[MichG]

Dear Foamers,

I want to simulate a stirred tank (rotor-stator-system using a sliding mesh with a cyclicAMI BC). I use OpenFOAM 2.3.x and run the simulation in parallel on a cluster with 96 cores. For the simulation I use the kOmega-SST turbulence model. The checkMesh shows no errors and the quality of the mesh regarding orthogonality and skewness is okay.

 
Mesh stats
    points:           1423216
    internal points:  1254741
    faces:            11352709
    internal faces:   11015967
    cells:            5242115
    faces per cell:   4.26711
    boundary patches: 12
    point zones:      0
    face zones:       1
    cell zones:       3
Overall number of cells of each type:
    hexahedra:     0
    prisms:        1400186
    wedges:        0
    pyramids:      30
    tet wedges:    0
    tetrahedra:    3841899
    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: 2
    The mesh has multiple regions which are not connected by any face.
  <<Writing region information to "0/cellToRegion"
  <<Writing region 0 with 4084840 cells to cellSet region0
  <<Writing region 1 with 1157275 cells to cellSet region1
Checking patch topology for multiply connected surfaces...
    Patch               Faces    Points   Surface topology
    heiss               84322    42161    ok (closed singly connected)
    kalt                84548    42274    ok (closed singly connected)
    rotor               80940    40481    ok (non-closed singly connected)
    wand                70816    35545    ok (non-closed singly connected)
    welle               3774     1912     ok (non-closed singly connected)
    oben                4468     2385     ok (non-closed singly connected)
    ami_stator_mantel   1237     687      ok (non-closed singly connected)
    ami_stator_unten    801      436      ok (non-closed singly connected)
    ami_stator_oben     988      588      ok (non-closed singly connected)
    ami_rotor_oben      1661     924      ok (non-closed singly connected)
    ami_rotor_mantel    1736     935      ok (non-closed singly connected)
    ami_rotor_unten     1451     760      ok (non-closed singly connected)
Checking geometry...
    Overall domain bounding box (-9.08279e-18 -0.61 -0.61) (1.22 0.61 0.61)
    Mesh (non-empty, non-wedge) directions (1 1 1)
    Mesh (non-empty) directions (1 1 1)
    Boundary openness (9.98975e-17 -2.10661e-15 -5.01181e-16) OK.
    Max cell openness = 5.07085e-16 OK.
    Max aspect ratio = 9.82912 OK.
    Minimum face area = 3.50454e-07. Maximum face area = 0.00173417.  Face area magnitudes OK.
    Min volume = 1.18683e-10. Max volume = 2.12797e-05.  Total volume = 1.37666.  Cell volumes OK.
    Mesh non-orthogonality Max: 69.3312 average: 17.8113
    Non-orthogonality check OK.
    Face pyramids OK.
    Max skewness = 2.63992 OK.
    Coupled point location match (average 0) OK.
Mesh OK.

When I start the simulation, everything is fine. The pimple-loop converges within 12 iterations giving me physical senseful results. Unfortunately the time step continuity errors increase in every time step until the simulation fails after ~200 time steps while solving the Ux-equation. I added my BC and the most important dictionaries as well as a picture to see geometry.

I read many posts in the forum and therefore want to know:

- time step continuity errors: are my BC, fvSolution and fvOptions-file okay? I have read to "sqeeze" the p-Equation and that these errors occur using wrong BCs
- How to avoid those time step continuity errors? Does it have an effect?
- Is my velocity BC for the rotor okay (movingWallVelocity) ?
- Do i have to preserve the cyclicAMI-patches in the decomposePar?
- What else can I do to get a stable simulation?

Details are attached.

Kind regards
Michael

[MichG]

And of course the files.

From the log file:

