Hi,
I am trying to modelise a bluff body flamme (Sandia laboratories) using reactingFoam.
I began with a non-reacting jet of CH4/H2 at 118 m/s in a air coflow
at 40 m/s in order to modelise the species mixing first. But, after several iterations,
I get this message :

FOAM FATAL ERROR : attempt to use janafThermo<equationofstate> out of temperature range 200 -> 5000; T = 145

From function janafThermo<equationofstate>::checkT(const scalar T) const
in file /home/dm2/henry/OpenFOAM/OpenFOAM-1.2/src/thermophysicalModels/specie/lnInclude/ janafThermoI.H at line 73.

Just before the bug, I noticed a iteration with : bounding k, min:-1.396588e+11 max: 3.28261e+11 average: 4.01455e+08
bounding epsilon, min:-2.495888e+12 max: 8.493524+13 average: 1.925648e+12


When I observed the registred data before it began to bug, i can see a cold point and a hot point in the middle of the box even
if the chemistry is off.

Before this bug, the species mixing seemed to be well done and the velocity profile was logical,
so I have difficulties to understand what is going on.

First, I thought the bug was due to the important jet velocity 118m/s but I tried with an other flamme with a jet velocity
equal to 42 m/s and I get the same problem

Has someone an explanation for this problem ?

Best regards

Julienh

[lucchini]

Hi,
Check the mesh with checkMesh and see if everything is OK. If you have high skewness or non-orthogonalities results can be seriously influenced by that.

Then check the value of courant number, and try to find a time step to keep it below, let's say, 0.2

let me know.
regards
Tommaso

Hi,
thank you for this quick answer. I checked the mesh and I get :

Mesh non-orthogonality Max:39.7779 average :6.71887

Max skewness = 39.7391 percent

what is the meaning of the skewness? Is 39.7391 percent correct ?

I tried also a jet of propane at 118 m/s in a coflow of propane at 40m/s, and it works.

The problem I had with the cold and the hot point is perhaps due to a problem of mixing ?

Best regards

Julienh

Hi,
thank you for this quick answer. I checked the mesh and I get :

Mesh non-orthogonality Max:39.7779 average :6.71887

Max skewness = 39.7391 percent

what is the meaning of the skewness? Is 39.7391 percent correct ?

I tried also a jet of propane at 118 m/s in a coflow of propane at 40m/s, and it works.

The problem I had with the cold and the hot point is perhaps due to a problem of mixing ?

Best regards

Julienh

[rajesh010439]

Hi,
Can somebody help me out how to implement eddy-dissipation model in openFOAM.Actually, I have to find out the dependence of Flame extinction on chemical time scale for turbulent lifted jet diffusion flame and I am searching how to incorporate chemical time-scale criteria for flame extinction.
Thanks in advance for your responses.
Best Regards,
Rajesh

[stefanke]

You do not need to implement a eddy-dissipation model because it is already there. What I mean is the PaSR combustion (an extension of the classical eddy dissipation concept) model which was proposed by Golovitchev.

If you are looking for combustion models which are able to simulate flame extinction you have to look for more advanced combustion models because eddy-dissipation models are not able to reproduce this type of physics.

[rajesh010439]

Thanks Stefan,
I tried my simulation with PaSR model, but it doesn't capture properly lift-off heights of lifted flames which I am looking for.
I am trying to implement five step reaction mechanism for propane given by Westbrook and Dryer, in which two reactions are based on arrhenious model and the rest are eddy dissipation.I don't know how to use both models together.Can somebody help me in this regard?

regards,
Rajesh

[mavimo]

I downloaded ReactingFoamCase form Wiki, but when I run it obtain:
<font color="0000aa">
debug::switchSet(const char*, dictionary*):
Cannot find DimensionedConstants in dictionary /home/simulation/OpenFOAM/OpenFOAM-1.3/.OpenFOAM-1.3/controlDict
</font>
Anyone can help me?

bye
Marco

PS: sorry for my bad english

[lucchini]

Ciao Marco,
have a look at the dieselFoam tutorial (it is basically reactingFoam with spray), you will probably find what is wrong in your case setup.
bye
Tommmaso

[mavimo]

Ciao Tommaso,
if I run <font color="0000aa">dieselFoam</font> case obtain the same result. It's possible I have a error in installation step? I had execute this passage (my guide, italian only). All run but not reactingFoam and dieselFoam!
bye
Marco

PS: I'm a student of polimi :D

[mavimo]

To solve "my" problem I must add the code [¹] to /openfoam/installation/directory/OpenFOAM-1.3/.OpenFOAM-1.3/controlDict

[¹]
DimensionedConstants
{
// Universal gas constant [J/(kmol K)]
R 8314.51;

//- Standard pressure [Pa]
Pstd 1.0e5;

//- Standard temperature [K]
Tstd 298.15;

}

[tak]

Hi, all

Now, I try to calculate methane diffusion jet flame, where fuel(methane) and air is supplied at temperature 298K, using reactingFoam.

My question is how to add an ignition condition into the reactingFoam and the jet flame case.

In the reactingFoamCase of OpenFOAM wiki, fuel can self-ignite because air and fuel are completely preheated.

Thanks a lot for the answers in advance.

[mavimo]

See the file constant/combustionProperties, and add this line:
ignitionProperties1
{
ignite on;
ignitionPoint ignitionPoint [ 0 1 0 0 0 0 0 ] ( 0.01 0 0 ) ;
timing timing [ 0 0 1 0 0 0 0 ] 0.0e-1 ;
duration duration [ 0 0 1 0 0 0 0 ] 1.0e-0 ;
}
to add a ignition point (set correct point, time and duration).

