[bluejojoba]

Hi!

I need to model methane/air combustion in a furnace. I am using "species transport" and EDC model. Methane an air is travelling into furnace through pipes, inner pipe is air and outer pipe has air. inner pipe is placed in the outer pipe.

My methane and air is entering at 300 K, 6 m/s and 4,2 m/s, respectively. I am using 2-step methane air which is already given by FLUENT. I am using Coupled solver and PRESTO! for pressure.

I just cannot get ignition and combustion does not occur. I put a tiny wall and defined it as 1000 K. Also after a hybrid initialize, I mark a region near to mixing of fuel/air and patch that area as 1500 K and 0.05 co2, 0.05 co and 0.05 h2o.

But just the area near to tiny wall has a temperature gradient which is like 0.7 mm, but the whole inner area is still blue (300K)

Where I am doing wrong? How can I get ignition?

[Andrea1984]

Hi bluejojoba,

did you manage to solve this? I am facing the same problem.
My model works fine when using ED and Finite rate/ED models but I see no reaction when I switch to EDC.
Don't know what the problem is...

Thank you

Andrea

[Zaktatir]

you have to initiate the combustion with a plug or something (high temperature/Products mass fraction/..) and perhaps to start with the eddy breakup model as start up.

[Andrea1984]

Hi Zaktatir,

thank you for your reply.

I have tried to start the EDC simulation both from the results obtained with the Eddy Dissipation model (that are quite reasonable) or from a "cold" flow field without reaction (i.e. with the "Volumetric" option in the Species Transport window swithced off to simulate only the species diffusion without reaction) and with a mass fraction of 0.01 for the reaction products patched in the reaction zone (with this kind of initialization the Eddy Dissipation model works fine), but in both cases i had similar results: no reaction at all and the maximum temperature in the domain is equal to the air inlet temperature in the combustor (about 700 K).

Do you have any clue of what is going on?

Thanks

Andrea

[Zaktatir]

try to patch a very high temperature location/ very high loading of the the reagant. If it does not work the whole setup has to be checked by fluent support

[Muh.Hesan]

I have the same problem, in "eddy dissipation" combustion occurs but in "EDC" does not. How did you fix the problem?

[zhuoyue89]

the same problem,how did you solve it? thanks.

[Andrea1984]

Hi zhuouyue89,

I overcame this problem by using a solution calculated with another combustion model (non-premixed, premixed or partially premixed model depending on your specific application) as initial condition for the EDC.

Andrea

[zhuoyue89]

Andrea，thanks a lot.
I'm simulating the internal combustion engine,which is an unsteady case,so I have no idea how to calculate a initial result before use the EDC model,do you have any advice?Besides,my english is poor since I'm from Asia.Sincerely.

[Andrea1984]

Is it a spark-ignition or a compression-ignition (Diesel) engine?

[zhuoyue89]

spark-ignition,I use the EDC model with methanol(it's mechanism includes 21 species,93reactions) and CH4(it's mechanism includes 53 species,over 300 reactions),but the resullt is not reasonable,for example,the temperature and pressure in the cylinder is much lower than the true conditions.And it seems like the flame spread very very slow,but around the ignition plug,there some CO2/H2O generated,also the temperature reached even 3000K around the plug.

[Andrea1984]

You should be able to compute an initial solution using the partially premixed model (it is not necessary to reach fully convergence at this stage), then switch to EDC

[zhuoyue89]

ok,I'll consider how to get an initial solution using the partially premixed model or the finite-rate/eddy-dissipation model about the unsteady case .thank you very much.

[Andrea1984]

How many reactions do you have in your chemical mechanism? Because the FR/ED model is useless with a number of reactions > 2

[zhuoyue89]

Well,I mean，my ultimate purpose is to use EDC with the detailed mechnism,if i use FR/ED to get an initial solution,the mechanism will be the fluent itself's mechanism.if i have represent my idea clearly?

[Andrea1984]

I would suggest you to use your detailed mechanism with the equilibrium partially premixed model to calculate your initial condition, and then switch to EDC using the same mechanism

[zhuoyue89]

but if i can get the results(such as the middle species's concentration distribution ) use the detailed mechanism with the equilibrium partially premixed model,is it still necessary to use the EDC model to calculate the case again?

[Andrea1984]

If you are satisfied with your results with the partially premixed model I don't see why you should switch to EDC (which is significantly more expensive in terms of computational cost)
If you want to stick to the partially premixed I suggest you to use the steady diffusion flamelet or the FGM approach instead of the equilibrium hypothesis to get more realistic results

[zhuoyue89]

however,what i have been told was that only EDC and PDF combustion model can couple the detailed mechanism in fluent,so it's not true?you mean partially premixed model can also calculate with detailed mechanism in #16?

[Andrea1984]

The partially premixed model combined with the flamelet approach can handle detailed mechanisms but it is suitable only for specific combustion regimes (i.e. flamelet regime).
My understanding (confirmed by the results I have obtained so far with different combustion models on for a gas turbine combustor) is that EDC and transported PDF can handle detailed mechanisms as well, and are suitable for all combustion regimes