GEOTHERMAL TRAINING PROGRAMME Reports 2010

Orkustofnun, Grensásvegur 9, Number 20

IS-108 Reykjavík, Iceland

CONTROLLED DIRECTIONAL DRILLING

IN KENYA AND ICELAND

Thomas Ongau Miyora

GDC South Rift - Geothermal Development Company, Ltd.
P.O. Box 1453-20117
Naivasha
KENYA
tmiyora@gdc.co.ke

ABSTRACT

Directional drilling is the most widely used method for drilling geothermal wells
due its various advantages. Drilling multiple wells from the same pad allows for
fewer rig moves, less surface area disturbance as well as making it easier and
cheaper to exploit the resource being drilled for. Current technology allows the
driller to steer the well to the target with high precision and this allows exploitation
of resources that would otherwise be difficult or impossible to reach. The study in
this report shows that almost 50% of the total time in directional drilling is spent on
activities that are not related to actual cutting of the formation by the drill bit.
Minimising time spent on these activities will reduce total drilling time per well
and, hence, reduce drilling costs. This study highlights the calculations of well
trajectory and an analysis of the actual time taken for all activities in drilling of 12
directional wells in Kenya and 14 directional wells in Iceland. The results show
that the average depth drilled per day for Iceland is about 56 m, and for Kenya it is
about 48 m. The average depth of the Icelandic wells is 2379 m, taking about 41
days to drill, and the average depth for Kenyan wells is 2830 m, taking about 58
days. Comparison of drilling times in Iceland and Kenya indicates that for a well
of 2830 m, it will take about 54 days to drill in Iceland while it will take about 58
days to drill in Kenya. These drilling rates are similar, although the difference of 4
days is significant considering the large costs involved in drilling per day.

The government of Kenya has started an ambitious programme to increase its

power production from geothermal energy. The capacity of geothermal power in
Kenya is estimated to be about 7000 MWe, valued at about USD30 billion.
Efficient drilling practices will be required in order to optimise exploitation of its
more than 14 prospects. The Geothermal Development Company’s ten year
business plan indicates that it intends to drill about 500 wells in Olkaria, Menengai,
Silali and four other fields and realise at least 2000 MWe by 2019. Within this
period it intends to procure 12 deep drilling rigs with a depth capacity of 5000-
6000 m. The rigs should have a capacity to drill deep directional geothermal wells.
The study carried out in this report indicates what has been achieved in the past and
will act as a benchmark for planning for future drilling, as it addresses good
practices that can assist in improving cementing, logging and bit selection. Also,
the computer programs generated in this report will be used in monitoring actual
drilling to ensure the target is reached within acceptable limits.

365
Miyora 366 Report 20

1. INTRODUCTION

Directional drilling is a special drilling operation used when a well is intentionally curved to reach a
bottom location (Vieira, 2009). Directional wells are drilled in different patterns with the inclined
angle of the wells varying from a few degrees to more than 90°. The shapes of the trajectory of
directional wells also vary depending on the position of the target to be reached. The government of
Kenya, through the Geothermal Development Company Limited (GDC), has started a very ambitious
programme of electrical power generation from geothermal. GDC’s ten year business plan and
strategy is to generate at least 2000 MWe of electricity over the next ten years (2009-2019), and at
least 4000 MWe by 2030 through an accelerated development plan (GDC, 2009).

During this period GDC plans to procure

12 deep drilling rigs for the new fields
and hire three more rigs for drilling in
Olkaria. There will be extensive
exploration and appraisal drilling for all
fields to mitigate initial risks. There are
plans to undertake early generation by
investment in well head generation units
which will produce electricity and be
connected to the national grid, as a
preliminary production before the large-
scale power plant.

Directional drilling will reduce the

capital investment cost of the wellhead
generation unit by making it possible for
multiple wells to be drilled on a single
pad, hence reducing the temporary pipe-
work required for a sizeable unit (Ngugi,
2002). GDC has already acquired two
large rigs that are rated with
1,000,000 lbs nominal hook load with a
depth rating of 5000-6000 m.

There are more than 14 high potential

geothermal fields along the Kenyan rift
valley. The estimated generation
potential of these fields is more than
FIGURE 1: Geothermal fields in Kenya 7000 MWe (GDC, 2009). Figure 1
(adapted from GDC, 2010) shows the location of these fields.

2. LITERATURE REVIEW

2.1 Brief history of directional drilling

Initially all wells were drilled vertically downwards. Directional drilling evolved out of the need to
drill wells in other directions. The most common need was a fish in the hole that was impossible to
recover. Instead of abandoning the well, drilling around and bypassing the fish was adopted since it
was cheaper rather than losing the well. Other reasons that led to the development of directional
drilling is the discovery of some crooked holes previously thought to be vertical holes and also the
need to drill into more productive areas under adjacent acreage where ownership may have been in
Report 20 367 Miyora

question. Almost all oil and gas wells drilled offshore are directional. Geothermal wells, both high-
and low-temperature ones, are also very commonly directionally drilled.

The history of directional drilling can be traced back to 1895 when wells were drilled at angles or
curved for the purposes of sidetracking equipment stuck in the hole. After the invention of
inclinometers and the survey of many wells, which were assumed to be vertical, it was noted that they
were actually directional (Vieira, 2009). However, formal directional drilling is thought to have begun
in the 1930s in California when a drilling contractor drilled a slanted well from land into a reservoir
that was offshore (Vieira, 2009; Inglis, 1987).

The technology of drilling directional wells has dramatically improved since the 1930s when it was
first officially documented that a contractor drilled a directional well in Huntington Beach, California
(Vieira, 2009) and in southeast Texas where a directional well was drilled to reach a blown out well at
a point near the bottom. Fluid was pumped through the directional well into the formation, stopping
the blow out (Short, 1993). The general principle of drilling a directional well to a given direction is
to point the drilling bit in that direction. This may be achieved by using different tools and equipment
which will be discussed later on.

2.2 Types/patterns of directional wells

There are different types of profiles for

directional wells. The main profiles
commonly adopted for directional
drilling in the oil industry are listed
below (Figure 2).

(i) Build and hold: This type of profile Type 1 Type II Type III Horizontal Slant
Build and hold Build, Deep KOP well well
may also be referred to as type I profile hold and & Build
or J-profile. It is used for moderately Drop
deep wells where the oil bearing rock is FIGURE 2: Well profiles (based on Inglis, 1987,
in a single zone and no intermediate and Short, 1993)
casing is required. It is the pattern most
commonly adopted for directionally drilled geothermal wells. The well has three sections, i.e. the
vertical section from the surface to the kick off point (KOP), the curved section where the well angle
is built up to the planned inclination and finally the inclined section drilled to the target.

This type of profile is applied when large horizontal offsets are required. The casing is done through
and over the curved section of the well trajectory. The angle of inclination varies from 15-55° (Inglis,
1987), but most geothermal wells have an inclination of 20-45°.

(ii) Build hold and drop: This type of profile may also be referred to as type II profile or S-profile. It
is similar to type I but the difference comes when the angle of the inclined section is reduced, often
drastically, to become vertical as the target is reached.

This type of profile is applied in several cases, such as, when completing a well that intersects multiple
producing zones; as a relief well, parallel to a wild well, for the purpose of quenching the reservoir;
and finally it is used for accurate bottom hole spacing where many wells are drilled from the same
surface location as in offshore drilling (Vieira, 2009).

(iii) Deep kickoff and build: This profile is also referred to as the type III profile. It is used when
drilling away from an obstacle such as when sidetracking to drill away from a lost-in-hole fish. Deep
kick off poses challenges in that deflection may be less responsive due to hard formations. Due to
Miyora 368 Report 20

great depth, more time is required to trip and change the bottomhole assembly (BHA) for deflecting
the well. The angle build up of this profile may be difficult to control.

(iv) Horizontal wells: This type of profile has an angle of inclination reaching 90° within the
reservoir. The purpose of this profile is to improve production due to unfavourable factors such as low
permeability. Many formations contain oil and gas but produce low volumes from vertical and
directional wells because of low permeability. Horizontal wells have increased flow rates because of
the increased flow area.

(v) Slant holes: Slant wells start at an angle (30-45°) from the surface and then are drilled by slant-
hole rigs. Slant holes are characteristically shallow, reaching depths of about 1200 m (true vertical
depth) and 1800 m (measured depth). Slant holes are drilled by first pointing the drill string in the
correct horizontal direction toward the target. Then it is raised 30-45° from the vertical. The general
design of the pattern and casing strings is similar to other directional holes with allowances made for
the angles and tubular compression due to the pull-down system.

