Method of predicting radiator temperature distributions for thermal fatigue analysis

The purpose of this study was to develop an accurate method of predicting radiator surface temperatures for fatigue life estimation due to thermal loading. Two methods are currently used for this purpose. The first method involves fabricating a prototype, mounting it on a thermal load fatigue tester, imaging the surface temperature of the heat exchanger using an infrared camera, and mapping it to a Finite-element analysis (FEA) model. The second method involves mapping the surface temperature of the heat exchanger considering the louver fins by performing computational fluid dynamics analysis using an FEA model. However, both of these methods yield poor predictions of the heat exchanger surface temperature, so the accuracy and usefulness of these techniques for fatigue life estimation are low. In this study, an effective method of predicting fatigue lives due to thermal loading was developed in a virtual laboratory environment. A complete unsteady analysis of one thermal load cycle was performed, and the simulation results agreed closely with the experimental results. The louver fin on the air side of the heat exchanger was assumed to be a porous medium in local thermal non-equilibrium. A modified Darcy equation was utilized as the momentum equation of the porous medium together with the permeability and Ergun constant for the louver fin that were obtained in a previous study. Since the permeability, Ergun constant, and interfacial heat transfer coefficient for the louver fin are given as a functions of the louver fin parameters, no additional experiments or corrections are required when the fin parameters are changed. Therefore, an efficient and accurate numerical method of determining automobile radiator temperature distributions for fatigue life prediction without requiring a prototype model was developed.