Species Transport Model Using FInite Rate

Ali_95 · July 12, 2023, 11:07

I've been trying to simulate a methane burner in Fluent using the species transport model with Finite rate chemistry. I tried using the in-built Chemkin mechanism for single stage combustion, however i was only able to get ignition using the eddy dissipation model since it uses fast rate kinetics. I tried to increase temperature inside the domain to get Finite rate Arrhenius model to work but i always end up getting a floating point error. Can anyone help redirect me to where i am going wrong with this?

alainislas · July 14, 2023, 12:15

Are you using the EDM /FR model? Please post some screenshots of your species transport panel

Ali_95 · July 14, 2023, 23:44

No. I am using the Finite rate solver only and trying to ignite by provding a high temperature patch. i have also tried a spark ignition both dont work for me.

alainislas · July 16, 2023, 05:17

PLease try to use the EDM /FR model and report back...
Also if you can post some pictures of your domain and boundary conditions..that would be helpful

Ali_95 · July 17, 2023, 11:40

The EDM/FR model does not work either. Also, i want to use the Finite rate Model for my case since it is more realistic to the problem. The domain is simply a square chamber with pure methane inlet from an opening in the bottom wall, both sides have a low air flow coming in and top is outlet. Imagine something close to a methane burner inside a room.

alainislas · July 17, 2023, 12:12

Please check your boundary conditions were set appropriately, did you define the CH4 inlet in mass or mole fractions? Please plot some pictures of the cold-flow (velocity, species, temperature).

Also I suggest to first start with a global reaction mechanism, with 1 or 2 reactions. This will help to deduce if the non-ignition is attributable to the kinetic mechanism or not. I'd look at Westbrook and Dryer or Jones-Lindstedt mechanisms as initial step, and then I will move to GRI 3.0.

Ali_95 · July 17, 2023, 12:21

I started with one and two step reactions (WD and JL mechanisms) and even went up as far as GRI 3.0 but none of them seem to work.
Methane is specified as mole fractions.
Everything works with eddy dissipation since it does not require source of ignition. Its just the finite rate that has problems.

alainislas · July 17, 2023, 14:56

So if WD nor JL did not work, why did you moved to GRI mechanism? It's harder and takes longer to detect issues with such mechanism. About the finite rates, did you make sure that the value and units of the pre-exponential factor are consistent with the rate exponents? I'm not sure about Fluent, but in case of OpenFoam, the kinetic rates are not in standard SI units, but in cm & Kmol I believe. So conversion is necessary.

Ali_95 · July 20, 2023, 10:47

I have used the inbuilt chemkin mechanism for fluent for single and two step methane air combustion. So, Its very unlikely that there will be a problem with the unit conversion. There seems to be a solver problem that i just can't seem to identify.

alainislas · July 20, 2023, 11:01

Ok, so please post some contours of your flow (velocity, TKE, temperature). This can help us to understand if the dynamics of the flow provoke local extinctions or convect the high temperature patch so fast In the downstream region.

Besides, what is your equivalence ratio? Is this air combustion? Oxy?

Ali_95 · July 20, 2023, 11:08

i dont need to specify equivalence ratios since i'm not doing a premixed flame. Its just methane entering a region that has air throughout the domain. i will post the contours in a bit but i do think that the patch provides ignition, its just that the flame quenches.
Also for the patch, i specify both high temperature and a mixture with methane and air.

alainislas · July 20, 2023, 11:22

Regardless if you are doing premixed or.non-premixed combustion, you should know the equivalence ratio of your case. In Fluent I know there is no dedicated place to input an ER, but you can control this by varying the mass fractions of the reactants / oxidizer being injected. In the species transport model you know you have control over all the species.

So, Based on your response I understand that you inject 100% CH4 in the burner? If not, please specify. Besides...
How is this burner defined? Is this coflow ? Oxygen is injected as well?

I'd create a CFF of the ER to see if there are local zones where the ER is excessively high so the temperature patch can't simply ignite the mixture.

