John M. Gahl

Investigation of a multichanneling, multigap Marx bank switch

There is increasing interest in the pulsed power community for understanding the complex phenomena of multichannel spark gap switches. Extracting the potential advantages of these switches is necessary for upgrades to large, high current, high voltage machines. A previously designed, multichanneling spark gap (MCSG), with toroidal electrodes, for Marx bank applications, is investigated to determine which factors effect its multichannel operation. To date, thorough research does not exist for multichanneling gas switches with toroidal electrode geometry. The following results are the first of their kind. Gas type, gas pressure, trigger voltage, and trigger polarity are varied to study their effects on multichanneling in the MCSG. The fill gas is varied between air and an Ar/SF6 (90%/10%) mixture, from 3 to 10 psig. There are more channels for the mixture than for air for similar pressures. Trigger voltage amplitude and polarity are varied negatively from −60 to −100 kV and positively from +75 to +95 kV. Tr...

Development of a terawatt test stand at the University of Missouri for fast, multichannel switching analysis

The University of Missouri Terawatt Test Stand (MUTTS) began assembly in January 2003. Construction of MUTTS is progressing rapidly with the design and development of its high energy Marx bank. The Marx bank consists of 32, 100 kV, 0.7 /spl mu/F capacitors switched by 16 Physics International T508 spark gaps. The Marx is switched into two parallel 7 nF, intermediate storage capacitors, which are fired into a dummy load through a fast multi-channeling output switch. The Marx stores 100 kJ and can deliver a voltage of 2 MV at 500 kA into a 4 /spl Omega/ load delivering 1 TW to the load. Initial testing will be of a multichanneling 2 MV output switch, which scales nicely to a 6 MV switch design for future very high energy machines at Sandia National Laboratories. The output switch is to reliably multichannel, or close with many parallel arc channels. The goal is to adapt an existing multichanneling switch to create a multichanneling output switch with significant operational advantages, including lower inductance, compared to existing multichannel switches. The target switch inductance is 100 nH or less. The facility and tank were assembled from January to June 2003, with testing to begin in July 2003. Simulations of the test stand and specifications of the output switch will be presented. Electrode configurations and switch augmentations that will facilitate a reliable multi-channeling switch will be introduced. Details describing the development of the MUTTS facility will be included.

Study of laser target triggering for spark gap switches

The use of laser targets as a method to decrease the required laser energy to trigger a laser triggered gas switch has been investigated at the University of Missouri. Target materials were identified based on durability, melting point, reactivity and reflection coefficient. Laser targets were placed into a cathode of a laser triggered gas switch. The switch was pulse charged by the Tiger pulsed power machine to between 185 kV and 330 kV. The switch was triggered by directing a 1064 nm or 266 nm wavelength laser pulse from an Nd:YAG laser onto a laser target to ablate material and create plasma, closing the switch. The goal of the project was to trigger a high voltage gas switch with less than 1 mJ of laser energy while maintaining a switch jitter comparable to present electrically triggered switches for LTD based systems. The study successfully demonstrated that triggering the switch using a 1 mJ infrared pulse and a graphite target resulted in a jitter less than 5 ns. Findings will be used in the design of switches for LTD based systems.

Effects of laser triggering parameters on runtime and jitter of a gas switch

Parameters affecting the runtime and jitter of a laser triggered gas switch have been studied. Experiments tested a variety of switch parameters including percentage of selfbreak and switch pressure. The effects of laser beam parameters were also considered, including focal length, laser energy, laser spark length, and laser wavelength. Experiments were performed on the Tiger pulsed power machine. Measurements were taken on a spark gap switch built from the trigger section of a Rimfire switch. A Marx bank consisting of 32, 3.1 uF, capacitors that fed into a 7 nF intermediate storage capacitor was used to drive the switch into a 4 ¿ resistive load. The test switch was pressurized to 306 kPa (30 psig) with SF6 and operated near 1 MV. A New Wave Tempest Nd:YAG laser was used to trigger breakdown of the switch. The laser was focused at the mid-gap between the switch electrodes using lenses with focal lengths between 30 cm and 100 cm. Focused laser energy in the switch ranged from <5 mJ to 80 mJ. The effects of switch and laser beam parameters on the runtime and jitter of a laser triggered gas switch are presented. The end goal of the research is to determine optimal conditions for improved switch performance.

