The development of Explosively Formed Fuses along two separate lines is discussed. One design, which has previously been demonstrated to conduct a 9.5 MA 350 .mu.s risetime pulse and interrupt it in 1.2 .mu.s. This scaled up design should operate at up to 15 MA with 20 nH loads. A second design with enhanced performance characteristics is being examined and will be tested on a small scale. This design includes opening switch inductance as part of the inductive store and, as a result, should have shorter pulse transfer times and should be able to be scaled to handle currents up to approx. 25 MA with 20 nH loads.
Pulsed power technology, in the simplest of terms, usually concerns the storage of electrical energy over relatively long times and then its rapid release over a comparatively short period. However, if we leave the definition at that, we miss a multitude of aspects that are important in the ultimate application of pulsed power. It is, in fact, the application of pulsed power technology to which this series of texts will be foc~sed. Pulsed power in today's broader sense means "special power" as opposed to the traditional situation of high voltage impulse issues related to the utility industry. Since the pulsed power field is primarily application driven it has principally engineering flavor. Today's applications span those from materials processing, such as metal forming by pulsed magnetic fields, to commercial applications, such as psychedelic strobe lights or radar modulators. Very high peak power applica tions occur in research for inertial confinement fusion and the Strategic Defense Initiative and other historical defense uses. In fact it is from this latter direction that pulsed power has real ized explosive growth over the past half century. Early thrusts were in electrically powered systems that simulated the environment or effects of nuclear weapons detonation. More recently it is being utilized as prime power sources for directed energy weapons, such as lasers, microwaves, particle beam weapons, and even mass drivers (kinetic energy weapons).
High explosive pulsed power (HEPP) systems are capable of generating very high energies in magnetic fields. Such stored energy is usually developed on time scales of a few tens or hundreds of microseconds. Many applications require shorter pulses and opening switches provide one way to use the large energy available for faster applications. With current flowing in an inductive circuit, introducing resistance produces voltage that can be used to drive current into a load. For an opening switch with a fast rising resistance, the load current rise time is determined by the R/L time constant of the circuit. A significant fraction of the circuit energy must be dissipated in the process, and in applications where very large energies must be dealt with only a few types of switches can be used. Experiments with high explosive driven opening switches have produced a few switches that can carry tens of MA current, and open on the time scale of one or a few [mu]s. [sup 1] We have specialized in a type of switch that we call an explosively formed fuse (EFF) switch at levels of [approximately]3 TW for 2[mu]s has become routine, and we will describe its characteristics and give data from a number of tests.
In this paper we describe results from tests of an explosive pulsed power system designed to deliver 15--16 MA to a plasma flow switch (PFS). The PFS, in turn, has the goal of switching current to a z- pinch load to produce a 1-MJ implosion for x-ray generation experiments. The system consists of a MK-IX magnetic flux-compression generator, a coaxial inductive store, an explosively formed fuse (EFF) opening switch, and a vacuum power flow/PFS assembly. Figure 1 shows a completed assembly ready to test. Computational modeling of this system is described in another paper in this conference, and important design considerations have been previously published. Vacuum diagnostics are also discussed in a separate paper in this conference as are results from a test in which a conventional foil-fuse opening switch replaced the EFF. We have performed two development tests of the Procyon system. A preliminary reduced energy test (Shot 1) delivered (approximately)13.6 MA to a 25-nH PFS load, and imposed a large voltage spike on the EFF at nominal pinch time without failure. In a full-energy test (Shot 2), the system delivered 20 MA to the EFF without suffering unexpected losses, and demonstrated the proper onset of EFF opening. In the 20-MA test, mistiming between the EFF and the load isolation switches led to transmission line failure that disguised late time opening switch performance and diverted most of the current pulse away from the PFS load. These two tests have provided important system characterization information. In some cases design expectations are confirmed and in others adjustments to initial expectations are called for. Performance details are presented below. 8 refs., 13 figs.
Explosive-driven magnetic flux compression generators (explosive generators) provide for the generation of large amounts of energy compactly stored in a magnetic field. Opening switches for use in explosive generator circuits allow the energy to be used for applications requiring higher power than can be developed by the generators themselves. We have developed a type of opening switch that we describe as an explosively formed fuse (EFF). These switches are well suited to explosive generator circuits and provide a considerable enhancement of explosive pulsed-power capability. 10 refs., 14 figs.
