|
[Sponsors] | |||||
|
|
|
#1 |
|
New Member
Join Date: Jan 2013
Location: New Zealand
Posts: 5
Rep Power: 2 ![]() |
Hi everyone,
I'm learning Fluent and I'm trying to model transient heat transfer from a metal cylinder to air by natural convection. I'm having trouble with interfaces and I'm not sure on the best approach, hopefully you can help. I am using Workbench 14.5, DesignModeler, ANSYS Meshing and of course Fluent. Here is what I have done. 1. Create a cylinder(solid) in DM 2. Create a box (fluid) which is larger than, and surrounds the cylinder 3. Boolean subtract of the cylinder from the box whilst preserving the tool body (the cylinder) 4.Created 3 named selections; cylinder, env, hotface. env represents the environment which is air and hotface is one end of the cylinder which will have a heat flux applied to it. 5. I mesh this using the Workbech Meshing program using default settings, a contact region with 3 faces is detected 6. In Fluent, 3 wall boundary conditions are created; hotface, wall-cylinder and wall-env. 7. I apply gravity, incompressible-ideal-gas (as an air property) and a heatflux to hotface. Result: there is heat transfer to the cylinder but none to the environment. Ideas: I have changed the Zones 'wall-cylinder' and 'wall-environment' to interfaces, created a mesh interface using these and got convection happening. Fluent creates 3 walls and a single corresponding shadow wall when I do this. This seems a bit messy as when I create the mesh interface there are 3 faces for 'wall-cylinder' and 9 faces for 'wall-env' (3 from the cylinder and 6 from the box. What is the best way to approach this problem? Thanks in advance
|
|
|
|
|
|
|
|
|
#2 |
|
New Member
Join Date: Jan 2013
Location: New Zealand
Posts: 5
Rep Power: 2 ![]() |
Alright so I figured out that a much simpler way is to form all bodies into a single part in DesignModeler and then coupled walls and shadow walls are created as there is a conformal mesh. Much easier!
|
|
|
|
|
|
|
|
|
#3 |
|
New Member
Join Date: Oct 2012
Posts: 27
Rep Power: 2 ![]() |
You could also mesh only the fluid (do not preserve booleans tool body) an use a shell conduction approach to model conduction in the cylinder.
|
|
|
|
|
|
|
|
|
#4 | |
|
New Member
Join Date: Oct 2012
Posts: 27
Rep Power: 2 ![]() |
Quote:
|
||
|
|
|
||
|
|
|
#5 |
|
New Member
Join Date: Jan 2013
Location: New Zealand
Posts: 5
Rep Power: 2 ![]() |
Many thanks for your advice Markat, so much to learn and sometimes the User's guide doesn't spell out the basic things like this.
Cheers! |
|
|
|
|
|
![]() |
| Tags |
| fluid-solid, heat-transfer, interfaces |
| Thread Tools | |
| Display Modes | |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Heat Flux at Internal walls or Fluid Solid Interface | Mahi | CFX | 3 | October 1, 2012 02:18 |
| goddamnit: how to set up heat transfer from fluid to solid | macfly | FLUENT | 7 | June 16, 2012 11:50 |
| CFX13 Post Periodic interface | EtaEta | CFX | 7 | December 8, 2011 17:15 |
| Wind turbine simulation | Saturn | CFX | 23 | April 16, 2011 07:34 |
| mass flow in is not equal to mass flow out | saii | CFX | 2 | September 18, 2009 08:07 |