I am having trouble in several models I run where contacts are involved. I am applying a contact between two surfaces and when the simulation is run the following error message appears:
No convergence could be reached. If you used manual timestepping try to decrease the timestep or use auto-timestepping. Otherwise check your settings again or contact our support team for problem specific help.
Is this due to the fact that one surface is given a much finer mesh than the the other when I use automatic tetrahedral meshing?
I have seen in some simulations such as Transient Thermal Analysis of a PCB that sections of one surface, the PCB in this, are defined as their own topological entity set.
How is this done? Are there several ways to resolve the issue I am having.
getting a converged solution is of course dependent on the mesh, so your question is very valid. It is not the only important factor though. Having two faces with a different fineness involved in a contact is often used even intentionally as the slave surface is often meshed with a higher fineness as the master surface. As you do not have any public projects I assume that you are working on a private project. If your data is not confidential you can simply make the project public and share the link here, then I can have a look. Otherwise you can also share your project for example with the SimScale support. Then you do not need to make your project public. In this case please mention this forum post in your question.
In the PCB analysis the top face of the plate has been splitted in the CAD pre-processing to have matching faces with the components. This makes it much easier to adapt the mesh fineness in the contact regions while not increasing the overall mesh size by a lot when meshing the complete board surface with fine parameters.
