt = Transmission efficiency, assumed to be 0.9 (for motor).

It is important to note that these losses depend upon fluid viscosity, density and the flow rate, so we
must consider this effect as we design the fluid. Higher flow rates, viscosities and density imply
higher pressure losses, hence a more expensive drilling fluid system. The system losses are normally
higher for turbulent flow than for laminar flow; turbulent flow occurs at high velocity and smaller
annular space, while laminar flow tends to occur at lower velocities. From the above calculations, we
can select the right pump for the drilling activities.

2.5 Drilling fluids and challenges in geothermal drilling

Dayan 140 Report 11

Drilling challenges vary in type and severity depending upon the well. However, the main challenges
encountered in most situations are stuck pipe, loss of circulation and well control. Most of these
challenges are due to the geology of the formation.

Drilling challenges can either be due to formation challenges or equipment failure. The most difficult
challenges are normally those associated with the formation since many times we are not able to see
what is in the formation and have to rely on the data to try and visualise the actual situation in the
borehole. One of the worst challenges is a stuck pipe situation. Many times this leads to loss of
drilling time, expensive fishing operations, formation damage and possibly a loss of the drill string.
Apart from depleting the moral of the crew, a stuck pipe leading to fishing operations normally
escalates the price of the well. As such, any well advised drilling team will do all in its power to avoid
a stuck pipe situation. Appreciation of drilling fluids and their interaction with various formations can
help save a stuck pipe situation from escalating.

2.5.1 Loss of circulation

This is the total or partial loss of drilling fluids due to highly permeable zones, cavernous formations
and natural or induced fractures during drilling. Often this affects well control, borehole stability and
may lead to formation damage. Loss of circulation is often dependent on how the drilling fluid
interacts with the formation during the drilling process. The main causes are:
1. Formation pore spaces are too large, or the particles in the fluid are too small to allow filter
cake formation;
2. Hydrostatic pressure is sufficient enough to force wellbore fluids into the pore spaces;
3. Hydrostatic pressure causes a natural fracture to open up and take wellbore fluid; and
4. Hydrostatic pressure induces fractures in weak formations.

From the above causes, we notice that most of them are related to the hydrostatic pressure, which is a
property of the drilling fluid, highlighting the significance of drilling fluids in managing loss of
circulation challenges. There are other causes related to how fast the drill pipes or casings are run in
hole, among other things (Baker Hughes, 1995).

The severity of the loss is classified by the volume of fluid loss per unit time: seepage losses with
losses of up to 1.6 m3/hr; partial losses with losses ranging from 1.6 m3/hr to 79 m3/hr; complete losses
when no returns are got on surface. In order to determine the right treatment for the drilling fluid to
heal the loss, we must identify the type of loss zone. The table below describes the common loss
zones and the possible type of losses (Ava, 2004).

TABLE 1: Loss zones

Loss zones Type of loss

Porous and permeable Losses start as a gradual reduction in pit volume. If drilling proceeds, the
sands and gravel losses could become complete. These zones usually occur near the
surface.
Natural fractures May occur in any type of rock. Usually these losses are partial but may
progress to complete loss as drilling proceeds or if the fluid density
increases.
Induced fractures Sudden and complete losses.
Vugular formations Usually located in limestone. Losses here can be sudden (located at the
bit) and complete. On occasion, the bit may drop a few inches before the
loss.

Losses in geothermal drilling are mostly due to encountering permeable zones or fractures, either
natural or induced. There are several ways of handling loss of circulation, and many loss-of-
Report 11 141 Dayan

circulation materials have been developed to suit various situations. However, in general the action
taken depends upon the location of the loss zone within the well.

There are various loss-of-circulation policies applied in various situations, but the general tendency is
to switch to water and drill blind whenever a complete loss is encountered. Minor or seepage losses
can be controlled by adding 10-60 kg/m3 of LCM to the drilling fluid. Increasing the quantity of LCM
can be done to tackle higher losses. However, if the loss persists, then three main options can be
explored:
1. Drill on blind with water;
2. Switch to aerated drilling; and
3. Stop and plug the well.

Table 2 gives a summary of some actions that can be taken to handle losses:

TABLE 2: Loss of circulation policy

Severity of loss Loss section Action

Seepage losses Upper well section Add LCM to the drilling fluid, ranging from 10 to 60
with losses of up kg/m3.
3
to 1.6 m /hr Main hole Usually this section is drilled with water or aerated fluids.
Continue drilling but spot polymer pills at every
connection.
Partial losses Upper well section 115-230 kg/m3 of LCM is spotted in pill form.
from 1.6 m3/hr to Main hole Switch to aerated drilling or drilling on with water but spot
79 m3/hr polymer pills at every connection.
Complete loss Upper well section 230-430 kg/m3of LCM is spotted in pill form. If loss
persists, consideration of stopping the drilling to plug the
loss zone should be looked into, to avoid hole cleaning
challenges which could lead to stuck pipe.
Main hole Switch to aerated drilling or continue drilling with water
and polymer pills at each connection. Proper hole cleaning
and wiper trips should be done before any new connection.

