PSI - Issue 37

Carl Fällgren et al. / Procedia Structural Integrity 37 (2022) 948–955 Carl Fällgren / Structural Integrity Procedia 00 (2019) 000 – 000

952

5

4. Fatigue life calculations As mentioned before, the process of autofrettage has multiple effects on the components' fatigue lives. They arise from the residual stress distribution after the autofrettage process. The total fatigue life of the specimens is a combination of the crack initiation lives and the crack propagation lives. For the non-autofrettaged specimens, as no residual stresses are retarding the fatigue crack growth, it can be assumed that the crack initiation lives largely dominate the overall fatigue lives. 4.1. Crack initiation life calculation The crack initiation life was calculated using the local strain approach in combination with the damage Parameter RAM according to Fiedler and Vormwald (2018) which is based on the Smith-Watson-Topper parameter with an extension to include the effect of the mean stress sensitivity. The knee point for the endurance limit was set at 10 5 cycles. Additionally, according to the method used, the influence of the stress gradient and the highly-stressed surface were taken into account. 4.2. Crack propagation behaviour The crack propagation behaviour was evaluated in order to estimate whether the initiated cracks are prone to arrest. This was done by using a linear-elastic fracture mechanics approach. For this purpose, the range of the stress intensity factor was calculated by a weight function established for the thicker specimen geometry h/d = 2.5 by Schlitzer, see Vormwald et al. (2018). The approach takes into account the residual stresses along the bisector line, the pressure onto the bore surface and the pressure on the crack edges. The results of the evaluation are shown in Fig. 5 for the thicker specimen geometry for the two applied autofrettage pressures. The threshold value of the stress intensity factor is shown as a black, solid line. The method assumes that a drop of the range of the stress intensity factor below the threshold value (after crack initiation, which is marked with an x) causes a crack to arrest. On the left side the range of the stress intensity factor is shown for three different pressures on the specimen after an autofrettage simulation with 850 MPa. It can be seen, that for any sufficiently high pressure range to which the specimen is subjected, the applied stress intensity factor range never drops below the threshold value. Therefore, for the specimen subjected to 850 MPa of autofrettage pressure, no crack-arrest is predicted. On the right hand side in Fig. 5 the range of the stress intensity factor is shown for the specimen simulation of autofrettage with the maximum autofrettage pressure of 1700 MPa. Here it can clearly be seen, that the range of the applied stress intensity factor rises after crack initiation but subsequently drops below the threshold value with increasing crack length. Here, crack-arrest is predicted for pressure ranges as high as 647 MPa.

Figure 4: Range of the stress intensity factors for the thick specimen geometry h/d = 2.5 after the simulation of autofrettage with 850 MPa (left) and 1700 MPa (right).