PSI - Issue 68

Martin Nesládek et al. / Procedia Structural Integrity 68 (2025) 527–533 Martin Nesládek et al. / Structural Integrity Procedia 00 (2025) 000–000

531

5

3. Numerical analysis of temperature increase due to friction in contact The presented estimates of &' by using the SH method are in a good agreement with the conventional values. Nevertheless, in order to obtain SH data unaffected by contact friction, numerical simulations of the temperature increase due to this heat source at various axial stress amplitudes were conducted.

Fig. 6. Workflow of FE simulations for analyzing the specimen temperatures due to contact friction.

The workflow for the numerical simulations is illustrated in Fig. 6. The uncoupled thermo-mechanical finite element (FE) analysis was performed in two stages: 1) A 2D elastic-plastic mechanical analysis to determine contact shear stress and slips within a stabilized cycle, and 2) A 3D transient thermal analysis to capture the temperature evolution in the specimen’s fatigue process zone, using surface heat flux derived from the contact shear stress and slips, as shown in Fig. 6. Surface heat flux values were mapped as boundary conditions for the thermal analysis. To achieve the highest possible resolution, the 2D model was meshed with linear elements as small as 10 µm.

Tab. 1. Comparison of stabilized temperatures from FEA due to friction in contact and those from IR measurement. Case of single-surface fretting pads.

Θ [°C] from FEA COF = 0.8

Θ m [e°aCs]u frreomme InRt 0.04

σ a [MPa]

COF = 0.5

COF = 1.0

87.6

0.08 0.26 0.66 1.81 4.29 N/A

0.05 0.16 0.39 0.72 1.31 1.91

0.05 0.14 0.32 0.59 1.10 N/A

131.6 175.4 219.4 266.7 296.9

0.17 0.53 1.29 2.90 3.80

Tab. 1 and Tab. 2 summarize the calculated stabilized temperature differences for different values of the coefficient of friction (COF). The rightmost column then shows the values from the SH test for comparison. It can be seen that for the bridge pads especially, the simulated temperatures are significantly higher than the measured temperatures and only fall below them for the non-physical COF values. It is obvious that even lowering the value of (i.e., the percentage of frictional work transformed to heat flux) from 1.0 to 0.8 will not lead to a significant improvement.