The treatment given to loss of circulation in the main hole is applied in a manner such as to avoid any
clogging of the cavities, since they are key to the well’s production. Often the tendency is to drill
blind. Hole cleaning challenges are handled by high viscosity polymer sweeps.

2.5.2 Stuck pipe

Stuck pipe can be described as a situation where both axial motion and sometimes rotary motion of a
drill string in a borehole is lost. This, at times, may be accompanied with blocking of the bit such that
circulation through the drill pipe is not possible. There are several causes of stuck pipe, however,
most are due to geological challenges which are dependent on the nature of the formation. The most
common formation challenge is instability. An unstable formation for whatever reason is more likely
to cave in on the drill string and possibly lead to a stuck pipe. The other common cause of stuck pipe
is poor hole cleaning situations, either due to loss of circulation or poor drilling fluid design and
differential sticking.

Differential sticking is a result of the drill pipe getting stuck on the wall of the well. This is due to the
formation of a sticky layer called a filter cake. The filter cake is a good property of any drilling fluid
because it prevents losses of the drilling fluid. However, there is a limit to the thickness that is
considered beneficial (Bourgoyne et al., 1991).
Dayan 142 Report 11

The best solution to handling stuck pipe is to avoid any stuck pipe situations. This can be done by
keenly monitoring well lithology to appreciate the formation. Drilling parameters such as ROP,
formation pressures, drilling fluid properties and losses are some of the key things to be checked. The
common indication of an abnormality is an increase in torque during drilling, or an over pull
encountered when tripping out of the hole. Proper data collection and interpretation is a sure way to
minimize stuck pipe situations or to quickly solve such situations.

The role of drilling fluids in handling stuck pipe is more important in the prevention of stuck pipe than
in freeing the stuck string or retrieving the fish. Since most stuck strings are due to formation
challenges, it is important to appreciate the type of formation being drilled through, before redesigning
the drilling fluid to combat it. The main types of problem formations are:
1. Erodible formations: These include soft tertiary sequence evaporates, permafrost and some
highly fractured formations; these could also include unconsolidated formations of sand and
gravel. Erosion of these formations may lead to an over-gauge borehole which leads to poor
hole cleaning, one of the causes of stuck pipe. In some cases, if the formation is fractured, this
could lead to slippage and, hence, stuck pipe.
2. Geo-pressured formations: These are formations with pore pressures higher than the
hydrostatic pressure of the drilling fluid. These, if not permeable, tend to cave into the
borehole once drilled through.
3. Dipping formations: These are formations that lie at an angle to the horizontal plane being
drilled through. The challenge with this is if they are plastic (more common in slates and
shales), they end up being mobile and flow into the borehole when drilled through.
4. Reactive formations: These are naturally occurring bentonitic shales which contain clays that
react with the mud filtrate and hydrate. The hydrated shells then fall and swell into the
borehole.

Other formation-related challenges could include collapsing cement blocks from an upper cased
section and green cement (Ava, 2004).

2.5.3 Well control

Abnormal formation pressures are rarely encountered in geothermal systems. However, this still
remains an area that must be watched in order to ensure proper well control. Most times, the drilling
fluid pressure exceeds that of the formation in geothermal drilling. However, when drilling using
aerated fluids, the density of the drilling fluid is reduced, and then the formation fluid pressure exceeds
the drilling fluid pressure.

One of the challenges encountered in well control in geothermal drilling is poor cooling of the well,
leading to high temperatures and, hence, high pressures. This may occur when there is continuous loss
or a sudden total loss of circulation (Moore, 1974).

A key issue in well control is monitoring the drilling fluid to keep losses minimal and keep the drilling
fluid pressure as near to the formation pressure as possible. This requires knowledge of the formation
pressure. One way of estimating the formation pressure is using the d exponent, a dimensionless
number. The d exponent is related to the differential pressure between the drilling fluid and the pore
fluid. This value is used to adjust the drilling fluid density. The d exponent usually increases with
depth but, as the formation becomes over pressured, it will decrease (Moore, 1974). The d component
is given by:

(47)
Report 11 143 Dayan

where R = ROP;
N = Rotary RPM;
W = Force on bit;
d b = Bit diameter; and
dexp = Drilling exponent.

The dexp is corrected for the effect of mud density changes, as well as changes in WOB, bit diameter
and rotary, as follows (Moore, 1974):

(48)

where , the mud density, is equivalent to a normal pore pressure gradient and is the equivalent
mud density at the bit while circulating. The new dmod is then used to calculate the equivalent mud
density at that differential pressure and, finally, the formation pressure from the following equations:
7.65 16.5 (49)
0.052 ∗ (50)
where ρe = Is the equivalent mud density at the bit while circulating;
dmod n = Equivalent mud density at the normal pressure;
dmod abn = Equivalent mud density at abnormal pressure.

Once the formation pressure is determined, we compare it with the fluid density and adjust
accordingly.

