LES, Courant Number, Crash, Sudden

Alhasan · January 22, 2018, 17:07

Hey All,

I have gone through other threads with simlar questions about crashing it hasnt helped me hence asking again here

I have two LES simulation that are acting up with me, let me talk about one here, I am using pimpleFoam and I have set up time step such that to main a CFL value of 2 but suddenly it starts crashing

I have a time step of 4e-7 lower than which it will take way too long to complete the simulation

The simulation runs without any issues for 0.1638s after which it crashes how do i solve the issue any idea

I have used also tried using this runtime modifiable the simulation crashes much sooner with the time step going out of propotion.
runTimeModifiable yes;
maxDeltaT 1e-7;
maxCo 6;//0.9
adjustTimeStep yes;

still the simulation crashes if I keep the CFL under 1 and the simulation will take forever to complete. I dont mind having a high CFL number like 8-10 but I dont want the simulations to crash what do i do about this ?

check Mesh Results:

Code:

Mesh stats
    points:           7229736
    faces:            21118815
    internal faces:   20597481
    cells:            6945066
    faces per cell:   6.00661
    boundary patches: 6
    point zones:      0
    face zones:       1
    cell zones:       1

Overall number of cells of each type:
    hexahedra:     6911856
    prisms:        0
    wedges:        0
    pyramids:      0
    tet wedges:    0
    tetrahedra:    0
    polyhedra:     33210
    Breakdown of polyhedra by number of faces:
        faces   number of cells
            7   26865
            8   2106
            9   3186
           11   1053

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
                   CURV1    78240    79112  ok (non-closed singly connected)
                   INLET     5859     6104  ok (non-closed singly connected)
                     TOP    39690    41216  ok (non-closed singly connected)
                  OUTLET     5859     6104  ok (non-closed singly connected)
                     SYM   195843   197368  ok (non-closed singly connected)
                    SYM1   195843   197368  ok (non-closed singly connected)

Checking geometry...
    Overall domain bounding box (-0.5 -0.3 -1.38675e-17) (1.5 0.3 0.036)
    Mesh (non-empty, non-wedge) directions (1 1 1)
    Mesh (non-empty) directions (1 1 1)
    Boundary openness (4.44428e-17 2.79828e-17 1.34681e-14) OK.
    Max cell openness = 2.21707e-07 OK.
    Max aspect ratio = 231.527 OK.
    Minimum face area = 1.63904e-10. Maximum face area = 4.73102e-05.  Face area magnitudes OK.
    Min volume = 7.23718e-15. Max volume = 6.30802e-08.  Total volume = 0.0431456.  Cell volumes OK.
    Mesh non-orthogonality Max: 66.8039 average: 9.43428
    Non-orthogonality check OK.
    Face pyramids OK.
    Max skewness = 2.38317 OK.
    Coupled point location match (average 0) OK.

Mesh OK.

End

These are the place where it crashes: the K stays between -1.xxx and 75-80 for thousands of time steps until 0.1638 after which it just crashes

Code:

Courant Number mean: 0.00476427 max: 2.39386
Time = 0.16831

smoothSolver:  Solving for Ux, Initial residual = 6.17299e-06, Final residual = 2.37401e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66203e-05, Final residual = 1.96127e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000423954, Final residual = 7.96985e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.49493, Final residual = 0.00339957, No Iterations 4
time step continuity errors : sum local = 2.14037e-10, global = -1.12945e-13, cumulative = 8.27299e-10
GAMGPCG:  Solving for p, Initial residual = 0.436823, Final residual = 8.13999e-08, No Iterations 27
time step continuity errors : sum local = 5.1405e-15, global = 2.77939e-16, cumulative = 8.27299e-10
bounding k, min: -1.70026 max: 143.203 average: 0.201061
ExecutionTime = 13111.3 s  ClockTime = 13419 s

