Investigation of non-uniform thickness section effect on twist springback behaviour
Loading...
Date
2018-07-01
Authors
Muhammad Nazmi Nashrudin
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Nowadays, automotive manufacturers are facing major challenges such as the increasingly strict emission regulations and the competitive need to manufacture fuel-efficient vehicles. These issues can be partly addressed by reducing the weight of a vehicle by using automotive components with non-uniform thickness. However, stamped metal parts produced using forming processes experience springback defects, which would result in components that do not conform to the required product tolerances and twist springback is among the most difficult to predict. Past studies on the measurement and prediction of twist springback have mostly focused on uniform thickness sections, and studies on non-uniform thickness sections are still lacking. Non-uniform thickness sections experience uneven strain hardening, making the prediction of twist springback behaviour difficult. In this study, we have investigated the effect of twist angle, material type and thickness ratio of several types of non-uniform metal profiles. The twist forming process was modelled in a commercial FEA software, ANSYS Workbench, using a static structural analysis procedure. The twist forming simulations were conducted for flat, tapered, stepped, convex and concave profiles. A graphical method was used to measure the springback angle using MATLAB software. The results have shown that increasing the twist angle would decrease the twist springback, while a low thickness ratio value increases the twist springback. Additionally, material with a low Young’s modulus value and high yield strength has a high value of twist springback. Example, the Aluminium Alloy 6061 specimen, having the lowest Young’s modulus as compared to the other metals chosen, obtained the highest springback value. In conclusion, all the parameters chosen had significant effects on the springback behaviour. The most significant parameter was the thickness ratio, with the twist angle being the last influencing. For experimental validation of the simulation work, only the flat and tapered profiles were selected. Comparison between the simulation and experimentation have shown good agreement for the flat profile, proving the reliability of the simulation model developed. However, some variations were observed for the comparison between the simulation and experimentation of the tapered profile. This difference could be attributed to the strain hardening effect in the production of the non-uniform thickness section profiles (Merklein et al., 2014).