Geothermics 53 (2015) 113

Contents lists available at ScienceDirect

Geothermics
journal homepage: www.elsevier.com/locate/geothermics

Economic comparison of a well-head geothermal power plant and a

traditional one
Carlos Atli Crdova Geirdal , Maria S. Gudjonsdottir, Pall Jensson
Reykjavik University, School of Science and Engineering, Menntavegi 1, Reykjavik 101, Iceland

a r t i c l e

i n f o

Article history:
Received 6 May 2013
Accepted 2 April 2014
Available online 8 May 2014
Keywords:
Geothermal
Wellhead power plant
Project development
Economical comparison

a b s t r a c t
The objective of this paper was to do an economic comparison between the traditional approach to
geothermal projects and a well-head method, where smaller power plants were installed on each well
to considerably reduce the time until energy production begins.
The two methods were compared in a hypothetical steameld, based on their NPV and net power production. The comparison showed that wellhead power plants benet geothermal projects by increasing
the power output and NPV by as much as 5% and 16% respectively, depending on how early they can start
production and the rate of installation.
2014 Elsevier Ltd. All rights reserved.

1. Introduction
1.1. Advantages of geothermal energy
Geothermal energy for power production has many advantages
over other sources of energy, which include:
High capacity factor (above 90%); higher than other renewable energy technologies and comparable to fossil fuel power
plants. This allows geothermal power plants to provide stable
and reliable base load power output, usually for several decades
(Gehringer, 2012, p. 2).
Flexibility; geothermal power plant loads can be increased or
decreased depending on demand. This can help an energy system where other intermittent sources of energy are used (e.g.
solar or wind) remain stable.
Low cost of energy produced (LCOE); between 410 US cents
per kWh with current prices (Cross and Freeman, 2009), mainly
because there is low operational cost and a high capacity factor,
offering an economically attractive power operation.
Low land use per unit of energy produced; low CO2 emissions and
relatively small environmental impact compared to other energy
sources.
Development of a domestic energy source that reduces the risks
related to the price of imported fuels.

Yet with all these advantages, electricity produced from

geothermal sources is only around 0.03% of global electricity production (Gelman, 2010, p. 53).
1.2. Obstacles for the use of geothermal energy
The main obstacles for increasing the utilization of geothermal
resource are:
The availability of geothermal resources; it is estimated that
geothermal resources are only available for utilization on one
quarter to one third of the planets surface.
Difculty in raising capital. This difculty is basically caused by
the following three reasons:
The large initial capital cost;
High risk in the early stages of development of geothermal
projects, mainly in the drilling stage;
The long period before geothermal projects start producing
energy and revenue, as one has to drill and test the wells before
designing the power plant, as well as the lead time it takes to
commence energy production.
This study addressed the second obstacle and focused on creating revenue earlier in the project by incorporating the use of smaller
wellhead power plants.
1.3. Types of geothermal power plants

Corresponding author. Tel.: +354 8449395.

E-mail addresses: [email protected], [email protected]
(C.A. Crdova Geirdal).
http://dx.doi.org/10.1016/j.geothermics.2014.04.003
0375-6505/ 2014 Elsevier Ltd. All rights reserved.

Geothermal power plants work similarly to traditional thermal

power plants in that they convert thermal energy to electricity

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

WH

WH

WELLHEAD
POWER PLANT

WELLHEAD
POWER PLANT

ELECTRIC LINES
STEAM
GATHERING
LINES

CENTRAL
TRANSFORMER
STATION

TRANSMISSION
LINE TO GRID

WH

WELLHEAD
POWER PLANT

WH

WELLHEAD
POWER PLANT

WH

WELLHEAD

TRANSMISSION
LINE TO GRID

Fig. 1. Schematic layout of wellhead power plants.

using a turbine and a generator in an energy process. The difference is the source of heat: in geothermal power plants geothermal
uids provide the heat from the hydrothermal system.
Geothermal power plants are basically divided in two groups:
steam cycles and binary cycles.
In the steam cycle the geothermal uid is allowed to boil or
ash above boiling point by lowering the pressure and becoming
a two-phase uid, and then the steam is separated from the brine
and expanded in a turbine. The process of lowering the pressure to
boil the uid is called ash process.
The single ash power plant is the mainstay of the geothermal
power industry (DiPippo, 2008) with more than 40% of the installed
capacity of geothermal power plants (Bertani, 2012). The double
ash power plant is basically the same as a single ash power plant;
the only difference is that the liquid phase from the rst ash process is separated again to extract more steam at a lower pressure.
The double ash power plant can usually obtain between 2025%
additional power output compared to a single ash power plant
(DiPippo, 1999) but with additional components and cost. Backpressure power plants also use the steam produced from a ashing
process, the difference is that their turbines discharge the steam to
the atmosphere, they produce less energy than condensing turbines
but they are the most simple of all, they do not have condensers,
gas extraction systems and cooling towers, thus making them more
portable. They are also of the lowest cost (Hiriart, 2003).
The binary cycles use a secondary working uid in a closed cycle.
Heat exchangers are used to transfer heat from the geothermal uid
to the working uid, the working uid is vaporized and expanded
in a turbine, and the cooled geothermal uid is reinjected to the
reservoir.
In this study the power plants used were:

Single ash power plant

Double ash power plant
Backpressure power plant
Binary power plant

1.4. Wellhead vs traditional power plants

Geothermal steamelds can have wells distributed across large
areas, with wells located far from one another. This raises a question

of whether to install wellhead power plants next to each well (see

Fig. 1), or a single, larger central power plant that uses the geothermal uid from several wells collected through a steam gathering
system (see Fig. 2).
The answer to which option is the best is not always a simple one. There are many factors that affect this decision, such as
site topography and the conditions of the geothermal uid. The
most important factor is the project cost, specically which option
makes the project a more protable investment. Project protability was the main objective of this study; an economical comparison
of traditional central power plants versus wellhead power plants.

