Texas Tech University
Center for Pulsed Power & Power Electronics

Authors: Tyler Watson; James Dickens; Andreas Neuber; John Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310794

Abstract: Flux compression generators (FCGs), are unable to drive high impedance loads efficiently. To ensure that the output from an FCG can drive a high impedance load, an impedance transformation is necessary. This impedance transformation is known as the power conditioning stage of the FCG. Air core transformers were developed to eventually be used as a power conditioning stage for an FCG. Two types of air core transformers were developed, known as the C-TACT and COAX-ACT. The design and fabrication of these types of ACTs are discussed, along with experimental results and simulation. Finally, a comparison is made between the developed ACTs.

Authors: Max Flynn; John Agan; Andreas Neuber; Jacob Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310706

Abstract: A genetic algorithm procedure used to generate and optimize electron-neutral collision cross-sections for C4 F7N is detailed. The method creates a set of cross-sections from parameterized shapes available from literature, Born approximations, and cubic splines. The most fit cross-sections generated in this way are those which, when input to a multi-term Boltzmann equation solver, best reproduce experimental drift velocity data. The cross-sections generated in this way are compared with cross-sections calculated using the R-matrix method software Quantemol-EC.

Authors: David Saheb; Travis Wright; John Mankowski; James Dickens; Andreas Neuber; Emily Schrock; James Schrock; Jacob Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310677

Abstract: Nonlinear transmission lines (NLTLs) are a promising technology for high power microwave (HPM) generation. However, NLTLs are also typically characterized by relatively short RF pulse widths, on the order of tens to hundreds of nanoseconds. One approach to increasing the overall average microwave power is the application of short excitations in very high pulse repetition frequency, thus yielding a high effective RF duty cycle. This research details the development of a compact pulsed power driver capable of delivering up to 3 kV high voltage excitation, with pulsed widths between 100 – 500 ns, and multi-MHz pulse repetition frequency to a PCB-based NLTL HPM source.

Authors: Jake Slattery; Austin T. Hewitt; Brandon Bywater; John Mankowski; James Dickens; Andreas Neuber; Donald Friesen; David Hattz; Craig Noltensmeyer; Neil Koone; Crystal Nelson; Jacob Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10376245

Abstract: Mitigating electric and magnetic fields induced by lightning strike events is critical for the security and longevity of electrical components. Electromagnetic interference originating from a lightning strike is categorized into the low to medium frequency ranges, commonly referred to as the “magnetic coupling’’ regime. This low-frequency characteristic brings about the problem of shielding magnetic flux lines. Such shielding can be accomplished with highly conductive materials via the generation of eddy currents or by diverting the flux lines using materials with a large relative permeability. The shielding effectiveness of materials with varying thicknesses and geometry are tested using a pulse forming network having comparable characteristics to a scaled lighting strike. A four-stage Marx generator with a peak current of $\sim$33kA and a risetime of 1.4 $\mu$s is used as the pulse forming network to emulate the scaled electric and magnetic fields. The data collected from these experiments provide information and techniques that can be implemented into the design of a portable, deployable shielding enclosure.

Authors: Brandon Bywater; John Mankowski; James Dickens; Andreas Neuber; Matthew Capps; Jacob Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310949

Abstract: Sulfur Hexafluoride’s (SF6) high dielectric strength makes it heavily relied upon as a high voltage insulating medium. However, its atmospheric lifetime of 3,200 years and overall high global warming potential has motivated a search for alternatives. One possible alternative, C4F7N (NOVEC 4710) features a dielectric strength more than twice that of SF6 yet features an atmospheric lifetime one-hundredth that of SF6. Previous research concerning breakdown characteristics of C4F7N has been primarily limited to long-timescale DC and 60 Hz AC conditions. In this study, the breakdown strength of gas mixtures comprised of 5%, 10%, and 20% C4F7N in a N2 or CO2 background were tested under pulsed, microsecond high power microwave (HPM) conditions over a range of 16 psia to 32 psia at room temperature. To achieve this, a 3 MW S-band (2.85 GHz) high-power magnetron is used to excite a traveling-wave ring resonator, which provides a nominal effective HPM power level of ~20 MW. A stepped impedance transformer is used to further increase the electric field to approximately 160 kV/cm RMS. For comparison, the reported dielectric strength of C4F7N is compared against the dielectric strength of pure SF6.

Authors: M. Flynn; A. Neuber; J. Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10481159

Abstract: Anisotropic scattering is implemented in a multi-term Boltzmann equation solver (MultiBolt v3.1.0) for the modeling of electron-kinetic behavior considering both tabulated and analytic differential cross sections (DCSs). Collision terms for isotropic scattering, forward scattering, and screened-Coulomb scattering are presented and employed for elastic and inelastically scattered electrons as well as secondary electrons generated from ionization events. Swarm parameters (electron mobilities, velocities, reaction rates, and diffusion coefficients) are calculated and compared against experimental data to assess the influence of different scattering treatments on the macroscopic observable quantities. The assumption of isotropic and forward scattering models, as an approximation of the DCS more broadly, is evaluated for these conditions.

Authors: M. Mounho; R. Clark; W. Brooks; M. Hopkins; A. Neuber; J. Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10481136

Abstract: Anode-initiated high-voltage vacuum surface flashover is a prominent failure point in large-scale (megavolt and mega-ampere) pulsed power machines. To investigate the underlying basic mechanisms of anode-initiated vacuum surface flashover, a moderate-scale test platform was developed at the hundreds of kilovolt level to emulate the conditions in large-scale machines. This system comprises a positive 45-degree wedge of insulator placed between a hemispherical anode with a 2-cm radius and a planar cathode, approximating a normalized electric field profile found in large-scale pulsed power insulator stacks. An 8-stage Marx generator that produces a pulse up to 240 kV with a ~50 ns risetime is used to provide the high voltage excitation; experiments are conducted across 2 to 6 mm sized flashover gaps at a ~ 10−6 Torr background pressure. This study reports the statistical characterization of high voltage vacuum surface flashover, high speed and time-integrated imaging of flashover development, and temporally resolved emission spectroscopy from 200 nm to 600 nm of self-produced light emanating from these flashover events. Future steps towards an upgraded experimental platform with 600 kV capability are discussed.

Authors: D. Wright; A. Gregory; H. Spencer; J. Mankowski; J. Stephens; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10481093

Abstract: The increased need for multi-carrier signals and higher power requirements has made it essential to study the multipactor (MP) phenomenon in practical structures. A 3-stage rectangular waveguide filter has been designed and implemented in a plug-and-play test fixture for X-band frequencies. The test source for this system is a coaxial magnetron, which yields a peak power output of 150 kW at a frequency of 9.4 GHz and a pulse width ranging from 0.25 to 2.5 μs. Global power diagnostics and phase detection methods were employed to detect MP in the system.

Authors: Matthew Hopkins; William Brooks; Raimi Clark; Zakari Echo; Ronald Goeke; Christopher Moore; Michael Mounho; Andreas Neuber; Jacob Stephens; Jacob Young

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10200954

Abstract: This presentation describes a new effort to better understand insulator flashover in high current, high voltage pulsed power systems. Both experimental and modeling investigations are described. Particular emphasis is put upon understand flashover that initiate in the anode triple junction (anode-vacuum-dielectric).

Authors: John Mockert; John Mankowski; Benedikt Esser; Donald Friesen; Blake Havens; David Hattz; James Dickens; Crystal Nelson; Jacob Stephens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310963

Abstract: A system to map surface charge densities on dielectric objects – e.g. PTFE, PMMA, and PA6 – is described to study maximum surface charge limits discharge and decay characteristics. Two electromechanical movements are used to achieve these goals: a three-axis robotic arm with an electrostatic probe and a three-axis auxiliary movement fitted with a current viewing resistor and laser displacement sensor. Dielectric objects no larger than a cylinder of eight-inch diameter and twelve inches in height are analyzed to create surface point clouds and generate a set of scan points. The arbitrary object is positioned upon a rotational axis with a laser displacement sensor oriented radially towards the rotational axis of the platform to determine the object’s physical limits.

