PSI - Issue 33

A. Kansy et al. / Procedia Structural Integrity 33 (2021) 1173–1180 A. Kansy, M. Kaynak, C. Bleicher, H. Kaufmann / Structural Integrity Procedia 00 (2019) 000–000

1175

3

The chemical compositions of the casting materials slightly differs, as shown in Table 1.

Table 1. Chemical composition of the used material EN-GJL-300

C Cu Cast samples 3.08 2.07 0.56 0.14 0.02 0.33 0.08 Component 3.13 1.56 0.90 0.02 0.02 0.06 0,06 Si Mn S P Cr

2.2. Specimens Unnotched and notched round specimens with diameters in the test cross section of 6 mm and 12 mm were used for the fatigue tests. The specimen types as well as the notch factor K t and the corresponding HSV 90% are shown in Table 2. Notch factors and HSV 90% were determined by FEM-simulation.

Table 2. Specimen types with notch factors and specific HSV 90%

Notch factor K t

HSV 90% [mm³]

Diameter [mm]

Specimen geometry

Specimen

Large specimen

12

1.05

3240

Small specimen

6

1.06

378

Notched specimen

6

2.84

0.104

The removal of the specimens from the Y-shaped cast samples was carried out from the lower area with a wall thickness of 75 mm. For the component, areas of constant wall thickness were identified and used for sampling as marked in Figure 1. Large specimens were removed from an area with a wall thickness of 36 mm, small unnotched and notched specimens from areas with a wall thickness of 18 mm. Due to the wall thickness differences, the technological size effect must also be considered. 2.3. Fatigue tests Force-controlled fatigue tests were carried out on an electric resonance test rig at test frequencies f between 95 and 155 Hz depending on the specimen geometry. The tests were performed at room temperature under alternating stress (R σ = -1) until crack initiation or until reaching the limit number of cycles N lim = 1ꞏ10 7 . When N lim is reached, the specimen is considered as a run out and subsequently tested with an increased stress amplitude to estimate a tendency for hardening or softening of the material due to the cyclic pre-stressing. The results were then evaluated to determine the cyclic material properties and presented as S-N curves. The evaluation was based on the Maximum-Likelihood method [5] for a probability of survival of P S = 10 %, 50 % and 90 %. Furthermore, due to the low number of specimens, a slope of the S-N curve of k* = 44.9 is assumed for high cycle fatigue after the knee point following [6], which corresponds to a stress drop of 5 % per decade.