[mighelone]

Dear all,

i'm trying to solve a common test-case for gas combustion (the CH4 bluff-body flame of Sandia).

I'm considering a skeletal mechanism with 16 species and 41 reactions (but I've obtained the same errors using a simple one-step mechanism) for CH4 combustion. I've solved the cold flow, and I've patched the initial temperature field, using setField creating an ignition zone near the bluff body with a temperature of 1000 K.

After some iterations I've obtained the following error message:

DILUPBiCG:  Solving for h, Initial residual = 0.993298, Final residual = 4.85856e-07, No Iterations 58
[0] 
[0] 
[0] attempt to use janafThermo<equationOfState> out of temperature range 200 -> 3000;  T = 170.194#0  Foam::error::printStack(Foam::Ostream&) in "/home/michele/OpenFOAM/OpenFOAM-1.6/lib/linux64GccDPOpt/libOpenFOAM.so"
#1  Foam::error::abort() in "/home/michele/OpenFOAM/OpenFOAM-1.6/lib/linux64GccDPOpt/libOpenFOAM.so"
#2  Foam::specieThermo<Foam::janafThermo<Foam::perfectGas> >::H(double) const in "/home/michele/OpenFOAM/OpenFOAM-1.6/lib/linux64GccDPOpt/libreactionThermophysicalModels.so"
#3  Foam::hPsiMixtureThermo<Foam::reactingMixture<Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas> > > > >::calculate() in "/home/michele/OpenFOAM/OpenFOAM-1.6/lib/linux64GccDPOpt/libreactionThermophysicalModels.so"
#4  Foam::hPsiMixtureThermo<Foam::reactingMixture<Foam::sutherlandTransport<Foam::specieThermo<Foam::janafThermo<Foam::perfectGas> > > > >::correct() in "/home/michele/OpenFOAM/OpenFOAM-1.6/lib/linux64GccDPOpt/libreactionThermophysicalModels.so"
#5  main in "/home/michele/OpenFOAM/OpenFOAM-1.6/applications/bin/linux64GccDPOpt/reactingFoam"

It seems an error calculating the temperature field from enthalpy.

Any idea how to solve this problem?

I'm starting from the reactingFoam tutorials to define chemistry dictionary.

Thanks in advance for the attention

Michele

[mighelone]

No ideas how to solve this problem?

I'm trying to modify some simulation parameters with no success, like for example the time step and others.

Thank you for the attention

Michele

[roger.jove]

Hi Michele,

I am sorry but going out of temperature range is a quite common problem. I have found this problem with different solvers that use janafThermo.

Sometimes I have solved it using a finer mesh. I have tried other options that I found in the link below but they didn't worked for me (I am still a newbie!). However, take a look, it will be provably interesting for you.


Roger


http://www.cfd-online.com/Forums/ope...naf-range.html

[mighelone]

Hi Roger,

thank you for your suggestion, in the topic I've found some interesting suggestions to solve the problem.

I guess the CHEMKIN chemistry solver is not enough robust to work at low temperature range, that could be found during the ignition of combustion problem.

Probably considering another chemistry solver (http://openfoamwiki.net/index.php/Co...ateReactinFoam) the problem can be solved.

Michele

[dohnie]

Mighelone,
can you locate where in your field this temperature occurs?
A non-suitable boundary condition (reaction close to a BC interfers with BC) is a frequent error source.
Moreover, a too high Courant Number in reactingFoam can also give problems with the reaction.

[mighelone]

I'm simulating a bluff body flame. The flame is developed downstream of the bluff body. I have modeled a little part of the duct upstream, therefore the flame is not attached to the inlet boundaries.

I've tried to modifiy the Courant number, reducing it until 0.01, but without success. On the contrary the Janaf error seems to appear before reducing the Courant number.

Tomorrow I'm trying to modify the mesh, reducing the size and improving the quality.

Michele