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

[GAMBIT] Forcing a new mesh to conform to a previously generated mesh

Register Blogs Community New Posts Updated Threads Search

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   July 2, 2012, 13:27
Thumbs down Forcing a new mesh to conform to a previously generated mesh
  #1
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
The problem that I am really working on is much more complicated, but I find it is easier to explain it with a simpler instance.



Suppose I imported a mesh from a file into GAMBIT (vertical rectangle). I need to attach a new geometry at the top of the original geometry (horizontal rectangle) and mesh it such that the new mesh coincides with the first mesh at their interface. In other words, I want the second mesh to recognize the existence of the nodes in vertical mesh where they intersect, so that the new mesh will use these nodes.

Obviously for this simple case, I know that I could unite the two faces together. However in my actual problem, the imported (meshed) geometry is TOO complicated to both unite and remesh. Also, in this picture, there are only three nodes that connect the two geometries. In my actual problem, there are many of them, and it is extremely cumbersome to select them one by one or even in groups.

How can I force my new geometry's mesh to conform to the imported mesh using Gambit?
Attached Images
File Type: jpg mergemesh.jpg (20.0 KB, 5 views)
pmdelgado2 is offline   Reply With Quote

Old   July 3, 2012, 01:40
Default
  #2
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
The two topologies have to match together, in other words you need to connect them.
In your simple case, you need to split the horizontal rectangle with the 2 upper vertices from vertical rectangle.
Then connect the 2 superposed edges (upper edge from vertical rectangle)
Once it is done, the 2 rectangles are connected and you can mesh your horizontal rectangle.
The 2 meshes will match each other.
I am also handling complicated geometries, and I "glue" old mesh with new one.
But I don't import mesh. I open old dbs, then delete what I don't need, I import new topology (step), finally I do the procedure I explained...
No problem
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   July 3, 2012, 01:52
Default
  #3
Far
Senior Member
 
Sijal
Join Date: Mar 2009
Location: Islamabad
Posts: 4,558
Blog Entries: 6
Rep Power: 54
Far has a spectacular aura aboutFar has a spectacular aura about
Send a message via Skype™ to Far
You can also use non conformal interface in fluent to connect. That would be easy method for complex cases.
Far is offline   Reply With Quote

Old   July 3, 2012, 12:10
Default
  #4
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
@Far: Can you describe how to create a non-conformal interface?
pmdelgado2 is offline   Reply With Quote

Old   July 3, 2012, 12:18
Default
  #5
Super Moderator
 
diamondx's Avatar
 
Ghazlani M. Ali
Join Date: May 2011
Location: Tokyo, Japan
Posts: 1,385
Blog Entries: 23
Rep Power: 29
diamondx will become famous soon enough
let's suppose you want to merge two bodies, of curse they need to share a same surface. you will load the first mesh in fluent. then you click on mesh -» zone -» append case file and select the second body.
You will then click on "mesh interface", give a name to you interface and select the surfaces above and click on apply.
__________________
Regards,
New to ICEM CFD, try this document --> https://goo.gl/KAOIwm
Ali
diamondx is offline   Reply With Quote

Old   August 22, 2012, 17:24
Default How?
  #6
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
Sorry for the long delay in my reply. I saw in your previous message that I will need to "connect" two superimposed edges. Could you elaborate on this? How do accomplish this on gambit?

Quote:
Originally Posted by -mAx- View Post
The two topologies have to match together, in other words you need to connect them.
In your simple case, you need to split the horizontal rectangle with the 2 upper vertices from vertical rectangle.
Then connect the 2 superposed edges (upper edge from vertical rectangle)
Once it is done, the 2 rectangles are connected and you can mesh your horizontal rectangle.
The 2 meshes will match each other.
I am also handling complicated geometries, and I "glue" old mesh with new one.
But I don't import mesh. I open old dbs, then delete what I don't need, I import new topology (step), finally I do the procedure I explained...
No problem
pmdelgado2 is offline   Reply With Quote

