25
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021
L. Lamberson et al. (eds.), Dynamic Behavior of Materials, Volume 1, Conference Proceedings of the Society
for Experimental Mechanics Series, https://doi.org/10.1007/978-3-030-59947-8_5
Chapter 5
Experimental Investigation of the Failure Rate Dependency
of Composite Materials Using Off-Axis Tensile Tests
Thomas Fourest and Julien Berthe
Abstract The present study investigates the mechanical response of T700/M21 Carbon Fiber Reinforced Polymers submitted to coupled shear and transverse tension at intermediate strain-rates ranging from 0.001 to 15 s
−1
. The experimental investigation is performed using transverse and oblique tabs off-axis tension specimens with a low aspect ratio. The axial
stress-strain curves are successfully simulated using Finite Elements simulations with a viscoelastic model. The numerical
simulations enable the calculation of corrections to take into account the effect of end constraints for such low aspect ratio
specimens. Finally, an increase of the material strength with the strain-rate is observed.
Keywords Off-axis tensile test · Unidirectional composite · Failure criteria · Viscoelasticity
5.1 Introduction
Reliable predictions of crash and impact scenarios rely on the knowledge of the material behavior at high solicitation speeds.
In the case of laminated composites, the physical justification of the different rate effects introduced in these models is classically based on compressive high strain-rate tests on off-axis specimens. Off-axis specimens permit to induce a coupling
between tension and shear in the specimen by using a single actuator and simple specimens which explain the interest of the
different authors for the dynamic testing.
5.2 Off-Axis Tension Tensile Tests
The advantage of off-axis specimens is the apparent simplicity of the test; however, their drawback is that they induced an
unwanted transverse load to the testing device. In addition to increasing the complexity of the analysis, it could prevent the
use of such specimens on servo-hydraulic jacks since they are usually not built to sustain high transverse loads. One way of
dealing with this drawback is the use of long specimens and of corrective factors. However, such long specimens are not
suitable for high strain-rate tests, since it would lead to smaller strain-rates for the same actuator speed and it would amplify
resonance of the load chain and pollute the load signal recording. It would also lead to larger times to reach mechanical
equilibrium. The investigated solution in this study is the use of oblique tabs on such specimens (Fig. 5.1). Several studies
[1, 2] have shown that the use of oblique tabs results in more homogeneous stress fields. In these specimens, the tab angle ϕ
is determined to follow the iso-displacement line induced by the testing of a specimen with fibers oriented with an angle θ
from the specimen length. Table 5.1 summarizes the numerical values used for both angles in this study.
T. Fourest (*) · J. Berthe
DMAS, ONERA, Lille, France
e-mail: thomas.fourest@onera.fr; julien.berthe@onera.fr
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021
L. Lamberson et al. (eds.), Dynamic Behavior of Materials, Volume 1, Conference Proceedings of the Society
for Experimental Mechanics Series, https://doi.org/10.1007/978-3-030-59947-8_5
Chapter 5
Experimental Investigation of the Failure Rate Dependency
of Composite Materials Using Off-Axis Tensile Tests
Thomas Fourest and Julien Berthe
Abstract The present study investigates the mechanical response of T700/M21 Carbon Fiber Reinforced Polymers submitted to coupled shear and transverse tension at intermediate strain-rates ranging from 0.001 to 15 s
−1
. The experimental investigation is performed using transverse and oblique tabs off-axis tension specimens with a low aspect ratio. The axial
stress-strain curves are successfully simulated using Finite Elements simulations with a viscoelastic model. The numerical
simulations enable the calculation of corrections to take into account the effect of end constraints for such low aspect ratio
specimens. Finally, an increase of the material strength with the strain-rate is observed.
Keywords Off-axis tensile test · Unidirectional composite · Failure criteria · Viscoelasticity
5.1 Introduction
Reliable predictions of crash and impact scenarios rely on the knowledge of the material behavior at high solicitation speeds.
In the case of laminated composites, the physical justification of the different rate effects introduced in these models is classically based on compressive high strain-rate tests on off-axis specimens. Off-axis specimens permit to induce a coupling
between tension and shear in the specimen by using a single actuator and simple specimens which explain the interest of the
different authors for the dynamic testing.
5.2 Off-Axis Tension Tensile Tests
The advantage of off-axis specimens is the apparent simplicity of the test; however, their drawback is that they induced an
unwanted transverse load to the testing device. In addition to increasing the complexity of the analysis, it could prevent the
use of such specimens on servo-hydraulic jacks since they are usually not built to sustain high transverse loads. One way of
dealing with this drawback is the use of long specimens and of corrective factors. However, such long specimens are not
suitable for high strain-rate tests, since it would lead to smaller strain-rates for the same actuator speed and it would amplify
resonance of the load chain and pollute the load signal recording. It would also lead to larger times to reach mechanical
equilibrium. The investigated solution in this study is the use of oblique tabs on such specimens (Fig. 5.1). Several studies
[1, 2] have shown that the use of oblique tabs results in more homogeneous stress fields. In these specimens, the tab angle ϕ
is determined to follow the iso-displacement line induced by the testing of a specimen with fibers oriented with an angle θ
from the specimen length. Table 5.1 summarizes the numerical values used for both angles in this study.
T. Fourest (*) · J. Berthe
DMAS, ONERA, Lille, France
e-mail: thomas.fourest@onera.fr; julien.berthe@onera.fr