 
Create time
　
Create mesh for time = 0
　
Selecting dynamicFvMesh solidBodyMotionFvMesh
Selecting solid-body motion function rotatingMotion
Applying solid body motion to cellZone region1
　
PIMPLE: max iterations = 50
field "(U|k|omega)" : relTol 0, tolerance 0.0001
field p : relTol 0, tolerance 0.001
　
Reading field p
　
Reading field U
　
Reading/calculating face flux field phi
　
AMI: Creating addressing and weights between 1237 source faces and 1736 target faces
AMI: Patch source sum(weights) min/max/average = 0.999836, 1.00065, 1.00026
AMI: Patch target sum(weights) min/max/average = 0.999852, 1.00068, 1.00019
AMI: Creating addressing and weights between 801 source faces and 1451 target faces
AMI: Patch source sum(weights) min/max/average = 0.986399, 1, 0.999513
AMI: Patch target sum(weights) min/max/average = 0.991773, 1, 0.999798
AMI: Creating addressing and weights between 988 source faces and 1661 target faces
AMI: Patch source sum(weights) min/max/average = 0.986501, 1, 0.999611
AMI: Patch target sum(weights) min/max/average = 0.990178, 1, 0.999789
Selecting incompressible transport model Newtonian
Selecting turbulence model type RASModel
Selecting RAS turbulence model kOmegaSST
kOmegaSSTCoeffs
{
alphaK1 0.85;
alphaK2 1;
alphaOmega1 0.5;
alphaOmega2 0.856;
gamma1 0.555556;
gamma2 0.44;
beta1 0.075;
beta2 0.0828;
betaStar 0.09;
a1 0.31;
b1 1;
c1 10;
F3 false;
}
　
Reading field C
　
Reading transportProperties
　
Reading/calculating face velocity Uf
　
No finite volume options present
　
Courant Number mean: 0 max: 0
　
Starting time loop
　
forces forces:
Not including porosity effects
　
Courant Number mean: 0 max: 0
Time = 0.001
　
solidBodyMotionFunctions::rotatingMotion::transformation(): Time = 0.001 transformation: ((0 0 0) (0.999956 (0.00941986 0 0)))
AMI: Creating addressing and weights between 1237 source faces and 1736 target faces
AMI: Patch source sum(weights) min/max/average = 0.99983, 1.0007, 1.00025
AMI: Patch target sum(weights) min/max/average = 0.999836, 1.00074, 1.00018
AMI: Creating addressing and weights between 801 source faces and 1451 target faces
AMI: Patch source sum(weights) min/max/average = 0.986731, 1, 0.999506
AMI: Patch target sum(weights) min/max/average = 0.990875, 1, 0.999794
AMI: Creating addressing and weights between 988 source faces and 1661 target faces
AMI: Patch source sum(weights) min/max/average = 0.969973, 1, 0.999087
AMI: Patch target sum(weights) min/max/average = 0.970237, 1, 0.99948
PIMPLE: iteration 1
DILUPBiCG: Solving for Ux, Initial residual = 1, Final residual = 0.0284447, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 1, Final residual = 0.0342313, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 1, Final residual = 0.0348894, No Iterations 1
GAMG: Solving for p, Initial residual = 1, Final residual = 0.00833246, No Iterations 8
GAMG: Solving for p, Initial residual = 0.836537, Final residual = 0.00428101, No Iterations 3
time step continuity errors : sum local = 1.05925e-05, global = -1.84807e-09, cumulative = -1.84807e-09
DILUPBiCG: Solving for omega, Initial residual = 0.00491671, Final residual = 5.84322e-05, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 1, Final residual = 0.00113274, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 2
DILUPBiCG: Solving for Ux, Initial residual = 0.396998, Final residual = 0.0212716, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.435977, Final residual = 0.0202887, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.433552, Final residual = 0.0198901, No Iterations 1
GAMG: Solving for p, Initial residual = 0.295205, Final residual = 0.00246142, No Iterations 6
GAMG: Solving for p, Initial residual = 0.405041, Final residual = 0.00205782, No Iterations 3
time step continuity errors : sum local = 5.09199e-06, global = -6.09731e-09, cumulative = -7.94538e-09
DILUPBiCG: Solving for omega, Initial residual = 0.00109439, Final residual = 6.77543e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.181064, Final residual = 0.000223387, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 3
DILUPBiCG: Solving for Ux, Initial residual = 0.210643, Final residual = 0.0112387, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.227965, Final residual = 0.0106337, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.224639, Final residual = 0.0103029, No Iterations 1
GAMG: Solving for p, Initial residual = 0.273154, Final residual = 0.00173302, No Iterations 5
GAMG: Solving for p, Initial residual = 0.171607, Final residual = 0.000936492, No Iterations 3
time step continuity errors : sum local = 1.92538e-06, global = -1.07954e-08, cumulative = -1.87408e-08
DILUPBiCG: Solving for omega, Initial residual = 0.000340442, Final residual = 3.48788e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.0710379, Final residual = 0.000123145, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 4
DILUPBiCG: Solving for Ux, Initial residual = 0.115488, Final residual = 0.00618736, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.123694, Final residual = 0.00589422, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.121163, Final residual = 0.00567863, No Iterations 1
GAMG: Solving for p, Initial residual = 0.286487, Final residual = 0.00150616, No Iterations 6
GAMG: Solving for p, Initial residual = 0.378669, Final residual = 0.00273378, No Iterations 3
time step continuity errors : sum local = 4.71711e-06, global = -1.17942e-08, cumulative = -3.0535e-08
DILUPBiCG: Solving for omega, Initial residual = 0.000122516, Final residual = 1.70868e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.0369097, Final residual = 7.32202e-05, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 5
DILUPBiCG: Solving for Ux, Initial residual = 0.0555343, Final residual = 0.00294634, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.0567241, Final residual = 0.00274847, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.0550717, Final residual = 0.00262941, No Iterations 1
GAMG: Solving for p, Initial residual = 0.271987, Final residual = 0.00193688, No Iterations 6
GAMG: Solving for p, Initial residual = 0.534424, Final residual = 0.00355508, No Iterations 3
time step continuity errors : sum local = 5.48211e-06, global = -1.134e-08, cumulative = -4.1875e-08
DILUPBiCG: Solving for omega, Initial residual = 5.57167e-05, Final residual = 1.42868e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.0217183, Final residual = 4.68592e-05, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 6
DILUPBiCG: Solving for Ux, Initial residual = 0.0192802, Final residual = 0.000938646, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.021321, Final residual = 0.00107651, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.0208575, Final residual = 0.00105196, No Iterations 1
GAMG: Solving for p, Initial residual = 0.241582, Final residual = 0.00190977, No Iterations 6
GAMG: Solving for p, Initial residual = 0.559528, Final residual = 0.00354837, No Iterations 3
time step continuity errors : sum local = 4.4691e-06, global = -9.18418e-09, cumulative = -5.10592e-08
DILUPBiCG: Solving for omega, Initial residual = 3.08845e-05, Final residual = 2.04467e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.0128353, Final residual = 3.08331e-05, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 7
DILUPBiCG: Solving for Ux, Initial residual = 0.0228589, Final residual = 0.00122298, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.0312112, Final residual = 0.00163726, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.0313295, Final residual = 0.00161643, No Iterations 1
GAMG: Solving for p, Initial residual = 0.217324, Final residual = 0.00159824, No Iterations 6
GAMG: Solving for p, Initial residual = 0.470618, Final residual = 0.0026836, No Iterations 3
time step continuity errors : sum local = 2.48011e-06, global = -6.14519e-09, cumulative = -5.72044e-08
DILUPBiCG: Solving for omega, Initial residual = 2.01208e-05, Final residual = 1.30397e-06, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.00791098, Final residual = 2.33801e-05, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 8
DILUPBiCG: Solving for Ux, Initial residual = 0.0358105, Final residual = 0.0021287, No Iterations 1
DILUPBiCG: Solving for Uy, Initial residual = 0.0448533, Final residual = 0.00235797, No Iterations 1
DILUPBiCG: Solving for Uz, Initial residual = 0.0450134, Final residual = 0.00232139, No Iterations 1
GAMG: Solving for p, Initial residual = 0.227131, Final residual = 0.00190435, No Iterations 5
GAMG: Solving for p, Initial residual = 0.248351, Final residual = 0.00159494, No Iterations 3
time step continuity errors : sum local = 1.24288e-06, global = -3.51663e-09, cumulative = -6.0721e-08
DILUPBiCG: Solving for omega, Initial residual = 1.4326e-05, Final residual = 4.90266e-07, No Iterations 1
DILUPBiCG: Solving for k, Initial residual = 0.00558223, Final residual = 2.04897e-05, No Iterations 1
DILUPBiCG: Solving for C, Initial residual = 0, Final residual = 0, No Iterations 0
PIMPLE: iteration 9