2.3 Directional drilling for petroleum

The most common application of directional drilling is in the development of offshore oil and gas
reservoirs. This is due to the single site requirement. It is more economical to drill many directional
wells from one platform than it is to build a costly platform for each vertical well (Short, 1993).
Platforms may take 2-3 years to construct and position which means an equivalent time delay before
production can begin. A single large platform can support more than 50 wells (Inglis, 1987).

Modern directional and extended reach techniques may drill into large areas containing oil and gas
from one surface location. A vertical well penetrates the reservoir at one point. Directional drilling
increases coverage substantially, depending on the angle of inclination. Increased coverage of the
reservoir increases production per well (Short, 1993).

2.4 Directional drilling for geothermal energy

Geothermal energy is utilised in certain areas of the world with good geothermal gradients found in
some rocks, often associated with volcanic intrusions. The heat is mined from the rocks through
drilling. The source rock is generally impermeable except for near-vertical fractures. Extracting heat
from this rock requires drilling wells. The wells are directionally drilled to increase the likelihood of
intercepting the vertical fractures.

2.5 Benefits of directional drilling

Drilling directional wells will cost about 22-41% more and may increase the cost of a 64 MWe project
by 2.8-6% if well productivity does not improve (Ngugi, 2002). In spite of these cost implications,
directional drilling is still preferred because of its benefits as discussed below.

(i) Better chances of intersecting near vertical targets

(ii) Side tracking: May be necessitated by a fish in the hole; a re-drill/re-completion where a well was
drilled in an unproductive part of the reservoir and did not locate the anticipated target; the hole can be
plugged and the well sidetracked towards a new target.

(iii) Drilling to avoid geological problems: This is mostly applicable in petroleum reservoirs.
Petroleum reservoirs are sometimes associated with salt dome structures that may be directly in the
Report 20 369 Miyora

path of the well. Drilling through the structure may pose serious corrosive problems later on. In this
situation it is prudent to avoid the salt formation by drilling a directional well. In case of a blowout
that makes it impossible to cap a well from the rig, relief wells (usually 2) are drilled directionally and
controlled to reach targets less than 3 m from the blown out well at the subsurface; this helps kill the
blowout.

(iv) Drilling beneath inaccessible locations: This may be a result of manmade or natural obstructions
such as buildings, lakes or mountains. Rigging up over these sites is not possible and the only way to
reach the resource is by drilling directional wells.

(v) Offshore development drilling: Offshore drilling requires the construction of drilling platforms
that are either fixed to the sea bed or floating on the sea. Drilling vertical wells from each platform
would not be economically viable. The normal practice is to construct a permanent platform from
which more than 50 wells can be directionally drilled (Inglis, 1987).

(vi) Environmental considerations: Directional drilling helps conserve the environment by causing
less surface disturbance since several wells can be drilled from the same pad. Vertical wells require a
new drilling pad for each well drilled, hence more disturbances. Also, directional drilling makes it
possible to use less extensive steam gathering pipe work, thus minimising land requirements.

3. DIRECTIONAL WELL PLANNING

There are many factors to be considered when planning the drilling of directional wells; these are
discussed here below.

3.1 Reference coordinates

Directional drilling involves measurements of depth, inclination and azimuth. These measurements
have fixed references so that the course of the well trajectory can be calculated. In directional drilling,
these references mainly refer to depth reference, inclination reference and azimuth reference.

For depth measurements, they could either be Measured Depths (MD) or True Vertical Depth (TVD).
MD is the actual depth along the well bore whereas TVD is the vertical distance from a reference to a
point on the well bore. Usually the rotary table is used as the reference. Depths may be given with
reference to the Below Rotary Table (BRT) or the Rotary Kelly Bushing (RKB). For inclination
measurements, the vertical reference is the direction of the gravitational vector which can be indicated
by a plumb bob. Azimuth reference can be one of the following: magnetic north, true geographic
north or grid north. These references are discussed in more detail below.

(i) True geographic north (Meridian direction): The direction is with reference to the geographical
North Pole. In maps, the direction is shown by meridians of longitude.

(ii) Magnetic north (compass direction): The compass gives a direction referenced to magnetic north.
The position is time dependant and varies accordingly. Correction from magnetic north to true
geographic north varies geographically.

(iii) Grid north: The earth’s surface where drilling occurs is curved but maps represent it as a flat
surface. For the purpose of planning a directional well, it is convenient if the curved surface of the
earth is projected onto a flat surface on which maps can be drawn. One such system is known as the
Universal Transverse Mercator (UTM). This is a projection of the section of the earth’s surface that
contains the area of interest. When projections are done, there is a distortion of axes such that the
Miyora 370 Report 20

UTM north is offset slightly from true north. Over large distances this offset can bring significant
differences and must be corrected for when converting coordinates from one system to another.

True North The most common method of fixing the position of a point
GN GN GN N GN GN GN on the earth’s surface is to give its latitude and longitude.
Latitudes run parallel to the equator and are denoted by a
number of degrees 0-90° north or south of the equator.
Longitudes are perpendicular to the equator and they pass
through the North Pole and South Pole and are denoted by
Central Meridian

a a
a number of degrees 0-180° east or west of the Greenwich
meridian.

The relationship between True North and Grid North is

0
indicated by a convergence angle. This is the angle
difference between grid North and True North for the
FIGURE 3: Relationship between location on the surface of the earth being considered.
true north and grid north Figure 3 shows the relationship between true north and
(Baker Hughes INTEQ, 1995) grid north. GN refers to grid north and ‘a’ refers to the
convergence angle.

Baker Hughes INTEQ (1995) states that the world is divided into 60 zones under UTM. The division
is based on reference meridians which are 6° apart. The zones are numbered 0-60. The zones are
further subdivided into grid sectors by latitudes. Each grid covers 8° latitude starting from the equator
and ranging from 80°South to 80°North. The sectors are given letters ranging from C to X, excluding
I and O. This means that a surface on the earth can be identified by a unique number and letter with
reference to a given sector.
Grid North
Coordinates in UTM are measured in Central Meridian
0
metres. North coordinates are measured
from the equator. For the northern 550027.0E
hemisphere, the equator is taken as 0.00 m 6253.194N
north, whereas for the southern hemisphere
the equator is 10,000,000 m north (to avoid
negative numbers). East coordinates for
each sector are measured from a line
500,000 m west of the central meridian for
that sector. In other words, the central 500,000m E
meridian for each zone is arbitrarily given
the coordinate 500,000 m east (to avoid
negative numbers). UTM coordinates are
always Northings and Eastings, and are
always positive numbers (Baker Hughes To the Equator
INTEQ, 1995). Figure 4 shows how
coordinates for a point are given. FIGURE 4: Northings and Eastings
(Baker Hughes INTEQ, 1995)

3.2 Targeting

The target is the main objective of drilling. The size of the target will determine the cost of hitting the
target because of the time spent in reaching it. The smaller the target zone, the greater the number of
correction runs necessary to ensure the target is intersected, consequently higher drilling time and cost.
To save on time and cost, the geologist or reservoir engineer should give as large a target zone as
possible. This will allow the driller to place the well trajectory within the target zone at minimum
cost.
Report 20 371 Miyora

3.3 Other adjacent wellbores

Directional drilling requires drilling several wells within the same pad; they may be separated by small
distances. In petroleum drilling, the separation distance from one well to another may be as small as
2-4 m. This is due to space constraints and the need to drill as many wells as possible. In geothermal
drilling, there may be a need to drill 3-6 wells from the same pad and, in some circumstances, the
separation distance from one well to the next may be quite small. Under such circumstances, precise
control is required and great care must be taken to avoid collisions under the surface. As each new
well is drilled, the separation distance of all adjacent wells must be calculated.

Survey results for the wells can be plotted using a computer program to aid in 3D visualisation of the
trajectories. This will help the driller to nudge a well being drilled away from the existing ones, thus
avoiding collision. Sometimes this may mean drilling away from the target; when the well is a safe
distance away, it may be directed back towards the target.

