Finite Element Modeling of Elastohydrodynamic Lubricated Rolling Contact Fatigue

Neil R Paulson, Purdue University


Rolling contact fatigue (RCF) is a complex tribological problem due to its interdependence on multiple physical phenomena. Fluid mechanics, solid mechanics, damage mechanics are all relevant to the development of RCF failures. Due to computational requirements of modeling the multiple physical phenomena that govern RCF, typical RCF problems are simplified to only include a few physical phenomena. However, without addressing all the mechanisms at work in the system, the model is limited to particular applications and problems. Recognizing the multiple physical phenomena at play in rolling contacts, this preliminary dissertation presents several models which extend the current RCF models to include the interdependence of physical phenomena previously not considered in RCF research. The first model developed integrates a topological microstructural model with a damage mechanics model of fatigue crack growth. By including microstructural variation with the damage model, it is capable of predicting fatigue life variations in RCF. The coupled microstructure topology and damage mechanics model is based on previously developed RCF models; however, it was extended to study the effects of refurbishment on RCF failure. Using the RCF topology model, after a set number of initial fatigue cycles, refurbishing was simulated by removing a layer of surface material yet retaining the subsurface damage that accumulated from the original fatigue cycles. After refurbishment, simulated microstructures were subjected to additional fatigue cycles until failure occurred. It was observed that the increase of fatigue cycles prior to refurbishing decrease the refurbished life while increased refurbishing depth increases the life. Next, an extended microstructural model is presented which accounts for crystal anisotropy and orientation in the microstructure. The polycrystalline nature of steel components is known to create stress concentrations between crystals of incompatible orientation. The model developed determines how crystal anisotropy and microstructure texture affect the initiation of RCF cracks. Next a model was developed which couples the elastohydrodynamic lubrication with a damage mechanics model for fatigue initiation and propagation. The model shows the interdependence of the EHL pressure profiles and fatigue damage. As fatigue damage occurs, the pressure profile is affected which in turn modifies the direction of future fatigue damage. While these two models have extended the current research in RCF, the next step was to integrate both models together in one combined model accounting for anisotropic material properties as well as elastohydrodynamic lubrication. Merging these two models allows the study of the effect of anisotropic properties on EHL contact pressure, film thicknesses as well as subsurface stresses. Results demonstrate that contact pressures are significantly affected by the anisotropic properties while film thickness profiles show minimal variation. The subsurface stresses show similar variation to those observed using the Hertzian pressure assumption with stress concentrations at the grain boundaries and significant scatter matching what is typically observed in experimental literature.




Sadeghi, Purdue University.

Subject Area

Engineering|Mechanical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server