Developing Low Temperature Geothermal Projects in
Developing Low Temperature Geothermal Projects in
Developing Low Temperature Geothermal Projects in
Email: [email protected]
1. Background
For electricity, geothermal production efforts have historically been focused on developing high
temperature and high enthalpy resources utilising conventional steam (either dry steam or flashed steam)
technology, and more recently binary and hybrid cycle plants (e.g., Ngatamariki, Olkaria III, Sarulla,
and Cerro Pabellon). While low and medium temperature fields are common amongst newly developed
projects internationally (e.g., Turkey, USA, and Honduras), none have yet been developed in Indonesia.
Whether these can be commercially competitive against traditional higher temperature project options
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
is of significant interest given plans for increasing power generation from geothermal resources in
Indonesia.
Most of Indonesia’s high temperature resources are associated with volcanic areas and have either
been developed, are caught up in ‘red tape’ or are slow to move due to various financial or permitting
issues. However, at least 2,000 MWe of low to medium temperature reserves are identified in Indonesia
[2]. In fact, water-dominated systems with temperatures between 110 and 160°C are believed to be the
most abundant geothermal energy resources globally [4]. Increasingly the projects being investigated
in detail by Badan Geologi (Geological Agency of Indonesia) are reported to have modest temperatures
and/or a higher degree of resource temperature uncertainty.
If Indonesia is to achieve its climate change commitments it will need to considerably increase the
renewable energy contribution to the electricity grid. Given the importance of geothermal energy to
Indonesia’s renewable energy portfolio, this will likely require a significant commitment to utilizing
low and medium temperature geothermal resources for electricity generation.
In this paper we consider the viability of using pumped technology for developing geothermal
systems in Indonesia with temperatures less than ~200°C.
2. International Trends
While most geothermal development has been at systems with resource characteristics that allow for
self-flowing wells, the introduction of binary cycle plants over 35 years ago in the USA led to the
development of several systems that were pumped and now have long operating history (e.g. East Mesa:
1980, Heber: 1985, Steamboat: 1986). The number of lower temperature (<210 °C) resources and
pumped systems developed has been increasing in recent years so that now approximately 7 - 10% of
all global geothermal generation comes from these types of systems, as documented in Figure 1.
Figure 1. Regions where <210°C geothermal reservoirs have been developed for power generation.
Showing average reservoir temperature and net installed capacity (world basemap with highlighted
geothermal regions from www.ThinkGeoEnergy.com).
2
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
Investigation of recently developed geothermal projects in the USA shows that the last known steam
turbine plant in the country was commissioned in 2012. All nine projects developed since 2012 appear
to be low temperature pumped projects and have tariffs lower than the final 50 MW steam turbine plant
(USc/kWh 13), even when many have smaller units or stages (Table 1).
Table 1. Recent pumped geothermal projects in the USA and Honduras (partially derived from [6]).
From this data, it is apparent that not only are pumped projects proving to be both technically and
financially feasible, but also in some circumstances they may even have a lower cost than conventional
flash plants. An important consideration is that developed case studies indicate that the average net
MWe/well achieved are reasonable (even after a subtraction of the parasitic well pump load).
Many of these developments are producing from quite shallow reservoirs (<1000 m), which reduces
the cost of drilling compared to conventional higher temperature systems. Some pumped developments
utilize very shallow wells of less than 300 m depth (Salt wells: 150m, Mammoth: 200m) that are
outflows from some upflow area, but that the outflow has proven most economical to develop.
While there will be geological and market cost differences between the US and Indonesia, the recent
development history for pumped systems in the USA and other countries is indication that such systems
may be feasible to develop in Indonesia, especially in the higher tariff regions.
3
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
The reservoir water is kept pressurised to avoid any flashing (to maintain original temperature and
avoid calcite scaling) and pumped to a binary power plant where the heat is transferred to a closed-cycle
working fluid (with a lower flash point temperature) that flashes and passes through a turbine, which
drives an electric generator. The cooled geothermal fluid is then reinjected and the working fluid is
4
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
cycled through an air- or water-cooled condenser. Lower temperature binary developments typically
utilize a production to injection well ratio of roughly 1:1.
The key resource performance indicators for pumped well systems are temperature, well productivity
index and inter-well permeability. The well productivity index has a direct bearing on how hard the
pump has to work for a required flow and as such dictates the parasitic power lost in driving the system.
The pump impeller must be set below the dynamic water level and this dictates the pump setting
depth for reliable operation.
Equipment performance indicators include the mean time between failures (MTBF) of the pump.
Fluid chemistry has a significant influence on the MTBF, as outlined in the experience of [12] at the
Soultz EGS site in France.