3. DRILLING FLUID DESIGN FOR DRILLING A 2500 M GEOTHERMAL WELL

The main interest in drilling a geothermal well is to access the steam resource located several hundred
meters down, or more. To do this, we must drill through sections that are of less significance and may
even block the producing reservoir with colder fluids. The most important part of the well is the steam
zone, normally called the main hole. This is the area believed to be producing the required steam in
the well. Wells normally have four main sections: surface hole, anchor hole, production hole and
main hole. The first three sections are normally cased using steel casings and cemented off to seal off
unwanted aquifers and also to stabilize the area as we drill on. The main hole, which can run from 800
m to 1500-3000 m depending on the well prognosis, is normally set with slotted liners. Figure 6
illustrates a 3000 m well casing design for a regular size well (Hole, 2006). There are many methods
for determining the casing depths. However, as a general rule, the depths are set such that at least 1/3
of the entire well drilled is cased.

A typical well design would include:

Conductor casing: 30” set at a depth of 24 m, either driven or drilled.
Surface casing: 20” casing setting 26” diameter hole, drilled to 100 m depth.
Anchor casing: 13 3/8” casing set in a 17½” hole drilled to 300 m depth.
Production casing: 9 5/8” casing set in a 12¼” hole drilled to 850 m depth.
Open hole: 7” slotted liner set in an 8½” hole drilled to 2500 m – total depth.

Once the well design is done, the next stage is to identify the drilling fluid to be used and prepare an
appropriate drilling fluid program.
Dayan 144 Report 11

3.1 Drilling fluid materials and properties

The final desired property of any drilling fluid will determine the materials it constitutes. As earlier
discussed, the constitution of a drilling fluid is guided by optimising costs and performance; these are
considered optimal when we have good hole cleaning and lowest circulation system pressure losses.
Other issues, such as loss of circulation, are best handled when they arise. But, at the design stage, a
guide is made on how to handle them. To optimise a fluid for hole cleaning and least pressure drop,
we must appreciate the fluid’s rheology, which describes the behaviour of a fluid when a force is
applied to it. Once we characterise this behaviour, we can select what materials to add to our drilling
fluid to improve upon its performance in hole cleaning and minimizing pressure drop.

The behaviour of fluids under force/stress can be characterised as: Newtonian - where the shear rate
increases linearly with shear stress; these fluids include water and oil; non-Newtonian, which exhibit a
non-linear relationship between shear stress and the shear rate. The behaviour of Newtonian fluids is
undesirable in drilling as it leads to higher pressure loss in the system, due to increased viscosity, and
also poor gel strength, which is key in hole cleaning. To correct these problems, various materials can
be added, depending on which property is being corrected. The main properties checked are viscosity,
yield point and gel strength (Darley and Gray, 1988).

Non-Newtonian fluids can be classified into four

different types, depending on the additives. For
geothermal drilling, the main additives are either
bentonite or polymer, which are viscosifiers,
leading to either a Bingham plastic fluid or
pseudo-plastic / power law fluid. These fluids
display a zero shear rate at the initial change in
shear stress, after which the relationship is either
linear or non-linear (Darley and Gray, 1988).
Figure 7 illustrates the behaviour of various
drilling fluids (Darley and Gray, 1988):

As can be seen, Newtonian fluids do not exhibit

the gelling property required. Bentonite-based
mud normally exhibits a Bingham-plastic fluid
model, which has a higher viscosity and high
yield strength; the yield strength is a measure of FIGURE 7: Fluid rheological behaviour
its gelling strength. This is important for (Baker Hughes, 1995)
suspending cuttings when circulation stops. The water-polymer drilling fluid exhibits a power law of
behaviour which has poorer gel strength but lower viscosities. However, increasing the concentration
of polymers, such as starch in bentonite, results in a Dialant behaviour, in which there is an adverse
rise in fluid viscosity (Darley and Gray, 1988).

Another key area in designing a fluid system is its flow regime. A fluid flow can be categorised as:
plug flow, laminar flow, transitional flow or turbulent flow. The fluid’s flow regime affects both the
cutting carrying capacity and pressure losses in the system. In general, laminar flow leads to lower
pressure loss but poorer hole cleaning. Turbulent flow leads to better hole cleaning but higher
pressure losses.

From this we see that any drilling fluid design must consider viscosity, yield and gel strengths, and the
density of the fluid.
Report 11 145 Dayan

3.1.1 Viscosity

Viscosity is a description of the thickness of a drilling fluid and, hence, its resistance to motion. It is
normally measured in centipoises. In the field, it is a common practise to measure viscosity in terms
of funnel viscosity, (secs/qt), which is how long it takes one quart of fluid to pass through the funnel.
This gives a view of how thick a fluid is, but is not used in calculations regarding viscosity. For
Bingham fluids (water-bentonite), we use plastic viscosity, which is the viscosity at which the fluid is
past its yield point. This is measured using a viscometer. Figure 8 shows a Bingham plastic
Rheogram.

Plastic viscosity is calculated once the viscometer readings at RPM of 300 and 600 are taken.

(51)
(52)
where PV = Plastic viscosity, (cP);
YP = Yield point (dynes/cm2); and
, = Dial readings at 600 and 300RPM.