Authors: Michael Giesselmann; Jon Mayes

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310878

Abstract: We are reporting on a capacitor charger with a rated peak voltage of 50kV. The charger is fed from a standard 208V, 3-phase AC utility outlet. The HV package, consisting of a high frequency transformer, HV rectifier and HV feedback voltage divider was provided by Stangenes Industries and is contained in a 19 inch, 5U rack mountable package. The primary of the transformer is driven by an H-bridge inverter. Inverter current control is achieved through a Digital Signal Controller (DSC) with a fast PWM controller and a programmable analog comparator. The DSC provides cycle-by-cycle peak current mode control as well as current loop stabilization through digital slope compensation. The charger is controlled by an Industrial controller with a touch screen interface. The touch screen interface involves multiple programmable and customizable screens. The controller monitors the output voltage of the load capacitor and closes the voltage loop. A back-up safety function limits the charge time should the voltage loop be broken or not work properly.

Authors: Michael Mounho; Raimi Clark; William Brooks; Matthew Hopkins; Andreas Neuber; Jacob Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310812

Abstract: Statistical characterization of vacuum surface flashover is achieved by electrically stressing a positive 45° wedge of cross-linked polystyrene (Rexolite) with a fast, high-voltage pulse. The insulator is placed between a hemispherical anode with a 2 cm radius extending into a parallel plate planar cathode. This configuration is intended to confine the flashover plasma to a ~1 cm wide region on the surface of the insulator for repeatable optical access. The voltage source comprises an 8-stage Marx generator that can provide a pulse up to 264 kV with a ~50 ns risetime. Two capacitive voltage dividers are used as voltage diagnostics for the system. They are placed on the high-voltage coaxial cable entering the vacuum chamber and on the ground side of the flashover gap. A current viewing resistor (CVR) is placed into the outer shield of the high-voltage coaxial cable as the current diagnostic. A DSLR camera is employed to capture a time-integrated image of the flashover event. Pulse voltage, current, and risetime statistics are collected and reported for a statistically significant sample size for a 2.5 mm gap.

Authors: Raimi Clark; Michael Mounho; William Brooks; Matthew Hopkins; Jacob Stephens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10310987

Abstract: Temporally resolved emission spectroscopy capturing the light emission from 200 to 800 nm is utilized to assess the behavior of pulsed anode-initiated surface flashover in vacuum. The limiting failure mechanism at vacuum-insulator interfaces tends to be surface flashover since it occurs at field thresholds lower than bulk insulator breakdown or bridging plain vacuum gaps. Of particular interest are insulator geometries whose surface is angled to the electric field such that the electrons are pulled away from the surface. This often-dubbed “positive” surface geometry is known to outperform planar or negative angle geometries under most circumstances. This research examines a positive 45° insulator system subject to conditions that emulate those experienced in the insulator stacks of large-scale pulsed power machines, such as Sandia National Laboratory’s Z-machine: voltage risetimes in the tens of nanoseconds, pressures in the single microTorr, and peak electric fields of hundreds of kilovolts per centimeter at the anode triple junction. Cross-linked polystyrene (Rexolite) insulators were tested under these conditions and investigated with spatiotemporally resolved spectroscopy. Carbon ions are identified which originate from either the insulator itself or from adsorbed gases, as well as aluminum and magnesium ions from the metal electrodes. Simulated spectral lines of the carbon ions show agreement with the measured spectra for a Boltzmann temperature of 2.75 eV, while metal species require a reduction in temperature of at least 1 eV in order to match. These results lend some insight into the non-equilibrium nature of the early stages of anode-initiated flashover.

Authors: R. Bhatta; R. Shrestha; C. Negri; K. Schmitt; M. Musraf; M. Chamana; O. Illham; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9817507

Abstract: Microgrids are in transition from the laboratory to real-world deployment. The key driving factor includes the improvement in the technology used, a significant reduction in cost factor, well-established and proven results, and the increasing recognition of utility towards green and eco-friendly generation. The advantages of microgrid adoption range from resiliency, availability, scalability to diversity of generation. The current electric framework is starting to 'decentralize, decarbonize, and democratize' from top to bottom. The factors driving microgrid development and deployment towards in-site locations with existing electrical grid infrastructure fall into three sectors: Availability, Cost, and Clean Energy. The novelty of this work is to analyze the benefits of microgrids in extreme weather events that may cause brownout or backout events. This paper presents a post-mortem analysis of the Texas Grid Outage event. It investigates the advantages of a real-world test microgrid adoption to meet the resiliency benefits under extreme conditions. Besides resiliency, the work presents the benefits of microgrid participation in electricity markets during critical events.

Authors: Henry John Gaus III; David H. Barnett; Andreas A. Neuber; John J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10099398

Abstract: An experiment for imaging detonators used in explosively driven pulsed power applications with high-speed, short-exposure-time cameras will be described in the work to follow. Three, commercially available, high intensity, pulsed xenon light sources (> 107 candela intensity) yielded unsatisfactory image quality with a minimum exposure rate (~320 k frames per second). Above 320-k frames per second, the combined output intensity of the light sources was too dim. Due to this limitation, a lamp system was designed that would be capable of delivering higher light intensity to the target. Two types of lamp arrays were designed and tested. The first was a large lamp array comprised of a two high-energy flash lamps, while the second was a small lamp array comprised of many low-energy flash lamps. The large lamp array was intended for multiple shot use and was placed behind a protective sheet of polycarbonate to separate the bulbs from the detonator. The second small lamp array with low-cost flash lamps was intended for one-time use and will be placed closer to the detonator. Multiple five-stage, Rayleigh Pulse Forming Networks (PFNs) were developed to find the optimal energy for the flash lamp array. Each PFN was modeled using LTSpice circuit simulator to verify proper operation and help with optimization. Experimental measurements were taken of the PFN voltage and current outputs and compared to simulated values. A photodiode was used to measure relative light intensity from the different lamp arrays.

Authors: A. Balal; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9772041

Abstract: This paper makes use of electric vehicles (EVs) that are simultaneously connected to the Photovoltaic Cells (PV) and the power grid. In micro-grids, batteries of the electric vehicles (EVs) used as a source of power to feed the power grid in the peak demands of electricity. EVs can help regulation of the power grid by storing excess solar energy and returning it to the grid during high demand hours. This paper proposes a new architecture of micro-grids by using a rooftop solar system, Battery Electric Vehicles (BEVs), grid connected inverters, a boost converter, a bidirectional half-bridge converter, output filter, including L, LC, or LCL, and transformers. The main parts of this micro-grid are illustrated and modeled, as well as a simulation of their operation. In addition, simulation results explore the charging and discharging scenarios of the BEVs.

Authors: Nicholas Harrison; Kaylee Allen; James C. Dickens; Andreas A. Neuber; John Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10099412

Abstract: The RADAN series-based MG-4 Microwave Generator is a compact, high power microwave system developed by the Institute of Electrophysics in Ekaterinburg. The system features the RADAN high voltage generator which is a SINUS-series device featuring a Tesla transformer charger and a Blumlein pulse forming line. The MG-4 microwave head is a mm-band relativistic backward wave oscillator (BWO) that operates at 35 GHz with a 5 to 10 MW peak output power and a pulse width of 3 ns. To confirm the RF output power level of the MG-4, a commercially available RF envelope detector was employed. Analog Devices ADL6012 is a broadband envelope detector that operates from 2 GHz to 67 GHz at input powers up to $+15 \ \text{dBm}$, It also features a 500 MHz envelope bandwidth and 0.6 ns output risetime capability. The diagnostic setup features the ADL6012-EV ALZ, an evaluation board with the ADL6012 offered by Analog Devices, shielded in a fitted brass box located in the far field $(\sim 60$ cm) from the microwave output horn. The output mode of the MG-4 is nominally TM0l, but a mode convertor allows for a TE11 output mode as well. The positive and negative envelopes of the pulse are captured separately. Based on the peak differential output voltage of the positive and negative signal, the input power of the detector can be determined by the typical performance characteristics curves in the ADL6012 data sheet. Lastly, accounting for the attenuators, antennas, and free space path loss, the peak output power of the MG-4 can be accurately determined.

Authors: L. Silvestre; R. Joshi; J. Stephens; J. Dickens; J. Mankowksi; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9813001

Abstract: The impact of secondary electron yield (SEY) variations on multipactor in a parallel plate geometry is probed. In this contribution, electron swarm dynamics are simulated via the continuum approach across a variety of SEY curves. The objective is to determine which changes and shifts in the SEY curve are most sensitive to the final multipactor outcome. For instance, will a variation in the maximum SEY yield while retaining the first crossover energy produce significant changes in the MP susceptibility? To parameterize, the probed SEY curves are approximated by suitable square, triangular, or trapezoidal shapes and fed into the Vlasov-Poisson-based MP model to evaluate the impact of chosen energy-dependent deviations on multipactor. A shaped approximation of the SEY curve has already shown to produce signifigant changes in multipacotr susceptibility especially in the 1 st order regime. Variations in the 1 st and 2 nd crossover points of the SEY curve are also compared within the same susceptibility graphs. The results of this Vlasov-Poisson method are benchmarked against commercial software, particularly Spark3D. The results obtained and implications of SEY deviations on multipactor will be presented and discussed.