The two main types of downhole pumps in use are lineshaft vertical turbine pumps (LSP) and
electrical submersible pumps (ESP), distinguishable by the location of the motor. The former have been
utilized extensively in USA since the 1970s (refer Figure 3), the latter are less widely used in geothermal
applications but have been used in France, USA and Germany (refer Figure 4).
Both types have typical flow rate limitations related to well/pump diameter, motor power and
temperature. Until recently, the typical flow limits for LSPs were about 150 kg/s, however, larger
diameter LSPs capable of pumping >225 kg/s have recently been manufactured and installed at a few
projects in the USA and Central America. Pumped geothermal wells are not presently used in Indonesia,
but some of the companies that develop these projects in the USA and Europe are actively seeking or
developing projects in Indonesia. It is reasonable to assume that they can bring or attract the services
required for this type of production if they see the development opportunities pumped systems in the
country.
There are relative advantages and disadvantages between the two types of well pumps as outlined in
Table 2 (adapted from [1]), and these aspects can be the basis in selecting which pump type is preferred
for a particular system.
5
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
The ESP temperature limit arises because the pump motor is submerged and the geothermal brine
acts as a cooling fluid. Vendors have claimed pumps can operate higher than 160°C but this can translate
into derating of the motor output and/or reduced MTBF.
6
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
Figure 5. Muraoka [11] noted the relationship of geothermal systems with major structures in
Sumatra
3. Lateral outflows from geothermal systems developed on the flanks of central volcanoes. Some high
temperature convecting geothermal systems have considerable hot fluid outflows at shallow levels
(often with strong lithologic influence), which is sometimes manifested as hot springs occurring
kilometers away from a central volcanic edifice. A conceptual illustration of this type of system is
provided in Figure 6.
Figure 6. High temperature systems can have lower temperature outflows that may be developable
with pumped wells.
7
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
While the “quality” of these resources in terms of temperature and resource capacity is often less
than high temperature systems, the structural (2) and volcanic outflow (3) type systems can occur at
relatively shallow depth and are likely to be in much more accessible terrain than many of the high
temperature systems that have been developed to date in Indonesia. These important characteristics may
make them more commercially viable than deep and remote high temperature systems in some regions.
While Indonesia has considerable surface volcanic activity, regionally-elevated mantle heat flow and
co-located major geological structures (such as the Great Sumatra Fault System) also provide favorable
host settings for low and medium temperature geothermal systems. The results of Badan Geologi’s
efforts to explore Indonesia’s geothermal resources reveal that many of the identified geothermal
systems are lower temperature. Thus, further efforts to understand these systems in terms of their
reservoir characteristics plus how they can be produced and developed for power generation is
warranted.
Some examples of lower temperature systems and regions in Indonesia include:
Ulubelu in South Sumatra is presented as example of a volcanic outflow system is noted by [7].
Although much of this system is higher temperature where it is presently feeding two power
plants, it has downstream areas with lower temperature. Typically, such lower temperature
outflows have been ignored as not being useful for power generation.
Riogilang [13] note that two wells drilled into sediments and limestones hosting an outflow of
Kotamobagu in North Sulawesi found temperatures of about 160°C and so would not flow
naturally.
Muraoka [11] note the presence of many geothermal systems particularly associated with 13
identified pull-apart basins along the Sumatra Fault Zone. There are often volcanic features in
proximity to these systems because some of the magmatism is also influenced by these
structures. However, many of the fields along the Great Sumatran Fault System appear to be
dominantly structurally-controlled, rather than directly associated with central volcanoes.
Herdianita [5] describe the low temperature system at Cisolok-Cisukarame in Java which may
have once been a hotter convective system, but which now may be a ~160°C outflow system
associated with permeable limestones.
Humaedi [9] describe wells that have drilled a highly permeable outflow with temperature about
200°C from a hotter part of the Rantau Dedap system in Sumatra. This thick and permeable
outflow is an example of the strong outflows that can occur from volcanic systems. If targeted
at low elevations, these sorts of resources could present an attractive development target in some
fields.
On the basis of temperature, an estimate of the potential resource for utilizing pumped wells for
power generation can be inferred from [2] classification of geothermal resources in Indonesia, as shown
in Table 3.
8
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
5. Well Productivity
[8] and [3] showed that the economic feasibility of a pumped well development depends on resource
extent and temperature, permeability (in terms of well productivity), drilling depth and parasitic load,
and pump reliability. For a case in Indonesia with reasonably positive parameters they estimated that a
levelized cost of electricity could be around 10 to 12.5 US¢/kWh. This estimate was calculated making
some assumptions regarding well productivity. We have conducted further analysis to determine what
range of well productivity may be achieved in the Indonesian context.
