CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

Interface generation

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   March 7, 2001, 14:10
Default Interface generation
  #1
Karl Kevala
Guest
 
Posts: n/a
Hello,

I have a 3-D mesh with a cylindrical shaped interface between two zones (zone 1 = a solid cylinder, zone 2 = a cylinder enclosing zone 1).

Whenever I go to create an interface between the two zones, fluent is creating "ghost" wall boundary zones in the create interface panel.

When I try to display the walls, nothing shows up.

The interfaces between the two zones were made to match exactly when I created the two separate meshes in Fluent. (Even though I created the meshes separately, I was very careful to define coordinates exactly the same for the interface faces -- Creating the meshes simultaneously, using identical faces is in theory possible, but I want to avoid this because it will be extremely difficult and time consuming for my overall geometry, and in any case, I may still have the same problem after spending days if not weeks to do this).

Does anyone know why I am getting ghost walls even though the two interface faces match exactly? Is it ok for me to just delete the walls and run my simulation -- How are these walls used in the computational algorithm?

Any help is greatly appreciated.

Karl
  Reply With Quote

Old   March 8, 2001, 06:58
Default Re: Interface generation
  #2
Flav
Guest
 
Posts: n/a
* I carried out several calculations with interfaces, and walls were automatically created, as in your case. I did not see any problem caused by those "walls", that I can't even display too.

* One solution to remove the interface in your case, would be to merge the nodes when you do the assembly of the differents meshes. I suppose you do it with Tgrid. If so, try Boundary/Nodes. Leave unchecked the "Only free nodes" options. Don't increase the tolerance value too much, that might corrupt the grid (Tgrid will tell you if this happens). Select the faces on which you want to merge nodes. If you select all the faces, that should also work fine. Then Merge. You should be able to see that one of the two coincident faces has been deleted. That means that the 2 faces are now connected.
  Reply With Quote

Old   March 8, 2001, 09:32
Default Re: Interface generation
  #3
Kevala
Guest
 
Posts: n/a
Thanks. I am not clear about your suggestion to merge nodes on the two faces in t-grid. The two interfaces for my meshes are geometrically coincident, however the nodes between the two faces do not line up. Can I still merge the "nodes" in tgrid? Is the merging procedure simply making the two faces geometrically the same and creating hanging nodes on each side of that face?
  Reply With Quote

Old   March 9, 2001, 03:16
Default Re: Interface generation
  #4
Flav
Guest
 
Posts: n/a
You are right, in the case the nodes between the 2 faces do not line up, you cannot merge them. Merge the nodes in Tgrid only if the faces are geometrically coincident AND the nodes are coincident too.
  Reply With Quote

Old   March 13, 2001, 18:45
Default Re: Interface generation
  #5
Raza Mirza
Guest
 
Posts: n/a
I too use non-conformal interfaces on real problems. Just like you I too have done some testing on simple cases where the nodes exactly matched on the two interfaces. And just like you I too got these "ghost" wall zones. They did not seem to affect the results. But I would be interested to know if someone from Fluent development team can explain what is the significance of these wall zones. May be I should call up their customer support.
  Reply With Quote

Old   March 14, 2001, 10:08
Default Re: Interface generation
  #6
Kevala
Guest
 
Posts: n/a
It is a strange problem. If you find anything more about it please let me know. By the way, I had an old case that I had run in fluent/UNS that was a two dimensional sliding mesh. I checked it in fluent 5 and noticed that the ghost walls were present. However, if I remember correctly, when I ran it originally in fluent/UNS the ghost walls were not there. It seems these ghost walls are a problem particular to fluent 5.

I am wondering about a related issue and if you also note it. My present mesh is 3-d hexahedral, and when I create the interface, not only do I get ghost walls but it also seems to create a large number of triangular faces and tetrahedral cells. I notice this when I rewrite the case after making the interface. Any idea about why this happens? Does it happen with your cases?
  Reply With Quote

Old   March 14, 2001, 17:05
Default Re: Interface generation
  #7
Jerry Lim
Guest
 
Posts: n/a
The extra walls are automatically created to hold non-overlapped faces. Not all cases will have "left-over" faces depending on the interface zones. However, there are cases where these walls may initially be empty but get filled as the interfaces slide with respect to each other. There is no impact on the solution if the walls are empty. In the solver loops, they will be visited though but since there are empty, nothing will be done. Hope this helps.
  Reply With Quote

Old   March 15, 2001, 23:20
Default Re: Interface generation
  #8
Raza Mirza
Guest
 
Posts: n/a
I have not paid attention. My mesh is a combination of hexes and tets. When I create the interface and the ghost wall is created, I don't know if my triangular face count and tet cell count go up or not. I can check though.

  Reply With Quote

Reply


Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Wind turbine simulation Saturn CFX 58 July 3, 2020 01:13
RPM in Wind Turbine Pankaj CFX 9 November 23, 2009 04:05
Convective Heat Transfer - Heat Exchanger Mark CFX 6 November 15, 2004 15:55
Replace periodic by inlet-outlet pair lego CFX 3 November 5, 2002 20:09
Latest news in mesh generation Robert Schneiders Main CFD Forum 0 March 2, 1999 04:07


All times are GMT -4. The time now is 06:40.