This relationship is true whenever

bentonite is used as the viscosifier.
However, the introduction of other
additives, such as starch polymers,
leads to a more complex drilling
fluid which is best analysed by the
power law model (Baker Hughes,
1995).

Viscosity decreases with an increase

in temperature, but increases with
cuttings content. For geothermal
drilling from various field results, a
plastic viscosity of 10-20
centipoises is found appropriate for
hole cleaning (Finger and
Blankenship, 2010). This can be
varied to optimise between hole FIGURE 8: Bingham plastic rheogram (Ava, 2004)
cleaning and pressure drop.
Efficient removal of solids at the surface is important in reducing the solid contents of a fluid. There
are graphs that illustrate the effect of total solid content on viscosity; these should be used when
designing for viscosity.

3.1.2 Yield point and gel strength

This behaviour indicates a resistance to flow until a certain minimum shear stress, the yield stress of
the fluid, is reached. Every fluid exhibits this behaviour, at least some minimal value. This property
determines the yield and gelling strengths of a fluid, properties which are of great significance in
cutting carrying and suspension. The suspension is more important when circulation stops, so that the
fluid can suspend the cuttings till circulation resumes; this property is not exhibited by water and is
one key area where bentonite plays a major role. Bentonite and other clays, when mixed with water,
result in a Bingham fluid with a higher gelling and yield point, as illustrated in Figure 7. Polymers, on
the other hand, have poorer gelling properties. There is, however, a limit to the useful value of
gelling; very high gel strength implies higher pressures needed to break the fluid to start flowing.
Hence, bentonite is often mixed with starch to reduce its gelling property to acceptable limits.
Dayan 146 Report 11

Increased solid content leads to higher gel and yield strengths; efficient surface removal should be
maintained to reduce this (Skalle, 2011). Yield strengths of 35-125 kPa are common in geothermal
drilling (Finger and Blankenship, 2010).

3.1.3 Density

When designing for fluid density in geothermal drilling, we must appreciate that most geothermal
systems have pore pressures lower than hydrostatic pressure, implying that we need to keep our
density as low as possible. However, low density has two main disadvantages: lower cutting carrying
capacity, and poorer filter cake formation. Filter cake is of high importance in geothermal drilling
since we need to form a slight impermeable layer to reduce fluid loss into the formation during
drilling; loss of drilling fluid is quite costly and can also lead to formation damage (Darley and Gray,
1988).

It is common to use a density ranging from 1.00 to 1.15 g/m3 in geothermal drilling (Finger and
Blankenship, 2010). However, if higher densities are needed, then barite is the most common material
used as a densifier. A good understanding of the formation pressures expected is paramount in
designing for density. The formation pressures can be estimated. Aerated drilling is one significant
way of reducing the density of the fluid. This is employed in drilling the main hole. However,
generally the lower the solids content of the drilling fluid, the lower the viscosity. Therefore, the
surface removal system is key. Table 3 is a summary of the key drilling fluid properties and their
values for geothermal drilling (Finger and Blankenship, 2010).

To achieve the parameters shown in Table 3 and

TABLE 3: Fluid properties
keep within the desired range, a good mixing must
be undertaken and keen monitoring of the fluid
Property Range
maintained. The graph in Appendix III shows the
Density 1.03-1.15 g/m3
relationship of water-based bentonite, density, yield
Funnel viscosity 35-55 Sec
point viscosity and solid contents. Wyoming
bentonite is normally used in geothermal drilling; Plastic viscosity 0.01-0.02 Pa-s
this is because it has the highest yield, hence does Yield 35-125 kPa
not result in a high increase in fluid density, pH 9-10
compared to other clays. This is illustrated in the
graph in Appendix III.

3.1.4 Drilling fluid materials

These are added to the base fluid, water, to achieve specific properties for a given purpose. The main
additives to water are weighting agents, viscosifiers, filtration control, LCMs, and conditioners for pH
control materials. Geothermal fluids are normally simply, constituted of the base fluid-water,
viscosifiers, either bentonite or polymer, and conditioners, usually caustic soda (Ava, 2004) (Table 4).

3.2 Calculation of key drilling fluid parameters

A drilling fluid program forms part of a well’s drilling program and is simply a guide on the
preparation and utilisation of the proposed drilling fluids for the various well sections. It is normally
prepared by the drilling engineer with knowledge of the well’s prognosis. The key part of the drilling
program is determination of the drilling fluid parameters to be used when drilling the various well
sections. This is important for good hole cleaning, minimising pressure drop in the system and
optimising bit hydraulics. Here, we shall assume a regular well with a profile as described at the
beginning of Section 3. The aim is to determine the drilling fluid parameters to be utilised for good
whole cleaning, minimal pressure drop and optimised bit hydraulics. The equations described in the
preceding sections will be applied. The following will be determined:
Report 11 147 Dayan

1. Minimum annulus velocity effective hole cleaning for the various well sections;
2. Minimum flow rate for the required annulus velocity for the various well sections for when
using either mud or water;
3. System pressure drop;
4. Equivalent circulating density for the various well sections;
5. CCI and Ca; and
6. Summary.

