[VIJAYA KUMAR]

Hi everyone

I am new to OpenFoam. I wanted to do 3D analysis of Jet Impingement on a flat plate using simpleFoam. The problem is my solver crashes after few iterations.

Please help me out !!

I am attaching my files for reference !!

i used ICEM_CFD unstructured meshing and imported in Open FOAM !!

My error message

#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::GAMGSolver::solveCoarsestLevel(Foam::Field<d ouble>&, Foam::Field<double> const&) const at ??:?
#6 Foam::GAMGSolver::Vcycle(Foam::PtrList<Foam::lduMa trix::smoother> const&, Foam::Field<double>&, Foam::Field<double> const&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::Field<double>&, Foam::PtrList<Foam::Field<double> >&, Foam::PtrList<Foam::Field<double> >&, unsigned char) const at ??:?
#7 Foam::GAMGSolver::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const at ??:?
#8 Foam::fvMatrix<double>::solveSegregated(Foam::dict ionary const&) at ??:?
#9 Foam::fvMatrix<double>::solve(Foam::dictionary const&) at ??:?
#10 Foam::fvMatrix<double>::solve() at ??:?
#11 ? at ??:?
#12 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#13 ? at ??:?
Floating point exception (core dumped)

[mprinkey]

It has been a while since I worked with openfoam, but that seems to be erroring out in the Preconditioned CG solver that is called to solve the coarsest AMG grid and that system is singular or otherwise produced a floating point exception...divide by zero, probably.

Does your domain have an outlet? This looks like the continuity equation (aka pressure (correction) equation) is singular likely because you have inconsistent boundary conditions set. If you have, say, an inlet for the jet and no outlets, your boundary conditions violate the global mass conservation condition. This violation can manifest in a pressure (correction) system that cannot be solved--because no pressure field will make the mass fluxes locally conservative if the overall system is non-conservative due to boundary conditions.

[VIJAYA KUMAR]

There is an outlet that I have named it as pressureoutletwall.
Is this a problem

herewith I am attaching my case.

log file is attached here

[vijy]

Hey vijay kumar ,My name is Vijay singh...I am also solving same problem as you....
and same error is occuring to me also...if you found any solution of the problem...please help me....

[nandhakumar]

can you explain your case setup briefly. It seems your your epsilon values are blowing up. How did you calculate your k and epsilon values for your model. The schemes used. try running in lower order schemes with low relaxation factors.

[vijy]

Thanks for the reply...@nandhakumar
I have attatched all the requied files includinlog files...I am using simpleFoam solver...
Renolds number is 35000..
formulas used are...
ε =(Cμ)^ 0.75*( k)^ 1.5/l
l=0.20D, where D=diameter of nozzle(40mm)
k = (U x ′ 2 + U y ′ 2 + U z ′ 2 )/2

I am eagerly waiting for your reply...

[nandhakumar]

Hello

I have read your boundary condition files.certain things are bit confusing. swirl flow rate inlet velocity i never used so i cannot comment on that.

Everything looks fine try reducing the tolerance value to 10e-5 or 10e-4 in fvSolution file and try changing the under relaxation values. If that doesnt work try calculating k and epsilon value from other formulas.

[Teosim]

Hi viji,
I am not confident with your case setup and OF version you are using but here are some thoughts.

How about the quality of your mesh? Could you please upload checkMesh log?

Noticed the simulation starts at time step 40: is that a "clean start" or is the solver resuming a previous simulation?

On the pressure BC: is fixedValue=0 exactly what you wanted for patch named "wall1" and "wall2"?

On the U BC: I never used swirlFlowRateInlet Velocity but is the velocity inlet vector (13 0 0) correctly oriented? (it might be just me but sometimes i overlook that)

You could try to increase relative tolerance for solvers (i.e. 0.01 for p and 0.001 or less for U etc.). You could also add a minIter entry (setting minimum number of iteration to 3 or 5)

Finally you could try starting the solver with turbulence switched off and turn it on after it stabilizes a little.

[JonW]

hi folks
if you are working with difficult mesh combined with difficult material model, you should not be shy in reducing the tolerance value.
In some cases, I am using tolerance between 0.1e-9 and 0.1e-11.
This is actually when I am using interFoam, with difficult grid and difficult material model. But the point is, when I started, I was very shy in reducing these tol values, and the solver always crashed. Now I am not, and everything is working perfectly.

If you are afraid that the solver is calculating something incorrect after a full run, make a case in which there exists an analytical solution for comparison.

If you are doing time dependent calculation, be careful on these relaxation factors. These can actually make your simulation results incorrect. Relaxation is actually best for time independent problems. Can be used for time-dependent problems as well, but just be careful (for example, run several runs with different relaxation factors to see if there are any difference).

cheers
J.

[ashishmagar600]

Hii

Please refer this post, you may find some direction to correct your case:
Time-step continuity error with diverging p and U

I had similar issues. Common mistakes are bad meshing and incorrect flow physics. If all is sure then there is a possibility of a crash in the mesh.

Regards,
Ashish