TMF Life Prediction for Aluminum Alloy Piston Head

Abstract

Recently, various lightweight materials have been applied to automobile engine components according to the development of high fuel efficiency and high performance internal combustion engine, and these materials must have the durability against the high pressure and high temperature of the operating environment. Especially, the piston head generally made of aluminum alloy takes thermo-mechanical fatigue (TMF) loading which is combination of cyclic thermal loading due to engine on/off and cyclic mechanical loading in the cylinder. Hence, the failure analysis and estimation of fatigue life considering TMF are important to design piston head. Numerical analysis using finite-element program is often conducted for appropriate design, as it could reduce the expensive and time-consuming bench tests. In order to ensure more accurate analytical reliability, it is necessary to develop the methodology of TMF failure analysis and life evaluation based on the material behavior model that appropriately reflects the mechanical behavior characteristics of aluminum alloys [1]. In this study, the low cycle fatigue (LCF) tests and the TMF tests were conducted using aged aluminum alloy specimens, and the fatigue life prediction model with the cyclic visco-plasticity behavior theory was implemented into the finite-element program ABAQUS to secure the high reliability of durability assessment. Then, the model was verified by comparing with the experimental results. The aluminum alloy specimens were aged at 250 degrees for 200 hours to account for changes in the mechanical properties of the aluminum material in the operating environment of the piston. The LCF tests were carried out at 25°C, 150 °C, 250 °C and 300 °C respectively, and the TMF tests were performed at 25~ 350 °C. In order to consider the time-dependent cyclic plasticity behavior of material, the Chaboche model was adopted and to predict the TMF life, the mechanism based life prediction model was adopted [2]. The numerical analysis describes cyclic hardening, softening, stress relaxation and strain rate sensitivity behavior well, within 10% of error. Also, the numerical analysis results of the fatigue life also correspond well with the experimental results as shown in Fig. 2. As a result, a TMF life prediction model coupled with cyclic visco-plasticity theory for piston head aluminum alloy was developed and found to be valid, and it is expected that the model can be sufficiently applicable to the numerical evaluation of TMF life as well as LCF for aluminum alloy.