JEC COMPOSITES MAGAZINE - Issue #107 - August/September 2016 - 63

Structural health of polymer composites

placed in front of the specimen in order to measure its surface temperature in real-time during the mechanical loading (passive thermography, hereafter abbreviated IRT-p) as well as during the heating of the
sample by two halogen lamps (320 W, with a three-cycle sinusoidal
signal of 0.1 Hz), and then during its cooling (active thermography,
hereafter abbreviated IRT-a).
The mechanical loading was applied at room temperature with a crosshead speed of 0.5 mm/min using a standard tensile machine according to the loading profile shown in Figure 2, which consists in a stepwise tensile test of 50 MPa each time (7 ramps and 6 levels of 4 min
duration). The thermal scene (specimen under loading, lamps and IR
camera) was isolated from the external environment using an opaque
and insulating curtain. During the ramps, the AE and passive thermal
activity of the composite specimen were acquired in real-time by, respectively, the AE1-AE2 sensors and the IR camera. The GW, LW and
IRT-a measurements were applied sequentially (GW before LW, then
IRT-a to avoid potential harmful interference) during the constant
stress levels and before the first ramp. The sample cooling phase under
constant stress, for 125 s after heating by the lamps, was monitored by
the infrared camera and the cooling rate was measured for each constant applied stress.

Damage assessment
Raw data analysis
Glass fibre-reinforced plastics are not very heat-dissipative under static tensile loading (see Figure 3). During the first five loading ramps,
almost no damage was detected on the thermal maps, which were
uniform over the entire surface of the specimen. During the 6th ramp,
three damaged areas were identified, labelled 1, 2 and 3 on the corresponding thermal map. Both areas 1 and 3 were rather large and diffused heat for a long time (over 102 seconds), while area 2 was thin
with very brief heat diffusion (flash lighting for about 4 seconds). The
two damaged zones 1 and 3 may be ascribed to localized delamination
between the 0° and 90° plies, whereas the longitudinal damaged zone

Fig. 3: Thermal maps under the applied tensile stress ramps

Fig. 4: Acoustic emission amplitude under the applied tensile stress ramps

2 corresponds to a fibre/matrix debonding area located on the 0° ply
on the rear face of the specimen. During the 7th ramp, the final failure
of the specimen gave rise to a very high heat emission.
The acoustic activity during the applied stress ramps is illustrated
in Figure 4 in terms of amplitude vs time. Once loading started, the
acoustic emission began with low amplitudes (38 to 50 dB) and gradually became denser and higher as the load increased. The growth of
the acoustic emission events in terms of quantity and amplitude indicates that the damage induced in the material was heterogeneous,
including different mechanisms such as matrix cracking, fibre/matrix
debonding, delamination and fibre rupture [12].
The guided wave signals (S0 Lamb mode) are shown in Figure 5 for
stress-free conditions and for the six applied stress levels. The signal
amplitudes decrease sharply and continually until the transmitted ultrasonic energy at the highest stress level totally disappears. This amplitude decrease is due to the damage induced in the material, result-

Fig. 5: Guided waves during the applied tensile stress levels

No107 August - September 2016

/ jec composites magazine

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #107 - August/September 2016