Geothermal Binary Cycle (ORC) Power Simulator

Design a binary geothermal plant that drives an Organic Rankine Cycle with a secondary working fluid (R245fa, Isobutane, n-Pentane or Ammonia/Kalina) using 100-200 °C brine that is too cool for a flash plant. Adjust brine conditions, fluid and condenser temperature and watch heat duty, net power, annual generation and LCOE update in real time.

Parameters

Brine inlet temperature T_in

°C

Temperature of brine arriving at the surface from the production well

Brine mass flow ṁ

kg/s

Total brine mass flow from all production wells

Brine outlet temperature T_out

°C

Temperature of brine returned to the injection well (lower bound set by silica scaling)

Working fluid

Sets the critical temperature and GWP

Condenser temperature T_cond

°C

Heat-sink temperature (air-cooled 30-45 °C, water-cooled 20-30 °C)

Target cycle efficiency

Cycle efficiency you do not want to fall below

Well depth D

Average production well depth (used in the CAPEX estimate)

Results

—

Brine heat duty Q (kW)

—

Carnot efficiency (%)

—

Cycle efficiency (%)

—

Net electrical output (MW)

—

Annual generation (GWh)

—

LCOE (USD/kWh)

—

Plant schematic — well → heat exchanger → ORC loop

Hot brine (red) rises from the production well on the left, vaporises the working fluid (purple) in the heat exchanger, drives the turbine and is condensed by the air cooler before the feed pump closes the loop. Cooled brine (grey/orange) returns through the injection well on the right.

Cycle efficiency vs brine inlet temperature

Per-fluid cycle efficiency comparison (same conditions)

Theory & Key Formulas

$$Q_{brine} = \dot m\,c_{p,brine}\,(T_{in}-T_{out}), \qquad \eta_{Carnot}=\frac{T_h-T_c}{T_h}$$

Brine heat duty Q (kW) and Carnot efficiency (T in K). c_{p,brine}=4.18 kJ/kg·K is approximated to liquid water.

$$\eta_{cycle} = \eta_{Carnot} \cdot f_{fluid} = \frac{T_h - T_c}{T_h} \cdot f_{fluid},\quad P_{net} = Q_{brine} \cdot \eta_{cycle} \cdot (1 - f_{aux})$$

f_fluid ≈ 0.55 (typical binary), f_aux ≈ 0.15 (pumps and the air-cooled fan).

$$LCOE \approx \frac{CAPEX}{n_{yr}\,E_{yr}} + OPEX_{kWh}$$

25-year life, OPEX 0.02 USD/kWh and CAPEX = depth × 5000 + P_net × 3000 USD in this simplified model.

Geothermal Binary (ORC) — designing the two-loop secondary fluid cycle

🙋

Geothermal plants run a turbine on the hot water coming out of the ground, right? I had never heard of a "binary cycle" before. How is it different from a normal geothermal plant?

🎓

Great question to start with. Conventional geothermal stations are "flash" plants that pull 200-300 °C steam from the reservoir and feed it straight into a steam turbine. But more than 70% of the world geothermal resource sits in the 100-200 °C band, where a flash plant produces almost no steam. That is where binary comes in: a heat exchanger takes heat out of the brine and boils a low-boiling-point secondary fluid like R245fa or isobutane, and the secondary fluid drives the turbine. Two loops — brine and working fluid — hence "binary". This tool is built around exactly that design space.

🙋

OK, so low-temperature resources become useful, that is convenient. But the default already shows "Cycle efficiency 15.6%". Coal and gas turbines easily top 40% — is binary that bad?

🎓

It is just thermodynamics. Carnot efficiency η = (T_h − T_c)/T_h is set by the temperature ratio, so brine at 150 °C and condensation at 30 °C caps you at 28.4%, and a real cycle reaches about 55% of that, around 15.6%. A coal boiler can run at 1500 °C, which is why it does better. But the fuel is free: the Earth keeps the heat in the ground for you, capacity factor sits above 90%, and you get a year-round baseload that solar and wind cannot deliver. Low efficiency is the price of having no fuel bill.