Runtime and Jitter on a Laser-Triggered Spark-Gap Switch

The University of Missouri has completed a new facility, named Tiger, for pulsed-power experimentation. Tiger consists of a 2.8-MV 450-kJ Marx bank that charges up to four 7-nF intermediate storage capacitors (I-stores) in parallel. When charged, the storage capacitors are switched into a resistive load through an SF6-filled laser-triggered gas switch. This switch has been designed to study the factors affecting runtime and jitter of spark-gap switches. All experiments presented in this paper were performed with a single I-store. The test switch was operated from about 500 kV up to 1.25 MV, at switch pressures from 10 to 50 psig. A 30-mJ 266-nm Nd:YAG laser was focused between the switch electrodes to initiate breakdown in the switch. The University of Missouri has examined laser energy, percentage of self-break, and focal length to determine their relation to runtime and jitter. A short discussion of the Tiger facility is presented with experimental results of jitter and runtime tests. The end goal of this paper is to understand the factors contributing to increased jitter and runtime and, thereby, provide paths to improved switch performance.

Modeling and analysis of the rimfire gas switch

Many accelerators at Sandia National Laboratories utilize the Rimfire gas switch for high-voltage, high-power switching. Future accelerators will have increased performance requirements for switching elements. When designing improved versions of the Rimfire switch, there is a need for quick and accurate simulation of the electrical effects of geometry changes. This paper presents an advanced circuit model of the Rimfire switch that can be used for these simulations. The development of the model is shown along with comparisons to past models and experimental results.

Dielectric flashover with triple point shielding in a coaxial geometry.

Increasing performance of vacuum insulator barriers is a common goal in pulsed power. Insulating performance is continually being improved while new methods are developed. Triple point shielding techniques have been shown to increase flashover voltage, but the role of cathode versus anode shielding is still not fully understood. Open circuit flashover characteristics were obtained for a coaxial geometry to view the effects of triple point shielding for this geometry. The tests included applying various combinations of triple point shields on zero and +45 degrees insulators. Shielding was tested at the cathode triple point outside of the dielectric and at the anode triple point inside the dielectric. The role of anode versus cathode triple point shielding was examined. Flashover voltage was observed to increase when either a cathode or anode triple point shield was applied; however, adding a shield to both regions lowered the flashover threshold. Both triple point regions were found to be important and dependent on each other for some coaxial geometries.

Multichannel and Impedance Analysis of the Laser-Triggered Rimfire Gas Switch

Many pulsed power applications require multichannel switching because of stringent low erosion and low impedance requirements. Rimfire, which is a multigap multichanneling switch, has been implemented extensively for this purpose at Sandia National Laboratories for several decades, but with incomplete understanding. This paper presents a thorough experimental analysis of impedance and multichanneling characteristics for a multigap switch under the influence of a laser trigger and in an SF6 environment. The implications of these results on the future design of multichanneling multigap gas switches are also presented

Particle Swarm Optimization of Pulsed Power Circuit Models

Circuit modeling is ubiquitous throughout the pulsed power discipline. Both plasma processes and systems can be modeled with circuits of varying complexities. Sometimes, circuit models need to be generated to match experimental waveforms. Particle swarm optimization (PSO) is a technique which can be utilized to automatically generate a circuit model to match experimental data. This paper details the PSO algorithm as well as two case studies of the implementation.

Efficiency and Scaling in DC Electromagnetic Launchers

Efficiency and scaling relationships for DC (i.e., non- induction) electromagnetic launchers (EMLs) are presented and discussed. Efficiency and scaling relationships for these types of launchers are easily generalized since their principle of operation is the same, namely the production of an electromagnetic force via a spatial change in inductance. Electromagnetic force, efficiency, back-voltage, and kinetic power are given in terms of electrical circuit parameters. A comparative analysis of the efficiency and scaling relationships for these types of launchers is given and includes EML geometries such as conventional railguns, augmented railguns, and helical coilguns. The comparative analysis results are performed in terms of the launchers inductance gradient, physical dimensions (i.e., armature- stator diameter and length), resistive losses, local energy storage (i.e., inductive), and armature velocity. A survey of recent electromagnetic launcher results is performed and compared to the theoretical predictions.