Characterization of Operating Parameters' Authority on the Flow-Field Mixedness of a DISI Engine

Abstract

A study designed to characterize the authority of engine operating parameters? contribution to the mixedness of the non-reacting flow-field of an optically accessible directinjection spark ignition (DISI) engine was conducted. Planar laser-induced fluorescence was used in conjunction with a high resolution imaging system to gather two-dimensional, planar images of the in-cylinder flow-field of the engine. The statistical pixel population, bounded by a rectangular region of interest that included the spark plug, of 100 images was used to characterize the bulk flow-field. Prior to the statistical analysis, an image correction procedure was applied to the flow-field images. Two statistical metrics, the normalized mean scalar spatial variation and normalized mean scalar dissipation, were used to characterize the flow-field?s mixedness. The relative effects that the fuel spray had on the mixedness of the in-cylinder flowfield was tested by using two fuel injectors: a pressure-swirl fuel injector operated at fuel pressures of 5.2 and 10.4 MPa, and an air-assisted fuel injector operated at a delivery gas pressure of 550 kPa. Each spray?s momentum flux was characterized by impinging the spray onto a force transducer. Additionally, the fuel injector-related effects of the mass of fuel injected and in-cylinder fuel vaporization were also investigated, the latter with a gaseous jet. The extent that the bulk flow-field contributes to in-cylinder mixing was assessed by separating it into two categories; intake flow momentum and intake flow direction. Shrouded intake valves were used to manipulate the momentum and direction of the in-cylinder flow, which were characterized by swirl and tumble coefficients. A symmetrically shrouded intake valve was used to differentiate the mixing effects associated with the intake flow direction and flow momentum. The mixing effect associated with intake flow momentum was also assessed by throttling the engine. Additionally, a valve deactivation mechanism was used to explore the relative contribution of bulk flow motion on the flow-field mixedness. The contribution of in-cylinder turbulence responsible for small-scale scalar advection was investigated by operating the engine at three operating speeds: 300, 600, 1200 RPM. The evolution of flow-field mixedness was assessed by phasing the end of fuel injection and obtaining flow-field images at three image acquisition times of 40, 60, and 80 CAD bTDC. It was found that the direction and magnitude of the angular momentum of the incylinder flow, especially tumble, had the greatest authority on flow-field mixedness and the rate at which the flow-field mixed. Fuel injector type had a minor influence on flow-field mixedness and the rate of mixing for early EOI timings, but yet had minimal influence at late EOI timings. The effects of in-cylinder fuel vaporization on flow-field mixedness were notable. At early EOI timings, the flow-fields? stratification slightly increased compared to that of a gaseous jet; however, as EOI timings retarded, flow-field mixedness comparatively increased. For quiescent in-cylinder flows, in-cylinder fuel vaporization increased the flowfield?s stratification and decreased the rate at which the flow-field mixed as compared to that of a gaseous jet. Engine speed had a significant effect on flow-field mixedness, as decreasing engine speeds produced flow-fields with comparatively increased levels of stratification; however, the flow-fields? spatial variation decreased with decreasing engine speed. Intake manifold pressure phasing revealed that decreasing intake manifold pressures produced flow-fields with increased levels of stratification. The evolution of flow-field mixedness, as assessed with the phasing of image acquisition time, indicated that flow-field stratification increased with advancement of the image acquisition time.