Negative diffusion !!
Hello,
I am running a case which involves diffusion. I am getting following error: Time = 4.68 smoothSolver: Solving for Ux, Initial residual = 7.20439e08, Final residual = 7.20439e08, No Iterations 0 smoothSolver: Solving for Uy, Initial residual = 6.83487e08, Final residual = 6.83487e08, No Iterations 0 smoothSolver: Solving for Uz, Initial residual = 1.9183e07, Final residual = 1.9183e07, No Iterations 0 DILUPBiCG: Solving for h, Initial residual = 9.49839e06, Final residual = 9.49839e06, No Iterations 0 GAMG: Solving for pd, Initial residual = 5.64058e08, Final residual = 3.39985e09, No Iterations 1 time step continuity errors : sum local = 4.40307e12, global = 2.87747e13, cumulative = 5.94087e05 rho max/min : 0.721684 .676527 DILUPBiCG: Solving for T, Initial residual = 6.34211e07, Final residual = 6.34211e07, No Iterations 0 #0 Foam::error::printStack(Foam::Ostream&) in "/usr/local/OpenFOAM/OpenFOAM1.5.x/lib/linux64GccDPOpt/libOpenFOAM.so" #1 Foam::sigFpe::sigFpeHandler(int) in "/usr/local/OpenFOAM/OpenFOAM1.5.x/lib/linux64GccDPOpt/libOpenFOAM.so" #2 ?? in "/lib/libc.so.6" #3 Foam::PBiCG::solve(Foam::Field<double>&, Foam::Field<double> const&, unsigned char) const in "/usr/local/OpenFOAM/OpenFOAM1.5.x/lib/linux64GccDPOpt/libOpenFOAM.so" #4 Foam::fvMatrix<double>::solve(Foam::Istream&) in "/usr/local/OpenFOAM/OpenFOAM1.5.x/lib/linux64GccDPOpt/libfiniteVolume.so" #5 Foam::fvMatrix<double>::solve() at /usr/local/OpenFOAM/OpenFOAM1.5.x/src/finiteVolume/lnInclude/fvMatrixSolve.C:187 #6 Foam::lduMatrix::solverPerformance Foam::solve<double>(Foam::tmp<Foam::fvMatrix<doubl e> > const&) at /usr/local/OpenFOAM/OpenFOAM1.5.x/src/finiteVolume/lnInclude/fvMatrix.C:1279 #7 main at ~/OpenFOAM/jagmohan1.5.x/applications/solvers/Scalar_Transport_combinations/buoyantTransportSimpleFoamNew/buoyantTransportSimpleFoamNew.C:85 #8 __libc_start_main in "/lib/libc.so.6" #9 _start at /build/buildd/glibc2.9/csu/../sysdeps/x86_64/elf/start.S:116 I checked my solver and at 85 line, it is scalar transport equation for diffusion which has problem. I did post processing and found that there are negative values for diffusion which is not possible. I think negative values arise due to numerical error while solving. But once the negative values occur, simulation diverges. At outlet, I am using zeroGradient condition, where it diverged first. Please tell me how to define diffusion term strictly non negative or appropriate boundary condition? I am attaching 3 pictures at different time steps which are explaining about this!! Regard, JM 
3 Attachment(s)
here are the attachments !

All times are GMT 4. The time now is 13:07. 