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

609

10

5. Discussion The initiation parameters (i.e., 0 , and 0 ) determined from the AE measurements show very good correlation. Furthermore, they compare well with the parameters computed using the (calibrated) deviation from linearity criterion (in fact, the 2 of the fitted 0 is larger for the parameters obtained using the deviation from linearity criterion). It remains to be shown whether the deviation from linearity can be applied for similar materials without calibration. 5.1. Experimental observations In general, all the AE parameters (rise, count, etc.) give the same onset time, 0 . However, the energy measurements often produce large discontinuities in the cumulative plots, which can be helpful to identify the transition from region I to region II. In contrast, the amplitude signals provide smoother variations, which makes it difficult to obtain the knee directly from the cumulative plot. However, because of this smoother transition the derivative of this parameter provides with a good visualization of the onset (see Fig. 3). Other combinations of the time signal features described in (Muir C. et al. 2021) such as the energy per hit, duration divided by the amplitude, the energy divided by the amplitude, and the counts per duration were analysed, however, no clear indication of crack initiation was found. The energy divided by the amplitude and the energy divided by the counts would often show a discontinuity at crack onset, however, they also showed other large discontinuities making it difficult to assess initiation. The duration divided by the amplitude and the amplitude divided by the counts did not show any clear visualisation of delamination initiation. The specimens tested with only one bending arm ( 1 / 2 =0) showed lower acoustic emission, than that recorded from other tests. The same observation can be found in (Goutianos S. 2019). A plausible explanation is that some of the acoustic emission picked up comes from mechanical vibrations from the lever-arms surpassing the threshold imposed. The presence of a region I in the cumulative plot where the crack front is not advancing suggests that initiation is more a process than an instant. Although, this observation holds valid for all mixed mode tests, it is easier to appreciate in shear dominated tests. 5.2. Limitations It should be noticed that in the present study the deviation from linearity was obtained using the AE measurements i.e., calibrated, as such, the plotted initiation parameters shown in Fig. 7 are not independent. It remains to be confirmed whether the deviation from linearity holds up for an independent experiment. It would be particularly interesting to investigate if the deviation from linearity is a material/interface dependent parameter. 6. Concluding remarks In the present paper it is shown that AE measurements can be used to identify the initiation of mixed mode delamination crack of glass/epoxy DCB composites in a systematic, and repeatable way. There is a clear correlation of the initial fracture resistance, 0 and the initial opening 0 with the mixed mode (here represented with the phase angle ). The acoustic emission signals from two sensors were also utilised to localise the crack front for different mixed mode values. Two regions were identified both in the accumulated sum of the AE features, as well as in the localisation plots. In region I, a slow increase in the cumulative sum of AE features is observed, while a fast increase in observed in region II. Although, all the AE features provided with a similar onset time, it can be note that the cumulative plot of Energy, and Amplitude are particularly useful. The cumulative Energy plot display an abrupt transition from region I to region II. The cumulative amplitude plots show a soft transition from region I to region II, which can be fitter and differentiated to visualise the transition from region I to region II. Furthermore, the AE