PSI - Issue 57

Felix-Christian Reissner et al. / Procedia Structural Integrity 57 (2024) 411–419

417

F.-C. Reissner et al. / Structural Integrity Procedia 00 (2023) 000–000 7 The estimation method takes two competing areas of failure into account. The first area is the surface layer (SL) and the second area is the transition point (TP) from compressive to tensile residual stresses. Depending on the material and the distribution of the residual stresses, both can be critical for failure. For cast iron, the estimation of the ultimate tensile strength R m needs to be adjusted. Based on the data from Bartels et al. the equation is adjusted to R m = 2 . 2 · HV . The next parts of the estimation method which need to be adjusted are the support effects. For steel, the support effect of notches is only considered for the SL. Based on the theory of critical distance according to Luca Susmel (2008), cast iron has a much higher material parameter L than steel, which can be interpreted as the effect depth of the support effect. Therefore, for cast iron the support effect needs to be considered for the SL and the TP. In deviation to the original approach, which is tailored for steel, the support factor for cast iron is determined according the Siebel&Stieler approach of the FKM-Guideline based on Siebel and Stieler (1955). In addition, the input parameter ultimate tensile strength needs to be calculated with the local hardnesses at the SL and the TP. The last adjustment concerns the mean stress sensitivity. While the mean stress sensitivity for steel depends on the ultimate tensile strength, the mean stress sensitivity for cast iron is constant at a value of M = 0 . 4 , according to the FKM-Guideline. Therefore the mean stress sensitivity is kept constant. 4.2. Results and Discussion In the FKM-Guideline the slope k and the knee point N k of the S-N curve is constant and depends on the material and manufacturing processes. Therefore, an adjustment of the slope depending on e.g. the residual stresses is not possible within the framework of the FKM-Guideline. Consequently, the estimation method introduced in section 4.1 only adjusts the S-N curve in the load direction. Given this constraint, the load level of the S-N curve which has the highest change in residual stresses must be selected as the design point to ensure a conservative design. In the presented example, the design point for steel is 2000 Nm and for cast iron 900 Nm. At these load levels the residual stresses are completely transformed. However, the method gives the possibility to adjust the design to an arbitrary load level which can be used to design a S-N curve based on multiple load levels and different residual stresses. The method shows a conservative estimation of the fatigue strength at the design points for both materials, Figure 6. The S-N curve of the cast iron specimens has a conservativeness of approximately 100 Nm in the very high cycle fatigue domain ( N > 10 6 ). This conservativeness is probably a result of the remained residual stresses, which are stable at this load level. In order to obtain a less conservative result, the design point for the S-N curve must be changed, e.g. to 600 Nm. However, this would lead to a non-conservative estimation at 900 Nm. In accordance with the experimental results, the failure location is at the surface. In the high cycle fatigue domain as well as in the very high cycle fatigue domain, the estimation of the fatigue strength of the steel specimens is conservative. Based on the estimation the expected location of failure is the TP. However, this is not in accordance with the experimental results. This is an indication that the support effect at the TP should also be taken into account for steel. Presumably, the support effect is not taken into account by the FKM-Guideline to ensure a conservative estimate. 5. Conclusion This study addresses the accurate estimation of design S-N curves for shot-peened components by con sidering the change in residual stresses due to loading. It has been shown, that the residual stresses have an influence on the fatigue strength and that they can change under loading. Therefore, the remaining residual stresses must be evaluated and taken into account to ensure an accurate estimation of the design S-N curves. The proposed combination of the estimation method from the FKM-Guideline chapter 5.5 and simulated residual stresses resulted in a conservative agreement between experimental results and the estimated design S-N curve. However, the results are dependent on the quality of the shot peening simulation, which still remains a challenge. In addition, the application range of the method from FKM-Guideline was successfully extended to cast iron.