Learning Forum

[Weiqiang Liu]

hello, all,

I am doing a steady micro-channel methane air premixed catalytic combustion simulation with fluent.

The gas phase mechanism is GRI 3.0 and surface mechanism is DETCHEM. I am using species transport with laminar finite rate model. 

To help with convergence, I calculated the case with the following procedures:

1. turn off volume and surface reaction. turn off species equations of water and carbon dioxide.

2. calculate with 1000 iterations to get a cold flow field.

3.turn on volume and surface reaction. turn on species equations of water and carbon dioxide.

4. patch a 2000K temperature to the rear part of the channel. meanwhile some concentration of methane and oxygen are also patched to this region.

5. continue iteration.

my problem is the outlet temperature of the channel keeps decreasing until room temperature. namely, I can not ignite the mixture. 

I have tried a lot ways recommended by literature. However, nothing succeed. 

below is the picture of residuals. Looks like the case doesn't converge very well.

below is the computation domain of my case

can anybody give me some suggestions or hints? I really really need your help !!!

best regards

weiqiang

[Amine Ben Hadj Ali]

Either provide the spark or heat required to overcome the energy barrier for the reactions to occurs via patching a hot spot/zone or via providing non adiabatic walls.

 

Best regards,

Amine

[Weiqiang Liu]

hello Amine,

I just removed the gas phase reaction and the mixture can be ignited. I don't know why. Some literature claim that gas phase mechanism can be neglected under their investigated condition. However, I am wondering is it because they have difficulty in igniting the mixture with gas phase mechanism.

 

weiqiang

[Amine Ben Hadj Ali]

Check every gas phase reaction if it's correct or not. You can use finite rate plus ebu model ignite with temperature and patch eith some products if you still have issues. You can contact the authors of the papers you are referring to.

[Weiqiang Liu]

hello, Amine

I just abandoned all the gas phase reactions and left the gas thermodynamic and transport data blank. I guess fluent will use default values in its database regarding to the gas thermodynamic and transport data.

With the above-mentioned procedures, all the divergence and failure of ignition problems just disappear and I can reproduce results in literature also.

However, if I import any chemkin thermo and transport data, divergence happens again?

I really can not understand why this happen. Have you ever met this problem before?

Best regards

Weiqiang.

[Amine Ben Hadj Ali]

Check if the properties imported are correct. This would justify why gas phase reactions are working with default Fluent materials and thermo data.