July 12, 2023, 11:07	Species Transport Model Using FInite Rate	#1
Ali_95 New Member Ali Join Date: Jul 2023 Posts: 6 Rep Power: 2	I've been trying to simulate a methane burner in Fluent using the species transport model with Finite rate chemistry. I tried using the in-built Chemkin mechanism for single stage combustion, however i was only able to get ignition using the eddy dissipation model since it uses fast rate kinetics. I tried to increase temperature inside the domain to get Finite rate Arrhenius model to work but i always end up getting a floating point error. Can anyone help redirect me to where i am going wrong with this?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Fixed Temperature Constraint problems for coal combustion	yglovella	OpenFOAM	0	February 16, 2023 01:00
species transport and non premixed model	Weiqiang Liu	FLUENT	2	June 22, 2020 08:15
[swak4Foam] swakExpression not writing to log	alexfells	OpenFOAM Community Contributions	3	March 16, 2020 18:19
A question about species box of species transport model	tjushang	FLUENT	2	June 25, 2014 07:56
How to use Eddy Dissipation rate only When using Finite rate/Eddy Dissipation model	Dziyad	FLUENT	2	February 19, 2014 17:25

July 14, 2023, 12:15		#2
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	Are you using the EDM /FR model? Please post some screenshots of your species transport panel

July 16, 2023, 05:17		#4
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	PLease try to use the EDM /FR model and report back... Also if you can post some pictures of your domain and boundary conditions..that would be helpful

July 17, 2023, 11:40		#5
Ali_95 New Member Ali Join Date: Jul 2023 Posts: 6 Rep Power: 2	The EDM/FR model does not work either. Also, i want to use the Finite rate Model for my case since it is more realistic to the problem. The domain is simply a square chamber with pure methane inlet from an opening in the bottom wall, both sides have a low air flow coming in and top is outlet. Imagine something close to a methane burner inside a room.

July 17, 2023, 12:12		#6
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	Please check your boundary conditions were set appropriately, did you define the CH4 inlet in mass or mole fractions? Please plot some pictures of the cold-flow (velocity, species, temperature). Also I suggest to first start with a global reaction mechanism, with 1 or 2 reactions. This will help to deduce if the non-ignition is attributable to the kinetic mechanism or not. I'd look at Westbrook and Dryer or Jones-Lindstedt mechanisms as initial step, and then I will move to GRI 3.0.

July 17, 2023, 12:21		#7
Ali_95 New Member Ali Join Date: Jul 2023 Posts: 6 Rep Power: 2	I started with one and two step reactions (WD and JL mechanisms) and even went up as far as GRI 3.0 but none of them seem to work. Methane is specified as mole fractions. Everything works with eddy dissipation since it does not require source of ignition. Its just the finite rate that has problems.

July 17, 2023, 14:56		#8
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	So if WD nor JL did not work, why did you moved to GRI mechanism? It's harder and takes longer to detect issues with such mechanism. About the finite rates, did you make sure that the value and units of the pre-exponential factor are consistent with the rate exponents? I'm not sure about Fluent, but in case of OpenFoam, the kinetic rates are not in standard SI units, but in cm & Kmol I believe. So conversion is necessary.

July 20, 2023, 10:47		#9
Ali_95 New Member Ali Join Date: Jul 2023 Posts: 6 Rep Power: 2	I have used the inbuilt chemkin mechanism for fluent for single and two step methane air combustion. So, Its very unlikely that there will be a problem with the unit conversion. There seems to be a solver problem that i just can't seem to identify.

July 20, 2023, 11:01		#10
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	Ok, so please post some contours of your flow (velocity, TKE, temperature). This can help us to understand if the dynamics of the flow provoke local extinctions or convect the high temperature patch so fast In the downstream region. Besides, what is your equivalence ratio? Is this air combustion? Oxy?

July 20, 2023, 11:08		#11
Ali_95 New Member Ali Join Date: Jul 2023 Posts: 6 Rep Power: 2	i dont need to specify equivalence ratios since i'm not doing a premixed flame. Its just methane entering a region that has air throughout the domain. i will post the contours in a bit but i do think that the patch provides ignition, its just that the flame quenches. Also for the patch, i specify both high temperature and a mixture with methane and air.

July 20, 2023, 11:22		#12
alainislas Senior Member Alain Islas Join Date: Nov 2019 Location: Mexico Posts: 142 Rep Power: 6	Regardless if you are doing premixed or.non-premixed combustion, you should know the equivalence ratio of your case. In Fluent I know there is no dedicated place to input an ER, but you can control this by varying the mass fractions of the reactants / oxidizer being injected. In the species transport model you know you have control over all the species. So, Based on your response I understand that you inject 100% CH4 in the burner? If not, please specify. Besides... How is this burner defined? Is this coflow ? Oxygen is injected as well? I'd create a CFF of the ER to see if there are local zones where the ER is excessively high so the temperature patch can't simply ignite the mixture.