CFD Online Discussion Forums - simpleFoam for complicated geometry, non-orthogonal mesh, omega wall function

Hello everyone,
I am using simpleFoam on complicated geometry (non orthogonal mesh). Here is the checkMesh report:
--------------
Mesh stats
points: 6487723
faces: 19200050
internal faces: 18938278
cells: 6356388
boundary patches: 16
point zones: 0
face zones: 0
cell zones: 0

Overall number of cells of each type:
hexahedra: 6356388
prisms: 0
wedges: 0
pyramids: 0
tet wedges: 0
tetrahedra: 0
polyhedra: 0

Checking topology...
Boundary definition OK.
Cell to face addressing OK.
Point usage OK.
Upper triangular ordering OK.
Face vertices OK.
Number of regions: 1 (OK).
.
.
Mesh (non-empty, non-wedge) directions (1 1 1)
Mesh (non-empty) directions (1 1 1)
Boundary openness (9.64958e-17 2.34786e-15 -6.13025e-16) OK.
Max cell openness = 4.58304e-15 OK.
Max aspect ratio = 299.485 OK.
Minumum face area = 3.41806e-11. Maximum face area = 0.000249108. Face area magnitudes OK.
Min volume = 1.22828e-14. Max volume = 4.16394e-07. Total volume = 0.20005. Cell volumes OK.
Mesh non-orthogonality Max: 87.0638 average: 19.2898
*Number of severely non-orthogonal faces: 18931.
Non-orthogonality check OK.
<<Writing 18931 non-orthogonal faces to set nonOrthoFaces
Face pyramids OK.
Max skewness = 2.88371 OK.
Coupled point location match (average 0) OK.

Mesh OK.
-----------------
I am using SST turbulence model. I am using 'flowRateInletVelocity' for few inlets and it's mentioned like this:
INLET
{
type flowRateInletVelocity;
flowRate constant XXXX; // volume flow rate
value uniform (0 0 0);
}
The boundary conditions for k and omega are:
---
Feature k, omega
Turbulent inlet 0.4, 4000
Other inlets 1e-6, 1e-2

Walls type kqRWallFunction; type omegaWallFunction;
value uniform 0.4; value uniform 4000;

Outlet zeroGradient, zeroGradient
internal field 0.4, 4000
--------
I am using DICPCG for pressure, smoothSolver for omega and DILUPBiCG for others.

My fvSchemes were as follows:
-----
ddtSchemes
{
default steadyState;
}

gradSchemes
{
default Gauss linear;
}

divSchemes
{
default none;
div(phi,U) Gauss linearUpwindV grad(U);
div(phi,k) Gauss upwind;
div(phi,omega) Gauss upwind;
div((nuEff*dev(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
default Gauss linear corrected;
}

interpolationSchemes
{
default linear;
}

snGradSchemes
{
default corrected;
}

fluxRequired
{
default no;
p;
}
----
The simulation diverges after 150 steps. I tried Gauss linear limited 0.333 for laplacian schemes which didn't work. gradSchemes was changed to 'cellLimited Gauss linear 1' and it was not good either.

Could someone give me tips to improve the simulation?

Note: When i use epsilonWallFunction(as wall boundary condition) for omega, it converges to some decent results (not accurate though) with the 1st set of fvSchemes. It's strange!!

Kind regards,
Achinta