SAMPE Journal - January/February 2023 - 59

Figure 8. (a) micrograph picture of a section of
the stiffener beam sample (b) zoomed in portion
of micrograph picture to show layer boundaries.
The constituent volume fraction results do not
indicate the presence of any significant trends in
the fiber volume fraction or void volume fraction
through the height of the stiffener beam. The
fiber volume fraction graph remains relatively
horizontal, suggesting that there was little change
in the fiber volume fraction of the composite
through the height of the stiffener beam. The void
volume fraction graph appears less linear and
shows a slight decrease towards the top of the beam.
This suggests that the void volume fraction slightly
decreased through the height of the stiffener beam
sample. These results suggest that the composite
quality slightly improved through the height of the
tall stiffener beam.
Microscopic Inspection
A micrograph specimen taken from the first loop
print sample was inspected using a microscope
set to 5x magnification. Images were taken of the
sample using a Nikon D5300 DLSR camera fitted
with a 2x magnification eyepiece adapter. Due to
the size of the polished specimen, multiple pictures
were captured to cover the entire width and height.
Images were then spliced together to show the full
sample. A section of the sample is shown in Figure 8.
Figure 8a shows a micrograph section of the
high aspect ratio stiffener beam sample which
includes nominally 40 E-glass/PET layers. It can
be seen in the image that layer boundaries are not
easily distinguishable. A magnified portion of the
www. sampe.org
same section is shown in Figure 8b and better shows
the layer boundaries of the sample. While linear
patterns in the reinforcement can be discerned,
which indicate the boundaries between print layers,
clear matrix-rich bands at the layer boundaries are
not present. This suggests that the reinforcement
is well distributed throughout the height of the
composite stiffener beams during manufacturing. It
can also be seen that the layers at the bottom of the
specimen appear thinner and wider, which suggests
that the consolidation force experienced during
printing was greater in the initial layers than those
towards the top. This increasing layer thickness with
increasing vertical position is related to the chosen
fixed Z-step of 0.4 mm, with the constant Z-step
leading to changing consolidation dependent on
position and process parameters. Furthermore, the
wider initial layers provide evidence that the tow
width varied between the initial layers of the print
sample, suggesting that higher consolidation forces
produced wider print layers. These wider print layers
would have larger areas over which the consolidation
force was applied, directly affecting the consolidation
pressure experienced during printing of these
layers. However, further research is required to fully
understand the relationship between consolidation
force and consolidation pressure with changing tow
width.
The micrograph specimen shown in Figure 8
was taken from the same loop print sample that the
constituent volume fraction analysis was performed
on. When looking at the voids in the sample, which
are shown as dark spots in the image, it can be
seen that the vast majority of the voids occur on
the sides of the sample. When looking only at the
voids on the sides of the sample, the concentration
appears to stay relatively consistent through the full
height. However, the same cannot be said for voids
located towards the center of the sample. When
looking at central voids, it can clearly be seen that
few voids are present in the initial layers, and more
and more central voids appear with increasing
height through the sample. This gradient of central
voids through the height contradicts the results of
the constituent volume fraction test results. This
suggests that the consolidation force experienced
during printing gradually decreases, corresponding
to the reduction in composite quality, with
increasing height. The lack of evidence of this trend
in the average void volume fraction data suggests
that the void volume fraction did not adequately
reflect the actual distribution of the voids at a given
height within the beam. The increasing trend in
central voids through the height of the sample was
obscured by the relatively constant concentration
JANUARY FEBRUARY 2023
|
SAMPE JOURNAL | 59
http://www.sampe.org
SAMPE Journal - January/February 2023

Table of Contents for the Digital Edition of SAMPE Journal - January/February 2023
