PSI - Issue 43

Robert Szlosarek et al. / Procedia Structural Integrity 43 (2023) 41–46 Author name / Structural Integrity Procedia 00 (2022) 000 – 000

45 5

Fig. 4. contact pressure distribution after the pretension of the bolt.

3.2. Comparison of experimental and numerical results Using the results from Szlosarek et al. (2022) it will be evaluated if it is possible to predict the results by a numerical model. Fig. 5 shows the results for the load cycles to crack in dependency of the preload. All tests were done with a tension force fluctuating between 13 and 130 kN, hence a force amplitude of 58.5 kN and a load ratio of R=0.1. The blue markers are for the experiments and the red ones for the numerical results. Only for the test without a bolt (preload of 0 kN) a good agreement between the simulation and the experiment can be stated. For all other preloads the numerical results show a too few load cycles to crack initiation. Only the tendency of the increase in load cycles with increasing preload is predicted by the simulation model. The comparison of the crack initiation locations in the experiment and simulation is shown in Fig. 6.

Fig. 5. comparison of the experimental and numerical results in dependency of the bolt preload.

Fig. 5 as well as Fig. 6 indicate that the crack initiation location is not always predicted correctly. In the simulation the crack always starts at the borehole of the bolt. This is obvious by the maximum of the damage value shown in Fig. 6. The contour plot describes the distribution of the damage value per cycle. The maximum can be interpreted as the crack initiation location. In the simulation the crack always starts at the borehole of the bolt whereas the location changes in the experiments. For low preloads like 0 kN, 10 kN, 20kN and 30 kN it is at the borehole and for 110 kN and 225 kN the cracks start in the fretting zone between the nut and the sheet. Hence, for 110 kN and 225 kN the simulation models shows a wrong location as well as a wrong number of load cycles.