SAMPE Journal - November/December 2021 - 58

FEATURE / STRUCTURAL COMPONENTS
Figure 2. (left) Geometry of the test lugs; (right) Set-up for mechanical characterization of
the lugs.
using an Anton Paar MCR702 instrument equipped
with a three point bending set-up (TPB40). Tests
were performed at a (logarithmic) frequency sweep
of 10 to 0.1 Hz and a stress of 1.5 MPa. Flexural
moduli are determined at 1 Hz.
Printing and Mechanical Testing of 3D
Products
3D products, in the form of test lugs, were printed
using similar printing conditions as for the flexural
test samples. The lugs consist of two tube sections,
one of which is placed almost perpendicular (at an
angle of 82.5 degrees) to the other (Figure 2, left).
Samples are printed layer-by-layer in the orientation
as shown in Figure 2, so that all integrated carbon
fibers are in a horizontal plane. To maximize the
length of the fibers as well as the fiber fraction, they
are positioned along the perimeters of each layer
in multiple lines. For one set of samples, the test
lugs are printed in an orientation tilted 40 degrees
counterclockwise so that the fibers are positioned
under an angle compared to the other samples.
Mechanical characterization of the 3D printed
lugs was performed using an Instron tensile tester
with a customized set-up, as shown in Figure 2.
The vertical tube of the lug is fixated to the bottom
part of the machine, while the horizontal tube of
the lug is fixated to the top part of the machine
and is pulled upwards with a speed of 10 mm/
min (after pre-loading to 50 N at 1 mm/min),
while the displacement and force is recorded. As
reference, part of a conventional carbon bicycle
frame (thermoset carbon composite) is used,
which has the same geometry. For all geometries
and materials, 2 or 3 test lugs specimens were used.
Modeling
In order to try to understand the observed response
we have built a model for calculating the effective
mechanical behavior of printed parts as a function
58 | SAMPE JOURNAL | NOVEMBER/DECEMBER 2021
www. sampe.org
of the properties of the individual constituents as
specified in the supplier's datasheet. In the case of
prediction of the elastic modulus of the composite
material under tensile stress we have made use of
the rule of mixtures under uniform unidirectional
strain for both fibers and polymer matrix:
[1]
where af
and apoly
are the volume fractions
occupied by the fiber and polymer respectively.
The same applies for the tensile modulus Ef
Epoly of both materials.
The flexural
and
three point bending test was
simulated by finite element analysis using COMSOL
Multiphysics11
. Only one quarter of the geometry
was simulated as shown in Figure 3 and quasistatic
conditions were used. The distance between
the rollers was set to 40 mm as in the experiments,
contact points were frictionless but also remained
Figure 3. (left) Depiction of the material model used for the prediction
of mechanical properties of the printed composites; (middle) Equivalent
material representations for sandwich and perimeter layouts; (right)
Simulated 3-point bending (flexural) test where only the enclosed part is
actually modeled.

http://www.sampe.org

SAMPE Journal - November/December 2021

Table of Contents for the Digital Edition of SAMPE Journal - November/December 2021