TABLE 4: Drilling fluid additives

Type Description/function Material

Densifier These are compounds dissolved or suspended in Commonly used is barite.
the drilling fluid to increase its density; can be
any substance denser than water.
Viscosifiers Improve on the drilling fluid viscosity, hence Mainly used in geothermal is
enhancing drilling fluid’s ability to remove bentonite; polymers are also used,
cuttings from the wellbore and to suspend however they lack gel strength, so
cuttings and weight materials during periods of do not suspend cuttings.
no circulation.
Filtration These reduce the amount of filtrate lost from the
Bentonite, polymers, starches, and
control drilling fluid into a subsurface formation. thinners or deflocculates; all
material function as filtration-control
agents.
Conditioners These are alkalinity and pH-control additives NaOH, KOH, Ca(OH)2 and Mg
which are used to optimize pH and alkalinity in (OH)2.
water-base drilling fluids.
LCM These can be broadly defined to include any In geothermal drilling, the most
material that seals or bridges against permeable common is mica flakes.
or fractured formations to inhibit the loss of
whole drilling fluid.

Assumptions:
1. Mud density of 1150 kg/m3 is used for mud drilling and water density of 1000kg/m3;
2. Turbulent flow is assumed;
3. Drill pipe size of OD 5” is used in the calculations;
4. Aerated drilling commences in the main hole; and
5. The upper well section is drilled using either water or mud; no aerated drilling.

Steps of calculation:
1. Calculation of minimum annular velocity: Section 2.3.1 will be used to calculate the
minimum annular velocity required for effective hole cleaning for the various well sections.
2. The minimum flow rate for the drilling fluid to achieve the minimum velocity is then
calculated.
3. Calculate cuttings carrying index, and cuttings concentration ratio.
4. Calculate the ECD (equivalent circulating density) and BHCP (bottom hole circulation
pressure).
5. The results of these are shown in Tables 5 and 6.
Dayan 148 Report 11

Results and analysis:

TABLE 5: Fluid parameters

Water: MW=1, Mud: MW=1,

Hole Hole Viscosity = 1.17 cP Viscosity = 15 cP
diameter depth Annular Flow Annular Flow CCI Rt Ca
(inches) (m) velocity rate velocity rate
(m/s) (l/m) (m/s) (l/m)
26” 100 0.7 13000 0.48 9500 1.5 0.65-0.78 0.021
17-1/2” 300 0.7 7980 0.5 5700 1.7 0.63-0.7 0.015-0.023
12-1/4” 850 0.7 4650 0.5 3618 1.6 0.67-0.78 0.011-0.017
8-1/2” 2500 0.78 2044 - - 1.6 0.73 0.013

TABLE 6: Fluid parameters

Water: MW = 1, Mud: MW = 1.15,

Hole Hole Viscosity 1.17 cP Viscosity = 20 cP
diameter depth Annular Flow Annular Flow CCI Rt Ca ECD BHCP
(inches) (m) velocity rate velocity rate
(m/s) (l/m) (m/s) (l/m)
26” 100 0.7 13000 0.4 6620 - 0.55- 0.021- - -
0.78 0.053
17-1/2” 300 0.7 7980 0.3 2898 1.4 0.52- 0.015- 9.6 33
0.7 0.055
12-1/4” 850 0.7 4650 0.26 1560 1.3 0.45- 0.011- 8.36 83
0.78 0.058
8-1/2” 2500 0.78 2044 - - 1.3 0.73 0.013 8.59 252

Discussion of tables
From Tables 5 and 6 we see the effects of changes in viscosity and density of mud on the various
drilling parameters. Assuming a fixed
size of the cuttings generated, we can
compare the annular velocity, density and
viscosity to the Rt, CCI, Ca, ECD and
BHCP.

Transport ratio, Rt vs. annular velocity

and density
This is an indication of hole cleaning
efficiency and helps to indicate whether
the annular velocity selected is sufficient
to move the cuttings out of the hole for
trouble-free drilling. As discussed earlier,
this ratio should not be less than 0.5 for
trouble-free drilling. From the two tables,
we notice that an increase in annular
velocity results in an increase in the
transport ratio, well above the minimum
of 0.5, and an increase in fluid density and
viscosity equally results in an increase in
the transport ratio. A plot of annular
FIGURE 9: Graph of annular velocity velocity against transport ratio is shown in
vs. transport ratio Figure 9. The plots show the effects of
Report 11 149 Dayan

density on Rt. It is, however, important to note that this increase in density must not be due to an
increase of solid particles (cuttings) in the fluid, but as a fluid property in itself. This effect of density
is important, more in the upper well sections. Where achieving the high flow rates for the required
annular flow rates is difficult we can, instead, design a fluid with higher viscosity and density to
increase the transport ratio.

Transport ratio vs. cuttings concentration

These two have an inverse relationship
such that the increase in one leads to a
reduction in the other. This creates a
dilemma in selecting the fluid parameters
and properties. From the results, it is clear
that the increase in density and viscosity
results in a reduction in Ca, yet the same
results in an increase in Rt, due to a
reduction in flow rates. Figure 10 shows
plots illustrating this.

