[MechaZ]

Hello everyone, i'm a student and kind of new to CFD. I properly ran a few 2D airfoil simulations using GMSH, OpenFOAM, then Paraview.

Now i'm trying to use ANSA in my internship to mesh, i'm having trouble exporting to OpenFOAM.
I'm getting floating point exceptions after 380 increments of simulation. I tried matching the exact boundary condition, and initial conditions i used previously in the working simulation but i think the generated polyMesh file by ANSA is missing something. I noticed the cellZones file is empty compared to the functioning one :

I don't know how to set those "cell zones" in ANSA and when i look at the first few increments of the simulation in Paraview it looks broken and the P, U values are completely off.
I'd be grateful if someone could help me figure out how to solve this issue.

*new*

FoamFile
{
	version 2.0;
	format ascii;
	class regIOobject;
	location "";
	object cellZones;
}
0
(
)

*old*

FoamFile
{
    format      ascii;
    class       regIOobject;
    location    "constant/polyMesh";
    object      cellZones;
}
1
(
Fluid
{
    type cellZone;
cellLabels      List<label> 
31595
(
0
1
2
3
.
.
*etc*
.
31592
31593
31594
)
;
}
)

the console log :

Time = 395s

smoothSolver:  Solving for Ux, Initial residual = 0.000900726, Final residual = 4.9224e-05, No Iterations 4
smoothSolver:  Solving for Uy, Initial residual = 0.00448011, Final residual = 0.000240954, No Iterations 4
GAMG:  Solving for p, Initial residual = 4.81564e-08, Final residual = 4.81564e-08, No Iterations 0
time step continuity errors : sum local = 321.603, global = 5.64817, cumulative = 484.134
#0  Foam::error::printStack(Foam::Ostream&) at ??:?
#1  Foam::sigFpe::sigHandler(int) at ??:?
#2  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#3  Foam::pow3(Foam::Field<double>&, Foam::UList<double> const&) at ??:?
#4  Foam::tmp<Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> > Foam::pow3<Foam::fvPatchField, Foam::volMesh>(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
#5  Foam::RASModels::SpalartAllmaras<Foam::incompressibleMomentumTransportModel>::fv1(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&) const at ??:?
#6  Foam::RASModels::SpalartAllmaras<Foam::incompressibleMomentumTransportModel>::correct() at ??:?
#7  ? in "/opt/openfoam11/platforms/linux64GccDPInt32Opt/bin/foamRun"
#8  ? in "/lib/x86_64-linux-gnu/libc.so.6"
#9  __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#10  ? in "/opt/openfoam11/platforms/linux64GccDPInt32Opt/bin/foamRun"
Floating point exception

note : I am using OpenFOAM 11 and ANSA exports to OF 8 version but that's in case you want ANSA to set up the case for you so that you just press run while selecting the solvers and parameters in the ANSA interface. I don't know if downgrading would solve this issue as i'm trying to export the mesh only.

[tomf]

Hi,

ANSA used to export a cellZones file even if all cells were in the same (and only) volume in ANSA. In more recent versions the export does not do it like that, it only exports cellZones if there are more than 1 solid pid in ANSA.

You should run checkMesh to see if the newly exported mesh has the same quality as the original one. Unless you have a setup where you run some constraints in this cellZone that stabilize the simulation, I would not expect any differences.

Regards,
Tom

[MechaZ]

Thank you for your answer Tom,

The case I was working on was using an Inlet/Outlet BC and it somehow didn't set the PIDs properly which was giving me floating point exceptions when I run the simulation, I now found another tutorial that uses the far field BC approach and apparently it's solving correctly and not giving me weird pressure and velocity values.

I thought that the export format of ANSA didn't work with the future versions of openFOAM at first, I don't know exactly what fixed the issue but as you said cellZones was not the issue here.

Regards,
Zack