[Kina]

Hi all,

I have a quite simple setup. I want to investigate the flow past a flat plate (c = 0.1m, t = 1% of c). The Simulation should be steady state and density based. The flow is pressure driven and compressible with air as fluid. BCs are:

- pressure and temperature given at inlet
- pressure at outlet
- velocity pressureInletOutletVelocity (inlet) and zeroGradient (outlet)
- no slip walls at top and bottom

Re = ~50.000
Ma = 0.59

My reference calculation with fluent takes around 100 iterations (~5 min.) for a converged steady state solution. However, rhoSimpleFoam and rhoPimpleFoam with LTS (localEuler as time scheme) refuse to give me a converged solution. Instead, they are spending countless iterations on the slight vortex shedding at the end of the plate.

checkMesh output:

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).

Checking patch topology for multiply connected surfaces...
    Patch               Faces    Points   Surface topology                  
    INLET               220      442      ok (non-closed singly connected)  
    OUTLET              260      522      ok (non-closed singly connected)  
    TOP                 715      1432     ok (non-closed singly connected)  
    BOTTOM              715      1432     ok (non-closed singly connected)  
    PLATE               790      1580     ok (non-closed singly connected)  
    frontAndBackPlanes  490400   493100   ok (non-closed singly connected)  

Checking geometry...
    Overall domain bounding box (-1.19239e-05 -0.000141202 -0.0117087) (1.1003 0.400178 0.0117087)
    Mesh has 2 geometric (non-empty/wedge) directions (1 1 0)
    Mesh has 2 solution (non-empty) directions (1 1 0)
    All edges aligned with or perpendicular to non-empty directions.
    Boundary openness (-5.52111e-18 1.68288e-17 1.47372e-19) OK.
    Max cell openness = 3.32425e-16 OK.
    Max aspect ratio = 46.1598 OK.
    Minimum face area = 6.00025e-11. Maximum face area = 0.00016199.  Face area magnitudes OK.
    Min volume = 1.40511e-12. Max volume = 8.99034e-07.  Total volume = 0.0103017.  Cell volumes OK.
    Mesh non-orthogonality Max: 43.5127 average: 9.91558
    Non-orthogonality check OK.
    Face pyramids OK.
    Max skewness = 0.347688 OK.
    Coupled point location match (average 0) OK.

Mesh OK.

Mesh is generated by ICEM and converted with fluentMeshToFoam.
Pictures of the mesh attached.
Does anyone have a clue on how to get a nice convergence for problems like this with density based OpenFOAM? Schemes are all set to first order. I tried limiting the relaxation for rho to a very small value as that was a recommendation on the forums. Didn't work very well.

Cheers
Alex

[Artur.Ant]

It is difficult to give a reasonable answer.
From my experience wrong BC (or not suitable) (in OF you have more than 100 different BC) can slow down the solution.
Also fluent require less iteration to converge compared to OpenFoam (I did several comparison with spray, subsonic aerodynamics, supersonic nozzles)