Valve recession: From experiment to predictive model

Increasing demands on engine performance and cost reductions have meant that advances made in materials and production technology are often outpaced This frequently results in wear problems occurring with engine components. Few models exist for predicting wear, and consequently each wear problem has to be investigated, the cause isolated and remedial action taken. The objective of this work was to carry out experimental studies to investigate valve and seat insert wear mechanisms and use the test results to develop a recession prediction tool to assess the potential for valve recession and solve problems that occur more quickly. Experimental apparatus has been developed that is capable of providing a valid simulation of the wear of diesel automotive inlet valves and seats. Test methodologies developed have isolated the effects of impact and sliding. A semi-empirical wear model for predicting valve recession has been developed based on data gathered during the bench testing. A software program, RECESS, was developed to run the model. Model predictions are compared with engine dynamometer tests and bench tests. The model can be used to give a quantitative prediction of the valve recession to be expected with a particular material pair or a qualitative assessment of how parameters need to be altered in order to reduce recession. The valve recession model can be integrated into an industrial environment in order to help reduce costs and timescales involved in solving valve/seat wear problems.