1.4.1. Wellhead power plants

Wellhead power plants have been used in geothermal projects
for several decades but mainly in a limited capacity (Hiriart Le Bert,
1986). However, other applications that offer benets to geothermal projects have been explored more recently, such as peaking
units for larger geothermal projects (Sutter et al., 2012). These
smaller power plants can be built and brought online as soon as
each well is drilled and tested instead of having to wait for all
the wells in the steam eld to be drilled and tested, which typically takes three years for a traditional power plant (Cross and
Freeman, 2009, p. 17). Also, producing energy early in the project
makes policy makers more likely to support geothermal energy.
Another application of the small wellhead power plants is to use
them as complementary power plants to exploit the high pressure
(HP) wells or the low pressure wells (LP) in a steam eld, this application allows for the optimum power production from each well
instead of needing to balance the pressure of the entire steameld.
The main advantages of wellhead power plants are:
Wellhead power plants use steam from only one well, it is not
necessary to wait for all the wells from the steameld to be drilled
and tested, this allows for the power plant to start producing
energy at an earlier timescale than a traditional power plant.
There is no need for a large steam gathering system, this reduces
the pressure drops in the piping system and also lead to important
savings in the project.
Wellhead power plants can be used to produce energy
from remote wells in a steameld, as it is not practical to

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

WH

WH

STEAM
GATHERING
LINES
CENTRAL POWER
PLANT

TRANSMISSION
LINE TO GRID

WH

WH

WH

WELLHEAD

TRANSMISSION
LINE TO GRID
Fig. 2. Schematic layout of a central power plant.

transport high temperature two phase mixtures or steam over

long distances.
Wellhead power plants are designed for each well individually
and can produce optimum power from each well. However, it
is important to consider the pressure decline of the reservoir
throughout its lifecycle to decide on the optimum wellhead pressure.
The installation of wellhead power plants can serve as a monitoring tool to help in the design of other power plants for the
same steameld and also to conrm outcomes from reservoir
engineering that would help geothermal projects become more
sustainable.
In the case of a technical problem with any of the components
in the power plant, only the energy from one well would be
suspended and not of the entire steameld.
Wellhead power plants are more easily transportable, making
them a better option for cases in which they are going to be used
temporarily.
Some disadvantages of wellhead power plants are:

Longer transmission lines from each wellhead power plant as

they are scattered throughout the steameld.
Higher cost per installed kW, because of smaller sizes.
More cost in transformers and related equipment.
A separation station for each power plant instead of having only
one in the central power plants.
The effect of the reinjection of the separated geothermal uid of
the wellhead power plants on the reservoir while drilling is still
in progress in the rest of the steameld.
It is more likely that in cases where wellhead power plants are
used the reservoir is shared with other developers and this could
bring legal issues that should be considered.
1.4.2. Central power plant
Central power plants use geothermal uid from several wells.
The main advantages of a central power plant are:
Lower cost per installed kW.
Components are designed to the optimum for the steam conditions of the entire steameld.

Main disadvantages of central power plants:

Long time to get power output, mainly because the design of the
components begins until most of the wells are drilled and tested.
Central power plants need a steam gathering system to deliver
the geothermal uid to the power plant; this has an impact on
the cost of the geothermal project.
The gathering system has to balance the pressure from all the
wells and the design pressure of the separator going into the
power plant is that of the well with the lowest pressure, this is
less efcient for the high pressure wells.
Long and Harvey (2012) presented a nancial evaluation of a
geothermal project with the insertion of wellhead power plants but
limited their use to 5 MW in a 30 MW project. Elasson and Smith
(2011) compared central power plants and wellhead power plants
focusing on costs and nancial risk.
In this study a method was developed to compare a central
power plant and a wellhead power plant in economic terms; this
was done by calculating the net present value of cash ows based
on the power production from a hypothetical steameld and in different scenarios where the use of wellhead power plants may bring
benets to a geothermal project.
2. Methodology
2.1. Hypothetical steameld
The productivity of a well in steameld is commonly estimated
by dividing the installed capacity of the geothermal power plant
by the number of active wells. This available data is in units of
MW per well and is called well productivity (Bertani, 2005, p.
688). The data required for this study was the properties of the
wells (specically enthalpy and productivity curves) but this data
is more difcult to nd than well productivity data. For this reason a
hypothetical steameld was created using data from the steameld
in Hellisheidi, Iceland (Sigfusson et al., 2012). To have a balanced
steameld, 10 wells from that steameld were chosen, 3 wells with
high enthalpy, 3 wells with low enthalpy and 4 wells with medium
enthalpy.

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Table 1
Hypothetical steameld well properties: productivity curve parameters and
enthalpy (h).
Well

h (kJ/kg)

Type of well

1
2
3

0.003
0.008
0.0052

0.032
0.262
0.101

10.03
37.5
50.76

2660
2500
1990

High enthalpy

4
5
6
7

0.038
0.047
0.034
0.009

0.942
1.131
0.739
1.56

61.23
19.35
78.03
82.95

1800
1750
1740
1500

Medium enthalpy

8
9
10

0.179
0.038
0.059

3.265
0.578
1.261

15.27
22.78
43.33

1220
1170
1110

Low enthalpy

Table 2
Scenarios and subscenarios.
Scenarios
Permanent

Subscenarios
1
2
3

(1)

well is the mass ow of geothermal uid coming from each

where m
well and P is the wellhead pressure (see Fig. 3). In this study absolute
pressure values were used and the unit of measure was bar-a.
Productivity curves are different for each individual well and are
created during a testing period that usually takes between 3 and
6 months (Thorhallsson, 2012). The well properties of the hypothetical steameld created for this study (productivity curves and
enthalpies) are listed in Table 1. The function in Eq. (1) has been
represented by a 2nd order polynomial:
j = aj P 2 + bj P + cj
m

Complementary

9
10

Wellhead PP with condensing turbine

and traditional PP after all wells are
drilled and tested
Wellhead binary PP and traditional PP
after all wells are drilled and tested
Wellhead PP with backpressure
turbine and traditional PP after all
wells are drilled and tested
Wellhead PP for high pressure wells
with backpressure equal to inlet
pressure of traditional PP and
traditional PP for rest of wells.
Wellhead PP with condensing turbine
for low pressure wells and traditional
PP for rest of wells
Wellhead binary PP for low pressure
wells and traditional for rest of wells
Wellhead PP with backpressure
turbine for low pressure wells and
traditional PP for rest of wells

Table 2). All of the subscenarios utilized the geothermal uid from
the hypothetical steameld created, as dened in Section 2.1.

