Virtual Fuel Testing Targets More-Efficient Injectors

Researchers at the University of Alabama-Huntsville create a computer model that targets higher power and cleaner combustion in the entire range of engines from those in motorcycles to rockets.

Chien-Pin Chen, chairman of the university’s UA Huntsville's chemical engineering department, headed the research effort to create a model that reduces the expense of evaluating fuel-efficiency calculations. “That's the reason we are so excited about this research,” he says.

“If somebody wants to do a numerical diagram of an internal-combustion engine, the first thing they need to study is the fuel,” Chen says. Usually a supercomputer would be required to analyze gasoline because it contains numerous substances with different evaporation rates and ignition points, he notes.

“So we designed a surrogate fuel with three components instead of hundreds,” he says. “It performs the same but it is not as complex to study. This allows the fuel to be mathematically stated and allows virtual testing of all fuels, including conventional ethanol/gasoline and E85.

UA-Huntsville's model permits evaluation of how fuel droplets from different injector designs will behave, revealing evaporation characteristics and combustion efficiency in the combustion chamber, Chen says. His team used the fuels database maintained by the National Institute of Standards and Technology to validate the research results.

“We are already changing injector designs,” Chen says. The research allows engineers to find the best injector design for optimal flame propagation, as well as insights into fuel turbulence in the combustion chamber.

The goal is to eliminate pinging caused when fuel explodes because of uncontrolled burning and expansion. This can cause premature wear and engine failure.

Chen says even propagation occurs in today's gasoline engines when fuel is layered to produce a rich mixture of fuel and air near the spark plug and swirls to a leaner mixture near the top of the piston. This allows the plug's spark to more easily ignite combustion in rich fuel and burn it more efficiently in the leaner mix.

Chen's team is seeking to develop a flame-propagation model that will allow computer testing before real-world evaluations start. This could result in efficiency gains, he says.