Residual gas homogeneity measurements

Abstract Planar laser-induced fluorescence (PLIF) of a dopant added to the fuel, which was well mixed with the intake air upstream of the engine, was used to investigate inhomogeneities that arise from the incomplete mixing of the intake charge with the residual gas left in the cylinder from the previous cycle. The residual gas fraction was independently measured using a fast-acting solenoid valve and a cylinder dumping technique. The experiments were performed on a four-stroke homogeneous-charge spark-ignition engine operating at light load. The PLIF data were shot-noise-limited, permitting filtering to reduce some noise artefacts; and to further reduce system-induced biases associated with pulse-to-pulse laser variations the data were ray-mean-normalized, i.e. normalized by the mean of the data along the laser propagation direction. The filtered results had excellent noise characteristics, with a signal-to-noise ratio in excess of 50:1, and allowed the low levels of mixture unmixedness to be identified. The fresh charge was found to exhibit inhomogeneities up to 30° before top dead centre, the latest time investigated. The second moment of the ray-mean-normalized intensity distribution was found to be relatively insensitive to the total residual fraction up to a residual fraction of approximately 30 per cent, having a value of ∼0.04. At higher levels of residual there was a monotonic increase in the intensity distribution's second moment.