Authors: A. Demir; L. F. Gutiérrez; S. Bayne; A. Bilbao

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9842564

Abstract: This paper presents a case of study of land-level temperature prediction using LSTMs in the context of renewable energy generation in microgrids. We explore the temperature forecast through LSTMs trained with pre-processed datasets consisting of historic temperatures along with relevant historic features, such as humidity, dew point, pressure, and wind speed. The results shows that the LSTM models are able to fit our available data and generate reasonable future values of temperature according to the obtained loss values. Moreover, we explore the forecast of temperature considering its historic values in an unsupervised manner using rolling LSTMs. The findings suggest that the approach presented in this paper can be used as a foundation of a machine learning-based microgrid control.

Authors: S. Daneshvardehnavi; C. A. Negri; K. E. K. Schmitt; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9458523

Abstract: This paper presents a control strategy for reactive power support using Plug-in Electric Vehicles (PEV) as a case study for the Global Laboratory for Energy Asset Management and Manufacturing (GLEAMM) microgrid. The control topology is divided into two levels with a dynamic limitation. This Electric Vehicle (EV) station Charger has a topology including a full-bridge AC-DC rectifier and a bidirectional half-bridge DC-DC converter. In the first level, a PQ and power control are used for the AC/DC inverter and DC/DC converters, receiving the set-points for active and reactive power from the upper level. The second level is a central controller that is used to manage the reactive power at the microgrid and to provide the necessary power setpoints for the PEV. A dynamic limitation strategy is proposed to maximize the reactive power support maintaining a reasonable amount of active power to charge the EV battery, considering the maximum limitation of the AC/DC inverter. To validate the effectiveness of the proposed control strategy, simulation results from a three-phase system are simulated in Matlab/Simulink environment.

Authors: K. E. K. Schmitt; I. Osman; R. Bhatta; M. Murshed; M. Chamana; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9595716

Abstract: This paper proposes a computationally efficient building energy management algorithm for demand response that can serve as a grid-ancillary system. The controller aims to regulate flexible loads and intelligent switches, complying with the utility"™s request. The control algorithm dynamically optimizes the load"™s configuration of the building. This optimization is based on the required power consumption level and the resident"™s actual comfort constraints. Since the load-matrix considered by the proposed algorithm is computationally expensive, a novel region-selection approach is incorporated in the algorithm to make the strategy computationally efficient. The proposed algorithm is validated through OPAL-RT Real-Time Digital Simulation with Raspberry Pi. The test results show that the algorithm is capable of curtailing controllable loads during emergencies and outage scenarios to maintain an uninterrupted supply to the critical loads and respect the power limit request of the building.

Authors: M. Flynn; A. Neuber; J. Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733141

Abstract: Low temperature plasma (LTP) models, such as fluid and global models, frequently rely on Boltzmann equation (BE) solvers for the fast calculation of rate and transport coefficients from cross section data. While multi-term BE solvers currently see strong support in the kinetic modeling community, two-term BE solvers remain the preferred choice in the greater pulsed power and plasma science community. The limitations of the two-term approximation are well known, yet it is still regularly assumed that the two-term approximation is suitable for applications in LTP models. Recent studies have demonstrated otherwise where, even in a simple 1D fluid model, the use of multi-term derived rate and transport coefficients yielded superior fluid model results compared to the same model using two-term derived rate and transport coefficients.Much of the reluctance in adopting multi-term BE solvers may be attributed to a lack of publicly available tools. Here, we report on the public release of MultiBolt v3.0.0 as a free and open-source C++ header library as well as its pre-compiled terminal-based executable and graphical user interface. We also report on the inclusion of new functionality in temperature dependence, superelastic collisions, and the object-oriented treatment of both discrete and analytic cross sections.

Authors: I. Zaman; M. He

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9458408

Abstract: The recent spike in microgeneration of renewable energy is demanding a smart, reliable, secured and efficient technology to enable Peer to Peer (P2P) energy trading. Due to the inherent characteristics, blockchain has been a preferred technology for realizing P2P energy trading. However, blockchain implementations for P2P energy trading so far are suffering from critical challenges such as security, privacy and scalability. In this paper, we introduce a P2P energy trading platform that leverages the popular blockchain technology and addresses these concerns. In particular, a multilayered semi-permissioned blockchain based platform along with a Quality of Transaction (QoT) module is proposed as a trading platform that can be used for transaction of energy. A two stage blockchain architecture, backed by QoT, ensures proper verification and validation of the participants and transactions. We present two use cases that demonstrate two different attack scenarios to highlight the resiliency of the proposed framework. Finally, a qualitative analysis shows the effectiveness of the system with respect to security, privacy and scalability.

Authors: A. Balal; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9639359

Abstract: Shortage of power generation results in unplanned load shedding. Active participation of the people on the demand side provides a solution by moving loads from peak time to off/peak hours during the day. This paper presents a Demand Side Management (DSM) that includes a rooftop photovoltaic system, a Battery Storage System (BSS), a hybrid control system, and a grid-tie inverter to reduce peak demands. The hybrid controller ensures that the maximum solar power generated is used or stored in the batteries for a given load, and any excess energy feeds the power grid. In this system, essential loads and generated power from rooftop PV systems are constantly monitored, and power generated by solar panels is preferred above grid electricity. According to the results, this system provides 100 % of each residence"™s required energy, plus an annual 6497 kWh to feed the AC grid and store in the BSS, resulting in monthly cost savings and the elimination of peak loads, particularly during the summer months, as well as a variety of environmental benefits.

Authors: Zachary Cardenas; Ben Esser; Ivan Aponte; James Dickens; John Mankowski; Jacob Stephens; Donald Friesen; Crystal Nelson; Neil Koone; David Hattz; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9733134

Abstract: Characterization of surface charge decay for varying materials – PTFE, Acrylic, and metals – of 100 mm diameter, radially symmetric bodies was performed. In a gaseous environment, charged materials will lose their charge depending on gas type, humidity, and other environmental factors. Eventually, the charge will be (1) neutralized by ions or electrons in the surrounding gas (2) distributed by conduction on the material surface or in its volume (3) or neutralized by ions or electrons in the material surface and volume. Metal and polymer samples are charged to a known potential in the tens of kilovolts range, and the decaying voltage is monitored with a non-contact probe until the potential reaches near zero. Humidity has an impact on both groups of materials. As a baseline, in air at 42% relative humidity in an enclosed chamber, a positive charge equivalent to 20 kV potential on a metal sphere of 50 mm radius decays to near zero within 25 minutes. In contrast, the same sphere negatively charged decays slightly faster within 20 minutes. Overall, the humidity has a significant effect on the charge decay time, such that an order of magnitude longer times (hours) are observed at low humidity. Effects of chamber size were also studied, drawing a connection to charge carriers in the air available for decay, with chambers of volume(s): 3.3 105 cm3 and 1.4 106 cm3, respectively.

Authors: Rawat, A; Tiwari, A; Gour, S; Joshi, R

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9688519

Abstract: The investigation of the design and analysis of a Metamaterials loaded Substrate Integrated Waveguide (SIW) antenna for Multiband Applications. A ground arrangement through a radiating rectangular patch conflicting the feed line as well as a mixture of Substrate Integrated Waveguide and Metamaterials is presented for the proposed antenna. These are utilized to improve the antenna's bandwidth and Radiation pattern while also reducing its size. To improve directivity, gain, and bandwidth, a complementary square split ring resonator and Substrate Integrated Waveguide are utilized. The proposed antenna structure is made of FR-4 epoxy with a epsilon(r)=4.4. The antenna functions over the frequency range of 6-18 GHz, with a resonant frequency for Wireless LAN and WIMAX.

