1
Mechanical properties balance in novel Z-pinned sandwich panels: -
out-of-plane properties
Andrea I Marasco
1
, Denis D R Cartié
1
, Ivana K Partridge
1
and Amir Rezai
2
1
School of Industrial and Manufacturing Science, Cranfield University, Bedford, UK,
2
BAE SYSTEMS ATC, Bristol, UK
ABSTRACT
The paper presents the first complete set of test results obtained from quasi-static
measurements of the out-of-plane tension, shear and compression properties of novel X-
Cor™ and K-Cor™ flat sandwich panels. The cured panels were composed of 0.75mm
thick quasi-isotropic IM7/8552 skins, separated by 12.7 mm thick Rohacell
®
foam core
containing 0.51 mm diameter carbon fibre pins (Z-Fiber
®
), arranged in a truss pattern at
pin angle of either 22° or 30° to the vertical. To obtain a suitable baseline comparison,
the equivalent set of properties was measured for Nomex honeycomb core panels
sandwiched by the same composite skins. The novel Z-pinned cores are found to
exhibit higher specific stiffness than conventional sandwich cores, but lower strength.
Keywords List
X-Cor
TM
- K-Cor
TM
sandwich structures; D-Mechanical testing; A-3-Dimensional
reinforcement; A- Polymer-matrix composites (PMCs)
INTRODUCTION
X-Cor
TM
and K-Cor
TM
sandwich panels
These novel polymer composite structures consist of composite skins separated by a
layer of Rohacell
®
foam into which Z-Fiber
®
pins (thereafter referred to as Z-pins) are
Composites Part A: Applied Science and Manufacturing, 37, pp295-302, 2006
Submitted to Composites Part A – revision March 2005
2
inserted at a specific angle to form a truss. The version of Z-Fiber
®
used here is a cured
pultruded carbon fibre/bismaleimide rod. Initially the pins inserted in the foam extend
beyond each surface of the foam, for a so called “reveal length”. The core at this stage is
called a X-Cor
TM
preform (see Fig.1a). For the X-Cor
TM
sandwich construction this
preform is then pressed between two uncured composite skins, the pins enter into the
surfacing prepreg plies and create a mechanical fastening between the core and the skin,
without the need for any adhesive (see Fig.1b).
In the case of the K-Cor™ preforms, the Z-pins which extend beyond the foam surface
are only partially cured. They can therefore be folded back under the action of moderate
heat and pressure, flush with the foam surface (see Fig.2). The required sandwich panel
skins, which can be pre-cured composite or metallic plates, are then usually adhesively
bonded onto the core. The additional heat supplied in the bonding process serves to
complete the cure of the Z-pins. Further detail of the manufacturing processes involved
in the production of the preforms can be found in a recent review [1]. The possibility of
producing these preforms to net shapes and required pinning densities is attractive
where complex sandwich construction is required [2]. The Rohacell
®
foam is a ‘closed
cell’ type and is highly resistant to ingress by water. This confers a clear advantage to
these sandwich panels in situations where water absorption and freeze-thaw cycles may
otherwise present a problem.
Out-of-plane properties of sandwich panels
Relatively little experimental work has been carried out so far on the above described
novel sandwich materials. The earliest published work by Vaidya and colleagues was
prompted by considerations of whether these new materials could completely replace
traditional honeycomb sandwich materials in aerospace applications and consequently it
Submitted to Composites Part A – revision March 2005
3
concentrated on their resistance to impact loading and compression after impact [3 - 6].
The basic conclusion was that the mechanical properties studied were strongly
influenced by the angle at which the pins had been inserted into the foam core, with an
angle of 10° exhibiting better in out-of-plane indentation resistance than equivalent
samples with insertion angle of 20°. In a later study O’Brien and Paris found the
strength of the bond between the core and the skins to be the limiting factor under
uniaxial tension, three-point bending and combined tension and bending loading
conditions [7]. The work reported here was carried out in the context of considering a
wide spectrum of applications for these sandwich structures; automotive and naval as
well as aerospace. For this reason the emphasis was placed on obtaining basic quasi-
static properties in the first instance and on their comparison to the response of Nomex
honeycomb panels under equivalent modes of loading.
