Stress intensity factors (SIFs) are obtained for a surface oblique crack under normal and tangential traction and remote extension loads. The surface oblique crack is modeled as the pseudo-dislocations. The Airy stress functions are derived for an edge dislocation in a semi-infinite plane. Stress field due to surface tractions is obtained by integrating the Flamant solution. The SIFs of Mode I and Mode II ($K_I$ and $K_{II}$) are then obtained. Finite element analysis is performed with a commercial code ABAQUS and compared with the analytical results. The analytical results are in good agreement with the previously published results for simple configurations as well as the results of FEM.
From investigating the SIFs and their ranges, following results are obtained. The growth rate of a surface oblique crack decreases due to the reduction of the SIF ranges. The crack driving force depends on the obliquity, the normal traction and the ratio of crack to traction length. The range of tangential traction is directly related with the crack growth rate. The peak value of tangential traction is found to be a key parameter to accelerate the crack growth.
A fretting fatigue experiment is conducted with Al 2024-T4 pad and specimen. The pad has a rectangular foot to simulate a punch pressure on the contact surface between the specimen and the pad. The pad length, the normal pressure and the bulk stress are varied for the experimental conditions. The load ratio is set to be zero during the experiments. A simple calibration method for measuring the contact forces is developed. Special specimens are designed for the calibration. The method developed use the strain values of the pad and the loading bars as correlating parameters. Linear relationships are observed between the fretting parameters, and used as in situ calibration curves during fretting fatigue experiments.
Fretting fatigue cracks are monitored and analyzed by image processing. The results with 10 and 5 mm pads reveal a partial slip regime. A gross slip regime is shown with 2 mm pad. All the cracks grow obliquely to some extent and then changes the direction to the surface normal. The contact surfaces, the crack initiation sites and obliquities are examined and measured with a microscope. A crack initiates at the stick and slip boundary for the specimen of partial slip, while at the boundary of slip and non-contact region for the specimen of gross slip. The crack initiation life is investigated in viewpoint of the slip regimes and the size of slip region.
The behavior of the crack growth rate is shown. It decreases at first and then increases so that a minimum crack growth rate occurs. So the shape looks like a well. The crack length at the minimum crack growth rate discriminates the fretting effective and the bulk fatigue effective regions. To compare with the plain fatigue crack growth rate, a plain fatigue experiment with CT specimen is also carried out. In the range of the oblique growth of a fretting fatigue crack, the crack growth rate of fretting fatigue is much higher than that of plain fatigue, which is a typical growth behavior of a fretting fatigue crack.
Analyses of the fretting fatigue experiments are carried out along the mid plane of specimen to deal with a plane strain condition. The elastic solution of square punch is utilized to estimate the crack initiation site with assuming the elastic-perfectly plastic behavior of a material. The predictions are in good agreement with the experimental results regardless of the slip regimes. It is found that the difference of the materials in contact by normal force only hardly affect the slip zone size in partial slip regime. Therefore, the present theory can be applied to predict the crack initiation site generally. The surface traction fields are suggested, from which SIFs are evaluated. The obliquity dependence of a crack driving force is evidenced.
