[Nek5000-users] n2to3 to create a tube

nek5000-users at lists.mcs.anl.gov nek5000-users at lists.mcs.anl.gov
Mon Apr 11 08:57:40 CDT 2016


Hi Neks,

I am trying to use n2to3 to create an axisymmetric mesh. I have created 
a 2D mesh using gmsh and it works. Now, the idea is to extrude the 2D 
mesh to create a tube (circular sweep). However, n2to3 doesn't seem to 
work.

In the 2d file the geometry is rectangular with SYM boundary conditions 
in the x axis. Is it possible to use n2to3 to create the tube, or half 
of the tube? If it is possible what are the files that I need? I suppose 
that with the gemoetry_2d.rea file should be enough, shouldn't it?

I realized that:

If I set the number of planes to, let's say -10, and
  input z min:
0
  input z max:
90

I have problems with genmap

NOTE: smaller is better, but generous is more forgiving for bad meshes.
0.2
  reading .rea file data ...
  start locglob_lexico:           8       47680      381440  
0.20000000000000001
  locglob:           1           1      381440
  locglob:           2         666      381440
  locglob:           3        2536      381440
  locglob:           1       51469      381440
  locglob:           2       51644      381440
  locglob:           3       51644      381440
  locglob:           1       51644      381440
  locglob:           2       51644      381440
  locglob:           3       51644      381440

      1     2     4  Matrix:  SELF!!
        1 SELF!!    11362    11361    11362    11361
        2 SELF!!     9722     9721     9729     9728
  cont: SELF!!           1   ??

  ABORT: SELF-CHK            1           5           1           0
  Try to tighten the mesh tolerance!

            0  quit


However, if I increase the number of planes, let's say -80, it seems 
that genmap also works. But, when I run the numerical simulations and I 
try to plot some snapshots, I cannot see anything. It seems that there 
is a problem with the connectivity. What I am missing?

Thanks in advance.
Cheers
SL




More information about the Nek5000-users mailing list