Measurements and ALE3D Simulations for Violence in a Scaled Thermal Explosion Experiment with LX-10 and AerMet 100 Steel
We completed a Scaled Thermal Explosion Experiment (STEX) and performed ALE3D simulations for the HMX-based explosive, LX-10, confined in an AerMet 100 (iron-cobalt-nickel alloy) vessel. The explosive was heated at 1 C/h until cookoff at 182 C using a controlled temperature profile. During the explosion, the expansion of the tube and fragment velocities were measured with strain gauges, Photonic-Doppler-Velocimeters (PDVs), and micropower radar units. These results were combined to produce a single curve describing 15 cm of tube wall motion. A majority of the metal fragments were captured and cataloged. A fragment size distribution was constructed, and a typical fragment had a length scale of 2 cm. Based on these results, the explosion was considered to be a violent deflagration. ALE3D models for chemical, thermal, and mechanical behavior were developed for the heating and explosive processes. A four-step chemical kinetics model is employed for the HMX while a one-step model is used for the Viton. A pressure-dependent deflagration model is employed during the expansion. The mechanical behavior of the solid constituents is represented by a Steinberg-Guinan model while polynomial and gamma-law expressions are used for the equation of state of the solid and gas species, respectively. A gamma-law model is employed for the air in gaps, and a mixed material model is used for the interface between air and explosive. A Johnson-Cook model with an empirical rule for failure strain is used to describe fracture behavior. Parameters for the kinetics model were specified using measurements of the One-Dimensional-Time-to-Explosion (ODTX), while measurements for burn rate were employed to determine parameters in the burn front model. The ALE3D models provide good predictions for the thermal behavior and time to explosion, but the predicted wall expansion curve is higher than the measured curve. Possible contributions to this discrepancy include inaccuracies in the chemical models, integration of the momentum equation, and representation of the interfaces in the gaps. Two model problems were used to explore the effects of parameter variations on the fracture and fragmentation of AerMet 100 tube sections driven by the deflagration of LX-10. For the range of parameters considered, the model fragment sizes are of the same scale as the measured sizes.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 877809
- Report Number(s):
- UCRL-CONF-212828; TRN: US200608%%707
- Resource Relation:
- Conference: Presented at: Joint Army-Navy-NASA-Air Force 40th Combustion Subcommittee/28th Airbreathing Propulsion Subcommittee/22nd Propulsion Systems Hazards Subcommittee/4th Modeling & Simulation Subcommittee Meeting, Charleston, SC, United States, Jun 13 - Jun 17, 2005
- Country of Publication:
- United States
- Language:
- English
Similar Records
Measurement and ALE3D Simulation of Violence in a Deflagration Experiment With LX-10 and Aermet-100 Alloy
ALE3D Model Predictions and Experimental Analysis of the Cookoff Response of Comp B*