Modal analysis with point mass and rotational inertia (Lumped mass approach) in CCX

naveen_rg

Hello all,

Can someone help me around with the modal analysis using lumped mass approach with mass applied on a node which is connected to a other part using rigid tie.

In calculix I'm unable to use rotational inertia and with internal solver I'm unable to use node to surface tie

Victor

Here's an example. The mass's node doesn't appear in the solution but it still exists for the solver.

disla

I would say one could adjust the rotational inertia with an offset of that ref node with respect the axis of rotation. I have done a couple of tests and the natural oscillation frequency of the system varies (Increase of the rotational inertia)

naveen_rg

Thank you Victor,

But I'm clear regarding the mass definition but I'm not sure regarding the inertia definition.

Victor

Sorry about that. For the internal solver, you could use high-stiffness beam, truss, or spring elements to support a remote node with rotational inertia.

disla

Deleted.

Victor

@disla, yea, I agree you can just use mass for rotational inertia, like how reality does it. But be careful with just two of them because it'll behave differently for rotations about other axes.

Does it even make sense to have a remote rotational inertia like I suggested? I think the angular velocity being the same both at the surface and away from it means the position of a zero-translational-mass rotational inertia doesn't matter.

disla

You are right, the other transverse and longuitudinal modes will change value.

I have try extruding a thin layer and assigning the equivalent inertia by means of adjusting that layer density and seems a more accurate substitute. Again, just for torsion.



Not so sure now the rigid body approach is accurate enough. It doesn’t allow that surface nodes to expand freely in radial direction or even rotate more or less circumferentially depending on the radio.