[piotr.mecht]

I wanted to find the easiest way to simulate flow in the channel with one patch being specified as a constant heat flux or constant heat power source.

There is a BC called externalWallHeatFluxTemperature which provides varying temperature field according to required heat flux/power. It seems to work well, unless you turn turbulenece model on. It crashes on the first iteration with division by zero exception. I've tried to reduce relaxation factors to 0.1 and changed my schemes to upwind for max stability, but it didn't help at all. Is there any chance to make it work this way ?

Starting time loop

Time = 1

DILUPBiCGStab:  Solving for h, Initial residual = 0.986233, Final residual = 0.00220438, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.999392, Final residual = 0.0018794, No Iterations 1
GAMG:  Solving for p_rgh, Initial residual = 0.196606, Final residual = 0.00138496, No Iterations 9
GAMG:  Solving for p_rgh, Initial residual = 3.1198e-05, Final residual = 1.18706e-07, No Iterations 10
time step continuity errors : sum local = 0.00354071, global = -0.000519767, cumulative = -0.000519767
rho max/min : 997 978.866
DILUPBiCGStab:  Solving for epsilon, Initial residual = 0.112274, Final residual = 8.83977e-05, No Iterations 1
bounding epsilon, min: -0.00162638 max: 0.27328 average: 0.000906018
DILUPBiCGStab:  Solving for k, Initial residual = 1, Final residual = 8.26542e-05, No Iterations 1
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  ? in "/lib/x86_64-linux-gnu/libm.so.6"
#4  pow in "/lib/x86_64-linux-gnu/libm.so.6"
#5  Foam::compressible::alphatJayatillekeWallFunctionFvPatchScalarField::Psmooth(double) const at ??:?
#6  Foam::compressible::alphatJayatillekeWallFunctionFvPatchScalarField::updateCoeffs() at ??:?
#7  Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh>::Boundary::evaluate() in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/buoyantSimpleFoam"
#8  Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foam::CompressibleTurbulenceModel<Foam::fluidThermo> > >::correctNut() at ??:?
#9  Foam::RASModels::kEpsilon<Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foam::CompressibleTurbulenceModel<Foam::fluidThermo> > > >::correctNut() at ??:?
#10  Foam::RASModels::kEpsilon<Foam::EddyDiffusivity<Foam::ThermalDiffusivity<Foam::CompressibleTurbulenceModel<Foam::fluidThermo> > > >::correct() at ??:?
#11  ? in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/buoyantSimpleFoam"
#12  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#13  ? in "/opt/openfoam7/platforms/linux64GccDPInt32Opt/bin/buoyantSimpleFoam"

Temperature

Heater
    {
        type            externalWallHeatFluxTemperature;        
        mode            power;
        Q               300;
        value           uniform 298.15;
        kappaMethod     fluidThermo;      
    }

Turbulence model k-epsilon, Re~3000

[HPE]

any setup you can provide? so that the readers can try to replicate the issue?

[simrego]

Hi!


I have used this BC many times with turbulent flow without any problems... The problem should be something else not the BC itself..
Check your BCs related to the turbuelnce (alphat, nut, etc..) since your solver crashes there... But you can try to just simply set this BC to some fixedValue to see if the problem is still there...

[piotr.mecht]

Thanks for your reply


I pepared a version with some sharable, similiar geometry.



blockMesh + snappyHexMesh (any layers)

[HPE]

I don't use `groovyBC`, I'm afraid. So couldn't replicate the issue on the test case you provided.

[piotr.mecht]

Originally Posted by HPE

I don't use `groovyBC`, I'm afraid. So couldn't replicate the issue on the test case you provided.

This file set works without groovyBC (i generated non-uniform list, so it might not fit other mesh settings). I used groovyBC to provide non-uniform velocity field with ease, but my issue applies to uniform inlet velocity field too, so you can delete non-uniform list as well and set Inlet velocity to uniform (0.12 0 0), still crashes at the first time step.