3.4 Kick off point (KOP)

The KOP is the point on the well trajectory where deviating or sidetracking begins. The KOP should
be at least 30 m below the bottom of the last casing in the hole, preferably 60 m or more, especially
below surface or shallow intermediate casing. This reduces the risk of excessive casing wear or
splitting the casing shoe (Short, 1993). When sidetracking, the KOP should be at least 15 m,
preferably 45 m, above the fish. This will prevent drilling back into the fish or re-entering the old
hole.

Choosing the point of the kickoff requires knowledge of the type of formations at that point. Very soft
formations may increase the difficulty in deviating and building up the angle. Such formations may
not have sufficient strength to provide the reaction force on the fulcrum of the directional assembly.
Thus, the drill string may partially enter the wall of the hole. Very hard and abrasive formations are
difficult to drill. The deviation BHA is less rugged with less weight on the bit. This reduces the
action of the bit on a formation, hence increasing the time spent deviating. The use of mud motors
allows fast drilling. It is important to avoid very soft, very hard, abrasive or laminated formations.
The KOP should be selected for medium soft or medium drillability massive formations when possible
(Short, 1993).

3.5 Build-up Rate (BUR)

This is the rate of angle build-up per drilled length. This is usually expressed in degrees/30 m. If the
change of angle build-up is too rapid, severe dog-legs can occur in the trajectory. These sharp bends
may prevent the drilling assemblies from passing through and cause more wear on the drill string. Too
slow a build-up may result in a long interval for the trajectory to reach the required inclination.
Commonly used build-up rates range from 1.5 to 2.5° per 30 m (Inglis, 1987).

3.6 Dog leg severity

In directional drilling, a dog leg refers to an abrupt change in the hole angle or direction that causes
sharp turns in the well bore trajectory. It is detected by increased torque and drag on the drill string.
The size of a dog-leg can be calculated using survey results over 20-30 m intervals, using the
following mathematical equation. Gabolde and Nguyen (2006) give the formulas shown in Equations
1 and 2:
cos cos cos sin sin cos (1)
Miyora 372 Report 20

where βn = Dog leg angle at station n; and

θn, ϕn = Inclination and azimuth, respectively, at station n.

Dog leg severity is expressed as the change in angle per 30 m drilled.

30 (2)
∆

where DLSn = Dog leg severity for segment n; and

∆Ln = Segment length between stations n and n+1.

Severe dog-legs cause the drill string to undergo cyclic loading because of the pipe bends containing a
dog leg. On the inside of the bend, the wall of the pipe will be under compression and, on the opposite
wall, there will be tension. The loading is reversed when the pipe is rotated 180°. This will encourage
fatigue and reduce the operational life of a drill pipe.

3.7 Directional well design

There are several parameters to

consider when designing a
directional well. These are surface
coordinates, target coordinates,
true vertical depth of target, depth Z0 KOP
to KOP, and the build-up rate.
Geologists usually give the target TVD1
in terms of inclination and azimuth
and allowable margins. The driller TVD2
will then make the trajectory 1
TVD3
design for the wellbore to hit the
TVDn Segment 1 = ΔL1
target. Below is a typical Z1
calculation. Formulas were θ1 2
derived to generate a computer h1 Segment 2 = ΔL2
programme for the visualisation of 3
Z2 θ2 Segment 3 = ΔL3
the trajectory. h2 n
Segment n = ΔLn
Z3 θ3
Figure 5 shows a directional well h3
profile. The various parameters in Zn θn hn
the diagram are explained as
follows: FIGURE 5: A directional well profile

Segment n, ΔLn = Distance along the trajectory between stations n and n+1;
Z0 = Vertical depth to KOP;
Zn = Vertical depth between stations n and n+1;
TVDn = True vertical depth to station n+1;
hn = Horizontal distance between stations n and n+1;
Hn = Resultant horizontal distance from the vertical axis to station n+1;
θn = Angle of inclination for the well over segment n; and
1, 2, 3, ..., N = Survey stations.

(i) Calculation of true vertical depth (TVD) and horizontal displacement for the segments
The results of a directional survey yield the inclination and azimuth of the trajectory at a given depth.
This survey must be analysed to calculate the actual position of the survey station with respect to the
surface location. Various kinds of geometrical models have been used, with each model generating a
Report 20 373 Miyora

number of equations. The accuracy of a given model depends on how close it comes to the actual
trajectory. For consistency, however, it is important to adopt one model for planning and monitoring
wells drilled from the same pad. The geometric models commonly used to analyse and calculate the
survey results are the tangential method, the balanced tangential method, the average angle method,
the radius of curvature method and the minimum curvature method (Inglis, 1987). In this paper, the
method used is the average angle method.

Referring to Figure 5, the first station surveyed is 2. But the angle θ1 at KOP is 0. Since the angle is
gradually increasing, we use the average of two stations, i.e. the upper and lower ends of the segment
to calculate the TVD for that segment. This makes it more representative for the entire segment. For
example, to calculate Z1 we use the average of θ1 and θ2. The well is assumed to be vertical down to
the KOP. By trigonometry,

∆ cos (3)
2

∆ sin (4)
2
This is repeated for each segment. In general, therefore, vertical depth and the horizontal
displacement for a given segment n is given as:

∆ cos (5)
2

The true vertical depth for the well at a given station will be the summation of vertical depths for the
segments up to that station. Thus, for the nth segment, TVD will be given as:

(6)

(ii) Calculation of northings, eastings and resultant horizontal displacement

Figure 6 shows a horizontal plane with the main directional axes crossing at the KOP. 2 refers to the
first station of survey and 3 to the second station of survey. Parameters are as follows:

φn = Azimuth at station n;
nn = Displacement to the northern direction (northings);
en = Displacement to the eastern direction (eastings); and
hn = Horizontal displacement between two consecutive points of survey.

Consider a well with azimuth ϕ1 at the KOP

and ϕ2 at the first station of survey. The
3 e2
h2
distances shown in Figure 6 are measured
from the stations along the main directional
e1 axes in the plane. The same case of using
2 h1 n1 the average of the angles between two
consecutive sections applies. Using
1
trigonometry, n1 and e1 will be given as:
KOP (1)

cos (7)
2

FIGURE 6: Vertical view of the well profile sin (8)

2
Miyora 374 Report 20

The northing and easting displacements for a given station will be the summation of all displacements
behind that point, as shown in Equations 9 and 10. For northings, DNn refers to the total displacement
in the northern direction at station n+1:

(9)

For eastings, DEn refers to the total displacement in the eastern direction at station n+1:

(10)

The resultant horizontal displacement Hn at station n is a vector quantity which has both magnitude
and direction. The magnitude is given by Equation 11:

(11)

4. CALCULATIONS AND VISUALISATION OF SELECTED WELLS

In this report, Equations 3-11 were used to generate a computer program in Excel that calculates all
the parameters at any given point of the wellbore trajectory, based on survey results. To help in
visualising the wellbore trajectory as it was being drilled; graphs were drawn to show the
designed/planned trajectory and the boundaries. The drilling of well OW-35A was used as an
example. The boundaries define the limits within which the actual wellbore trajectory must stay in
order to intersect the target.
The program Grapher 7 was
used to visualise the wellbore
trajectory in both two
dimensions (2D) and three OW-35A
dimensions (3D). 2D graphs OW-36A
help visualise the actual OW-37A
trajectory in relation to the
planned trajectory and
Elevation(masl)

boundary limits. 3D graphs

help to visualise the actual
trajectory with respect to
existing adjacent trajectories.
The resultant graphs from the
tables generated by the
computer programme are
shown in Figures 7, 8 and 9.
Figure 7 shows a 3D plot of
three well bore trajectories
near each other, while
Figures 8 and 9 show the
actual wellbore trajectory of Northings(m)
well OW-35A against the Eastings (m)
planned trajectory and the
boundaries. FIGURE 7: 3D view of three wells in Kenya
Report 20 375 Miyora

0 1000
Side view of well trajectory Vertical view of well trajectory

Designed trajectory
Designed trajectory Upper & lower limit 800
Actual trajectory Actual trajectory
Upper & lower limit boundary

1000

True vertical depth (m)

600

Northings(m)
400

2000

200

3000 0

3000 2000 1000 0 -1000 -800 -600 -400 -200 0

Horizontal displacement (m) Eastings (m)

FIGURE 8: Side view of the well profile of FIGURE 9: Vertical view of the well profile of
the directional well, OW-35A, showing the the directional well OW-35A, showing the
planned trajectory and boundary limits planned trajectory and boundary limits

5. EQUIPMENT FOR DIRECTIONAL DRILLING

5.1 Equipment and instruments for changing course

For directional drilling to be possible, tools are needed that will enable a change in the course of the
well bore from vertical to the desired direction. The principle is to orient the drill bit in the direction
required at the point of kick off. Although the bent sub and the drill bit can be turned at the surface to
the right direction of drilling, this needs to be confirmed by surveying after the drill string has been
tripped into the hole.

