PSI - Issue 64

M. Esmaelian et al. / Procedia Structural Integrity 64 (2024) 2091–2100 M. Esmaelian/ Structural Integrity Procedia 00 (2024) 000 – 000

2099

9

4.2.3. Influence of concrete confinement Unconfined concrete has a brittle behavior and is crushed at relatively low axial compressive strains. On the other hand, the crushing strain of confined concrete is considerably higher. Due to the rocking behavior of precast segmental columns, high axial compressive strains are expected at the toes of the base segment. Therefore, several confinement methods have been proposed in previous studies to prevent damage at these regions. This study investigates the effectiveness of five confinement techniques, including variable spacing of spirals and installation of a 6 mm steel jacket at the base segment by conducting analysis on 5 specimens denoted as S8 to S12. Three different spiral spacings of 37.5, 75, and 150 mm were considered to understand the effect of spiral spacing on the confinement of the base segment. The results show that by reducing the spiral spacing by 4 times, the residual drifts significantly reduced, as shown in Figure 9 (a). For instance, the maximum residual drift at a drift ratio of +5% decreased from 1.45% to around 0.1%, which is well below the self-centering limit. However, smaller residual drifts imply less concrete damage and therefore less energy dissipation. This can be clearly seen in Figure 9 (b), where reducing the spiral spacing reduced the maximum EVDR from 15% to 8%. The installation of the steel jacket further improved the self-centering behavior by reducing the residual drifts to the order of less than 0.1%, as shown in Figure 9 (a). As expected, the maximum EVDR was reduced to about 7.5%. In other words, the results of specimens S11 and S12 in Figure 9 (a) show that the steel jacket can provide more effectively concrete confinement than closely spaced spirals, although it may be more expensive. By comparing the results of specimens S10 and S11, which have identical spiral spacing for the bottom segment and different spacing for the top segments in Figure 9 (a) and (b), it can be concluded that these two specimens have completely similar behavior, indicating that the confinement of the top segments is not as effective as the confinement of the bottommost segment.

Figure 9 Effect of concrete confinement on (a) the residual drift (b) the equivalent viscous damping ratio of Fe-SMA prestressed segmental columns

5. Conclusions This paper discusses the feasibility of self-centering existing and new RC columns using prestressed Fe-SMA reinforcement. Experimental results show that Fe-SMA significantly reduces residual drifts in retrofitted existing RC columns. Columns with an ED/Fe-SMA ratio of 0.15 maintained residual drifts below the satisfactory limit of 1% up to a drift ratio of 4%. In addition, numerical models were developed to investigate the feasibility of self centering behavior of segmental columns using the unique self-prestressing properties of Fe-SMA bars rather than conventional tendons. The results showed that Fe-SMA bars provide both self-centering and higher energy dissipation capacity to segmental columns. In addition, the effect of three parameters on the cyclic behavior of Fe SMA prestressed segmental columns was investigated. The results of the parametric study show that increasing the ED/Fe-SMA ratio from 0.18 to 0.85 at a total axial load ratio of 17.5% increased the average residual drifts by 2.9 and 4.7 times and the EVDR by 89% and 87% at 3% and 5% drift ratios, respectively. Bonded Fe-SMA bars improved energy dissipation but caused stress concentration, resulting in early loss of SMA PT force and larger residual drifts. Concrete confinement of the bottom segment improved self-centering, while confinement of the top segments did not affect column performance. The use of a steel jacket, FRP wrap, or closely spaced spirals in the first segment is recommended to delay concrete damage.