PSI - Issue 57

6

Author name / Structural Integrity Procedia 00 (2019) 000 – 000

Luc AMAR et al. / Procedia Structural Integrity 57 (2024) 217–227

222

3. Experimental results The reproducibility of the coefficient of friction was about 4% of the mean value. Regarding wear indicators, the standard deviation was between 1% and 27% of the mean value. The figure 4 shows the summary of experimental results. The depth wear is below the hardened depth for induction, nitrided and carburized specimens which confirms that the core was not subject to fretting wear and that the test this accepted (see figure 4.d). The coefficient of friction is presented in the figure 4.a. The values range from 0.64 to 0.70 (-). These results show that the coefficient of friction of the six material-treatment combinations is very similar. The volume wear is presented in the figure 4.b. High dispensaries are observed between the 50µm and the 100µm test and for between every tested material and heat treatment configuration. The less worn configuration at 50µm slip peak-to-peak amplitude is the annealed material while the less worn configuration at peak-to-peak 100µm slip amplitude is the carburized material. Despite the white layer, the nitride shows a significant volume wear. The 42CrMo4 through hardened configuration shows the minimum coefficient of friction and the minimum volume wear. For this configuration, the wear is almost the same at 50µm and 100µm peak-to-peak slip amplitude. The carburized material volume wear seems also few dependent to the peak-to-peak amplitude. The section surface wear and the depth wear are presented in the figure 4.c and 4.d. Differences of surface and depth wear value between 50µm and 100µm peak-to-peak amplitudes are lower than observed for volume wear. Depth wear are in the same order for all tested configurations, within 5 to 15µm (except for 100Cr6 with 100µm peak-to peak amplitude). The figure 5 presents sections of the measured profiles from the worn plate. A bead is observed around the worn area, especially for the 100µm peak-to-peak amplitude. The wear diameter appears to be similar between the 50µm and the 100µm peak-to-peak amplitude. The figures 6 and 7 presents the plate and the ballfretting wear pictures taken with a microscope. Since the contact zone (from 186 µm at the start of the test to more than one mm at the end of the test) is clearly greater than that of the displacement amplitude (50 and 100 µm), the contact regime is therefore in the partialslip regime. The central zone in relief on the traces of the plate and the ball which corresponds to the “Stick” zone and a band at the periphery the “Slip” zone. The “Stick” area embo ssed on the ball is from the transfer of material from the plate. This area is significantly larger with annealed 42CrMo4 than with Through hardened 42CrMo4, which shows the adhesion mechanisms are more important for annealed 42CrMo4. Streaks on the trace of the 100 µm balls according to the direction of the friction are observed, whereas these traces are barely invisible on the 50 µm traces. The length of these streaks matches well to the range of motion. As the streaks are caused by abrasion which is a more important mechanism at 100µm than at 50µm. This may give some elements to explain why some heat treatments are not sensible to the variation of amplitude.