Am solving species transport, melting and evaporation problem and I keep getting this error which reads "Error: Divergence detected in AMG solver: species-0". How do I deal with this?
AMG solver is turned on by default. And other option available is FAS ( which is only for density based explicit formulation).
By the way AMG is algebraic multigrid and FAS is full approximation storage, both are powerful tools used to reduce the computational time and robustness by multigrid concept. It reduces the matrix at coarser levels, to solve for error, thus error is reduced at coarser grids. And it is interpolated back to actual grid. Thereby reducing the overall computational time.
The error you are getting is a divergence issue. Please check the following:
problem formulations, boundary conditions, initialization, mesh quality etc. The problem might by with any of the above.
if it is not clear let me know.
Hope this helps you.
Regards!
AMG solver is turned on by default. And other option available is FAS ( which is only for density based explicit formulation).
By the way AMG is algebraic multigrid and FAS is full approximation storage, both are powerful tools used to reduce the computational time and robustness by multigrid concept. It reduces the matrix at coarser levels, to solve for error, thus error is reduced at coarser grids. And it is interpolated back to actual grid. Thereby reducing the overall computational time.
The error you are getting is a divergence issue. Please check the following:
problem formulations, boundary conditions, initialization, mesh quality etc. The problem might by with any of the above.
if it is not clear let me know.
Hope this helps you.
Regards!
I have the same problem. i'm working on a reacting flow with the effect of magnetic field (using MHD modul). My mesh is uniform. I reduce URF, using stabilation methods BCGSTAB ,but this problem persist .
When I change all of the equation cycles to F-cycle, I encounter 1.#QNBe+00 in residuals. when I change all of them to F-CYCLE except B_y ,I encounter AMG divergence with uds_scalar1 .And when I change all of them to F-CYCLE except on of the species ,I encounter AMG divergence with species.
I would be grateful if anyone can help me.
Anyway Daniel as i read in user guide , if you set Max Coarse
Levels to 0 [in Advanced solution control], the multigrid solver (AMG) would be turned off.
I hope it helps you
Thanks Rajesh and Javad for your concern and technical suggestions. I have tried some of the suggestions given but still same AMG solver problem. Guess i will change my mesh then. For more details and accessing my case file, please check the link below so you could advice me better.
https://www.dropbox.com/sh/nz25uxkth3wueac/AACpAEcjJMA9lgMskMPEBJ6Ja
Thank you
change the pressure discretisation to standard, it should not give this error, if it gives then examine your case closely because then the problem is with set up not with discretisation. Fluent is very stable solver.
- Badges
- Science topic