JEC COMPOSITES MAGAZINE - Issue #99 - September 2015 - 30

simulation
Tab. 1: Preform properties of CFM225, balsa wood
and biaxial fibre [2]

CFM225

Balsa wood

Biaxial

K11 [m2]

1.16E-11

0.44E-10

1.80E-11

K22 [m2]

1.59E-11

1.47E-10

1.23E-11

K33 [m2]

5.44E-10

1.47E-10

1.11E-12

0.81

0.11

0.44

Fig. 2: Moldex3D simulation: flow front time at the
end of the filling stage for a sandwich structure

Permeability tensor
The impregnation of a fibrous material by
a resin flow can be described by Darcy's
law. Darcy's law establishes a relation
between velocity, pressure, permeability
and viscosity that is used for the current
simulation approach.
When analyzing the dynamic behaviour
of resin, we need to know the permeability
tensor in porous media. The details of the
permeability tensor can be found in [4].

where Kij (i, j = x, y, or z) are the components of the permeability tensor, K11, K22
and K33 are the principal permeability in
porous media, and lij are the directional
cosines of local coordinates.

Composition of the
sandwich structure

Fig. 3: Flow front evolution for a sandwich
structure [2]
Tab. 2: Material parameters for the carbon fabric
and the distribution media

Carbon fabric

Distribution
media

0.9653

0.9263

K11 [m ]

8.387E-12

1.071E-9

K22 [m2]

8.387E-12

1.071E-9

K33 [m2]

8.400E-13

1.071E-9

0.399

0.491

Thickness [mm]
2

Fig. 4: Mould geometry design for a wind turbine blade

As shown in Figure 1, the preform of
a sandwich structure is made of a core
material and a skin material. Balsa wood
[5] is placed as the core and biaxial
material on each side as the skin. A layer
of CFM225 material is placed on top of
the preform in order to reduce the filling
time; the length of CRM225 is 20mm.
The inlet has the same height (1.5mm)
as CFM225. Table 1 lists the preform
properties of CFM225, balsa wood,
and biaxial fibre, with biaxial having the
lower permeability property.

Results and discussion
Validation of RTM simulation:
sandwich structure
In this study, the
Moldex3D CAE
software was used
to simulate the
sandwich structure in the RTM
process. The results
show that, due to
the differential permeability between
the fibre mats, the

30 jec composites magazine / No99 September 2015

resin flows more slowly in the wall region than in the core region. Figure 2
shows the flow front time results at the
end of the filling stage, as simulated by
Moldex3D. It shows that the flow rate
decreases with time under constant
injection pressure.
The filling time and the flow front
distance are plotted in Figure 3, where
the red and blue lines represent the
simulation results. The results are in
good agreement with observations
from the experiment done in reference
[2]. It is obvious that the effect of a
sandwich fibre mat structure cannot
be ignored in this study, because it will
affect the filling result and lead to a
significant non-uniform filling behaviour in the RTM process. In addition,
according to the simulation results,
the sandwich structures in the RTM
process are correctly predicted.
Validation of RTM simulation:
fabric lay-up design for a wind
turbine blade
Figure 4 shows the mould geometry
design for a wind turbine blade. The
infusion design for this wind turbine
blade consists of 5 layers of carbon
fabric with four inlets. The entire region is laid up with a fibrous preform
and the dark region expresses the
placement design for the distribution media, which is used to provide
efficient resin flow throughout the
injection process. The cavity is filled
with the epoxy resin (Swancor 2502A/B) under constant injection pressure (P=1 atm). The sequence of inlet
opening times is tightly controlled, the
concept being to initiate resin filling
through the first inlet, Inlet E1. When
that part of the cavity is filled with
resin and the resin front reaches the
next inlet (Inlet E2), the next inlet
opens and the previous one (Inlet E1)
closes simultaneously. This sequence
is repeated incrementally until the cavity is completely filled.
The material properties of the carbon
fabric and the distribution media are
listed in Table 2. Figure 5 shows the
comparison between the experiment

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #99 - September 2015