Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Comsol terminates in strange ways
Posted Apr 26, 2017, 6:01 a.m. EDT Electrochemistry, Results & Visualization, Studies & Solvers Version 5.2a 1 Reply
Please login with a confirmed email address before reporting spam
Hi! I've been wondering about this for some time now, as it happens quite a lot with me. How is it, that the "Termination criterion" is set to "Solution or residual", and the "Relative error" is set let's say to 0.001. ("Temination technique" is "Tolerance".) It happens quite a lot that the SolErr is e.g. 1e-10 and still the Fully Copuled solver terminates with an error saying something like:
"Failed to find a solution for all parameters,
even when using the minimum parameter step.
No convergence, even when using the minimum damping factor.
There was an error message from the linear solver.
The relative error (0.0027) is greater than the relative tolerance.
Returned solution is not converged."
(I'm also using an auxiliary sweep.) The minimal damping factor is set to 1e-6, the one that shows up in the solver log is much greater (it even happens that it's exactly 1.0). The last parameter step that shows up in the log is also too large, around 0.1, whereas the minimum step is around 0.001. Can somebody explain these strange phenomena?
"Failed to find a solution for all parameters,
even when using the minimum parameter step.
No convergence, even when using the minimum damping factor.
There was an error message from the linear solver.
The relative error (0.0027) is greater than the relative tolerance.
Returned solution is not converged."
(I'm also using an auxiliary sweep.) The minimal damping factor is set to 1e-6, the one that shows up in the solver log is much greater (it even happens that it's exactly 1.0). The last parameter step that shows up in the log is also too large, around 0.1, whereas the minimum step is around 0.001. Can somebody explain these strange phenomena?
1 Reply Last Post Apr 26, 2017, 8:47 a.m. EDT
Please login with a confirmed email address before reporting spam
Posted:
7 years ago
Hi! I've been wondering about this for some time now, as it happens quite a lot with me. How is it, that the "Termination criterion" is set to "Solution or residual", and the "Relative error" is set let's say to 0.001. ("Temination technique" is "Tolerance".) It happens quite a lot that the SolErr is e.g. 1e-10 and still the Fully Copuled solver terminates with an error saying something like:
"Failed to find a solution for all parameters,
even when using the minimum parameter step.
No convergence, even when using the minimum damping factor.
There was an error message from the linear solver.
The relative error (0.0027) is greater than the relative tolerance.
Returned solution is not converged."
(I'm also using an auxiliary sweep.) The minimal damping factor is set to 1e-6, the one that shows up in the solver log is much greater (it even happens that it's exactly 1.0). The last parameter step that shows up in the log is also too large, around 0.1, whereas the minimum step is around 0.001. Can somebody explain these strange phenomena?
I'm not sure that I fully understand your problem but it sounds like it could be a couple of different things; if you uploaded a model, it would really help to see your solver configuration. Also, what parameters are you sweeping over? It looks like one of your parameters is just causing a greater relative error than the tolerance. Check the log to see this.
--
Nick