[GiDlist] 3D structured mesh with tetrahedron
Posted: Thu Aug 02, 2007 11:17 pm
Hi,
I got a surprising results during meshing of a simple 3D rectangular plate
with tetrahedrons. The mesh is structured according to a uniform distribution
of elements along the three cartesian directions. If the option Normal is
specified as the Quadratic element option, I got 600 tetrahedrons and 1155
nodes. If I specify Quadratic, I got 2400 tetrahedron and 678 nodes.
This is totally different from the behaviour during meshing using hexahedron.
The same number of element is obtain for both Normal and Quadratic option.
Only the number of nodes is modified.
After some search in the preference menu, I switch off the option Symmetrical
structured tetrahedra in the mesh sub-menu. This change as fix the problem.
My question is:
- Why this option only applied to Normal interpolation and is inactive for
Quadratic and Quadratic9
A similar option (Symmetrical structured triangle) performs a good behaviour
on both Normal, Quadratic and Quadratic9 interpolation for triangle.
Thanks,
Daniel Marceau
I got a surprising results during meshing of a simple 3D rectangular plate
with tetrahedrons. The mesh is structured according to a uniform distribution
of elements along the three cartesian directions. If the option Normal is
specified as the Quadratic element option, I got 600 tetrahedrons and 1155
nodes. If I specify Quadratic, I got 2400 tetrahedron and 678 nodes.
This is totally different from the behaviour during meshing using hexahedron.
The same number of element is obtain for both Normal and Quadratic option.
Only the number of nodes is modified.
After some search in the preference menu, I switch off the option Symmetrical
structured tetrahedra in the mesh sub-menu. This change as fix the problem.
My question is:
- Why this option only applied to Normal interpolation and is inactive for
Quadratic and Quadratic9
A similar option (Symmetrical structured triangle) performs a good behaviour
on both Normal, Quadratic and Quadratic9 interpolation for triangle.
Thanks,
Daniel Marceau