One important aspect to be carefully considered and set is the tool face. The tool face is the direction
in which the drill bit tends to drill. It is important to remember that the direction of the tool face is not
necessarily the same as the required direction of the well trajectory. There are different types of tools
for deflecting the hole. Vieira (2009) identifies the following five factors that determine the choice of
a particular tool:

(i) Degree of deflection needed;

(ii) Formation hardness;
(iii) Hole depth;
(iv) Temperature;
(v) Presence or absence of casing; and
(vi) Economics.

In this report, discussion will be limited to two tools for changing the course: the downhole motor
and the bent sub.
Miyora 376 Report 20

(i) Downhole motors

Downhole motors can also be referred to as positive displacement motors
(PDMs) or mud motors. Downhole motors can be used for drilling both
vertical and directional wells. The motors use the pressure and volume of the
drilling fluid to rotate the bit. In the past, downhole motors were used in
combination with bent subs attached to the top of the motor (Figure 10).
Modern mud motors have a bend on the lower end of the motor that allows
drilling in “sliding” and rotating mode. In sliding mode, the drill string is
locked and does not rotate, thus allowing the bit to build up the angle. In
rotation, the whole drill string is rotated, thus adding to the mud motor’s
speed of rotation. Then the drill bit, facing a different direction, is rotated and
thus drills straight.
FIGURE 10: Mud
(ii) Bent subs
motor with a bent sub
Deflection of the hole when using a mud motor is achieved by a special sub
(Inglis, 1987)
placed above the motor to create a side force at the bit. The bent sub is a
short length of drill collar which is about 0.6 m long. The axis of the lower
pin connection is machined slightly off vertical. Figure 11 shows a bent sub Muleshoe
(below), orienting sub (middle) and the muleshoe stinger which Stinger
accommodates the measuring instrument. The pin is inclined to angles that
vary from ½ to 3° depending on the rate of angle build up needed. The bent
sub causes the motor and the bit to drill in a specific direction depending on
the tool face. The amount of deflection is controlled by the amount of offset Key
angle on the pin of the bent sub, the stiffness of the motor and the hardness of
the formation.

5.2 Surveying equipment without measuring while drilling (MWD)

systems

In directional drilling, it is of paramount importance to know the position of

the hole as drilling progresses. Both the drift angle and the direction
(azimuth) must be determined at various depths to compare the actual
trajectory to the planned trajectory of the well. The information gathered
from these measurements will help in monitoring and making appropriate
changes to keep the trajectory on course. Inglis (1987) states that the
objectives of directional surveying are: Offset

(i) To monitor the actual well path as drilling continues to ensure that the
target will be reached;
FIGURE 11: Bent
(ii) To orient deflection tools in the required direction when making
sub, orienting sub
corrections to the well path;
and muleshoe
(iii) To ensure that the well being drilled is in no danger of intersecting an
(mod. from
existing well nearby;
Inglis, 1987)
(iv) To determine the true vertical depths of the various formations
encountered for geological mapping purposes; and
(v) To determine the exact bottom hole location of the well for the purpose of monitoring
reservoir performance and also for relief well drilling.

There are many types of equipment and tools used for surveying, but they can be generally classified
into two groups: those with MWD and those with wireline logging tools. MWD systems can be used
to survey when drilling is ongoing. The equipment without MWD systems are those that cannot be
used when drilling is ongoing. The equipment can be classified as single/multi shots and steering
tools. Single-shot surveys can be done during routine drilling operations like just before tripping to
Report 20 377 Miyora

change the bit or BHA. If there is a need to take a survey, drilling can be stopped to make single shot
surveys and then resume drilling. On the other hand, steering tools or MWD systems furnish the
driller with real-time directional data on the rig floor. Some of the tools are discussed below.

(i) Magnetic single-shot: The magnetic single-shot measures both the

drift and the compass direction of the wellbore. The instrument barrel
of a magnetic single-shot has several components, for example a
precision floating compass, a device to superimpose the concentric
circles calibrated in degrees with a plumb bob indicator (for measuring
inclination). A camera photographs the plumb bob and compass face
to record both drift and direction (Short, 1993). The magnetic single-
shot cannot record compass directions inside steel pipe or casing
because they shield off the earth’s magnetic lines of force. It records
only in open holes or inside nonmagnetic drill collars. The instrument
is checked, the timer is set and the tool is lowered down the drill string
on the wireline. Sinker bars may be attached to the instrument to aid
its travel through the mud into the baffle plate. After waiting for the
camera to operate, the instrument is pulled out of the hole. The
recording disk is removed and readings taken from it.

The readings from the magnetic single-shot must be corrected for

magnetic declination. This is the difference between magnetic north
and true north. The standard practice is to report directions as true
bearings. The amount of declination depends on the geographical
location. A magnetic single-shot is shown in Figure 12.

(ii) Magnetic multi-shot: The magnetic multi-shot instrument works

on the same principle as the single-shot but is capable of taking a series
of pictures at pre-set intervals. This is due to a built in film-wound
camera with a timer that automatically exposes and advances the film
at preset intervals. Using a stop watch, the surveyor keeps track of the
depth at which the preset timer takes a shot. Only the shots taken at
known depths with the drill string stationary are used in plotting the
trajectory. This instrument is used to survey the entire well trajectory
while tripping out. The tool must be inside a nonmagnetic drill collar
for it to be able to measure direction in the cased hole (Vieira, 2009).

(iii) Gyroscopic multi-shot: Gyroscopic instruments measure compass FIGURE 12: Magnetic
directions without using the earth’s natural lines of magnetic force. A single-shot instrument
gyroscopic compass is not affected by the presence of magnetic fields. (Vieira, 2009)
Gyroscopes can also measure drift angles using regular or modified drift measuring instruments.

The components of a gyroscopic instrument include a spinning wheel driven by an electric motor at
40,000 revolutions per minute (Vieira, 2009). The working principle of a gyroscopic multi-shot is
based on its spinning. The direction of the spinning wheel is maintained by its own inertia and the
axis of rotation of the wheel is kept in one direction, irrespective of how the other axes are rotated.
This property of the gyroscope is used as a reference for measuring the azimuth. Before running the
gyro instrument into the well, the gyro must be aligned with a known reference direction which is
usually true north. The gyro takes a survey at preset intervals as the tool is run into the wellbore. The
survey results are conveyed to the surface via a wireline. The readings do not need to be corrected for
declination because they are taken with reference to true north.
Miyora 378 Report 20

The disadvantage of gyroscopic instruments is that they tend to drift gradually out of alignment when
an unbalanced force acts upon them. This effect is reduced by taking measurements while running the
instrument into the hole. When taking readings, allowance for drift is made with the results.

5.3 Surveying/logging equipment with MWD systems

MWD records measurements on a real time basis while drilling is in progress. Vieira (2009) states
that the measured data may be transmitted to the surface by one of the three methods stated below:

(i) Via a wireline;

(ii) Through pressure variations in the drilling mud/fluid (mud pulse); and
(iii) Through electromagnetic or radio waves.

When an instrument uses a wireline to convey survey data to the surface, drilling is usually done with
a downhole motor which eliminates the need for the drill string to rotate. There is a provision for the
wireline to pass out of the drill string at the side-entry sub. It allows the wireline to pass from the drill
string out to the annulus at some point beneath the rig floor. Since the wireline does not interfere with
the pipe connection, the instrument need not be pulled out (Inglis, 1987).

The science of measuring at a distance is known as telemetry. In drilling, the three most important
parameters for the directional driller are the inclination, the direction (azimuth), and the direction of
the tool face. Inclination is measured by accelerometers which can measure the components of the
earth’s gravitational field. The direction of the hole is measured by magnetometers which can
measure the earth’s magnetic field. Three types of MWD instruments will be discussed here: the
steering tool, mud pulse telemetry and electromagnetic telemetry.