Authors: X. Chen; J. Zhao; M. He

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9458556

Abstract: Numerical weather predictions used for wind power forecasting might not be updated in a timely manner in practice, due to its high computational complexity and complicated postprocessing. Thus, the accuracy of wind power forecasts could be significantly compromised especially during wind ramp events. This paper presents an innovative method for improving regional wind power ramp forecasting through ensemble learning of numerical weather prediction models, by using real-time weather measurements as the supervisory data. The numerical weather prediction models are combined to minimize the discrepancy between the forecast values and the real-time measurements in the trend of wind ramps, and the weights of the linear combination are calculated through gradient boosting. The proposed method is non-intrusive and could be efficiently carried out online. The proposed method is evaluated on historical ERCOT wind power ramp events, and compared with existing ensemble aggregation method using simple averaging. The results reveal the effectiveness of the proposed method for improving wind power forecasting during wind ramp events.

Authors: M. Chamana; K. Schmitt; R. Bhatta; I. Osman; S. Liyanage; M. Murshed; S. Bayne; J. MacFie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9611816

Abstract: Community-scale microgrids play an essential role in serving critical loads during emergency conditions, involving the operation of breakers, tie-switches, distributed energy resources (DERs), and loads. Electric loads are primarily considered as lumped loads without many granular levels of controls. Flexible buildings offer the central microgrid management system an opportunity to shed multiple noncritical loads at granular levels by adopting Internet-of- Things (IoT) based controls. This work presents a novel bi-level optimal sequence of operations for managing the controllable devices in microgrids to serve loads, based on a priority scheme in community scale-scale microgrids. The proposed methodology is formulated as a mixed-integer linear programming (MILP) model and adapts to various operating conditions. The proposed method is validated through case studies that are performed on the Banshee microgrid benchmark model.

Authors: A. Gregory; Z. C. Shaw; D. Wright; J. Mankowski; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733122

Abstract: The multipactor effect is detrimental in space-based RF systems through detuning, heating, and causing permanent component damage. Studying thresholds and suppression of multipactor utilizing surface geometries in structures akin to WR-90 waveguide are of specific interest in the presented work. Operating in the dominant TE10 mode, a copper stepped impedance transformer transitions the waveguide to a 1.5 mm height, providing a frequency-gap product conducive for multipactor ignition. The stepped impedance transformer housing is designed as a plug-and-play system, allowing for quick swapping different stepped impedance transformer heights or materials. Input power is injected into the test gap from a coaxial magnetron, operating at 9.4 GHz with a peak power of 250 kW and 2.5 µs pulse width.The setup will enable the measurement of base thresholds for a machined and polished copper surface (material that has undergone a simple cleaning process only), as well as thresholds for multipactor suppression geometries. The testing apparatus details, including the phase-sensitive diagnostics, are presented

Authors: H. O"™Brien; A. Ogunniyi; S. -H. Ryu; T. Tsoi; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733054

Abstract: The Army Research Laboratory (ARL) has funded the development of high-voltage silicon carbide (SiC) thyristors and diodes for pulsed power switching, culminating in the novel 1.0 cm², 15-kV SiC thyristor with n-type doping in the drift layer. N-type thyristors have been predicted to achieve faster switching speeds and lower switching losses, but were only recently realized following the development of novel fabrication techniques. These devices are targeted to reduce volume and increase reliability of pulsed switches in high-energy systems. ARL and Texas Tech University characterized the first fabrication lot of these devices for high-voltage DC-blocking capability (<1 µA leakage at 15 kV), optimal turn-on controls (4 A gate pulse), and on-state resistance at high current densities (up to 3 kA/cm²). This paper presents recent analysis of the turn-on speed and dI/dt capability for low-kHz pulse repetition of the n-type SiC thyristors as compared to previously reported 15-kV p-doped SiC thyristors. The current through the n-thyristor peaks 50 ns earlier, reaches 10% higher amplitude, and has significantly faster dI/dt as compared to the similarly designed p-thyristor.

Authors: V. Roy; S. S. Noureen; S. Atique; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9467414

Abstract: Some of the recent reports show that Power Grid is a target of attack and gradually the need for understanding the security of Grid network is getting a prime focus. The Department of Homeland Security has imposed focus on Cyber Threats on Power Grid in their "Cyber Security Strategy,2018" [1] . DHS has focused on innovations to manage risk attacks on Power System based national resources. Power Grid is a cyber physical system which consists of power flow and data transmission. The important part of a microgrid is the two-way power flow which makes the system complex on monitoring and control. In this paper, we have tried to study different types of attacks which change the data propagation of Synchrophasor, network communication interruption behavior and find the data propagation scenario due to attack. The focus of the paper is to develop a platform for Synchrophasor based data network attack study which is a part of Microgrid design. Different types of intrusion models were studied to observe change in Synchrophasor data pattern which will help for further prediction to improve Microgrid resiliency for different types of cyber-attack.

Authors: J. A. Rodriguez; F. Salcedo; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733138

Abstract: Non-linear loads have become more prevalent due to the increasing use of power electronics for power conditioning and regulation. The current harmonics generated from non-linear loads can cause long-term effects on generator sets, such as the degradation of their lifetime due to higher operating temperatures and component failures due to high current transients. A testbed was developed to run a 3-kW military tactical generator set under non-linear loads. The 3-kW generator set includes a permanent magnet alternator (PMA) and a power electronic converter to deliver a 120VAC/240VAC output at 60 Hz. A range of non-linear loads was developed to create several load profiles for the generator set. The generator set was tested in four-hour and two-hour intervals for several weeks under different load profiles to stress its alternator and internal power electronics. In addition, several measurements were taken during testing, such as the output power, output voltage and current, and stator winding temperature. The collected data and testing suggest that the power electronics converter is more susceptible to the current harmonics than the PMA. These issues can result in premature failures and overall degradation of the lifetime of the generator set.

Authors: A. V. Bilbao; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733153

Abstract: The purpose of this paper is to show the results obtained by performing maximum power transfer in inductive resonant wireless power transfer (WPT) systems. The research highlights the importance of power electronic systems on the receiving side of a WPT system. Before getting into the specifics of the research, a fundamental theoretical analysis about WPT systems is performed followed by an LTSpice simulation setup and results. The results presented show that power electronic DC/DC converters can be successfully used to increase the amount of wirelessly delivered power through impedance matching.

Authors: Michael Giesselmann; Jon Mayes

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9733157

Abstract: We are reporting on the design of a compact capacitor charger with a rated output voltage of 50kV and a power rating of 10kJ/s. The input source is a 208V, 3-phase AC utility supply. The charger shall fit in to an 19inch, 5U enclosure. The core of the charger is a HV-power pack containing a high-frequency transformer with a nano-crystalline core, multiple output windings using litz-wire and integrated HV rectifiers. The primary of the transformer is driven by an H-bridge inverter with advanced IGBTs and matching High Power Drivers. The control is through a state-of-the-art Digital Signal Controller (DSC) with fast cycle-by-cycle Peak current mode control, leading edge blanking, and high resolution, fast, automated slope compensation for current loop stability. The DSC runs at 5 times the speed of the main processor core for ns level PWM resolution.

Authors: T. Tsoi; B. Westmoreland; S. Bayne; S. Jadva

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733135

Abstract: Silicon Carbide (SiC) Merged PiN-Schottky (MPS) diodes have the benefit of conduction modulation under high current events while achieving low forward voltage and zero reverse-recovery under normal operations. Thus, the SiC MPS diodes can sustain surge currents several times larger than their average current rating, avoiding oversizing components and resulting in a more compact power electronic device. Two SiC MPS diodes were evaluated using a non-repetitive surge current testbed that delivers a square current pulse of 800 A. Five devices from each group were subjected to a ten µs current pulse every 20 seconds. The first device from each group started at lower current levels and was increased until device degradation occurred. Subsequent devices were then tested at the highest current level until degradation. Both groups have sustained currents up to 2.5 times their rated surge current rating. These devices were subjected to several thousand pulses, and their electrical characteristics, such as forward IV and reverse blocking voltage, were measured between testing intervals. Device degradation was observed as the reverse-blocking voltage has significantly decreased from preliminary measurements, but no degradation of the forward-IV curve was observed. The collected data demonstrate the device’s ability to operate under non-repetitive surge current events. Each device has sustained several hundred pulses above their rated surge current rating before any sign of degradation was detected. Device degradation becomes apparent when the leakage current increases as the MPS diode is blocking voltage. They eventually become prone to short-circuit failure due to a reduced reverse blocking voltage capability.

Authors: D. Z. Graves; A. V. Bilbao; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733117

Abstract: The purpose of this paper is to show the results obtained from novel research performed in the area of inductive-resonant wireless power transfer (WPT) safety. Traditionally a communication link between the transmitter and receiver power electronic system is required to ensure that power is delivered only to a suitable target system. This research aims to eliminate the communication hardware required to increase the system’s volumetric power density and mass. Machine learning is used to perform current waveform analysis to detect the receiver’s signature and enhance system safety.

