Effects of Fuel Composition on In-Cylinder Air/Fuel Ratio During Fuelling Transients in an SI Engine, Measured Using Differential Infra-Red Absorption 961204

Departures from optimum stoichiometry during transients (acceleration and deceleration) and cold start can lead to significant degradation in driveability and emissions control. Such departures occur as a result of a complex interplay between fuel transport mechanisms and the fuelling strategy. The relative contributions of several of these mechanisms are affected by fuel composition. To help understand these effects an open-path differential infra-red absorption technique has been used to monitor the transient evolution of the fuel vapour phase directly within the combustion chamber.

The sensor projected a narrow infra-red beam which traversed the cylinder of an optical access engine along an open path under the head, and measured the path-integrated attenuation caused by absorption of the infra-red radiation by the fuel vapour. It operated in the near infrared (NIR) spectral region around 2.3 μm, an absorption band in hydrocarbon species containing methyl groups. The signal was ratioed against a collinear reference measurement made at 2.1 μm, to correct for attenuation due to liquid phase droplets and deposit accumulation on windows. The dual-wavelength differential absorption scheme, coupled with phase sensitive detection, provided a robust measurement which was readily adapted to the noisy engine environment.

The technique was used to measure the effect of fuel distillation characteristics on the response of the in-cylinder air/fuel ratio to a simple fuel shut-off transient. A full boiling range gasoline was compared with three single component fuels with different boiling points: isopentane, iso-octane, and xylene, representing the volatilities of the light end, mid-range and heavy ends of a typical gasoline. The delay between fuel shut-off and in-cylinder AFR response increased with increasing boiling point of the component. It was very short for iso-pentane and iso-octane, longer for toluene and longer still for xylene. The results were analysed using a simple physical model of mixture preparation which considers the relative contributions of inlet port liquid wall film and entrained fuel transport. This method enabled measurements to be made of the steady state wall film mass. Very little fuel was stored in the wall film reservoir when the lighter components were used, but the mass stored in the reservoir was much higher for xylene. Comparison with the results for gasoline suggests that the heavy ends made the dominant contribution to the wall film reservoir mass; the contribution from lighter components was small.