🙋

The dropdown has R245fa, Isobutane, n-Pentane and Ammonia. How do I pick? Switching it actually changes the per-fluid comparison chart, which is fun.

🎓

The first rule is to pick a fluid whose critical temperature Tc is close to the brine inlet. R245fa with Tc=154 °C is the classic for medium-temperature resources, but its GWP of 858 is putting it on phase-out lists in Europe. The leading replacement is Isobutane (Tc=135 °C, GWP=3) — yes, it is flammable, but it dodges the HFC regulations and is now standard in new low- to medium-temperature plants. n-Pentane (Tc=196 °C) suits hotter resources and is the workhorse of Ormat, the largest binary supplier. Ammonia is not used alone — it goes into an ammonia-water mixture in the Kalina cycle, where temperature glide lets the working fluid follow the brine cooling profile and pull out more heat. Husavik in Iceland is the famous example.

🙋

You can also set the well depth. Deeper means hotter, right? Why does LCOE go up when I drill deeper?

🎓

Good catch. The geothermal gradient averages 30 °C/km, so deeper does mean hotter — but drilling cost grows non-linearly with depth. This tool uses a flat 5000 USD/m, but beyond 3 km drilling rate drops, high-strength drill pipe is needed, and real cost climbs to 7000-15000 USD/m. New Zealand and Japan reach plenty of heat at 2-3 km, which is why geothermal is cheap there. Enhanced Geothermal System (EGS) projects like Soultz-sous-Forêts and Habanero aim at 5 km dry hot rock, but commercialisation is still pending because the wells are so expensive and induced seismicity is hard to manage.

🙋

If I push the brine outlet temperature down to 50 °C, the heat duty should go up. Why don't you let it cool further?

🎓

Physically you are right — a wider ΔT extracts more heat. But geothermal brine carries dissolved amorphous silica (SiO2) and, as the brine cools, it becomes super-saturated and precipitates inside pipes and the injection well. At Hatchobaru and Matsukawa in Japan the floor is around 80-90 °C; go below and you spend the gains on scale-removal maintenance. One reason the Kalina cycle is interesting is exactly because its temperature glide lets brine be cooled to about 70 °C without precipitation. So design is not pure thermodynamics — chemistry and maintenance set the real floor.

Frequently Asked Questions

A flash plant sends high-temperature (200-300 °C) steam straight to a steam turbine and is the workhorse of conventional geothermal generation. A binary or Organic Rankine Cycle (ORC) plant instead passes medium-low temperature brine (100-200 °C) through a heat exchanger and boils a secondary working fluid such as R245fa or isobutane to drive the turbine. The brine never touches the turbine, so scaling and corrosive brines can be tolerated, and roughly 70% of the world geothermal resource that is too cool for flash becomes usable. That is the main reason binary capacity now grows faster than flash.

Pick a fluid whose critical temperature is close to the brine inlet, with high thermodynamic performance in that range, and acceptable GWP / flammability / toxicity. R245fa (Tc=154 °C, GWP=858) is the historic medium-temperature benchmark but is being phased out in Europe. Isobutane (Tc=135 °C, GWP=3) is rapidly replacing it for new low- to medium-temperature plants. n-Pentane (Tc=196 °C, GWP=4) is favoured by Ormat for higher-temperature resources. Ammonia in an ammonia-water mixture (the Kalina cycle) uses temperature glide to follow the brine cooling line and squeezes out more heat, but the equipment is expensive and field experience is limited.

Real geothermal binary plants typically reach 50-70% of the Carnot upper bound. This tool uses a fluid-quality factor f_fluid = 0.55, so η_cycle = η_Carnot × 0.55. With brine at 150 °C and condenser at 30 °C, Carnot is 28.4% and the cycle efficiency lands near 15.6%. About 15% of the gross output goes to brine and circulation pumps and the air-cooled fan (parasitic load), leaving the net output. The 0.55 factor depends on the fluid, turbine efficiency, heat-exchanger pinch and superheat; state-of-the-art units exceed 0.7.