[MichG]

The Dict Files

[piu58]

Your log ist for time=0.001 where everything is or at least seems ok. It would be helpful to have a log snippet from later times, where the simulation gets slow.

[Moa_miche]

I attached the end of the log file with the last time step which was ok.

[piu58]

From your log

forces forces output:
    sum of forces:
        pressure : (1.45681e+12 6.68787e+10 4.42744e+10)
        viscous  : (2.2505e+07 1.02148e+08 -5.4285e+08)
        porous   : (0 0 0)
    sum of moments:
        pressure : (-1.34002e+08 -2.0655e+10 2.72405e+10)
        viscous  : (-1.2777e+07 2.25572e+08 4.14318e+07)
        porous   : (0 0 0)

Courant Number mean: 12685.4 max: 1.20778e+06

This step is far away from ok. Giant forces!

The Courant number is HUGE. That means your time step is magnitudes to high.

All i all I assume your case has general problems and is set up in a non physical manner. I recommend to write results at every time step and look what happens and at which boundary it starts to arise.

I possible, use Courant number control instead of a fixed time step.

[MichG]

In all the time steps before the simulation is diverging the Courant number is almost constant and quite okay

Time = 0.179
Courant Number mean: 0.040338 max: 26.9502

After the simulation diverges I of course get huge time steps

I tried it also with a smaller timestep (Co_max ~ 10), i thought this is no problem because im using the solver in Pimple-mode.

[piu58]

I don't recommend even a max Co of this size (27). Keep it lower than 1, if you have problems, even lower.

If large gradients arise during your simulation, and your max Co ist this large, it is easy to explain how your simulation crashes.

[MichG]

I adapted the divSchemes and now the simulation is running without problems. Due to my mesh (max nonOrhogonality ~ 70°) the linear scheme maybe caused the crash.

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

I changed the schemes as follow:

    div(phi,K)      Gauss upwind;
    div(meshPhi,p)  Gauss upwind;

Now the Simulation is running good with maxCo ~ 25. I hope this is not making my results worse and smearing my gradients / lowering the Forces.