6. Conclusions
This paper addresses the characterization of the mechanical
response of short-fiber reinforced thermoplastic tubes. The tubes
are manufactured by helical winding of a composite made of highdensity
polyethylene (HDPE) with short glass-fibers as reinforcement.
The mechanical behavior of the composite lamina was
characterized by means of monotonic and loading/unloading tensile
tests at different deformation rates. The tests were performed
on specimens extracted in the axial and circumferential directions
of the tube
The tests allowed observing that the material presents a viscoelastic
behavior. The elastic response is transversally isotropic, with
the isotropy plane perpendicular to the direction of the fibers; the
hysteresis effect diminishes with the strain-rate; and the hysteresis is
larger in the transversal than in the longitudinal direction. There are,
for all the range of deformation strains studied, acceptable linear
correlations between the stress and the stress-rate, and a linear lneln
correlation between the viscosity and the total strain-rate.
Based on the experimental results, a three-dimensional
Maxwell model is proposed for the laminas. The model is tailored
to account for the elastic anisotropy of the material while the
viscous behavior is defined via a non-linear isotropic law written in
terms of the strain-rate. This resultant model has eight parameters
(five for the elastic response and three for the viscous response),
which are retrieved from the test data via a least square fitting.
The comparison of the model results to the experimental data
shows that: the model predictions for the elastic specimen stiffness
is very accurate (differences around 3%) in the circumferential direction
(this is the direction nearly coincident with the direction of
the fibers) but there are important deviations (around 27%) for the
longitudinal direction. The predictions of the Poisson's ratios are
good; discrepancies between the model predictions and the test
data are around 15%. The model properly captures the non-linear
anisotropic material response under loading conditions;
maximum loads are predicted accurately. For the unloading conditions,
the model properly predicts the material response in the
longitudinal direction, but it results in important deviations from
the experimental measurements in the circumferential direction.
These results put in evidence the anisotropic nature of the material
viscoplastic behavior and thus provide valuable hints for the
further development of the model.
The model for the material behavior is validated by comparison
to experimental measurements for a full-scale test that combines
the bending and punching of a tube. To this end, the material model
was implemented as part of an in-house developed finite element
code. The results compare well in terms of forces and displacements.
Maximum discrepancies are around 15%. These discrepancies
are mainly attributed to the deficiencies of the model to
accurately reproduce the material behavior during unloading.
The non-linear characterization of the mechanical response of a
short-fiber reinforced polyethylene tube introduced in this work is
a step forward with respect to classical linear-elastic analyses. The
model has been developed bearing in mind the ease of application
for industrial problems. The model has the capability to capture the
main features of the material behavior under monotonic loading
conditions, it can be set-up using data from tensile tests, and it is
compatible for the implementation into standard finite element
codes.