The vibration of the plate only moves the shaft two-tenths of a millimeter per cycle, so not much helium is being moved with each cycle. But the engine achieves a rapid 500 cycles per second. "If you go to high-frequency, you can get more power out of it," says Greg Swift, a thermoacoustics expert at Los Alamos National Laboratories who saw an early version of Etalim's engine. "Steiner has really done a good job of taking a different [design] direction and not making any mistakes."

The company has ambitious goals. A first prototype, completed last year, demonstrated that the concept works, but relatively low heat was used, so its efficiency was only 10 percent. A second prototype that aims for 20 to 30 percent efficiency at 500 °C is expected this spring.

A commercial product with 40 percent efficiency running at 700 °C is targeted for 2012. It will initially be sold as a cheaper and longer-lasting competitor to fuel cells used for residential cogeneration. The company believes it can manufacture the engine for less than $1 per watt, and has a long-term target of 15 cents per watt, which would make it less expensive than a comparable internal combustion engine.

"Everything to get us to 40 percent efficiency is right in line with what we've seen from our prototype so far," says Klopfer, adding that 50 percent is the longer-term target. "To get from 40 percent to 50 percent, we need to raise the temperature to 1,000 °C, and that requires some use of ceramics."

Mike Hayden, a professor of physics at Simon Fraser University, says Etalim's design is promising, but a lot of engineering works lies ahead to prove that the device can handle high temperatures and achieve the kind of efficiencies that would make it stand out. "But there's no doubt these guys have something interesting," he says.