Headed Reinforcing Bars: CCT Node Tests, Design Provisions, and Evaluation of a Granular Micromechanics Model for Use in Finite Element Analysis of Bond

Abstract

The anchorage behavior of headed bars at the end of beams within compression- compression-tension (CCT) nodes subjected to monotonic loading is assessed. The test parameters for the 10 specimens tested include the embedment length, number and spacing of the anchored bars, and the presence or absence of a head at the anchored end of the bar. The nominal compressive strength of the concrete was 5000 psi. The test results, along with that of members other than beam- column joints available in the literature, are compared with strengths based on a descriptive equation developed for headed bars anchored in beam-column joints. The test results for specimens with bars without heads are compared with anchorage strengths predicted for straight bars. More broadly, test results for headed bars anchored in beam-column joints are compared with design provisions for the development length of headed bars in ACI 318-14 and ACI 318-19 and those 1 proposed by Shao et al. (2016) and Darwin and Dolan (2021) , along with the anchorage provisions in Chapter 17 of ACI 318-19. These comparisons are based on tests of 178 beam-column joint specimens containing headed bars with bearing areas between 3.8 to 9.4 times the area of the bar. In comparisons with the anchorage provisions in Chapter 17 of ACI 318-19, three modes of failure were checked–breakout, side-face blowout, and strength of the anchor reinforcement. Forty of the specimens (18 without confining reinforcement and 22 with confining reinforcement), had a ratio of effective depth of the beam to embedment length of 1.5 or more. In addition to the design provisions for development and anchorage, test results for these specimens are compared with strengths based on the strut-and-tie method in ACI 318-19. Finally, a granular micromechanics model and associated model for reinforcing steel-concrete interaction are evaluated for their general applicability for use in finite element modeling of anchorage of headed and straight reinforcing bars to concrete. A key point in the evaluation is to determine the importance of representing the local interaction between deformed bars and the surrounding concrete, which is not represented in this case, to obtain a fully objective model. Finite element results are compared with those from tests of headed bars embedded in slabs and straight bars embedded in concrete blocks. The strength of the CCT node specimens was limited by anchorage failure, either side-face blowout for headed bars or pullout for straight bars. Anchorage type (headed bars and straight bars) had a minimal effect on initial load-deflection behavior, but did affect strength. The descriptive equation developed for the anchorage strength of headed bars in beam-column joints is very conservative (test-to-calculate ratios ranging from 1.37 to 2.68 with an average of 2.05 for the current study test-to-calculated ratios ranging from 1.67 to 2.21 with an average of 1.89 for a study by Thompson 2006a) for headed bars in CCT nodes that have a compressive force placed perpendicular to the bar, as is the descriptive equation developed by ACI Committee 408 (ACI 408R-03) for straight bars (test-to-calculated ratios ranging from 1.72 to 2.76 with an average of 2.25). The provisions in ACI 318-14 for the development length of headed bars do not accurately estimate the anchorage strength of headed bars with high steel strength or concrete compressive strength. The equation, however, is generally conservative. The development length design provisions proposed by Shao et al. (2016) can be safely used for the design of the development length of headed bars for steel strengths at least up to 120 ksi and concrete compressive strengths at least up to 16,000 psi, while those in ACI 318-19 for headed bars do not fully capture the effects of confining reinforcement, bar spacing, and concrete compressive strength for compressive strengths above 6000 psi. The provisions proposed by Darwin and Dolan (2021) accurately reflect the anchorage strength of headed bars and provide a similar level of accuracy as that provided by those proposed by Shao et al. (2016). The strut-and-tie method in ACI 318-19 should be used to design joints with ratios of effective depth to embedment length of 1.5 or greater. The anchorage provisions in ACI 318-19 are very conservative when compared to any of the other methods evaluated in this study and, if used, would lead to nearly unbuildable designs. The granular micromechanics model and associated model from reinforcing steel-concrete interaction provide a good representation of the anchorage strength in cases where behavior of these specimens is dominated by the compressive and tensile properties of concrete, which are well represented by the granular micromechanics model. The combined model, however, does not provide a good representation of the behavior from members with strength that depends on splitting of concrete caused by to slip of the bar. Lack of representation of the local interaction between deformed bars and the surrounding concrete prevents the model from being generally applicable for use in representing reinforced concrete members, especially in cases where strength is governed by bond between straight reinforcing steel and concrete.