[lcj66]

Hi,

I am pretty new to OpenFOAM and I am attempting to run a membrane mass transfer solver that I obtained from a git repository.

I am getting this error:
#0 Foam::error:rintStack(Foam::Ostream&) at ??:?
#1 Foam::sigFpe::sigHandler(int) at ??:?
#2 in "/lib/x86_64-linux-gnu/libc.so.6"
#3 Foam::explicitFOmembraneVelocityFvPatchVectorField ::updateCoeffs() at ??:?
#4 Foam::GeometricField<Foam::Vector<double>, Foam::fvPatchField, Foam::volMesh>::GeometricBoundaryField::updateCoef fs() at ??:?
#5 Foam::fvMatrix<Foam::Vector<double> >::fvMatrix(Foam::GeometricField<Foam::Vector<doub le>, Foam::fvPatchField, Foam::volMesh> const&, Foam::dimensionSet const&) at ??:?
#6 Foam::fv::gaussLaplacianScheme<Foam::Vector<double >, double>::fvmLaplacianUncorrected(Foam::GeometricFi eld<double, Foam::fvsPatchField, Foam::surfaceMesh> const&, Foam::GeometricField<double, Foam::fvsPatchField, Foam::surfaceMesh> const&, Foam::GeometricField<Foam::Vector<double>, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
#7 Foam::fv::gaussLaplacianScheme<Foam::Vector<double >, double>::fvmLaplacian(Foam::GeometricField<double, Foam::fvsPatchField, Foam::surfaceMesh> const&, Foam::GeometricField<Foam::Vector<double>, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
#8 Foam::fv::laplacianScheme<Foam::Vector<double>, double>::fvmLaplacian(Foam::GeometricField<double, Foam::fvPatchField, Foam::volMesh> const&, Foam::GeometricField<Foam::Vector<double>, Foam::fvPatchField, Foam::volMesh> const&) at ??:?
#9
at ??:?
#10
at ??:?
#11
at ??:?
#12 __libc_start_main in "/lib/x86_64-linux-gnu/libc.so.6"
#13
at ??:?
Floating point exception (core dumped)

from what I have read on the forums, this often happens in using the turbulence model when k or epsilon is set to 0. however, I'm not using a turbulence model. I think it is an issue in the velocity BC that is affecting the gaussLaplacianScheme, but not sure.

Was wondering where is a good point to go about looking for errors, and maybe someone knows what variable or parameter I may have set to 0 or some other bad value.

Thanks!!
Lori

[alexeym]

Hi,

Look at line #3:

#3 Foam::explicitFOmembraneVelocityFvPatchVectorField::updateCoeffs() at ??:?

The error happens in updateCoeffs of explicitFOmembraneVelocityFvPatchVectorField. So, yeah, the error has nothing to do with turbulence model. Why it is happening there? I do not know, show your source code, show your case.

[lcj66]

Hi Alexey,

Thank you for taking the time to reply!

I have attached the boundary conditions (which includes the explicitFOvelocity BC where the problem is occuring. I compile the BCs into a separate library which I call in the controlDict file.

I was going to attach the solver and the case files, but they are too big of attachements.

Let me know if you get time to look at them, and what you think. If I should attach something else let me know.

Thank you so much!
Lori

[alexeym]

Hi,

Though you can follow the code yourself (and substitute values, to check if FPE happened there). Just to show and examples of points of failure (all excerpts are from explicitFOmembraneVelocityFvPatchVectorField.C):

1. Line 240

operator[](facei) = vfnf[facei] * (A_*(ppsf[facei]-ppsf[fm_[facei]])) / rho

What is the value of rho?

2. Line 293

v *= (1.0 + rho_mACoeff_.value() * feedMem) / (1.0 + rho_mACoeff_.value() * drawMem)

What is the value of denominator?

3. Lines 428-437

        scalar numerator    = A()*pi_mACoeff().value()*drawm_A + B();
        scalar denominator  = A()*pi_mACoeff().value()*feedm_A + Jvalue + B();
        
        // To avoid floating point exceptions
        if( denominator > SMALL ){
            return Jvalue - ( 1/K() ) * log( numerator / denominator );
        }
        else{
            return 0;
        }

Can numerator be equal to 0? Can ratio be negative?

Since I do not know neither the nature of your solver, nor the case you are trying to run, I can not answer above questions.

[lcj66]

So, the following outputs before the error:

/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.3.1 |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.3.1-bcfaaa7b8660
Exec : pisoSaltTransport
Date : Jul 02 2015
Time : 13:56:40
Host : "Inferno"
PID : 3150
Case : /home/lori/OpenFOAM/lori-2.3.1/run/reTestCase
nProcs : 1
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Allowing user-supplied system call operations

// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time

Create mesh for time = 0


Reading g
Reading field p

Reading field m_A

Reading field U

Membrane_master: forward direction specified by user

Membrane_slave: forward direction specified by user

Reading transportProperties

Creating rho

Reading/calculating face flux field phi

Flow is assumed to be laminar
Turbulence is currently unsupported in this solver

Calculating field (g.h)f

Courant Number mean: 2.539428089 max: 20.28108117
m_A: Initial mass = 3.491218911e-05


Starting time loop

Courant Number mean: 8.61067278e-05 max: 0.0007020358824
deltaT = 2.070895363e-06
Time = 2.070895363e-06

Membrane_master: Ridders' Method - Total iterations = 100
Water flux, advanced: 6.229807669 kg/(h*m2)
Draw/Feed m_A estimate: 0.05150560985 / 3.279628002e-05
Max Slip Velocity: (0 0 0) with slip Coeff: 1 and under-relax factor: 1
A: 1e-12 / B: 1e-07 / K: 500000


So the Jw term (the flux) is being calculated I'm pretty sure, so it doesn't seem like the numerator or denominator are giving errors.

I also set rho to initial values given from a potential solver.

My case files are too large to attach here, so I uploaded it here:
https://www.mediafire.com/?8i3b9us6gbd243p

If you have time to take a look, it would be greatly appreciated. I'm just totally at a loss of what may be the problem.

Thanks!
Lori

[alexeym]

Hi,

so the error appears after

A: 1e-12 / B: 1e-07 / K: 500000

line?

No, I do not want to go through all these banners, popup windows, and other stuff. Use normal sites for case files; for example, bitbucket.org, github.com will not flood me with advertisement shit.