JEC COMPOSITES MAGAZINE - Issue #82 - July 2013 - 57

nodes on the yarn paths at each crossover point between warp/
binder and weft yarns. Extra nodes are then positioned along
the binder yarns to follow the contour of the outer weft yarns.
At this stage, all yarns are assumed to have constant cross-sections.
With no refinement applied
yet, there are intersections
between the warp and weft
yarns and the through-thickness binder yarns. On examination of the µ-CT images,
it is also apparent that, even
in the uncompacted fabric,
the cross-sections of the outer
weft yarns differ from those
in the middle of the fabric. A
refine option has, therefore,
been implemented for the
case of orthogonal weaves in
order to model the geometry
more realistically both in the
uncompacted and compacFig. 2: TexGen 3D Wizard: Orthogoted states. Throughout the
nal weave pattern dialogue
process, the volume fractions
of the yarns are monitored so that they are maintained within
realistic limits. To facilitate this, yarn properties are required as
input for the warp, weft and binder yarns.
Firstly, the cross-sections of the through-thickness binder yarns
are changed, if necessary, to fit into the available through-thickness spaces between warp and weft yarns.
Then, the heights of the warp and weft middle layers are adapted to model compaction on mould closure. An initial check of
the space available will show whether it is possible to achieve
a volume fraction for the yarn below the maximum specified.
If not, TexGen will flag an error stating that the thickness
requested cannot be achieved without exceeding the maximum
volume fraction. It has been observed that the height of these
yarns decreases in proportion to the total initial thickness of
warp and weft layers. The height of the yarns is reduced accordingly and then the width is increased to maintain the original
volume fraction. If this cannot be achieved, the power of the
power ellipse section used is increased, thus increasing the area,
potentially to a rectangular section filling the space available.
The cross-sections of the outer weft yarns are then changed to
give a section which more accurately represents the real textile,
at the same time reducing the height to reflect the requested
compaction. Where necessary, the height is reduced further to
allow for the, similarly reduced, thickness of the binder yarn,
which is incorporated into the height available for the weft yarn.
Finally, the outer binder yarn cross-sections are changed to give
a section which is flat on the outer fabric surface. The widths
of the binder yarns are increased as necessary to maintain the

volume fraction. If the required height cannot be achieved,
the height of the weft yarn is reduced further, volume fraction
allowing. If this fails, crimp is introduced in the weft yarn to
achieve the overall textile thickness, and the cross-section of the
warp yarn below is adjusted to remove resultant intersections in
the yarns.
Figure 3a shows the orthogonal weave with the refine option
selected but no change to the initial fabric thickness of 6.315
mm. Here, the refinement is limited to the binder yarns, which
are adjusted through the thickness, and the outer weft yarns.
Figures 3b and 3c show the fabric compacted to thicknesses
of 5.03 mm and 4.432 mm. Figure 3c shows the addition of
a small amount of crimp in the outer weft yarn, necessary to
achieve this degree of compaction. Comparison with the µ-CT
images shows that, based on the assumptions mentioned above,
TexGen is capable of modelling the geometry realistically
down to a fabric thickness of 5.03 mm (Vf = 55 %). At a higher
compaction level (thickness 4.43 mm), deviations between
the TexGen model and the real geometry occur, noticeably in
surface yarn cross-sections. Manual modification is required in
this case.

Fig. 3: Orthogonal weave generated using TexGen 3DWizard refine option; A:
original fabric thickness, H = 6.32 mm; B: H = 5.03 mm; C: H = 4.43 mm

Flow modelling
To determine the textile permeability, which affects impregnating resin flow in composites processing, flow though the textile
unit cell was simulated using Computational Fluid Dynamics
(CFD) software. The yarns in the unit cell were modelled as
porous media. The yarn permeability as input parameter was
calculated using Gebart’s analytical model [4]. Conservation
of fluid mass and momentum was assumed at the interfaces
between the porous yarns and the inter-yarn flow channels.
Periodic boundary conditions were set on opposite faces of the
textile unit cell both in weft and warp directions. Non-slip walls
were specified for top and bottom faces to simulate the boundary conditions during the in-plane permeability test [5]. A
pressure drop was applied on opposite faces of the unit cell. Dry
air at 25°C under atmospheric pressure was selected as fluid for
the simulation of steady-state laminar flow through the fabric. A
voxel mesh was chosen for the current study as it is robust and
can be generated automatically. Properties of either the flow
channel domain or yarn volume were attributed to the voxel
elements. The mesh can be exported directly from TexGen to
ANSYS® CFX 12.0. The in-plane permeability in warp and weft
No82 July 2013 /

jec composites magazine 57

JEC COMPOSITES MAGAZINE - Issue #82 - July 2013

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #82 - July 2013