Courant Number mean: 0.00476427 max: 2.40022
Time = 0.1683104

smoothSolver:  Solving for Ux, Initial residual = 6.17539e-06, Final residual = 2.2862e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66265e-05, Final residual = 1.97553e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424722, Final residual = 7.98927e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.494715, Final residual = 0.00372712, No Iterations 4
time step continuity errors : sum local = 2.35334e-10, global = -7.43158e-13, cumulative = 8.26556e-10
GAMGPCG:  Solving for p, Initial residual = 0.437231, Final residual = 9.57638e-08, No Iterations 34
time step continuity errors : sum local = 6.05988e-15, global = 2.28986e-16, cumulative = 8.26556e-10
bounding k, min: -2.40413 max: 135.996 average: 0.201166
ExecutionTime = 13125.1 s  ClockTime = 13433 s

Courant Number mean: 0.00476427 max: 2.22604
Time = 0.1683108

smoothSolver:  Solving for Ux, Initial residual = 6.16541e-06, Final residual = 2.34176e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66186e-05, Final residual = 1.99606e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000425189, Final residual = 8.04212e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.494838, Final residual = 0.00405709, No Iterations 4
time step continuity errors : sum local = 2.5623e-10, global = -5.33397e-13, cumulative = 8.26023e-10
GAMGPCG:  Solving for p, Initial residual = 0.438004, Final residual = 9.12675e-08, No Iterations 31
time step continuity errors : sum local = 5.76987e-15, global = 1.11952e-15, cumulative = 8.26024e-10
bounding k, min: -1.69341 max: 146.946 average: 0.201293
ExecutionTime = 13138.4 s  ClockTime = 13446 s

Courant Number mean: 0.00476427 max: 2.31384
Time = 0.1683112

smoothSolver:  Solving for Ux, Initial residual = 6.16279e-06, Final residual = 2.38931e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66167e-05, Final residual = 1.95725e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000425232, Final residual = 8.02747e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.494657, Final residual = 0.00373473, No Iterations 4
time step continuity errors : sum local = 2.35479e-10, global = 2.51002e-14, cumulative = 8.26049e-10
GAMGPCG:  Solving for p, Initial residual = 0.437436, Final residual = 9.06846e-08, No Iterations 26
time step continuity errors : sum local = 5.72547e-15, global = -9.29442e-17, cumulative = 8.26049e-10
bounding k, min: -1.68987 max: 168.615 average: 0.20146
ExecutionTime = 13150.1 s  ClockTime = 13458 s

Courant Number mean: 0.00476427 max: 2.38364
Time = 0.1683116

smoothSolver:  Solving for Ux, Initial residual = 6.16212e-06, Final residual = 2.40648e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.6612e-05, Final residual = 1.9752e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.00042509, Final residual = 8.20654e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.493943, Final residual = 0.00313908, No Iterations 4
time step continuity errors : sum local = 1.97732e-10, global = 8.09449e-13, cumulative = 8.26858e-10
GAMGPCG:  Solving for p, Initial residual = 0.436808, Final residual = 9.37087e-08, No Iterations 24
time step continuity errors : sum local = 5.91186e-15, global = -3.492e-16, cumulative = 8.26858e-10
bounding k, min: -2.00964 max: 199.534 average: 0.201708
ExecutionTime = 13161.3 s  ClockTime = 13469 s

Courant Number mean: 0.00476427 max: 2.4211
Time = 0.168312

smoothSolver:  Solving for Ux, Initial residual = 6.15957e-06, Final residual = 2.41124e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66092e-05, Final residual = 2.04054e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424818, Final residual = 8.40124e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.492384, Final residual = 0.00329825, No Iterations 4
time step continuity errors : sum local = 2.07622e-10, global = 2.00169e-13, cumulative = 8.27058e-10
GAMGPCG:  Solving for p, Initial residual = 0.436385, Final residual = 7.85718e-08, No Iterations 29
time step continuity errors : sum local = 4.95646e-15, global = 2.0603e-16, cumulative = 8.27059e-10
bounding k, min: -1.68348 max: 277.497 average: 0.202139
ExecutionTime = 13173.4 s  ClockTime = 13482 s

