MAT9ORT for honeycomb core

Fernando Hernández Alonso_20813

Hi there, 

I am currently working on a rectangular sandwich structure using PCOMPP/MAT8 for the carbon fiber plies and PSOLID/MAT9ORT for the honeycomb core. I'm trying to run an analysis with a simple load applied in one corner and one of the sides of the rectangle fully constrained, just to see if everything works the way it should. However, I keep on getting some very strange displacement results.

I have modelled the core alone, in an attempt to locate the root of the problem, and I have come to the conclussion that it lies within the use of MAT9ORT, for two reasons: 

-In the core model: When using MAT9ORT, the analysis won't even finish and Optistruct displays "ERROR#153" (Note that when running the sandwich analysis, the solver does finish without showing any error codes). However, when using MAT1 everything goes as expected (the analysis finishes running and the displacement obtained makes perfect sense). 

-In the sandwich model: Again, using MAT1 makes the resulting displacement much more like it. 

 

I have already assigned a coordinate system to the PSOLID property (PSOLID -> CORDM options -> USER), which hasn't changed anything.

I attach both models. I would appreciate any help, because I have run out of ideas on how to make it work...

Thank you so much,

Fernando

Nucleo (3D).hm

Núcleo-Laminado.hm

Gildas_GUILLY

Hello Fernando,
Your definition of yout MAT9OR is not complete, in particular if you use it alone in a model.
You defined E1-E2-E3, giving stiffness in tension/compression
But you did not define any shear stiffness (G12=G13=G23=0), meaning your material has absolutely no strength in shear, reason why your model full 3D cannot work (In your hybrid model, the shells will support shear).

Those fields must be set to model correctly your honeycomb.
Those values are important, as in general in a composite sandwish, the honeycomb support most of the shear forces.

Gildas