PSI - Issue 52

Ruben I. Erives et al. / Procedia Structural Integrity 52 (2024) 600–610 Ruben Erives et al./ Structural Integrity Procedia 00 (2019) 000–000

602

3

prediction as AE signals are very sensitive wherein the separation of true damage from spurious noise is indistinguishable. The current work focuses on combining the AE data recorded during the DCB testing and the measured crack openings to detect the delamination initiation. More precisely, the externally applied moments are plotted against the effective end-openings to check the deviation from linear relation and further, compared with the AE cumulative amplitude and energy to accurately detect the onset of crack initiation. The paper is structured as follows: Section 2 briefly discusses the DCB test setup with the instrumentation used and the geometry of the test specimen. Section 3 gives theoretical background of the data treatment. Section 4 discusses the results in terms of cumulative AE features recorded during the testing and applied moments against effective end-opening from various mixed mode tests, to predict the delamination onset. Lastly, limitations are discussed in Section 5 and the conclusions are summarised in Section 6. 2. Experimental setup The present study uses experimental results from a test campaign of double cantilever beam specimens with uneven bending moments (DCB-UBM) carried out in an earlier study (Erives, Sørensen, & Goutianos, 2023). The DCB specimens used in a such study were manufactured from a glass/epoxy laminate composed of 20 unidirectional (UD) fabrics. Details regarding laminate manufacturing and test specimen geometry are provided in the study. Several tests were carried out at different moment ratios, i.e., different mixed modes. The data is available as a data article (Erives 2023). Nevertheless, a brief description of the test setup is provided for the sake of completeness.

a)

b)

Fig. 1. Schematic illustration of a) test setup, and b) test specimens

The specimen is fixed in the un-cracked end and the moments are applied at the cracked end using a wire system pulled by a displacing beam at 10 mm/min. The lever- arms used to generate the moment are offset by, θ 0 , for testing specimens under (near) shear opening displacement. A schematic of the experimental setup, and the DCB specimen is shown in Fig. 1. The test specimens were instrumented with two acoustic emission (AE) sensors, two load cells, an extensometer, a linear variable differential transformer (LVDT), and two cameras. The AE sensors were placed, with sensor 1 (channel 1) approximately at the location of the artificial crack (made via an insert of Teflon foil), and sensor 2 (channel 2) at 250 mm apart from sensor 1. The AE system used during the testing was a two daisy-chained 1283 USB AE Nodes from MISTRAS Group (Physical Acoustics) each with R15-alpha sensors. These sensors