High Strain Fatigue Crack Growth and Crack Closure

Abstract: Understanding of the growth of fatigue cracks is of utmost importance since such growth often has a profound influence on the life of components subjected to cyclic loading. Thus, reliable fatigue life models enable a more efficient use of materials and improve the performance and efficiency in many applications. This thesis deals with the growth of fatigue cracks subjected to high load amplitudes. One mechanism that is known to have a strong influence on fatigue crack growth is crack closure, i.e. premature contact between the crack surfaces, caused by for instance residual plastic deformation, crack surface asperities or oxidation of the crack surfaces. Closure can reduce the effective load driving the crack, thus influencing the crack growth rate. In this thesis the use of the electrical potential drop (PD) technique for crack closure measurements has been investigated by a combination of numerical simulations and experiments. It has been shown that crack closure has a strong influence on the variations of the PD-value during a load cycle. Also, crack opening measured with the PDtechnique is consistent with closure measurements made from in situ fatigue crack growth observations using a scanning electron microscope. Furthermore, the growth of fatigue cracks subjected to high load amplitudes in Ti-6Al-4V at room temperature and Inconel 718 at 650°C has also been studied. It has been shown that crack growth under high strain amplitudes can be analysed using a strain intensity approach or by using the cyclic J-integral. In both cases crack closure plays an important role and must be accounted for by using an effective strain intensity range or an effective cyclic J-integral. For Inconel 718 at 650°C, crack growth occurs by a combination of cyclic and time dependent growth, and thus the load frequency is of importance for the crack growth rate. A crack growth law based on the product between the effective cyclic J-integral and a function compensating for the frequency was proposed. Finally an in situ SEM study of crack growth in an aluminium alloy was performed. It was shown that on a micro scale crack growth is a highly irregular process that is strongly influenced by the local microstructure at the crack tip. Also, there is no correlation between the local crack opening and the crack growth rate. Thus, it is difficult to predict the crack growth during an individual load cycle, but for crack growth rate on a macro scale the stress intensity range, compensated for crack closure, is a proper measure.