PSI - Issue 37

Grzegorz Wójcik et al. / Procedia Structural Integrity 37 (2022) 179–186 Grzegorz Wo´ jcik / Structural Integrity Procedia 00 (2022) 000–000

184

6

The first tool was a low-cost guillotine (OMRON E39). According to the manufacturer, the razor blade of this tool shall be used only several times. After that, this tool has to be changed because the blade gets worn out. This guillotine was used in two di ff erent scenarios: without stripping o ff the outer jacket of the cable and with stripping it o ff . The second tool was a high-end cleaving tool with a razor blade that rotates after each cut (RENNSTEIG 8002 0004 3). In other words, the tool has a limited number of cuts and each new cleaving is done using a sharp part of the razor blade. After the sensor was fabricated, both ends of the fiber were cleaved with a particular tool. After the cleaving, microscopy images of the end-faces were taken and the sensor was characterized. This procedure was repeated several times for each tool. The same cleaving tool was not used more than once in a row. Fig. 4 presents the quality of the polymer optical fiber after cleaving with a particular tool. Upper images present fiber surfaces after cleaving with a high-end tool, i.e. the RENNSTEIG 8002 004 3. Visual inspection shows that the surfaces do not have any deformations, nor cracks. The surfaces look smooth and only some small amount of polymer contamination gathered on the final part of the cleaving. Middle pictures represent the POF surfaces cleaved with a low-cost guillotine, without stripping o ff the outer jacket. The surface of the end faces presents some artifacts, and it looks as if the cutting speed was not constant. Bottom images show end-faces of a guillotine cleaving with outer jacket stripped o ff . The fiber is deformed and looks, similarly as before, as if the cleaving speed was not constant. Furthermore, some cracks are visible. Its quality is very poor because the fiber was also squeezed due to the lack of the outer jacket.

Fig. 5: Sensor characteristics: (a) RENNSTEIG 8002 004 3; (b) Omron E39-F4; (c) Omron E39-F4 (outer jacket removed); (d) comparison.

Obtained characteristics for both end-faces of the sensor cleaved with the high-end tool are shown in Fig. 5(a). Fig. 5(b) presents characteristics for the end-faces cleaved with a low-cost guillotinne. Fig. 5(c) contains charts for the POF cleaved with the guillotinne as well, but with the outer jacket stripped o ff . Top columns of the Fig. 5(a),(b), and (c) represent generated photocurrents. Middle columns show the data after the normalization. The average of normalized measurements with standard deviations are shown by bottom columns. The comparison between all three sets of characteristics are presented in Fig. 5(d). Derived sensitivites indicate, that the better the quality of end-faces, the higher the sensor reproducibility by means of lower standard deviation. Surprisingly, the sensitivity was found