[bragostin]

Dear OpenFoam users,

I have a simple forced convection + heat transfer case where a hotter jet is blowing into a fluid domain at room temperature. I use RANS + k-omega SST with buoyantBoussinesqPimpleFoam and I am not able to obtain convergence. I set gravity and thermal expansion coefficient to 0 because I have a highe inlet velocity (125 m/s).

However with pimpleFoam (so wthout the temperature field) I obtain convergence.

I am scratching my head on this for a few days now and I feel I have tried everything.

Any idea what I am doing wrong here?

Case in attachment.

[bragostin]

Even if I set the initial temperatures in °0/T" at 300 K everywhere, buoyantBoussinesqPimpleFoam will diverge while PimpleFoam will converge with the same "fvShemes* and "fVSolution* files.

Should not buoyantBoussinesqPimpleFoam and PimpleFoam behave similarly if there is a uniform temperature field, g=0 and thermal expension set to 0?

[Alczem]

Hey,


Try using Euler instead of backward for ddtSchemes. Also, are you sure your CFL is ok? You are using a fixed timestep, which is okay if you are confident about max velocities in your domain (btw, you can probably use a higher writePrecision and timePrecision, like 8 or 10, just to be sure).


Turn on gravity, it will help the solver, and since beta is set to 0, it should not change anything.


I would also ramp up velocity instead of applying 125m/s at t=0 (poor solver probably has a hard time managing this sudden change ). You can use uniformFixedValue for this. If you are ramping up the velocity, it will be smoother to compute. Adapt your turbulence quantities correspondingly, and I find that inletOutlet is better than zeroGradient for turbulente quantities.


As a last resort, you can also try to use 1st order schemes for everything and once your case is running, go back to 2nd order progressively.

[bragostin]

Hi,

I have tried everything you suggested:
- I would think CFL and time step are OK because the same case converges nicely with pimpleFoam
- using Euler instead of backward for ddtSchemes
- higher writePrecision and timePrecision
- Turn on gravity
- Use a very low velocity of 0.1 instead of 125 m/s (it should be equivalent to ramping-up for convergence purposes?)
- using inletOutlet instead of zeroGradient
- use 1st order schemes Gauss upwind

None of these improved convergence:

/*---------------------------------------------------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:                                        |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
Build  : OPENFOAM=1912 patch=200626
Arch   : "LSB;label=32;scalar=64"
Exec   : buoyantBoussinesqPimpleFoam
Date   : Mar 31 2024
Time   : 11:18:49
Host   : debian
PID    : 184002
I/O    : uncollated
Case   : /media/data_ntfs/OpenFoam/run/3D/Puffer_Knife/tests/MWE/buoyantBoussinesqPimpleFoam_NOK
nProcs : 1
trapFpe: Floating point exception trapping enabled (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster (fileModificationSkew 10)
allowSystemOperations : Allowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


PIMPLE: Operating solver in PISO mode

Reading thermophysical properties

Reading field T

Reading field p_rgh

Reading field U

Reading/calculating face flux field phi

Selecting incompressible transport model Newtonian
Creating turbulence model

Selecting turbulence model type RAS
Selecting RAS turbulence model kOmegaSST
Selecting patchDistMethod meshWave
RAS
{
    RASModel        kOmegaSST;
    turbulence      on;
    printCoeffs     on;
    alphaK1         0.85;
    alphaK2         1;
    alphaOmega1     0.5;
    alphaOmega2     0.856;
    gamma1          0.55555556;
    gamma2          0.44;
    beta1           0.075;
    beta2           0.0828;
    betaStar        0.09;
    a1              0.31;
    b1              1;
    c1              10;
    F3              false;
    decayControl    false;
    kInf            0;
    omegaInf        0;
}

Reading field alphat


Reading g

Reading hRef
Calculating field g.h

No MRF models present

Radiation model not active: radiationProperties not found
Selecting radiationModel none
No finite volume options present
Courant Number mean: 3.7781692e-10 max: 6.845495e-05

