Buckling cellular structures - negative eigenvalues

[stjc]

I'm investigating buckling of cellular structures and started off with a simple two-cell case which worked as expected. I've started effectively building up combinations of these two cell units but even by the time I get to a six-cell structure I get negative eigenvalue results. Whenever this problem has arisen in the past it has been because I have had insufficient restraints, resulting in a rigid body motion at some point, but that doesn't seem to be the case now. Given that I'm building up from a model which does work as expected and is adequately restrained, I'm confused as to what is going wrong.

Any advice welcome.

Thanks.

 

[johnhors]

Nothing is going wrong !

Negative eigenvalues and their associated eigenvectors are simply valid "mathematical solutions" to the eigen problem and can be safely ignored as they are usually of no real world significance.