Authors: J. Williams; E. Weeks; R. Clark; S. Watkins; J. Dickens; J. Mankowski; J. Brinkman; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733143

Abstract: Explosive driven pulsed power performance benefits from modern polymer-bonded explosives. In practical application, high precision in the dimensional shape is required, and high machining speeds are desired. The limits of machining speed, driven mainly by the thermal response of the energetic material, are investigated. Specifically, the thermal response of PBX 9501 (95% HMX, 2.5% Estane, 2.5% BDNPA/BDNPF by weight) and PBX 9502 (95% TATB, 5% Kel F-800 by weight) under conventional milling methods is examined. The presented work focuses on face milling performed with dry machining on a CNC, remote-controlled milling machine. The primary parameters of interest are the spindle speed, feed rate, step size, and depth of cut, with additional consideration given to endmill diameter, milling technique, and the number of flutes. The temperature of the system is monitored via high-speed IR videography and near the cutting interface with a K-type thermocouple inserted into the endmill’s through-coolant hole. Operational forces and torques are recorded by a 6-axis force sensor mounted beneath the HE samples.An empirical relationship between temperature and the milling parameters is presented. Overall, milling regimes exist outside of DOE-STD-1212-2019 for which milling temperatures remain well below the HE critical temperatures.

Authors: T. Tsoi; C. Whitworth; M. Kim; S. Bayne; H. O&#x2019;Brien; A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733125

Abstract: Silicon Carbide (SiC) is a wide-bandgap semiconductor with a wider bandgap, higher critical electric field, higher saturation velocity, and higher thermal conductivity than silicon, making it desirable for pulsed power applications. The n-type Gate Turn-off thyristor (nGTO) is a controllable solid-state switch with high blocking voltage and high current conduction capabilities. However, its device structure is challenging to develop using SiC. Wolfspeed has developed a 15.0 kV SiC nGTO that withstood peak current up to 1.0 kA. A testbed was developed to evaluate the long-term reliability of these SiC nGTOs. The pulser operates with a capacitor bank charged to 10.0 kV to deliver a 35.0 µs ring-down current waveform through the device. The SiC nGTOs were pulsed with a repetition rate of 0.5 Hz. It was observed that the device dissipated 700.0 J peak energy during pulsing. COMSOL Multiphysics simulated the SiC nGTO during a single pulsing event. The simulation models the thermal flow and current density in the nGTO. The thermal result shows the peak device temperature rising to 360.1 K after 17 µs into the pulsing event. The peak magnitude of current density reached 15 kA/cm2

Authors: M. LaPointe; B. Esser; I. Aponte; Z. Cardenas; J. Dickens; J. Mankowski; J. Stephens; D. Friesen; C. Nelson; N. Koone; D. Hattz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9733129

Abstract: Undesired accumulation of charge on dielectric materials causing electrostatic discharges can be an issue in pulsed power systems and electrical systems in general. As such, an understanding of surface charge distribution, charge accumulation, and decay is required. An in-house no-contact electrostatic probe designed with a 2 mm resolution was fabricated to measure and map surface charge distribution. Designed as a contactless instrument, the probe consisting of a metal rod and low leakage amplifier circuit probes the electric potential near the surface. A scan of the surface then provides the raw data, a potential distribution. The actual probe response - i.e., the surface charge to voltage transfer function, is accounted for in post-processing. This is accomplished via an Inverse Wien Filter - a technique often applied in image processing - to deconvolve the probe response from the measured data. A commercially available electrostatic probe, the Trek 341B meter with a 3455ET probe, capable of measuring +/- 20 kV was compared to the in-house probe that is designed to cover a wider range from +/- 35 kV. A resolution better than the simple probe resolution is achieved through the distribution’s scanning voltage method and deconvolution. Applying repeated scans, surface charge decay was tracked on various polymer materials to determine the material and environment dependence; materials included were PA6, PTFE, and others. As an example of material dependence, samples charged to 20 kV at 65% humidity experience full charge decay in approximately 45 seconds for PA5 (152 mm dia.), while it took about 100 times longer for PTFE (51 mm dia.) to observe the same decay/redistribution of charge.

Authors: Raimi Clark; Jacob Young; William Brooks; Matthew Hopkins; John Mankowski; Jacob Stephens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9733139

Abstract: Early light emission provides information about the dominant mechanisms culminating in vacuum surface flashover (anode-initiated vs. cathode-initiated) for particular geometries. From experimental evidence gathered elsewhere, for the case of an insulator oriented at 45° with respect to the anode, anode-initiated flashover is believed to dominate since the field at the anode triple point is roughly three times that of the cathode. Similar to previous work performed on cathode-initiated flashover, light emission from the voltage rise through the impedance collapse is collected into two optical fibers focused on light emanating from the insulator in regions near the anode and cathode. The optical fibers are either connected to PMTs for spectrally integrated localized light intensity information or to a spectrograph used in conjunction with an ICCD camera. Challenges associated with localizing the flashover for optical diagnostics and incorporating the optical diagnostics into the high-field environment are discussed. Initial results for cross-linked polystyrene (Rexolite 1422) support the premise that flashover is initiated from the anode for these geometries, as early light from the anode leads cathode light up to photocathode saturation. Early spectroscopy results show promise for future characterization of the spatio-temporal development of emission from desorbed gas species across the insulator surface and identification of bulk insulator involvement if it occurs.

Authors: S. Ramabhotla; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9176792

Abstract: Microgrid is considered as the future power systems due to the demand in the power supply and also due to its capability of integrating with the renewable energy sources. Based on the demand and location, the microgrid operates independently in an islanded mode by disconnecting from the power grid. A microgrid improves the efficiency of power consumption, along with the reliability of the power quality and power supply. The reliability of the microgrid is an important aspect to improve for the efficient power quality and availability of the electricity. This paper provides a review on reliability of microgrid considering various techniques and methods. The paper discusses about the background and concept of microgrid, effect of reliability in power systems, different scenarios of reliability in microgrids. Also, the challenges and solutions are elaborated from the implementation of various optimization techniques for the microgrid systems.

Authors: I. A. Aponte; B. Esser; J. C. Dickens; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717642

Abstract: DC and RF breakdown at 3.3 MHz was studied in centimeter size gaps − 1 − 10 mm – with comparison to small gaps in literature and Monte Carlo simulations in atmospheric conditions. As a point of reference, DC breakdown using stainless-steel Bruce-profile electrodes were measured to compare with RF measurements. RF breakdown with a slow rising envelope ($\sim 5\ \text{mV}/\mu\mathrm{s}$) yielded approximately 80% of DC measurements (∼25 kV/cm in 5 mm gap), which agrees with Monte Carlo simulations and results found in literature. Increasing the envelope rise time to greater than $1000\ \mathrm{V}/\mu \mathrm{s}$ yielded ∼120% of DC measurements (∼37 kV/cm in 5 mm gap). Comparisons with Monte Carlo simulations which included photon processes – theorized to be critical to obtaining accurate results – furthered understanding of the processes involved in pre-ionization before breakdown occurs.

Authors: T. Sugai; Z. Shaw; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717525

Abstract: Multipactor is a resonant nonlinear electron multiplication effect that may occur in high power microwave devices at very low pressures, such as those operating in particle accelerators and satellite subsystems. Its effects range from signal degradation to the damage and destruction of microwave components. Thus, multipactor physics has been studied through theoretical analysis, numerical simulation, and experiment. Previously, we developed a direct electron observation system using an Electron Multiplier Tube (EMT) and succeeded to directly detect multipactoring electrons in the center of the broadwall of rectangular waveguides ^{1, 2}. Here, we provide a method for evaluating the electric charge density and secondary emission yield (SEY) in waveguides. The experimentally obtained EMT signal is analyzed with the extensive usage of the numerical simulation software Spark3D. The software was utilized to analyze multipactor onset in waveguide structures, where the electric field distribution without multipactor was carefully simulated, employing high-frequency solvers. The EMT signal and the charge density were simulated for the same conditions as the experiment. As a result, a calibration line indicating the proportional relation between the EMT voltage and the charge density, which is independent of some conditions, i.e., input power and gap size, was obtained. Further, after adjusting the SEY curve imported to Spark3D, the rising shape of the experimental EMT signal pulses fit with the simulated one, and the experimental threshold power for the EMT signal generation was consistent with the simulated multipactor threshold power. Since the simulation matches the experiment in threshold power and signal shape, one expects that the charge density and SEY curve deduced from the simulation are accurate.