Data on the out-of-plane properties of a sandwich construction or of a type of core are
required for design purposes. A sandwich structure is usually designed to bear out-of-
plane shear stresses caused by bending moments acting on the skins. The overall
bending deformation is dominated by the deformation of the skins while the core
stabilises the faces against global and local buckling.
The out-of-plane tension test characterises primarily the strength of the core-skin
interface. The out-of-plane shear test characterises the core behaviour as a part of the
whole sandwich structure, as the skins play a minor part under this loading condition.
The out-of-plane compression test determines the compression stiffness and strength of
the core. It would be expected that the mechanical properties of the X/K-Cor™
products are mainly influenced by the Z-pin densities used, the pin insertion angle and
the pin lay-out in the core. The angle of the pins in the truss will alter the balance
Submitted to Composites Part A – revision March 2005
4
between the shear and compression properties [1-2]. The testing reported here was
carried out to explore and quantify this balance.
MATERIALS AND METHODS
Table 1 summarises the types of cores used in the study, together with their relevant
attributes. The ‘hollow’ X-Cor and K-Cor were obtained by chemically removing the
Rohacell foam, by a process which does not affect the Z-pins. These hollow cores were
then combined with 6-ply skins of IM7 tape, prepregged with 8552 resin, in quasi-
isotropic lay-up configuration, and cured to a nominal final skin thickness of 0.75mm.
The film adhesive used to co-cure the prepreg skins onto K-Cor preform was Redux 322
and a 180° C cure with a 120° C dwell was used.
In the tests in which the sandwich specimen facings were required to be bonded to other
metal parts the adhesive used was Redux 420A Araldite and the best performance was
obtained for 150 minute cure at 70°C under slight pressure. The surfaces to be glued
were abraded with sandpaper and cleaned with acetone.
Out-of-plane tension test
The test was carried out according to ASTM C297 standard, the load being transmitted
to the sandwich through thick loading blocks bonded to the sandwich skins. These
loading blocks must be sufficiently stiff to keep the bonded facings flat under load.
Accurate alignment of the specimen and of the fixtures is critical. The schematic of the
test setup is shown in Fig.3. The tests were carried out at a constant crosshead speed of
0.5mm/min. Failure of the bond between the loading blocks and the facings is not
considered a valid failure for the purpose of this test.
Submitted to Composites Part A – revision March 2005
5
The out-of-plane tensile stress
σ
is calculated by dividing the load P by the effective
area of the specimen A
eff
for the pinned cores (see Results and Discussion section) or by
the original cross-sectional area for the Nomex. The elastic modulus is determined from
the slope of the linear elastic portion of the load-extension curve. All the results
presented here have been corrected for the compliance of the test machine and of the
fixtures.
Out-of-plane shear test
The test method used, ASTM C 273, allows the determination of shear properties of
sandwich constructions or cores, associated with shear distortion of planes parallel to
the facings. As for other shear tests developed in recent years the objective is to
maximise shear stress and minimise extraneous induced stresses in the specimen [8].
The test configuration does not produce a pure shear stress state in the specimen, but the
specimen geometry is prescribed so as to minimise secondary stresses [9]. The test can
be conducted on a core bonded directly to the loading plates or on the sandwich with its
skins bonded to the plates. For the pinned cores the presence of the skins is necessary to
guarantee representative constraint at pin ends. Fig.4 shows the fixture with all the
main parts labelled. According to the test protocol the specimen should have a thickness
equal to the thickness of the sandwich and a width not less than 50mm. The requirement
on the length in order to minimize secondary stresses is to be at least 12 times the
thickness and having the line of action of the direct tensile force passing through the
diagonally opposite corners of the sandwich, as shown in Fig.4. The nominal
dimensions chosen for the specimen are listed in Table 2.
