Study of the wear and rolling contact fatigue of class B wheel against the R350HT and R260 rail steels using the twin-disc wear simulator

Abstract

Maintenance due to the replacement of damaged wheels and rails as a result of wear and rolling contact fatigue (RCF) has been found to be the major problem to rail operating companies. This problem tends to lead to poor availability of railway networks. In order to solve this problem, cost effective wear simulators are developed to predict the wear behaviour of the rails and wheels in order to improve the preventive maintenance in pursuit of operational efficiency. Therefore, more studies that simulate a combination of rolling and sliding wear, together with RCF, are required, specifically for the Southern African region, where good rail wear simulators are not readily available. The problem with wear and RCF simulators is high production costs and, therefore, part of this work was to solve this problem by developing a cost-effective wear test rig that would be able to simulate RCF, sliding and rolling wear as experienced by the wheel and rail during movement of train on rail tracks. For this work, it was decided that twin-disc concept would be used, since literature clearly demonstrated that the method was successful in simulating the three damage mechanisms mentioned. The developed twin-disc wear test rig was successful in simulating both rolling and sliding wear and parameters such as applied load and speed (slip) were easily varied so to simulate the actual contact conditions between the wheel and rail. Outputs such as coefficient of friction and wheel disc temperature were obtained. The results obtained from the developed test rig agreed with literature as they are repeatable and comparable. To validate the performance and accuracy of the rig, typical South African wheel and rail materials were used. The wear and RCF performance of AAR class B wheels against BS EN 13674 R350HT and R260 rail steels under different slip ratios and applied loads were investigated. The results showed that the severity of wear is heavily dependent on slip ratio i.e., increased with slip ratio. Severe plastic deformation was also observed at high values of slip ratio. The AAR class B wheels performed better against the softer R260 rail as compared to the harder R350HT rail. As expected, the R350HT performed better than the R260 due to it having higher hardness values and finer interlamellar spacing. Three wear regimes were identified from the plots of wear rate versus wear index (Tγ/A) namely; mild wear, severe wear and catastrophic wear. The effects of introducing water and oil on the contact surface were investigated. It was found that wear was much lower when water or oil was introduced at the wheel-rail interface compared to dry conditions, for all slip ratios. When water was used, the main cause of RCF was found to be fluid crack pressurisation. The RCF cracks were also observed under dry contact. Therefore, the rig was successful in simulating wear and RCF at the wheel/rail contact under different contact conditions as experienced by the wheel and rail during movement of train on rail tracks. For the wear rates model, a data correlation coefficient to link twin disc and field wear rates was determined and used to predict the reprofiling times of wheels for two different distances of 100 000 and 200 000 km. Therefore, the purpose of this study was achieved in providing a tool for the predictive maintenance for the local rail industry that uses AAR class B wheels against R350HT or R260 rail combinations.