Statistical Distribution of Fatigue Life for Cast TiAl Alloy

Abstract

Statistic distribution of fatigue life data and its controls of cast Ti-47.5Al-2.5V-1.0Cr-0.2Zr (atom fraction/%) alloy were investigated. Fatigue tests were operated by means of load-controlled rotating bending fatigue tests (R=-1) performed at a frequency of 100 Hz at 750 ℃ in air. The fracture mechanism was analyzed by observing the fracture surface morphologies through scanning electron microscope,and the achieved fatigue life data were analyzed by Weibull statistics. The results show that the fatigue life data present a remarkable scatter ranging from 103 to 106 cycles, and distribute mainly in short and long life regime. The reason for this phenomenon is that the fatigue crack initiators are different with different specimens. The crack initiators for short-life specimens are caused by shrinkage porosity, and for long-life ones are caused by bridged porosity interface and soft-oriented lamellar interface. Based on the observation results of fracture surface, two-parameter Weibull distribution model for fatigue life data can be used for the prediction of fatigue life at a certain failure probability. It has also shown that the shrinkage porosity causes the most detrimental effect to fatigue life.

Keywords

• TiAl alloy
• fatigue life
• casting defects
• Weibull statistics