Search

Axial current discharges were used to implode spherical aluminum shells with thickness tapered to match the external magnetic pressure. These implosions were used to compress hydrogen plasmas with an initial pressure $g100\mathrm{bar}$ and an initial temperature $g1\mathrm{eV}$. The implosion and compressed plasma effect on a central target were observed radiographically. The target's radius vs time indicates that the plasma pressure reaches $\ensuremath{\sim}1\mathrm{Mbar}$ prior to liner-target contact.

Electromagnetic implosions of shaped cylindrical aluminum liners that remain at solid density are discussed. The approximate liner parameters have an initial radius of 3 to 4 cm, are 4 cm in height, and are nearly 0.1 cm thick. The liners are driven by the Shiva Star 1300-{mu}f capacitor bank at an 84-kV charging voltage and an nearly 30-nH total initial inductance (including implosion load). The discharge current travels along the length of the liner and rises to 14 MA in nearly 8 {mu}s. The implosion time is nearly 12 {mu}s. Diagnostics include inductive current and capacitive voltage probes, magnetic probes, and radiography. Both right-circular cylinder and conical liner implosion data are displayed and discussed. Radiography indicates implosion behavior substantially consistent with two-dimensional magnetohydrodynamic calculations, which predict inner surface implosion velocities exceeding 20 km/s, and compressed density of two to three times solid density. Less growth of perturbations is evident for the conical liner (nearly 1% thickness tolerance) than for the right-circular cylindrical liner (nearly 3% thickness tolerance). 12 refs., 8 figs.

This paper presents scaling results based on simulations with the CALE hydrodynamics code [1] of aluminum projectile impacts on typical aluminum Whipple shields at speeds of 6 to 14 km/s.The objective was to determine the extent of projectile and target material melting. The approach was to perform a matrix of computer simulations varying the impact speed from 6 to 14 km/s and varying the areal density of the shield from 5 percent to 80 percent of the centerline areal density of the projectile. The projectile radius was fixed at 9.5 mm (mass = 1.27 grams). The melt state of the projectile material and the shield material was assessed after release of the initial shock. The post-release specific energy in the projectile and in the shield was compared with the enthalpy of incipient melt and the enthalpy of complete melt provided in the Hultgren Tables [2]. Material with specific energy greater than the enthalpy of complete melt was assumed to be fully melted; material with specific energy greater than the enthalpy of incipient melt but less than that of complete melt was assumed to be partially solid and partially melted mixed phase material with no strength; and material with specific energy less than the enthalpy of incipient melt was assumed to be in a solid state with strength. It is likely that this solid material is in a highly fragmented state as a result of the initial shock.

A Ranchero flux compression generator (FCG) was recently tested at the 76 MA level. Ranchero generators were designed to be cost effective high current devices, and a variety of configurations have been tested. The Ranchero armature is a 152 mm diameter aluminum cylinder with a 6 mm thick wall. The high explosive (HE) is detonated simultaneously on axis, and the armature is expanded by a factor of two. At the final 300 mm diameter, the circumference is over 950 mm, which has been anticipated to accommodate currents approaching 100 MA. The test was performed with a 43 cm long module having an initial inductance of 56 nH, and a load inductance of 0.55 nH, consisting of a short load feed slot and probe grooves. A 12 mF capacitor bank at ~16.5 kV provided the initial 3.75 MA seed current. The performance of the generator with this load was calculated using a 2D MHD ALE code, and agreement was excellent. Generator design, test data, and details of the MHD calculations are given. The most important aspect of the results is verification of codes at the 76 MA level. The same codes do not show excessive losses up to the 100 MA level, and it is a much smaller extrapolation to that level at this time.

Ranchero is a coaxial magnetic flux compression generator (FCG) initiated simultaneously on-axis, which is intended to produce currents approaching 100 MA. We continue with both applications and development, striving for both cost effectiveness and high performance. In this paper we discuss on-going work, provide details of the sophisticated detonation system, and show high current predictions from our active modeling effort. We intend further tests that will push the limits of Ranchero performance.

Experiments with coaxial plasma guns at currents in excess of ten megamperes have resulted in the production of high‐voltage pulses (0.5 MV) and hard x radiation (10–200 keV). The x‐radiation pulse occurs substantially after the high‐voltage pulse suggesting that high‐energy electrons are generated by dynamic processes in a very high speed (≳106 m/s), magnetized plasma flow. Such flows, which result from acceleration of relatively low‐density plasma (10−4 vs 1.0 kg/m3) by magnetic fields of 20–30 T, support high voltages by the back electromotive force‐u×B during the opening switch phase of the plasma flow switch. A simple model of classical ion slowing down and subsequent heating of background electrons can explain spectral evidence of 30‐keV electron temperatures in fully stripped aluminum plasma formed from plasma flows of 1–2 × 106 m/s. Similar modeling and spectral evidence indicates tungsten ion kinetic energies of 4.5 MeV and 46 keV electron temperatures of a highly stripped tungsten plasma.

