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

Coupled meshes

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   August 11, 2005, 11:02
Default Coupled meshes
  #1
Vyatscheslav
Guest
 
Posts: n/a
Is it possible to mesh each part of an aircraft (eg:wings,fuselage,...: decomposition into 6 or 7 parts) and to couple these objects to compute the whole airflow pattern. Any idea, advice, or reference (paper) is welcome.

Thank you in advance. Vyatsheslav
  Reply With Quote

Old   August 11, 2005, 14:36
Default Re: Coupled meshes
  #2
steve
Guest
 
Posts: n/a
Yes it possible. The "CPMATCH" tools exist specifically to join up disimilar meshes.
  Reply With Quote

Old   August 11, 2005, 15:07
Default Re: Coupled meshes
  #3
William Blake
Guest
 
Posts: n/a
It is possible but not at all a good way to do it.

The following link shows a good way to solve your problem: :

http://www.gridpro.com/gridgallery/airplane.html

  Reply With Quote

Old   August 12, 2005, 05:14
Default Re: Coupled meshes
  #4
Hubert Janocha
Guest
 
Posts: n/a
Yes, cpmatch works, but be carefull with high speed / high compressible flows. For this flows you need a very high quality mesh because of the sensitive physics / nature of the flow.

The way William Blake told is the best one (merged vertices between to meshes / blocks), but it is not really suitable in very complex geometries, and it is not a rapid way, but sometimes necessary. The keyword is block structured meshing.

  Reply With Quote

Old   August 12, 2005, 07:08
Default Re: Coupled meshes
  #5
William Blake
Guest
 
Posts: n/a
Gridpro is multiblock structured and the topologies and meshes for the examples in the grid gallery can be downloaded (you have to write an email and ask for the login data).

So this might be a fast start.
  Reply With Quote

Old   August 12, 2005, 07:28
Default Re: Coupled meshes
  #6
Vyatscheslav
Guest
 
Posts: n/a
Thanks for these elements of answer.

But let me precise what Iīm looking for. My work consists roughly of analysing the airflow pattern around the whole aircraft comparing different wings models. Iīd like to mesh fuselage and wings independently, and to couple both grids. That would allow me to mesh only my different wings models and not to have to remesh the fuselage each time.

Do you confirm what you said? Is it a way to do it with StarCD? Is GridPro (free) available on the internet?

Once again, thanks for your help. Vyatscheslav
  Reply With Quote

Old   August 12, 2005, 08:15
Default Re: Coupled meshes
  #7
William Blake
Guest
 
Posts: n/a
It is possible to do it in your way but based on my 10 years of experience with starcd I would not advise you to do it.

GridPro is not free but the most suitable tool to do this kind of work. As long as you only change some values of the geometry like the thickness or the shape of the wing, the diameter of the engine, position of the fuselage ... , remeshing is fully automatic and takes you only some minutes.
  Reply With Quote

Old   August 17, 2005, 02:13
Default Re: Coupled meshes
  #8
Vyatscheslav
Guest
 
Posts: n/a
OK, thank you. And how does it works for the connections of the blocks? For example, if the mesh of the fuselage is done, and if I want to test a new wing: I will import the new CAD file, mesh it, and what about the connection? On the linking plane, how to do to make the points correspond (in fact to avoid to do like on starCD: couples,which implie loss of precision for the code). Also: where can I find some valuable documentation of GridPro, and how much is a license?

Vyatscheslav
  Reply With Quote

Old   August 17, 2005, 06:12
Default Re: Coupled meshes
  #9
William Blake
Guest
 
Posts: n/a
If you mesh in gridpro, you mesh the complete geometry and not parts of it. After changing some part of the geometry you simply run the grid generation for some iterations.

It is also possible to mesh parts and use the "weld" utility to get a 1 by 1 connectivity but I never used that feature. http://www.gridpro.com/products/weld.html

For the licensing costs you should contact them directly. But it is usually not the cheapest code (mainly used by the NASA).

  Reply With Quote

Old   August 17, 2005, 09:42
Default Re: Coupled meshes
  #10
William Blake
Guest
 
Posts: n/a
I tried to send you some tutorials via email but your email doesn't work.

  Reply With Quote

Old   August 17, 2005, 10:31
Default Re: Coupled meshes
  #11
Vyatscheslav
Guest
 
Posts: n/a
Thanks a lot. In fact, itīs .fr and not .com ...

Yes, the weld seems to be a possible solution. When you say connection 1 by 1, do mean itīs possible to mesh a part (eg fuselage), to keep in memory the nodes on the connecting plane, and to use these datas to mesh the other part (eg wing)? I think itīs an important feature, allowing to use higher order of discretisation for the solver.

Other question: itīs not clearly stated on their site, but: when the mesh is done, is there an output file format to solve with starCD.

Vyatscheslav
  Reply With Quote

Old   August 17, 2005, 11:31
Default Re: Coupled meshes
  #12
William Blake
Guest
 
Posts: n/a
I never used weld myself. It should be much easier and faster do create the whole mesh at once again every time you change someting.

Export to StarCD is no problem. You can use Nastran or SrarCD format.

  Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

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
coupled DPM Peter Kolb FLUENT 5 October 4, 2012 10:27
Coupled solver weiss OpenFOAM Running, Solving & CFD 4 April 19, 2005 13:17
PDE coupled !! Doche Main CFD Forum 11 December 4, 2004 03:43
?-->FAS and AMG for coupled solver Harry Qiu FLUENT 0 April 3, 2001 21:08
coupled ben Main CFD Forum 4 January 18, 2001 09:57


All times are GMT -4. The time now is 02:00.