# Conjugate Heat Transfer Problem Coupled Boundary Condition

 Register Blogs Members List Search Today's Posts Mark Forums Read

 March 3, 2016, 09:37 Conjugate Heat Transfer Problem Coupled Boundary Condition #1 New Member   Edmond Lam Join Date: Jun 2015 Location: Geneva, CH Posts: 15 Rep Power: 11 Hi everyone, I am a beginner in solving conjugate heat transfer problems in FLUENT (heat transfer between solid and fluid). I have successfully created a mesh and solved my problem, but would like to know more about the details of the boundary conditions involved. When I import a mesh created in ICEM CFD and read the mesh in FLUENT, FLUENT automatically creates 'shadow' surfaces for surfaces that separate the solid region and fluid region, and automatically set the boundary condition for the original surface and the 'shadow' surface to "coupled". I have tried to look for the definition of a "coupled" boundary condition but failed. Does anyone have a clear definition of this boundary condition Thank you in advance!!

 March 3, 2016, 14:47 #2 New Member   Christoph Ferk Join Date: Feb 2016 Posts: 17 Rep Power: 10 Hi, In general there are two approaches for conjugate heat transfer: 1) two sided wall: this is what you've done 2) thin wall ad 1) This approach always uses the coupled boundary condition, because there are cells on both sides of the wall. It is the most accurate way for conductive heat transfer but requires more meshing effort, because the solid zones (wall thickness) must be meshed. So as you mentioned Fluent will automatically create a "shadow" zone of the wall if there are fluid or solid regions on EACH side of the wall. The coupled option is also selected automatically and there is no addiational thermal B.C. required, because the solver will calculate heat transfer directly from the solution in the adjacent cells. ad 2) The thickness of the wall is artificially modeled (specified in the wall B.C.). By default walls have zero thickness. In this approach conduction is onyl calculated in thickness direction. I refer also to ANSYS User's Guide Chapter 6.3.14.3.7 and 6.3.14.3.8. Go through it, there are quite good definitions of Thermal B.C. at walls. Hope it helps you. Regards

 March 4, 2016, 12:00 #3 New Member   Edmond Lam Join Date: Jun 2015 Location: Geneva, CH Posts: 15 Rep Power: 11 Hi chrisf90, Thank you for your reply. I have gone through the stated sections of the FLUENT User's Guide before and understand that FLUENT will calculate the solution directly. However, what I do not understand and wish to figure out is the actual boundary condition FLUENT uses when it calculates "automatically", such as setting the heat flux on both sides of the wall equal, or setting the temperature on both sides of the wall equal to each other. Thank you for your input anyway Does anyone have more info on my questions Any help is appreciated Thank you in advance

March 4, 2016, 22:09
#4
Senior Member

Lucky
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 5,685
Rep Power: 66
A coupled wall is not a boundary but an interface. Hence it's meaningless to talk about boundary conditions.

The wall and shadow-wall are the two artificial boundaries for the two separate domains.

Fluent applies energy balance across the interface. If you utilize the wall thickness model then a temperature jump at the boundary is permitted. You can also specify heat generation, which is also a temperature jump condition. The temperature is whatever it needs to be to satisfy the energy balance (the temperature is solved for).

Quote:
 Originally Posted by edmondlam However, what I do not understand and wish to figure out is the actual boundary condition FLUENT uses when it calculates "automatically", such as setting the heat flux on both sides of the wall equal, or setting the temperature on both sides of the wall equal to each other.
The heat flux nor the temperature, neither need be equal.

 March 6, 2016, 07:47 #5 New Member   Edmond Lam Join Date: Jun 2015 Location: Geneva, CH Posts: 15 Rep Power: 11 Hi LuckyTran, Thank you for your reply. Are there any special situations where there should be equal heat flux and/or temperature? For example, if there is no heat generation, is it that the heat flux should be equal? What about in physics in reality? You guys have assisted me a lot so far, and thank you again for your assistance!!

 March 6, 2016, 10:04 #6 Senior Member   Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,685 Rep Power: 66 If there is no heat generation then heat fluxes should be equal. If there is no thermal resistance at the interface (perfect thermal contact) then there is no temperature jump across the interface and the temperatures of the solid and fluid on either side of the interface will be equal. In reality there is always some thermal resistance because at the micro-molecular level there is space between molecules (but the magnitude may be negligible in many problems). Be careful when you ask about reality... Because the laws of physics are phenomenological laws (observed). In reality things do whatever they want to do. Concepts such as energy balance as merely observations and reality does not need to conform to these beliefs. Physical reality is a bit of an oxymoron. Rohtiwari likes this.

 March 9, 2016, 03:17 #7 New Member   Edmond Lam Join Date: Jun 2015 Location: Geneva, CH Posts: 15 Rep Power: 11 Hi LuckyTran, Thank you so much for your reply!! Now I understand them much better Thanks again

 March 9, 2016, 03:37 surface_heat_flux #8 New Member   asmita Join Date: Feb 2016 Location: Mumbai Posts: 11 Rep Power: 10 hello, Q.1 q(heat flux)=h(Tamb-Tw) q- surface heat flux Tw - wall temp q,Tw unknown how to update Tw and q through udf in fluent?