Many pulsed power experiments need pulse shaping to optimize the power flow from a flux compression generator (FCG) to an experimental load. In a laboratory environment this can be a simple task where the switches are not destroyed. However, in experiments with high explosives, where a large amount of damage occurs, a single use EFF opening switch may be a good choice. In an EFF, explosives are used to thin a current carrying sheet of aluminum as it is forced into a grooved dye. The current is modified by the time dependent changes in resistance as the aluminum is stretched. We will correlate the hydrodynamic effects with resistance. The hydrodynamic profile is determined by Mesa-2D, a well proven hydrodynamics computer code, and MA THEMA TICA is used convert material contours into total resistance using the resistivity as a function of time from various sources. Experimentally, we will determine the actual resistance and compare it with the calculated values. We have used these switches for decades but still do not understand the details of the physics. The resistance change may be due to several processes but in this paper we will concentrate on stretching as the most important contribution. Also, in this paper we will compare the details of the hydrodynamics with the details of experimental and calculated resistance and hopefully generate a predictive model for future designs with other geometries and materials.
At Los Alamos, the authors have primarily applied Explosively Formed Fuse (EFF) techniques to high current systems. In these systems, the EFF has interrupted currents from 19 to 25 MA, thus diverting the current to low inductance loads. The magnitude of transferred current is determined by the ratio of storage inductance to load inductance, and with dynamic loads, the current has ranged from 12 to 20 MA. In a system with 18 MJ stored energy, the switch operates at a power up to 6 TW. The authors are now investigating the use of the EFF technique to apply high voltages to high impedance loads in systems that are more compact. In these systems, they are exploring circuits with EFF lengths from 43 to 100 cm, which have storage inductances large enough to apply 300 to 500 kV across high impedance loads. Experimental results and design considerations are presented. Using cylindrical EFF switches of 10 cm diameter and 43 cm length, currents of approximately 3 MA were interrupted producing [approximately]200 kV. This indicate s the switch had an effective resistance of [approximately]100 m[Omega] where 150--200 m[Omega] was expected. To understand the lower performance, several parameters were studied, including: electrical conduction through the explosive products; current density; explosive initiation; insulator type; conductor thickness; and so on. The results show a number of interesting features, most notably that the primary mechanism of switch operation is mechanical and not electrical fusing of the conductor. Switches opening on a 10 to 10 [micro]s time scale with resistances starting at 50 [micro][Omega] and increasing to perhaps 1 [Omega] now seem possible to construct, using explosive charges as small as a few pounds.
Taking advantage of the high energy density attainable in the magnetic field of an inductor requires a prime power source capable of producing very large currents and a means of extracting the energy as a fast current pulse from the inductive store. Many existing high current sources have pulse risetimes of hundress of microseconds, while most pulsed power applications have submicrosecond pulse requirements. In principle, high current opening switches represent a good solution to the problem. An inductor is charged over a long period of time by a relatively slow current supply with a closed switch completing the circuit. At a desired time, the switch is opened and the voltage produced transfers current rapidly to a load circuit. In practice, building opening switches that will carry multimegampere currents for hundreds of microseconds and then open on a submicrosecond time scale has posed an extremely difficult problem. Explosive-driven opening switches have been used in long-pulse applications for some time, but until recently the explosives systems used to drive these devices would not produce a rapidly opening switch. We have developed a fast technique for interrupting large currents by using explosives to extrude short sections of relatively thick conductors into long thin fuse-like conductors. Although the formation of the fuse requires about 2 .mu.s, the switch will sustain considerable voltage as its resistance rises, and it is feasible to deliver pulses with approx. 1 .mu.s risetimes to low inductance loads. We discuss here the small scale proof of principle experiments and 2-D hydrodynamics calculations that have led to an optimized cylindrical opening switch design. In addition, we will describe the results of testing a cylindrical switch at currents up to 4.6 MA, and give extrapolations for device designs for the 15 to 20 megampere range.
