Issue 51

G.S. Serovaev et alii, Frattura ed Integrità Strutturale, 51 (2020) 225-235; DOI: 10.3221/IGF-ESIS.51.18 230 For all FBGs the split of the reflected spectrum is observed. As the peak values are close to each other it is difficult to detect the initial Bragg wavelength, relative to which the wavelength shift will be calculated. Also application of the most popular algorithm of Bragg wavelength detection, based on measuring the maximum value of the spectrum can lead to significant measurement errors because the maximum value can shift from one peak to another under loading. For studied case the biggest distance between two separated peaks is observed for the FBG2 and is 0.39 nm, which can lead to strain measurement error of 327 µε that is unacceptable for most practical cases. A similar situation was observed for CFRP specimen. The obtained results on the distortion of the reflected optical signal from FBG sensors embedded into composite materials with a woven reinforcement structure are consistent with well-known studies for unidirectional composites [18,29,30]. This result shows that this problem is common for different types of composite materials and types of reinforcement. Spectral distortions can be caused by curing pressure, which leads to residual transverse strains of the optical fiber, and/or the non-uniform strain distribution along the length of the optical fiber that also occurs during the manufacturing process. There are several ways to overcome such problem, among which are the implementation of more advanced algorithms of calculating the spectrum change [31] and the use of additional mechanical protection of the Bragg grating area [32]. A NALYSIS OF MICROSCOPIC IMAGES OF COMPOSITE SAMPLES WITH EMBEDDED OPTICAL FIBER o study the internal structure of the composite material with an embedded optical fiber, fiberglass and carbon fiber samples were cut into pieces using a round cutter. For each of the materials, 10 cross-sectional elements were investigated. The photos of the GFRP and CFRP samples intended for the microscopic analysis are shown in the Fig. 8. Figure 8 : GFRP (a) and CFRP (b) samples. In order to assess which type of surface is most suitable for study with an optical microscope, one cross sectional surface of each sample was polished, and one left in its original state after the cutting (Fig. 9). Figure 9 : Surface treatment of samples. The experiment was carried out with the help of microscope KH-7700 (Fig. 10), which allows to achieve a magnification up to 7000x. In the experiment an MX (F) -10C lens with an OL-140 adapter and possible magnification up to 140х, 280х, 420х, 560х, 700х was used. T