what is the meaning of giving a fixed temperature to the wall

Weiqiang Liu · January 16, 2019, 13:51

hi all,

I am doing a micro-channel catalytic methane combustion simulation with fluent. after trying many methods recommended by literature, I failed to ignite the mixture, namely the simulation results are like cold flow.

somebody told me to remove the solid computation domain and give a fixed high temperature to the catalytic wall in order to ignite the mixture.

what confuses me is what is the meaning of giving a fixed temperature to the wall. The wall temperature is supposed to be computed by CFD calculation. How can I know the wall temperature in advance. What if the calculated temperature is not the same with the pre-defined wall temperature.

I am a new learner of CFD. Can anybody give me some help?

thanks

weiqiang.

LuckyTran · January 16, 2019, 16:10

In addition to mass transport and momentum transport equation, there is an energy transport equation. Let's suppose you take temperature as the transport variable (instead of internal energy, or enthalpy, or something else). Then you need to provide boundary conditions for temperature.

The canonical boundary conditions are: fixed temperature, fixed gradient (specified heat flux). So yes you can specify and should specify a wall temperature or a wall heat flux. When you specify a heat flux then you can say the temperature is calculated from the solution. But fixing the wall temperature is an even more canonical way to solve the pde (a fixed wall temperature constraint is a stronger constraint than fixing the heat flux).

When you simulate a coupled fluid and solid domain, you are coupling the temperature/gradient boundary condition of the fluid domain with the solid domain. That is, you say my wall temperature of the fluid is the same as the solid or you say the heat flux of the fluid is the same as the heat flux of the solid.

Removing the solid domain is a good idea since you are having trouble. Simplify your model and figure out why you cannot get sustained reactions.

Weiqiang Liu · January 16, 2019, 21:59

Quote:

Originally Posted by LuckyTran

In addition to mass transport and momentum transport equation, there is an energy transport equation. Let's suppose you take temperature as the transport variable (instead of internal energy, or enthalpy, or something else). Then you need to provide boundary conditions for temperature.

The canonical boundary conditions are: fixed temperature, fixed gradient (specified heat flux). So yes you can specify and should specify a wall temperature or a wall heat flux. When you specify a heat flux then you can say the temperature is calculated from the solution. But fixing the wall temperature is an even more canonical way to solve the pde (a fixed wall temperature constraint is a stronger constraint than fixing the heat flux).

When you simulate a coupled fluid and solid domain, you are coupling the temperature/gradient boundary condition of the fluid domain with the solid domain. That is, you say my wall temperature of the fluid is the same as the solid or you say the heat flux of the fluid is the same as the heat flux of the solid.

Removing the solid domain is a good idea since you are having trouble. Simplify your model and figure out why you cannot get sustained reactions.

yes, I boundary temperature or heat flux is necessary for solving pde. what confuses me is how can I know the appropriate boundary temperature. obviously, different boundary temperature would yield different results. am I right?

LuckyTran · January 17, 2019, 06:17

Yes I think your friend is trying to help you get an ignited solution. Since you say cold stuff just flows out, setting the walls to high temperature might help keep it hot enough to stay ignited. Maybe after you get an ignited mixture with slightly different BC's, then you can try to change it back.

You are in debugging stage. Don't complain that this isn't the setting that you want. Because right now, with the settings that you want, your results are garbage. You already ran 200 000 iterations with the correct boundary conditions and got the wrong result.

Weiqiang Liu · January 17, 2019, 10:14

Quote:

Originally Posted by LuckyTran

Yes I think your friend is trying to help you get an ignited solution. Since you say cold stuff just flows out, setting the walls to high temperature might help keep it hot enough to stay ignited. Maybe after you get an ignited mixture with slightly different BC's, then you can try to change it back.

You are in debugging stage. Don't complain that this isn't the setting that you want. Because right now, with the settings that you want, your results are garbage. You already ran 200 000 iterations with the correct boundary conditions and got the wrong result.

yes, I get it. Thanks for your patience

January 16, 2019, 13:51	what is the meaning of giving a fixed temperature to the wall	#1
Weiqiang Liu Senior Member Weiqiang Liu Join Date: Feb 2018 Posts: 278 Rep Power: 9	hi all, I am doing a micro-channel catalytic methane combustion simulation with fluent. after trying many methods recommended by literature, I failed to ignite the mixture, namely the simulation results are like cold flow. somebody told me to remove the solid computation domain and give a fixed high temperature to the catalytic wall in order to ignite the mixture. what confuses me is what is the meaning of giving a fixed temperature to the wall. The wall temperature is supposed to be computed by CFD calculation. How can I know the wall temperature in advance. What if the calculated temperature is not the same with the pre-defined wall temperature. I am a new learner of CFD. Can anybody give me some help? thanks weiqiang.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Commercial meshers] Fluent3DMeshToFoam	simvun	OpenFOAM Meshing & Mesh Conversion	50	January 19, 2020 15:33
Shadow Wall and temperature	norger	FLUENT	10	September 28, 2019 11:43
Fixed Wall Temperature Pipe With Odd Results	khoopes	CFX	5	June 15, 2015 07:58
Wall BC: fixed Temperature Distribution	camoesas	OpenFOAM Pre-Processing	0	February 6, 2012 05:19
Question about bcdefw.f for wall temperature bc.	Jimmy	Siemens	10	March 18, 2008 15:28

January 16, 2019, 16:10		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	In addition to mass transport and momentum transport equation, there is an energy transport equation. Let's suppose you take temperature as the transport variable (instead of internal energy, or enthalpy, or something else). Then you need to provide boundary conditions for temperature. The canonical boundary conditions are: fixed temperature, fixed gradient (specified heat flux). So yes you can specify and should specify a wall temperature or a wall heat flux. When you specify a heat flux then you can say the temperature is calculated from the solution. But fixing the wall temperature is an even more canonical way to solve the pde (a fixed wall temperature constraint is a stronger constraint than fixing the heat flux). When you simulate a coupled fluid and solid domain, you are coupling the temperature/gradient boundary condition of the fluid domain with the solid domain. That is, you say my wall temperature of the fluid is the same as the solid or you say the heat flux of the fluid is the same as the heat flux of the solid. Removing the solid domain is a good idea since you are having trouble. Simplify your model and figure out why you cannot get sustained reactions.

January 17, 2019, 06:17		#4
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	Yes I think your friend is trying to help you get an ignited solution. Since you say cold stuff just flows out, setting the walls to high temperature might help keep it hot enough to stay ignited. Maybe after you get an ignited mixture with slightly different BC's, then you can try to change it back. You are in debugging stage. Don't complain that this isn't the setting that you want. Because right now, with the settings that you want, your results are garbage. You already ran 200 000 iterations with the correct boundary conditions and got the wrong result.