[hughmorgan]

Hello CFD Online community,

I have a 19 injector nozzle with a converging-diverging nozzle and oxygen-methane combustion using rhoReactingFoam. rhoReactingFoam is breaking fairly soon after the simulation begins if I turn on either chemistry(laminar<rhoChemistryCombustion>) or turbulence(kOmega), but it runs forever with both of those turned off. My first thought was the mesh, but checkMesh says everything is fine. My trouble is determining what quantity is blowing up where and why.

I have looked at:

• slowly ramping up the inlet velocity (I don't like this because then I can't apply the inletOutlet BC)
• refining the mesh
• smaller ODE tolerances
• fireFoam and reactingFoam
• Having N2 be the inert gas in the chamber initially
• changed constant/reactions to the reactingFOAM example
• both OpenFOAM-2.4.0 and the latest release OpenFOAM-4.0
• pretty much everything I could find on CFD online including, but not limited to: Combustion modelling in OpenFOAM – Difficulties, Problem With reactingFOAM, reactingFoam boundary conditions.

All without success.

[7] #0  Foam::error::printStack(Foam::Ostream&) at ??:?
[7] #1  Foam::sigFpe::sigHandler(int) at ??:?
[7] #2  ? in "/opt/openfoam240/platforms/linuxGccDPOpt/lib/libOpenFOAM.so"
[7] #3  Foam::EulerImplicit<Foam::chemistryModel<Foam::rhoChemistryModel, Foam::sutherlandTransport<Foam::species::thermo<Foam::janafThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleEnthalpy> > > >::solve(Foam::Field<double>&, double&, double&, double&, double&) const at ??:?
[7] #4  double Foam::chemistryModel<Foam::rhoChemistryModel, Foam::sutherlandTransport<Foam::species::thermo<Foam::janafThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleEnthalpy> > >::solve<Foam::UniformField<double> >(Foam::UniformField<double> const&) at ??:?
[7] #5  Foam::chemistryModel<Foam::rhoChemistryModel, Foam::sutherlandTransport<Foam::species::thermo<Foam::janafThermo<Foam::perfectGas<Foam::specie> >, Foam::sensibleEnthalpy> > >::solve(double) at ??:?
[7] #6  Foam::combustionModels::laminar<Foam::combustionModels::rhoChemistryCombustion>::correct() at ??:?
[7] #7  ? at ??:?
[7] #8  __libc_start_main in "/lib/i386-linux-gnu/libc.so.6"
[7] #9  ? at ??:?

Are there any suggestions for debugging this or changing the model. I've noticed that bad boundary conditions can drastically affect the model, do I have any bad BCs? I've included a link to the tar-ed up files here. I appreciate any help! I've been stuck on this for about 6 months and it's driving me crazy!

[KarenRei]

Ah, the joys of simulating a reacting mixture

Two things I usually do to try to figure out why a model is dying of instability:

1) Resume from the last run before the crash with a miniscule write timestep so that you write out every last timestep that it's considering. Then step through the final timesteps in paraView looking at each of the parameters to see if you see any unusual oscillation or values. Often there's something that's escalating out of control somewhere. Those areas often need finer subdivision.

2) Play around with your fvSchemes, trying to choose a less oscillatory (if less accurate) set of parameters.