SAMPE Journal - July/August 2020 - 24

FEATURE / FIBER BED ARCHITECTURE

Figure 7. Bar graph showing the time to full sorption for all five fabric types.

stopped changing. These values are outlined in Figure 7. Of note is the significant difference from the
standard UD-NCF samples to the woven fabrics.
This was expected, following the fact that the nonUD samples have a significant number of pores
and gaps in the fiber bed surface that are much
larger than the small capillaries that form between
individual tows in the unidirectional fiber beds.
Looking at the two UD-NCF fabrics, the difference in time to full sorption can be correlated to
fiber areal weight: The heavier fabric, sample 1, allows more fluid to flow in the out-of-plane direction before reaching full saturation.
Between the three non-unidirectional weaves,
the fastest time to full sorption is for the plain weave
fabric. This is likely due to the high number of pinholes, and how the pinholes provide full clearance
between the top and bottom of the fabric.
Figure 8. Surface area covered by fluid over time for the different fabric
samples.

SAMPE JOURNAL

J U LY/AU G U S T 2 0 2 0

Surface Coverage Over Time
Using the top-down view, the area covered by the
droplet was measured. This was done by selecting
images at specific points in time and manually covering the droplet, then using the image processing
code to determine the covered area. Due to lighting difficulties, some data points are missing. Furthermore, the droplet became absorbed into the
fabric for some weaves, becoming too faint to be
measured. The surface area vs time data is shown
in Figure 8.
Some things to note from this data is that, much
like the height vs time data, the two UD-NCF samples followed a very similar curve, with both having
the surface area covered continue to expand over
time until the time of test termination (t=10000s).
The other weave styles exhibited a maximum coverage value before the surface area covered began
to decrease, as the droplet became fully absorbed
into the fabric.
This behavior is expected for fluid flow over
porous media, where the wetting phenomena of a
droplet on a surface drive the surface area covered
to increase, while the capillary effects of the porous
surface drive fluid into the material, causing the
covered area to decrease18.
In the case of the UD-NCF fiber beds, the capillary effect is significantly lower due to the smaller
size of the capillaries, resulting in the time needed
to observe the maximum surface coverage exceeding testing time.
The effect of each individual weave, however,
is better seen by the shape the fluid takes when
covering the droplet, and not merely the value of
the coverage that Figure 8 shows. For this, Figure 9
shows a general guideline, displaying contours of
the droplet over time.
What the contours show is that the fluid follows
the details of the weave closely: for the UD- NCF
fabrics it spreads mostly along the direction parallel to the fibers; for PW the fluid flows around the
pinholes, until it wraps around the pinhole and
goes over; the TW sample shows the fluid unable to
go over the sides of tows and flow into the tow edges; this same behavior can be seen in the Quilted
UD sample, where the fluid spreads significantly in
the gap between two adjacent tows.
In the case of the plain weave fabric, it is predicted that the droplet was large enough that positioning would not affect the spread significantly,
the ratio of droplet size to weave detail size was
large. However, for the other two woven fabrics, the
size of the weave details is significant compared
to the size of the droplet. Looking particularly at
the Quilted UD sample, the droplet is positioned

w w w. s a m p e . o r g

http://www.sampe.org

SAMPE Journal - July/August 2020

Table of Contents for the Digital Edition of SAMPE Journal - July/August 2020