A probabilistic Kitagawa-Takahashi diagram for fatigue strength assessment of cast aluminium alloys

Aluminium cast components containing intrinsic flaws were investigated in terms of fatigue strength. Therefore, quasi-static, fracture mechanical and fatigue tests were conducted at specimens exhibiting varying microstructural characteristics. Subsequently, the specimens were metallographically and fractographically analysed to evaluate spatial extent of the fracture initiating defects as well as the statistical distribution of defect sizes. Furthermore, the near defect-free fatigue strength was investigated to evaluate the upper boundary in fatigue strength for the Kitagawa-Takahashi diagram. In addition, the fracture mechanical test results were statistically evaluated as probabilistic values. Further on, a probabilistic evaluation of the crack resistance curve-concept was set up, taking the statistically distributed long crack threshold into account. The fatigue tests were finally utilized as probabilistic, defect correlated, fatigue design approach. The experimental results revealed that the fracture elongation of HIPped material increases by an average of 146% and the fatigue strength elevates of about 35%. The evaluated parameters from extensive crack propagation tests propose a long crack threshold of approximately 4 MPa root m and an intrinsic threshold of slightly beneath 1 MPa root m. The probabilistic Chapetti and El Haddad approaches lead to a sound correspondence with the experimental test data. Therefore, the presented fatigue assessment scientifically supports the lightweight design process especially for utmost probabilities of survival.