[Gmsh] Hex-Mesh problem

[Martin.Raedel at dlr.de]

Hi Christophe,

thanks for the reply. My mesh however does not require to be purely hex-based. There may also be some prisms as shown in the attached example. The nodal positions and elements of the two “layers” in the front and the back are the same. I imagine, the extrusion does the same thing, here based on the back layer. It creates a copy and basically transforms it to the front layer position and then uses the identical node information to create the mesh. Thus, it is possible to create a hex-dominant mesh with the occasional prism in between based on the 2D mesh of the initial layer.

[cid:image001.png at 01D3C52B.A7FA2310]

In my models I basically use the same principle. However, I define all “layers” in my tool and write them as a geo-file. Thus, each layer has its own but identical 2D mesh. Is there a way to use the gmsh-algorithm for connecting layers in an extrusion but with me specifying the mesh of each layer myself (the meshes of both layers remain identical except the node z-coordinate) ?

Best regards
Martin

Von: Christophe Geuzaine [mailto:geuzaine at gmail.com]
Gesendet: Sonntag, 25. März 2018 09:14
An: Rädel, Martin
Cc: gmsh at onelab.info
Betreff: Re: [Gmsh] Hex-Mesh problem


Hi Martin,

General (high-quality) full-hex meshes are not available: it's a current research topic. For high-quality full-hex meshes without extrusion, your best bet is to generate a fully structured grid - see attached file. Otherwise you'll have to either accept a hex-dominant mesh (we are actively working on this) ; or a low-quality hex mesh (using the subdivision algorithms).

Christophe




On 22 Mar 2018, at 18:24, Martin.Raedel at dlr.de<mailto:Martin.Raedel at dlr.de> wrote:

Hello,

are there any new possibilities for this problem using the new OpenCascade-options? Or is there any option to specify the layers of an extrusion by surface meshes?

Regards
Martin

Von: gmsh [mailto:gmsh-bounces at ace20.montefiore.ulg.ac.be] Im Auftrag von Martin.Raedel at dlr.de<mailto:Martin.Raedel at dlr.de>
Gesendet: Mittwoch, 16. November 2016 16:30
An: gmsh at onelab.info<mailto:gmsh at onelab.info>
Betreff: Re: [Gmsh] Hex-Mesh problem

Hello again,

I got the desired result using an Extrusion:

<image001.png>

However, using Extrusion greatly disagrees with my modelling approach, since I create all required geometry (points, lines, surfaces, volumes) already in advance, before calling gmsh. Afterwards, I want to mesh the model with gmsh and  in a third step assign the created mesh to the existing geometry. Is there any way to create a (semi-)structured mesh using volumes already defined in my geo-file?

Best regards
Martin

Von: gmsh [mailto:gmsh-bounces at ace20.montefiore.ulg.ac.be] Im Auftrag von Martin.Raedel at dlr.de<mailto:Martin.Raedel at dlr.de>
Gesendet: Dienstag, 15. November 2016 17:23
An: gmsh at onelab.info<mailto:gmsh at onelab.info>
Betreff: [Gmsh] Hex-Mesh problem

Hello,

I experience difficulties creating a pure or at least Hex-dominant mesh for the model in the text file. I tried a lot of solver options but even with the frontal algorithm and AllHexas subdivision algorithm I cannot get a proper mesh. Volume elements are only created for the left and right volume while the boundary surfaces of the middle volume are meshed but no volumes are created and different errors are thrown by gmsh, e.g.

“Cannot use frontal 3D algorithm with quadrangles on boundary” for frontal 3D

Is there any way that I am missing? Do I somehow have to mesh the middle volume first?

Best regards
Martin Rädel

————————————————————————
Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR)
German Aerospace Center
Institute of Composite Structures and Adaptive Systems | Structural Mechanics | Lilienthalplatz 7 | 38108 Braunschweig | Germany

Dipl.-Ing. Martin Rädel
Telephone +49 (0)531 295-2048 | Telefax +49 (0)531 295-2232 | martin.raedel at dlr.de<mailto:martin.raedel at dlr.de>
www.DLR.de<http://www.dlr.de/>

_______________________________________________
gmsh mailing list
gmsh at onelab.info<mailto:gmsh at onelab.info>
http://onelab.info/mailman/listinfo/gmsh

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://onelab.info/pipermail/gmsh/attachments/20180326/c9cad865/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 26661 bytes
Desc: image001.png
URL: <http://onelab.info/pipermail/gmsh/attachments/20180326/c9cad865/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: geometry_minimal_3D.geo
Type: application/octet-stream
Size: 5649 bytes
Desc: geometry_minimal_3D.geo
URL: <http://onelab.info/pipermail/gmsh/attachments/20180326/c9cad865/attachment-0002.geo>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: geometry_minimal_3D_Extrusion.geo
Type: application/octet-stream
Size: 1997 bytes
Desc: geometry_minimal_3D_Extrusion.geo
URL: <http://onelab.info/pipermail/gmsh/attachments/20180326/c9cad865/attachment-0003.geo>