At the moment I am performing a mesh convergence study for specific areas in a model. It works fine, until I get to a point of refinement, where the solver produces an error:
Warning: AMG solver might encounter difficulty (tolerance=0.10E-05)
Please inspect model for stability or try direct solvers.
Warning: Found trivial solution (|X| <= 1.0E-38).
Missing all loads?
**** End AMG Iterative Solver Solution
Basically, the only thing I do is to increase the refinement. I have a mesh of 80 and a refinement factor of 5 is no problem. Everything finer produces the error. What could be a reason for this??
I am sure that the loads are connected to the model and the boundary conditions do exist as they do in the coarser versions of the mesh.
Solved! Go to Solution.
Solved by klnkai. Go to Solution.
After a day of receiving error messages I finally found a solution by using Microholes.
In addition to the error message described above the solver also indicated a warning that said a zero jacobian element was found. At first I did not see a connection, but obviously there was one. However with microholes it eventually worked.