(i) Steering tool: Vieira (2009) defines a steering tool as a wireline telemetry surveying instrument
that measures inclination and direction while drilling progresses. Steering tools, as the name suggests,
help the driller by providing him with the necessary information to steer the bit in the correct direction.
Steering tools are especially important during the critical period of kicking off the well when surveys
of the well need to be taken at close intervals.

The tool consists of an electronic probe that is run into the hole on a conducting wireline. It is usually
put at the orienting sub just above the bent sub. Because the wireline of the steering tool needs to be
stationary, the tool can only be used with a downhole mud motor that turns the bit while the drill string
does not rotate. The electronic probe has sensors (accelerometers and magnetometers) that measure
hole inclination, azimuth and tool face orientation. The survey data are transmitted as signals from the
probe through the wireline to the surface, where a computer converts them and displays the directional
data on the screen monitor. The use of steering tools in the market has decreased because of their
limitations (used only when the drill string does not rotate) as compared to mud-pulse instruments.

(ii) Mud-pulse telemetry: This instrument transmits signals uphole through the drilling fluid, allowing
the driller to obtain real time readings on the monitoring screen while drilling is in progress, including
when the drill string is rotating. Vieira (2009) identifies five components in mud-pulse telemetry as
follows:

a) Downhole unit that senses direction and inclination;

b) Mud pulse generator, also called a pulser;
c) Surface transducer that decodes mud pressure variations and transforms them into electric signals;
d) Computer that receives the electric signals and interprets them; and
e) Rig floor unit that displays the data.
Report 20 379 Miyora

The mud pulse generator contains a microprocessor that converts the survey data (inclination,
direction and tool face orientation) into a series of positive and negative pressure pulses. Positive
pulse indicates pressure increase and negative pulse indicates pressure decrease. The pulses are
usually transmitted to the surface coded in a binary signal. The mud-pulse telemetry unit inside the
hole contains accelerometers and magnetometers for measuring inclination and direction. To avoid
magnetic interference, it is placed within the nonmagnetic drill collar.
The downhole unit is powered by a turbine which is rotated by the drilling fluid. The unit can also be
powered by batteries. Some mud pulse units contain both batteries and turbine. Measurements are
made continuously as drilling goes on, whether there is circulation or not. The data is transmitted to
the surface only when there is circulation or circulation resumes. The limitation of mud pulse
telemetry is that it cannot transmit data when drilling with air or aerated drilling fluids or when there is
no continuous column of drilling fluid.

(iii) Electromagnetic (EM) telemetry systems: EM telemetry can transmit data where mud-pulse
telemetry cannot. This is made possible when a two way communication link between the downhole
equipment and the surface equipment is established. Data can be transmitted through any formation
using low-frequency electromagnetic waves. EM telemetry is especially important when doing
underbalanced drilling. Because the EM system does not have moving parts to produce pressure
pulses, it is more reliable than mud pulse telemetry.

6. COMPARISON OF DIRECTIONAL DRILLING IN KENYA AND ICELAND –

TIME ANALYSIS

The drilling histories of 12 wells from Olkaria, Kenya and 14 wells from Hengill field, Iceland were
analysed for comparison. All of these wells were directionally drilled and the casing and bit sizes used
were the same.

The Hengill high-temperature field rates as one of the largest in Iceland. It is located 30 km east of
Reykjavik. The Hengill volcanic system is composed of crater rows and a large fissure swarm. It is
located on the eastern border of the Reykjanes Peninsula, SW-Iceland. The Hengill volcanic system
has a 100 km long NE-SW axis, 3-16 km wide, extending from Selvogur in the southwest to
Ármannsfell in the northeast. The Hengill central volcano covers an area of about 40 km2 (Björnsson
et al., 1986).

The greater Olkaria geothermal area in Kenya is divided into seven fields, namely Olkaria East
(Olkaria I), Olkaria Northeast, Olkaria Central, Olkaria Northwest, Olkaria Southwest, Olkaria
Southeast and Olkaria Domes. It is situated south of Lake Naivasha on the floor of the southern
segment of the Kenya rift (Figure 1). The wells studied in this report are from the Olkaria I and
Olkaria Domes field. Olkaria I has been producing power since 1981 when the first of the three
15 MWe units was commissioned (Lagat, 2004). Olkaria Domes is currently undergoing production
drilling. It will eventually be developed as Olkaria IV, where a 140 MWe power plant is scheduled to
be developed (Cherutich, 2009).

6.1 Time analysis-raw data

This section deals with raw data and the tables and figures represent the actual work done. The
drilling of the wells under study was done in steps; each step is identified by a different casing. The
Icelandic naming style for the stages has been adopted for the Kenyan wells. In this report, for
Kenyan wells, surface drilling (0–60 m) was called step 0. Table 1 shows the steps of a typical well
for both Kenya and Iceland.
Miyora 380 Report 20

0
OW-35A
TABLE 1: Drilling phases
OW-36A
OW-38A Kenyan wells Icelandic wells
OW-903A
OW-903B Steps Depths Steps Depths
OW-904A
OW-904B (m) (m)
OW-905A
OW-906A 0 0-60 Predri. 0-90
1000
OW-907A
OW-908A
1 60-300 1 90-300
OW-910A 2 300-1000 2 300-800
Measured Depth (m)

3 1000-2800 3 800-2300

For the comparison to be as close

2000
as possible, step 0 and pre-drilling
times were not included in the
analysis except in determination of
the average depth drilled per day
(ADD/D), which did include step 0
3000 for the Kenyan wells. This is
because a smaller rig was usually
used for pre-drilling to 80–90 m
for Icelandic wells. The wells
0 20 40 60 80 100
Days studied, their depths and the time
FIGURE 13: Depth vs. days graph for drilled Olkaria wells taken to drill them are shown in
Table 2. Figure 13 shows the
actual drilling progress (depth versus days) for the Kenyan wells. Wells OW-903A and 903B took the
longest time to drill. The reason is that cementing took a long time and there was a gas blowout that
delayed drilling.

TABLE 2: Summary of wells studied; pre-drilling for Iceland and step 0 for Kenya are not included.

Icelandic wells Kenyan wells

Drilled Drilled
Well no. Depth Days Well no. Depth Days
depth depth
HE-03 1887.00 1792.00 39 OW-903A 2810.89 2748.39 71.33
HE-04 2008.00 1930.40 45 OW-903B 2800.00 2740.00 76.04
HE-05 2000.00 1904.00 45 OW-904A 2799.31 2731.14 57.08
HE-06 2013.00 1935.00 37 OW-904B 2820.00 2755.20 62.48
HE-13 2397.00 2318.70 44 OW-905A 2800.00 2738.00 56.21
HE-26 2688.00 2596.00 54 OW-906A 2804.49 2742.22 57.23
HE-36 2808.00 2703.00 45 OW-907A 2588.22 2527.72 58.92
HE-51 2620.00 2520.00 34 OW-908A 3000.00 2937.65 45.38
HE-53 2507.00 2438.00 57 OW-910A 2881.73 2819.38 54.75
HE-54 2436.00 2342.00 29 OW-35A 2763.00 2697.61 52.71
HE-55 2782.00 2681.00 37 OW-36A 2880.00 2817.00 48.77
HE-57 3118.00 3023.00 40 OW-38A 3010.00 2949.30 46.92
NJ-24 1928.60 1849.60 35
NJ-25 c 2098.00 1993.00 28
Average 2378.90 2287.60 40.64 2829.80 2766.97 57.32

6.2 Normal distribution curve for average depth drilled per day (ADD/D)

The acronym ADD/D is not usually used in drilling. For convenience in this report, it has been
adopted since it will be used many times. Average depth drilled per day (ADD/D) indicates the
average depth in metres that is drilled in 24 hours. High ADD/D indicates that the well was drilled in
Report 20 381 Miyora

a shorter time, hence at lesser cost. A normal distribution curve indicates how spread the ADD/D time
is and indicates the average ADD/D for all the wells. The results in Tables 3 and 4 were calculated
using Equations 12–14.

ADD/D becomes variable x which is the parameter being studied. The range is the difference between
the maximum ADD/D and the minimum ADD/D. It is used in finding the size of the class interval.
The mean is the average of the sample taken. Variance and standard deviations (S2 and S) are
statistical parameters found as shown in Equations 12 and 13. The function f(x) is the so-called
probability function, which is based on normal distribution giving the same values for mean and
standard deviation as calculated from our samples. It is calculated from Equation 14 (Chatfield,
1983):

∑
(12)
1

(13)

1
(14)
√2

where S2 = Variance;
x = Variable under study;
= Sample mean;
n = Sample size/population;
S = Standard deviation; and
e = Euler’s number, a constant given as (2.71828...).

