[gr135]

Hi everybody

I have been trying to run Combustion of Methane-air mixture using species transport eddy dissipation model. I happen to see rise in temperature to 2400 in steady state which is expected but in transient state I see no combustion in the flow or any trace of product species.

Is there something I am missing in transient combustion? I have learnt from other forum that in steady state solver sets many higher order and non-linear terms to zero which are time dependent that might be the reason why I see agreeable results. Can someone guide on how to simulate Transient combusting flows?

[LuckyTran]

Did you start your transient simulation from the steady results? Play a video (easy to say, harder to day) and watch what happens to the flame. Probably it couldn't stabilize and burnt out, much like lean blow-off.

Large time-steps in transient are also prone to blow off.

[gr135]

Hi,

Thanks for the response

Not from results but I have run transient analysis with the same setup I have used for steady analysis. I see no flame in transient analysis nor do I see any production of CO2 and H20 at any point in the total flow time of 1 sec that verifies that there is no lean blow off I guess.

flow is not diverging even at time step of 0.01, that takes out large timestep out as probable reason.

[LuckyTran]

Yes but in your transient analysis you must have started with some initial condition. There should have been some sort of flame, or maybe you started with a cold-flow with an igniter.

A mixture of methane and air doesn't spontaneously ignite.

If there's no flame ever, then there won't be any flame ever. Combustion isn't spontaneous. Obviously if there's no flame you won't see any flame blow-out because there's no flame.

[gr135]

Okay, does it mean that I should start with steady analysis for few iterations and then continue with transient analysis? But would that simulation give me a video on how the flow the flame is developing in the flow?

My reasoning: I have given air inlet above 1000 degrees which is higher than auto-ignition temperature of methane so I expected flame without any igniter.

Thank you for the response.

[LuckyTran]

I'm not sure why it didn't ignite while above the autoignition temperature, but it's highly dependent on the combustion model. Some models just can't simulate any ignition events at all (i.e. flame speed models).

I would use the steady case and switch over to transient, since you are sure there is a flame in that case. Just check and make sure the flame doesn't get blown out.

[gr135]

Yeah I have tried doing that. In fact I have run the same case in CFX and it worked well, observed stable flame in transient analysis.

I don't know if its the right place to ask the question and I have posted this in CFX forum but If you can help it would be pretty helpful.

I have been trying to simulate combusting flows using EDM and apply Electromagnetic model to it. I want to capture the effect of magnetic and electric fields on combusting flow field in a duct combustor. The ions generated in the flow due to chemical reactions and the high temperatures should be effected my the body forces and I intend to measure the field strengths required to effect the flow structures in the field.

I have tried to couple all the physics in comsol but I do not have the level of understanding required to do it in comsol.

I wanted to know if the same can be done in CFX activating EM model along with the flow. I am unable to find the electrical conductivity properties of reactants and products of the reaction that I want to simulate in order to see what the model does. (hydrogen- Air Mixture)

I did not understand which combustion model is the right pick as you say its highly dependent on the model we pick. I have learnt from tutorial that PDF Flamelet can generate information regarding the intermediate ions in the reactions but I do not have license for CFX RIF to generate PDF libraries? if there way I can find Pdf libraries in the internet for Hydrogen-air mixture.

So, Can it be done in CFX with EM model? Things I should try in sequence to be able to do order to do it? or softwares I should try to learn

Thank you for your time and valuable suggestions.

[gr135]

Hey!

Anyone who can help me with aforementioned query?

Thank you for your response in advance.

[souza.emer]

Originally Posted by gr135

Hey!

Anyone who can help me with the aforementioned query?

Thank you for your response in advance.

Hey Gr135,

I'm trying to simulate a transient combustion too, with the reaction CO + 1/2O2 = CO2

So, I was also having trouble with my transient simulation, I wasn't getting any CO2 production and my temperature didn't change, but in the steady state condition, the reaction happened. With this first steady state result, I changed to transient and the reaction proceeded, although it didn't converge (10^-2 residuals at continuity). So it was one try that I made. Someone knows a more accurate approach?

[acimino]

Originally Posted by gr135

Okay, does it mean that I should start with steady analysis for few iterations and then continue with transient analysis? But would that simulation give me a video on how the flow the flame is developing in the flow?

My reasoning: I have given air inlet above 1000 degrees which is higher than auto-ignition temperature of methane so I expected flame without any igniter.

Thank you for the response.

The problem is in the equations used to model chemical reactions, i.e. the rate equations. Some depend also on the mass fractions of the products of combustion.
To start the flame I used a simple model that represents a pilot flame: a small inlet at a high temperature (say 2000 K) and with the mass fractions of the burned products of combustion of the fuel oxidizer mixture that you are trying to model.

I calculated those using NASA's CEA code:
https://cearun.grc.nasa.gov/index.html

Hope it helps