PSI - Issue 17

Kristýna Čápová et al. / Procedia Structural Integrity 17 (2019) 726 – 733 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

732

7

All the samples were subjected to testing inside the climate chamber (Fig. 6 (a)) with temperature range from -20 to 50 °C. The temperature change (increase) was performed in 10 °C steps. The samples were tested one after another. To ensure the constant temperature throughout the sample, samples were left in the chamber for more than 12 hours for each temperature step. Furthermore, relative strains were observed throughout the measurement; when the measured values began to oscillate around a certain value, it was noted that the temperature was constant throughout the sample. In Fig. 6 (b) the values of relative strain detected for all the tested FBG sensors are compared. In a bare FBG sensor, the detected elongation increases linearly with temperature. Concerning the sensors laminated inside GLT beams, there is a significant shift in measured values when the temperature changed from -20 °C to -10 °C. This shift is probably caused by the change in the properties of wood at low temperatures when the water is converted to ice and backwards. The rest of the plot is approximately linear. If the temperature changes by 70 °C, relative strain of approximately 25% of the maximum calculated value for the tested sample is detected, which can cause a false alarm in structural health monitoring. 2.4. Third phase of testing in 2018 The third phase of testing consisted of two experiments – Test 01: testing of FBG sensors in polyimide and ormocer primary coating glued in GLT beams and Test 02: testing of FBG sensors in Glass Fibre Reinforced Polymer (GFRP) coating inside concrete beams. The temperature and humidity inside the climate chamber were recorded throughout the tests in two different ways. The first is the recording of these quantities by the climate chamber itself. However, this data is rather distorted, especially if there is higher moisture change in the chamber. This measurement was therefore supplemented by different temperature and humidity sensor located directly in the chamber next to the tested samples. This control measurement has shown that the actual response of the climate chamber to the set humidity is not perfect. The parameter of either constant or controlled variable humidity proved to be irrelevant during testing. An overview of parameters of realized and evaluated test is presented in Table 5.

Table 5. Configuration of realized and evaluated tests. No. of cycles

Temperature

Humidity

Test 01-a: Standard operating conditions – timber sample

10 °C – 30 °C – 10 °C (in 2 °C / min ramp)

3

50 %

Test 01-b: Standard operating conditions – timber sample

10 % – 70 % – 10 % (in 5 % / min ramp) 50 % (no control below 0 °C) 65 % – 95 % – 65 % (in 5 % / min ramp)

3

21 °C

Test 02-a: Standard operating conditions – concrete sample 3

-15 °C – 35 °C – -15 °C (in 3 °C / min ramp)

Test 02-b: Standard operating conditions – concrete sample 3

21 °C

In the first part of Test 01, when the air humidity supposed to be constant and the temperature was changing, this did not happen due to the malfunction of the chamber. Thus, with the changing temperature, air humidity also changed. The response of FBG sensors to the changes were very good, however, there was a strong hysteresis effect probably caused by the moisture penetration into the structure. The same conclusion was reached for the second part of Test 01; the temperature was constant and the air humidity had all the effect on the sensor behaviour. The results of measuring the concrete sample (Test 02) are analogical. 3. Mechanical loading tests The GLT beam equipped with FBG sensors and extensometers was subjected to cyclic loading in four-point bending test, Fig. 7 (a), performed with the same loading arrangement in June 2016, March 2017 and March 2018. The beam is made from eight lamellae and has dimensions of 140 x 315 x 6110 mm. Three optical fibres are laminated between the lamellae; each optical fibre contains three FBG sensors. The first fibre is placed between the two top lamellae, the second one is between the lamellae in the mid-height of the cross-section and the last one is placed between the two bottom lamellae. There are three extensometers in the middle of the span; their location in