Experimental derivation of a predictive model for shear stress-induced surface cracks in railway wheels and rails

• Applicant: Professor Dr. Raphael Pfaff
• Subject Area: Engineering Design, Machine Elements, Product Development
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 571471889

In the wheel–rail contact, shear stresses occur due to traction, braking, and curve negotiation, which can lead to surface cracks. These surface cracks are the cause of spalling on the wheel surface and rail fractures. The risk to operational safety posed by surface cracks is currently under control through rail grinding and adapted rail profiles, known as “anti-head-check profiles.” However, continuous rail grinding is associated with high costs, while anti-head-check profiles negatively affect vehicle dynamics. To enable a more targeted use of mitigation measures, predictive models for the development of surface cracks are required. However, the results of currently available predictive models vary significantly. The parameterisation of existing models has mainly been conducted on scaled-down test rigs under simplified conditions. As part of the DFG funding programme “New Instrumentation for Research,” a novel large-scale test rig has been built, in which a counter-roller moves along the inner surface of a hollow wheel. This configuration creates a convex–concave contact situation that replicates the wheel–rail interface at full scale. The newly constructed large-scale test rig is intended to support the further development of existing predictive models for surface cracks. Initially, the existing models are used to define a parameter space within which the occurrence of surface cracks on the test rig is expected. Subsequently, tests are carried out to compare the results with the predictions of the existing models and to identify potential for optimization. A second test series is conducted to establish a broader data basis for optimizing the crack prediction models. Depending on the outcome of these tests, either an existing model is modified or an entirely new empirical approach to crack prediction modelling is introduced. The newly developed and improved predictive models can then help enable more targeted maintenance of railway wheels and rails and make the wheel–rail system more resistant to damage caused by surface cracks.

• DFG Programme: Research Grants
• Co-Investigator: Professor Dr. Christian Schindler