Wear simulation for boundary lubricated, radially loaded, spherical roller bearings

The wear of railway axle bearings is an important phenomenon. Rail companies must keep their rolling stock in a serviceable condition as excessively worn bearings can contribute to safety issues such as vibrations, excessive clearances and possibly even derailments. At present, maintenance scheduling is typically performed at predetermined time intervals. Predicting the wear and degradation of the bearings, would enable operators and bearing manufacturers to 1) optimise the design and maintenance of the bearings, and 2) quantitatively determine the best time to replace the bearings. This paper presents a model developed for predicting the wear in radially loaded spherical roller railway axle bearings. It uses a slice method to calculate the roller load, traction and creep distribution along the roller for the contact mechanics, and predicts wear using a frictional power model. The model is first validated with the wear in a thrust bearing in the literature and then used to predict wear in the radial bearing. The wear contribution from each roller in the loading zone due to the radial loading is calculated and combined for an overall wear profile.