View Issue Details
|ID||Project||Category||View Status||Date Submitted||Last Update|
|0002290||OpenFOAM||[All Projects] Bug||public||2016-10-11 09:21||2016-10-12 18:55|
|Priority||low||Severity||minor||Reproducibility||have not tried|
|Fixed in Version|
|Summary||0002290: tmp deallocation in twoPhaseEulerFoam and reactingTwoPhaseEulerFoam with implicitPhasePressure|
|Description||In twoPhaseEulerFoam if implicitPhasePressure is on and nAlphaCorr > 1, the solver crashes because alpha1alpha2f gets deallocated during the first iteration of solving alpha. Similar thing happens in reactingTwoPhaseEulerFoam, but instead of crashing, the implicit solution step is omitted in the later iterations because alphaDbyA is no longer valid.|
Is there a reason that nAlphaCorr shouldn't be > 1 with implicitPhasePressure or that the implicit step should be solved only once? If not, both problems can be fixed by using () after the terms in laplacian so that they don't get cleared.
|Tags||No tags attached.|
Thanks for the report and suggested fix. While the combination of implicitPhasePressure with nAlphaCorr > 1 is unlikely to be very useful there is no reason why it should not be possible.
Resolved in OpenFOAM-4.x by commit 1e03d68d4f4ebc698c702a5cba367fa28648c6b8
Resolved in OpenFOAM-dev by commit c12573c79967a50b7db2439f898f3a1e6b7eaf41