How to evaluate sandwich panel FWC performance thro FEM?

[pcomp]

I am analyzing a cowling of sandwich construction (carbon skins and honey comb core). It is idealized in FEA as shell (pcomp) mesh, presuressied by air loads. How should I evlauate the compressive strength of the core for the pressure loads, through the f06 element forces (Fx, FY, Fxy, Mx, My, Mxy, Qx and Qy) from nastran runs?

 

[rb1957]

i'd've thought that pressure would apply shear to the core, and bending loads onto the faces (tension on one, compression on the other).

 

[SWComposites]

You can't do it with shell elements - there are no thru-thickness normal results. You can either make a model with the core modelled as solids (but in most cases this is gross overkill) or take the shell loads and do a curved beam analysis to determine the thru-thickness normal stresses. But if you have a non-uniform stress field, such as at the top of a core ramp, you are going to have to get some sandwich panel test data with ramps that fail in that mode in order to get a correlation to simple FWC test specimen data (or just be very conservative).

 

[rb1957]

doesn't he get thru thickness shear from the PCOMP card ? i can remembering entering core as one ply in a sandwich panel ...

 

[SWComposites]

yes, you get thru-thickness shear forces from the shell elements, through they are often not very accurate, but you do not get thru-thickness normal (sigma_z) forces/stresses.

 

[rb1957]

yes, but how much out-of-plane normal stress is in the core ? i thought the core is usually shear effective only, and the faces are in-plane effecive only ?

 

[SWComposites]

rb - for a flat panel, yes there is typically very litter out of plane stress. But in a curved panel with bending moments, there are crush/pulloff thru-thickness normal forces, and when there are "kinks" in a facesheet, such as at the top of a core ramp, there are also crush/pulloff thru-thickness normal forces resulting from the change in direction of the in-plane facesheet loads. Now whether the OP has either of these cases I have no idea.

 

[pcomp]

Thanks all for the reply. I have quick question. why flat sandich panel does not show out of plane stress, if uniform pressure applied normal to flat surface?

 

[SWComposites]

"why flat sandwich panel does not show out of plane stress, if uniform pressure applied normal to flat surface?" > it does, it is usually just so low as to be negligible; even 3# Nomex core as a flatwise compression strength above 100 psi.