Applying rotation BC to a solid model

[Anonymous]

Hi,

Can anyone help us on the following simulation : it is a solid aassembly, on which we have to apply rotation BC to be applied to 0 or free.

I thought about creating a cylindrical model, but on this kind there is only 1 rotation.
Also a sherical could be used, but only 2 can be used.


Thanks for your reply

[Anonymous]

Hi,

I'm not sure I understand your question. Solid elements in Algor do not have rotational degrees of freedom. You mention that you want a free BC? Or do you mean constrain the rotation?

Björn

[Anonymous]

Yes this, is the problem : some parts have their rotation allowed and some parts no ; and since there are no rotation dof, it seems to be difficult to modelize

ALAIN

[shakeel.mirza]

Bonjour Alain,

There is a "Remote Load" option in Algor that you can apply on a surface.

What it does is that creates a node on the coordinates that you'll define and then connect that to all the nodes on the surface that you selected with lines (that you can define as rigid or beam elements later). You are then free to do whatever you like with that node.

In your particular case you can apply your Rot. BC To that node.

Let me know if this responds to your question.

Regards,

Shakeel Mirza

Autodesk France

[S.LI]

To apply rotation BC on solid part, I usually attach several beam/pipe elements to the part, and apply rotation BC/constrains on these beam/pipe elements. The reason is obvious that there is no rotation DOFs on solid elements. As you know, the "remote loads/constrains for rotation" are also applied by similar ways.

[xli]

Generally talking about rotational DOF, there are two different concepts: nodal and structural. Brick element made part does not have nodal DOFs but structural DOFs, which are discussed here to constraint and to apply load etc. in this thread. This is a very general and interesting topic in FEA modeling for long time.

Structural rotation is formed by multiple nodes’ translational displacements, in ideal case, we may build up a relation between a rotation DOF and a group of translational DOFs in perfect positions in the part. Then proper corresponding BCs on translational DOFs would achieve the purpose of restraining the rotational DOF. The idea behind of the current "remote load" in FEA editor that is mentioned in this thread previously is based on it: Using strong beams or rigid element to tie up a common node having rotational DOFs with a group of nodes on a specified brick part surface, it will build the relationship automatically and then rotational constraint or load may apply on that node to obtain “equivalent” effect on brick part.

But in general we refer only the rigid portion deformation as the structural rotation that is mixed with elastic/plastic displacements. So those extra built beams or rigid element (above in “remote load") will over-constraint those surface nodes of the brick part, i.e. that surface will be "rigid" in deformation. Unless you expect results are close to "rigid" there this is not a good approximation.

Because the relationship I talked about need to be a quantity condition about multiple nodes, MPC (Multiple Points Constraint) is a good way to impose such constraint. MPC is currently available in linear stress analysis. When you are able to write down such relation, you can use the feature “user MPC” to build up it.

There are still some further considerations like: how many brick nodes would be involved to the rotational DOF, whole surface or only minimum nodes (at least three none collinear nodes)  that can form the rotation, or some nodes in between that two cases need user to decide based on practice; and shell-brick/beam-brick connection etc.

-xli