Measurements of NO Emissions From a Stratified Charge Engine: Comparison of Theory and Experiment

Abstract

Measurements of NO exhaust emissions have been made on a single-cylinder engine with and without stratification. The results were compared with predictions based on an existing computer model of Blumberg (1973). Combustion duration, a critical parameter of the model, was inferred from experimental pressure-time records for each condition studied. In premixed operation at low NO levels the post-flame Zeldovich kinetics, as used by Lavoie et al, (1970), could not account for the observed concentrations. The discrepancy was ascribed to flame-formed NO, and a correction to the chemical kinetics was made based on the measurements of Fenimore (1970). With this correction good qualitative agreement between theory and experiment was obtained for the premixed case. On the basis of these results the modified kinetics were also employed in the stratified charge calculations. For stratified operation, calculations were made with a number of assumed stratification functions and the results compared with experimental measurements. The best fit to the data was obtained for a strongly stratified fuel distribution with equivalence ratio variations from the mean on the order of the mean itself. Theory and experiment agreed to within a scale factor of ∼2, Further improvements in the absolute accuracy of the model would require consideration of heat transfer effects and a better estimate of flame-formed NO. The results show that the usefulness of stratification for NO control is highly dependent on the overall equivalence ratio. At equivalence ratios near stoichiometric and for equal EGR rales stratified charge operation gives lower NO concentrations than premixed operation. For mixtures leaner than φ = 0.8 however, stratification results in a loss of NO control compared to the premixed case.