Authors: S. Basu; V. Roy; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9342862

Abstract: The extensive integration of renewable sources of energy has caused the number of conventional power plants to go down. This in turn causes a reduction of the system resilience, i.e. the ability of a system to recover from a fault. Synchronous condensers, when included in the grid, contribute to the fault current and provide voltage support thereby enhancing dynamic voltage recovery. This article summarizes the capacity of a synchronous condenser, when connected to the power grid with wind or solar power plants to compensate for reactive power compensation and injection of active power at their point of common coupling. A case, based on a local distribution network, is developed on a 33kV transmission line that feeds into several inductive loads. Simulations are performed in PSIM to display the enhancement of system stability and reduction in power losses while supplying reactive power by activating synchronous condensers to the grid with given practical grid rules.

Authors: M. Flynn; A. Neuber; J. Stephens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9717463

Abstract: Insulating gases, such as SF6, and N2, are modeled using a multi-term Boltzmann equation model. Owing to their large vibrational cross-sections, modeling such gases using the conventional two-term Boltzmann approach leads to near singularities in the electron phase space at the velocity-space origin. Thus, a large error in predicted in swarm parameters for two-term Boltzmann models, even at low-to-moderate reduced electric fields.

Authors: L. Silvestre; R. Joshi; J. Stephens; J. Dickens; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717616

Abstract: Multipactor discharge is a resonant phenomenon that can be initiated in vacuum under RF excitation, giving rise to charge growth over time. The electron dynamics under such collisionless conditions has been researched by kinetic Monte Carlo and magnetohydrodynamic models in the past. As an alternative, we develop and present studies of a Vlasov equation based numerical model to calculate multipactor susceptibility in common microwave structures [1]. In contrast to kinetic models, utilization of the Vlasov equation permits the continuous treatment of the electron distribution in phase space, thereby capturing all statistical outcomes in a single calculation. To address the computational demand of the Vlasov equation, parallel computing techniques are utilized.

Authors: R. P. Joshi; J. P. Tarapure; T. Shibata

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9142645

Abstract: In rapidly aging societies, robotic solutions for clothing assistance can significantly improve the quality of life of the elderly while coping with the shortage of caregivers. Previously, we proposed a framework for the same by employing imitation learning from a human demonstration to a compliant dual-arm robot. As the robot has a limited workspace, this framework involves a manual movement of the wheeled chair by pushing it while coordinating with the robot to stay within the workspace of the robot [1]. To avoid the manual push and coordination, we facilitate the automatic movement of the chair based on the trajectory of the robot's dual arms. In this paper, we present an approach for the collaboration of an electric wheelchair and a humanoid robot to achieve the clothing assistance task. Our approach incorporates Manifold Relevance Determination (MRD) to learn an offline latent model from the simultaneous observations of the clothing assistance task as well as the movement of the wheelchair. We trained and tested the latent model on different human subjects by dressing a sleeveless T-shirt. Experimental results verify the plausibility of our approach. To the best of our knowledge, this is the first work addressing collaboration between wheelchair and robot to perform clothing assistance.

Authors: Jacob Stephens; Guy Rosenzweig; John Tucek; Ken Kreischer; Michael Shapiro; Richard Temkin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9520451

Abstract: This paper details recent progress on the experimental demonstration of a W-band klystron amplifier completed at the MIT. The amplifier utilizes a square lattice photonic bandgap (PBG) structure that permits the use of a highly oversized beam tunnel of diameter ~λ/4. Cold test measurements of the PBG klystron cavities revealed successful fabrication of the device. In hot test, a small-signal gain of 26 dB was measured at 93.7 GHz, with a saturated output power of 30 W.

Authors: T. Buntin; M. Abide; J. Dickens; A. Neuber; R. Joshi; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717481

Abstract: Most high power microwave sources, such as the Magnetically Insulated Transmission Line Oscillator (MILO) being developed at Texas Tech, utilize cold cathodes that generate electrons via explosive emission. These highly emissive cathodes can generate current densities and currents greater than kA/cm² and 10 kA, respectively which are required for devices that can output RF power greater than 100 MW. Typical explosive emission cathode material includes metal, velvet, carbon fiber, and CsI coated carbon fiber. In order to optimize the MILO performance, we have begun fabricating carbon fiber velvet and comparing the performance with other commercially available materials. Fabrication was done on a manual, mechanical loom using commercially available carbon fiber thread.

Authors: B. Esser; Z. Shaw; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717900

Abstract: A pulsed 2.85 GHz RF source design with high average power is presented for use as the principle exciter in an experiment to directly detect multipactor. With a rated output power of 700 W each in long pulse mode (∼ 100 □s), four Cree GaN HEMTs are used to achieve a maximum of 2.8 kW rated output. A fifth HEMT is used to drive the four output devices with approximately 40 W each. A free running VCO with low tuning sensitivity is used which ensures a stable frequency output despite a noisy environment. Buffering and initial amplification is provided by an LNA with a TTL RF switch used to generate the pulses. With a rise time of 35 ns, this switch is fast enough to create reasonably square pulse edges. A microwave amplifier with 45 dB of gain provides the majority of the gain in the system and the power necessary for the final GaN stages, approximately 9 W. An in-line micrometer adjusted, continuously variable attenuator between LNA and TTL switch provides level control.

Authors: M. Giesselmann; S. Daneshvardehnavi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9158658

Abstract: We are presenting an LT-Spice Model for 3-phase Induction Generators that are operating in off-grid mode using a 3-phase capacitor bank for self-excitation. To model the self-excitation, we include the effects of a non-linear magnetizing inductance with residual magnetic flux and saturation. Our model represents both the electrical as well as the mechanical characteristics of the machine.

Authors: L. Arce; M. Chamana; I. Osman; B. Ren; Q. -C. Zhong; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9330668

Abstract: This work proposes the large-scale adoption of self-synchronized universal droop controller (SUDC)-based inverters to enable ancillary services for different modes of distribution system operations. The IEEE 123 bus system was modeled on a real-time simulator to study the performance of large-scale adoption of SUDC inverters in a distribution system. The resulting data collected shows that the voltage and the frequency were regulated within ranges, such as less than 5% for voltage and less than 0.5% for frequency, under different load variations and grid operations. Also, the black start was achieved within 0.4 s without any voltage overshoot. Through the simulation and validation on a small microgrid and the IEEE 123 bus distribution system, it can be concluded that the SUDC was successfully adopted to regulate the voltage and the frequency within the given ranges, and black start achieved within 1 s without voltage overshoot for different modes of distribution system operations.

Authors: Zachary C. Shaw; Benedikt Esser; James C. Dickens; John J. Mankowski; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9520510

Abstract: A planar test cell was designed and implemented to observe the multipactor effect in waveguide structures. This plug and play device allows for multiple geometries to be machined and easily replaced within the test structure. A direct detection method was used to observe the multipactor effect while the upper and lower thresholds were measured for a 2.1 mm gap at 2.85 GHz. While there is an obvious lower limit to multipactor (2 kW), there was no observable upper limit even at powers over 200 kW. This is attributed to the transverse electric field distribution in the dominant TE10 mode which is not taken into account in most multipactor theoretical models.

Authors: X. Chen; J. Zhao; M. He

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9289816

Abstract: Wind power ramps are the abrupt yet significant change in wind power productions. The information on the ordinal levels of impending wind power ramp could help power system operator to arm operation or ramping reserves in a timely manner. This paper presents novel approaches for regional wind power ramp level forecasting using real-time meso-scale wind speed measurements. Motivated by the correlation of the meso-scale wind speed measurements with the regional wind power data, the proposed approach utilizes multinomial logistic regression for wind power ramp forecasting. An approach that combines the probabilistic output of individual regressive models in a weighted manner is proposed, with the weights calculated by minimizing the Brier skill score of the combined model. The proposed methods are tested by using real-world data, and is compared with benchmark methods. The results reveal the effectiveness of the proposed approaches.

