This work details an analytical assessment of heat generation in a turbine aero-engine main-shaft bearing and the development of a model to predict that heat generation. The new model is based on an empirical model, previously developed by the Air Force Research Laboratory (AFRL), which features physics based terms multiplied by empirical regression coefficients. That model proved to be limited in that portions of the terms were essentially an extension of the regression coefficients due to the fact that the experimental data was limited to that of one bearing. Additionally, there were separate models for each rolling element material. To develop the new model, the validated bearing analysis code ADORE was used to generate power loss data for angular contact ball bearings of various sizes. The effects of speed, thrust load, pitch diameter, element diameter, number of rolling elements, lubricant inlet temperature, lubricant flow rate, and rolling element material (AISI M50 bearing steel and silicon nitride) are examined. Speed and thrust load are addressed at four levels each. Number of elements, bore diameter, and element diameter as well as lubricant temperature and flow rate are each addressed at three levels. These effects are captured in the model through traction (friction), churning (drag), and shearing (viscous) terms and their respective regression coefficients. The material effect is address through the use of an effective elastic modulus within an estimate of raceway to rolling element contact area. The performance of the model was then compared with experimental data collected in the AFRL High Mach Engine (HME) Bearing Rig. The model created in this work provides designers with an effective tool to examine bearing heat generation during the early engine design phases, avoiding the significant computational and front end expense of other, more detailed methods.
