Buckling of honeycomb composite panel

[Skysurfer76]

Hello All!

I analyze airstructure (elevator) which coitains among other composite sandwich panels. Layup is balanced and symmetrical. In Sol.105 I obtain very strange forms of buckling (see 1.jpg)

http://files.engineering.com/getfile.aspx?folder=f0e0161f-06e5-4c3e-bdd3-b717e6fead99&file=1.JPG

After some investigation I discovered that this phenomena occurs as I set G1Z and G2Z of composite materials to definite value. After I had set these values to blank field I obtained quite realistic form (see 2.jpg).

http://files.engineering.com/getfile.aspx?folder=932d3645-4200-48c8-bf58-42358a53800b&file=2.JPG

I checked this phenomena also with a simple rectangular specimen. This occurs exactly in the same way.
Moreover result with G1Z and G2Z set to blank corresponds with analytical formulas (which is not surprising though)
So question is: is this result with G1Z and G2Z set to blank realistic and beleivable?
Or just this strange form of buckling indicates some problem with structure (like wrinkling, dimpling etc.)

Thanks

 

[ESPcomposites]

How is the core modeled?

What were you setting the G1Z and G2Z as? Was that for the facesheet or the core?

It is possible that transverse shear deformation of the core can significantly affect the result (depending on many variables). If you don't account for it, you may not get an accurate result.

You should be able to model the simple rectangle properly. If you can't get a good result, then you are probably doing it wrong.

You can check it against this solution:

http://www.espcomposites.com/software/eStability.html

Brian

http://www.espcomposites.com

 

[Skysurfer76]

It is modelled with PSHELL element, LAMINATE property. The core is just on layer in this laminate. This phenomena occurs when G1Z and G2Z sre set either for core or for facesheets, simultaneously as well.
Thanks for the link!

 

[SWComposites]

It is probably a shear crimping buckling mode, which occurs when the core thru-thickness shear stiffness is low. Essentially it is zero wavelength buckling mode. If you set G1z, G2z blank you will miss the shear crimping failure mode, unless you make a separate calculation by hand using the local applied in-plane loads on the panel.

 

[ESPcomposites]

Good point, I agree that it is probably shear crimping. It wouldn't be dimpling since you would have to model the actual honeycomb to witness that. Wrinkling shouldn't behave like this (at least not for standard designs).

My software produced a similar result if the core transverse shear stiffness is low.

Brian

http://www.espcomposites.com

 

[Skysurfer76]

Brian,

I try to download ESP solver, but unfortunately zip file is corrupted

 

[ESPcomposites]

Weird. It works fine on my end and for others as well. You can contact me directly as well with more a more detailed description of the problem.

Brian

http://www.espcomposites.com