PSI - Issue 11

Petr Stepanek et al. / Procedia Structural Integrity 11 (2018) 12–19 Petr Stepanek at al. / Structural Integrity Procedia 00 (2018) 000–000

16

5

was subjected to a combined load with normal force and bending moment due to the deformation of the beam. Different load direction had an impact also in monotonic tests, where bare bars reached higher tensile strength. The aim of our experimental program was the creation of S-N curves for axially loaded composite bare bars and bars encased in concrete blocks; for detailed description and results see Janus at al. (2018). E-CR GFRP 10 mm diameter bars with the same composition as described at chapter 2 were tested. The short-term mechanical properties of the tested GFRP reinforcement were determined experimentally before the fatigue tests (Table 3). The tensile testing of FRP bars was performed in accordance with ISO 10406-1 (2015). The aim of the inter laminar shear test was primarily to quantify matrix and matrix/fibre interface strength, which is very important in view of the long-term reliability of the composite subjected to cyclic loading. The configuration of the test and its implementation was based on ASTM D 4475 (2016). Table 3. Tensile test results of GFRP AR reinforcement immersed in alkaline solution. Bar diameter declared by manufacturer [mm] Bar diameter, including sand coating [mm] Average tensile strength [MPa] ± standard deviation Average modulus of elasticity [GPa] ± standard deviation Ultimate strain [%] Average strength in interlaminar shear [MPa] ± standard deviation 3.1. The fatigue tensile test of GFRP bars (without concrete environs) Fatigue test parameters were selected according to ACI 440.3R-12 (2012). The samples were force-control loaded by sinusoidal function. One of the key parameters influencing the obtained results is the loading frequency. This affects the fatigue life, which decreases with the increasing frequency. Higher loading frequencies result in material heating due to energy dissipation. According to ACI 440.3R-12 (2012), the acceptable load frequency range is 1-10 Hz, the recommended value is 4 Hz. This value was also chosen for this study. Also, increasing the difference between the minimum and maximum stress (the stress ratio R) leads to a reduction in the fatigue life, see Demers (1998). Due to the ratio of live load and dead load on conventional building structures, it is advisable to consider the stress ration equal to 0.1, see ACI 440.3R-12 (2012). To create the S-N curves, the recommended procedure (c), according to ACI 440.3R-12 (2012), was used; the stress ratio is equal to 0.1 for all levels and the maximum and minimum load values in the cycle are always adjusted. The samples were loaded until failure occurred or until the fatigue life of 2 million cycles was reached. So far, three stress levels have been tested for two test configurations of GFRP reinforcement samples: bare bars and bars encased in concrete. Levels with a maximum stress value in the cycle equal to 40%, 50% and 60% of the monotonic tensile strength were tested. During the loading, the machine’s actuator displacement was recorded on all samples. In addition, some samples were fitted with LVDTs to determine the potential change of the modulus of elasticity at the beginning of the loading, but the recording was only performed for the first 100 cycles. It can be seen from Fig. 4 that the strain of the reinforcement determined by LVDTs fitted at a known length and by means of a calculation from the actuator displacement for the whole sample do not correspond with each other. The strain determined by the calculation of the known actuator displacement seemingly overestimates the degree of damage to the reinforcement and indicates its gradual accumulation. This fact, however, is caused by a significant influence of sample anchoring on the measured values. The cyclic load affects not only the properties of the reinforcement but also the grouting material used to fix the reinforcement in the anchor. It is obvious that the degree of failure (reduction of the modulus of elasticity of the reinforcement) with this configuration cannot be directly determined by observing the actuator displacement and should be determined by direct reading of the strain using LVDTs. 3.2. The fatigue tensile test of GFRP bars encased in concrete Fatigue tests of GFRP bars in the concrete blocks were performed according to Rahman at al. (1996), see Fig. 5. A middle part of 100 mm length was separated from the end blocks. To prevent stress concentrations and the undesirable failure of end blocks at the beginning of the bond area, the bond of the bar with concrete was separated at the length of 5d b . At the same time, the end blocks were compressed by steel frames. To eliminate possible 10 11.03 1018.8 ± 5.2 52.2 ± 0.3 1.95 63.32 ± 2.32