PSI - Issue 3

Stavros K. Kourkoulis et al. / Procedia Structural Integrity 3 (2017) 316–325 S. K. Kourkoulis and I. Dakanali / Structural Integrity Procedia 00 (2017) 000–000

323

8

between them was measured as a function of time. It can be seen from Fig.9a that the agreement between DIC and clip-gauges is almost perfect and the differences recorded are well below 1% for the whole duration of the experiment. 3.3. Acoustic emission analysis The acoustic emission technique provided also interesting data, in excellent correlation with the respective slope changes of the crack opening- and load-time curves (Figs.8a, 9a). Indeed the time evolution of the acoustic char acteristics (duration and AE energy) exhibited clear quantitative changes which divide the test’s duration in three distinct intervals, denoted as A, B, C in Fig.11, providing interesting indications about the succession of the failure mechanisms activated. Especially for the intervals B and C, the events’ location was identified and is shown in Figs. 11b, c. It is evident that the events during interval B are concentrated along the fault’s plane and are safely attributed to gradual failure of the cement paste layer between the fragments. On the other hand, the events recorded during the time interval C, are accumulated around the reinforcement bars (Fig.11c) of the lower and the intermediate level.

(b)

(a)

C

B

B

`

(c)

Α

C

Fig. 11. (a) The time variation of the duration and energy (chromatic scale) of the acoustic hits; (b) The location of acoustic events for the time interval B; (c) The location of acoustic events for the time interval C.

The dependence of the overall load exerted on the crack opening is plotted in Fig.12, in juxtaposition to the respective dependence of the cumulative AE energy. It is seen that during the linear segment of the load curve the AE energy increases smoothly. Its rate changes at the end of load linearity and again at the end of the load’s non-linear segment. Finally, it is interesting to note that while the load remains constant there is almost no acoustic activity at all. 4. Discussion The mechanical response of a fragmented marble epistyle, restored with three pairs of threaded titanium bars, according to the technique adopted on the Acropolis of Athens work site, was studied. The epistyle was subjected to ten-point bending with the aid of a proper assembly of metallic bars and beams. The innovation of the study is that the response of the epistyle was monitored with the aid of a complex sensing system including a combination of traditional and innovative sensing techniques. Although the analysis of the data gathered is still in progress, some interesting preliminary conclusions were drawn, related both to the epistyle’s response and the failure mechanisms which were activated during loading as well as to the efficiency of the innovative sensing techniques used.