[itsthaniyal]

I have been trying to simulate flow over an RAF airfoil. Using upwind scheme, I am able to achieve convergence in about 170 iterations when I initialize the field using potentialFoam. However when I change my schemes from div(phi,U) Gauss linear to div(phi,U) Gauss linearUpwind Gauss linear, the pressure equation residual oscillates around 0.4 like this:


I have been trying to figure out what has been causing this problem for a couple of days now, does anybody know what could be causing this? I have attached the case files below.
The mesh appears to be fine, checkMesh is perfectly happy.

Checking geometry...
This is a 2-D mesh
Overall domain bounding box (-4 -5 -7.20229e-159) (6 5 1)
Mesh (non-empty, non-wedge) directions (1 1 0)
Mesh (non-empty) directions (1 1 0)
Mesh (non-empty, non-wedge) dimensions 2
All edges aligned with or perpendicular to non-empty directions.
Boundary openness (2.20399e-17 -5.50998e-18 1.14217e-16) Threshold = 1e-06 OK.
Max cell openness = 2.47687e-16 OK.
Max aspect ratio = 647.241 OK.
Minumum face area = 1.92e-05. Maximum face area = 0.806645. Face area magnitudes OK.
Min volume = 1.92e-05. Max volume = 0.237668. Total volume = 99.8347. Cell volumes OK.
Mesh non-orthogonality Max: 50.6816 average: 5.66521 Threshold = 70
Non-orthogonality check OK.
Face pyramids OK.
Max skewness = 1.8656 OK.

Using GAMG as solver for pressure equation crashes the solver. I am using foam-extend-5.0 but have been able to reproduce the issue in OpenFOAM-2212.

[geth03]

First order upwind schemes are diffusive, so convergence is more likely. Second order upwind is more accurate and resolves the physiques better, if your case is not stationary at all, you can not expect a stationary solution.