It has long been recognized that self-flowing wells tend to have greater capacity at high temperatures
than at lower temperatures. This is because the higher energy content at high temperatures helps drive
the well flow (driven by fluid flashing and volume expansion) and also the reservoir fluid itself has more
energy per unit mass. The result is that resources with reservoir temperatures <210°C typically cannot
be developed using self-flowing wells as the obtainable flow rates are too low – exceptions are systems
with elevated gas contexts (e.g., systems in the Menderes Graben, Turkey) and systems with shallow or
artesian piezometric surfaces (e.g., Beowawe, USA).
Further to this, modelling of lower temperature self-flowing wells has highlighted that they can
achieve useful production where water levels are high and good reservoir permeability is encountered.
The implication is that lower temperature projects with self-flowing wells need to be developed in low
elevations close to the hydrostatic water level of the reservoir, and not be designed to have wellpads on
the ridgelines of steep terrain, for example.
Modelling of pumped and self-discharge well flows reported in the IFC report “Success of
Geothermal Wells: A Global Study” from 2013 (Figure 7) was based on earlier work but showed a step
change in well capacity as temperature increases and pumps reach their temperature limit (then identified
at 190°C) and the production from self-flow wells was lower than pumped wells in the middle region
around 200°C.
Figure 7. Well productivity variation with temperature. From [10], Success of Geothermal Wells:
A Global Study. Note the “break” between pumped and non-pumped wells could now be shifted to
a higher temperature due to recent technology advancements.
That work assumed a maximum pump set depth of 457 m and a maximum pump flowrate of 2,500
gpm (about 150 kg/s). This tended to limit the maximum capacity of any pumped wells to about 7.3
MW (net excluding pumped power – the IFC report is not clear about this distinction). Subsequently,
9
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
it is now common to set pumps deeper and also some manufacturers are working towards about 50%
greater flows. Also, ambient conditions in Indonesia are somewhat different to the US (affecting the
efficiency of the condensing cycle in the binary plant), so we have conducted some modelling to update
these estimates of capacity for application in Indonesia.
Figure 8. Productivity of pumped wells as a function of temperature and productivity index (PI).
This assumes a water level in the reservoir that is in hydrostatic equilibrium with the surface.
10
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
The MW presented are net after power plant parasitic power and pumping power are deducted from
the gross generation of a typical binary plant.
These results differ from the [10] results as some of the deeper limits in terms of pump set depth are
increased allowing for more generation at high temperatures, and at lower temperatures, the higher
ambient temperature in Indonesia curtails the net generation compared to the IFC work based on US
conditions (we assume). We note that two of the largest projects in the US and Honduras (see
Table 1) are exceeding the MW/well results with an average of 6.5 MW/well at McGinness Hills
(165°C) and 7.5 MW/well average at Platanares (177°C). This indicates that the modelling done here is
reasonably conservative and does not rely on particularly optimistic assumptions.
The results of this modelling are plotted over the [10] results (see Figure 9). The assumed average
successful well MW/well numbers used in recent tariff modelling by GT Management are also included
for reference. As a working guideline, 4 MW/well can be expected from 160°C systems and 6 MW/well
from 180°C systems. It can be seen that the MW/well that may be expected for pumped wells with
reasonable permeability reservoirs are comparable to the values obtained from self-flowing wells in low
and medium temperature systems.
6. Financial Implications
The possible viability of pumped well geothermal projects in Indonesia using the well productivity
estimated in this work has been tested with a financial model that calculates the tariff required by a
developer in order to achieve their required IRR.
Using typical financial assumptions for a State Owned Enterprise (SOE) developer which has a lower
required IRR and access to concessional finance, and assuming the following main input assumptions,
our preliminary modelling resulted in a calculated tariff requirement of about 12 US¢/kWh:
11
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
This is about 10% higher than a similar sized project high temperature project using typical costs and
well productivity assumptions for that type of development.
Testing the sensitivity of the different key parameters showed that the well capacity (MW/well),
project size (MW) and well depth (m) were the parameters that have most effect on required tariff
(Figure 10).
7. Conclusion
This paper has outlined the current status of pumped well technology, typical areas in Indonesia where
this technology is appropriate and estimates of expected well productivity.
Indonesia potentially has a large amount of low and medium temperature geothermal resources that
traditionally have been overlooked in favor of exploiting high temperature magmatic resources.
