No description available.

The paper presents a computer-based guide of the effect of layout and loading (tension and torque) on the timing belt life and uses it to show the sensitivity of life to changed conditions in an automotive camshaft drive. The predictions are in line with experience. The guide requires belt property information, such as the tooth and tension member stiffness, the friction coefficient between the belt lands and pulleys and the pitch difference from the pulley, in order to calculate the tooth deflections caused by the belt loadings on the various pulleys in the layout. It also requires information on how the belt life depends on the tooth deflections. Experimental data are presented on the life±deflection relations of a commercial automotive timing belt tested between 100 and 140 8C, although the bulk of the data has been obtained at 120 8C. Four different life laws have been found, depending on whether the failure-initiating deflection occurred on a driver or a driven pulley, and whether at entry to or exit from the pulley. Theoretical analysis of the tooth loading in the partial meshing state shows that, in three cases out of the four, the different life±deflection laws transform to a single relation between the life and the tooth root strain. The exception is failure caused by driven entry conditions; work is continuing to understand better the causes of failure in this circumstance

The work described here has been carried out to obtain a better understanding of the tooth root cracking failure mode of timing belts. Previous work has demonstrated the close dependence of this on the tooth deflections of fully meshed teeth, generated by torque transmission, but has not considered the additional distortions generated in the partially meshed conditions at entry to and exit from a pulley groove. Approximate compatibility and constitutive equations are combined with a rigorous consideration of tooth equilibrium in partial meshing to show how bending moments are generated at both exit from a driven pulley and entry to a driving pulley. Experimentally determined belt lives correlate very well with a combined measure of fully meshed tooth strain and strain due to bending at entry or exit. The analysis also shows that this strain measure reduces with increasing belt tooth stiffness, confirming the importance of a high tooth stiffness for a long belt life. Tooth force variations through the partial meshing cycle have also been predicted and compared with measurements obtained from a special strain gauge instrumented pulley. A greater pulley rotation than is predicted is required for a belt tooth to seat in a pulley groove. There is room for improvement in the modelling