Courant Number mean: 0.00476427 max: 2.6113
Time = 0.1683124

smoothSolver:  Solving for Ux, Initial residual = 6.15557e-06, Final residual = 2.49922e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66031e-05, Final residual = 2.07838e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424472, Final residual = 8.41308e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.492943, Final residual = 0.00398935, No Iterations 4
time step continuity errors : sum local = 2.51402e-10, global = 1.85571e-13, cumulative = 8.27244e-10
GAMGPCG:  Solving for p, Initial residual = 0.437938, Final residual = 8.82808e-08, No Iterations 27
time step continuity errors : sum local = 5.56824e-15, global = -2.41116e-16, cumulative = 8.27244e-10
bounding k, min: -1.6807 max: 452.143 average: 0.202836
ExecutionTime = 13185.5 s  ClockTime = 13494 s

Courant Number mean: 0.00476427 max: 2.90376
Time = 0.1683128

smoothSolver:  Solving for Ux, Initial residual = 6.16156e-06, Final residual = 2.51566e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66091e-05, Final residual = 2.05866e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424155, Final residual = 8.46042e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.494805, Final residual = 0.00412264, No Iterations 4
time step continuity errors : sum local = 2.60347e-10, global = 2.3208e-13, cumulative = 8.27476e-10
GAMGPCG:  Solving for p, Initial residual = 0.439378, Final residual = 9.88441e-08, No Iterations 26
time step continuity errors : sum local = 6.2528e-15, global = -1.20021e-16, cumulative = 8.27476e-10
bounding k, min: -1.67949 max: 681.507 average: 0.203954
ExecutionTime = 13197.5 s  ClockTime = 13506 s

Courant Number mean: 0.00476427 max: 3.30989
Time = 0.1683132

smoothSolver:  Solving for Ux, Initial residual = 6.17584e-06, Final residual = 2.41709e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66272e-05, Final residual = 2.03654e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424275, Final residual = 8.59482e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.498405, Final residual = 0.00364243, No Iterations 4
time step continuity errors : sum local = 2.31002e-10, global = -9.33976e-14, cumulative = 8.27382e-10
GAMGPCG:  Solving for p, Initial residual = 0.441554, Final residual = 8.01972e-08, No Iterations 28
time step continuity errors : sum local = 5.09744e-15, global = -4.27189e-16, cumulative = 8.27382e-10
bounding k, min: -1.67916 max: 998.119 average: 0.205817
ExecutionTime = 13210 s  ClockTime = 13519 s

Courant Number mean: 0.00476427 max: 3.86591
Time = 0.1683136

smoothSolver:  Solving for Ux, Initial residual = 6.18222e-06, Final residual = 2.35813e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66425e-05, Final residual = 2.1045e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000424955, Final residual = 8.9451e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.501903, Final residual = 0.00342728, No Iterations 4
time step continuity errors : sum local = 2.18593e-10, global = -2.33473e-13, cumulative = 8.27149e-10
GAMGPCG:  Solving for p, Initial residual = 0.444615, Final residual = 8.72048e-08, No Iterations 27
time step continuity errors : sum local = 5.58744e-15, global = -4.67401e-16, cumulative = 8.27148e-10
bounding k, min: -2.75056 max: 1607.42 average: 0.209344
ExecutionTime = 13221.8 s  ClockTime = 13531 s

