Search Results

Reliability research in hybrid ceramic bearings involves life testing of hybrid bearings and ceramic specimens. New materials for bearings like advanced ceramics have emerged for evaluation in recent years. In fatigue testing to determine the stress-life relationship, the number of sample size in life testing can be limited by consideration of cost and testing time. In the testing of ceramics, some researchers have relied on the use of a stepwise multiple loading approach to increase the failure data points. In this paper, a maximum likelihood method is applied to test data with multiple loads to estimate the stress-life exponent. This method treats the data at different loads or steps at once. Test data from three fatigue experiments using silicon nitride materials have been analyzed to obtain the stress-life exponents. Also, Weibull plots of the ‘equivalent lives' have been presented for all test specimens tested at different loads and load steps.

In an effort to reduce the design cycle time and to meet increasingly demanding applications, an improved procedure for bearing retainer design has been introduced. This paper discusses a methodology which allows the designer to predict the life and failure modes of a retainer under application conditions. Specific attention is given to the case of fatigue of the retainer due to the dynamic interactions between the retainer, rolling elements and races. The methodology which has been developed for the life prediction of retainers is based on the dynamic loads and retainer structural integrity. Central to this technique is the ability to predict the loads imposed on the retainer as a function of design and application conditions. The bearing analysis code ADORE has been used for this purpose. The technique will be discussed by means of an example.

This paper presents a method of measuring the actual load distribution within a rolling element bearing. This method has been used successfully to measure the load distribution in both cylindrical and tapered roller bearings, with accuracies of 15% or less when comparing the applied load with the computed resultant load. An example using a tapered roller bearing is given.