Fatigue of Engineered Metallic Materials under Constant- and Variable-Amplitude Loading (#12)
1 Mississippi State University, Aerospace, Mississippi State, Mississippi, United States of America
In 1961, the classic paper by Paris, Gomez and Anderson on “A Rational Analytical Theory of Fatigue” was a major development in the study of fatigue. The newly emerging field of Fracture Mechanics, driven by the works of Griffith and Irwin, began to help engineers characterize fracture of brittle materials; and to provide a crack-tip parameter, the stress-intensity factor (K), to correlate fatigue-crack-growth-rate data on metallic materials for different crack configurations, and provided a methodology to predict the failure of cracked structural components.
In the early 1970’s, attempts to predict “fatigue” of metallic materials using the stress-intensity-factor concept was unsuccessful. Several things had to be developed before fatigue, especially under spectrum loading, could be predicted using Fracture Mechanics: (1) stress-intensity factors for small surface cracks, (2) crack-closure theory, (3) plasticity effects on crack-driving parameters, (4) constraint effects on crack growth and closure, and (5) small- and large-crack data in the "threshold" regime without load-history effects. After several decades of research, these concepts began to merge together and the vision that Paris et.al had could now be achieved on "engineered" metallic materials. These are materials that nucleated cracks at constituent particles, inclusions, grain boundaries, voids, and, also, manufacturing defects.
For engineered materials, test and analysis programs on “small-crack” behavior have shown that the majority of the fatigue life is consumed by crack propagation from a micro-structural feature. Thus, the nucleation life in classical fatigue is actually crack propagation from a micro-structural feature to about 250 micro-meters. Therefore, the “crack-propagation” approach gives a unified theory for the determination of fatigue life.
For spectrum loading, the influence of load-history effects is key to predicting the fatigue life. Rainflow like-methods are required to predict material damage when damage rules are non-linear. At a small or large crack front, damage (crack growth) is a function of the current loading and load history; and the current damage is not a function of future loading. Thus, a “rainflow-on-the-fly” methodology was programed into the FASTRAN life-prediction code, based on Elber’s crack-closure concept.
The current paper reviews the application of the FASTRAN code to the fatigue behavior of notched and un-notched coupons made of aluminum alloys, titanium alloys, steels and additive manufactured materials. In general, the code worked very well using the same initial micro-structural flaw size for a given material under both constant- and variable-amplitude loading.
Keywords: cracks, fatigue, crack propagation, plasticity, spectrum loading