Using Equations 12 to 14, various parameters were calculated for Icelandic and Kenyan wells,
presented in Tables 3-6. For Icelandic wells (Table 3) the range is 38.45 (i.e. 80.76-42.31), while for
Kenyan wells (Table 5 and 6) the range is 30.87 (i.e. 60.15-37.01).

TABLE 3: ADD/D for the TABLE 4: Frequency of

studied Icelandic wells different ADD/Ds, Iceland

ADD/D (x) f(x) Interval classes Frequency

42.31 0.0161 37-41 0
42.77 0.0167 41-45 3
42.90 0.0169 45-49 3
45.95 0.0205 49-53 3
47.60 0.0223 53-57 0
48.07 0.0228 57-61 0
52.30 0.0265 61-65 0
52.70 0.0268 65-69 0
52.85 0.0268 69-73 2
71.18 0.0174 73-77 2
72.46 0.0158 77-81 1
74.12 0.0139
75.58 0.0122
80.76 0.0071
Miyora 382 Report 20

TABLE 5: ADD/D for the TABLE 6: Frequency of

studied Kenyan wells different ADD/Ds, Kenya

ADD/D (x) f(x) Class interval Frequency

32.34 0.0104 31-33 1
37.01 0.0219 33-36 0
41.14 0.0338 36-39 1
42.77 0.0379 39-42 1
45.15 0.0423 42-45 1
46.90 0.0438 45-48 2
48.07 0.0440 48-51 3
49.08 0.0435 51-54 0
50.50 0.0420 54-57 1
56.31 0.0281 57-60 0
60.15 0.0172 60-63 1
63.21 0.0102 63-66 1

The resulting sample distribution for the ADD/D for Icelandic wells is shown in Figure 14 and the
corresponding normal distribution in Figure 15. Similar figures for the Kenyan wells are shown in
Figures 16 and 17. For Kenyan wells, the average depth drilled per day is about 48 m and for
Icelandic wells it is about 57 m. For Kenyan wells the standard deviation is 9 m, and for Icelandic
wells it is 14 m.

0.06 0.06
Probability density function
Sample distribution

0.05 0.05
0.04 0.04
0.03
0.03
f(x)

0.02
0.02
0.01
0.01
0.00
0.00
35 40 45 50 55 60 65 70 75 80
Average depth drilled per day, x (m/d) Average depth drilled per day, x (m/d)

FIGURE 14: Sample distribution for ADD/D FIGURE 15: Normal distribution for ADD/D
for Icelandic wells for Icelandic wells

0.09 0.09
Probability density function f(x)

0.08 0.08
0.07 0.07
Sample distribution

0.06 0.06
0.05 0.05
0.04 0.04
0.03 0.03
0.02 0.02
0.01 0.01
0.00 0.00
31-3333-3636-3939-4242-4545-4848-5151-5454-5757-6060-6363-66 30 35 40 45 50 55 60 65 70

Average depth drilled per day, x (m/d) Average depth drilled per day, x (m/d)

FIGURE 16: Sample distribution for ADD/D FIGURE 17: Normal distribution for ADD/D
for Kenyan wells for Kenyan wells
Report 20 383 Miyora

Tables 7 and 8 show various activities undertaken during drilling and the time spent doing them. The
data for Icelandic wells was adopted from the work of Sveinbjörnsson (2010). Figures 18 and 19 give
graphical representations of how the total work time was consumed by various activities during
drilling. In order to improve efficiency in drilling, it is important to reduce the time spent on activities
other than actual drilling.

TABLE 7: Time analysis (days) for different activities

during directional drilling of Kenyan wells

plug jobs
Well No.

bit/BHA

cleaning
Wait on
Cement
Drilling

Drilling

Change
Fishing
Casing

Repair

Other
Ream

water
depth

Stuck

Total
Cem.
steps
Well

Hole

Log.
(m)

Step 0 60.5 0.73 0.29 2.08 1.38 0.00 0.00 0.00 0.00 0.00 0.15 0.00 0.00 0.00 4.63
Step 1 309.4 9.65 1.13 5.21 0.00 0.00 1.00 0.00 0.38 0.50 0.00 0.19 0.00 1.29 19.33
OW-903A

Step 2 1319.2 10.19 0.50 1.54 1.63 6.25 1.46 0.00 0.25 1.71 0.25 0.17 0.19 1.81 25.94
Step 3 2810.89 16.88 0.71 0.00 0.00 0.00 0.46 0.00 0.00 3.77 0.63 0.00 2.22 1.41 26.06
TOTAL 37.44 2.63 8.83 3.00 6.25 2.92 0.00 0.63 5.98 1.02 0.35 2.41 4.51 75.96
(%) 49.29 3.46 11.63 3.95 8.23 3.84 0.00 0.82 7.87 1.34 0.47 3.17 5.94 100
Step 0 60 2.13 0.13 7.50 0.00 0.00 0.54 0.00 0.00 0.00 0.00 0.04 0.00 0.21 10.54
Step 1 319.19 16.02 0.50 7.63 0.00 0.00 0.63 0.00 0.00 0.54 0.04 0.15 0.00 0.63 26.13
OW-903B

Step 2 1204 11.58 0.58 2.08 0.00 0.00 0.88 0.00 0.00 0.75 0.00 0.00 0.21 5.83 21.92
Step 3 2800 18.65 0.71 0.00 0.00 0.00 0.90 0.00 0.00 3.12 0.25 0.58 3.29 0.50 27.99
TOTAL 48.38 1.92 17.21 0.00 0.00 2.94 0.00 0.00 4.41 0.29 0.77 3.50 7.17 86.58
(%) 55.88 2.21 19.88 0.00 0.00 3.39 0.00 0.00 5.09 0.34 0.89 4.04 8.28 100
Step 0 68.17 2.17 0.54 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.46 0.00 1.00 4.92
Step 1 316 2.79 0.83 1.46 0.00 0.00 0.21 0.00 0.00 0.00 0.00 0.08 0.00 1.58 6.96
OW-904A

Step 2 1259 11.88 0.63 4.29 1.54 0.00 5.17 0.33 0.00 1.21 0.21 0.67 0.25 2.96 29.13
Step 3 2799.31 10.77 1.29 0.00 0.00 0.00 0.19 0.00 0.00 4.69 0.50 0.13 2.48 0.96 21.00
TOTAL 27.60 3.29 6.50 1.54 0.00 5.56 0.33 0.00 5.90 0.71 1.33 2.73 6.50 62.00
(%) 44.52 5.31 10.48 2.49 0.00 8.97 0.54 0.00 9.51 1.14 2.15 4.40 10.48 100
Step 0 64.8 1.54 0.29 0.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.88 3.46
Step 1 322 4.71 0.54 1.25 0.00 1.21 0.00 0.88 0.00 0.25 0.17 0.33 0.00 1.17 10.50
OW-904B

Step 2 1206 13.00 1.75 2.46 0.00 0.00 0.38 0.00 0.00 1.88 0.17 0.27 0.52 0.83 21.25
Step 3 2820 16.75 1.04 0.00 0.00 0.00 0.88 0.75 0.00 4.40 0.75 2.50 2.50 1.17 30.73
TOTAL 36.00 3.63 4.46 0.00 1.21 1.25 1.63 0.00 6.52 1.08 3.10 3.02 4.04 65.94
(%) 54.60 5.50 6.76 0.00 1.83 1.90 2.46 0.00 9.89 1.64 4.71 4.58 6.13 100

6.3 Time analysis - processed data

To compare the time spent in each of the activities for the wells, the number of workdays must be
normalised with respect to a reference well (Sveinbjörnsson, 2010). The reference well was found by
obtaining the average of the parameters under study and using it as a standard. Table 9 shows the
reference wells against which the study wells were normalised. The normalising method was adapted
from a previous work by Sveinbjörnsson (2010). In order to normalise the work days for each activity
in each phase, interpolation is carried out using Equation 15 (Sveinbjörnsson, 2010). This was done
for both Kenyan and Icelandic wells.

