Cam-roller followers are widely used in the valve train of internal combustion engines in recent years. They can significantly reduce the power loss and improve the lubrication state. However, the cam-roller pair still faces harsh working conditions as the load, the entraining velocity and the radius of curvature change drastically. This often causes severe wear problems on the cam surface and cause a complex lubrication state.

In the lubrication analysis of the cam-roller pair, the kinematic analysis of the cam-roller unit was carried out for solving the entraining velocity. It was found that the contact point not only rotated around the center of the camshaft as a fixed axis of rotation, but also oscillated from side to side as the roller moves up and down. This swinging produced a convected motion which effect the result of entraining velocity. The use of the synthetic motion method of points simplified the calculation process by transforming complex motions into a superposition of simple motions. The central idea is that relative velocity equals absolute velocity minus convected velocity.

In lubrication analysis of the cam-roller pair, a simplified method that ignored the contact point motion was used by some people in the calculation of entraining velocity, but the resulting errors had rarely been studied. In order to clarify the kinematic relationship between cam and roller, the eccentric wheel-roller was taken as a special case of cam-roller for kinematic analysis, and the analytical solution of the entraining velocity considering the contact point motion was obtained. In order to explore the effect of contact point motion on the results of the entraining velocity, the relative error ψ was defined. It can be seen that the relative error was 8.89% when rotation angle was 0°. The difference was obvious that could not be ignored, so the contact point motion must be taken into account in the calculation of the entraining velocity.

In the process of calculating the entraining velocity, there were four parameters that would affect the results of calculations: eccentricity, the radius of eccentric wheel, the radius of roller and the rotating speed. Therefore the effects of the four parameters on the calculation results had been discussed. It could be seen that the eccentricity had the most obvious impact on the relative error, and when the eccentricity increased from 5.0 mm to 7.5 mm, the relative error increased from 8.89% to 14.33%.

It was known that the entraining velocity effected the film thickness directly. The higher the entraining velocity, the more lubricant would be involved into the contact zone. Therefore, the impact of the contact point motion on the lubrication state had been studied. A non-Newtonian transient lubrication model of the eccentric wheel-roller pair was established to simulate the effect of the contact point motion on the lubrication state. Since the central film thickness was a strong function of the entraining velocity, the trend of central film thickness was similar with the trend of entraining velocity. The relative error was 6.30% when rotation angle was 0°. Accordingly, the effect of the contact point motion could not be ignored on the analysis of lubrication state.