Vol 19, Issue 39

September 26, 2005

Car Engines Get Efficient by Being Smart

By Kevin Krewell

Microprocessors have had an impact on most areas of our lives—from our appliances, to our heating and cooling systems, to our work, to our entertainment devices, and so on. One area of our lives that is getting a lot of attention these days is the high cost of gasoline (or petrol for our European friends), which is driving up the cost of driving. But this is another area in which the microprocessor has helped in the past and can help even more in the future.

At a recent Infineon press and analyst briefing, we heard a number of industry executives discuss new developments in power-train design that could significantly boost gas efficiency. The use of microprocessors, along with greater control over the combustion process, has made significant efficiency improvements while still offering power on demand. Present auto engines are about 35–40% efficient, with another 5% efficiency possible without major changes. This present loose closed-loop system of air/fuel mix, valve and ignition timing, and oxygen-sensor feedback will have to become more complex in future engines.

The next generation of combustion-engine efficiency will come from a technology you probably have never heard of called homogeneous charged compression ignition (HCCI). As described by John Pinson, group manager of Diesel Engine Research at General Motors Research and Development Center, HCCI combustion in many ways resembles a diesel engine.

HCCI combustion could add another 20% improvement in fuel efficiency, without the high costs of hybrid technology. That is why all the major auto manufacturers have research into HCCI engines. Many issues must still be sorted out to take HCCI engines out of the labs and onto the street.

HCCI Puts Gas Under Pressure

Like its diesel cousin, the HCCI engine doesn't use a spark plug to ignite the fuel in the cylinder; rather, it uses pressure. HCCI is different in that the combustion process doesn't produce a flame—the fuel burns cleanly and clearly in the cylinder. Without the flame, the combustion is more efficient and produces much cleaner exhaust gases.

To achieve this nirvana of combustion, the engine needs more precise control over fuel quality, valve actuation (including, possibly, multiple open/close cycles per combustion cycle), and direct fuel injection. To close the loop, the engine controller needs to monitor the combustion process. That is one area of research, as the solutions range from precise pressure sensors for each cylinder (the optimal solution) to simple knock sensors (used in some engines today to detect detonation). The precise control algorithms for HCCI are still under intense development.

Unfortunately, the various auto manufacturers are pursuing their research programs independently. In our industry, we see consortiums formed when a potentially disruptive technology (such as EUV) needs extensive research. If the HCCI research could be centralized, we might be able not only to get the efficiency of HCCI engines sooner but also to get it from a much broader range of manufacturers. As it is, the auto manufacturers view engine development as a potential competitive advantage and a potential licensing opportunity. This is clearly the way Toyota views its hybrid power-train technology, and the company is benefiting from sharply increased sales of its Prius line of hybrid cars in the United States and from licensing its technology.

Although the U.S. Department of Energy has funded research into HCCI, a government-sponsored consortium like Sematech could more quickly spread the benefits to multiple manufacturers and achieve the maximum impact. Japanese manufacturers Toyota and Honda have a significant lead in hybrid technology, but there is a new opportunity to take the lead in HCCI development. And driving both technologies will be faster, more capable microprocessors.