Glass breakage by blasts: A prediction model and real-world validation.

Abstract

The structural elements most sensitive to blast damage are usually windows. A classic paper by Hershey et al. [J. Acoust. Soc. Am. 55, 1009–1017 (1974)] developed a powerful statistical model for damage to windows from impulsive (sonic boom) pressure loads. The key part of the model depended on both the pressure load and failure stress having a log normal distribution; that is, the log of these variables was normally distributed. This basic concept, applied by the author in several studies on blast and sonic boom damage, is briefly reviewed here for application to blast damage to windows. Validation of the model is provided by data from an accidental explosion of an equivalent 20 lbs of TNT near a nearby nine-building apartment complex. 58 windows were broken, fortunately without any injuries. The predicted damage was augmented by results of a model study of diffraction around a cubical obstacle to approximate the changes in the net pressure loading on the apartment building windows located at various orientations to the blast wave front. The predicted window breakage rate, varying from 0.5% to 61% compared well with the observed values in the range of 0–64%. [Work supported by Wyle Labs.]