Members subjected to combined axial load and bending moment - Reinforced Concrete with FRP Bars: Mechanics and Design - page 156

COMMENTARY
The experimental work by De Luca et al. [10] included a total of five
specimens: one steel RC benchmark, and four GFRP RC columns. The
specimens had a square cross section with 24 in. (610 mm) sides and length
of 10 ft (3.0 m). The GFRP RC columns were subdivided into two sets of
two; each set was identical to the other, but bars from different manufac-
turers were used.
The steel RC benchmark column was designed using the minimum amount
of longitudinal reinforcement and the minimum tie cross-sectional area at
maximum spacing as mandated by ACI 318-11. In particular, the total area
of longitudinal bars was taken as 1.0% of the gross section area, A g , choos-
ing eight no. 8 (25 mm) bars. No. 4 (12 mm) ties were used as transverse
reinforcement at a spacing of 16 in. (406 mm) on center. For each set of
two GFRP RC columns, bar size and total area of longitudinal reinforcement
were adopted as for the steel benchmark. For the GFRP ties, the same bar
size as their steel counterpart was used, but the spacing was reduced to
12 in. (304 mm) and 3 in. (76 mm) when compared to the steel case. The
12 in. (304 mm) spacing was selected to prevent longitudinal bars buckling,
while the 3 in. (76 mm) spacing was chosen as the minimum practical spac-
ing for GFRP ties.
The GFRP RC specimens with the large tie spacing behaved similarly to the
benchmark steel specimen. Failure typically initiated with vertical cracks—
followed, first, by lateral deflection of the longitudinal bars contributing to
the splitting of the concrete cover and, then finally, by crushing of the con-
crete core and buckling of the longitudinal bars. In the case of the GFRP RC
specimens with the smaller tie spacing, energy absorption and deformability
were greatly increased as capacity decreased steadily after the peak load until
the test was intentionally interrupted.
In all columns, the concrete compressive stress at peak was close to 0.85 f ′ c ,
which is the value defined in ACI 318-11 as the average concrete compressive
stress when an adequately tied column reaches its axial strength. The aver-
age load carried by the longitudinal GFRP reinforcement (assuming an equal
modulus of elasticity for GFRP in tension and compression) ranged between
about 2.9% and 4.5% of the peak load, whereas the average load carried by
the longitudinal grade 60 (413 MPa) steel reinforcement was about 11.6% of
the peak load.
The failure of the steel RC specimen appeared to be ultimately
caused by the buckling of the longitudinal bars preceding crushing of