Validation of semi-empirical blast pressure predictions for far field explosions - is there inherent variability in blast wave parameters?

A considerable amount of scientific effort has been expended over many decades on developing means of predicting the loading generated when a blast wave impinges on a structure. Semi-empirical ‘look-up’ predictive methods, such as those incorporated in the UFC-3-340-02 manual, the ConWep code or the *LOAD_BLAST module of LS-DYNA, offer a simple means for predicting the blast loading generated in geometrically simple scenarios. However, reported test data frequently show considerable spread and lack of repeatability, which is often attributed to some inherent variability in the blast waves developed from detonations, although no definitive physical interpretation has been forwarded as to the source of such inherent variation. As such, the semi-empirical predictions are often viewed as only ‘ball-park’ or ‘order of magnitude’ estimations. This paper presents experimental measurements of reflected pressure-time histories from a series of well-controlled small scale blast tests. Data fitting techniques are used to obtain experimental reflected pressure and impulse values which are compared to corresponding semi-empirical predictions. We find that it is possible to produce reliable and highly consistent, repeatable results that match predictions remarkably well and therefore show that existing semi-empirical blast predictions can be used with confidence as a first-order approach for quantifying the blast load a structure will be subjected to. Our results presented here suggest that for small scale far-field loading in simple geometrical scenarios, test-to-test variability can be reduced by ensuring that test parameters are tightly controlled.