View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0000054 | OpenFOAM | Bug | public | 2010-10-07 10:42 | 2010-10-07 17:58 |
| Reporter | Assigned To | henry | |||
| Priority | high | Severity | major | Reproducibility | always |
| Status | resolved | Resolution | fixed | ||
| Summary | 0000054: The buoyancy force term in the momentum equation of the buoyantBoussinesqPimpleFoam solver is incorrect. | ||||
| Description | In OF 1.6.x (buoyantBoussinesqPisoFoam) the buoyancy force term in Ueqn.h is UEqn == fvc::reconstruct ( ( fvc::interpolate(rhok)*(g & mesh.Sf()) <=== - fvc::snGrad(p)*mesh.magSf() ) ) This changed in OF 1.7.x (buoyandBoussinesqPimpleFoam) to UEqn == fvc::reconstruct ( ( - ghf*fvc::snGrad(rhok)*mesh.magSf() <=== - fvc::snGrad(p)*mesh.magSf() ) ) which is incorrect. The buoyancy force should be F=rhok*g instead of F~grad(T)*g.h | ||||
| Additional Information | In my calculations (Rayleigh-Benard configuration) the UZ residual practically does not drop with increasing number of pimple iterations. Changing back to the buoyancy force term from OF 1.6.x solves this problem. As far as I see, also the other solvers in heatTransfer are affected. | ||||
| Tags | Modelling, Solver | ||||
|
|
This is already fixed in OpenFOAM-1.7.x, the error is in using p rather than p_rgh. The correct form of the momentum predictor is solve ( UEqn == fvc::reconstruct ( ( - ghf*fvc::snGrad(rhok) - fvc::snGrad(p_rgh) )*mesh.magSf() ), mesh.solver(U.select(finalIter)) ); |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2010-10-07 10:42 |
|
New Issue | |
| 2010-10-07 10:52 | henry | Assigned To | => henry |
| 2010-10-07 10:52 | henry | Status | new => assigned |
| 2010-10-07 10:57 | henry | Note Added: 0000077 | |
| 2010-10-07 10:57 | henry | Status | assigned => resolved |
| 2010-10-07 10:57 | henry | Resolution | open => fixed |
| 2010-10-07 17:58 |
|
Tag Attached: Modelling | |
| 2010-10-07 17:58 |
|
Tag Attached: Solver |
