Monitoring and modelling of false brinelling for railway bearings

An adaptive finite element model was developed to predict false brinelling in a cylindrical bearing, during the transportation of new trains and then compared with experimental measurements. The model was developed, based on the Archard wear equation, using ABAQUS and developing an ABAQUS UMESHMOTION code. A false brinelling monitoring system was designed and installed to record the vibration and rotational motion of the bearing, during the train's road and sea transportations and were used as inputs for the FE model. A set of laboratory experiments were conducted to determine wear and friction coefficients and the threshold energy for the bearing that were used in the simulation process. To compare the FE results with a real case, a profilometry measurement of false brinelling marks in a damaged bearing was performed. According to the FE wear profile results, rotational displacement of the bearing is the most likely cause for false brinelling during transportation. Due to the wear energy being below the threshold, it was predicted that no false brinelling occurred due to lateral and axial vibrations at the roller. The existence of the partial slip area, in the contact, causes the development of W shape wear marks that was seen in the profilometry measurement of the damaged sample.