07-07-2012, 12:49 PM
Buckling of Honeycomb Sandwiches: Periodic Finite Element Considerations
Buckling of Honeycomb Sandwiches.pdf (Size: 1.28 MB / Downloads: 39)
Abstract:
Sandwich structures are efficient
lightweight materials. Due to there design they exhibit
very special failure modes such as global buckling,
shear crimping, facesheet wrinkling, facesheet dimpling,
and face/core yielding. The core of the sandwich
is usually made of foams or cellular materials, e.g.,
honeycombs. Especially in the case of honeycomb
cores the correlation between analytical buckling predictions
and experiments might be poor (Ley, Lin, and
Uy (1999)). The reason for this lies in the fact that
analytical formulae typically assume a homogeneous
core (continuous support of the facesheets). This work
highlights problems of honeycomb core sandwiches
in a parameter regime, where the transition between
continuous and discrete support of the facesheets is
studied.
Introduction
Sandwich structures exhibit very high structural efficiencies
(ratio of strength or stiffness to weight) and, therefore,
are of interest in the use of aerospace structures.
A sandwich consists of two thin load bearing facesheets
glued on a lightweight core that prevents the facesheets
from buckling individually. Quasi homogeneous cores
(e.g., made of foam) and cellular cores (e.g., honeycomb
cores) respectively, are typically used.
Failure modes of sandwich structures, as stated in standard
literature, are: global buckling, shear crimping,
facesheet wrinkling, facesheet dimpling, face/core yielding,
core-face debonding (Plantema (1966), Zenkert
(1995)).
Methods
In this section the homogenization as well as the analytical
buckling formulae for different sandwich failure
modes - global buckling, facesheet wrinkling of sandwiches
with thin and thick cores, facesheet dimpling,
face/core yielding, cell wall buckling under multi-axial
in-plane compression - are considered, and the periodic
finite element models are described.
Discussion
Dimpling can only be expected for very thin facesheets
(in the considered cases approximately tf < 0.1 mm).
High deviations of the analytical dimpling predictions
from finite element results are visible (Figure 14, A), because
the dimpling formula does not take different core
materials into account, i.e., the stiffness of the core material
influences the accuracy of the analytical dimpling
loads. A new analytical dimpling approach has to be developed.
Furthermore, the interaction formula (Equation
(30)) should be improved. In a first step the coefficents
p might be reduced (p =1. . . 3). A comprehensive study
on this topic should be a future task.