 August 10, 2016, 11:28 conjugate heat #9 New Member   hussein Join Date: Aug 2016 Posts: 5 Rep Power: 9 Hi!! thank a lot my model in ansys fluent is solid with heat flux on the upper face and fluid domain in the lower face ,but there are no heat transfer between solid and liquid, the interface is defined as wall and shadow

 April 18, 2017, 13:55 CHT boundary conditions #10 New Member   RAG Join Date: Mar 2017 Posts: 5 Rep Power: 9 HI All, I am modelling a conjugate heat transfer problem, My model is cylinder head, I have taken out the fluid volume for easy solving. I have confusion in giving boundary conditions, weather to use "wall interface" or thin walled model to properly model heat transfer from "Wall to fluid". I have another doubt, i have divide the wall in to two regions, bottom face is "wall heat" where heat is applied, and the side faces are simply wall boundaries. here, how to model the problem or to give the boundary conditions such that the heat can transfer from "wall-heat" to fluid and "wall" both simultaneously. Is it possible to apply as such. Thank U, __________________ d_Rag

April 18, 2017, 17:29
#11
Senior Member

Lucky
Join Date: Apr 2011
Location: Orlando, FL USA
Posts: 5,685
Rep Power: 66
Quote:
 Originally Posted by Rag HI All, I am modelling a conjugate heat transfer problem, My model is cylinder head, I have taken out the fluid volume for easy solving. I have confusion in giving boundary conditions, weather to use "wall interface" or thin walled model to properly model heat transfer from "Wall to fluid". I have another doubt, i have divide the wall in to two regions, bottom face is "wall heat" where heat is applied, and the side faces are simply wall boundaries. here, how to model the problem or to give the boundary conditions such that the heat can transfer from "wall-heat" to fluid and "wall" both simultaneously. Is it possible to apply as such. Thank U,
In a conjugate heat transfer problem your wall boundary condition is coupled. Did you remove all the solids and convert it into a non-conjugate problem?

If it is decoupled, then you just need to specify the appropriate BC, wall temperature, wall heat flux, heat transfer coefficient, etc.

April 19, 2017, 08:25
#12
New Member

RAG
Join Date: Mar 2017
Posts: 5
Rep Power: 9
Quote:
 Originally Posted by LuckyTran In a conjugate heat transfer problem your wall boundary condition is coupled. Did you remove all the solids and convert it into a non-conjugate problem? If it is decoupled, then you just need to specify the appropriate BC, wall temperature, wall heat flux, heat transfer coefficient, etc.
Hello Lucky Tran,

Thanks for your reply, if I am not wrong "decoupling" and "coupling" means it's the solution method in pressure based solver!!??

and regarding my problem, while extracting the fluid volume in hypermesh, I removed all the solid material and the resulted one is hollow, bounded by a surface with inlet and outlets and meshed using tet elements.

while solving in ansys, I am not getting proper temperature profiles of water inside the domain as the surface temperature of "wall-heat" in my problem is 426K and fluid temperature at the inlet is 393K, at an operating pressure of 1.1 bar and I have problems like

1. continuity equation is not converging, when I decrease courant number to 20/15 it is becoming a straight line at 1.2e-1 other residuals are converging up to e-4 ( I checked mesh quality, boundary conditions)

2. I want to study nucleate boiling in it, but still unable to figure out the problem

3. I am confused regarding the quality of the mesh for turbulent flow analysis with CHT, my model quality is 5.085e-2 -- min ortho in fluent, and the solution initialization going up to e-7, but have convergence problem whether the quality is fine or not,

4 I tried doing nucleate boiling for a simple model, using the Eulerian boiling model with phases water and water vapor but my solution getting diverged even before 1st iteration, what might be the problem?

Sorry, for a bit lengthy description,

Thank U,
__________________
d_Rag

 October 22, 2019, 09:59 #13 New Member   Join Date: Oct 2019 Posts: 1 Rep Power: 0 Hello, i am a beginner in ansys fluent.let me expalin my Task first. I Need to find the temperature Distribution along the fire brick storage medium. for that a fire brick medium is heated for 50 hrs using heating coil. after the heating hours air is passed through the medium. How can i create a time dependent heating coil

October 29, 2019, 07:08
#14
Member

sudhir
Join Date: Mar 2009
Location: india
Posts: 65
Rep Power: 17
you could create a UDF input for heat flux with respect to time. UDF's are bascially C language structured file. there are lot of docs available online to do this..

Quote:
 Originally Posted by MOOLAYIL Hello, i am a beginner in ansys fluent.let me expalin my Task first. I Need to find the temperature Distribution along the fire brick storage medium. for that a fire brick medium is heated for 50 hrs using heating coil. after the heating hours air is passed through the medium. How can i create a time dependent heating coil

 Tags conjugate, coupled, fluent, fluid, solid