|
[Sponsors] |
[ANSYS Meshing] Solid block inside a tunnel for smoke production (3D) |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
December 8, 2017, 16:44 |
Solid block inside a tunnel for smoke production (3D)
|
#1 |
New Member
Join Date: Dec 2017
Posts: 3
Rep Power: 9 |
I'm trying to create a good quality mesh for a problem involving smoke dispersion in a tunnel fire.
I have attached the geometry for the tunnel which includes the heat source block (a slice through the original air cavity) which is to be used later for production of the smoke. Obviously, the mesh will need to be fine around the heat source as well as the walls and ceiling and the space between the heat source and the ceiling. Due to the nature of smoke, a coarser mesh could be used at the floor of the cavity and at the outlet (the farthest away wall). I have tried the inflation method but I can't get it right. The mesh always ends up a fine grid around the boundaries and coarse in the middle of the tunnel, where a lot of turbulence will occur. Any advice on how to build this mesh would be greatly appreciated. I can attach more images if it helps. Thank you for everything you do! |
|
December 15, 2017, 14:27 |
|
#2 |
Senior Member
Gwenael H.
Join Date: Mar 2011
Location: Switzerland
Posts: 392
Rep Power: 20 |
Hi Freddy,
This is quite an easy fix. You should create planes on the 5 fluid faces around the heat source. You can then slice the geometry in DM, group the resulting bodies into a single part (select all bodies in DM > rmb > form new part). You will then be able to fully "control" the mesh generation by adding edge sizing with bias to create your BL. Have fun |
|
December 15, 2017, 15:50 |
|
#3 |
New Member
Join Date: Dec 2017
Posts: 3
Rep Power: 9 |
Hi Gwenael, thanks for the advice! I have done something similar since I made the post but still haven't found the solution. I will do as you have described above.
Something I found when I tried it my way - of course it was probably wrong: the new sub-sections in the part added a lot of boundaries in the solver (FLUENT) that I couldn't identify or define with a named selection (named wall-23, wall-45, etc for example). Most of these boundaries were "interfaces". This made setup entirely confusing! Additionally, I couldn't properly edit the solid block as a boundary for smoke production as it was just an interface. Are there any settings or features you think I may have missed there? Thanks again for the advice. Freddy |
|
December 15, 2017, 17:03 |
|
#4 |
Senior Member
Gwenael H.
Join Date: Mar 2011
Location: Switzerland
Posts: 392
Rep Power: 20 |
Hi Freddy,
Well, one thing you need to pay attention to is to put all your bodies inside a single part (in DM select all the bodies > rmb > form new part). As you split your domain, you end up with sub-domains and the interfaces represent the fluid “link” between adjacent faces so that the fluid can flow in between. If you upload your model as a workbench archive I can have a look. |
|
December 15, 2017, 17:39 |
|
#5 |
New Member
Join Date: Dec 2017
Posts: 3
Rep Power: 9 |
I can't seem to attach the file (too big) so I have put it in my OneDrive folder. Let me know if you would rather I upload it somewhere else!
https://1drv.ms/u/s!AsCvXYVS5GeotGMdTygEkcXGw2xV You'll notice that there are so so many extra(?) boundaries if you open in the solver. Thanks |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[blockMesh] 3D laminar flow in a pipe with a solid object inside | clint | OpenFOAM Meshing & Mesh Conversion | 1 | November 28, 2016 13:45 |
Freely solid rotation inside a fluid | Juan Maria | CFX | 0 | April 10, 2008 08:59 |
Howto Sphere solid as a subdomain inside a domain? | jakjak | CFX | 0 | October 25, 2007 00:31 |
block geometry inside fluid domain | jeff | Main CFD Forum | 18 | April 12, 2004 12:37 |
contours with block inside. | Paul | Main CFD Forum | 0 | November 20, 2000 05:20 |