The automotive industry and the design of engines are strongly ruled by performance and legislation demands. In the valve train, mechanical components such as camshafts and rollers are defined by specifications including more and more constraints concerning their ability to face wear thus prolonging lifetime.

The aim of the thesis is to develop tools facilitating the choice between different manufacturing processes for wear optimization purposes of cam and roller components for IC engine valve trains. Tools are both experimental and theoretical.

For the experimental part, statistical and relocated studies of wear have been performed. It is shown that measuring the very same surface before and after experiments is preferable to understand wear mechanisms of cams and rollers. A set of analysis tools for describing changes between unworn and relocated worn surfaces is developed. As results, it is found that the predominant mechanism of wear for cams and rollers is a flattening of asperities: surfaces are pressed and plastic deformations occur.

In parallel, simulations have been developed to explain theoretically the wear observed. Micro and macro simulations are developed to predict the ability of a given manufacturing process to resist wear. For the microscopic simulation, a rough contact model including elasto-plastic behavior of materials is used and shows good correlations with experiments. Concerning the macroscopic simulation, a model including form deviations due to manufacturing is developed and computes oil film thicknesses and deformations. The different parameters computed by both simulations are indicators of the wear performance of different surfaces. It is shown that such simulation can rank different manufacturing processes in terms of ability to face wear.