PSI - Issue 35

Vera Friederici et al. / Procedia Structural Integrity 35 (2022) 106–114 V. Friederici et al. / Structural Integrity Procedia 00 (2019) 000–000

109

4

Schenck AG, Darmstadt, Germany) at a constant strain rate of 0.5 mm / min (until R p0.2 ) and 5 mm / min (until failure) at room temperature and in ambient air. For fracture toughness measurements compact tension specimen with W = 45.5 mm and B = 25 mm were used. 5 samples per batch were tested. The initial crack is introduced by cyclic loading with a lowest amplitude loading of 8 kN and stress ratio of 0.1. The specimens are then strained quasi-statically until failure while load-displacement curves are simultaneously recorded. The fracture toughness K IC was calculated according to ASTM E-399. 2.2. Crack propagation Crack propagation was performed on a resonant testing machine (Rumul Testronic 50 kN, Russenberger Prüfmaschinen AG, Neuhausen am Rheinfall, Schweiz) according to ASTM E-647 on compact tension samples using positive stress ratio ( R = K min / K max ) of 0.1, 0.3 and 0.5 and a test frequency of 60 - 75 Hz. Each R-curve was obtained by testing at least 3 samples. The parameters C and m of the Paris’ law were calculated (see equation 1). Preliminary experiments with R = - 1 were conducted on compact tension samples using a specially modified servo hydraulic testing machine (Hydropuls PSA, Carl Schenck AG, Darmstadt, Germany) and a frequency of 5 Hz. The range of threshold of crack propagation ∆ K th was determined by using the linear-extrapolation method. 2.3. Fatigue tests Test specimens with a diameter of 5.9 mm and ground surface were used for rotation-bending tests. For push-pull tests specimens with a diameter of 2.6 mm and ground surfaces were used. The rotation-bending experiments were performed on a rotation-bending machine (Type Punz, Carl Schenck AG, Darmstadt, Germany) with a frequency of 60 Hz. Push-pull tests were carried out using a servo-hydraulic testing machine (PC160, Carl Schenck AG, Darmstadt, Germany) with a frequency of 20 Hz. In addition to the influence of non-metallic inclusions on the fatigue behavior, another important influencing factor is corrosion of the surface. To illustrate this, some rotation-bending specimens in core condition were exposed to a salt spray test for several hours before testing. The duration of this targeted corrosion was selected as 3 hours and 24 hours. 2.4. Simulation of crack propagation Simulation of crack propagation was done by developing an Abaqus XFEM model. First, a model of rotation bending specimen was built up. During axial loading the crack will propagate in the plane of the highest stress. For meshing C3D8R element type was used and nodes were seeded on the outer perimeter of the specimen in the area of the crack. A starting crack of 0.5 mm into the surface was introduced and a bending load was applied. The resulting stress intensity factor K I was read and averaged over eight contours (starting from contour 3) for every node of the crack front. With the specification that the crack grows a specific maximum interval da , the cycles for crack propagation can be calculated by using the Paris-equation for all nodal points. The cycle number calculated for the node with highest stress intensity factor K I gives the limitation, so that for all other nodes the crack propagation is limited to values smaller than da . The calculation is continued step by step until the stress intensity factor reaches or exceeds the value of the fracture toughness K IC . Adding up all propagation steps, the simulation of the crack propagation finally gives the cycle number for crack propagation N p . By correlating data from the rotation bending experiment with the simulation results, a crack initiation SN-curve was calculated. 3. Results and Discussion Chemical analysis revealed that all elements of the 42CrMo4 steel meet the specification according to DIN EN 10083 (Table 1). The largest expected inclusion for an area of 150,000 mm 2 can be estimated to 325 µm (similar to ASTM E-2283-06) when taking all types of inclusions (sulfides, oxides, slacks) that were found into account. The heat treatment parameters to mimic the hardened raceway and the transition region between core and raceway are shown in Table 1 together with the resulting hardness and microstructure.