Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
About applying fixed constraint in Structural Mechanics
Posted 24 oct. 2010, 14:42 UTC−4 Version 5.2a 4 Replies
Please login with a confirmed email address before reporting spam
I currently encountered a problem of using fixed constraint. Basically, in my model, fixed constraint is applied on part of a surface. This is how the structure was fixed and held in reality. It is just a single geometry without any contact pair issue. The solution is converged and I get a pretty reasonable stress distribution. However, the maximum principal stress is mesh dependent. It exists right at a corner around the fixed constraint. With my workstation computer, I can compute with very fine mesh. But the meshing dependent problem still exists.
Have anyone encountered this kind of problem before? Any method to solve it? The maximum stress is a very important information for me to see how good the design is. But the mesh dependence really makes me suffer.
Thanks
Qiulin
Hello Qiulin Ma
Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.
If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.
Please login with a confirmed email address before reporting spam
Please login with a confirmed email address before reporting spam
I am also having same problem
Is there any solution found for it?
Please login with a confirmed email address before reporting spam
This is "normal" as a fixed constraint means u=v=w=0 that is NO displacements, but this is not really "physical", your part is fixed to something with a non-infinite Young modulus, and a given Poisson coefficient (the latter links perpendicular swell or contraction to the normal compression/traction) so in reality it will allow some minor displacements at the "fixed" boundary, if not you will get "numerical" stress hotspots for sure.
if you want to study "true stress" when applying theoretical loads toa cube of material (in compression, or in shear etc) then you need to use symmetry conditions, or cut your cube by mid-planes and apply *roller" conditions on these middle surfaces (in 3D you need 3 roller conditions for tensile-compression but only 2 roller conditions for shear forces).
You might also apply more sophisticated weak conditions to your "fixed boundary" to keep it "flat" etc
Two more caveats:
if you solve for non-linear geometrical displacement, then you should NOT use symmetric BC as this will give you further artificial numerical stress values, since the non-linear theory adds an anti-symmetric factor to your displacement calculations, hence in collision with a symmetric BC.
If you solve for thermal expansion, be aware that many of the "fixed, or Rigid Connector" BC have now an optional thermal expansion sub-node, that allows to compensate for the local thermal expansion effect on these "fixed" boundaries
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
This is also discussed in
www.comsol.com/blogs/singularities-in-finite-element-models-dealing-with-red-spots/
Regards,
Henrik