A year and a half ago, Ausra, a solar thermal company backed by $115 million in venture capital, talked about its plans to generate over a gigawatt of power in the U.S. The electricity it would produce would be cheaper than coal or natural gas after mass manufacturing kicked in.

A 92-mile-by-92-mile square in the Arizona desert stocked with its equipment could provide all the electric power needed in the U.S., mused Peter Lievre, then CEO.

Last week, Ausra scaled back those ambitions (see Ausra Update: Layoffs and Change of Business Plans Confirmed). It will continue to make mirrors, pipes and the other equipment necessary for harnessing solar thermal energy. However, it is not going to own and manage large power plants in the desert. In fact, Robert Fishman, who became CEO in September 2007, says now that solar power plants were likely beyond its reach from the beginning.

Power plants "are way beyond" the capabilities of a startup, he said.

So what happened? Chalk it up to a change of circumstances, Fishman says. Detractors also like to add that a dash of hubris didn't hurt – several other startups, after all, are winning contracts to build multimegawatt plants.

Either way, the Ausra flip – like the Imperium Renewables meltdown from 2007 – will be one of those touchpoints entrepreneurs likely study for the next few years. If the switch to becoming an equipment supplier succeeds, it will be hailed in case studies. If not, it's tsk, tsk, tsk.

Although the recent credit crunch had a huge impact on Ausra's power ambitions, the shift actually began in the first part of 2008 when utilities began to examine the potential for owning their own solar plants, according to Fishman. Southern California Edison, for instance, unveiled a rooftop solar program (with PV panels) in March 2008, a move that solar power providers saw as a shot across the bow.

The new version of the investment tax credit passed in October built the momentum. Under earlier versions of the laws, utilities couldn't build their own solar thermal plants and qualify for the tax. Only independent power providers – like Ausra, Optisolar, BrightSource Energy and others – could.

"Utilities began to say, ‘We'd really like it better if you didn't compete with us,' " Fishman said.

It would also be a lopsided content. Utilities have sterling credit ratings. A solar thermal field backed by a startup would inherently be at a disadvantage to one built by a utility.

Before the utility interest, the startups didn't really have a choice anyway. Utilities were reluctant. Back in 2005 and 2007, "there was a belief that [the solar thermal startups] had to do it," Fishman said.

Some of the projects also didn't make sense once the planning was underway. Ausra, for instance, signed a deal to provide 300 megawatts solar thermal energy to Florida Power & Light in 2007 with a proportion of the deal going to a thermal plant in Florida (FPL also has operations in California). Then it started studying hurricane patterns.

"We walked away from that deal," Fishman said. "Our product is rated for about 95 mile per hour winds. We realized that Florida is not a good location for our technology." 

During this period, some oil company representatives near Ausra's facilities in California's central valley came calling. Petroleum companies require large amounts of steam for enhanced oil recovery. "Could Ausra's equipment work?" they asked. (Ausra started talking about steam in its press releases in June 2008. By July, the company began to list steam ahead of electricity in its company description.)

This new market is potentially huge, said Fishman. In some ways, it is even better because the steam generated by the solar heat doesn't have to be converted into electricity. The cost per kilowatt/hour for power for a steam boiler is about two-thirds the cost for producing solar thermal electricity because less equipment is needed, he said. (Less energy gets lost in the conversion process as well.)

Steam projects aren't nearly as big. A steam project might require five to 50 lines of Ausra's solar thermal equipment. A single line, which consists of mirrors and pipes and other devices, can generate 7 megawatts of heat and 2.5 megawatts of power.

A solar power plant might start at 150 lines and go up from there. On the other hand, the customer base is far larger. And erecting a five-line steam project can be done quickly. A solar thermal power plant takes years of planning and strategy meetings.

"We are a solar steam boiler company. There are a whole bunch of industrials that need steam – food processers, hospitals, power plants," he said. "We'd like to get revenue now."

Fishman also claimed that Ausra's technology is better than what the competitors – a list that includes Solel, Schott, BrightSource and Stirling Energy – can offer.

"We can get twice as much steam per acre as power tower [BrightSource's technology] and twice as much as trough [Solel, Schott]," he said. "We can generate steam cheaper than anyone."

The Reality Check

Although startups may not be ideal candidates to build solar thermal power plants, many are. BrightSource won a contract to build a 500-megawatt thermal power plant for PG&E that could be expanded to 900-megawatts (see Is a PG&E Solar Rooftop Investment on the Horizon). Stirling has a 500 to 850 megawatt deal with SCE and another 300 to 900 megawatt deal with San Diego Gas and Electric. Solel, one of the older companies in the field, has a 533 megawatt deal with PG&E. Both BrightSource and Solel are largely managed by people who built plants in California two decades ago.

Ausra bid on these deals, according to sources.

In terms of performance, all of them have said at one time or another that their technology is better than the others. Power towers are less subject to thermodynamic losses than other technologies, proponents have said.

FPL, meanwhile, is still interested in solar thermal. It is building a 75-megawatt hybrid solar thermal plant in Martin, Fla. that will have 180,000 mirrors spread over 500 acres. FPL has a lot of experience in the field: it owns 310 megawatts of solar thermal capacity in the Mojave.

Some also doubt that the company can easily retool its products for the varied steam market.

"How does one downsize a system that has been optimized to produce 10s of megawatts of power down to a system that maybe produces .5 megawatts," wrote one VC.

One thing Ausra will have going for it, at least, is factory footprint. It has a facility in Las Vegas capable of churning out 350 megawatts worth of equipment a year. The company also raised $60.6 million in October.

And Fishman can claim that history to some degree is behind him. Back in the early ‘90s, Solel, Luz and other then-startups laid plans for several more solar thermal fields. The state of California, however, let real estate tax exemptions lapse. Credit dried up and the effort died.