Convective heat transfer using Fluent

lamboram · August 27, 2015, 05:22

Hello all,

I am starting to work with Fluent for convection heat transfer (Natural convection) between a solid which is generating heat and a surrounding volume of air (the volume is fixed). I have tried two methods so far

1. defining these two bodies as a single part and by doing so I get a conjucate mesh on the boundry (between solid and fluid)
2. deining these two bodies as a individual parts, the individual faces of the solid as a named selection, upon updating it in Fluent I get respective shadow walls.

Now, in both cases I have no success so far. I know there is a problem in wall BC definition. Now I have the basic settings right. Can you tell me some material where I could understand the wall BC definitions better.

one more question. I have worked relatively a lot in Ansys Thermal, can I model the fixed volume of air around the body as a 3D Model and simulate it in FEM? Will it make good sense? because the air in our case is not moving around and the volume might also be same because its a closed environment.

Looking forward your suggestions!
Ram

LuckyTran · August 27, 2015, 13:22

Method #2 is the correct way. You should have two cell zones, one for the solid and one for the fluid. You should have a coupled wall (an interface) between at the solid-fluid boundary. A shadow wall will be automatically generated when you create this coupled wall. There are no boundary conditions on this interface (it is not a boundary).

You need to specify the correct heat generation rate in your solid domain. Then finally you need to specify sensible boundary conditions for the outside of the fluid domain.

Air is moving if it's natural convection, it's driven by density gradients instead of being forced but it is still a type of bulk motion. If you have truly stagnate problem then you could treat it as a solid body but you should confirm first whether or not you expect there to be any bulk motion.

lamboram · September 2, 2015, 06:13

Thanks LuckyTran,

I ran a simulation with reasonable results with the Method #2. You are right regarding the air bulk. I think it is sensible to simulate using the CFD method. But I have simulated using Pressure-based. I am running one using Density-based now. I will compare the results and keep you posted.

Do you know any materials where I can read in detail about these walls (Shadow, interface, coupled)?

Cheer,
Ram

Quote:

Originally Posted by LuckyTran

Method #2 is the correct way. You should have two cell zones, one for the solid and one for the fluid. You should have a coupled wall (an interface) between at the solid-fluid boundary. A shadow wall will be automatically generated when you create this coupled wall. There are no boundary conditions on this interface (it is not a boundary).

You need to specify the correct heat generation rate in your solid domain. Then finally you need to specify sensible boundary conditions for the outside of the fluid domain.

Air is moving if it's natural convection, it's driven by density gradients instead of being forced but it is still a type of bulk motion. If you have truly stagnate problem then you could treat it as a solid body but you should confirm first whether or not you expect there to be any bulk motion.

August 27, 2015, 05:22	Convective heat transfer using Fluent	#1
lamboram New Member Lamboram Join Date: Jul 2011 Location: Munich Posts: 16 Rep Power: 15	Hello all, I am starting to work with Fluent for convection heat transfer (Natural convection) between a solid which is generating heat and a surrounding volume of air (the volume is fixed). I have tried two methods so far 1. defining these two bodies as a single part and by doing so I get a conjucate mesh on the boundry (between solid and fluid) 2. deining these two bodies as a individual parts, the individual faces of the solid as a named selection, upon updating it in Fluent I get respective shadow walls. Now, in both cases I have no success so far. I know there is a problem in wall BC definition. Now I have the basic settings right. Can you tell me some material where I could understand the wall BC definitions better. one more question. I have worked relatively a lot in Ansys Thermal, can I model the fixed volume of air around the body as a 3D Model and simulate it in FEM? Will it make good sense? because the air in our case is not moving around and the volume might also be same because its a closed environment. Looking forward your suggestions! Ram

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Heat Transfer Problem in FLUENT	eng_yasser_2020	FLUENT	3	February 19, 2019 03:13
Setting up heat transfer beetween wall heated zone and flow channel zone in Fluent	tafa	Fluent Multiphase	0	May 25, 2014 06:08
Fluent 14.5 Help (Heat Transfer)	har1s	FLUENT	0	March 29, 2014 05:37
The fluent stopped and errors with "Emergency: received SIGHUP signal"	yuyuxuan	FLUENT	0	December 3, 2013 23:56
UDF convective heat transfer correlations	sarah	FLUENT	0	March 29, 2007 10:52

August 27, 2015, 13:22		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,753 Rep Power: 66	Method #2 is the correct way. You should have two cell zones, one for the solid and one for the fluid. You should have a coupled wall (an interface) between at the solid-fluid boundary. A shadow wall will be automatically generated when you create this coupled wall. There are no boundary conditions on this interface (it is not a boundary). You need to specify the correct heat generation rate in your solid domain. Then finally you need to specify sensible boundary conditions for the outside of the fluid domain. Air is moving if it's natural convection, it's driven by density gradients instead of being forced but it is still a type of bulk motion. If you have truly stagnate problem then you could treat it as a solid body but you should confirm first whether or not you expect there to be any bulk motion.