honeycomb modelisation 3

[floch]

Hi,

First, sorry for the mistakes, I'm not a god in english.

I begin in the composite structure. Actually, i try modelling a equivalent sandwich model with Cosmos software.
I get on the equivalent modulus for the skins (carbone,epoxy: 6 layers) but,for the honeycomb (nomex), i don't go the right way about it. Cosmos ask me a poisson ratio and young modulus to perform the analysis, coefficients that i don't know.
I have just Compression modulus and shear mudulus.

How is a equivalent model for a honeycomb on software?

Thank you for your help.

 

[rb1957]

i think you want to model with a "sandwich", or laminate, element. cosmos manual and examples should be able to help.

i think you might get more responses in the FE forum.

good luck.

 

[RPstress]

If modeling honeycomb as a layer in a shell element it's usual to put in very low dummy values for the in-plane properties, as the in-plane properties of honeycomb are arguably unmeasurably low.

Putting in zero is asking for trouble, so it's usual to put in a nominal value such that the honeycomb won't pick up any significant in-plane stress. Try giving it 1-, 2- and 12-direction moduli of maybe 0.1 MPa. As long as your skins have a modulus measured in GPa that should work ok-ish. The poisson's is irrelevent, but it's common to put in a nominal value of about 0.3. (The actual poisson's of a honeycomb is not relevant, but for naked core it's theoretically 1, as the hexagonal cells form a mechanism.)

Keep an eye on the results to make sure that there is no significant in-plane load in the core.

If you're modelling the core as solid elements then you must follow the same philosophy, using elements that permit 3D orthotropy or anisotropy. The 13- and 23-direction shear moduli and the 3-direction direct modulus should all be the correct value, and the 1-, 2- and 12-direction properties should be nominal and very small.

rb1957 is right, an FE forum would be a slightly better place for this thread.

 

[floch]

Thank you for your solutions,

I think I understand the process, the module E1, E2 and the poisson ratio V12 and V23 which does very little about the final result.So, a little value is acceptable,they allow just the feasibility of the analysis by the solveur.

You're right, the FE forum is more appropriate for the topic.


"L'intelligence est le seul outil qui permet à l'homme de mesurer l'étendue de son malheur"
Pierre Desproges

 

[SWComposites]

For h/c core use the following in the FE model:

E1 = E2 = 100 psi (too early in the morning to do the conversion to SI units in my head)
E3 = the compression modulus you have
G12 = 100 psi
G13, G23 = the shear moduli you have
nu12 = nu13 = nu23 = 0