Explosives exposed to conditions above the Chapman-Jouget (CJ) state exhibit an overdriven response that is transient. Reactive flow models are often fit to the CJ conditions, and they transition to detonation based on inputs lower than or near CJ, but these models may also be used to predict explosive behavior in the overdriven regime. One scenario that can create a strongly overdriven state is a Mach stem shock interaction. These interactions can drive an already detonating or transitioning explosive to an overdriven state, and they can also cause detonation at the interaction location where the separate shocks may be insufficient to detonate the material. In this study, the reactive flow model XHVRB utilizing a Mie-Grüneisen equation of state (EOS) for the unreacted explosive, and a Sesame table for the reacted products, will be used to examine Mach stem interactions from multi-point detonation schemes in CTH. The effect of the overdriven response driven by PETN-based explosive pellets will be tracked to determine the transient detonation behavior, and the predicted states from the burn model will be compared to previously published data.
The Extended History Variable Reactive Burn model (XHVRB), as proposed by Starkenburg, uses shock capturing rather than current pressure for calculating the pseudo-entropy that is used to model the reaction rate of detonating explosives. In addition to its extended capabilities for modeling explosive desensitization in multi-shock environments, XHVRB's shock capturing offers potential improvement for single shock modeling over the historically used workhorse model HVRB in CTH, an Eulerian shock physics code developed at Sandia National Labs. The detailed transition to detonation of PBX9501, as revealed by embedded gauge data, is compared to models with both HVRB and XHVRB. Improvements to the comparison of model to test data are shown with XHVRB, though not all of the details of the transition are captured by these single-rate models.
The use of S2 glass/SC15 epoxy woven fabric composite materials for blast and ballistic protection has been an area of on-going research over the past decade. In order to accurately model this material system within potential applications under extreme loading conditions, a well characterized and understood anisotropic equation of state (EOS) is needed. This work details both an experimental program and associated analytical modelling efforts which aim to provide better physical understanding of the anisotropic EOS behavior of this material. Experimental testing focused on planar shock impact tests loading the composite to peak pressures of 15 GPa in both the transverse and longitudinal orientations. Test results highlighted the anisotropic response of the material and provided a basis by which the associated numeric micromechanical investigation was compared. Results of the combined experimental and numerical modeling investigation provided insights into not only the constituent material influence on the composite response but also the importance of the plain weave microstructure geometry and the significance of the microstructural configuration.
Explosive shock desensitization phenomena have been recognized for some time. It has been demonstrated that pressure-based reactive flow models do not adequately capture the basic nature of the explosive behavior. Historically, replacing the local pressure with a shock captured pressure has dramatically improved the numerical modeling approaches. A pseudo-entropy based formulation using the History Variable Reactive Burn model, as proposed by Starkenberg, was implemented into the Eulerian shock physics code CTH. Improvements in the shock capturing algorithm in the model were made that allow reproduction of single shock behavior consistent with published Pop-plot data. It is also demonstrated to capture a desensitization effect based on available literature data, and to qualitatively capture multi-dimensional desensitization behavior. This model shows promise for use in modeling and simulation problems that are relevant to the desensitization phenomena. Issues are identified with the current implementation and future work is proposed for improving and expanding model capabilities.
