PSI - Issue 42

T. Fekete et al. / Procedia Structural Integrity 42 (2022) 1684–1691 T. Fekete et al.: Extending reliability of FEM simulations… / Structural Integrity Procedia 00 (2019) 000–000

1690

7

Fig. 2. Measured and simulated curves: (left) original flow curve, (right) Choung-Cho correction with n =0.145.

In this section, it is demonstrated –based on the results of a tensile test on S560 material performed on 28. 10. 2021. with the experimental setup described above and evaluated using the DT of the tested specimen– that the Choung Cho corrected flow curves have an accuracy that makes them promise for further verification and use. The experimental setup was implemented on a Gleeble 3800 thermomechanical simulator. Force and crosshead displacement were observed by in-system sensors, while standards-compliant elongation was detected by clip-type extensometers. Based on the raw data, collected by the material testing machine after the experiment, a first version of the flow curve corresponding to Eq. (6) was developed. The second stage of data processing was the determination of different contours of the test specimen from the series of images, which provided all the information related to necking and the areas of cross-sections perpendicular to the specimen’s longitudinal axis. The processing was done using an in-house developed software. Based on these data, in a few iteration steps, (1) the uncorrected and (2) the Choung-Cho corrected flow curves were determined, having the following parameters: σ y = 435 [MPa], K = 854 [MPa] and n =0.145. Post test simulations were performed on the DT with the two kinds of flow curves. The results of simulations were compared with the measurements in several aspects. One of the most important of these –the comparison of thickness reduction at necking vs. Force– is presented in Fig. 2. As clearly shown on the figure, the results of the measurements, and simulation results on the DT –based on the uncorrected flow curve– show a very good agreement in the range of 0-0.6 [mm] thickness reduction. In the larger range of thickness reductions, computed forces systematically overestimate the observed ones: at rupture at ≈ 3.15 [mm] the deviation is ≈ 13%. Comparing the measurement results with the simulation result achieved on the DT –using the Choung-Cho corrected flow curve–, it can be observed that they are in very good agreement over the range of 0 – 1.6 [mm] thickness reduction. In the range of thickness reductions larger than this, the calculated forces are systematically underestimated: at rupture, at ≈ 3.15 mm, the deviation is ≈ 10%.

Fig. 3. Cushioning effect for 2 mm thickness reduction (left), 3 mm thickness reduction (right).