PSI - Issue 80

Marilyne Philibert et al. / Procedia Structural Integrity 80 (2026) 65–76 Author name / Structural Integrity Procedia 00 (2019) 000–000

74 10

composite plate showed reduced residual strain, suggesting that the residual strain was related to bonding quality or wiring.

Fig. 8. Strain testing of printed strain gauges (3 and 4) compared with commercial strain gauges (1 and 2) attached to (a) metal plate and (b) composite plate.

Fatigue testing was further performed for both metal and composite plates to evaluate fatigue resistance of the monitoring performances of the printed strain gauges. The plates were clamped at 5 cm from the edges of the plate and a cyclic load of 20 kN was applied with a frequency of 1 Hz for 1000 cycles. Position data, ( ) , was recorded as displacement induced to the specimen between clamping, allowing calculation of the averaged strain applied to the specimen according to the equation: = ∆ . . ! = .(+)-. ! . ! (5) The unstrained length of the specimen, # , was 200 mm taking into consideration clamping. The sampling frequency for resistance measurement was about 9.4 Hz, limited by the data acquisition system with multiplexer but sufficient for good resolution. The unstrained resistance values and position values are set at the beginning of the fatigue testing. As the printed strain gauge (numbered as 4) on the metal plate underwent permanent damage, its resistance results are not shown. Similar resistance changes were observed for printed and commercial strain gauges as plotted in Fig. 9 for both metal and composite plates. The improved bonding quality and wiring connection on composite plate showed improved monitoring performances under fatigue testing. Therefore, printed strain gauges showed similar strain response than commercial strain gauges with consistent performances.