Old   August 27, 2012, 04:23
Default
  #7
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
did you fix your problem?
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 27, 2012, 10:21
Default Not yet...
  #8
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
I tried, but I'm not quite sure what you mean by "connecting the two superimposed edges". How do I do this on gambit? (i'm using version 2.4.6)

Quote:
Originally Posted by -mAx- View Post
did you fix your problem?
pmdelgado2 is offline   Reply With Quote

Old   August 27, 2012, 10:28
Default
  #9
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
let say you have 2 cubes with one "common" face
to be sure that the 2 volumes are connected, you can try to move (not copy) one volume. If it is failed, then both volumes are connected (and gambit will explain in the error, that your entity is connected to another one)
If it is successfull then you can imagine that you have 2 superposed surfaces (common faces), and you will be able to connect them with the connect-face tool
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 27, 2012, 11:02
Default It makes sense, but...
  #10
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
The concept makes sense, but the implementation is what I'm struggling with. Where do I find the command to 'connect' two faces together? Other than the union operation, I'm not sure how else to connect them?

Quote:
Originally Posted by -mAx- View Post
let say you have 2 cubes with one "common" face
to be sure that the 2 volumes are connected, you can try to move (not copy) one volume. If it is failed, then both volumes are connected (and gambit will explain in the error, that your entity is connected to another one)
If it is successfull then you can imagine that you have 2 superposed surfaces (common faces), and you will be able to connect them with the connect-face tool
pmdelgado2 is offline   Reply With Quote

Old   August 27, 2012, 11:05
Default
  #11
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
http://aerojet.engr.ucdavis.edu/gamb...connect_discon
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Old   August 27, 2012, 15:33
Default Error
  #12
New Member
 
Paul
Join Date: May 2012
Posts: 23
Rep Power: 14
pmdelgado2 is on a distinguished road
Thanks for your help. I looked at the tutorial and found the connection button that I was looking for. I also tried it on a 2D geometry (connecting edges together) with the mesh/geometry shown in the image I posted earlier. I selected edge 1 (top of the vertical rectangle) and edge 7 (the bottom of the horizontal rectangle) and applied the following command on gambit:

edge connect "edge.1" "edge.7" virtual

I then received the following error:

ERROR: The connected vertices include vertex.7 which has a short edge edge.7 attached to it (edge of length close to connect tolerance). In order to avoid degenerate edge construction include vertex vertex.8 (the other end vertex at the edge) in the connect. (Enable connect with degenerate edge removal.)

Am I approaching this correctly?
Let me know your thoughts.
Thanks,
Paul
pmdelgado2 is offline   Reply With Quote

Old   August 28, 2012, 01:53
Default
  #13
Super Moderator
 
-mAx-'s Avatar
 
Maxime Perelli
Join Date: Mar 2009
Location: Switzerland
Posts: 3,297
Rep Power: 41
-mAx- will become famous soon enough
edges have to be exactly the same: in your case, you need to split the bottom edge (horizontal rectangle) with respect of top edge (vertical rectangle)
This operation will split the top edge in 3 segments (edges). The middle one will be exactly the same as top edge (but will not belong to verticle rectangle).
Then you will have 2 superposed edges, which you can connect
__________________
In memory of my friend Hervé: CFD engineer & freerider
-mAx- is offline   Reply With Quote

Reply

Tags
conforming meshes, gambit mesh, merge face mesh


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
[snappyHexMesh] No layers in a small gap bobburnquist OpenFOAM Meshing & Mesh Conversion 6 August 26, 2015 10:38
[ICEM] Problem making structured mesh on a surface froztbear ANSYS Meshing & Geometry 4 November 10, 2011 09:52
[snappyHexMesh] snappyHexMesh won't work - zeros everywhere! sc298 OpenFOAM Meshing & Mesh Conversion 2 March 27, 2011 22:11
2d irregular grid Remy Main CFD Forum 1 December 22, 2008 05:49
unstructured vs. structured grids Frank Muldoon Main CFD Forum 1 January 5, 1999 11:09


All times are GMT -4. The time now is 14:18.