Flow rate vs. ECD and BHCP

The flow rate used is dependent upon the
annular velocity and the well cross-
section. If the cross- section is kept
constant, the only variable we have is
annular velocity. The higher the annular
velocity required, the higher the flow rates
needed. The annular velocity, as shown
earlier, depends on the cuttings’ slip
FIGURE 10: Graph of transport ratio
velocity. If we have higher annular
vs. cuttings concentration
velocity and flow rates, we end up with
higher ECD and eventually BHCP. This
is because the pressure losses in the
system are dependent on flow rate, density
and viscosity of the fluid. The higher
these parameters are, the higher the losses
and pumping costs will be. In addition,
higher BHCP implies higher chances of
loss of circulation and formation damage.

3.3 Air and aerated drilling fluid

Aerated drilling is used in geothermal

drilling with the aim of reducing the
formation damage resulting from
increased drilling fluid pressure on the
formation wall. As drilling progresses, the
density of the fluid increases because of
cuttings and other solid content in the
fluid; this increase results in higher BHCP
with depth. Figure 11 shows a typical
well, indicating the drilling window and
the effect of increased mud density
(Vollmar et al., 2013). Introduction of air
in the fluid helps lower the density and FIGURE 11: Drilling window (Vollmar et al., 2013)
Dayan 150 Report 11

subsequently reduces the damage on the formation.

As can be seen in Figure 11, the higher the fluid density, the greater the formation damage. When
drilling the main hole, the tendency is to try and keep the fluid pressure below the pore pressure,
which results in an underbalanced system. The ratio of air to water pumped into the hole, and the back
pressure applied to the ‘exhaust’ or flow line from the well, allow the down-hole pressures in the hole
to be ‘balanced’ with the formation pressure in the permeable zones, thus allowing for the return of the
drilling fluids to the surface and, therefore, maintaining drilling fluid circulation. In fact, the term
‘under-balanced’ drilling, as applied to this form of geothermal drilling, is a misnomer (Hole, 2006).

Several graphs have been developed (Lyons et al., 2009), based on the criteria described earlier in the
hole cleaning section, for selecting the minimum volumetric flow rates for various well profiles. One
such graph is shown in Appendix IV. These can be used in selecting the adequate velocity for the well
profile.

For aerated drilling, we must have knowledge of the formation pressures so that we intentionally
design the drilling fluid to have lower pressure at the desired points. We first design for the minimum
flow rate for optimum hole cleaning, as earlier described. Then we select the appropriate air flow
rates for the system’s mud flow rate at the desired depth. As to whether we achieve the desired
underbalanced drilling can only be assessed by comparing the formation pressures at depth to the
drilling fluid pressure.

4. CASE STUDIES

4.1 Drilling fluid design – Olkaria-Kenya scenario

Olkaria geothermal field is one of the largest geothermal production fields in Kenya, with over 300
MWe installed capacity. Several deep geothermal wells have been drilled in this field. A study of one
of the wells, OW 717, was considered to appreciate the application of drilling fluid design in
geothermal drilling. Well OW 717 is a vertical geothermal well drilled to 3000 m TVD. The well has
a regular profile and was drilled in a total of 45 days. The parameters used for the various well
sections are shown in Table 7 (KenGen, 2012):

TABLE 7: Fluid parameters

Water Mud
Hole Hole Pressure
Annular Flow Annular Flow ECD BHCP
diameter depth loss CCI Ca
velocity rate velocity rate (bar) (bar)
(inches) (m) (bar)
(m/s) (l/m) (m/s) (l/m)
26” 100 - - 0.3 4792 59
17-1/2” 303.5 - - 0.3 3600 76 1.3 0.002 9.6 29
12-1/4” 749.5 0.9 3900 - - 105 3.2 0.003 8.5 84
8-1/2” 3000 2.1 3300 - - 137 7.3 0.018 8.59 252

Drilling upper well section hole

Drilling the 26” hole in Olkaria was done using water-based bentonite. Because of the larger annular
cross-section, this section of the well required higher flow rates in order to achieve the minimum hole
cleaning requirements. It is impractical to achieve high pumping rates of up to 10,000 lpm. The
solution here was to adjust the drilling fluid to higher density and viscosity, hence lowering the flow
rates. This, however, increased the bottom hole pressures encountered. It is common practise to spud
with mud of viscosity of up to 15 cP. This is achieved by mixing high yield bentonite, like Wyoming
bentonite, with water, commonly used in geothermal drilling.
Report 11 151 Dayan

A look at the CCI and Ca achieved with these parameters indicated excellent hole cleaning. The Ca
was well below the maximum allowed for trouble-free drilling, implying an even higher rate of
penetration could be achieved before the limit was reached.

The actual drilling parameters for drilling the 12-1/4” and 8-1/2” sections showed higher flow rates
and velocities than recommended by previous calculations. The effect was a higher BHCP which
could lead to formation damage. The higher pressure losses resulting from the higher flow rates also
implied more fuel being utilised and poor bit hydraulics.

