Seismic Retrofit of Lap Splices in Nonductile Square Columns Using Carbon Fiber-Reinforced Jackets

ACI Structural Journal, Nov/Dec 2006 by Harries, Kent A, Ricles, James R, Pessiki, Stephen, Sause, Richard

Specimen L1

Specimen L1, as noted previously, was retrofitted with four plies of CFRP material oriented transversely over the lower 500 mm (19.7 in.) height of the column and two plies from 500 to 1000 mm (19.7 to 39.4 in.). Figure 4(b) presents the applied load-displacement response for the specimen. First cracking at the interface of the column and the footing was observed at a lateral load of 80 kN (18.0 kips). As the test continued, yielding was observed at a lateral load of 181 kN (40.7 kips) (greater than the nominal flexural capacity of the cross section). The cycles at the yield displacement produced several transverse cracks in the jacket at heights of approximately 250 and 530 mm (9.8 and 20.8 in.) from the column-footing interface. The cracks located at 530 mm (20.8 in.) are approximately at the top of the lap splice (490 mm [19.3 in.]). These cracks may also be influenced by the change in jacket stiffness (4 to 2 plies) just below this point. Following yield, opening of the cracks at the interface of the column and the footing accounted for most of the tensile deformation in the hinge region. These crack openings increased significantly as the ductility demand increased (similar to the crack opening shown in the photograph of Specimen L2 at 6δ^sub y^ shown in Fig. 4(c)). The first observable indication of lap splice slip in Specimen L1 occurred at a deflection of 38 mm (1.5 in.) on the unloading cycle from the second excursion to 3δ^sub y^ as noted in the applied lateral load-displacement response of Specimen L1 shown in Fig. 4(b). This slippage was sudden and marked by a very audible "popping and grinding" noise. During subsequent cycles to 3δ^sub y^, bulging of the jacket began 25 mm (1.0 in.) above the column-footing interface on the south face of the column. Similar bulging was noted on the north face beginning in the cycles to 5δ^sub y^. With continued cycling beyond 3δ^sub y^, increased pinching and deterioration of the hysteretic response was observed (refer to Fig. 4(b)). This behavior is discussed in the following, Lap splice failure designated in Fig. 4 and Table 3 is defined as the point at which splice capacity begins to degrade (the end of Region 2 in Fig. 1). Lap splice tension capacity was determined from extensive investigation of the strain distributions and thus force transfer between the column and starter bars (Harries et al. 1999). Figure 4(b) shows Specimen L1 at -6δ^sub y^. Despite the splice slippage observed, Specimen L1 maintained a capacity greater than its nominal flexural capacity through the initial cycles to 5δ^sub y^.

Specimen L2

Specimen L2, as noted previously, was retrofitted with four plies of CFRP material oriented transversely over the lower 500 mm (19.7 in.) height of the column and 2 plies from 500 to 1000 mm (19.7 to 39.4 in.). Additional longitudinally (vertically) oriented plies applied under the transverse jackets consisted of four plies of longitudinal CFRP were applied over the lower 500 mm (19.7 in.) height of the column and two plies from 500 to 1000 mm (19.7 to 39.4 in.). Figure 4(c) presents the applied lateral load versus displacement response of Specimen L2. First cracking at the interface of the column and the footing was observed at a lateral load of 80 kN (18.0 kips). As the test continued, yielding of the longitudinal reinforcement was observed at a lateral load of 184 kN (14.4 kips) (greater than the nominal flexural capacity of the cross section). Through a displacement of 2δ^sub y^, no cracks were observed in the jacket. The column-footing interface crack absorbed most of the rotational strains. Hairline cracks were observed on the exposed concrete above the jacketed region. Due to the moment gradient, however, these did not develop as the test progressed. The lack of cracking in the jacket illustrates the effect of the longitudinal plies on the tension faces of the column. During the cycles to 2.5δ^sub y^, some compressive distress in the CFRP matrix was observed in the two-ply region, which was above the level of the lap splice. Due to the presence of the longitudinal plies, most of the deformations were concentrated at the column-footing interface (refer to Fig. 4(c)). Pinching of the hysteretic behavior, due to bond slip at this location, is evident in the response of this specimen as it was in Specimen L1. This slip led to a lap splice failure at a displacement of 5δ^sub y^, as noted in Table 3 and Fig. 4(c). Similar to Specimen L1, deformation was concentrated at the column-footing interface (refer to Fig. 4(c), resulting in relatively rapid deterioration of the specimen lateral load carrying capacity. Specimen L2 also maintained its nominal flexural capacity through the initial cycles to 5δ^sub y^.