Learning Forum

[felix512]

Hello,

I already checked others forum but it couldn't fix my problem, so I would like to know if someone can help me please.

I have an ignition issue with my combustion simulation.

My simulation uses the GRI 3.0 mechanism, for non premixed air-methane combustion, but the ignition doesn't start... When I do the same simulation with methane-air mixture (in species transport model), it works.

Do you know how I can solve this issue please ? 

Here are my parameters :

Steady combustion, with energy model, k-e viscous model and Species Transport model (using GRI 3.0 + Eddy Dissipation Concept)
Air inlet velocity (I tried 2,4 and 6,1, but it doesn't works...), 300K
Fuel inlet velocity 5 m/s, 300K, 100% methane
Outlet pressure of 101kPa
Coupled Pressure scheme, and 2nd order for spacial discretisazions.
I'm trying to reproduce a simulation I found on a paper, to learn and improve in combustion simulations. Hence, normally these parameters are correct (according to the paper, which succeed to do simulations with these parameters...).

So I don't get why my mixture doesn't ignite, can you help me please ?

PS : The mesh is normally correct and is more precise than the mesh of the paper.

Thanks.

[felix512]

Anyway, I just put a wall at 2000K to start the ignition, and it works.

That's not the ideal way I guess but it's giving good results.

[Vijay Narayan]

HI Felix,

If i understand correctly your FR(species transport case is working) whereas the non premixed model is not igniting as expected ?. Are you using the Equillbrium model or Steady diffusion flamelet model ? Not sure how patching a temp triggered reaction (this approach would have worked for the FR model but not for PDF/flamelet based models. You could also try the partially premixed model (FGM based) and patch the progress variable to trigger combustion.

 

[felix512]

 

 

Hi Vijay, thanks for yor answer.

I don't understand your question; the "Chemical Equilibrum" or "Steady Diffusion Flamelet" are available for Non-premixed models. Here I'm using the Species Transport model, with GRI Mech 3.0 and EDC. 

In the 1st case I had ignition : with a wall at 2000 K starting the ignition. The geometry had a bluff, which increase turbulence and so enable ignition.

In the 2nd case, I just removed the bluff, and keep the wall at 2000K. With this geometry, I don’t have ignition. I guess this is because there is less turbulences due to the bluff removed, while EDC needs turbulences to start chemical reactions (= ignition). I don’t know how to do… 

I want to use the GRI Mech (so I need species transport model). I already tried to patch a high temperature, and all species at 0.01, but it’s also not working.

Even when I don’t put a volumetric reaction, or just the Eddy Dissipation model, I don’t have ignition.

 

Thanks for your help

 

 

[Vijay Narayan]

HI Felix,

Sorry for the delayed response. I think the notification got lost in my outlook emails.

From your original post, i was of the impression that you were doing both FR and flamelet based modelling. But now from the last post, it is clear that you have using only FR approach and the difference between geom is mainly in the presence of bluff body. For the reaction to happen, Fuel and oxidizer need to be mixed within the flammability limit: this would have been achieved in the recirculation region close to bluff body and hence the system ignited with the hot temp trigger. You could compute the equivalence ratio using fuel and oxidizer mass fraction for the design w/o bluff body and the best guess is- the system isnt properly mixed there.

Based on your post in another Forum thread - i have a confusion in your original statement: why do you state that air-methane doesnt ignite while methane-air system ignites ? arent they the same.

 

[felix512]

Hi Vijay,

I'm no longer working on this project, because I couldn't make it work, but thanks for your answer.

Just to clarify, in both case I was trying to burn a blend of Methane-Air. But in one case, I was using the GRI Mechanism (no ignition obtained), and in the other case, I was using the default mechanism provided by Fluent named "Methane-Air", with only 1 or 2 reactions (here I had an ignition).

Then I guess you're right, the problem might be that the systme wasn't properly mixed. I based my simulations on eq. ratio and tested various different cases, but it never worked...

Thank you