Courant Number mean: 0.00476428 max: 4.63102
Time = 0.168314

smoothSolver:  Solving for Ux, Initial residual = 6.17363e-06, Final residual = 2.46097e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66389e-05, Final residual = 2.25103e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000425524, Final residual = 9.87208e-10, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.504331, Final residual = 0.00363937, No Iterations 4
time step continuity errors : sum local = 2.33834e-10, global = -7.90824e-13, cumulative = 8.26357e-10
GAMGPCG:  Solving for p, Initial residual = 0.449269, Final residual = 9.41598e-08, No Iterations 30
time step continuity errors : sum local = 6.07633e-15, global = -1.16056e-15, cumulative = 8.26356e-10
bounding k, min: -3.24369 max: 3210.82 average: 0.216819
ExecutionTime = 13234.2 s  ClockTime = 13544 s

Courant Number mean: 0.00476428 max: 5.70862
Time = 0.1683144

smoothSolver:  Solving for Ux, Initial residual = 6.16872e-06, Final residual = 2.60786e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66395e-05, Final residual = 2.48458e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000425701, Final residual = 1.22359e-09, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.508905, Final residual = 0.00352578, No Iterations 4
time step continuity errors : sum local = 2.28998e-10, global = -7.03264e-13, cumulative = 8.25653e-10
GAMGPCG:  Solving for p, Initial residual = 0.456481, Final residual = 8.32348e-08, No Iterations 24
time step continuity errors : sum local = 5.43514e-15, global = -8.80714e-16, cumulative = 8.25652e-10
bounding k, min: -5.62525 max: 7858.24 average: 0.233945
ExecutionTime = 13245.3 s  ClockTime = 13555 s

Courant Number mean: 0.00476428 max: 7.32406
Time = 0.1683148

smoothSolver:  Solving for Ux, Initial residual = 6.17624e-06, Final residual = 2.89695e-10, No Iterations 1
smoothSolver:  Solving for Uy, Initial residual = 1.66664e-05, Final residual = 3.23421e-10, No Iterations 1
smoothSolver:  Solving for Uz, Initial residual = 0.000425773, Final residual = 1.89709e-09, No Iterations 1
GAMGPCG:  Solving for p, Initial residual = 0.51658, Final residual = 0.00304282, No Iterations 4
time step continuity errors : sum local = 2.00963e-10, global = 1.67691e-14, cumulative = 8.25669e-10
GAMGPCG:  Solving for p, Initial residual = 0.466839, Final residual = 7.83318e-08, No Iterations 22
time step continuity errors : sum local = 5.21692e-15, global = 2.25184e-16, cumulative = 8.25669e-10
bounding k, min: -15.909 max: 18005.5 average: 0.275304
ExecutionTime = 13256.4 s  ClockTime = 13566 s

Code:

ddtSchemes
{
    default         backward;
}

gradSchemes
{
    default         cellMDLimited Gauss linear 0.5;
    grad(p)         cellMDLimited Gauss linear 0.5;
    grad(U)         cellMDLimited Gauss linear 0.5;
}

divSchemes
{
    default         none;
    div(phi,U)      Gauss linearUpwindV grad(U);
    div(phi,k)      Gauss limitedLinear 1;
    div(phi,omega)      Gauss limitedLinear 1;
    div(phi,B)      Gauss limitedLinear 1;
    div(phi,nuTilda) Gauss limitedLinear 1;
    div(B)          Gauss linear;
    div((nuEff*dev(T(grad(U))))) Gauss linear;
}

laplacianSchemes
{
    default          Gauss linear limited 1.0;
}

interpolationSchemes
{
    default         linear;
    interpolate(U)  linear;
}

snGradSchemes
{
    default          limited 1.0;
}

fluxRequired
{
    default         no;
    p               ;
}

Code:

solvers
{
    p
    {
        solver    PCG;//GAMG
        preconditioner
        {
            preconditioner    GAMG;
            smoother    GaussSeidel;
            nPreSweeps    0;
            nPostSweeps    2;
            nVcycles    2;
            cacheAgglomeration    true;
            nCellsInCoarsestLevel    50;
            agglomerator    faceAreaPair;
            mergeLevels    1;
            tolerance    1e-07;
            relTol    0.01;
        }
        maxIter          100;
        tolerance    1e-06;
        relTol    0.01;//0.05
    }

