Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.
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.
"Axis to use as symmetry plane normal" settings
Posted 2013年11月29日 GMT-5 02:58 Structural Mechanics Version 4.3b 3 Replies
Please login with a confirmed email address before reporting spam
I have some inquiries about the stated subject. Under the postbuckling shell tutorial of the verification models of structural mechanics module, it is stated to that different symmetry edges are assigned different values of Axis to use as symmetry plane normal. What actually does this option do? I do not manage to find its definition in the Comsol Reference Manual.
I also notice there are two coordinate systems available: global coordinate system and local edge system. What is the local edge system necessary and only present in the shell module but not solid mechanics module?
Thank you.
Regards
Goh
Please login with a confirmed email address before reporting spam
I think you refer to the following instruction: "From the Axis to use as symmetry plane normal list, choose 1."
This instruction points out the axis of the selected coordinate system which acts as normal to the symmetry plane. The number of the axis (in the case of the global Cartesian system) is 1-X, 2-Y, 3-Z. So choosing "1" would make YZ be the symmetry plane.
The local edge system for a shell has the axes 1- Tangent to edge, 2 - normal to edge and in the plane of the shell, 3 - normal to shell. The default for a symmetry condition is 2.
There are two reasons why you should not use the symmetry condition as defined by the local edge system (which is convenient):
a) The symmetry plane is not aligned with an axis of the local edge system. This is the upper case in the appended sketch (where YZ is assumed to be a symmetry plane and the shell stretches out in the Z direction.)
b) Due to curvature, there may be a small error in the computed direction. This error will decrease with mesh density, and can often be ignored. This is the lower case in the appended sketch.
Regards,
Henrik
Attachments:
Please login with a confirmed email address before reporting spam
Since symmetry plane is defined on a boundary, why the symmetry plane is not aligned with an axis of the local edge system? (one of the axis of local edge system is tangent to the edge)
Please login with a confirmed email address before reporting spam
N indicates the normal to the shell. The symmetry line is the vertical dashed line. In the case in the lower sketch, the shell normal N is mathematically in the Y direction, but numerically not because of the discretization.
In the sketch, the shell edge in the symmetry plane goes in the Z direction. So the symmetry plane can have any rotation with respect the edge tangent. For most cases, however, the second axis of the edge tangent system coincides with the normal to the symmetry plane (at least mathematically).
Regards,
Henrik