Evaluation of a Mixing-Controlled Model for Engine Combustion

Abstract

Comparisons were made of computed and measured flame fronts, flame tip propagation rates, and pressures for eight configurations of a divided-chamber, stratified-charge engine differing in speed, equivalence ratio, and throat area. In the model, conversion from reactants to products was assumed to be mixing limited so that the controlling processes were turbulence and wall effects. Turbulence was represented by a k — e model and wall effects by the “law of the wall” and Reynolds analogy. The results were also compared with those from earlier models that attempted to account for finite rate kinetics too, through one overall irreversible reaction.

The flowfield is controlled mostly by the flame speed, except near the walls and in the throat area. All models, including the mixing-controlled one, reproduced the flame speed for the eight cases, but tended to overpredict the pressure, particularly for rich cases. However the mixing-controlled model was shown to be inherently inadequate to reproduce combustion near cold walls where kinetic rates eventually become controlling. In engine combustion, mixing-controlled models can also be expected to be inadequate for self ignition (knock and Diesel), pollutant formation, and rich and lean misfire limits. Thus, ultimately they would have to be supplemented with chemical kinetics information.