|
[Sponsors] |
FGMRES diverge for SU2_7.0.7, for 3D incompressible flow |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
December 3, 2020, 01:35 |
FGMRES diverge for SU2_7.0.7, for 3D incompressible flow
|
#1 |
New Member
Bikalpa Bomjan Gurung
Join Date: Mar 2013
Posts: 12
Rep Power: 13 |
Hello every one,
I am running an In-compressible RANS model for a 3D wing at Re 10^5. And I am facing an error with linear solver divergence in version 7.0.7, and 7.0.8. Whereas it was running fine in 7.0.6. I have also attached Log output for both cases. Is there an additional setting I need to add to the configuration file for SU2_7.0.7? (note: I am using same configuration file and mesh for both the version ) Thank You Regards Bikalpa Bomjan Gurung |
|
December 3, 2020, 19:09 |
|
#2 |
Senior Member
Pedro Gomes
Join Date: Dec 2017
Posts: 466
Rep Power: 13 |
Hello,
The case was not running very well, the residual only drops one order of magnitude in 2000 iterations. You can try using a lower CFL, or allowing more linear solver iterations, or using the adaptive CFL. The change in 7.0.7 likely to cause the slightly different behaviour is this one: https://github.com/su2code/SU2/pull/1059 You can try adding the following lines to the config: PARMETIS_TOLERANCE= 0.05 PARMETIS_POINT_WEIGHT= 1 PARMETIS_EDGE_WEIGHT= 0 But like I said I think your case is not converging properly which causes it to be sensitive to minor numerical aspects, for example changing the number of cores. |
|
December 4, 2020, 08:58 |
|
#3 |
New Member
Bikalpa Bomjan Gurung
Join Date: Mar 2013
Posts: 12
Rep Power: 13 |
Hello Pedro,
Thank you for your suggestion, I applied it regarding Partitioning, but the solution still diverged. Also, I did perform simulation at CFL = 5, for Linear solver, max iteration 10 and 50. Both the case diverged and the solution for both the max iteration 10 and 50 were exactly the same with the following error message. (I have attached the output log ) Error Message: Error in "void CSysSolve<ScalarType>::ModGramSchmidt(int, C2DContainer<unsigned long, ScalarType, StorageType::RowMajor, 64UL, 0UL, 0UL> &, std::vector<CSysVector<ScalarType>, std::allocator<CSysVector<ScalarType>>> &) const [with ScalarType = double]": ------------------------------------------------------------------------- FGMRES orthogonalization failed, linear solver diverged Thank you Bikalpa |
|
Tags |
incompressible divergence, incompressible solver, su2_7.0.7 |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
compressible vs incompressible flow | Nick R | Main CFD Forum | 1 | March 15, 2017 12:29 |
starting compressible flow from incompressible flow results | mihaipruna | OpenFOAM Running, Solving & CFD | 2 | December 3, 2012 11:02 |
Reynold's number and incompressible flow | Neil | Main CFD Forum | 3 | October 24, 2006 09:11 |
Can 'shock waves' occur in viscous fluid flows? | diaw | Main CFD Forum | 104 | February 16, 2006 06:44 |
incompressible flow - prescribing pressure drop - how best to do it? | M. Gerritsen | Main CFD Forum | 4 | January 10, 1999 10:53 |