[Nek5000-users] Setting up a case from scratch
nek5000-users at lists.mcs.anl.gov
nek5000-users at lists.mcs.anl.gov
Wed Nov 28 09:16:05 CST 2012
Try running 'mbsize' on the gmsh file, and check the number of vertices to make sure it's getting all of them. Or, run
mbsize -ll, piping the output to a file, and check one of the hexes to make sure it's 27-noded. If that's the case,
we're all set (modulo how to handle the BC's).
- tim
On 11/28/2012 09:12 AM, nek5000-users at lists.mcs.anl.gov wrote:
> Thanks Tim. Gmsh+moab was my next question and you answered it.
>
> I will try to see if it works. I made a mesh in gmsh with quadratic elements and mbconvert was able to convert it,
> atleast I get no error. Does this mean that nek will be able to use the curved element information contained in the mesh
> ? I see that gmsh can generate upto 5'th order elements (degree 5).
>
> In gmsh I can group boundary faces into sets which are identified by an integer tag, which should help in specifying
> boundary condition.
>
> praveen
>
> On Wed, Nov 28, 2012 at 8:17 PM, <nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov>> wrote:
>
> Note that this should be relatively trivial to do with the MOAB-based version of Nek (MOAB can read gmsh meshes
> already). I'm not sure whether you can specify higher-order hexes in gmsh, but if you can, start by generating
> those (27-node hex). Building and running Nek on MOAB is described in the Nek user's guide and on the nek wiki.
> The examples/moab problem is a MOAB-based one, showing how to do this for a fluid-only problem; moab_conjht
> demonstrates it for a fluid/solid problem. Note that the user documentation on text input is a bit out of date, so
> refer to the moab example rather than the user's guide for that (will take this as an indication I need to update
> the user's guide and wiki).
>
> As for boundary conditions, I'm not sure how this is represented in the gmsh files (whether groups of nodes/faces
> are indicated in the file, or whether a field on nodes/faces/sides is used to indicate this). If the former, these
> should read into MOAB sets marked with NEUMANN_SET or DIRICHLET_SET tags; if the latter, you'll have to interpret
> this data in the .usr file to set the boundary conditions.
>
> MOAB's gmsh reader is not parallelized, so if you want to run this problem in parallel, you'll need to translate
> into MOAB's .h5m file format, then partition, as described in the wiki instructions. Building MOAB builds also
> 'mbconvert', which converts between any file formats read/written by MOAB (for that matter, partitioning with MOAB's
> tool would allow you to read the gmsh file directly and output the partitioning info with the mesh to .h5m).
>
> - tim
>
>
>
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
>
--
================================================================
"You will keep in perfect peace him whose mind is
steadfast, because he trusts in you." Isaiah 26:3
Tim Tautges Argonne National Laboratory
(tautges at mcs.anl.gov) (telecommuting from UW-Madison)
phone (gvoice): (608) 354-1459 1500 Engineering Dr.
fax: (608) 263-4499 Madison, WI 53706
More information about the Nek5000-users
mailing list