(2)

where P, pressure and j, number of well (1, . . ., 10) with the

coefcients a, b, c. These coefcients are determined after the test
period for each well in a 2nd order polynomial regression.
2.2. Scenarios
Three main scenarios were considered in this study. One in
which a type of power plant was installed from the beginning and
stayed through the entire planning horizon, this scenario is called
the permanent scenario. Another scenario where a combination of
power plants was used in sequential form, with wellhead power
plants in the early stage of the project and a central power plant
after that, referred to in this study as the early stage scenario. The
nal scenario is one in which the wellhead power plants are used
simultaneously with a central power plant; called a complementary
scenario. Each of these 3 scenarios has combinations of different
types of power plants, requiring the denition of subscenarios (see

Fig. 3. Example of a productivity curve of the hypothetical steameld.

The productivity curve is the relation between the pressure at

the wellhead and the mass ow of the geothermal uid from the
well, which is usually presented as a function of the pressure:
well = f (P)
m

Early stage

Single ash central power plant (PP)

with condensing turbine
Double ash central PP with
condensing turbine
Wellhead PP as a permanent option
with condensing turbine

2.2.1. Permanent scenario

In this scenario there are 3 subscenarios where the power plants
were installed on a permanent basis. The rst two: the single ash
central power plant and the double ash central power plant are
the traditional arrangements and were considered as a reference
to compare the other 8 subscenarios.
The third subscenario was the single ash wellhead power plant
with condensing turbine. The energy conversion cycle of this power
plant is the same as for the single ash power plant; the only difference is that this power plant is a wellhead power plant (see Fig. 1).
In this subscenario one such power plant was considered for each
of the wells in the hypothetical steameld. The power production
in this subscenario was considered to start progressively as each
well was drilled and tested.
The advantage of this third subscenario compared to the central
power plant was that the wellhead power plants start production
of energy and revenue as each well was drilled and tested instead
of having to wait for most of the wells in the steameld to be drilled
and tested.
2.2.2. Early stage scenario
In this scenario wellhead power plants were installed after each
well was drilled and tested, once the complete steameld was
drilled and tested a single ash central power plant replaced the
wellhead power plants.
The time interval between the start and the end of the drilling
stage can take several months and in some cases even years.
When the central power plant was ready to be operated the wellhead power plants were decommissioned, this means the wellhead
power plants were temporary. Within this scenario 3 subscenarios
were created. In the rst a wellhead single ash power plant with
a condensing turbine was used. In the second subscenario a binary
power plant was considered. The third subscenario is similar to the
rst, but with a backpressure turbine instead of a condensing turbine. Backpressure turbines do not produce as much electricity as

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Fig. 4. Schematic layout of complementary scenario for high pressure wells.

the condensing type, but are easier to move once the central power
plant is ready to be installed, and are cheaper due to having fewer
components. In this study the working uid chosen for the binary
power plant was methanol. Methanol was chosen because the temperature of the geothermal uid used to heat the working uid is at
a temperature close to 200 C, and it is important to use a working
uid that has a critical point above the temperature of the geothermal uid. The critical point of methanol is 240 C at 78.5 bar-a. Most
common working uids, like isopentane or isobutene, have critical
points lower than 200 C.
As mentioned in the introduction, one important advantage of
wellhead power plants is that they can produce optimum power
from each well. In these scenarios the separator pressure was calculated to optimize power output for each well.
An important factor in this scenario was to consider a resale
value or scrap value of the wellhead power plants for the nancial
calculations; this will be further discussed in the cost estimation
section.

amount of energy from the high pressure wells. An important thing

to consider in this subscenario is that the outlet pressure for the
wellhead power plants has to be the separator inlet pressure for
the medium pressure power plant, so in the arrangement for this
scenario the wellhead power plant would not be condensing at a
low pressure.
In the subscenarios where the wellhead power plants were used
for the low pressure (LP) and low enthalpy wells in the steameld the wellhead power plants were installed separately so that a
condensing turbine could be used in the central power plant that
used the rest of the wells, allowing for optimum power production
(see Fig. 5). Typically low pressure wells are discarded as dead
so that they do not lower the design pressure of the power plant.
By exploiting the low pressure wells with independent wellhead
power plants, some energy could be extracted from them without
any adjustments to the central power plant.

2.2.3. Complementary scenario

In this scenario the wellhead power plants were used at the
same time as the central power plant. The wellhead power plants
were used, either for the high pressure wells or for the low pressure wells in the steameld, while the central power plants were
used for the other wells. This scenario was divided into four subscenarios. In the rst one the wellhead power plants were installed
to utilize the geothermal uid from the high enthalpy wells and
in the latter three subscenarios the wellhead power plants utilized
the geothermal uid from the low enthalpy wells.
The high enthalpy wells in the hypothetical steameld are wells
13 and the low enthalpy wells are wells 810 (see Table 1). The
rest of the wells are utilized in a single ash central power plant.
In the rst subscenario, the wellhead power plants were used
to produce electricity from the wells with the higher pressure (HP)
and higher enthalpy, then the geothermal uid exiting the wellhead plants was reused, combined with the geothermal uid from
the medium enthalpy wells, in another power plant (the medium
pressure power plant) (see Fig. 4). This can help the overall power
production of the steameld because in this way the complementary wellhead power plant would be producing the optimum

The power output calculations were done using the EES software
(Engineering Equations Solver version 9.203) (Klein and Alvarado,
2002) with the equations governing the power cycles for each type
of power plant in every scenario (DiPippo, 2008).
It is important to note that the calculations results for the high
enthalpy wellhead power plants (subscenario 7) resulted in an optimum separator pressure that is above what currently available
wellhead power plants can achieve. Despite this, it was decided to
perform the calculations with these results to illustrate the potential of those wells.

2.3. Power output calculations

2.4. Calculation of time difference (TD)

Geothermal projects are known to take a long time to start production, approximately 5 years (Islandsbanki, 2011). The objective
of this section was to calculate how much earlier a wellhead power
plant can start production before a central power plant does. This
was dened in this study as time difference (TD) (see Fig. 6). The
calculation of the TD was achieved by focusing on the time it takes
to drill each well and the fact that each wellhead power plant can
be installed once each well is drilled and tested but a central power

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Fig. 5. Schematic layout of complementary scenarios for low pressure wells.

plant has to have most of the wells drilled and tested before it can
operate. This TD is dependent on the number of wells in a steameld and the number of drilling rigs operating in the steameld. For
this study one drilling rig was assumed.
The time it takes to drill one well is considered to be:

where WHT is the time for a wellhead power plant to start production and CT is the time for a central power plant to start production.
In all the 10 subscenarios the TD was the same.

