PSI - Issue 37

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

180

2

ity for distributed monitoring, as well as spark-free and long-term monitoring without susceptibility to drift (Kuang et al. (2005)). This makes them them useful for various areas, involving biochemical, structural health monitoring (SHM), and medicine (Bilro et al. (2012)). To exploit the properties of the light and its interaction with the surrounding medium, the scientific community developed several sensing techniques, including intensity-based, wavelength-based, and interferometry-based sensing. Chosen sensing technique a ff ects not only the sensor fabrication methodology, but also the measurement instruments that are used by the system. Intensity-based sensing is one of the earliest, the most common, and inexpensive technique that has found its use in multiple applications, such as the refractive index sens ing (Sequeira et al. (2016); Samavati et al. (2020); Wang et al. (2020)), curvature and strain measurement (Kuang et al. (2002); Chen et al. (2017)) or crack monitoring (Kuang et al. (2003); Yang et al. (2017)). The sensing principle relies on the monitoring of the light intensity, which varies due to the interaction between the fabricated POF sensor and the property that is measured. This methodology does not require expensive equipment and the sensor can be fabricated using simple procedures, which make this technique attractive for large-scale applications. Wavelength based sensing and interferometry pattern changes have been reported to be capable of measuring similar parameters, including refractive index (Hu et al. (2014); Ferreira et al. (2017)), strain (Chen et al. (2010); Statkiewicz-Barabach et al. (2017)), deflection (Yan et al. (2018)), temperature (Chen et al. (2010); Dong et al. (2018)), pressure (Bundalo et al. (2016); Oliveira et al. (2018)), and humidity (Oliveira et al. (2018)). These sensors require more sophisticated fabrication procedures in order to inscribe the fiber bragg gratings (FBG) in a POF. Moreover, the data acquisition is done using expensive instruments, such as high resolution spectrum analyzers for the wavelength shift detection. From the perspective of large-scale applications, there are other significant benefits of POF sensing solutions. In contrary to their silica counterparts, the POFs utilize cores of relatively large diameters (usually 1 mm). This results in lowered costs of the associated accessories and hardware (such as low-precision connectors). POF termination can be done using a razor blade and easily connected to a ferrule. Polymethyl methacrylate (PMMA) based POF is less susceptible to flexural damage due to its order lower Young’s Modulus when compared to glass optical fibers (GOF), and does not produce sharp edges when broken. Furthermore, the intensity-based sensing can be done us ing inexpensive electronics components, namely LEDs, photodiodes, operational amplifiers, which results in simple electronics circuit. All of this eases the fabrication and handling procedures. Although POF sensing is insensitive to electromagnetic interference, it is subjected to other factors that may influence its usage, especially when it comes to the intensity-based sensing. Most of the researched papers present applications under laboratory conditions. Real-life implementations are subjected to various external factors, such as vibrations, environmental conditions, light source fluctuations, component aging. Such factors may influence (or even limit) the fabricated sensor performance. Bilro et al. (2012) have identified the stability of the light sources as one of the causes of measurement errors, and a major disadvantage of intensity-based POF sensors. Kiesel et al. (2007) have experimentally measured mechanical nonlin earities of a POF intensity-based strain sensor and reported the importance of sensor calibration. Authors pointed out four mechanical and six opto-mechanical parameters that need to be calibrated. Their work does not include, but mentions the importance of temperature, creep, and hysteresis properties of the POF sensors that will likely play a role in the long-term response. Some of the errors induced by external factors can be compensated by splitting the optical signal into two POFs, before the light is distorted by the sensing area (Montero et al. (2009); Sequeira et al. (2016, 2019)). By doing so, the data processing algorithm can filter out factors that influence both of the fibers at the same time, such as light source fluctuations, vibrations, and other. This solution, however, increases costs dramatically for large-scale applications due to the presence of POF-splitters or second optical path (transmitter-POF-receiver). Also, not much attention has been paid to the sensor reproducibility issues. Rodriguez et al. (2016) demonstrated an electri cally passive cleaver for polymer optical fibers, especially for microstructured POFs. The cleaver achieves particular range of velocities due to proposed mechanical system, composed of a spring and a damper. Authors have also demon strated an annealing proces, which has successfully removed some of the cracks present on the fiber end-faces after cleaving. The annealing process is however a time-consuming process (24-hours at 70°C). The research conducted by Chapalo et al. (2020) has evaluated the e ff ective cleaving parameters for polymer optical fibers based on di ff erent core material (CYTOP). Researchers utilized razor blades of di ff erent manufacturers for manual as well as automated cleaving under di ff erent temperatures and cleaving speeds. Obtained results indicated that the fiber end-face after manual cleaving significantly depends on the razor blade type. Automated cleaving under controlled speed and tem perature has significantly improved the reproducibility, reducing the irregularity from 3.5% to 0.7% for a chosen razor blade. Furthermore, an extended razor blade lifetime was obtained due to slower quality degradation of the blade (up