THE EFFECTS OF AIR SWIRL AND FUEL INJECTION SYSTEM PARAMETERS ON DIESEL COMBUSTION (EMISSIONS)

Pollutant formation processes of a single cylinder direct injection diesel engine were investigated using analysis of exhaust emissions, ignition delay and heat release. The effects of air motion and injection system parameters were studied by varying the air swirl with a shrouded valve and the injection pressure (5000-15000 psi) with an American Bosch EFIS. Five different nozzle geometries were used with the EFIS.; Increased swirl caused the amount of premixed burning to increase but also caused an increase in the time required to complete combustion. Lower swirl had the opposite effect. Increasing the swirl increased particulate emissions. It is hypothesized that the small nozzle orifices (0.0070 in. - 0.0098 in.) and high swirl levels (five times engine speed) decreased the rate of spray penetration resulting in stratification of the fuel-air mixture and slow burning under rich, soot-producing conditions. Maximum NO(,x) production was observed to occur at the intermediate swirl condition.; No significant effect of swirl or injection pressure on ignition delay was observed. A small increase in ignition delay was observed for large hole nozzles (.0096 in. - 0.0098 in.). This effect was attributed to local cooling caused by high injection rates and not to any chemical effect of the local fuel-air distribution.; The NO(,x)-particulate trade-off was favorably influenced by increases in injection pressure. This effect is most pronounced for six-hole nozzles. Four-hole nozzles have a complex interaction with swirl and injection pressure so that the NO(,x)-particulate trade-off for variable swirl is not easily interpreted.; In-cylinder measurements of solid carbon were attempted using total cylinder contents sampling. The solid carbon was determined by using a carbon balance expressed as the difference between the carbon injected as fuel and the gaseous carbon present in the sampled products. The measurement was not successful because ten to forty percent of the gaseous carbon present in the sample came from lubricating oil and obscured the solid carbon measurement. No way was found to eliminate the lubricating oil.