Average 43.5 days with a standard deviation of 5.5 days

(Thorhallsson and Sveinbjornsson, 2012)
Time to test the wells: 3 months 90 days (Thorhallsson, 2012)

For this study the cost estimation was done to measure the feasibility of the subscenarios. The type of cost estimate used in this
study is what is commonly referred to as order of magnitude.
The costs considered for the power plant options discussed in
this study were:

The time was calculated in the following way:

WHT, time to drill 1 well with a (95% certainty) + time to test the
wells
CT, time to drill 10 wells with a (95% certainty) + time to test the
wells

2.5. Cost estimation

Cost of power plant

Cost of operation and maintenance of a geothermal power plant
(O&M)
Cost of transmission lines for the wellhead power plants
Cost of steam gathering system for the central power plants

And then
TD = CT WHT

2.5.1. Investment cost of power plants

From communication with industry participants it was found
that the cost of a geothermal single ash power plant of 5 MW with
condensing turbine is 1700 USD/kW, in the case of the binary power
plant the capital cost is considered 34.85% higher (Hance, 2005)
than the ash steam power plant and the back pressure power
plant is considered to be 1500 USD/kW for the 5 MW power plant
(Long and Harvey, 2012). These costs include: steam separators,
well connection, civil works, power houses, electric and mechanical installations, switches and controls, security facilities, generator
and everything needed for the installation of the power plants to
start running. To account for economies of scale, the following
equation was used (Sanyal, 2005):
CC = CPP e0.0025(W 5)

Fig. 6. Visual representation of the time difference (TD).

(3)

where CC is the capital cost in USD/kW, CPP is the cost per kW of the
power plant, and W is the gross power output of the power plant
in MW.
In the options where the wellhead power plants were installed
in the early stages of development while the central power plant
was being built, it was important to consider the resale value to
be used. The only information found on some resale value for

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

wellhead power plants was in a case where the wellhead power

plant is assumed to be sold after 10 years for 40% of its initial value
(Long and Harvey, 2012) and another where it is mentioned that
wellhead power plants claim to have a scrap value of 70% (Elasson
and Smith, 2011). Because the resale value was calculated for a
much shorter time than the cases mentioned (between 6 and 24
months), for this study the resale value considered was 90%, the
cost of uninstallation was not considered. The resale value was
considered in the cash ows at the same time that the central
power plant was installed.
2.5.2. Cost of operation and maintenance
This is the cost once electricity production starts, and it is also
affected by economies of scale. For this study the following equation
was used (Sanyal, 2005):
CO&M = 2 e

0.0025(W 5)

(4)

where CO&M is the cost of operation and maintenance (O&M) in US

cents/kWh and W is the gross power output of the power plant.
This assumes an O&M cost of 2 US cents per kWh for a 5 MW
single ash power plant in 2005, but after analyzing more recent
data it was found that O&M costs have declined over the last decade
(Entingh and McVeigh, 2003) and even in some cases the cost considered in recent papers as low as 1.6 US cents per kWh (Long and
Harvey, 2012). It was decided to use 2 US cents per kWh for all
plants in this study, except for the binary power plant, which was
given a 35% increase in the O&M costs based on communication
with the industry.
2.5.3. Cost of transmission lines
The transmission lines considered in this study were the ones
that go from each wellhead power plant to the central transformer
station. This study did not consider the transmission lines to the
grid.
The cost of transmission lines considered for this study was 100
USD/kW (Hance, 2005).
2.5.4. Cost of steam gathering system
This is the cost of the steam gathering system that transports the
steam or two-phase geothermal uid from the geothermal wells
to the central power plant. The steam pipeline system inside the
power plants was not considered in this study. The cost of the
steam gathering system can vary a lot between steamelds since
it depends mostly on the length of the pipelines. From research on
the cost of geothermal projects it was estimated that the cost of the
steam gathering system was 250 USD/kW (Hance, 2005).
2.6. Economic analysis
2.6.1. Revenue estimation
Together with the cost estimation results it was important to
estimate the revenues from each of the power plants according to
the net power output calculations. In order to do that the power
output results were converted to energy produced during one year
using a capacity factor:
Revenue
= NPO 365 24 c.f. price
year

(5)

where c.f., capacity factor; NPO, net power output = gross power
output parasitic loads; price, the price of electricity, in this study
the price considered is 0.1 USD/kWh.
For this study the capacity factor considered for all the power
plants was 90% which is well in the range of geothermal power
plants (Gehringer, 2012). Capacity factor is dened as the electricity
produced in a period of time divided by electricity the power plant

would produce at full nameplate capacity for the same period of

time.
c.f. =

electricity produced in a period of time

name plate capcity period of time

(6)

2.6.2. NPV comparison

The analysis in this study was focused on the differences in
power output and cost of each of the scenarios created. A net
present value calculation was used to compare all the scenarios and
determine which was better with regards to the nancial aspects.
The net present value (NPV) is a method that brings future cash
ows to present values using a given discount rate, the formula to
do that is (Park, 2002):
NPV =

n

i=0

CFi
(1 + dr)

(7)

where i is the time of the cash ow which is divided in periods,

CFi is the cash ow of the period and dr is the discount rate for the
period.
The discount rate is a key variable of this process and is a measure of the difference in value an investor puts on money in the
present vs. the future.
The net present value was calculated creating cash ows of the
different scenarios. Negative cash ows were investment costs and
O&M costs; and positive cash ows were revenues and the resale
value. The planning horizon considered was 20 years xed from
the start of production of the wellhead power plants. The cash ows
were divided into quarterly cash ows; i.e. cash ows are calculated
for periods of 3 months each, which means there are 80 periods in
this NPV analysis. The discount rate used for these calculations was
15% per year. In this study ination was not considered.
Cash ow for the wellhead power plants started in the rst
period and the central power plants started after the time difference
(TD).
2.6.3. Sensitivity analysis
The sensitivity analysis in this study was done to determine the
variation in the results of net present value due to changes in 3 key
factors. The key factors considered for the sensitivity analysis were:
1. The time difference (TD): This factor affects the net present value
calculation of subscenario 1 and 2 where central power plants
were considered and it means that these central power plants
would be operational, and revenue would start, after the TD
determined for this sensitivity analysis. The sensitivity analysis
was done for a TD of 6, 12, 18 and 24 months.
2. Order in which the wells were drilled: this factor affects only
the subscenarios where wellhead power plants were involved.
It is not relevant for central power plants since for this study the
central power plants require all the wells in the steameld to be
drilled before they start production. The sensitivity analysis was
done considering the following options of drilling order:
2.1. start with the high enthalpy wells (H)
2.2. start with low enthalpy wells (L)
3. Rate at which the wellhead power plants were installed: this factor is relevant to the subscenarios where wellhead power plants
are involved, and it has an impact on how fast the wellhead
power plants can start production and revenue. The sensitivity
analysis was done considering the following rates of installation:
3.1. One wellhead every 3 months (3)
3.2. One wellhead every 1 month (1)
The base case scenario considered in this study according to the
previous key factors was:

