Explosive Closure Computational Analysis

Note: Information and images here are intentionally vague due to the confidential nature of much work at Sandia National Labs.

Background:

The Z-machine is the world’s most powerful pulsed power facility. It compresses energy in space and time to create and study extreme environments in support of the country’s nuclear weapon stockpile. One way Z supports stockpile stewardship is through the study of special nuclear material (SNM) under extreme conditions. When using radiological material such as SNM, the material must be isolated inside a containment chamber immediately after the experiment to mitigate the risk of exposure. This must be done with explosives due to the microsecond pulse-shaping requirements of shots on Z. The existing solution, called theUltrafast Closure Valve (UCV) uses a device raised up from the transmission lines (Figure 3), which explosively closes upon firing of the Z. The geometry of the device increases the total inductance of Z, which limits the total current.There is an LDRD effort for a low inductance closure device that aims to increase the current. This involves explosively welding the final feed gap to fully seal and prevent the spread of debris associated with the target. Many complex multi-physics analyses and tests must be conducted in order to verify the functionality of the system prior to use with actual nuclear material.

Analysis of CTH Multiphysics Simulations for Weld Verification:

Explosive Welding Criteria has been established experimentally based on the following variables:

• Beta Angle (at contact)
• Flyer Velocity

There is a combination of Beta Angle and Flyer Velocity at which two plates will weld, called the weld window. A CTH simulation of the explosion tracked 100 tracer points over the period of interested. This meant that 2,000,000 data points had to be analyzed to produce beta angle versus flyer velocity.

Python was used to analyze the data. some numbers used for the calculation were variable, but the second plot shows changes to those variables had no effect on the overall weld plots. The results produced reasonable confidence that the plates would weld at some points, though not all.

Explicit Dynamic Abaqus Analysis of the Baffle:

The shock wave from the explosives necessary to weld the MITL produces a large pressure pulse that must be contained. Thus, a baffle that reduces the velocity and pressure imparted on the bottom plate was designed. To ensure the survivability of the containment system, a CTH analysis was run to predict the pressure on the bototm plate over time. This impulse was used as the imput for a FEA analysis to evaluate the structural response of the system.

A real explosives test was run weeks after these analyses were completed.

Due to the incredibly sensitive nature of explosive simulations the CTH analysis which produced the data the rest of the analyses were based on was somewhat wrong when compared with the real data. However despite this, the FEA analysis still displayed a high level of accuracy in predicting the deformation shape (eyeballed) and velocity (PDV data) of the bottom plate upon exploding. The plates which I predicted would weld between two certain radii did not weld, however it was evident that they attempted to weld but the right conditions were not met due to a number of extraneous circumstances during the actual explosives test.