When an explosive detonates, whether a high explosive, or a flammable gas or vapour, a high pressure shock wave is produced that expands rapidly away from the initiation point. For anything caught in the path of this shock wave, the initial peak pressure and the impulse loading can cause severe damage to occur. In order to facilitate a safer built environment, modern architectural design requires an accurate and complete understanding of the response of structures to shock waves caused by a variety of explosives and explosive fuels.
Abstract When modelling the range of possible explosives, a TNT equivalence factor is usually considered. This equivalence factor relates the amount of energy released of any explosive or fuel directly to the equivalent energy of TNT. To be able to accurately assess and effectively utilise this equivalence, it is imperative that the parameters are not only measured and calculated using accurate and reliable methods, but that the shortcomings and caveats of the TNT equivalence model is well understood. Effective modelling requires an understanding of how the different parameters measured with a pressure wave can vary relative to each other depending on factors such as the unique behaviour of different types of explosives; conditions such as the size or casing of the explosive; or the positioning of the explosive relative to the structure under consideration.
Abstract To enable this thesis to examine the behaviour of explosions in a significantly more thorough and scientific approach than previously possible, the author has conducted in excess of ninety explosive tests using ANFO, PE4 Nitromethane and Hydrogen, resulting in approximately 950 data points. The data produced from these trials permits an in depth examination of the sensitivities involved with measuring blast parameters during experimental trials, allowing an evaluation of the factors to be considered in order to produce accurate and reliable data.
Abstract Key findings of this research identifies the superiority of using pencil gauges for freefield pressure measurement; ensuring reflected gauges are installed within targets that are either appropriate for the trial or as large as possible to reduce clearing effects; and reducing interactions between targets in a single arena. This thesis also demonstrates the value of applying a Friedlander curve-fit to experimental pressure data as a way to produce a repeatable and reliable way to identify key parameters from the data.
Abstract Applying these recommendations, further trials were conducted, producing a second data set, shown through assessment of the uncertainty factors and coefficient of variation to be both highly repeatable and accurate. This data set was then used to validate Air3D and ConWep as computer models for use in predicting blast parameters, finding excellent agreement between the experimental data and the outputs from the models. This work stands as the largest experimental comparison seen in literature.
Abstract The expansive experimental data set was then used to validate the Hopkinson-Cranz scaling law, proving its use for a range of charge sizes and scaled distances for PE4 and ANFO. It was also used to validate the calculation of energy flux using the Grisaro method for PE4, ANFO, Nitromethane and Hydrogen. This method had only previously been investigated by Swisdak and Grisaro using very limited data and for neither commercial explosives nor gas. Using the calculated energy flux, the author compared these values to that from TNT to allow calculation of TNT equivalences. This produced values found to be independent of charge size and stand-off, and in good agreement with current literature: 1.2 for PE4, 0.81 for ANFO and 25 for Hydrogen.
Abstract In summary, this thesis explores how the sensitivities of blast parameters can vary within experimental trials and makes recommendations for producing reliable data. It also suggests a new method for producing a value for TNT Equivalence. While there remain a great number of complexities which offer potential for further investigation, it is believed that the work presented here will both enhance the reliability and accuracy of explosive testing, and will act as a springboard for further research.