Starting time loop

Courant Number mean: 3.7781692e-10 max: 6.845495e-05
Time = 1e-06

PIMPLE: iteration 1
DILUPBiCG:  Solving for Ux, Initial residual = 1, Final residual = 8.7320927e-17, No Iterations 2
DILUPBiCG:  Solving for Uy, Initial residual = 1, Final residual = 6.3892476e-17, No Iterations 2
DILUPBiCG:  Solving for Uz, Initial residual = 1, Final residual = 5.2846807e-17, No Iterations 2
DILUPBiCG:  Solving for T, Initial residual = 0.99999984, Final residual = 1.845995e-16, No Iterations 2
GAMG:  Solving for p_rgh, Initial residual = 1, Final residual = 0.096629228, No Iterations 4
GAMG:  Solving for p_rgh, Initial residual = 0.00080300067, Final residual = 6.358011e-05, No Iterations 3
time step continuity errors : sum local = 5.7819839e-12, global = -4.3219316e-12, cumulative = -4.3219316e-12
GAMG:  Solving for p_rgh, Initial residual = 0.30627109, Final residual = 0.0092989315, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.0083253952, Final residual = 0.00044624563, No Iterations 2
time step continuity errors : sum local = 6.6185617e-11, global = 2.4934788e-11, cumulative = 2.0612857e-11
GAMG:  Solving for p_rgh, Initial residual = 0.95625552, Final residual = 0.024676226, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.011365063, Final residual = 8.1338688e-11, No Iterations 19
time step continuity errors : sum local = 2.2691104e-16, global = -2.0223419e-16, cumulative = 2.0612655e-11
DILUPBiCG:  Solving for omega, Initial residual = 0.10158353, Final residual = 2.7347121e-12, No Iterations 5
DILUPBiCG:  Solving for k, Initial residual = 1, Final residual = 1.3850753e-11, No Iterations 5
ExecutionTime = 0.97 s  ClockTime = 1 s

Courant Number mean: 1.837939e-06 max: 1.6308975
Time = 2e-06

PIMPLE: iteration 1
DILUPBiCG:  Solving for Ux, Initial residual = 0.48251246, Final residual = 1.4501455e-11, No Iterations 5
DILUPBiCG:  Solving for Uy, Initial residual = 0.0023942959, Final residual = 2.2868351e-11, No Iterations 4
DILUPBiCG:  Solving for Uz, Initial residual = 0.0085650537, Final residual = 1.0874003e-13, No Iterations 5
DILUPBiCG:  Solving for T, Initial residual = 0.00015218494, Final residual = 9.0926168e-14, No Iterations 4
GAMG:  Solving for p_rgh, Initial residual = 0.99998981, Final residual = 0.030212572, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.0084934592, Final residual = 0.0006179973, No Iterations 2
time step continuity errors : sum local = 0.00032030567, global = -0.00011928623, cumulative = -0.00011928621
GAMG:  Solving for p_rgh, Initial residual = 0.99975453, Final residual = 0.033627093, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.014235237, Final residual = 0.00089822233, No Iterations 2
time step continuity errors : sum local = 3.7835138, global = -0.057893112, cumulative = -0.058012398
GAMG:  Solving for p_rgh, Initial residual = 0.99996294, Final residual = 0.034551342, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.015784975, Final residual = 6.0138784e-11, No Iterations 20
time step continuity errors : sum local = 0.0049351801, global = -0.00063566976, cumulative = -0.058648068
DILUPBiCG:  Solving for omega, Initial residual = 0.99209426, Final residual = 2.8003419e-11, No Iterations 33
DILUPBiCG:  Solving for k, Initial residual = 0.99997385, Final residual = 2.7428404e-11, No Iterations 48
ExecutionTime = 1.33 s  ClockTime = 1 s

Courant Number mean: 71017419 max: 1.1749655e+14
Time = 3e-06

PIMPLE: iteration 1
DILUPBiCG:  Solving for Ux, Initial residual = 0.99999687, Final residual = 9.6173061e-11, No Iterations 47
DILUPBiCG:  Solving for Uy, Initial residual = 0.9999992, Final residual = 3.7729611e-11, No Iterations 48
DILUPBiCG:  Solving for Uz, Initial residual = 0.99999605, Final residual = 6.736675e-11, No Iterations 48
DILUPBiCG:  Solving for T, Initial residual = 0.99997834, Final residual = 8.1354606e-11, No Iterations 48
GAMG:  Solving for p_rgh, Initial residual = 0.9999979, Final residual = 0.036830371, No Iterations 2
GAMG:  Solving for p_rgh, Initial residual = 2.8044998e-06, Final residual = 1.6242056e-07, No Iterations 3
time step continuity errors : sum local = 47185.242, global = -30815.243, cumulative = -30815.302
GAMG:  Solving for p_rgh, Initial residual = 0.0067741062, Final residual = 0.00023273762, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 6.7500252e-08, Final residual = 4.6582018e-09, No Iterations 2
time step continuity errors : sum local = 4679360.7, global = 1221094.1, cumulative = 1190278.8
GAMG:  Solving for p_rgh, Initial residual = 0.014418351, Final residual = 0.00081069134, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 9.8338202e-09, Final residual = 7.119612e-11, No Iterations 5
time step continuity errors : sum local = 5.777278e+09, global = -3.9586639e+08, cumulative = -3.9467611e+08
DILUPBiCG:  Solving for omega, Initial residual = 0.96193547, Final residual = 6.0207772e-11, No Iterations 41
DILUPBiCG:  Solving for k, Initial residual = 0.75525706, Final residual = 5.7498303e-11, No Iterations 53
ExecutionTime = 1.83 s  ClockTime = 1 s

Courant Number mean: 3.7951163e+13 max: 9.6732841e+18
Time = 4e-06

Should'nt a case converging with pimpleFoam converge also with buoyantBoussinesqPimpleFoam?

[bragostin]

It was solved by upgrading from openfoam-1912 (Debian 12 stable package) to openfoam-2312.