Would like to share some strange results I came across recently.
I was running a static study, with a solid draft quality curvature-based mesh, and started with a fairly coarse mesh. The study used a symmetry restraint and was based on one solid body., I found a possible singularity in the stress results and in order to investigate I started manually refining the global mesh, when I refined smaller than a certain size, the results varied dramatically and were definitly unrealistic.
I have included a screen shot below showing the original result (left) and the blotchy result (right), where it can be seen that the stress is also significantly reduced.
Has anyone come across this before?