Experimental and Analytical Reexamination of Classic Concrete Beam Tests

TLDR

The classic Bresler–Scordelis reinforced‑concrete beam tests, performed about 40 years ago, serve as a benchmark for calibrating finite‑element models. A nominally identical set of beams was recently tested at the University of Toronto, with load–deformation response, load capacity, and failure mode compared between the original and duplicate specimens. The duplicate tests largely replicated the original behavior, but revealed that crushing near loading plates, anchorage‑plate effects in unreinforced beams, and complex three‑dimensional disturbances challenge two‑dimensional FE modeling, while incorporating out‑of‑plane confinement improves accuracy and provides new insights into post‑peak behavior.

Abstract

The classic series of beam tests conducted by Bresler and Scordelis some 40 years ago to investigate the behavior of reinforced concrete in shear, is commonly regarded as a benchmark against which finite element analysis models can be calibrated. A nominally identical set of beams was recently tested at the University of Toronto. Aspects of behavior of the original and duplicate beams are compared and discussed, including load–deformation response, load capacity, and failure mode. Generally, it was found that most aspects of behavior were well replicated. Test observations reveal that the behavior of the beams is highly influenced by crushing of concrete beneath and adjacent to the loading plates. In the case of the beams containing no shear reinforcement, failure was influenced by the reinforcement anchorage plates. The disturbances around the loading plates and anchor plates introduce complex three-dimensional effects, making the modeling of these beams using two-dimensional finite element techniques difficult. However, accuracy is substantially improved if out-of-plane confinement effects are considered. In addition to some insights on the behavior of the original beams, and on factors that should be considered in their finite element modeling, the duplicate tests also provide information on postpeak behavior.