    pFinal
    {
        solver    PCG;
        preconditioner
        {
            preconditioner    GAMG;
            smoother    GaussSeidel;
            nPreSweeps    0;
            nPostSweeps    2;
            nVcycles    2;
            cacheAgglomeration    true;
            nCellsInCoarsestLevel    50;
            agglomerator    faceAreaPair;
            mergeLevels    1;
            tolerance    1e-07;
            relTol    0.01;
        }
       maxIter          100;
        tolerance    1e-07;
        relTol    0.0;
    }

    k
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-09;
        relTol          0.01;
    }
    kFinal
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-09;
        relTol          0;
    }

    omega
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-09;
        relTol          0.01;
    }
    omegaFinal
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-09;
        relTol          0;
    }
    U
    {
        solver      smoothSolver;
        smoother    DILUGaussSeidel;
        tolerance    1e-07;
        relTol    0.01;
    }
    UFinal
    {
        solver      smoothSolver;
        smoother    DILUGaussSeidel;
        tolerance    1e-07;
        relTol    0.0;
    }
    B
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-07;
        relTol          0;
    }

    nuTilda
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-07;
        relTol          0;
    }
    nuTildaFinal
    {
        solver          PBiCG;
        preconditioner  DILU;
        tolerance       1e-07;
        relTol          0;
    }
}

PISO
{
    nCorrectors     2;
    nNonOrthogonalCorrectors 0;
}
PIMPLE
{
    nOuterCorrectors 1;
    nCorrectors     2;
    nNonOrthogonalCorrectors 0;

    pRefCell        0;
    pRefValue       0;
}

relaxationFactors
{
    fields
    {
        "(p)"               0.3;
    }
    equations
    {
        "(U)"   0.7;
        "(UFinal)" 0.7;
    }
}

// ************************************************************************* //

so can one tell me how to run this simulation without crashing and essentially run a simulation with large courant number without crashing, some cases of mine are stable at even 8 CFL value but some time they crash at even 2, how do I work this can someone suggest some thing.

Kind Regards,
Hasan K.J

piu58 · January 23, 2018, 00:09

> Courant Number mean: 0.00476427 max: 2.39386

That may be to high. I have LES simulations which require a max Co of 0.1 or even lower. Sometimes a larger maxCo can be used, but the simulation runs faster with a lower one (less correction steps).

BlnPhoenix · January 23, 2018, 02:54

Your Courant Number is too high for LES. Even for implicit RANS something like CFL = 8 can be critical.

To be safe in LES you can use a DNS time step leading to CFL numbers <<1.

I also see that your not converging in each time step. SO watch out for that too.

calf.Z · October 31, 2019, 04:01

Quote:

Originally Posted by BlnPhoenix

Your Courant Number is too high for LES. Even for implicit RANS something like CFL = 8 can be critical.

To be safe in LES you can use a DNS time step leading to CFL numbers <<1.

I also see that your not converging in each time step. SO watch out for that too.

I am wondering how much Courant Number is acceptable for DNS simulation. The time step in my case is 3e-6 and I think it is very small. The current courant number is around 0.4.
Besides, I set the residual criteria of p for 1e-6 and u/h for 1e-7. I am wondering whether it is enough for DNS or not.

Thank you very much.

RKyle · November 21, 2019, 04:57

Probably too late to respond, but might be useful for future...

To use PIMPLE correctly you need to set "nOuterCorrectors" to something like 50 or 100. That's the maximum number of steady-state loops it will do before moving on in time, but will move on in time if it converges before then. If you have that set to 1, you're operating in PISO mode and so yes, a Co of more than 1 lead to issues.

