[Gmsh] Using Gmsh as a post-processor

Rami Ben-Zvi rami.ben-zvi at weizmann.ac.il
Thu Oct 21 10:00:06 CEST 2010


Dear Geordie,

Thank you so much for your prompt reply.

   1. I followed the example in
      http://www.geuz.org/pipermail/gmsh/2010/005676.html , and it
      worked fine. However, trying to apply it to my data (from the GUI)
      caused Gmsh to crush. I attach the relevant files: post-1c.msh is
      the mesh (nodes & elements), post-080.msh are the results (at
      nodes & elements) at the 80th iteration, and post-1c.msh.opt is
      the options file. Please see it you can draw the temperature
      distribution along z and a certain polar angle, 162 degrees say,
      for e.g., at the 1D elements Tfluid (views 0 and  8)  and the 2D
      elements Tw (views 19 and 55).
   2. I do not understand the suggestion how to use the "physical
      entity" tags to select and unselect parts of the geometry. Can you
      please demonstrate it, preferably on the files I had sent?

The following questions from my original message were not answered. 
Would you please refer to them as well?

   1. For the former issue, it would be beneficial for my geometry if
      results could be displayed in a cylindrical coordinate system. Is
      it possible? At present I generate an alternative nodes and
      element file, where the nodal coordinates are in r-q-z rather than
      x-y-z (BTW, I have noticed that the elements cannot be input on a
      separate file. Why?), but find it quite inconvenient and wasteful.
   2. Is there a way to scale differently the various coordinates? It
      would be convenient for slender geometries in general, e.g, a high
      length-to-diameter-ratio cylinder.
   3. A tiny-but-irritating problem - whenever I start Gmsh GUI and wish
      to open or merge files, it starts at a certain folder where Gmsh
      executable resides, rather that the folder I start from (where the
      post- files and a link to the Gmsh executable reside). How do I
      customize it to start browsing for these files from the current
      (or other) folder?

/*PS:*/ I also attach the files of the mesh in cylindrical coordinates 
(post-1r.msh) and the associated options file (post-1r.msh.opt).

Thanks a lot!
Rami


Geordie McBain wrote:
> 2010/10/19 Rami Ben-Zvi <rami.ben-zvi at weizmann.ac.il>:
>   
>> Is it possible in Gmsh to display results along line of nodes and elements
>> as x-y graphs (e.g., the solid and fluid nodes and elements temperature
>> fields along an axial line)? At present I do it manually, by importing parts
>> of the output into Excel, which is rather cumbersome. The answer to this
>> question is therefore the most urgent.
>>     
>
> Does the answer at http://www.geuz.org/pipermail/gmsh/2010/005676.html help?
>
>   
>> The geometry of the steam generator we develop is rather inconvenient to
>> display, since it comprises of several concentric cylinders having narrow
>> gaps between each other, and also having bases. In order to display the
>> results, it would be convenient if parts of the model can be unselected, so
>> that only desirable parts temperatures contours, say, are displayed. The way
>> I do it now is by defining many views (around 100), each devoted to a
>> certain group of the nodes or element and a certain field values. Is there a
>> more elegant and simpler way to do it? Are the "Physical Names" properties
>> useful for this? There is very limited documentation of their use in the
>> post-processing part of Gmsh (provided I do not use Gmsh geometry and mesh
>> processors).
>>     
>
> I think one way would be to set the "elementary entity" or "physical
> entity" tag of the elements in question.  You can select visibility by
> either of these.  These are two of the three or four integer tags that
> each element gets in the $Elements block.  Would it be possible to
> modify your output routine to do that?  I don't think you need to
> worry about Physical Names, although it is nicer to refer to "wall"
> than physical entity 2, e.g.
>   
-- 
_______________________________________________________________
Rami Ben-Zvi
Solar Research Facilities Unit           Tel.   +972-8-9343397
Weizmann Institute of Science            Fax    +972-8-9344117
Rehovot  76100,  ISRAEL                  Mobile +972-54-6279767
e-mail: rami.ben-zvi at weizmann.ac.il      Home   +972-8-9349223
_______________________________________________________________

_
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment.html>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: post-080.msh
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment.msh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: post-1c.msh
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment-0001.msh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: post-1c.msh.opt
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment.opt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: post-1r.msh.opt
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment-0001.opt>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: post-1r.msh
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20101021/76eab4ad/attachment-0002.msh>