JEC COMPOSITES MAGAZINE - Issue #83 - August/September 2013 - 46

Canada: country guest of honor

Fig. 3: The PermLab 2 permeability workstation

the permeability of stacks of 2D fibrous
preforms or thick 3D fabrics.
While saturated permeability is frequently
used to model saturated flows in porous media such as groundwater flows, unsaturated
(or process) permeability is the parameter
used to predict the evolution of the flow
front in mould filling simulations based
on Darcy’s law [6]. The PermLab 2 testing
bench allows measuring Darcy’s unsaturated
in-plane permeability by conducting unidirectional injection experiments in a rectangular mould as per the procedure published
by Ferland et al. [7] in 1996.

Measurement procedure
The statistical method used [7] provides a
simple and accurate way to process experimental data in order to evaluate the in-plane
permeability of fibrous reinforcements. Permeability is identified by a least square fit on
the measured positions of the flow front. By
taking into account the progressive pressure
build-up at the beginning of the injection,
it is possible to decrease the experimental
scatter. Another important point is the fibre
volume content of the fibrous preform,
which is never perfectly constant from one
experiment to the other. A correction based
on the Kozeny-Carman model can be used
to further reduce the experimental scatter.
In addition, a minimum injection length of
300 mm is required to ensure accuracy and
a minimum testing bench width of 100 mm
allows neglecting the effect of race tracking
on the edges of the mould cavity. Finally,
plotting the experimental results in real time

provides a fast and
simple means of controlling quality. Following this procedure,
a scatter of about 10%
is obtained for 2D
fabrics, which can be
further reduced for
3D fabrics when no
nesting occurs. In the
case of anisotropic
preforms, the two
principal in-plane
permeability values
and their orientation
are identified by a
special procedure [8].
Figure 4 illustrates the
three unidirectional
injection experiments
needed to identify the
permeability ellipse.

Fig. 4: Determination of the anisotropic permeability tensor from three unidirectional injection experiments [8]

The PermLab
2 permeability
workstation
The new PermLab
2 testing bench was
devised for inplane permeability
Fig. 5: Typical mould configurations used to measure preform permeability:
(a) unsaturated permeability in rigid mould; (b) saturated permeability in
measurements. The
rigid mould; (c) flexible permeability in resin infusion; (d) flexible permeabilirectangular mould
ty with measurement of injected flow rate and evolution of preform thickness
is composed of a
bottom retractable plate, a frame and a transexperiment. As illustrated in Figure 5, the
parent cover. Two sliders on the tool’s main
workstation can be used in several configuframe allow vertical motion of the mould
rations to measure various kinds of preform
cover. A pneumatic device was incorpopermeabilities for the RTM or resin infusion
rated to the system to lift the mould cover.
(VARI) processes.
The steel mould is designed to withstand a
closing pressure of 10 bars as this pressure
Data acquisition and processing
is required to close the mould and reach
A pressure gauge measures the injection
high fibre volume contents (close to 70% in
pressure near the inlet gate. The data acquisicertain cases). The thickness of the cavity
tion software records the times when the
can be set between 0 and 50 mm, allowing
flow front crosses a series of equidistant lines
the use of very thick laminates such as 3D
engraved on the mould cover. From this
fabrics. The plate motion is activated by hyinformation, the velocity of the flow front
draulic pressure in order to provide enough
can be calculated and hence the permeabilforce to close and seal the mould. Clamps
ity at each recorded position. As illustrated
are used to maintain the closing pressure. A
in Figure 6, the injection pressure and flow
computer is provided with the system to run
front positions in time, the front velocity
the data acquisition and processing software
and permeability at each recorded flow
and record images of the successive posifront position are drawn in a PermLab 2
tions of the flow front during an injection
window. Note that a flowmeter or a balance

46 jec composites magazine / No83 August - September 2013

JEC COMPOSITES MAGAZINE - Issue #83 - August/September 2013

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #83 - August/September 2013