Authors: Dinkhah, S; He, M

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9289787

Abstract: in this paper, we propose an improved control mechanism for a microgrid. This happens through adding a controlling measure and algorithm one step ahead of the combination of inertia emulation control technique and droop control. This controlling step is a self-adjusting control system designed for a stable electrical grid model. This control technique applied to a home that is furnished with Photo-voltaic (PV) system and Vehicle to Grid (V2G) capable Electric Vehicle (EV). This control could run in both grid-connected and islanded modes in a microgrid application. The model is used to achieve an improvement in frequency over the standard droop/inertia emulation control method, in case of load changes and faults. The provided control system can handle different scenarios such as sudden load changes and transient conditions through dealing with the power of the battery and PV to regulate the voltage and frequency in this microgrid system. This control technique will achieve a better result compared to the standard control technique. Both models are implemented in MATLAB/Simulink. The results for the simulations are presented, showing the improvements over the fixed values of the controller.

Authors: J. Williams; D. Barnett; J. Dickens; A. Neuber; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9717487

Abstract: The development of a solid state high-voltage pulse generator for a spark-gap triggered Marx generator is discussed. The control board features a microcontroller to provide added flexibility and safety. A PIC16 microcontroller and two fiber-optic cables are critical to the design. Optically isolated serial communication between the device and a remote PC is used to control every aspect of the output pulses. This allows for more specific configuration of the pulser and provides remote adjustment of the pulse generator. The design uses the microcontroller's digital I/O pins and SPI/UART peripherals to send and receive operating data over fiber-optic, control the output pulse repetition rate, interface with a digital potentiometer to change the signal amplitude, and activate components that send optically isolated signals to disable/enable the high voltage power supply. In the case of microcontroller failure, the design includes a backup circuit that reconfigures the fiber-optic receiver to trigger an output pulse instead of interpreting UART communication signals. This permits the operator to still trigger the Marx generator remotely in a “single-shot” mode. The successful operation of this device involves configuring the PIC16 to send a 4 kV pulse at repetition rates up to 1 kHz. The flexibility provided by the microcontroller allows customization of several aspects of the output pulse, leaving the design useful in almost any application requiring high voltage pulses.

Authors: S. Atique; S. Noureen; V. Roy; S. Bayne; J. Macfie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9289796

Abstract: In this paper, the potential of machine learning based methods for time series forecasting of total daily solar energy generation has been explored. Firstly, the time series is modeled using the seasonal version of well known classical method auto regressive integrated moving average (ARIMA) and its performance is later compared to two other popular machine learning methods, support vector machine (SVM) and artificial neural network (ANN). The potential of machine learning based methods in this line of work is demonstrated by the superior performance of SVM. However, the reasons behind the low yield of ANN need to be inspected to enhance our understanding. In spite of SVM's relative success in prediction of solar generation, the overall accuracy still needs to be improved and the methods to achieve this objective should be researched in future.

Authors: S. Noureen; S. Atique; V. Roy; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8885349

Abstract: This paper has presented the use of Auto Regressive Integrated Moving Average (ARIMA) method for forecasting of seasonal time series data. The dataset that has been used for modeling and forecasting is a small-scale agricultural load. ARIMA method can be applied only when the time series data is stationary. As seasonal variations make a time series non-stationary, this paper also presents analyses on testing stationarity and transforming non-stationarity into stationarity. Lastly, model has been developed with a specific selection of orders for autoregressive terms, moving average terms, differencing and seasonality and the forecasting performance has been tested and compared with the actual value. The results are encouraging, however there is scope of further research in refining the idea.

Authors: J. Forbes; F. Salcedo; C. Tchoupe-Nono; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9009635

Abstract: Silicon carbide (SiC) power semiconductor devices are experiencing an increasingly widespread adoption in many power electronics and pulsed power applications such as high-power DC-DC converters and inverters, battery chargers, industrial motor drives, as well as high-power solid-state pulse generators such as a Marx generator or a linear transformer driver (LTD). The wide-bandgap (WBG) and thermal properties of SiC provide inherent advantages over silicon (Si) power devices especially in high power density applications. These advantages include higher blocking voltages, increased switching speeds, physically smaller implementations of power electronics and pulsed power circuits, improved system efficiencies, and higher operating temperatures. To improve the overall confidence in the ability of SiC devices to reliably replace equivalent silicon solutions, independent reliability testing and analysis must be conducted. In this research, a short-circuit test board was developed to analyze the short-circuit ruggedness of 1200 V MOSFETs. Using the test board, commercially available 1200 V/10 A SiC MOSFETs from 3 different manufacturers were subjected to short-circuit events, and the short-circuit ruggedness of each device was measured and analyzed. The purpose of this research is to independently measure and report on the short-circuit capabilities of commercial off-the-shelf (COTS) 1200 V SiC MOSFETs.

Authors: S. S. Noureen; S. B. Bayne; E. Shaffer; D. Porschet; M. Berman

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8662186

Abstract: The emerging smart grid, cyber-physical infrastructure, provides a steady, secure, and reliable power system over the current power grid. Synchrophasor systems, like Phasor Measurement Units (PMUs), are a key element of smart grids. They have the capability to measure time-coherent phasors of a grid. The key advantage of PMUs is the fast sampling rate that they provide over traditional Supervisory control and data acquisition (SCADA) systems which can be in the range of 30-120 samples/second. These higher sampling rates come at the cost of higher data quantities. Generating large amounts of data per day poses a challenge in making the most efficient use of information. In this paper, this problem has been addressed utilizing machine learning techniques, Logistic Regression Analysis, on PMU data. Identifying system anomalies in smart power grids is the primary focus of this paper. The standard IEEE 39 Bus system has been modified using the RT-LAB environment to generate faults and to produce synthetic synchrophasor data. Archived/offline mode data from a Phasor data concentrator (PDC) database is being used to train and test the algorithm. Additionally, the algorithm has been tested in real-time using an OPAL-RT digital real-time simulator.

Authors: T. Buntin; M. Abide; D. Barnett; J. Dickens; A. Neuber; R. Joshi; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009681

Abstract: A low-impedance MILO is being developed at Texas Tech University, and a compact Marx generator was designed to drive it. The target design goals of the Marx are an output voltage greater than 500 kV and an output current greater than 40 kA. Risetime needs to be sub 150 ns and the pulsewidth must be greater than 100 ns. These performance goals were determined from PIC simulation of the MILO such that an RF efficiency (>10%) and RF peak power (> 1 GW) can be achieved. Tests using smaller 3 and 4 stage Marx generators with the same topology as the final design were used to determine a per-stage inductance of approximately 120 nH. From this derived inductance, multiple configurations were simulated to decide upon the ideal design for the desired performance goals. From these simulations, an 18-stage Marx with 2 capacitors per stage was chosen as the most optimal design, and from simulations into a 12 Ohm load a number of the criteria can be met with this configuration. The simulated peak voltage and current are 570 kV and 48 kA, respectively, while pulse risetime and pulsewidth are 170 ns and 540 ns, respectively. The designed Marx is being experimentally validated to confirm the findings of the simulation, firing into an approximately 12 Ohm water load to represent the low-impedance MILO that is being designed.

Authors: M. Chamana; S. Bayne; A. Swift

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8784605

Abstract: Phasor Measurement Unit (PMU) design and application requires knowledge of various mathematical and electrical engineering fundamentals. The fundamental theories are covered in lower level courses offered at the undergraduate level in a typical electrical engineering and energy related curriculum. Furthermore, laboratory equipment/device based courses help students to improve their practical skills. Since students find experimental, product development and group work related courses more engaging, they are motivated to enroll in such courses. This paper summarizes a PMU design and application course suitable for electrical engineering/renewable energy undergraduate and graduate students. The paper is aimed towards helping course developers who plan to develop similar courses in the future.

Authors: V. Roy; S. Noureen; T. Atique; S. Bayne; M. Giesselmann; A. S. Subburaj; M. A. Harral

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9020467

Abstract: Introduction of Synchronized phasor measurement unit (PMU) in power system is a solution for monitoring system wide disturbances. PMUs provide precise information about a system's phasors such as magnitude and phase angle of sine waves of the system. PMU data is a valuable source when determining the post-mortem of a fault, or a system disturbance. It can be unveiled what time the system began to falter and exactly what instabilities the system experienced. Knowing the cause of such instabilities will allow enhancement of the current system protection system. This will help in minimizing risk, disruptions or total system collapse. PMUs are synchronized via global positioning system (GPS). For wide area monitoring (WAM), synchronization of the PMUs allows for interconnected systems to be monitored simultaneously, giving real time records. This paper explores the impact of PMU in modern power system, deployment strategies of PMU network around Texas Tech University, importance of PMU data collected from the network, correlation of events by data analysis and focus on how the time stamped information is valuable for grid stability. The ongoing smart architecture of grid will increase dependency and importance on PMU based network in future.

