An assessment of systematic errors in beam tests on brittle materials

Abstract

In deriving the unit volume strength of brittle materials from beam bend test data, the shear and compressive stresses in the beam specimens are usually ignored. Depending on the span-to-depth ratio of the beam, these omissions may give rise to significant errors in the unit volume strength value, and in component failure probabilities derived from this strength value. In this paper, the effects of these systematic errors are considered for the 3-point beam as a function of span-to-depth ratio and for different values of the Weibull modulus. The relative errors are found to be small and conservative provided a sufficiently large span-to-depth ratio is used and the compressive/tensile strength ratio of the material is relatively high. © 1985 Chapman and Hall Ltd.