(15)

where Ti = The normalized number of workdays for activity i; and

ti = The actual number of days spent on activity i.
Miyora 384 Report 20

TABLE 8: Time analysis (days) for different activities

during directional drilling of Icelandic wells

depth (m)

plug jobs
Well No.

bit/BHA

cleaning
Wait on
Cement
Drilling

Drilling

Change
Fishing
Casing

Repair

Other
Ream

water
Stuck

Total
Cem.
steps

Well

Hole

Log.
plug
HE-03 1. step 324 4 1 1 1 1 8
2. step 812 5 1 1 1 2 10
3. step 1887 8 2 1 2 3 5 21
TOTAL 17 4 2 1 2 5 8 39
(%) 43.59 10.26 5.13 2.56 5.13 12.82 20.51 100
HE-04 1. step 305 4 1 1 1 2 9
2. step 789 4 1 1 4 3 4 17
3. step 2008 11 1 1 2 4 19
TOTAL 19 3 2 4 1 6 10 45
(%) 42.22 6.67 4.44 8.89 2.22 13.33 22.22 100
HE-05 1. step 303 3 1 1 2 1 8
2. step 802 6 1 1 1 1 10
3. step 2000 14 2 1 2 1 3 4 27
TOTAL 23 4 2 0 1 2 1 0 6 6 45
(%) 51.11 8.89 4.44 0.00 2.22 4.44 2.22 0.00 13.33 13.33 100
HE-06 1. step 310 4 1 1 4 2 1 13
2. step 813 4 1 1 1 1 2 10
3. step 2013 5 1 3 5 14
TOTAL 13 3 2 4 1 6 8 37
(%) 35.14 8.11 5.41 10.81 2.70 16.22 21.62 100
80 80
Other Measurement Other
Hole cleaning Repair
Measurement
Change bit/BHA Wait on water
Hole cleaning
70 Fishing Reaming
70 Repair
Stuck Cement plug jobs
Cement Casing Change bit/BHA

Drilling Wait on water

60 Fishing
60
Reaming
Stuck
Cement plug jobs
50
50 Cement
Casing
Drilling
Days

40
40

30
30

20
20

10
10

0
0

FIGURE 18: Time analysis for different FIGURE 19: Time analysis for different
activities for Kenyan wells activities for Icelandic wells
Report 20 385 Miyora

TABLE 9: Drilling phases

Reference Kenyan well Reference Icelandic well

Steps Depth (m) Steps Depth (m)
1 314.09 1 309.40
2 1073.89 2 854.90
3 2829.64 3 2377.90

Finally, in order to compare Kenyan wells with Icelandic wells, Icelandic wells were normalised into
Kenyan wells using the Kenyan well in Table 9 as the reference well. Because there were so many
tables generated in the process, the resultant graphs for each phase used as a basis of comparison are
shown in Figures 20-27.
16 16
Maximum Maximum
14 Average 14 Average

Minimum Minimum
12 12

10 Days 10
Days

8 8

6 6

4 4

2 2

0 0

FIGURE 20: Step 1 time analysis for FIGURE 21: Step 1 time analysis for
different activities for Kenyan wells different activities for Icelandic wells
35 35
Maximum Maximum

30 Average 30 Average

Minimum Minimum
25 25

20 20
Days

Days

15 15

10 10

5
5

0
0

FIGURE 22: Step 2 time analysis for FIGURE 23: Step 2 time analysis for
different activities for Kenyan wells different activities for Icelandic wells

6.4 Discussion

Table 10 shows the percentage of the work time consumed by the different activities during drilling.
Based on the time analysis, the following general deductions can be made from the study: The study
of 12 Kenyan wells and 14 Icelandic wells shows that about 58% of the work time for Kenyan wells
and 45% of the work time for Icelandic wells is spent on actual drilling, i.e. the bit cutting the
formation. The rest of the time is spent on other activities that make actual drilling possible, like
cementing, and also activities that hamper drilling, for example a stuck drill string or casing.
Minimising the time spent on activities other than drilling will improve the efficiency of drilling time
Miyora 386 Report 20

30 30
Maximum Maximum

Average Average
25 25
Minimum Minimum

20 20
Days

Days
15 15

10 10

5 5

0 0

FIGURE 24: Step 3 time analysis for FIGURE 25: Step 3 time analysis for
different activities for Kenyan wells different activities for Icelandic wells
80 80
Maximum Maximum
70 Average 70 Average
Minimum
60 Minimum 60

50 50
Days

Days

40 40

30 30

20 20

10 10

0 0

FIGURE 26: Overall time analysis for FIGURE 27: Overall time analysis for
different activities for Kenyan wells different activities for Icelandic wells

TABLE 10: Percentage of total time taken by each drilling activity

Drilling Casing Cem. Plug Stuck Ream. Fish WOW bit/BHA Repair Cleaning Meas. Other
Kenya 57.94 4.42 7.40 0.47 1.26 3.22 0.42 0.37 9.55 2.02 1.66 4.93 6.35
Iceland 45.31 8.33 5.29 4.45 4.99 2.16 0 0.12 0.95 1.16 9.43 17.52 0.28

and eventually lower the drilling cost per well. From Table 2, the average drilling time and well depth
are 41 days and 2379 m for Iceland and 57 days and 2830 m for Kenya. Note that the drilling time
excludes step 0 (Kenya) and pre-drilling time (Iceland). Analysis of ADD/D as shown Figures 14 and
15 indicates that for Iceland it is about 57 m per day, while for Kenya (Figures 16 and 17) it is about
48 m per day. Figures 20 to 27 show graphs generated after normalising all the wells with reference
wells. Each step can be compared with the other for similar activities because they have been put on
the same basis as much as possible. Here below, the areas of large disparities for similar activities
undertaken in drilling for Kenya and Iceland will be discussed.

(i) Bit or BHA change: Kenya spends ten times more time on a bit/BHA change as Iceland. Most of
this time is spent during tripping out to change worn out bits or to run in angle correction BHA.
The possible cause could be in the types of bits used. Using long life bits may save on time spent
to trip out of the hole in order to change the bit frequently. Tripping to change the angle correction
BHA is another possible cause. Adopting measurement-while-drilling technology (MWD) and
using a mud motor for a longer time after finishing the angle build up would greatly reduce the
Report 20 387 Miyora

need to trip out in order to run in angle correction BHA. In Iceland, drilling is done using the mud
motor for a big portion of the trajectory (until a total loss of circulation is encountered) after
building the inclination to the required angle. In Kenya, after building up the angle, the mud motor
is changed and there are cases where it is run in for angle correction. This takes a lot of time.

(ii) Well logging: Takes the second largest work time after drilling in Iceland. Iceland spends almost
four times as much time for well logging as Kenya. Mostly Kenya does measurements for
direction and inclination and well completion tests and well temperature and pressure logs during
completion tests. The most common well logging done in Iceland is for temperature and pressure.
Calliper logging is done before cementing and cement bond logging (CBL) after cementing. Then
there is a full set of lithological logs made in the open hole before running each casing string. Such
measurements help get information with which to better understand subsurface conditions. In
Iceland, several gyroscopic surveys were made to confirm the MWD readings, due to reversals in
polarity of the basalt lavas from different ages.

(iii) Cementing: Cementing wells in Kenya takes almost 1½ times longer than cementing in Iceland.
A possible cause for delay is the large amount of backfill cementing done in Kenya when the
primary cementing did not fill the annulus space up to the surface. Unlike in Iceland, where
plugging of major loss zones is done, Kenya continues to drill blind ahead of major loss zones
which eventually takes longer during cementing to fill up. Another possible cause is in the
cementing programme. No calliper logging is done to accurately ascertain the capacity of the
annulus so it could be that the amount of cement needed to be pumped into the hole during primary
cementing is being underestimated.

(iv) Running the casings: The analysis indicates that Iceland takes almost two times longer to run
casings in than Kenya.

(v) Stuck: The analysis indicates that the drill string/casing gets stuck more often or longer in Iceland
than in Kenya. There could be a close correlation with the time Iceland spends in cleaning the hole
to avoid getting stuck.

(vi) Hole cleaning: Iceland takes almost five times longer for hole cleaning than Kenya. It is
important to clean the hole to avoid cuttings from accumulating at the bottom of the well which
could cause sticking of the drill string and reduce the penetration rate due to regrinding of the
cuttings.

(vii) Other activities: This section covers activities that arise to delay the progress of drilling such as
accidents, the presence of H2S, installing well heads and blow out preventers (BOPs). Kenya
spends a lot of time in installing well head and BOPs when changing from drilling in one step to
another.

