PSI - Issue 75

Tomáš Karas et al. / Procedia Structural Integrity 75 (2025) 150–157 T. Karas et al. / Structural Integrity Procedia 00 (2025) 000–000

155

6

(a)

(b) Fig. 5. (a) K&V S-N curves obtained from fretting fatigue tests at R = 0; (b) LinExp method evaluation of σ FL for case (R = 0, r = 1mm).

3. Results and Discussion

3.1. Conventional Fretting Fatigue Results

For each R ratio and pad geometry, multiple samples were tested at varying stress levels. The resulting S-N curves and the fatigue limits were derived from them by fitting the Kohout and Veˇchet (2001) (K&V) function (Eq. 1): σ a = A C N + B N + C β , (1) where σ a is the stress amplitude, N is the number of cycles to the test termination, A , B , C and β are parameters of regression (see Table 2.). For the case of R = 0 loading, a plot of the S-N curve is shown in Fig. 5. (a) for both pad geometries. The lower fatigue life observed for the pads with r = 1 mm is to be expected and is caused by the bridge type geometry that inherently increases the slip amplitudes. The self-heating measurement data were evaluated by the LinExp method and included in Table 2., with percentage deviation in comparison to K&V σ FL from conventional fretting fatigue tests. On top of that, estimate by the LinExp method was plotted in Fig. 5. (b) for the case of R = 0 loading and r = 1 mm pad geometry. The plot depicts both the elastic (green dash-and-dot line) and plastic (yellow dashed line) parts of the fitted LinExp function. For an easier comparison between the fatigue limit determined from K&V regression and self heating estimate, the stress amplitude axis in Fig.5. (b) was normalised by the σ FL measured conventionally. 3.2. Self-Heating Results

Table 2. Summary of fretting fatigue test results, Kohout and Veˇchet regression parameters and fatigue limits. Case ID Pad type R-ratio Pad radius A B C β K&V σ FL

LinExp σ FL

Di ff [%]

–

–

[–]

[mm]

[–]

[–]

[–]

[–]

[MPa]

[MPa]

MNJ005 Single contact MNJ006 Bridge type MNJ007 Bridge type MNJ008 Single contact

-1 -1

50

45943.9 8701.4

5848 277768 -0.431 210

198.83 166.59 106.65 144.53

-5.3

1 1

1

642222 -0.301 158

5.4

0 0

3955520.6 43251 180181 -0.860 119 30294182.3 51947 169950 -1.007 164

-10.4 -11.9

50

The LinExp method results di ff ers, mainly depending on the type of loading. In the case R = − 1 loading, it estimates the fatigue limit in the range ± 5% around the measured FL. On the other hand, in the two cases where non-zero mean