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
15 nov. 2010, 14:50 UTC−5
Hi
For me with a CAD import and SolidWorks Livelink, I can read in 3D Solidworks, Parasolid, STEP, and IGES files. For 2D its limited to dxf. But 3D dxf is read in X-Y format.
However you must be sure you save the correct entities to in the CAD files, as these files can contain only volumes, or surfaces or only lines, or a mixture, and COMSOL cannot always reconstruct solids form simple lines, without help. YOu have a few options on the CAD file save side, as well on the COMSOL side. Finally the repair tolerance must be set accordingly to allow small features i these are wanted or needed.
Now concerning mixing Solids and shell, beams or trusses, you have already a few threads thereon in the FORUM, try a search. Basically you must ensure that you link these correctly, as solids have no rotational degrees of freedom defined by default, and the other three have rotations defined. Furthermore, there seem to be a issue on the material definitions. You should not use the same material for 3D and for 2D physics, rather duplicate the materials and use them separately for 3D solids and the other for the 2D. Apart from this workaround it'e OK for me
--
Good luck
Ivar
Hi
For me with a CAD import and SolidWorks Livelink, I can read in 3D Solidworks, Parasolid, STEP, and IGES files. For 2D its limited to dxf. But 3D dxf is read in X-Y format.
However you must be sure you save the correct entities to in the CAD files, as these files can contain only volumes, or surfaces or only lines, or a mixture, and COMSOL cannot always reconstruct solids form simple lines, without help. YOu have a few options on the CAD file save side, as well on the COMSOL side. Finally the repair tolerance must be set accordingly to allow small features i these are wanted or needed.
Now concerning mixing Solids and shell, beams or trusses, you have already a few threads thereon in the FORUM, try a search. Basically you must ensure that you link these correctly, as solids have no rotational degrees of freedom defined by default, and the other three have rotations defined. Furthermore, there seem to be a issue on the material definitions. You should not use the same material for 3D and for 2D physics, rather duplicate the materials and use them separately for 3D solids and the other for the 2D. Apart from this workaround it'e OK for me
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
16 nov. 2010, 03:27 UTC−5
Hi Ivar,
Thanks for replying. I have tried to search for related threads on coupling solid and shell structure. But there is no explicit example to teach one how to tie the two physics together. I would be grateful if by any chance you could provide a very simple example file of how to tie these two together. I'm rather new to comsol.
For the cad import module, I just cannot import any other file type except parasolid file format. It keeps saying 'licensing errro -5 no such feature exists in cadreader'. Maybe I have to contact support to find out whether there is anything wrong with my license. It does'nt happen on 3,5a and 4.0.
thanks in advance.
Hi Ivar,
Thanks for replying. I have tried to search for related threads on coupling solid and shell structure. But there is no explicit example to teach one how to tie the two physics together. I would be grateful if by any chance you could provide a very simple example file of how to tie these two together. I'm rather new to comsol.
For the cad import module, I just cannot import any other file type except parasolid file format. It keeps saying 'licensing errro -5 no such feature exists in cadreader'. Maybe I have to contact support to find out whether there is anything wrong with my license. It does'nt happen on 3,5a and 4.0.
thanks in advance.
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
16 nov. 2010, 09:47 UTC−5
Hi
I can only indeed suggest to contact support for such versioning issues, I'll try to reply further on the mixing solid-beam issue, but I need a day or two, sorry slightly skeezed these days ;)
--
Good luck
Ivar
Hi
I can only indeed suggest to contact support for such versioning issues, I'll try to reply further on the mixing solid-beam issue, but I need a day or two, sorry slightly skeezed these days ;)
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
16 nov. 2010, 13:28 UTC−5
Thanks Ivar,
Look forward to your reply soon.
Thanks Ivar,
Look forward to your reply soon.
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
16 nov. 2010, 15:34 UTC−5
Hi,
to start with
The main issue is to link the beam angles "thx, thy, thz" to the solid that has no default angular dependent variables. I nomally use the curl of the displacement field, that is valid for solid small deformation (small angles up to some degrees) i.e. (if I remember right)
Thx = 0.5[rad]*(wY-vZ)
Thy = -0.5[rad]*(wX-uZ)
Thz = 0.5[rad]*(vX-uY)
pls check the sign carefully, I easily mix them up ;)
These angles defined on the interface (or an average of a boundary interfacing the beam) are to be used as prescribed displacement for the Beam physics on the interface.
This links the solid to the beam, but you need the other direction too, this you can get via a weak constrsaint, but I do not rememnber the correct formula, have to find it back ;)
I wounder if there is not some examples on the model exchange ?
Thest this out on a simple case that you know or can esily check analytically, you will soon catch the trick
--
Good luck
Ivar
Hi,
to start with
The main issue is to link the beam angles "thx, thy, thz" to the solid that has no default angular dependent variables. I nomally use the curl of the displacement field, that is valid for solid small deformation (small angles up to some degrees) i.e. (if I remember right)
Thx = 0.5[rad]*(wY-vZ)
Thy = -0.5[rad]*(wX-uZ)
Thz = 0.5[rad]*(vX-uY)
pls check the sign carefully, I easily mix them up ;)
These angles defined on the interface (or an average of a boundary interfacing the beam) are to be used as prescribed displacement for the Beam physics on the interface.
This links the solid to the beam, but you need the other direction too, this you can get via a weak constrsaint, but I do not rememnber the correct formula, have to find it back ;)
I wounder if there is not some examples on the model exchange ?
Thest this out on a simple case that you know or can esily check analytically, you will soon catch the trick
--
Good luck
Ivar