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

1178

6

Figure 5. S-N curves of notched specimens, 6 mm diameter, K t = 2.84, HSV 90% of 0.104 mm³; red curve: component; blue curve: cast samples

Table 3 summarizes the evaluation of the fatigue tests. The scatter band increases with smaller HSV 90% . The fatigue strength of the large specimens is 14 % (component) and 17 % (cast sample) higher compared to the small specimens. The smaller increase of the fatigue strength in specimens from the component is due to the technological size effect resulting from wall thickness differences. Table 3. S-N curve parameters

Casting

Specimen geometry Kt = 1.05 ⌀ = 12 mm Kt = 1.06 ⌀ = 6 mm Kt = 2.84 ⌀ = 6 mm

HSV 90% [mm³]

S-N curve parameters

T σ

k

N k

σ a,n (N k ) [MPa]

σ a,n,lim [MPa]

1.18 1.10 1.31 1.36 1.43 1.48

Cast sample Component Cast sample Component Cast sample Component

3240

9.1

1ꞏ106

99 98 91 94 66 74

104 104

10.7 1ꞏ106

378

7.7 5.5 5.5 4.9

2ꞏ105 2ꞏ105 5ꞏ105 3ꞏ105

84 86 62 68

0.104

To determine the statistical and geometrical size effect, the local stress amplitudes  a,l obtained from the stress controlled tests are plotted against the highly stressed volume HSV 90% (Figure 6). The local stress amplitude is calculated as the product of the nominal stress amplitude  a,n,lim and the notch factor K t . The slope ν can be specified for each section of HSV 90% . According to equation (1) and using the HSV 90% diagram, it is now possible to interpolate the local stress amplitude to another highly stressed volume HSV 90%,II of a given component to assess its local stress amplitude  a,l,II . ������ � ����� �� ����� ������ � � (1)