Im simulating the behavior of anisotropic material using a UMAT subroutine. The UMAT works fine when using single element. However, when I use multiple elements (C3D8R), I get convergence problems.
The problem is as follows:
The simulation stops suddenly with error Too many attempts...... The point where the simulation stops depend on the elements size. When I increase the element size (coarser mesh) I get better convergence, but for finer meshes, the simulation isnt completed.
Ive tried to increase the hourglass stiffness, and I found that for higher values, the simulation is completed, but the resulting forces are higher !
Is this problem coming out from the definition of the DDSDDE matrix inside the UMAT, or its a result of using reduced elements (C3D8R)?
Thank you in advance.