An extensive experimental investigation of different temperatures and strain rates has been carried out on aluminum alloy Al319-T7. It is found that elevated temperature tends to decrease the strength in tensile and cyclic tests, whereas increasing strain rate appears to improve the strength, and in strain-controlled isothermal fatigue tests, higher temperature results in shorter fatigue life for a given stress amplitude. In addition, increased strain rates are found to improve the fatigue life in high-temperature tests, while their effect is negligible in room-temperature tests. In-phase (IP) and out-of-phase (OP) thermo-mechanical fatigue (TMF) tests with and without dwell time are also conducted in this paper. The results reveal that for the same applied total strain amplitude, OP tests, owing to their larger mechanical strain amplitude, are more critical than IP tests. Furthermore, stress relaxation happening in dwell time will help to increase the fatigue life, compared to those tests without dwell time.

Introduction

Nowadays, aluminum alloys are used more frequently to replace cast iron and steels in various mechanical and structural components of vehicles for the purpose of increasing fuel economy by reducing weight. One typical use is in the cylinder heads of internal combustion engines, the operating temperatures of which have been increased to improve engine efficiency. The peak temperature they may experience is up to 300˚C [1]. These components subjected to high cyclic thermal and service loads are prone to fatigue failure. Thus, proper understanding of the material behavior at high temperatures is required. Engler-Pinto et al. [2] investigated the thermo-mechanical fatigue behavior of cast 319 aluminum alloys under high temperatures and found the stress-strain behavior is similar between in-phase and out-of-phase TMF tests. Under a given mechanical strain amplitude, fatigue life in IP tests is lower compared to OP tests. By using a twostate-variable unified inelastic constitutive model proposed by Sehitoglu [3], Engler-Pinto et al. [4] determined the constants in this model systematically from isothermal experiments and verified its ability to simulate the material response, including cyclic softening and thermal recovery, by comparing it with TMF experimental results.