[Gmsh] Gmsh

Mikhail Artemyev artemiev.mikhail at gmail.com
Mon Jul 28 02:19:04 CEST 2014


Hi Ali,

Geometry definition and meshing are different processes, and Gmsh is rather
a meshing tool. That said, it has quite reach tools for geometry
definition. They are weaker than in mid-size CAD program though. So, if you
have a more or less simple model - you can use Gmsh for both geometry
definition and mesh building. For complex models one can use a third-party
CAD software to define a geometry, and then use Gmsh only for meshing.
That's why I supposed that the first model with spherical grains could be
modelled in Gmsh, whereas the second one - with grains of complicated
shapes - couldn't.

As for your second question - if you don't want to save the mesh inside the
circles (but the mesh will be built), comment the line describing the
corresponding volumes as "physical":

Physical Volume (t) = thehole ;

If you don't want the mesh in the circles be even built, comment the line
(as well as the previously mentioned one):

Volume(thehole) = theloops[t] ;


Best,
Mikhail


On Sun, Jul 27, 2014 at 3:35 AM, Ali Nasab <nrdpali at gmail.com> wrote:

> Hi Mikhail,
>
> Thank you for following up my question.
> I am a bit confused here. You mentioned that "Gmsh can build meshes for
> both cases". You, then stated that " For the second model - I honestly
> think that you'll not be able to define such a geometry in Gmsh". Can you
> please clarify about this sentence? Can I use Gmsh for the second case or
> not?
> I am not sure which one is case 1 or 2.
>
> The t5.geo is a good example. In this example, the mesh is created for all
> of the shapes. Now, can I exclude the circles from the mesh (no mesh inside
> the circles)?
>
> Regards,
> Ali
>
>
>
>
> On Sat, Jul 26, 2014 at 6:22 PM, Mikhail Artemyev <
> artemiev.mikhail at gmail.com> wrote:
>
>> Hi Ali,
>>
>> Gmsh can build meshes for both cases. For the first model you will be
>> even able to define a geometry with Gmsh - look at t5.geo tutorial for
>> example. For the second model - I honestly think that you'll not be able to
>> define such a geometry in Gmsh, so you'll need to use some CAD tool (best
>> of all - based on OpenCascade) and import the model from it into Gmsh in
>> STEP or BRep format.
>>
>> Best,
>> Mikhail
>>
>>
>>
>>>
>>> ??Dear Professor Geuzaine,
>>>
>>> ?I am a research scientist collaborating with NJIT. We are planning to
>>> develop a code to conduct simulations of water flow in pore scale. The
>>> first figure below, represents the porous media. In this figure you can
>>> see
>>> the grain phase in red circles and the void space in white (and black)
>>> color.
>>>
>>> [image: Inline image 1]
>>>
>>> ?
>>> ?I need to be able to create the 3D mesh for 1) only the void space
>>> (ignoring the grain phase) and 2) both the grain phase and the void space
>>> like the image below (grain phase is in light blue and void space in
>>> green).  ?
>>>
>>>
>>> [image: Inline image 2]
>>>
>>>
>>> ?Please let me know if Gmsh is suitable for our purpose.
>>> Please let me know if you need further clarifications.
>>>
>>> ?Regards,?
>>>
>>>
>>
>
>
> --
> Ali Abdollahi-Nasab, PhD
> Center for Natural Resources Development and Protection
> Department of Civil and Environmental Engineering
> Newark College of Engineering
> The New Jersey Institute of Technology
> 323 MLK Blvd, Newark, NJ 07102
> Ph: 267-237-2302
> Email: nrdpali at gmail.com
> http://nrdp.njit.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20140727/915aedbc/attachment.html>