6.5 Suggested improvements

Areas of improvement for Kenya include investing in better quality bits to reduce the tripping time
needed to change the bits; also an investment in MWD equipment which would reduce the time
needed for checking the direction and inclination of the well. Kenya should reduce cementing time by
doing calliper logs to help generate a more accurate cementing programme. Besides drilling, the
activity that takes the longest time for Iceland to accomplish is well logging. Taking the right set of
logs, sufficient for understanding the geology and the reservoir properties, may reduce the time spent
in logging. There is need to optimise in the time spent in taking logs to increase knowledge compared
to the time spent in acquiring it.
Miyora 388 Report 20

7. SPECIAL PROBLEMS IN DIRECTIONAL DRILLING

Generally, there are many problems encountered while drilling vertical wells, but directional wells are
more difficult to drill than vertical wells. This is because everything done by routine in vertical
drilling becomes more complex when the well has to be drilled directionally. The problems
encountered in directional drilling are related to factors such as the well profile and the reduced axial
component of gravity acting along the drill string. The proportion of difficulties in drilling a well is
usually reflected in the time taken to complete the well, which has a direct effect on the cost of the
well. Vieira (2009) identifies five special problems that occur in directional drilling as:

(i) More hoisting capacity is often needed to raise and lower the drill string;
(ii) Greater rotary torque is needed to overcome friction;
(iii) Mud and hydraulic system requirements are more critical;
(iv) Stuck pipe and equipment failures are more common; and
(v) Casing is harder to run and cement.

These problems are caused by several factors encountered during drilling. These factors are discussed
below.

7.1 Tortuosity

An abrupt rate of change in wellbore trajectory is the cause of many problems in directional drilling.
Inclination and direction should be changed gradually and be evenly distributed throughout the length
of the trajectory. Severe dog legs should be avoided because they may cause key-seating and increase
torque and drag. If the drill string has to pass through a severe dog leg, the pipe will make contact
with the side of the hole. As the drill string rotates, a small diameter groove in the side of the borehole
wall will result. A problem will result during tripping because the large diameter drill collars will get
stuck at the keyseat. To free the drill string, the keyseat must be reamed out by a stabilizer or keyseat
wiper which is usually installed on top of the drill collars. It is good practice to install a keyseat wiper
in the BHA when drilling directional wells where dog-legs can be expected.

7.2 Formation effect

Different types of formations are encountered when drilling. Some formations become unstable either
during drilling or some time later. This may cause fragments of the formation to fall into the hole and
around the drill collars or the bit. Borehole instability may result from such conditions as the presence
of a high percentage of swelling clays (sodium montmorillonite), the presence of steeply dipping or
fractured formations, or over pressured shale zones and turbulent flow of drilling fluids in the annulus
can cause washouts in soft formations. Most of the problems can be related to shale zones. Most
shales will absorb water to some extent, lowering the compressive strength of the rock and allowing it
to expand.

Sometimes the formation deflects the bit. Controlling the direction of the trajectory becomes more
difficult when drilling through laminar or thin layer formations that are not level. When the
formation’s angle off the horizontal plane (dip) is less than 45°, the bit tends to drill perpendicular to
the layers. If the angle is more than 45° the bit drills parallel to the layers. To overcome this problem,
a stiff BHA must be used.

The drill bit also tends to deviate horizontally parallel to tilted formation strata. This effect is called
wandering. Even where strata are horizontal, the right-rotating bit tends to walk to the right in
inclined holes. This is called bit walk. Stiff BHA may not solve this problem. A more effective
method is to use a steering system, discussed under steering tools in this report. If the driller
Report 20 389 Miyora

anticipates this problem, he can offset the bit in the opposite direction to compensate for bit wandering
and bit walk and then let the bit walk to the final target.

7.3 Differential sticking

To prevent the flow of formation fluids into the well bore, the hydrostatic mud pressure in the
borehole must balance or exceed the pore pressure. In a permeable zone, a natural filtration process
will take place whereby the fluid content of the mud will invade the formation while the solids will
build up on the wall of the bore to form a filter cake. If the filter cake becomes thick, the drill collars
may come into contact with it and become embedded. If the positive pressure differential is large
(about 1000 psi) it may be difficult to free the pipe. The risk of differential sticking is increased if the
pipe is allowed to stay static for a period of time (Inglis, 1987).

8. CONCLUSIONS AND RECOMMENDATIONS

A major project of exploratory and appraisal drilling will soon kick off in Kenya in a number of its
geothermal fields, hitherto not drilled into. The majority of the wells drilled will be directional wells.
There is a need for proper tools for close monitoring of the well trajectory to ensure that the target is
reached. The computer programs generated in this report will be handy in monitoring. The time
analysis done here, which includes getting the overall ADD/D and the overall average time taken, will
be helpful as a benchmark of what has been achieved, and will be useful in planning for the project.
The comparison of Icelandic and Kenyan drilling data is also important; Kenyans can learn from the
Icelandic experience and the comparison also points out areas where Iceland can improve.

It is recommended that Kenya invest in long life bits which, although expensive, will reduce the time
taken for tripping to change bits and, hence, prove to be more effective. It would also be better to
consider drilling with a mud motor after building up the angle in order to reduce angle correction re-
runs.

ACKNOWLEDGEMENTS

I am indebted to the Icelandic Government, United Nations University and Geothermal Development
Company Limited who financially made the study possible. I wish to specially thank the UNU staff:
Dr. Ingvar B. Fridleifsson, Director, Mr. Lúdvík S. Georgsson, Deputy Director, and Thórhildur Ísberg
who was efficient in all administrative matters. I thank Ingimar G. Haraldsson for his help. My
special gratitude goes to Markús A.G. Wilde who had solutions at his fingertips for every challenge I
encountered in Iceland. The lecturers and the Orkustofnun staff were very kind in passing on
knowledge to us. I wish to thank in a special way my supervisors, Sverrir Thórhallsson, Sigvaldi
Thordarson and Björn Már Sveinbjörnsson. Finally, I thank my family, my dear wife E, my daughter
M and son P who gave me the moral support to endure our separation while undertaking the study.

I wish to thank God, creator of heaven and earth, for enabling and sustaining me throughout the period
I have been a Fellow at UNU-GTP.
Miyora 390 Report 20

REFERENCES

Baker Hughes INTEQ, 1995: Drilling engineering workbook. A distributed learning course. Baker
Hughes INTEQ Inc., 410 pp.

Björnsson, A., Hersir, G.P., and Björnsson, G., 1986: The Hengill high-temperature area SW-Iceland:
Regional geophysical survey. Geoth. Res. Council, Transactions, 10, 205-210.

Chatfield, C., 1983: Statistics for technology (3rd Edition). Chapman and Hall NY, 381 pp.

Cherutich, S.K., 2009: Rig selection and comparison of top drive and rotary table drive systems for a
cost effective drilling project in Kenya. Report 8 in: Geothermal training in Iceland 2009. UNU-
GTP, Iceland, 65-84.

Gabolde, G., and Nguyen, J.P., 2006: Drilling data hand handbook (8th edition). IFP publications,
Editions Technip, 541 pp.

GDC, 2009: Ten year business plan (2009-2019). Geothermal Development Company, Ltd.,
unpublished document.

GDC, 2010: Geothermal sites in Kenya. Geothermal Development Company, Ltd., website:
www.gdc.co.ke/index.php?option=com_content&view=article&id=191&Itemid=163

Inglis, T.A., 1987: Directional drilling, vol. 2. Graham & Trotman Ltd., London, 260 pp.

Lagat, J.K., 2004: Geology, hydrothermal alteration and fluid inclusion studies of the Olkaria Domes
geothermal field, Kenya. University of Iceland, MSc thesis, UNU-GTP, Iceland, report 2, 71 pp.

Ngugi P., 2002: Technical, economic and institutional evaluation of adopting directional drilling by

KenGen, Kenya. Report 9 in: Geothermal training in Iceland 2002. UNU-GTP, Iceland, 113-144.

Short, J.A., 1993: Introduction to directional and horizontal drilling. Pennwell Publishing Co., Tulsa,
OK, 232 pp.

Sveinbjörnsson, B.M., 2010: Estimate of costs and uncertainties in high-temperature drilling in the
Hengill area. University of Reykjavik, MSc thesis, 170 pp.

Vieira, J.L., 2009: Controlled directional drilling (4th edition). Petroleum Extension Service, Austin
TX, 133 pp.