How to eliminate modes in Ansys modal analysis

[Mohamed_Wael]

I am studying the modal analysis of a mechanism and I am only interested in the transverse modes, I wonder if there is a way to formulate my problem so that the results show only the transverse modes without the torsional and longitudinal ones, this would help in the post-processing a lot.

 

[Dr.D]

While I am not an ANSYS expert in any sense, I doubt that this will be possible as long as you work in ANSYS. One of the "features" of the finite element method is that it captures all types of deformation.

As an alternative, if you avoid ANSYS, and simply formulate your problem as a conventional eigneproblem, you can choose what coordinates you use in the formulation; these will be the only ones that appear in the final result. You can use an eigensolver in Mathmatica, Maple, etc to get your modal results.

 

[Mech_Engineer]

The best way to do this [in FEA] is through boundary conditions. If you apply boundary supports that prevent motion in certain directions, you won't see any modes in that direction.

 

[Dr.D]

It is certainly true that particular boundary conditions will exclude specific modes. To use this idea, however, requires an intimate knowledge before the fact of the eigenvectors that are to be eliminated. This is usually not practical.

 

[Mech_Engineer]

It is certainly true that particular boundary conditions will exclude specific modes. To use this idea, however, requires an intimate knowledge before the fact of the eigenvectors that are to be eliminated. This is usually not practical.

I meant for excluding certain mode shapes in ANSYS. If you aren't interested in modes of a certain kind in an FEA model, one way to exclude them is to apply boundary conditions which don't allow motion in certain directions.

Take for example a square plate with mounting holes at the corners. By constraining the plate at the holes you will still get high-order modes which allow deflection in the middle of the plate between the mounting holes, which might not happen if the plate is bolted to a rigid structure. So to constrain these modes in the FEA model you might apply a boundary condition to the supported surface area rather than just the holes.

 

[Laurry]

There is an option to write an APDL code for post processing. Perhaps the code could check the directional deformations of each mode and
determine modes of interest.