[Edoardo1993]

Hello,
I am trying to simulate a flammable gas flow in atmosphere with reactingFoam. You can find the whole case directory attached at the end.

If I do not ignite the gas (/constant/combustionProperties --> active off and /constant/chemistryProperties --> chemistry off) the gas leak is totally diffusive and results are good.
However, if I try to ignite the flammable gas (/constant/combustionProperties --> active on and /constant/chemistryProperties --> chemistry on) the reaction does not occur. Therefore the results are exactly the same as withouth ignition.

The reaction is a single-step reaction, implemented in /constant/reactionGRI.

reactionGRI

elements
(
)
;

species
4
(
H2
O2
H2O
N2
)
;


reactions
{
    un-named-reaction-2
    {
        type            irreversibleinfiniteReaction; 
        reaction        "2H2 + O2 + 3.79N2 = 2H2O + 3.79N2";
    }
}

chemistryProperties

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v2012                                 |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      chemistryProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

chemistryType
{
    chemistrySolver   ode;
    chemistryThermo   psi;
    TDAC              off;
}

initialChemicalTimeStep 1e-07;

chemistry       on;

odeCoeffs
{
    solver          seulex;  // Rosenbrock34, seulex or rodas23
    absTol          1e-12;
    relTol           1e-7;
}

// ************************************************************************* //

combustionProperties

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v2012                                 |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      combustionProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

combustionModel  EDC;

active  on;

EDCCoeffs
{
    version v2005;
}

// ************************************************************************* //

thermophysicalProperties

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v2012                                 |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      thermophysicalProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

thermoType
{

    type            hePsiThermo;
    mixture        reactingMixture; //singleStepReactingMixture;
    transport       sutherland;
    thermo          janaf;
    energy          sensibleEnthalpy;
    equationOfState perfectGas;
    specie          specie;
}

inertSpecie N2;

chemistryReader foamChemistryReader;
foamChemistryFile "<constant>/reactionsGRI";
foamChemistryThermoFile "<constant>/thermo.compressibleGasGRI";

dpdt false;

 // ************************************************************************* //

I have the following question:


1) Why does the reaction not occur?

2) Is irreversibleinfiniteReaction in /constant/reactionGRI correct? Maybe is it not fast enough to show something in a short time?

3) In /constant/thermophysicalProperties should be selected mixture reactingMixture or singleStepReactingMixture?

4) Could be the problem the lack of ignition?




Really thank for your help.

[kerim]

Try to use another turbulence model. For example RNG instead of standard k-e-model.
Please see this post


reactingFoam SandiaD_LTS tutorial error.
May be it will help you.
Good luck!

[Edoardo1993]

Originally Posted by kerim

Try to use another turbulence model. For example RNG instead of standard k-e-model.
Please see this post


reactingFoam SandiaD_LTS tutorial error.
May be it will help you.
Good luck!

@kerim thank you for your answer. I followed your tips and switched standard k-eps model to RNG k-eps model in turbulenceProperties... but the results do not show any combustion, only diffusion.

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  v2012                                 |
|   \\  /    A nd           | Website:  www.openfoam.com                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      turbulenceProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

simulationType   RAS;

RAS
{
    RASModel        RNGkEpsilon;

    kEpsilonCoeffs
    {
        Prt 1;        
    }

    turbulence      on;
    printCoeffs     on;
}

// ************************************************************************* //

[tas38]

Edoardo1993,

Per your original post, I would suggest changing the reaction in reactionsGRI from irreversibleinfiniteReaction to irreversibleArrheniusReaction. You can then adjust the Arrhenius rate constants and activation temperature until you get a reaction. If ignition/combustion still does not occur, re-running with alternative RAS models per Kerim is also advised.

Notably, irreversibleArrheniusReaction is only used in conjunction with LES turbulence in the combustion tutorials.

[Edoardo1993]

Originally Posted by tas38

Edoardo1993,

Per your original post, I would suggest changing the reaction in reactionsGRI from irreversibleinfiniteReaction to irreversibleArrheniusReaction. You can then adjust the Arrhenius rate constants and activation temperature until you get a reaction. If ignition/combustion still does not occur, re-running with alternative RAS models per Kerim is also advised.

Notably, irreversibleArrheniusReaction is only used in conjunction with LES turbulence in the combustion tutorials.

Thank you tas38. I did not know about LES and irreversibleinfiniteReaction.


Now, my reaction file is:

reactions
{
    un-named-reaction-2
    {
        type         irreversibleArrheniusReaction;
        reaction        "2H2 + O2 + 3.79N2 = 2H2O + 3.79N2";
        A        5.2e16;
        beta     0;
        Ta       14906;
    }
}

Reaction does not occur yet, with both standard k-eps and RNG k-eps models. I think the problem is the lack of ignition, because there are not hot spots in the domain.

[tas38]

Edoardo1993,

Yes, you could try adding an ignition source (small box with high temperature) using setFieldsDict containing a boxToCell region as in DLR_A_LTS.

However, if I recall, the EDC model should not require a direct ignition source. Combustion is initiated simply due the presence of turbulence, i.e. and .

I just noticed in your combustionProperties file you have active on;, however, I believe this should be active true;. Not sure if no vs. true makes a difference, but I would try this first.

[Edoardo1993]

Originally Posted by tas38

Edoardo1993,
Yes, you could try adding an ignition source (small box with high temperature) using setFieldsDict containing a boxToCell region as in DLR_A_LTS.
However, if I recall, the EDC model should not require a direct ignition source. Combustion is initiated simply due the presence of turbulence, i.e. and .
I just noticed in your combustionProperties file you have active on;, however, I believe this should be active true;. Not sure if no vs. true makes a difference, but I would try this first.

tas38, thank you for your kind help.
I knew that active true and on are the same keyword.
Regarding ignition with EDC model, I created another directory for test and the combustion starts only when the flammable flow touches an hot spot, in my case a wall with T=1000K (probably you can get similar result with setFieldDict). I don't see any reaction to occur when this wall is cold.


What do you mean with "Combustion is initiated simply due the presence of turbulence, i.e. and "? The greater k and epsilon, the faster the reaction?

[Edoardo1993]

Hello. I hope someone can help me because I am still stucked with this issue.
I finally managed to ignite the gas with fvOptions setting some cells at 1200K. The problem is the simulation is running only when using CH4.
As soon as I switch the gas to H2 (changing the reaction with "H2 + 2O2 = 2H2O" and the thermo.compressibleGas ), the simulation starts but it stops because of several errors..

elements
(
O
C
H
N
);


species
(
    O2
    H2O
    CH4
    CO2
    N2
);

reactions
{
    methaneReaction
    {
        type     irreversibleArrheniusReaction;
        reaction "CH4 + 2O2 = CO2 + 2H2O";
        A        5.2e16;
        beta     0;
        Ta       14906;
    }
 }

Furthermore, I have another question. I attempted to integrate the GRI reactions, which I found in the SANDIAD_LTS tutorial, into the working model with CH4. Why does the reaction occur with the single-step reaction but not with the GRIMECH file?"

[Han Yu]

Hello. I want to simulate a combustion model through EDC model based on reactingFoam. Could you please show me a tutorial about EDC implementation? Thank you very much!