Return of the steam engine: cheap storage for solar
A group of Australian engineers have “re-invented” the steam engine and combined it with solar thermal energy to deliver a cheap solar storage solution. What’s more, it works on the distributed level and can operate behind the metre, and is far cheaper than PV combined with batteries.
Not long after Robert Mierisch finished up as director of thermal systems research at the solar thermal energy pioneer Ausra, he took himself to the Smithsonian in Washington, the world’s largest museum complex, to find out everything he could about the Skinner steam engine.
Mierisch was playing on a hunch. Or should that be a conviction. He had worked with solar thermal energy for years, and had been looking for a viable storage solution. He thought he could find the enabler in comparatively ancient technology.
Steam turbines, he reasoned, couldn’t do the job at the scale he had in mind, nor did they have the flexibility required to replace diesel power. But steam engines just might. This is what led him to the Smithsonian and the Skinner Unaflow steam engine, which were used in ships through the 1940s, the last commercial version of the steam engine that had begun at the start of the industrial revolution nearly 200 years earlier.
Now, a unique distributed generation technology with storage, the product of a 4-year collaboration between Mierisch and Steve Bisset, another Australian expat and Silicon Valley entrepreneur, is soon to see the light of day.
Terrajoule, the Redwood City, California-based company they co-founded, will soon bring a demonstration system up to full-power operation, generating 100 kW for 24 hours per day. This is a key milestone toward bringing the solar/steam/storage technology to market within the next 18 months.
Mierisch and Bisset say their technology is potentially revolutionary but deceptively simple. It combines inexpensive solar power with inexpensive storage and behaves like an electric motor plugged into the grid, or even like a diesel genset. In other words, it can operate 24 hours per day, but without the utility bill or the fuel cost. And they say it will be cheaper and far more efficient than alternatives such as solar PV combined with batteries.
RenewEconomy was invited in July to visit Terrajoule’s demonstration site near Oakdale, on an irrigated almond farm in the heart of the California Valley, about 150kms east of San Francisco.
There, Terrajoule has an array of parabolic trough receivers that collect solar energy to create steam to drive the engine. The breakthrough comes from the realization that the storage can be created by exploiting the difference between the high pressure and the low pressure cycles of the engine.
While the sun is shining, high pressure steam is created and used to power the high pressure stage of the steam engine. This high pressure stage produces power only while the sun shines, but its exhaust steam still contains over half its original energy, now reduced to an intermediate pressure. This remaining energy is captured by condensing the exhaust steam into an insulated tank of water, which heats and pressurizes the water.
When energy is needed above what the sunshine-driven high pressure engine stage is producing, to handle peaks in daytime demand, or nighttime demand, the pressurized water in the tank is flashed back to steam which then drives the low pressure stages of the steam engine. The combined output of the high and low pressure engine stages provides 24-hour power on demand, like a diesel generator, but with no fuel cost.
The energy lost in this steam-water-steam storage and retrieval process is negligible, and the net cost to store the energy is a small fraction of the equivalent batteries.
Bisset says they have made only minor adjustments to the 1930s steam engine in concept, although the format is changed for manufacture alongside modern internal combustion engines. “They were highly evolved machines, he says. “Over 300 years they figured out how to make them that good. What we’ve done is match new technology (solar thermal), and old technology (steam engine) and a thoroughly modern idea (storage) and come up with a solution. The core patent is in the architecture, the novel combination of these technologies.”
Bisset says it solves the storage problem because it is much more capital efficient than other ideas. “If we started from scratch, it would have taken us 100 years to refine it. We don’t have to build more factories to do it. There is lots of capacity to build piston engines in the world, in the US, Europe, China and India.” The company’s first steam engines are being developed in collaboration with Roush Industries of Detroit, and the first market will be developed with partner JKB Energy, the leader in solar power for agriculture in California.
The initial product units will deliver 300kW to 1.5 MW peak, with constant output of 125kW to 625 kW over 24 hours. Each unit will be delivered with a 600kWh to 5 MWh of storage capacity. The engine and storage units are shipped, installed and operated in two or more 40’ shipping containers.
Terrajoule’s initial target market is the more than 300,000 electric and diesel-powered irrigation pumps in the solar-rich western US. Bisset says the technology will work in other applications, for example manufacturing plants, off-grid locations including mines, and even entire towns.
Mierisch estimates less than a five year pay-back for customers using diesel engines, a three-year pay-back when that diesel is transported long distances, and a six year payback for grid-connected customers in Australia.
“If you consider just the cost per kilowatt-hour of electricity produced”, he says, “the Terrajoule systems will be similar in cost to wind turbines and solar PV panels, and those technologies now produce kilowatt-hours as cheaply as fossil fuels. However, a Terrajoule kilowatt-hour is much more valuable than kilowatt-hours from wind or PV, because you can generate them when you need them. To make wind or PV viable in the long run, you have to combine them with batteries or some other expensive storage technology, and Terrajoule will be a fraction of the cost of those power-with-storage systems”.
The irony is that a unaflow steam engine had been used to convert steam into energy at Australia’s first ever solar farm in the remote town of White Cliffs in NSW in 1984.
This actually provided the seed of an idea for Mierisch’s later investigation, although it took a lot of work from Mierisch and Bisset to identify exactly where the solution was to be found.
“I don’t know why I didn’t think of it 40 years ago,” Mierisch says. Had they started several years earlier, pre-GFC and pre-Solyndra, Mierisch says it might have been a lot easier to get funding. As it was, California, as it had been for Ausra, was a more fruitful place to get backing for “partially developed thinking” than Australia. “I went to see a VC firm in Australia – and they said, ‘when you have some sales traction I will see you again’, and I thought, ‘when I have sales traction, I won’t be talking to you’.”