Experimental research is presented to establish the characteristics of bond between plain steel reinforcing bars and concrete. Plain bars are regularly encountered in historical structures and criteria for assessing their bond are necessary. This study establishes:​ a statistical description of bond stresses for plain bars to calibrate design provisions;​ a realistic bond stress-slip model for pullout specimens;​ and documentation of the interaction between flexural behaviour, bond loss, and cracking in beams. This information provides a foundation for the development of design provisions for the evaluation of historical concrete structures reinforced with plain bars.

Abrams’ milestone investigation was reproduced to obtain a statistical description of bond stresses for plain bars. Overall coefficients of variation of 8% and 24% were calculated for the maximum and residual average bond stresses in pullout specimens, respectively.

Mechanics-based relationships were established between bond stress, bar force, slip at the unloaded end of the bar, and relative slip along the length of the bar in pullout specimens. Two 200 mm diameter by 800 mm long pullout specimens reinforced with instrumented built-up hollow plain bars were tested. Resulting observations showed that a two-step bond stress-slip model captures the essential features of bond behaviour, however, residual bond decreases gradually with increasing end slip.

Two concrete T-beams reinforced with instrumented built-up hollow plain bars were tested to document the interaction between flexural behaviour, bond loss, and cracking. The reinforcement ratio was varied to achieve failure by bond in the more heavily reinforced section and conventional failure initiated by yielding of the reinforcement in the more lightly reinforced section. Arch action occurred in one case while shear in the other beam was primarily carried by beam action. The flexural stiffness associated with arch action is reduced and so can indicate bond loss in beams with plain reinforcement