Search Results

This paper examines three strategies of detecting knock that are less dependent of engine noise. The first strategy uses the exhaust temperature, the second uses a dithering method (systematically advancing and retarding the timing), while the third uses the standard deviation of knock intensity as the indicator of knock intensity. The first strategy proves to be difficult to detect knock since the exhaust temperature is strongly dependent on the combustion efficiency instead of knock intensity. The second strategy uses a conventional accelerometer but discriminates against mechanical noise by subtracting the knock intensity during the retarded part from that of the advanced part of a dither cycle. This approach is found to require averaging the signals over large number of engine cycles and using large dither amplitude. The third strategy uses the Difference of Knock Intensity strategy where two cycle standard deviation is used.

A 0.5 liter single-cylinder, indirect-injection diesel engine has been fumigated with producer gas, a mixture of principally H2, CO, and N2 with a heating value of about 160 Btu/ft3. Producer gas is produced by air-blown gasification of coal or biomass. Measurements of power, efficiency, cylinder pressure, and emissions were made. At each operating condition, engine load was held constant, and the gas-to-diesel fuel ratio was increased until abnormal combustion (severe efficiency loss, missfire, knock, or preignition) was encountered. This determined the maximum fraction of the input energy supplied by the gas, Emax, which was found to be dependent upon injection timing and load. At light loads, Emax was limited by severe efficiency loss and missfire, while at heavy loads it was limited by knock or preignition.