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Table 3
Power output, revenue and cost results of the single ash and the double ash central power plants.
Subscenarios

Power output results

Revenue and cost results

Separator pressure
(bar-a)

Medium pressure
separator (bar-a)

Temp. ( C)

Mass ow of
wells (kg/s)

Net power
(KW)

Revenue
(103 $/year)

Power plant
cost (103 $)

O&M
(103 $/year)

Steam gathering
(103 $)

13

191
.6

438

115,047

90,703

142,732

14,006

29,438

15

198
.3

427
.9

127,227

100,305

152,152

15,025

32,592

Time difference equal to TD calculated in Section 3.1, the order

in which the wells were drilled was considered to have started
with the high enthalpy wells (H) and the rate of installation of the
wellhead power plant was 1 every 3 months (3).
2.7. Main assumptions
Main assumptions of all scenarios
Pressure drops in the power plants pipelines are not considered
Same type of components where applicable: i.e. vertical separators, axial centrifugal pump, direct contact condenser, induced
counter ow cooling towers.
The condensing method considered for the binary power plant
was an air cooled condenser.
Same dry bulb temperature and wet bulb temperature.
Condensing pressure is assumed to be 0.1 bar-a except in backpressure power plants were 1013 bar-a is considered.
Same steameld.
Installation times are considered equal for all scenarios, the time
difference (TD) considered in this study is determined by the time
taken to drill the necessary wells to start production for each type
of power plant: wellhead or central.
Non-condensable gases extraction system not considered.
No thermal losses considered.
No changes in potential or kinetic energy considered.
No pressure decline of the reservoir throughout the time horizon
of the project.
Turbine efciency of the turbine of 82%.
Although transmission lines and steam gathering systems are a
part of both wellhead and central power plants, for this study the
transmission lines were only considered for the wellhead power
plants and the steam gathering system was considered only for
the central power plants. The costs of transmission lines and the
cost of the steam gathering system were not considered to be
affected by economies of scale for this study.

The costs considered in this study were overnight costs; this is

the lump sum that would be paid up front for a project without
considering interests for the time it takes to nish it.
All costs are per kW of installed capacity, except cost of operation
and maintenance which are per kWh. The currency used is US
dollars (USD).
Assumptions for the central power plants.
The separator pressure was calculated using the geothermal uid
from all the wells in the hypothetical steameld.
One dual pressure turbine was considered for the double ash
power plant.
Assumptions for the Wellhead power plants:
The separator pressure is calculated for each power plant depending on the properties (productivity curve and enthalpy) of each
well.
The order in which the wells are drilled is considered to begin
with well number 1 and end with well number 10. This is relevant in the case where the wellhead power plants are considered
because each well has different power outputs.
The time to install each wellhead power plant after the rst one
is installed was considered to be 3 months.
3. Results and discussion
This section presents the main results for each scenario, being
the power output, the time difference (TD) calculation, revenue,
cost estimation, the net present values and the sensitivity analysis.
Since the time difference (TD) is an important input for the net
present value calculation and as it was also used in all scenarios it
will be calculated at the beginning of this results section.
The complete results of all the scenarios can be viewed in the
thesis which forms the basis for this paper (Crdova Geirdal, 2013).
3.1. Time calculation results
Following the method described in Section 2.4 the time difference calculated for this study was 12 months, this means that the

Table 4
Power output, revenue and cost results of the wellhead power plants with condensing turbines.
Well

1
2
3
4
5
6
7
8
9
10
Total

Power output results

Revenue and cost results

Separator
pressure (bar-a)

Temp. ( C)

Well mass
ow (kg/s)

Net power
(KW)

Revenue (103
$/year)

19
27
22
16
13
15
5
8
6
6

209.8
228.1
217.3
201.4
191.6
198.3
151.9
170.4
158.9
158.9

9.555
38.74
50.47
66.57
26.11
81.47
74.93
29.93
24.88
48.77

5583
21,608
18,483
19,760
7248
22,568
14,295
3838
2922
5040

4401
17,035
14,571
15,578
5714
17,792
11,270
3025
2303
3973

9676
35,599
30,768
32,812
12,519
37,167
24,323
6719
5142
8811

570
2203
1886
2020
741
2308
1473
394
300
518

897
3330
2873
3066
1162
3478
2267
622
476
817

451.4

121,345

95,668

203,540

12,417

18,993

Power plant
cost (103 $)

Transmission
cost (103 $)

O&M cost
(103 $/year)

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

wellhead power plants started production, and revenue, 12 months

before the central power plant was ready to be operational.

Table 5
Net present value results of the permanent scenario.
Net present value (103 $)

