Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
20 juil. 2012, 14:10 UTC−4
Hi
it depends if you have a plane that is defined in the geometry, hence is understood as a boundary for COMSOL will therefore have nodes and mesh elements attached, or it's a Data Set Cut plane cuting the results DB via a virtual plane, in which case there is NO mesh attached to this cut plane only interpoalted variables from the mesh elements that interfer with the "cut plane"
Sometimes to see the mesh inside you need to add a filter with a Boolean varaible of the type (X>0.5), or define a view with some hidden domains to reveal the internal boundaries
--
Good luck
Ivar
Hi
it depends if you have a plane that is defined in the geometry, hence is understood as a boundary for COMSOL will therefore have nodes and mesh elements attached, or it's a Data Set Cut plane cuting the results DB via a virtual plane, in which case there is NO mesh attached to this cut plane only interpoalted variables from the mesh elements that interfer with the "cut plane"
Sometimes to see the mesh inside you need to add a filter with a Boolean varaible of the type (X>0.5), or define a view with some hidden domains to reveal the internal boundaries
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
21 juil. 2012, 05:55 UTC−4
I'm using a data set cut plane... now i understood! thanks!
Monday at work i'll try to use the method described.
Sorry but i'm new to comsol and in FEM softwares like deform and forge the approach is little bit different :)
I'm using a data set cut plane... now i understood! thanks!
Monday at work i'll try to use the method described.
Sorry but i'm new to comsol and in FEM softwares like deform and forge the approach is little bit different :)
Ivar KJELBERG
COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
21 juil. 2012, 06:29 UTC−4
Hi
I agree that COMSOL has a more modern and logical approach to Physics simulation (before starting to think FEM), but for those of us having worked extensively with "older" FEM programmes, approaching the simulation issue purely from the FEM theory (math before physics, or mesh before physics), we are often disturbed when starting with COMSOL. Until we get the catch of the COMSOL approach (now also used by other FEM programmes) Geometry - Physics - BC/loads - (Mesh = discretization as for signal sampling) - Solving and Postprocessing.
I have now learned to build my physics and load models purely on the geoemtry or Entity base, meshing is the last step of "sampling" before solving, and for deciding theoptimum mesh density one must have an idea of the solution, hence loop a couple of times mesh - solve - postprocess
--
Good luck
Ivar
Hi
I agree that COMSOL has a more modern and logical approach to Physics simulation (before starting to think FEM), but for those of us having worked extensively with "older" FEM programmes, approaching the simulation issue purely from the FEM theory (math before physics, or mesh before physics), we are often disturbed when starting with COMSOL. Until we get the catch of the COMSOL approach (now also used by other FEM programmes) Geometry - Physics - BC/loads - (Mesh = discretization as for signal sampling) - Solving and Postprocessing.
I have now learned to build my physics and load models purely on the geoemtry or Entity base, meshing is the last step of "sampling" before solving, and for deciding theoptimum mesh density one must have an idea of the solution, hence loop a couple of times mesh - solve - postprocess
--
Good luck
Ivar