View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001441 | OpenFOAM | Bug | public | 2014-11-11 18:45 | 2014-12-28 16:51 |
| Reporter | projectionist | Assigned To | henry | ||
| Priority | high | Severity | major | Reproducibility | always |
| Status | closed | Resolution | no change required | ||
| Platform | Linux | OS | Ubuntu | OS Version | 14.04 |
| Summary | 0001441: twoPhaseEulerFoam-2.3.x breaks cases working under 2.3.0 | ||||
| Description | Simulating a case which worked perfectly well under OpenFOAM-2.3.0 with the most recent version of OpenFOAM yields unphysical results or even crashes the simulation. I strongly suspect the conservative formulation introduced in commit https://github.com/OpenFOAM/OpenFOAM-2.3.x/commit/abcc7428ac7ba5964160110ef655142ba6ec2d25 However, I didn't test this by reverting back to the state before and after the commit. Maybe, this is a simple matter of case setup. In the best case this problem is solved with different settings reflecting some consequences of the new formulation of the governing equations. In the worst case, twoPhaseEulerFoam isn't usable for simple stuff like the attached case. Please, have a look on this issue on comment on it. | ||||
| Steps To Reproduce | Run the attached case with OF-2.3.0 and a recent version of OF-2.3.x The reported behaviour was observed with Build : 2.3.x-590d57f32fed Build : 2.3.x-c854414e7a81 | ||||
| Additional Information | This behaviour is also described here: http://www.cfd-online.com/Forums/openfoam-solving/143429-twophaseeulerfoam-2-3-x-breaks-my-bubble-column.html | ||||
| Tags | No tags attached. | ||||
|
|
|
|
|
The convergence of the energy equations is more difficult when using the conservative form. I am experiencing some issue too, especially with cases involving MRF. Usually however convergence is achieved iterating over the system of equations, and using a sufficiently small time-step. You may want to set the number of outer correctors to 20 or so, and use a conservative value of the maximum Courant number (0.2 - 0.4). Also, set residualAlpha to 1.0e-4 everywhere. You are currently using 1.0e-6 for the momentum equation. I typically guarantee convergence by checking residuals (add residualControl to fvSolution/PIMPLE). |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2014-11-11 18:45 | projectionist | New Issue | |
| 2014-11-11 18:45 | projectionist | File Added: bubbleColumn.tar.gz | |
| 2014-11-11 21:12 | albertop | Note Added: 0003281 | |
| 2014-12-28 16:51 | henry | Status | new => closed |
| 2014-12-28 16:51 | henry | Assigned To | => henry |
| 2014-12-28 16:51 | henry | Resolution | open => no change required |
