On June 30, 2020, a 0.87 gram PETN charge being pressed in the Rapid Prototyping Facility (RPF), unexpectedly initiated, resulting in destruction of the pressing fixture but no injuries or facility damage. In response, the Safety Review Board (SRB) met on Aug. 13, 2020 and Oct. 1, 2020 to review information collected following the incident, consider likely direct causes, and form recommendations.
Combustion and Flame
Validated models of melt cast explosives exposed to accidental fires are essential for safety analysis. In the current work, we provide several experiments that can be used to develop and validate cookoff models of melt cast explosives such as Comp-B3 composed of 60:40 wt% RDX:TNT. We present several vented and sealed experiments from 2.5 mg to 4.2 kg of Comp-B3 in several configurations. We measured pressure, spatial temperature, and ignition time. Some experiments included borescope images obtained during both vented and sealed decomposition. We observed the TNT melt, the suspension of RDX particles in the melt, bubble formation caused by RDX decomposition, and bubble-induced mixing of the suspension. The RDX suspension did not completely dissolve, even as temperatures approached ignition. Our results contrast with published measurements of RDX solubility in hot TNT that suggest RDX would be completely dissolved at these high temperatures. These different observations are attributed to sample purity. We did not observe significant movement of the two-phase mixture until decomposition gases formed bubbles. Bubble generation was inhibited in our sealed experiments and suppressed mixing.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Society for Experimental Mechanics - SEM Annual Conference and Exposition on Experimental and Applied Mechanics 2010
The "DaMaGe-Initiated-Reaction" (DMGIR) computational model has been developed to predict the response of ideal high explosives to impulsive loading from non-shock mechanical insults. The distinguishing feature of this model is the introduction of a damage variable, which relates the evolution of damage to the initiation of a reaction in the explosive, and its growth to detonation. This model development effort treats the non-shock initiation behavior of explosives by families; rigid plastic bonded, cast, and moldable plastic explosives. Specifically designed experiments were used to study the initiation process of each explosive family with embedded shock sensors and optical diagnostics. The experimental portion of this model development began with a study of PBXN-5 to develop DMGIR model coefficients for the rigid plastic bonded family, followed by studies of the cast, and bulk-moldable explosive families, including the thermal effects on initiation for the cast explosive family. The experimental results show an initiation mechanism that is related to impulsive energy input and material damage, with well defined initiation thresholds for each explosive family. These initiation details will be used to extend the predictive capability of the DMGIR model from the rigid family into the cast and bulk-moldable families. © 2010 Society for Experimental Mechanics Inc.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
A simultaneous PVDF/VISAR measurement technique was used for isentropic-loading experiments with a polymethyl methacrylate (PMMA) specimen. The experiments used a graded density impactor accelerated onto a tantalum driver backed with PMMA and then lithium fluoride windows for each experiment. Simultaneous measurements made at each window interface provided precise transit time and particle velocity measurements which can be used to determine the stress-vs-strain loading path using Lagrangian analysis techniques. The experimental technique provides access to 40 GPa stress levels in PMMA under isentropic-loading conditions.
This paper describes recent work to make high quality quartz gauge (temporal and spatial) shock wave measurements in a pulsed ion beam environment. Intense ion beam radiation, nominally 1 MeV protons, was deposited into material samples instrumented with shunted quartz gauges adjacent to the ion deposition zone. Fluence levels were chosen to excite three fundamentally different material response modes (1) strong vapor, (2) combined vapor and melt phase and (3) thermoelastic material response. A unique quartz gauge design was utilized that employed printed circuit board (PCB) technology to facilitate electrical shielding, ruggedness, and fabrication @e meeting the essential one dimensional requirements of the characterized Sandia shunted quartz gauge. Shock loading and unloading experiments were conducted to evaluate the piezoelectric response of the coupled quartz gauge/PCB transducer. High fidelity shock wave profiles were recorded at the three ion fluence levels providing dynamic material response data for vapor, melt and solid material phases.
Piezoelectric polymer stress gauges in copper fixtures were used with te Sandia 2.5-inch bore gas gun to obtain time-resolved pressure measurements for two polytetrafluoroethylene powders having significantly different particle morphologies. The powders had approximate average particle sizes of 534 microns and 28 microns, respectively, and scanning electron microscopy revealed differences in the appearances of representative particle surfaces. The range of input stresses was from 0.13 GPa to 2.81 GPa, and the initial densities were 57% of the solid density. The ``crush strength`` (pressure required to compress the porous compact to solid density) was close to 1.0 GPa for the coarse material as compared to 0.6 GPa for the finer material. At an input stress of about 0.6 GPa, the risetime of the propagated stress waves in the coarse material was approximately 240 nsec compared to 50 nsec for the finer material. These measurements show the strongly rate-dependent deformation of the powders and that particle morphology has a significant effect on the shock compression.