JEC COMPOSITES MAGAZINE - Issue #88 - April 2014 - Special JEC Americas Spring Edition - 60

BVID

nation growth. For example, 0-40° surface
plies should be avoided for axially-loaded
laminates (see Figure 3). This has been
experimentally verified [7]. Conversely,
burying the load-carrying fibres deep within
the laminate protects them from stiffness
reductions associated with sublaminate
buckling and thus the structural integrity of
the laminate is better maintained.
High-stiffness core plies
Although delamination growth in surface
regions is strain driven, it is usually the stress
carrying capacity of the overall laminate that
matters to the aircraft designer. Maximisation of core stiffness, assuming that the
laminate is subject to uniform end-shortening (as is reasonable for wing cover panels),
means surface plies are less highly strained,
making delamination growth less likely.
Stability of propagation
Current aircraft design regulations negate
the possibility of delamination growth.
However, allowing benign stable delamination growth to occur could provide
significant potential for weight saving. The
Strip model is able to indicate the stability
of delamination growth, e.g. whether or not
increasing load is required for growth to
continue.

Current applications
Straight fibre components
The Strip model has been used to successfully predict the strain to failure (or
delamination growth initiation) of a number
of problems including: laminates artificially delaminated with a Teflon insert [1],
flat coupons manufactured from a variety
of materials and impacted at a variety of
energies (where individual delamination
sizes are either known or approximated from
an overall area) and impact to a skin panel
under a stiffener flange [7]. Typically, results
for standard ply angles (0°, ±45° and 90°) are
conservative and accurate to within 15%.
Stiffener edge impact
By adapting the VICONOPT model for
buckling strain to account for free-edge
delaminations seen in stiffener edge impacts,
it is possible to successfully apply the

Strip model and produce
predictions of failure strain
to within 15% of experimental results [8]. In some
cases however, the failure
of impacted stiffeners is
material (micro-buckling/
kink-banding/fibre failure)
rather than sublaminate
buckling driven. The factors
affecting the likely failure
mechanism are still topics
of research but thicker
laminates are more likely
to suffer material driven
failures.

Fig. 5: Comparison of experimental (black error bars) and Strip
model (block coloured columns) results for residual strength of C-ply
sandwich coupons subject to 9.76J impacts. Planar damage was
approximately 15mm diameter in all cases. Upper (red) bars relate to
the assumption that, owing to the resistance of the stitching, growth
cannot occur within a C-Ply™. Insets are stacking sequences, with
the exception of the right-hand bar which is a quasi-isotropic (QI)
laminate with a stacking sequence too complex to show fully

Steered fibre laminates
Steered fibre laminates offer
considerable opportunities
for weight saving through
improved load path distribution and thereby buckling performance
[9]. The spatially varying fibre angles and
unavoidable manufacturing features such
as tow gaps and overlaps seen in these laminates produce complex localised residual
strength issues. However, the Strip model
can be successfully adapted to such problems. In the former case by assuming that
locally steered fibres can be approximated by
a straight fibre problem and in the latter case
through the use of detailed descriptions of
damage structure obtained from non-destructive testing [10].
C-Ply™
The University of Bath's Composites Research Unit (CRU) has been working with
Prof. Tsai, Dr Alan Nettles and the Chomarat Group on our shared goals of improving
damage tolerance and producing rapid
validation/certification tools. The thin-ply
material (C-Ply™) being developed by Chomarat has many manufacturing advantages
but also offers significant damage tolerance
improvements over traditional thick-ply
materials. Figure 5 shows the improvement
achieved by unoptimised/homogenised
C-Ply™ laminates compared to baseline
thick-ply homogenised and damage-tolerant
laminates. Figure 5 shows the Strip model
is able to capture the stress to failure of the

60 jec composites magazine / No88 April 2014

C-Ply™ laminates and the uppermost line
in Figure 5 represents a C-Ply™ laminate
designed using Strip model principles. This
is a future target for experimental validation.
The superior damage-tolerant properties offered by C-Ply™ are attributed to its excellent
damage resistance. This is a consequence of
its stitched NCF construction (conferring
improved fracture toughness) and its greater
number of ply interfaces (in comparison
to an equivalent thick-ply laminate) across
which impact energy can be dispersed. An
improvement occurs because, assuming an
even dispersion through the thickness, lots
of small delaminations are preferable to few
large delaminations as they lead to higher
sublaminate buckling strains and hence
propagation strains.

Looking forward
The CRU is constantly striving to align itself
with industrial needs, whilst developing new
understanding of fundamental behaviour.
As far as the Strip model is concerned, this
means capitalising on its computational efficiency to produce an early stage design tool
for damage-tolerant structures. Key to this is
the development of a model for predicting
the extent of impact-induced delamination
on an interface-by-interface basis with computational efficiency matching that of the

JEC COMPOSITES MAGAZINE - Issue #88 - April 2014 - Special JEC Americas Spring Edition

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #88 - April 2014 - Special JEC Americas Spring Edition