Effect of Aggregate Size on Beam-Shear Strength of Thick Slabs

This paper describes an experimental program in which 10 large-scale and 10 geometrically-similar, small-scale, shear-critical reinforced concrete slab-strip specimens were loaded to failure. It was found that the major mechanism of shear transfer in these element types is aggregate interlock, and that the maximum aggregate size plays an important role in beam-shear capacity. The abilities of the ACI design method and a simplified design method based on the modified compression field theory (MCFT) to predict the failure loads are compared. It was found that the ACI design method is unconservative when applied to thick slabs or large wide beams constructed without stirrups, whereas the simplified MCFT design method is both safe and accurate. It is further found that the simplified method can accurately predict the effects of decreasing the maximum aggregate size on the beam-shear behavior of lightly reinforced concrete members. It is concluded that shear design methods should account for the fact that aggregate interlock plays a critical role in the beam-shear behavior of reinforced concrete members.

Keywords: aggregate; design; reinforced concrete; shear.

(ProQuest-CSA LLC: ... denotes formulae omitted.)

INTRODUCTION

Shear design methods for reinforced concrete structures should be simple, rational, and general. Above all, however, they should be safe and accurate. Unlike flexural failures, reinforced concrete shear failures are brittle and can occur without warning. Furthermore, they tend to be less predictable than flexural failures due to more complex failure mechanisms. Whereas flexural design provisions are based on the simple assumption that plane sections remain plane, the search for equally rational design provisions for shear continues.

To illustrate the need for safety and accuracy in shear design methods, consider the design of the thick one-way transfer slab in the condominium building shown in Fig. 1. The slab has been designed using the ACI code1 to transfer closely-spaced wall loads from the 12 upper stories to the more widely-spaced walls in the podium and parking levels below. The ACI code requires that narrow beams contain at least minimum shear reinforcement if the factored shear force V^sub u^ exceeds 0.5Φ V^sub c^. This requirement, however, does not apply to wide beams or slabs, so the code requires stirrups for these element types only when V^sub u^ exceeds Φ V^sub c^. The thickness of the slab shown in Fig. 1 has been chosen so that Φ V^sub c^ exceeds V^sub u^ and hence shear reinforcement is not required. If an error resulted in, for example, the placement of only 2/3 of the required top flexural reinforcement above Walls W2 and W3, this reinforcement would yield under the dead weight of the structure, causing wide cracks and excessive deflection. This structural distress would provide considerable warning that the slab was under-strength. Even if this structural distress was not noticed, however, it is unlikely that the slab would collapse. Strain hardening, moment redistribution, and membrane action in the slab would all contribute greatly to increasing the load required to cause flexural failure. On the other hand, if the shear design of the transfer slab is inaccurate, and the shear capacity is only 2/3 of that which is required, a brittle shear failure could occur with no prior warning, resulting in the collapse of the structure and possibly great loss of life.

The basic ACI 318-051 expression for Vc, Eq. (11-5), which was used to choose the required thickness of the transfer slab shown in Fig. 1, does not account for the fact that the shear stress to cause failure of members without shear reinforcement decreases as the depth of the member increases. Because of this, there are concerns2,3 that the current ACI shear design provisions are unconservative for thick slabs such as that shown in Fig. 1. These concerns can be investigated by constructing and load-testing 1 ft (300 mm) wide beam-strips extracted from the slab as shown in Fig. 1(c). Such elements are simpler to test than wide slabs, and because the web width b^sub w^ has been found to have no significant influence on the beam-shear failure stress,4,5 they can provide accurate information about the beam-shear behavior of slabs and wide beams.

The purpose of this paper is to explore the safety and accuracy of the ACI shear design method when applied to thick slabs by focusing on the size effect in shear and the role played by the maximum coarse aggregate size in transferring shear stress across cracks. An experimental program is described, in which a series of 10 large-scale and 10 geometrically similar small-scale slab-strip specimens were loaded to shear failure. A simplified shear design procedure based on the modified compression field theory (MCFT) is described, and its predictive capabilities compared with the ACI method.

RESEARCH SIGNIFICANCE

The research reported in this paper demonstrates that aggregate size influences shear strength and hence supports the theory that aggregate interlock plays an important role in shear behavior. It is shown that a new simplified method of shear design based on the MCFT is significantly more accurate than the current ACI code. The ACI code is shown to be unconservative in situations where the size effect dominates shear response.