Electronic Theses and Dissertations



Document Type


Degree Name

Doctor of Philosophy


Mechanical Engineering

Committee Chair

Ali Fatemi

Committee Member

Ali Fatemi

Committee Member

Gladius Lewis

Committee Member

Amir Hadadzadeh


Defects in metallic components can be formed in many manufacturing techniques. These defects can significantly affect the mechanical properties, especially fatigue behavior of industrial components. In this study, the effect of defects on fatigue behavior and life prediction of a cast aluminum alloy, as an illustrative material containing defects, was investigated. Defect characterization was performed by using metallography, X-ray radiography and micro-computed tomography techniques. The variability of defects between the specimens of two sizes, different porosity levels, and based on location within the specimens were studied. The maximum defect size within the specimens was also estimated using extreme value statistics. In addition to fully-reversed fatigue tests, tension-tension as well as rotating bending tests were conducted to study the effects of mean stress and stress gradient. Fatigue tests were also performed under torsion, and under in-phase and out-of-phase combined axial-torsion loadings. Fatigue life predictions were performed using long and small crack growth models based on the observed defects size from fracture surfaces, as well as the maximum defect size estimated by extreme value statistics. Considering the effect of mean stress and overall plastic deformation, a small crack growth model is proposed which predicts fatigue lives in very good agreement with experimental results. Variable amplitude loading fatigue test results representative of service load histories were also predicted following both S-N and fracture mechanics approaches and the results are compared and discussed with reference to the experiments conducted.


Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to ProQuest