On Gas Dynamics of Exhaust Valves

Sammanfattning: With increasing effects of global warming, efforts are made to make transportationin general more fuel efficient. When it comes to internal combustion engines,the most common way to improve fuel efficiency is through ‘downsizing’. Downsizingmeans that a smaller engine (with lower losses and less weight) performsthe task of a larger engine. This is accomplished by fitting the smaller enginewith a turbocharger, to recover some of the energy in the hot exhaust gases.Such engine systems need careful optimization and when designing an enginesystem it is common to use simplified flow models of the complex geometriesinvolved. The exhaust valves and ports are usually modelled as straight pipeflows with a corresponding discharge or loss coefficient, typically determinedthrough steady-flow experiments with a fixed valve and at low pressure ratiosacross the valve. This means that the flow is assumed to be independent ofpressure ratio and quasi-steady.In the present work these two assumptions have been experimentally testedby comparing measurements of discharge coefficient under steady and dynamicconditions. The steady flow experiments were performed in a flow bench, witha maximum mass flow of 0.5 kg/s at pressures up to 500 kPa. The dynamicmeasurements were performed on a pressurized, 2 litre, fixed volume cylinderwith one or two moving valves. Since the volume of the cylinder is fixed, theexperiments were only concerned with the blowdown phase, i.e. the initial partof the exhaustion process. Initially in the experiments the valve was closed andthe cylinder was pressurized. Once the desired initial pressure (typically in therange 300-500 kPa) was reached, the valve was opened using an electromagneticlinear motor, with a lift profile corresponding to different equivalent enginespeeds (in the range 800-1350 rpm).The results of this investigation show that neither the quasi-steady assumptionnor the assumption of pressure-ratio independence holds. This meansthat if simulations of the exhaustion process is made, the discharge coefficientneeds to be determined using dynamic experiments with realistic pressure ratios.Also a measure of the quasi-steadiness has been defined, relating the changein upstream conditions to the valve motion, i.e. the change in flow restriction,and this measure has been used to explain why the process cannot be regardedas quasi-steady.