Panel Buckling with Fixed Boundary Conditions and Pure Shear

[ngordon]

I have an sandwich with 1 inch core and bi-directional cloth at a 45 deg on top and bottom, Nxy = 1000 lbs/in, and fixed-fixed boundary conditions and the buckling eigenvalues appear very large when compared to a nastran solution.

However, when I do the same analysis with Simple-Simple support, I get a similar answer to nastran.

For an isotropic plate, Roarks would estimate the simple-support BC would be about 2x lower, but hypersizer is showing about a 10x lower eigenvalue for simple support

There is the possibility my nastran model is incorrect, but I am curious if anyone else has come across this?

[James]

I'm curious how you applied the shear load and BC. Those loads and BC are not easy to get right.

You could try to use HyperFEMgen to create the verification FE model.
See: https://hypersizer.com/help_7.3/#FemGen/fg-overview.php%3FTocPath%3DFeatures%7CFEM%2520Interface%7CHyperFEMgen%7C_____1

HyperFEMgen will create a simple smeared model for the honeycomb panel and apply the static and buckling cases. It supports fixed-fixed BC.

I ran a quick study to compare my HyperSizer results with FEA (see attached images). Notice that the FEA EV = 9.06, and the HyperSizer MS = 8.08. To compare to FEA, add '1' to the MS to get the EV (EV = MS + 1).

FEA EV = 9.06
HyperSizer EV = 9.08

Mode shape appears to be the same.

I hope this is helpful.

-James
 



-