Subscenarios

3.2. Permanent scenario

In this scenario the power plants were installed on a permanent
basis. The rst two subscenarios are central power plants and these
were installed after all the wells in the steameld were drilled and
tested. The rst subscenario is the single ash central power plant,
the second subscenario is the double ash central power plant, and
these two subscenarios serve as reference to compare the rest of
the subscenarios. In the third subscenario wellhead power plants
are used, in this case the wellhead power plants used condensing
turbines. The wellhead power plants were installed after each of
the wells in the steameld were drilled and tested.
3.2.1. Power output results of the permanent scenario
The power output results for the single ash central power plant
(subscenario 1) and for the double ash central power plant (subscenario 2) are shown in Table 3. The power output results of the
wellhead power plants with condensing turbines (subscenario 3)
are shown in Table 4. As expected the power output of subscenario
3 is higher, with an increase of 5% in power output, than the single ash central power plant (subscenario 1); the reason is that the
wellhead power plants were designed to get the maximum power
output for each well.
3.2.2. Revenue and cost results of permanent scenario
The revenues and cost results for the permanent scenario are
shown in Tables 3 and 4. The rst two subscenarios, where central power plants are used, include the cost of the steam gathering
system but not the costs of transmission lines, the cost of the
transmission lines is only considered for the subscenarios where
wellhead power plants are used as mentioned in Section 2.5.1.
3.2.3. Net present value results of the permanent scenario
The results of the net present value of the permanent scenario
are shown in Table 5.
3.2.4. Sensitivity analysis of permanent scenario
As mentioned in Section 2.6.3, the time difference (TD) has an
impact on the NPV results of the central power plants (subscenario 1 and subscenario 2). The results of the sensitivity analysis
of these subscenarios are shown in Table 6,where the NPV of both
central power plants decreases as the TD increases. Four different
results were obtained from the sensitivity analysis of subscenario
3.Of these 4 results, the lowest NPV was obtained when the rate
of installation was 1 wellhead power plant every 3 months (3H
and 3L). The NPV showed much higher results when the rate of

1
2
3

271,179
307,099
239,361

Table 6
Net present values (in 106 USD) calculated in the sensitivity analysis for subscenario
1 and subscenario 2.
Time difference (TD)
Subscenario 1
6
294

12
271

Subscenario 2
18
250

24
230

6
333

12
307

18
283

24
261

Table 7
Net present values (in 106 USD) calculated in the sensitivity analysis of subscenario
3.
Order of drilling

Rate of installation

Time difference (TD)

Subscenario 3
6

Starting with high

pressure wells (H)
Starting with low
pressure wells (L)
Starting with high
pressure wells (H)
Starting with low
pressure wells (L)

12

18

1 every three
months (3)
1 every three
months (3)
1 every month (1)

239

1 every month (1)

260

24

222
267

installation of the wellhead power plants was 1 wellhead power

plant every month (1H and 1L). In both cases of varying the rate
of installation, the NPV was higher when the high pressure wells
were drilled rst, rather than the low pressure wells (3H and 1H).
When the time difference (TD) is greater, the wellhead power plants
become a better option than the central power plants. Itwas interesting to see that for a time difference (TD) greater than 12 months,
the wellhead options NPV was between 4% and 16% greater than
the single ash central power plant if one wellhead power plant
was installed per month (see Tables 6 and 7).
3.3. Early stage scenario
As mentioned in Section 2.2.2, in this scenario the wellhead
power plants were installed after each well was drilled and tested
and when all the wells in the steameld were drilled and tested
a single ash central power plant replaced the wellhead power
plants. Three subscenarios (subscenarios 46) were created in this

Table 8
Power output, revenue and cost results of the binary wellhead power plants.
Well

1
2
3
4
5
6
7
8
9
10
Totals

Power output results

Revenue and cost results

Optimum separator
pressure (bar-a)

Temp. ( C)

Well mass
ow (kg/s)

Net power
(KW)

17.2
24.92
25.46
18.31
14.54
18.49
6.39
9.05
8.31
6.97

204.9
223.8
225
208
196.9
208.5
161.4
175.7
172
164.9

9.693
39.06
49.96
65.74
25.86
80.07
72.60
30.16
24.96
49.26

5364
20,948
21,009
24,113
8948
28,335
16,379
4517
3340
5770

447

138,723

Revenue
(103 $/year)

Power plant
cost (103 $)

Transmission
cost (103 $)

O&M cost
(103 $/year)

4228
16,515
16,563
19,010
7054
22,339
12,913
3561
2633
4549

14,357
53,404
53,580
60,463
23,628
70,011
42,299
12,121
9006
15,451

628
2471
2480
2828
1047
3324
1925
529
391
677

988
3709
3722
4207
1629
4884
2930
833
619
1063

109,369

354,322

16,304

24,587

10

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Table 9
Power output, revenue and cost results of wellhead power plants with backpressure turbines.
Well

1
2
3
4
5
6
7
8
9
10

Power output results

Optimum ash
Temp. ( C)
power (bar-a)

(kg/s)
Well m

Net power
(KW)

Revenue and cost results

Power plant
Revenue (103
$/year)
cost (103 $)

Transmission
cost (103 $)

23
32
31
20
16
20
10
10
10
10

9.179
37.69
48.89
64.87
25.41
79.21
66.45
30.02
24.76
50.04

3642
14,695
12,418
12,586
4475
14,334
7751
2171
1595
2749

2871
11,585
9790
9922
3528
11,300
6110
1711
1257
2167

5485
21,410
18,217
18,454
6723
20,907
11,530
3284
2417
4151

364
1469
1241
1258
447
1433
775
217
159
274

576
2261
1922
1947
706
2208
1213
344
253
435

447

76,416

60,246

112,581

7641

11,869

219.6
237.5
235.7
212.4
201.4
212.4
179.9
179.9
179.9
179.9

Total

O&M (103
$/year)

scenario. The only difference between the three subscenarios was

the type of wellhead power plant used in the early stage of the
project. In subscenario 4 the wellhead power plants considered
were with condensing turbines. In subscenario 5 binary wellhead
power plants were considered, and in subscenario 6 wellhead
power plants with backpressure turbines were used.

Table 10
Net present value of the early stage scenario.

3.3.1. Power output results of early stage scenario

The power output for this scenario was the combination of the
power output of the wellhead power plants for the early stage of
the project and the power output of the single ash central power
plant after all the wells were drilled and tested. Therefore the power
output result of subscenario 4 was the combination of the power
output result of subscenario 3 (see Table 4) for the early stage of
the project and of subscenario 1 (see Table 3) once all the wells
in the steameld were drilled and tested. For subscenario 5 the
power output was the combination of the power output result of
the binary wellhead power plant (see Table 8) for the early stage
and of subscenario 1 after that. The power output result of subscenario 6 was the power output result of the wellhead power plant
with backpressure turbine for the early stage (see Table 9) and of
subscenario 1 after that.

and cost of the single ash central power plant installed once all
the wells were drilled and tested (see Table 3). In this scenario
a resale value of the wellhead power plants was included in the
calculations when they were replaced by the central power plant.

