[Time4Tea]

I am fairly new to COMSOL, although I have a fair bit of prior experience with CFD/numerical methods. I am trying to run a stationary EM wave, frequency domain simulation of microwaves heating some dielectric material. However, I am seeing some very strange behavior with use of the iterative solver.



I am using COMSOL v6.0. For the stationary linear solver, I am using GMRES with a geometric multigrid preconditioner. For the pre- and post-smoothers, I am using SOR Vector and for the coarse solver, I am solving direct (using Pardiso). When I run the solver, the reported error on the convergence plot drops rapidly to below 10-7 within a few iterations; however, I can't get the solver to terminate. I have tried entering high values (such as 0.1) in the relative tolerance fields; however, regardless of what I put there, the solver just keeps on going and won't stop.



Does anyone know what might be the issue here? The level of error the solver is reaching is likely to be fine for what I need, but I just can't get the solver to terminate. I am very confused as to what is going on and how I can resolve this.