CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Coupling solid/fluid heat transfer (https://www.cfd-online.com/Forums/fluent/174403-coupling-solid-fluid-heat-transfer.html)

Hoppe July 11, 2016 22:50

Coupling solid/fluid heat transfer
 
1 Attachment(s)
Hello,

I have been trying to model convection between an array of cooling fins and air flow for awhile and am stuck on creating interface boundary conditions. I would like for there to be heat transfer between the solid and the fluid. When Fluent creates the interfaces, it merges my chip base and the cooling fin, and then creates 4 interfaces. I am not sure if I am specifying the boundary conditions correctly. Does anyone have advice on this? When I perform the calculation, I am unable to achieve convergence for continuity (it flattens out with residuals ~0.5) and for momentum the residuals flatten out higher than 10^-3. I have tried multiple ways to get continuity to have better convergence, but have not yet found a way. edit: the energy equation also completely diverges after ~ 100 iterations

Another spot where I may be doing this incorrectly is that I specify a velocity inlet, then call all other outlets pressure outlets with gauge pressure 0 Pa. Is this a good idea?

I have listed out my steps **only** for clarifying my questions above. I understand that it may be too long winded (I believe my error is in the physics set-up though).


Thank you in advance for any advice offered.



My steps are:
In Geometry:
1. Create base of solid by drawing rectangle with dimensions 50 x 50 mm

2. Extrude by 1 mm using “Add Material” operation

3. Create new coordinate system 1 mm high in the z direction, 25 mm in the x and y directions (ie, on the top-middle of the base)
a. Create new sketch:
b. Draw in fin (let’s just say 1 fin for simplicity). I draw in a circle on the xy plane.
c. Extrude circle using the “Add Material” operation.
4. Returning to the original plane, to create fluid domain:
a. Create new sketch
b. Draw large rectangle encompassing original rectangle, with lines 40 mm from original rectangle lines.
c. Extrude new rectangle 80 mm
5. Create a Boolean with the “Subtract” operation
a. Target Body: Fluid
b. Tool Bodies: Fin, base
c. Preserve Tool Bodies: Yes
At this point the model looks like the included image.



In Meshing:
1. Name one fluid Face “Inlet”
2. Name fluid wall that connects to base “fluid bottom”
3. Name all other fluid faces “outlet 1”, “outlet 2”, etc.
4. Name the fin walls “Fin Wall”
5. Name the base walls “Base Wall”


In Physics (Double Precision):
1. Energy – On
2. Viscous – Laminar
3. Materials:
a. Air: Ideal gas
b. Solid: aluminum
4. Mesh Interfaces
a. Create interface between interface zone 1: “base_wall-contact_region-src”,”fin_wall-contact_region-src”, and interface zone 2: “contact_region-trg”(why do
b. Make interface “coupled”
5. Boundary conditions
a. Inlet: velocity of 3 m/s
b. Outlets: pressure outlet with gauge pressure = 0 Pa
c. Fluid bottom: wall
d. I’d *like* to specify the base as having a convection coefficient and power generation (I am not sure how to do this since Fluent creates several walls for the coupled interface).
6. Solution methods
a. Scheme: coupled
7. Solution Initialization:
a. Calculate from inlet

AHF July 12, 2016 03:13

Quote:

Originally Posted by Hoppe (Post 608969)
Hello,

I have been trying to model convection between an array of cooling fins and air flow for awhile and am stuck on creating interface boundary conditions. I would like for there to be heat transfer between the solid and the fluid. When Fluent creates the interfaces, it merges my chip base and the cooling fin, and then creates 4 interfaces. I am not sure if I am specifying the boundary conditions correctly. Does anyone have advice on this? When I perform the calculation, I am unable to achieve convergence for continuity (it flattens out with residuals ~0.5) and for momentum the residuals flatten out higher than 10^-3. I have tried multiple ways to get continuity to have better convergence, but have not yet found a way. edit: the energy equation also completely diverges after ~ 100 iterations

Another spot where I may be doing this incorrectly is that I specify a velocity inlet, then call all other outlets pressure outlets with gauge pressure 0 Pa. Is this a good idea?

I have listed out my steps **only** for clarifying my questions above. I understand that it may be too long winded (I believe my error is in the physics set-up though).


Thank you in advance for any advice offered.



My steps are:
In Geometry:
1. Create base of solid by drawing rectangle with dimensions 50 x 50 mm

2. Extrude by 1 mm using “Add Material” operation

3. Create new coordinate system 1 mm high in the z direction, 25 mm in the x and y directions (ie, on the top-middle of the base)
a. Create new sketch:
b. Draw in fin (let’s just say 1 fin for simplicity). I draw in a circle on the xy plane.
c. Extrude circle using the “Add Material” operation.
4. Returning to the original plane, to create fluid domain:
a. Create new sketch
b. Draw large rectangle encompassing original rectangle, with lines 40 mm from original rectangle lines.
c. Extrude new rectangle 80 mm
5. Create a Boolean with the “Subtract” operation
a. Target Body: Fluid
b. Tool Bodies: Fin, base
c. Preserve Tool Bodies: Yes
At this point the model looks like the included image.



In Meshing:
1. Name one fluid Face “Inlet”
2. Name fluid wall that connects to base “fluid bottom”
3. Name all other fluid faces “outlet 1”, “outlet 2”, etc.
4. Name the fin walls “Fin Wall”
5. Name the base walls “Base Wall”


In Physics (Double Precision):
1. Energy – On
2. Viscous – Laminar
3. Materials:
a. Air: Ideal gas
b. Solid: aluminum
4. Mesh Interfaces
a. Create interface between interface zone 1: “base_wall-contact_region-src”,”fin_wall-contact_region-src”, and interface zone 2: “contact_region-trg”(why do
b. Make interface “coupled”
5. Boundary conditions
a. Inlet: velocity of 3 m/s
b. Outlets: pressure outlet with gauge pressure = 0 Pa
c. Fluid bottom: wall
d. I’d *like* to specify the base as having a convection coefficient and power generation (I am not sure how to do this since Fluent creates several walls for the coupled interface).
6. Solution methods
a. Scheme: coupled
7. Solution Initialization:
a. Calculate from inlet

actually i didnt work with ansys geometry but in gambit , when you create you geometry , there is a option named "zone" . we choose fluid region and solid region . after that when fluent read mesh , it automatically create shadow boundary condition . i dont see something like that in your modeling

Hoppe July 12, 2016 08:44

1 Attachment(s)
Quote:

Originally Posted by AHF (Post 608998)
actually i didnt work with ansys geometry but in gambit , when you create you geometry , there is a option named "zone" . we choose fluid region and solid region . after that when fluent read mesh , it automatically create shadow boundary condition . i dont see something like that in your modeling

I believe that zoning is performed automatically when the mesh is read into Fluent. It creates shadow boundary conditions only when I create a coupled interface.

These are the boundary conditions that are created. All of the ones starting with "coupled1" are created when I select that the interface is coupled.


All times are GMT -4. The time now is 02:52.