JEC COMPOSITES MAGAZINE - Issue #81 - May/June 2013 - 71

imperfections detected depends on
the inspection technique used. A costquality compromise needs to be found
in all cases. Designers have to take the
presence of these imperfections into
account. The established tolerance
criteria are the maximum size and
number of imperfections. Minimum
mechanical characteristics correspond
to the maximum size and number. The
design of the part takes these “degraded” characteristics into account
relative to the theoretical material
“without imperfections”. When part
inspection identifies imperfections in
terms of number or size greater than
those defined by the designer, we no
longer talk about “imperfections” but
“defects”.
The quantity of imperfections
tolerated in composites parts varies
enormously depending on the industrial sector concerned. As discussed
before, this can be directly linked to
the manufacturing cost. The traditional processes used in sectors such as
boiler-making generate a substantial
number of imperfections. This leads,
on one hand, to limited mechanical
characteristics and, on the other hand,
to a significant dispersion of the mechanical characteristics from one part
to the next. Moreover, on this type of
part, inspections are generally limited
to a visual inspection. In order to
take these uncertainties into account,
manufacturing standards for such
parts necessitate safety coefficients
that can be as high as 20 [1].
In contrast, the highly competitive
transport sector needs to reduce safety coefficients while at the same time
limiting manufacturing costs. In the
aerospace sector, the weight-gain factor takes precedence over manufacturing cost. As a result, the high-performance processes used make it possible
to manufacture parts containing very
low, controlled levels of imperfections. Detailed unit inspections ensure
high levels of mechanical performance
combined with a very low dispersion
of mechanical characteristics. These

requirements, which are essential for
guaranteeing the reliability of aircraft,
result in high manufacturing and nondestructive testing costs.
In the automotive industry, the constraints of manufacturing costs and
processing cycle-time encourage the
use of new processes and materials:
continuous fibre and thermoplastic
matrix composites. In this context,
knowing the impact of the different
types of imperfections on performance is a subject of major importance for the years ahead. This is one
of the areas studied by the Composite
Innovation OpenLab, a common
laboratory formed by CNRS, Ecole
Centrale de Nantes, ENS Cachan and
Cetim, that was inaugurated at the
JEC Show in March 2012.
There are two approaches to studying
the effect of defects. The first consists
in performing a large number of mechanical tests on a material containing
a range of imperfections that is statistically representative of the whole
manufacturing cycle. The dispersion
obtained is then taken into account in
the design. Such a study was conducted by Cetim on composite pipes.
A total of 1000 samples were analysed
for a statistical approach to apprehend
the various defects and their effects
on mechanical properties [2].
The second approach consists in testing composites containing calibrated
sizes of imperfections and measuring their impact on the mechanical
characteristics. This latter approach is
often combined with modelling.
The main factors that influence the
harmfulness of a particular type of
defect are:
- type of composite material (matrix,
reinforcement and interface),
- stacking sequence and direction of
the reinforcement,
- type of stress (tension, compression,
shear, fatigue, impact, etc.).
Therefore, it is difficult or even
impossible to generalize about the

harmfulness of defects in composites.

Damage
Damage refers to the internal degradation of a material due to different
types of stress or environments (mechanical, thermal, physico-chemical,
radiative, etc.). In addition to the
imperfections related to manufacturing, the onset of damage in composite
materials is very different from that
observed in metallic materials.
When a metallic material is subjected
to a stress, a crack generally starts
from a specific point and then propagates and leads to failure. The damage
to composite materials is completely
different: a large number of microscopic events develop very gradually
over a large volume of the material.
This is due to the heterogeneity of
the material on a microscopic scale,
as the matrix and reinforcement have
very different mechanical behaviours.
Understanding these mechanisms is
essential for modelling the behaviour
outside the elastic domain.
The first step of damage logically
begins in zones of lower strength than
the matrix/fibre interface and the
matrix itself, with failures over small
distances called intralaminar cracks.
Intralaminar damages mainly appear in the areas where fibres are not
oriented in the axis of the load, when
the strain in the matrix reaches its
breaking strain. In general, intralaminar cracks are parallel and regularly spaced. A crack density which
increases with the strain rate can then
be measured. This type of damage has
little impact on the ultimate strength
of the material.

Fig. 2: Delamination during tensile tests due to the
stiffness difference of plies

No81 May - June 2013 /

jec composites magazine 71

JEC COMPOSITES MAGAZINE - Issue #81 - May/June 2013

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #81 - May/June 2013