An engineering high cycle fatigue strength prediction model for low plasticity burnished samples

Abstract

High cycle fatigue performance is significantly affected by the surface conditions of mechanical components. Traditionally, one major effect of low plasticity burnishing process in promotion fatigue resistance has been ascribed to the generation of compressive residual stresses and the formation of gradient structure in the top-treated surface layer. However, the prediction model of fatigue performance considering the effects of roughness, microhardness and residual stress has not been well established. Thus, the aim of this work is to develop a model predicting the fatigue strength of high cycle fatigue for burnished metallic components. First, the main parameters affecting the fatigue performance of burnished components are investigated by dimension analysis method. Then, by introducing the influence factors of stress gradient and microhardness gradient, a fatigue strength prediction model is proposed to correlate the fatigue strength of untreated and treated samples. This model considers the effects of residual stress, microhardness, roughness, gradient and size. In order to verify the validity of the model, double faces low plasticity burnishing and tension-tension fatigue tests are carried out on TA2 alloy samples with center hole. In addition, to investigate the effect of low plasticity burnishing on the fatigue performance when the direction of crack propagation is perpendicular to the burnished surface, one face burnishing and three-point bending fatigue tests are also performed. The tension-tension fatigue strength and bending fatigue strength after low plasticity burnishing are 144MPa and 329MPa, respectively. The predicted tension-tension fatigue strength and bending fatigue strength after low plasticity burnishing are 133MPa and 313MPa, respectively. The error between the measured value and the predicted value is less than 15%. At last, fatigue comparison tests using new burnishing process parameters are carried out. The fatigue strength after treatment is 309MPa. The predicted fatigue strength after treatment is 322MPa. The prediction error is less than 10%. This demonstrates that the proposed model is effective in the high cycle fatigue strength prediction model for low plasticity burnished samples.