twoPhaseEulerFoam drift flows in non-orthogonal mesh and dilute phase
 

Hello,

when using twoPhaseEulerFoam (2.4.0, but also at least 2.3.x) I've encountered mysterious drift flows, seemingly induced at wall boundary, where mesh cells are non-orthogonal / wall angle changes. These flows are encountered in the limiting case, when there is mostly just one phase present.

Please find attached example based on laminar bubbleColumn 2.4.0 tutorial case, which shows the formation of the drift flows. Non-orthogonal solver corrections do no help. So far only way I have been able to eliminate the extra flows is to use purely cubical cells in the mesh. Has anyone else encountered this? Please let me know if you find another solution!

FYI: Seems like the new twoPhaseEulerFoam in OpenFOAM-dev repository corrects this issue.

Hi,

for follow-up readers and just in case you haven't read the following links yet (I believe they're related):

http://www.cfd-online.com/Forums/ope...ho-meshes.html

http://www.openfoam.org/mantisbt/view.php?id=1379

Cutter