IRENA and DOE put the global average LCOE for medium-low temperature geothermal binary at roughly 0.07-0.12 USD/kWh. Well drilling (30-50% of CAPEX) and the surface plant dominate the cost. This tool uses a simplified capex = depth × 5000 + P_net × 3000 USD and adds 25-year amortisation plus 0.02 USD/kWh OPEX. With the defaults (150 °C, 200 kg/s, 2500 m well) you get about 0.043 USD/kWh — a deliberately optimistic figure that ignores resource risk, permitting and grid connection. Use it to feel the order of magnitude, not as a bankable number.

Real-World Applications

Japanese geothermal binary fleet: Kyushu Electric's Hatchobaru unit (Oita, 2 MW, R114 → Isopentane retrofit), Tohoku Electric's Matsukawa auxiliary system (Iwate, 2 MW) and the Tsuchiyu Onsen plant (Fukushima, 400 kW) are examples of small/medium binary units built across Japan's volcanic islands. Sites near hot-spring resorts depend on community agreements, with 100% reinjection often a condition of operation, so designers have to keep the brine outlet above the silica-saturation floor from day one.

Ormat's global fleet: Israel-based Ormat Technologies is the world's largest binary supplier, with more than 3 GW installed across the US (Heber, Mammoth, Steamboat), Nicaragua, Guatemala, Kenya (Olkaria) and Indonesia. They tend to use n-pentane and ship modular 25-50 MW air-cooled units that are paralleled to build out plant capacity. Mitsubishi Power and Toshiba supply equipment in the same segment.

Co-produced oil-and-gas brine: Wells that used to discard 80-120 °C produced water in Wyoming and Alberta are now sending it through ORC units to recover power. Reusing the existing well saves the largest CAPEX item and decarbonises late-life oilfields. Japan's JOGMEC is running similar pilots in Akita and Niigata.

Enhanced Geothermal System (EGS): Conventional geothermal needs a natural reservoir, which limits sites. EGS hydraulically fractures hot dry rock to create an engineered reservoir, in principle making geothermal possible anywhere. Soultz-sous-Forêts (France), Habanero (Australia) and FORGE (US) are the flagship demonstrators, with induced seismicity being the main hurdle. If commercialised, EGS could grow the global geothermal resource by two orders of magnitude.

Common Misconceptions and Pitfalls

The biggest trap is choosing the brine outlet temperature purely on heat recovery. Dropping T_out in this tool makes Q_brine and net output look better on paper, but real brine carries dissolved amorphous silica (SiO2) that is super-saturated as the brine cools. Pushing T_out below the silica saturation temperature deposits a hard scale on the heat exchanger, piping and injection well, and within months the plant is offline for descaling. In Japan's high-silica brines (200-600 ppm) the practical floor is T_out ≥ 80-90 °C. Always run a fluid chemistry analysis on the reservoir and back-calculate the safe floor from the SiO2, CO2 and H2S levels.

Next, believing that a clever cycle can beat Carnot. Kalina and supercritical-ORC marketing sometimes quotes "30% efficiency" headlines. Those numbers actually mean "more heat is extracted from the brine than R245fa simple ORC would do under the same conditions", not that the cycle efficiency exceeds the Carnot limit. The honest figure of merit for Kalina is η_cycle × Q_brine (actual MW), not η_cycle alone. That is why switching working fluids in this tool barely moves η_cycle — the fluid changes how much heat you can grab, not the thermodynamic ceiling.

Finally, do not assume one well drives the plant forever. A geothermal reservoir is finite. Long-term operation reduces pressure and temperature, and output typically decays at 1-3% per year (the Geysers field in California lost half its output in 30 years). Mitigation needs (1) reinjection to maintain reservoir pressure, (2) drilling additional make-up wells in stages, and (3) sometimes EGS to enhance the reservoir. A realistic LCOE study assumes the average output over 25 years is 70-80% of the initial value — the simplified LCOE in this tool ignores that and therefore reads optimistically.

Geothermal Binary Cycle (ORC) Power Simulator

Geothermal Binary (ORC) — designing the two-loop secondary fluid cycle

Frequently Asked Questions

Real-World Applications

Common Misconceptions and Pitfalls

How to Use

Worked Example

Practical Notes