A variable compression ratio engine would be an ideal match for flexible-fuel vehicles (FFVs). Sensors would indicate the fuel ethanol concentration to the onboard diagnostics system, much like how FFVs currently adjust fuel and air ratios to automatically change the compression ratio.

Theoretically, variable compression ratio is a good idea, but for mechanical engineers, it poses a practical nightmare. "Variable compression ratio is sort of the Holy Grail of engine technology," says Paul Blumberg, a former Ford Motor Co. engineer who is now vice president of Ethanol Boosting Systems LLC (EBS). "Everybody's been looking at variable compression ratio for a hundred years. It's extremely mechanically complicated to do successfully, and no one has come out with a production-worthy system." While engineers scratch their heads, Blumberg says EBS has the virtual equivalent of variable compression ratio.

"Rather than variable compression, we're using variable octane through direct injection of a knock-suppressant fuel—E85," he says. Most engines today are port fuel-injected, and as torque increases, the engine may experience knock or premature combustion of the fuel, especially if the compression ratio of an engine is considered high and the fuel's octane isn't.

Evaporating ethanol requires more heat than evaporating gasoline, so under high loads, ethanol is injected directly into the cylinders to cool down the air/fuel mixture so it doesn't prematurely ignite (i.e., knock). "The amount of E85 that you need to control knock is relatively small, but the benefit is that you can design the engine to be substantially smaller," Blumberg says. "Then you can boost it with turbocharging by a factor of two or 2.5." This means the air entering the cylinders would be compressed prior to injection by up to 2.5 times that of ambient air pressure. "You can run compression ratios of 12:1 or 13:1," he says, compared with 10:1, the typical practical limit.

Engine and fuel delivery system designs and computer controls necessary to commercially assimilate this technology into production lines aren't a big departure from current systems, Blumberg says. A "lookup map" tells the system how much E85 is needed under a given rpm or load, similar to how spark timing and fuel injection are programmed today into electronic controls. This would be tailored with feedback from nitrogen oxide sensors, feed-forward controls from predetermined lookup tables, feedback information from real-time sensors and more. "The technology is known," he says. "What's unique here is octane on demand." Of course, two onboard fuel tanks—one for unleaded gasoline and a smaller unit for E85, which would be refilled every two months—would be required, much like urea containers on diesels with advanced nitrogen oxide controls.

Tests run by original engine manufacturer Ford, EBS and the Massachusetts Institute of Technology—where the idea originated—indicate that on a volume basis of total fuel consumed, only 5 percent of E85 used in this direct-injection system leads to an average fuel economy boost of 25 percent to 30 percent. Ford will receive a U.S. DOE grant to help commercialize this technology. While Blumberg stresses that Ford hasn't given a commitment date for commercial deployment, EBS would like to see its E85 direct injection system in production by 2012. Blumberg says the first likely engines/vehicle platforms would be the sport utility vehicle market, where a 25 percent boost in fuel economy would mean greater savings in overall fuel consumption. "The industry has a fairly long product development cycle," he explains. "It's about three years, and then you have to certify the engine with the U.S. EPA. That's why we say we'd like to go commercial by 2012—not because it will take that long to develop the concept, but it's just the normal product development cycle."