Holographic detection of fatigue-induced surface deformation and crack growth in a high-strength aluminum alloy

Changes in optical correlation intensity (I)_c are observed during fatigue cycling of 2024-T3 aluminum alloy. TheI _c measurements are made by transmitting light scattered from the specimen surface through a holographic filter containing information about the surface topography at an earlier time. Topographic changes such as slip band development, microcracking, and crack propagation are observed and recordedin situ during fatigue cycling of individual specimens and cause corresponding changes in correlation intensity. A three-stage curve log (I)_c vs number of fatigue cycles is observed for both unnotched and notched specimens. The overall shape of the curve is not affected by the applied stress levels in constant amplitude tests. Thein situ metallographic observations confirm that region A of the correlation intensity curve corresponds to progressive roughening of the specimen surface caused by slip during the early part of the fatigue life, together with a rapid increase in the number of microcracks of the order of a few micrometers in length. Few metallographic changes are observed during region B of the curve, where the correlation intensity remains relatively constant. The accelerating loss of correlation intensity in region C of the curve arises from the elastic and plastic displacements which occur as a crack or cracks grow beyond about 10 µm in length. The metallographic observations also show that for both notched and unnotched specimens, the correlation intensity readings in region C are sensitive to factors such as crack branching, crack-tip plasticity, and changes in crack growth direction as well as to the overall increase in crack length. The total loss of correlation intensity from the beginning of fatigue cycling to the development of a crack about 800 µm in length can be more than eight orders of magnitude at the present sensitivity of our experiments. The optical correlation technique is an extremely sensitive method of detecting remotely, in air, fatigue damage, and the propagation of fatigue cracks from ten to several hundred micrometers in length. The correlation intensity curve provides an indication of developing fatigue damage and impending fatigue failure in individual specimens, and detects the onset of crack propagation with no prior knowledge of the presence or precise location of particular flaws or cracks.