3.3.2. Revenue and cost results of early stage scenario

In a similar way to the power output results, the revenue and
cost results of the early stage scenario was calculated as the combination of the revenue and cost of the wellhead power plants used in
the early stage of the project (see Tables 4, 8 and 9) and the revenue

Subscenarios

Net present value ($)

4
5
6

268,628,886
256,519,329
270,979,452

3.3.3. Net present value results of early stage scenario

The results of net present value for the early stage scenario are
shown in Table 10.
3.3.4. Sensitivity analysis of early stage scenario
The results of the sensitivity analysis of the early stage scenario
are shown in Table 11. In this scenario the 3 key factors affect all
the subscenarios.
In the early stages scenario, if the time difference (TD) was
greater than 18 months the wellhead power plants with condensing turbines (subscenario 4) were a better option compared to the
single ash central power plant (subscenario 1) even when the rate
of installation of wellhead power plants was one every 3months,
between 2% and 8% greater NPV, and better than the double ash
central power plant (subscenario 2) when the rate of installation is
1 every month (see Tables 6 and 11), as much as 5% greater NPV.

Table 11
Net present values (in 106 USD) calculated in the sensitivity analysis of early stage scenario.
Order of drilling

Rate of installation

Time difference (TD)

Subscenario 4

Starting with high pressure wells (H)

Starting with low pressure wells (L)
Starting with high pressure wells (H)
Starting with low pressure wells (L)

1 every three months (3)

1 every three months (3)
1 every month (1)
1 every month (1)

Subscenario 5

Subscenario 6

12

18

24

12

18

24

12

18

24

290
293
284
286

269
269
276
269

255
250
275
269

249
236
275
268

287
292
266
276

257
264
250
246

235
239
245
240

223
217
240
235

292
294
289
288

271
271
278
271

256
251
270
265

246
236
263
258

Table 12
Power output results of the complimentary scenario with wellhead power plants for HP wells (subscenario 7).
Well
WHPP for high pressure wells

Medium pressure power plant

Total

Separator pressure (bar-a)

Well mass ow (kg/s)

Net power (KW)

1
2
3

27

97.15

10,459

410

10

443.25

109,821
120,280

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

11

Table 13
Power output results for subscenarios 810.
Well

Subscenario 8

Subscenario 9

Subscenario 10

Separator
pressure
(bar-a)

Well m
(kg/s)

Net power
(KW)

Separator
pressure
(bar-a)

Well mass
ow (kg/s)

Net power
(KW)

Separator
pressure
(bar-a)

Well m
(kg/s)

Net power
(KW)

Traditional single ash

power plant for High
pressure wells

17

14

334.6

105,093

14

334.6

105,093

14

334.6

105,093

WH power plants for low

pressure wells

29.93

3838

9.05

30.16

4517

10

30.02

2171

9
10

6
6

24.88
49

2922
5040

8.31
6.97

24.96
49.26

3340
5770

10
10

24.76
50

1595
2749

438

116,893

439

111,608

Total

439

118,720

Table 14
Revenue and cost results of complementary scenario.
Subscenarios

7
8
9
10

Revenue

Cost

Revenue ($/year)

Power plant cost ($)

O&M ($/year)

Transmission ($)

Steam gathering ($)

97,006,313
92,158,441
93,598,848
87,991,747

156,088,419
155,064,745
170,970,170
144,244,334

15,192,535
15,037,622
15,637,544
14,155,165

1,045,900
1,213,300
1,598,400
651,500

28,145,750
26,880,500
26,880,500
26,880,500

Of the three scenarios where the wellhead power plants were

installed in the early stages, the binary (subscenario 5) was the least
protable, although it had the highest power output (20% higher
power output than the single ash power plant), mainly because
of its high capital cost and high O&M costs (see Table 8). Another
factor that could affect this scenario was the assumption made in
this study of excluding the non-condensable gas (NCG) extraction
system. If it had been considered, the NCG extraction system would
reduce the power output of the steam cycle power plants relative
to the binary plants as NCG extraction is not required for binary
power plants.
If the time difference (TD) was as small as 12 months, the better
option was subscenario 6 which used wellhead power plants with
backpressure turbines (see Tables 11 and 6), the NPV was as much
as 14% greater than the single ash power plant (subscenario 1).
3.4. Complementary scenarios
In this scenario there were 4 subscenarios (subscenario 710)
and in all of these subscenarios the wellhead power plants were
used at the same time as the single ash central power plant. In
subscenario 7 the wellhead power plants were used with the high
pressure wells (see Table 1) while the rest of the wells were used
by a central power plant. In subscenarios 810 the wellhead power
plants were used with the low enthalpy wells (see Table 1) while the
rest of the wells were used by a central power plant. The only difference between the subscenarios where the wellhead power plants
were used for the low pressure wells was the type of wellhead
power plant, in subscenario 8 the wellhead power plants considered were with condensing turbines, in subscenario 9 the wellhead
power plants used were binary wellhead power plants and in subscenario 10 the wellhead power plants used were wellhead power
plants with backpressure turbines.
3.4.1. Power output results of complementary scenario
The power output results for subscenario 7 are shown in
Table 12, as mentioned in Section 2.2.3 the geothermal uid from
the high pressure wells (wells 13) were used in the wellhead
power plants and then, together with the geothermal uid from
the rest of the wells (wells 410), used in the central power plant.

Therefore the condensing pressure of the wellhead power plants

was assumed to be equal to the inlet pressure of the separator of
the central power plant. With this arrangement it was possible to
get 4% more power from the steameld.The power output results
of the subscenarios where wellhead power plants were used for the
low pressure wells are shown in Table 13.
3.4.2. Revenue and cost results of the complementary scenario
The revenue and cost results for the complementary scenario
are shown in Table 14.
3.4.3. Net present value results of complementary scenario
The net present value results of the complementary scenario are
shown in Table 15.
3.4.4. Sensitivity analysis of complementary scenario
Of the complementary scenarios, the only scenario where the
use of wellhead power plants had some benet was the option
where the wellhead power plants were used for the high enthalpy
wells (subscenario 7), in this subscenario the NPV was greater than
the single ash central power plant (subscenario 1) by 7% even with
the TD of 6 months. In this subscenario the only factor relevant in
the sensitivity analysis was the time difference (TD), as the other
factors which are: the order of drilling and the rate of installation,
did not affect the results (see Table 16).
For the other three options, where the wellhead power plants
were used for the low enthalpy wells (subscenario 810), the NPV
was worse than if those wells had been used in the single ash central power plant even though the optimum separator pressure for
the central power plant using the high pressure wells was higher
because the low pressure wells were handled separately but it did
not compensate for the additional cost of the low pressure power
Table 15
Net present value of the complementary scenario.
Subscenarios

