PSI - Issue 64

Quentin Sourisseau et al. / Procedia Structural Integrity 64 (2024) 893–900 Quentin SOURISSEAU/ Structural Integrity Procedia 00 (2019) 000 – 000

895

3

that the concept of equivalent interface strongly depends on the thicknesses of interfaces and/or involved interphases and should therefore be validated if another interface arrangement had to be studied. The steel plates are grade S355, the resin is a two-component hot-polymerizing epoxy, the glass fibers are in the form of woven plies and carbon fibers are in the form of unidirectional fabrics. For all specimens, the steel substrates were sandblasted and degreased before bonding. The test pieces were manually laminated before vacuuming. For each specimen, a Teflon film, used to create the beginning of a crack, was positioned at the edge of the specimen, over 10 cm, and between each interface.

Fig. 1: Studied equivalent interface samples.

2.2. Description of the experimental investigations used to characterize the equivalent interface Figures 2(a), (b), (c) represent the frames and tested specimens respectively in mode I (DCB test), mode II (ENF test) and mixed mode (MMB test). Two mixed mode ratios were tested (using different geometrical properties according to ASTM standard): 55 and 75 % (of mode II). For all the tests, the applied force and the displacement were recorded using a load cell with a capacity of 5kN (accuracy ± 0.145%) and a laser displacement sensor (accuracy 10µm). All the tests are monotonic at constant displacement rate (2 mm/min). A single mode polyamide coated optical fiber is bonded on the surface of the specimen to monitor the strain profiles. This allows, using the methodology developed in Sourisseau et al. (2022b), to monitor the propagation of the crack when it is stable. Indeed, according to classical beam theory, longitudinal strains on the surface of the metal substrate are maximum at the junction of the bonded and non-bonded parts (corresponding to the location of the crack front) and at the point of application of the force. This last position being known and not changing during the test, the modification of the local maximum on the strain profile gives access to the position of the crack front during the test.

(a)

(b) (c) Fig. 2: Experimental investigations on the equivalent samples using: (a) DCB, (b) ENF, (c) MMB.

An example of force/displacement curve obtained for a DCB type test is shown in Figure 3a. Figure 3(b) represents the evolution of the strain profile along the sample length, for different times. We observe a shift in the position of the maximum corresponding to the crack tip propagation. Crack propagation can thus be determined for each test (figure