Try setting nOuterCorrectors to 50, that might help. You'll then see many more iterations per time step, but as expected with PIMPLE (merged PISO-SIMPLE). Excellent explanation and demo of PIMPLE by T. Holzmann, Mathematics, Numerics, Derivations and OpenFOAM, 2018.. Link here.

cryabroad · November 22, 2019, 02:05

Does the "bounding k, xxx" message show up at the biginning of the simulation or from the time the simulation crashes? I doubt that it is there for the whole time, which suggests that something is wrong with your settings. Also, the residuals for pressure seem to not change at all, maybe the pressure boundary conditions need more caution?

I personlly find that setting maxCo is not as good as using a fixed time step (and carefully increase it if the courant number is below, say 1.0). For several cases I find that the max CFL number exceeds the maxCo I set so now I'm always using fixed time step for LES. By the way you obviously need more nOuterCorrectors.

January 23, 2018, 00:09		#2
piu58 Senior Member Uwe Pilz Join Date: Feb 2017 Location: Leipzig, Germany Posts: 744 Rep Power: 15	> Courant Number mean: 0.00476427 max: 2.39386 That may be to high. I have LES simulations which require a max Co of 0.1 or even lower. Sometimes a larger maxCo can be used, but the simulation runs faster with a lower one (less correction steps). faiazk likes this. __________________ Uwe Pilz -- Die der Hauptbewegung überlagerte Schwankungsbewegung ist in ihren Einzelheiten so hoffnungslos kompliziert, daß ihre theoretische Berechnung aussichtslos erscheint. (Hermann Schlichting, 1950)

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[snappyHexMesh] Error snappyhexmesh - Multiple outside loops	avinashjagdale	OpenFOAM Meshing & Mesh Conversion	53	March 8, 2019 09:42
Weird AMI Courant Number	Vyssion	OpenFOAM Running, Solving & CFD	3	April 13, 2016 02:31
Cluster ID's not contiguous in compute-nodes domain. ???	Shogan	FLUENT	1	May 28, 2014 15:03
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3	bookie56	OpenFOAM Installation	8	August 13, 2011 04:03
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues	michele	OpenFOAM Meshing & Mesh Conversion	2	July 15, 2005 04:15

January 23, 2018, 02:54		#3
BlnPhoenix Senior Member Join Date: Aug 2014 Location: Germany Posts: 292 Rep Power: 13	Your Courant Number is too high for LES. Even for implicit RANS something like CFL = 8 can be critical. To be safe in LES you can use a DNS time step leading to CFL numbers <<1. I also see that your not converging in each time step. SO watch out for that too.

November 21, 2019, 04:57		#5
RKyle New Member R Kyle Join Date: Feb 2017 Location: Scotland Posts: 11 Rep Power: 9	Probably too late to respond, but might be useful for future... To use PIMPLE correctly you need to set "nOuterCorrectors" to something like 50 or 100. That's the maximum number of steady-state loops it will do before moving on in time, but will move on in time if it converges before then. If you have that set to 1, you're operating in PISO mode and so yes, a Co of more than 1 lead to issues. Try setting nOuterCorrectors to 50, that might help. You'll then see many more iterations per time step, but as expected with PIMPLE (merged PISO-SIMPLE). Excellent explanation and demo of PIMPLE by T. Holzmann, Mathematics, Numerics, Derivations and OpenFOAM, 2018.. Link here.

November 22, 2019, 02:05		#6
cryabroad Senior Member Ruiyan Chen Join Date: Jul 2016 Location: Hangzhou, China Posts: 162 Rep Power: 9	Does the "bounding k, xxx" message show up at the biginning of the simulation or from the time the simulation crashes? I doubt that it is there for the whole time, which suggests that something is wrong with your settings. Also, the residuals for pressure seem to not change at all, maybe the pressure boundary conditions need more caution? I personlly find that setting maxCo is not as good as using a fixed time step (and carefully increase it if the courant number is below, say 1.0). For several cases I find that the max CFL number exceeds the maxCo I set so now I'm always using fixed time step for LES. By the way you obviously need more nOuterCorrectors.