Aerated drilling
Aerated drilling in Olkaria is introduced whenever a large loss in circulation is encountered.
However, the main well section that is designed for aerated drilling is the main hole section, drilled
from the production casing depth to total depth, normally 3000 m. The parameters used are 3300 lpm
as the flow rate of water and 1800 scfm.

Loss of circulation policy

Minor losses encountered when drilling the surface and anchor casing are treated by drilling blind and
spotting mud pills when making connections. Losses encountered in the production hole are handled
by drilling blind if the loss is partial, or switching to aerated drilling in cases of full loss. Plugging is
rarely done to heal losses unless the well is suspected to have collapsing formations. The main hole is
drilled with aerated water and foam.

4.2 Drilling fluid design – Iceland scenario

Table 8 lists the parameters used when drilling part of well RN-19 in Reykjanes.

TABLE 8: Fluid parameters, well RN-19

Hole Hole Water Mud

diameter depth Annular velocity Flow rate Annular velocity Flow rate
(inches) (m) (m/s) (l/m) (m/s) (l/m)
26” 84 - - - -
21” 349 0.8 3600
17-1/2” 746 0.5 4800
12-1/4” 2500 1.0 3900 - -

Drilling the upper well section

The surface hole was drilled to a 26” diameter using an air hammer to a depth of 84 m. The rest of the
upper well section was drilled using mud with flow rates ranging from between 50 to 60 lps in the 21”
hole to 80 lps in the 17-1/2” hole. The flow rates used in drilling the 17-1/2” section were high,
indicating the challenge of drilling using water in large well sections. This can be seen in both the
theoretical example and the Olkaria drilling scenario. The high flow rate implies higher frictional
losses, hence higher BHCP, which leads to formation damage and possible drilling challenges.

Drilling the main hole

In Iceland, unlike Kenya, water is used in drilling the main hole. This was applied in drilling this well
with a flow rate of 3900 lpm, giving an annular velocity of up to 1 m/s. Polymer pills were used in
each connection to reduce challenges with the cuttings.

Loss of circulation policy

In Iceland, minor losses encountered during drilling are handled by drilling ahead with water and
spotting with polymer or bentonite pills in every connection, when drilling the upper well section.
Dayan 152 Report 11

They can also be handled by using LCMs, such as mica flakes, introduced into the drilling fluid.
Major losses encountered while drilling the upper well sections are healed by plugging, using cement.

4.3 Comparison and remarks

The key areas to be noted in drilling fluid design are hole cleaning, pressure losses and bit hydraulics.
These, when optimised, will highly increase chances of successful drilling at lower drilling fluid costs.
From the scenarios studied, the parameters calculated using the minimum annular velocity for hole
cleaning resulted in the least pressure losses. This is because minimal flow rates were considered.
The low pressure implies less BHCP which reduces formation damage in cases where drilling through
a low pressure formation. The low pressure losses also imply less pumping power, hence saving on
costs for drilling fluid.

In geothermal drilling, it is important to design a fluid utilising minimum flow rates. As shown above,
the increase in pressure loss is highly dependent on the flow rates used, but it is advisable to begin
with the minimum required rates for good hole cleaning, and advance to higher rates in case
challenges are noted.

The two scenarios studied in Kenya and Iceland drilling used the actual field parameters used in
drilling at the different sites. They both indicate parameters close to the theoretical minimum
required, apart from in the main hole sections where much higher velocities were used. Whereas these
all ensure excellent hole cleaning, as illustrated by the CCI and the Ca, the higher flow rates result in
higher pumping pressures, which implies more fuel consumption. The higher pressure losses in the
annulus also result in higher BHCP, which may lead to formation damage.

5. CONCLUSIONS

Fluid design is a wide and complex field of study. However, the key issues pertain to good hole
cleaning and protection of a well’s formation. Application of the hole cleaning principles gives good
criteria for optimising the fluid system. It is important to keep the flow rates as near to the theoretical
ones as possible to ensure optimal performance. Prior to any fluid design, the engineers must
appreciate the formation profile, as well as the anticipated pore pressures and fracture pressures of the
well. This will help in understanding the drilling window and in designing the mud. In general, in
geothermal drilling, the tendency is to utilise fluid with densities as close to that of water as possible.
Using annular velocities of 0.3 – 0.4 m/s for mud and 0.6 – 0.7 m/s for water is appropriate for good
hole cleaning. A further study analysing the fluid parameters used in various wells and the formation
pressure would be useful to further appreciate the application of the hole cleaning principles described
herein.

ACKNOWLEDGEMENTS

I would like to express my deepest gratitude to the Government of Iceland and the United Nations
University Geothermal Training Programme for according me the opportunity to undergo this training.
My sincere gratitude to the UNU Geothermal Training Programme director, Mr. Lúdvík S. Georgsson,
the deputy director, Mr. Ingimar G. Haraldsson, and the entire UNU-GTP staff and lecturers for their
great support and assistance throughout the training.