Low temperature systems have been developed internationally, and pumped systems have been the
dominant type of geothermal technology utilized in projects in the USA over the last 5 years. In the
USA such projects appear to be cost competitive, even while operating in a very competitive tariff
market.
Modelling of pumped wells in Indonesian conditions shows that with reasonable productivity indices
(i.e., moderate permeability of at least 20 tph/bar) it should be possible to achieve 4 to 9 MWe net per
well over a resource temperature range of 160 to 200°C. Resources with temperatures less than 160°C
may not currently be economic in Indonesia.
The reasonable indicated well power capacities (in terms of MW/well) and potential for drilling
shallower wells at lower cost than is required at many typical high temperature projects indicates that
pumped systems may be cost competitive with similar sized high temperature projects.
Preliminary financial modelling indicates that, with reasonably favorable conditions, it may be
possible for a SOE developer to develop a project in a tariff regime of about 12 US¢/kWh. However,
further economic analysis comparing the development with other traditional geothermal developments
is warranted.
12
7th ITB International Geothermal Workshop (IIGW2018) IOP Publishing
IOP Conf. Series: Earth and Environmental Science 254 (2019) 012021 doi:10.1088/1755-1315/254/1/012021
The findings of this study indicate that lower temperature resources may be attractive targets (from
a technical standpoint) for pumped well developments in Indonesia if they meet the following technical
criteria: (1) Resource temperatures over 160°C; (2) shallow/artesian piezometric levels, and / or the
ability to site wellpads at low elevations relative to system hydrology.
Acknowledgement
The work to model pumped well productivity for Indonesian conditions, and compiling information on
recent global pumped well projects, was funded under the New Zealand Government’s Ministry of
Foreign Affairs GEOINZ program. GEOINZ has the objective of supporting the accelerated
development of geothermal projects in Indonesia.
References
[1] DiPippo R 2016 Geothermal Power Plants – Principles, Applications, Case, Studies and
Environmental Impact 4th Edition Amsterdam: Butterworth-Heinenmann, imprint of Elsevier
2016.
[2] Fauzi A 2015 Revision of Geothermal Resource Classification in Indonesia Based on Type of
Potential Power Generation Proceedings World Geothermal Congress 2015.
[3] Febrianto R Bydder J Hochwimmer A Ussher G 2016 Pumped Well Geothermal Power
Generation in the Indonesian Context Proceedings The 5th Indonesia International
Geothermal Convention & Exhibition (IIGCE) 2017.
[4] Franco V and Villani M 2009 Optimal design of binary cycle power plants for water-dominated,
medium-temperature geothermal fields Geothermics Volume 38 (4) December 2009 Pages
379-391
[5] Herdianita N and Mandradewi W 2010 Evolution of Cisolok – Cisukarame Geothermal System,
West Java – Indonesia, Based on Its Surface Manifestation Proceedings World Geothermal
Congress 2010.
[6] Hernandez K Richard C and Nathwani J 2016 Estimating Project LCOE – An Analysis of
Geothermal PPA Data Proceedings 41st Workshop on Geothermal Reservoir Engineering
Stanford University.
[7] Hochstein M P and Sudarman S 2015 Indonesian Volcanic Geothermal Systems Proceedings
World Geothermal Congress 2015.
[8] Hochwimmer A Ussher G Urzua L Parker C 2013 An Assessment of the Economic Feasibility of
Electricity Generation from Pumped Wells Tapping Lateral Outflows of Liquid Dominated
Geothermal Systems Proceedings 35th New Zealand Geothermal Workshop 2013.
[9] Humaedi M T Alfiady Putra A P Martikno R Situmorang J 2016 A Comprehensive Well Testing
Implementation during Exploration Phase in Rantau Dedap, Indonesia Proceedings 41st
Workshop on Geothermal Reservoir Engineering 2016.
[10] International Finance Corporation – IFC 2013 Success of Geothermal Wells: A Global Study.
[11] Muraoka H Takahashi M Sundhoro H Dwipa S Soeda Y Momita M and Shimada K 2010
Geothermal Systems Constrained by the Sumatran Fault and Its Pull-Apart Basins in Sumatra,
Western Indonesia Proceedings World Geothermal Congress 2010.
[12] Ravier G Graff J and Villadangos G 2015 Operating a Lineshaft Production Pump in a Small
Pump Chamber Under Highly Aggressive Geothermal Fluid Conditions: Results from the
Soultz EGS Site Proceedings World Geothermal Congress 2015.
[13] Riogilang H Itoi R Tanaka T and Jalilinasrabady S 2012 Natural State Model of the Kotamobagu
Geothermal System, North Sulawesi, Indonesia GRC Transactions 36 2012.
13