Repeating a Classic Set of Experiments on Size Effect in Shear of Members without Stirrups

The shear tests performed by Bazant and Kazemi in 1991 have been cited as strong evidence that the theory of fracture mechanics applies to the shear strength of beams without stirrups. It has also been noted in the past, however, that these results are not totally consistent with other size effect experimental series. This paper provides a detailed summary of duplicate experiments carried out to determine if the original 1991 results were reproducible. It was found that the repeat experiments were 31 to 70% stronger than the original results. All experimental results were stronger than the predictions of the ACI and CSA codes. Cracks patterns were observed to scale with the beam size. A comparison of historical data indicated that the higher strengths were likely not a result of random variation, but due to a systematic effect in the original testing, making them inconsistent with other tests. Potential reasons for the differences from the 1991 results are presented and discussed.