[PRATBHARAT]

Hi, I developed the solver using icoFoam with temperature gradient for the flow over multiple spheres. While I am running the test case. I am getting the following error/
Time = 1.1e-06

Courant Number mean: 0.0312314 max: 105.509
smoothSolver: Solving for Ux, Initial residual = 0.354868, Final residual = 1.0188e+222, No Iterations 1000
smoothSolver: Solving for Uy, Initial residual = 0.346559, Final residual = 6.00719e+221, No Iterations 1000
smoothSolver: Solving for Uz, Initial residual = 0.455163, Final residual = 6.6173e+221, No Iterations 1000
#0 Foam::error:rintStack(Foam::Ostream&) at ??:?
#1 Foam::sigFpe::sigHandler(int) at ??:?
#2 ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3 double Foam::sumProd<double>(Foam::UList<double> const&, Foam::UList<double> const&) at ??:?
#4 Foam::PCG::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#5 Foam::fvMatrix<double>::solveSegregated(Foam::dict ionary const&) at ??:?
#6 Foam::fvMatrix<double>::solve(Foam::dictionary const&) in "/home/cfdlab/OpenFOAM/cfdlab-3.0.1/platforms/linux64GccDPInt64Opt/bin/my_icoFoam"
#7 ? in "/home/cfdlab/OpenFOAM/cfdlab-3.0.1/platforms/linux64GccDPInt64Opt/bin/my_icoFoam"
#8 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#9 ? in "/home/cfdlab/OpenFOAM/cfdlab-3.0.1/platforms/linux64GccDPInt64Opt/bin/my_icoFoam"
Floating point exception (core dumped)


Kindly solve my issue. I am attaching the solver and test case too.
Thanks in advance.

[sagar_saroha]

Hi

The 'floating-point exception' error indicates a problem with mesh.

Run the following command and share the output from the log file:

checkMesh > meshReport.log

[PRATBHARAT]

Hi sagar_saroha
Thanks for replying...
Here, I am attaching the checkMesh report. Please, look in to it.

[CFD_10]

What about your time step deltaT?

[hxaxtma]

If you need help, please give us more information about your case. Next, provide your BCs for U and p. Pretty sure that your BCs are set wrong.

[PRATBHARAT]

object controlDict;


application my_icoFoam;

startFrom latestTime;

startTime 0;

stopAt endTime;

endTime 0.01;

deltaT 1e-7;

writeControl timeStep;

writeInterval 50;

purgeWrite 0;

writeFormat ascii;

writePrecision 6;

writeCompression off;

timeFormat general;

timePrecision 10;

runTimeModifiable yes;

adjustTimeStep yes;

maxCo 1.0;

maxDeltaT 0.1;




object U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions [0 1 -1 0 0 0 0];

internalField uniform (0 0 0);

boundaryField
{
inlet
{
type fixedValue;
value uniform (0.0002 0 0);
}
outlet
{
type zeroGradient;
value uniform (0 0 0);
}
side_walls
{
type fixedValue;
value uniform (0 0 0);
}
spheres
{
type fixedValue;
value uniform (0 0 0);
}



object p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions [0 2 -2 0 0 0 0];

internalField uniform 0;

boundaryField
{
inlet
{
type zeroGradient;
}
outlet
{
type fixedValue;
value uniform 0;
}
side_walls
{
type zeroGradient;
}
spheres
{
type zeroGradient;
}


Kindly, again look in to the controlDict and BCs.

[anon_q]

Could you just share your case?

[misospider]

This is most likely due to mesh and time step issue.

BCs are look ok.

Reduce time step and maxCo to less than 0.7, recommeding 0.5.

And save each time step until crash, check flow what is going wrong, especially at high Co>1.

If your setting are ok, this is problem of mesh quality, as you can see there are converted meshes (non-orthg.), if you can you need to mak mesh without that.

[sagar_saroha]

Originally Posted by PRATBHARAT

Hi sagar_saroha
Thanks for replying...
Here, I am attaching the checkMesh report. Please, look in to it.

Hi Pratyush

Please note that mesh non-orthogonality is high (>75) in 34 cases- please locate and try to correct these. [as also suggested by WJ].

[PRATBHARAT]

Hi sagar_saroha ,
I am new in CFD. Can you explain how can I correct these non orthogonality?

[sagar_saroha]

Hi Pratyush

Non-orthogonality is related to the angle of intersection of mesh lines. Using mesh elements of larger size (or using less element count) may lead to high non-orthogonality in case of unstructured grids.

Most of the commercial solvers have an 'inbuilt-mesh check' feature which highlights and suggests ways to repair mesh elements with non-orthogonality.
Otherwise, a crude way is using the hit-and-trial method by reducing element size in the suspected zones of bad quality mesh. A rough guess would be 'near sphere geometry' zone(s).

I would be able to comment better after seeing mesh snap-shot.

[PRATBHARAT]

https://drive.google.com/file/d/1Wjf...ew?usp=sharing


Here I am attaching my mesh file. Kindly look in to it and solve my problem.