When I use mesh adaptivity in my stamping simulation I don’t know how to define the proper adaptive frequency. Would you please give me some suggestions?

It depends on the specific case. If the initial mesh is too coarse, it is necessary to have a very high frequency; if the initial mesh is fine enough, it is possible to have a low adaptive frequency. Proper adaptive frequency should make sure that there are enough elements in the contacted region with curvature. As a rule of thumb, it is desired to have at least four (five to six is better) elements for a 90-degree radius.

If the mesh is not adapted enough, the deformable sheet metal may not have enough elements near small radii of the rigid tools. Therefore, the detailed features of the tool will not be properly captured. Accordingly, the stress distribution in the blank might not be correct and the strains will be under predicted.

A fine mesh in the intemediate forming stage is also important (usually people will pay more attention to the final stage mesh), this is especially important for a subsequent springback analysis. Sometimes, it has been found that even though the final mesh is suitably fine, the springback prediction is in the wrong direction. This result may be a direct consequence of not having enough mesh resolution during the intermediate stage.

However, it is not proper to adapt the mesh too frequently, because the problem must be initialized after each adaptive cycle is completed.