Of course VTU is not working directly.
One should modify fem/src/modules/ResultOutputSolve.F90
Code: Select all
FUNCTION Elmer2VtkElement( ElmerCode ) RESULT ( VTKCode )
SELECT CASE (ElmerCode)
CASE( 827 )
VTKCode = 29 ! VTK_TRIQUADRATIC_HEXAHEDRON 29
That's it, then from Gmsh to Paraview 827 looks like working, still trying to understand why 820 and 827 lloks weird in paraview.... like with a bad order....
Correction GMSH type 12 Elmer type 827 is a VTK TriQuad : code 29
cell represents a parabolic, 27-node isoparametric hexahedron
vtkTriQuadraticHexahedron is a concrete implementation of vtkNonLinearCell to represent a three-dimensional, 27-node isoparametric triquadratic hexahedron. The interpolation is the standard finite element, triquadratic isoparametric shape function. The cell includes 8 edge nodes, 12 mid-edge nodes, 6 mid-face nodes and one mid-volume node. The ordering of the 27 points defining the cell is point ids (0-7,8-19, 20-25, 26) where point ids 0-7 are the eight corner vertices of the cube; followed by twelve midedge nodes (8-19); followed by 6 mid-face nodes (20-25) and the last node (26) is the mid-volume node. Note that these midedge nodes correspond lie on the edges defined by (0,1), (1,2), (2,3), (3,0), (4,5), (5,6), (6,7), (7,4), (0,4), (1,5), (2,6), (3,7). The mid-surface nodes lies on the faces defined by (first edge nodes id's, than mid-edge nodes id's): (0,1,5,4;8,17,12,16), (1,2,6,5;9,18,13,17), (2,3,7,6,10,19,14,18), (3,0,4,7;11,16,15,19), (0,1,2,3;8,9,10,11), (4,5,6,7;12,13,14,15). The last point lies in the center of the cell (0,1,2,3,4,5,6,7).
Code: Select all
top
7--14--6
| |
15 25 13
| |
4--12--5
middle
19--23--18
| |
20 26 21
| |
16--22--17
bottom
3--10--2
| |
11 24 9
| |
0-- 8--1
I do like finite elements!!
Thanks a lot for your advices, i'll try to make it work. When it's done i'll produce a report in order to help other people who would love to work with Full quadratic hexahedron