[Gmsh] 3D Periodic Boundary Conditions

Dorin Niculescu niculescu_dorin85 at yahoo.com
Fri Mar 22 11:33:47 CET 2013


Hello Filipe,

Thank you for your quick answer. Yes, i've tried to use Periodic Surface too but the result is even worst, the mesh seems to be the same on all the faces but when you are testing the coordinates you can see that there is a difference (not really big, but it can introduce errors in the solution). To find the corresponding mesh points i had to code a subroutine for testing numbers which allows me to set a tolerance. When i use Extrude and Transfinite Surface i find perfectly corresponding mesh just on the first direction in which i extrude (i've tried to extrude in the other two directions hoping that it will work, but with no result). Is this a bug (i'm using Gmsh v 2.5.1) or is like this that it is supposed to work?

Thank you very much for your time!

Dorin


________________________________
 From: F. A. Portela <felipe.alves.portela at gmail.com>
To: Dorin Niculescu <niculescu_dorin85 at yahoo.com> 
Cc: "gmsh at geuz.org" <gmsh at geuz.org> 
Sent: Thursday, 21 March 2013, 23:34
Subject: Re: [Gmsh] 3D Periodic Boundary Conditions
 

Hi Dorin,

You can add this lines at the end of your .geo to force the opposing faces to have the same mesh:

  Periodic Surface 27 {4,13,11,22} = 19 {-2,14,-9,18};
  Periodic Surface 6 {2,3,4,1} = 28 {9,10,11,8};
  Periodic Surface 15 {14,8,13,1} = 23 {18,-10,22,-3};

However it seems to me that by making the faces transfinite you should not require this...

Cheers,
Felipe



--Felipe Alves Portela
MSc student in Aerospace Eng. at TU Delft
http://www.linkedin.com/in/felipealvesportela 


On 21 March 2013 11:28, Dorin Niculescu <niculescu_dorin85 at yahoo.com> wrote:

Hello all,
>
>
>I am trying to implement periodic boundary conditions for my 3D CFD code and i have troubles generating matching points on the opposite faces of a simple cube. I've used Extrude and Transfinite Surface as you can see below but the points seems to match perfectly only for the zyplan. Is there a way to make them match on the other two plans too?
>
>
>cl = 20.0;
>Point(1) = {-50, 50, 50, cl};
>Point(2) = {-50, -50, 50, cl};
>Point(3) = {-50, -50, -50, cl};
>Point(4) = {-50, 50, -50, cl};
>Line(1) = {1, 2};
>Line(2) = {2, 3};
>Line(3) = {3, 4};
>Line(4) = {4, 1};
>Line Loop(5) = {1, 2, 3, 4};
>Plane Surface(6) = {5};
>Extrude {100, 0, 0} {
>  Surface{6};
>}
>Extrude {0, 0, -100} {
>  Surface{15};
>}
>Extrude {0, 100, 0} {
>  Surface{19};
>}
>Transfinite Surface {6} = {1, 2, 3, 4};
>Transfinite Surface {28} = {5, 6, 10, 14};
>Transfinite Surface {15} = {1, 2, 6, 5};
>Transfinite Surface {23} = {4, 3, 10, 14};
>Transfinite Surface {19} = {2, 6, 10, 3};
>Transfinite Surface {27} = {1, 5, 14, 4};
>
>
>
>
>Thank you verry much for your time.
>
>Dorin Niculescu
>_______________________________________________
>gmsh mailing list
>gmsh at geuz.org
>http://www.geuz.org/mailman/listinfo/gmsh
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/gmsh/attachments/20130322/2688ca18/attachment.html>


More information about the gmsh mailing list