Net present value ($)

7
8
9
10

291,519,995
267,080,461
257,089,397
257,745,210

18

239
239
239
239

12

258
258
258
258

278
278
279
279

24

221
221
221
221

18

238
238
239
239
257
257
258
258

12
6

277
278
278
278
231
231
232
232

24
18

248
248
249
249
267
267
268
268
287
288
289
289
251
251
251
251

12
6
24
18

271
271
271
271
292
292
292
292

12
6

1 every three months (3)

1 every three months (3)
1 every month (1)
1 every month (1)
Starting with high pressure wells (H)
Starting with low pressure wells (L)
Starting with high pressure wells (H)
Starting with low pressure wells (L)

314
314
314
314

Subscenario 10
Subscenario 9
Subscenario 8
Subscenario 7

Rate of installation

Time difference (TD)

plants. It is possible that the results would be different in a steameld with wells with lower enthalpies or lower mass ows than
those from the hypothetical steameld. In these scenarios there
was no benet from a time difference (TD) between the wellhead
and the central power plants, because both types of power plants
were used simultaneously, so the NPV follows the same trend as
the NPV of the central power plants.
4. Conclusions

Order of drilling

Table 16
Net present values (in 106 USD) calculated in the sensitivity analysis of complementary scenario.

221
221
222
222

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

24

12

In the three main scenarios created in this study, important

benets were obtained from the use of wellhead power plants
depending on some factors:
Time difference between the start of production with a wellhead
power plant and a central power plant. The longer the time difference between the start of production of a wellhead power
plant and a central power plant, the more protable the wellhead
option became.
Another important factor was the rate at which the wellhead
power plants were installed. If the installation was at the rate of
one wellhead power plant every month, then the scenarios where
the wellhead power plants were installed in the early stages of
development were more protable than the central power plants
with time difference (TD) as small as 12 months. Also in the permanent scenario the benet was greater in the cases where the
wellhead power plants were installed once a month.
Of the three factors analyzed in the sensitivity analysis the less
relevant was the order of drilling. This factor was incorporated
to have two extreme cases. On the high end starting with the
high enthalpy wells, and on the low end starting with the low
enthalpy wells. The difference in NPV with this factor was less
than with the other two factors: time difference (TD) and rate to
install wellhead power plants.
In the complementary scenarios the only increase in NPV was
observed in the scenario where the wellhead power plants were
used for the high enthalpy wells.
This study was done considering one hypothetical steameld
with ten subscenarios, consisting of different congurations of
power plants. Some assumptions were made in order to make the
comparisons on an even basis and to establish the methodology.
In future work it is recommended to make a more detailed model
to calculate the power output and do a more detailed cost analysis
with a closer look at the resale value.
References
Bertani, R., 2005. World geothermal power generation in the period 20012005.
http://dx.doi.org/10.1016/j.geothermics.
Geothermics
34,
651690,
2005.09.005.
Bertani, R., 2012. Geothermal power generation in the world 20052010
update report. Geothermics 41, 129, http://dx.doi.org/10.1016/j.geothermics.
2011.10.001.
Crdova Geirdal, C.A., 2013. Economic Comparison between a Well-head Geothermal Power Plant and a Traditional Geothermal Power Plant. Reykjavik
University, Reykjavik, Iceland.
Cross, J., Freeman, J., 2009. 2008 Geothermal Technologies Market Report.
DiPippo, R., 1999. Small geothermal power plants: design, performance and economics. GHC Bull. (June).
DiPippo, R., 2008. Geothermal Power Plants: Principles, Applications, Case Studies
and Environmental Impact. Elsevier, Oxford, UK.
Elasson, L., Smith, C., 2011. When smaller is better-cost/size/risk analysis of geothermal projects. In: Proceedings of Kenya Geothermal Conference 2011. Presented
at the Kenya Geothermal Conference 2011.
Entingh, D.J., McVeigh, J.F., 2003. Historical improvements in geothermal power
system costs. Trans. Geotherm. Resour. Counc., 533538.
Gehringer, M., (Technical report) 2012. Geothermal Handbook: Planning and Financing Power Generation. Energy Sector Management Assistance Program.
Gelman, R., 2010. 2009 Renewable Energy Data Book US DOE (Energy Efciency &
Renewable Energy). National Renewable Energy Laboratory.

C.A. Crdova Geirdal et al. / Geothermics 53 (2015) 113

Hance, C.N., 2005. Factors Affecting Cost of Geothermal Power Development.
Hiriart, G., 2003. A new 100-MW Geothermal power project starts operations Near
Morelia, Michoacn in Central Mexico. Geotherm. Resour. Counc. Bull., 208211.
Hiriart Le Bert, G., 1986. Developments in geothermal energy in Mexico part
three: economics of wellhead versus central power plants. J. Heat Recov. Syst.
6, 191200.
Islandsbanki, 2011. US Geothermal Energy Market Report. Islandsbanki.
Klein, S.A., Alvarado, F.L., 2002. Engineering Equation Solver.
Long, M., Harvey, W., 2012. Staged Asset Deployment Commercial and Technical
Advantages of using a wellhead generation unit. In: Presented at the African Rift
Geothermal Conference 2012, Nairobi, Kenya.
Park, C.S., 2002. Contemporary Engineering Economics. Prentice Hall Upper Saddle
River, NJ.

13

Sanyal, S.K., 2005. Cost of geothermal power and factors that affect it. In: Proceedings
of the World Geothermal Congress.
Sigfusson, B., Kjartansson, G., sbjrnsson, E.J., 2012. Well productivity report
Hellishedi. Orkuveita Reykjavikur.
Sutter, J., Kipyego, E., Mutai, D., 2012. The use of portable geothermal wellhead generators as small power plants to accelerate geothermal development and power
generation in Kenya. In: Thirty-Seventh Workshop on Geothermal Reservoir
Engineering Stanford University, Stanford, CA.
Thorhallsson, S., 2012. Subject: Test Times for Geothermal Wells.
Thorhallsson, S., Sveinbjornsson, B.M.,2012. Geothermal drilling cost and drilling
effectiveness. In: Presented at the Short Course on Geothermal Development
and Geothermal Wells. UNU-GTP, Santa Tecla, El Salvador.