Index-of-refraction changes induced by hydrodynamic activity have been blamed for poor beam quality in photodissociation lasers. Until recently, the pump field, hydrodynamics, chemical kinetics, and the losing field have been decoupled in theoretical models because of the wide range of time scales on which they act. Although these mechanisms occur at varying time scales ranging over seven orders of magnitude (subnanosecond to milliseconds), each mechanism strongly affects the others. In this paper we present a model that first couples the pump field, chemical kinetics, and hydrodynamics in the form of implicit and explicit finite-difference equations. We then couple in the laser field to show that index-of-refraction changes induced by radiative transfer degrade peak beam intensity

Several two-dimensional hydrocode simulations have been performed and analyzed to support the design and interpretation of experiments involving hypervelocity projectile impact onto multi-shock shield configurations.1,2 These simulations, performed with the Lawrence Livermore National Laboratory (LLNL) CALE3 hydrocode, demonstrate the increased projectile heating produced by the multi-shock concept as compared to the projectile heating by the first sheet of a conventional dual sheet “Whipple shield.” Detailed comparisons have been made of simulation results with experimental results obtained at the NASA Johnson Space Center (NASA JSC) Hypervelocity Impact Research Laboratory1 and at the Impact Physics Laboratory of the University of Dayton Research Institute (UDRI).2 Also described are bumper thickness and projectile velocity scaling studies.

A general purpose, multi-dimensional material model typically used for metals has been used to characterize the shock loading response of B4C, SiC and TiB2 which are considered candidate ballistic armor materials. The model predictions show excellent agreement with measured particle velocity or stress waveforms recorded during uniaxial shock compression experiments. Unique strength behavior has been observed including significant pressure dependence for TiB2. Analysis of the ceramic material behavior suggests that a) the Hugoniot Elastic Limit (HEL) is not a useful figure of merit (FOM) for ranking ceramics as armor materials and b) ceramic modeling which assumes perfectly plastic behavior can lead to significant error in predicted stress levels and wave propagation velocities. A more useful FOM for ceramics that normalizes strength at an operating stress by the material's density (σ*/ρ) is proposed as an alternative.

A study of the atmospheres of OB supergiants in x-ray binary systems indicates that when the stellar surface is close enough to the saddle in the gravitational potential to provide a mass transfer rate adequate to power a compact x-ray source, large-amplitude variations in the rate of mass flow occur. 9 references. (auth)

The plasma flow switch utilizes the nonlinear and nonuniform dynamics of a plasma discharge in vacuum to accumulate magnetic energy in times of several microseconds and then release this energy to a load in times of a few hundred nanoseconds. Experiments have been performed with capacitor banks up to 6 MJ, providing currents in excess of 107 A and peak voltages over 0.5 MV. Theoretical models include simple slug dynamics coupled to lumped-circuit analyses, magnetoacoustic considerations of one- and two-dimensional aspects of the plasma flow, and two-dimensional magnetohydrodynamic code calculations. The present article reviews both experimental and theoretical efforts, discusses the use of the plasma flow switch to drive plasma liner implosions and high-energy ion flows, and indicates directions for plasma flow switch applications to very high current, high-energy inductive pulsed power systems.

We have performed numerical studies of self‐oscillation instabilities which can result from the upstream‐downstream feedback in a gas‐dynamic laser (GDL) with an unstable resonator. In this letter, which extends previous work by Dreizen and Dykhne, we show that inclusion of a simple model for CO2 pumping by N2 strongly damps any oscillations in the output intensity.

Using a 1313-?F, 3-nH, 120-kV, 9.4-MJ SHIVA Star capacitor bank, we have performed vacuum inductive store/plasma flow switch (PFS) driven implosions of low mass (200-400 ?g/cm2) cylindrical foil liners of 2-cm height and 5-cm radius. This technique employs a coaxial discharge through a plasma armature, which stores magnetic energy over 3-4 ?s and rapidly switches it to an imploding load as the plasma armature exits the coaxial gun muzzle. The current transferred to the load by the PFS has a rise time of less than 0.2 ?s. With 5-MJ stored energy, we have driven fast liner implosions with a current of over 9 MA, obtaining an isotropic equivalent 2.7-TW 0.5-MJ X-ray yield.

We have studied the ''beginning Roche lobe overflow'' phase of mass transfer. We find that before the primary fills its Roche lobe, radiation-driven density waves generated in the atmosphere produce a sufficient mass outflow rate to power a compact X-ray source. In particular, for a model of Cygnus X-1, if the radius of the photosphere is between 82 and 86 percent of the radius of the Roche equipotential, large-amplitude density waves with irregular periods of a few days are observed. These density waves, which are generated by radiation pressure effects, pass through the inner Lagrangian point with velocities approx.100 km s/sup -1/, and can easily provide sufficient material to power the compact X-ray source. (AIP)