I would also like to extend my gratitude to my employer, KenGen, under the directorship of Mr.
Albert Mugo, for granting me the opportunity to come and study. My deepest gratitude to Mr. Sverrir
Report 11 153 Dayan

Thórhallsson of ÍSOR for good guidance and sharing of valuable knowledge and experience
throughout my training. I would like to sincerely thank my supervisor from Mannvit Engineering, Mr.
Thóroddur Sigurdsson, for his good guidance and technical support throughout my project work.

Special thanks go to all my family members, especially my wife, son and daughter for their moral
support.

Above all, I would like to sincerely thank the Almighty God for good care, protection and guidance
during the six months training in Iceland.

REFERENCES

Ava, 2004: Drilling fluids manual. AVA S.p.A, Rome, Italy, 483 pp.

Azar, J.J., and Samuel, R.G., 2007: Drilling engineering (1st ed.). Pennwell Corp., Tulsa, OK, United
States, 486 pp.

Baker Hughes., 1995: Drilling engineering workbook: A distributed learning course (1st ed.). Baker
Hughes INTEQ, Houston, TX, United States, 395 pp.

Bourgoyne, A.T., Millheim, K.K., Chenevert, M.E., and Young, F.S., 1991: Applied drilling
engineering (2nd ed.). Society of Petroleum Engineers, Richardson, TX, United Sates, 502 pp.

Chemwotei, S.C., 2011: Geothermal drilling fluids. Report 10 in: Geothermal training in Iceland
2011. UNU-GTP, Iceland, 365-390.

Darley, H.C.H., and Gray, R.G., 1988: Composition and properties of drilling and completion fluids
(5th ed.). Gulf Publishing Company, Houston, TX, United States, 643 pp.

Finger, J., and Blankenship, D., 2010: Handbook of best practices for geothermal drilling. Sandia
National Laboratories, Albuquerque, NM, United States, report SAND2010-6048, 84 pp.

Gabolde, G., and Nguyen, J.P., 2014: Drilling data handbook (9th ed.). Editions Technip, Paris, 576
pp.

Guo, B., and Liu, G., 2011: Applied drilling circulation systems. Hydraulics, calculations and models
(1st ed.). Gulf Publishing Company, Burlington, VT, United States, 262 pp.

Hole, H., 2006: Lectures on geothermal drilling and direct uses. UNU-GTP, Iceland, report 3, 32 pp.

Hole, H., 2006: Aerated fluids for drilling of geothermal wells. Lecture 1 in: Lectures on geothermal
drilling and direct use. UNU-GTP, Iceland, report 2006-3, 1-12.

KenGen, 2012: Well OW-717 completion report. KenGen – Kenya Electricity Generating Co., Ltd.,
Kenya, unpublished report, 22 pp.

Kipsang, C., 2013: Cost model for geothermal development. Report 11 in: Geothermal training in
Iceland 2013. UNU-GTP, Iceland, 365-390.

Lapeyrouse, N.J., 2002: Formulas and calculations for drilling production and work over (2nd ed.).
Gulf Publishing Company, Burlington, VT, United States, 224 pp.
Dayan 154 Report 11

Lyons, C.W., Guo, B., Graham, L.R., and Hawley, D.G., 2009: Air and gas drilling manual (3rd ed.).
Gulf Publishing Company, Burlington, VT, United States, 308 pp.

Mitchel, R.F., and Miska, S.Z., 2011: Fundamentals of drilling engineering (1st ed.). Society of
Petroleum Engineers, Richardson, TX, United States, 696 pp.

Moore, P.E., 1974: Drilling practices manual. The Petroleum Publishing Company, Tulsa, OK, 228
pp.

Rahimov, E., 2009: Through tubing rotary managed pressure drilling. University of Stavanger,
Norway, MSc thesis, 84 pp.

Rehm, B., Haghshenas, A., Paknejad, A., Hughes, J., and Schubert, J., 2012: Underbalanced drilling
limits and extremes (1st ed.). Gulf Publishing Company, Houston, TX, United States, 629 pp.

Skalle, P., 2011: Drilling fluids engineering (1st ed.). Pal Skalle & Vervus Publishers, APS, 132 pp.

Thórhallson, S., 2014: Geothermal drilling technology. UNU-GTP, Iceland, unpublished lecture
notes.

Vollmar, D., Witting, V., and Bracke, R., 2013: Distributed parameters models. Geothermal drilling
best practices: the geothermal translation of conventional drilling recommendations - main potential
challenges. In: Bracke, R., Harvey, C., and Rueter, H. (editors), Geothermal exploration best
practices. IGA-Academy, report 0104-2013, 12 pp.

APPENDIX I: Friction factor for calculating Particle slip velocity (Mitchel and Miska, 2011)
Report 11 155 Dayan

APPENDIX II: K factor for power law fluid (Mitchel and Miska, 2011)

APPENDIX III: Effects of clay concentration on viscosity of fresh water

(Darley and Gray, 1988)
Dayan 156 Report 11

APPENDIX IV: Direct circulation minimum volumetric flow rates

(Mitchel and Miska, 2011)