Bye
Marco

[tak]

Hi Marco,

Thank you for your quick answer!!
I have one question; default application "reactingFoam" seems to not read the file "constant/combustionProperties".
Do I have to modify the source code of "reactingFoam" in order to ignite the jet flame?

Best regards,

taka

Hello all,

I have to simulate a combustion model with air,fuel mixture ,for which I want to try reactingFoam code.
But when I create a new case in FoamX, reactingFoam solver in not visible in the given classes.So, I am not able to implement the reactingFoam code.Should I use XiFoam and modify it,but am not familiar with modification to different solvers.

I am a beginner to openfoam and would appreciate help in this regard.

Thanks.
Mayank.

hello ,

Its a rather silly question but where are densities for fuel,O2,N2 entered in reactingfoam.
Since I want to use 'massFluxInletVelocity' type bc which uses the latest rho.

Any help would be appreciable.
Mayank

[mavimo]

Hallo all,
I have simulated a combustion model by reactingFoam with ethane/air mixture (non stoichiometric, laminar flame). If I use a very simple model (only one reaction) it ran and I obtain plausible result, but if I use a complex mechanism (developed by "Department of Mechanical Engineering - University of California, Berkeley") it don't work and I receive this error:

Reading thermophysicalProperties
Selecting thermodynamics package hMixtureThermo<reactingmixture>
Selecting chemistryReader chemkinReader


--> FOAM FATAL ERROR : attempt to use janafThermo<equationofstate> out of temperature range 300 -> 3000; T = 298.15
#0 Foam::error::printStack(Foam:stream&)
#1 Foam::error::abort()
#2 Foam::sutherlandTransport<foam::speciethermo<foam: :janafthermo<foam::perfectgas> > >::alpha(double) const
#3 Foam::hMixtureThermo<foam::reactingmixture>::calcu late()
#4 Foam::hMixtureThermo<foam::reactingmixture>::hMixt ureThermo(Foam::fvMesh const&)
#5 Foam::hCombustionThermo::addfvMeshConstructorToTab le<foam::hmixturethermo<foam:: reactingmixture> >::New(Foam::fvMesh const&)
#6 Foam::hCombustionThermo::New(Foam::fvMesh const&)
#7 main
#8 __libc_start_main
#9 __gxx_personality_v0 at /usr/src/packages/BUILD/glibc-2.3/csu/../sysdeps/i386/elf/start.S:122


From function janafThermo<equationofstate>::checkT(const scalar T) const
in file /home/dm2/henry/OpenFOAM/OpenFOAM-1.4/src/thermophysicalModels/specie/lnInclude/ janafThermoI.H at line 73.

FOAM aborting

I uploaded my case at this address.
Into chemkin folder there are 4 file:
- chem.inp and therm.dat are files with simple kinetic model (only one reaction)
- wf_mech.dat and wf_thermo.dat are files with complex kinetic model (approx 140 reactions).
You can switch between simple or complex model by commenting (and decommenting) lines into constant/thermophysicalProperties file.
Anyone can help me?
Thanks
Marco
PS: sorry, but I don't write English very well...

[jacques_kools]

Hi everybody,

I have the following question:
I am working on low pressure ,laminar gas flow in vacuum systems with large relative variations in pressure , and varying temperatures on the surfaces. Boundary conditions are time dependent, so I need a transient solver, compressible and laminar. Up to now I have been working only with single species (e.g. Ar), and thus sonicFOAM did the job very well.
Now I want to do multispecies ( e.g. O2 in Ar), and monitor things like diffusion of one species in another, and also flushing of one species by another. Chemistry is not needed at the moment, but it would be nice to have later on.

I noted some posts suggesting that reactingfoam would be the right solver for that, but I have a few questions. I looked at the code , and I found it quite opaque, so I wonder if anyone could help me with these ?

0) Is there any other solver that might be better suited than reactingfoam ?

1) I read in one of the posts that ReactingFoam is compressible. Can one of you guru's confirm that ?

2) I am interested in laminar flow, so I want to desactivate the turbulence model. What are the exact steps to do that ?

3) I played with the example posted in the wiki,and found I could get some kind of diffusion ( i.e. species move even when the pressure is constant, and U=(0,0,0) everywhere ). What is the underlying physics. Where do I input the diffusion coefficient ?

4) I had trouble working with a Chemkin file that contained no reaction. I faked my way out by reducing the rate constant by a factor 1e9, but is there a real way to do it ?

Thanks for your response.

Jacques Kools

[markusrehm]

Hello Marco,

your thermo data isn't suited for temperatures below 300 K (I guess polynomials for a few species are given in the range [300-1000] and [1000-3000] see example below). Does anyone know how that is treated in other codes? Is there some extrapolation or some trick to avoid it? For example in GRI-Mech the thermo.dat-lines for CH3O look like that:

CH3O 121686C 1H 3O 1 G 300.00 3000.00 1000.000 1
0.03770799E+02 0.07871497E-01-0.02656384E-04 0.03944431E-08-0.02112616E-12 2
0.12783252E+03 0.02929575E+02 0.02106204E+02 0.07216595E-01 0.05338472E-04 3
-0.07377636E-07 0.02075610E-10 0.09786011E+04 0.13152177E+02 4

Thank you.

Regards Markus

[mavimo]

Hello Markus

Ok, my thermo.dat-file don't support temperature below 300K but I dont have it into my case. Inlet BC and body temperature are set to 300K, other walls are zero gradient (or empty).
If I increase all temperature (Inlet BC and body) to 1000K it don't work again (the same one error). Why reactingFoam use T_a (298.15)? This morning I'll test a new thermal.dat with all polynomial for temperature range up to 200K (only for test).

Bye
Marco