Authors: Zhao, J; Chen, XM; He, M

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8662203

Abstract: Detection of impending front-induced ramp events is studied as a new class of change detection problem - change detection for multiple time series with spatial dependency. A critical step to ramp event detection is to capture the spatial dependency between neighbor turbines' power output. To this end, a graphical model is utilized to model the dependency of turbine-level ramp events. Then, change point detection is carried out for the time series of individual turbines' power output, by using the belief from neighbor turbines in the dependency graph. Once an impending ramp is detected, the magnitude of ramp is then forecasted by using current measurement data. A key observation is that due to the movement of front, the best predictors for individual turbines' power output vary across three different regions of the wind farm. With this insight, different predictive models are adopted for forecasting power output from each region. Through numerical experiments, the proposed detection-based wind power forecasting method is proven to outperform conventional methods for wind power ramps.

Authors: C. Negri; S. Daneshvardehnavi; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009899

Abstract: We are reporting on the formulation and performance of dynamic models for Pulsed Alternators for LTspice. The models are modular and hierarchical and cover both the electrical and mechanical aspects of the electric machine including the mechanical torque, speed and inertia. The models also include the effects of the damper cage and the excitation winding to accurately represent the sub-transient and transient behavior. The models can also be used to represent synchronous generators in steady state operation. LTspice is a powerful, widely available software package that can be used to model Pulsed Power circuits. We are presenting the detailed models as well as results of the simulations.

Authors: Y. R. Challapuram; G. M. Quintero; S. B. Bayne; A. S. Subburaj; M. A. Harral

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8767145

Abstract: Grid-scale energy storage systems have become popular for the growth of hybrid renewable power systems. Due to design flexibility, low manufacturing costs for large scale, indefinite lifetime and recyclable electrolytes Vanadium redox flow batteries (VRFBs) are a promising energy storage technology. In this paper, an electrical equivalent model of VRFB which is present in Distributed Asset Research Testing (DART) facility in Texas is proposed. The VRFB is connected to a 3.6kW resistive load. The effect of flow rate and pump power losses has been considered in modeling the VRFB. A control method for State of Charge (SOC) estimation is also proposed as it plays an important role in over-charge/ discharge of VRFB. The entire work is simulated in Matlab/Simulink environment and the results obtained prove the efficient operation of VRFB.

Authors: S. Atique; S. Noureen; V. Roy; V. Subburaj; S. Bayne; J. Macfie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8666481

Abstract: In this paper, a well known statistical modeling method named ARIMA has been used to forecast the total daily solar energy generated by a solar panel located in a research facility. The beauty of the ARIMA model lies in its simplicity and it can only be applied to stationary time series. So our time series data, which is seasonal and non-stationary, is transformed into a stationary one for applying the ARIMA model. The model is developed using sophisticated statistical techniques. The optimum model is chosen and validated using Akaike information criterion (AIC) and residual sum of squares (SSE). Error analysis is done to demonstrate the efficiency of the proposed method. The accuracy of the developed model can be further increased, which is subject to future research.

Authors: M. Powell; Z. Shaw; J. C. Dickens; J. J. Mankowski; A. A. Neuber; C. Scribner

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009615

Abstract: Pure N2 and SF6, as well as their mixtures, are evaluated for high electric field breakdown tested at pressures ranging from 750 torr to 1650 torr at 2.85 GHz. Previous research concerning the breakdown characteristics of pressurized SF6 and SF6 mixtures at S-Band frequencies is limited, likely due to the high electric fields required to breakdown pressurized SF6. A stepped impedance transformer is used in conjunction with a traveling wave resonator to obtain the high electric field amplitudes necessary to break down the gases. Starting with the output from a 3.5 MW coaxial magnetron the electric field amplitude in the test piece at the center of the stepped impedance transformer yielded a maximum of about 150 kV/cm RMS field. Using Pure SF6 as a baseline, the measured breakdown field shows a mostly linear dependence upon pressure in a range from 750 torr to 1350 torr, while some leveling out tendency is observed at pressures greater than 1350 torr. Since pure N2 exhibits a much lower breakdown threshold, ~ 60%, compared to pure SF6, mixing the two gases also results in a lower effective breakdown threshold; however, the reduction in the electric field breakdown threshold is not strictly proportional. For example, a 60/40 SF6 to N2 mixture resulted in a 90% breakdown field while a 20/80 mixture still yielded about 80% in the high-pressure regime.

Authors: R. Lee; A. Hewitt; R. Clark; H. Hudyncia; T. Buntin; D. Barnett; J. C. Dickens; J. J. Mankowski; W. A. Harrison; E. Tucker; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009758

Abstract: The impact of mechanical stresses on polymer bonded high explosives, HE, is investigated. High-Speed photography in the visible spectrum, VIS, as well as mid-wave infrared (MWIR) of HE during small diameter drilling and controlled skidding is presented. Controlled drilling into the HE enables recording the size and temperature of shavings under varying feed and speeds. Even at very high drill speeds, the HE phase transition temperature of approx. 180 degree Celsius is rarely exceeded. The MWIR signals radiated are recorded with FLIR's X6901sc High-speed MWIR camera, which uses InSb technology, with a wavelength range from 3.0 to 5.0 μm, and up to 1,004 fps at a resolution of 640 × 512 in the temperature range of interest. High-speed recording in the visible is obtained utilizing Phantom's VEO710s high-speed camera at a higher frame rate of 7,400 fps at a resolution of 1280 × 800 in the VIS. Observing the HE-grit interaction in the MWIR poses a great challenge, for IR is blocked by many glasses.

Authors: R. Clark; R. J. Lee; A. T. Hewitt; T. Buntin; D. Barnett; J. C. Dickens; W. A. Harrison; E. Tucker; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009974

Abstract: The drilling behavior of polymer bonded high explosives (HE) is investigated by varying drilling parameters and analyzing resultant forces and thermal response. A modified drill press enables remote operation and precise control of cutting speed, feed, and depth. To acquire temperature at the cutting interface a K-Type thermocouple is inserted in the coolant holes of thru-coolant drill bits and epoxied flush with the drill's flank face, which yielded a sensing accuracy of ±1°C and a resolution of 0.48°C. Cutting forces and torques are acquired with a sampling speed of up to 2,000 Hz. The comparison of downward directed forces across cutting operations is indicative of which speed and feed rate combinations limit excessive stressing of the HE, while cutting axis torques give indication in the case of drilling obstructions such as insufficient chip clearance. Drilling conditions in excess of the existing US DOE-STD-1212-2012 limitations are tested to determine safe but efficient machining limits for these materials. Drilling speed, feed rate, and peck depth are varied for drilling cycles with a 5 mm diameter drill bit, and further cycles are performed to determine the effect of increasing cut diameter. In peck drilling, clearance of chip from the drill flute is crucial and governs the drill's temperature rise.

Authors: M. LaPointe; L. Collier; T. Kajiwara; J. Dickens; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009986

Abstract: An innovative gating scheme for wide bandgap semiconductor switches is investigated to fully exploit recent advances of SiC MOSFET properties in hold-off voltage (from single digits to tens of kV) and low on-state resistance (tens of mΩ). Robust gate driving techniques are required to achieve fast risetimes on the order of 10-20 ns. Further, due to the high dI/dt, and subsequent inductive kickback, parasitic inductance may drastically affect the performance of commercially available totem-pole gate drivers. Further, traditionally packaged MOSFETs exhibit additional degradation of switching characteristics due to the introduction of parasitics primarily due to their lead geometry.

Authors: M. Abide; T. Buntin; D. Barnett; J. Dickens; R. Joshi; A. Neuber; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009939

Abstract: The development of a low-impedance magnetically insulated transmission line oscillator (MILO) driven by a compact Marx generator developed by Texas Tech University is discussed. The goals of the project aim to develop a MILO operating within the S-Band that can provide an RF peak output power of greater than 1 GW with greater than 10% efficiency. The device design followed a set of base design equations that were applied to a CST Studio Suite (CST) for a Particle-in-Cell, PIC, simulation to model the MILO. These simulation results then inform changes to the model to optimize the prospective performance of the device. The simulations were developed to account for realistic material properties that were then applied to critical surfaces of the device. Additionally, a circuit simulation was included to model a Marx generator feeding the input of the MILO to simulate the eventual experimental setup. Current results verify an expected RF peak power of approximately 4.5 GW at 2.5 GHz operating in the TM01 mode when excited with an input signal that has a peak voltage of 600 kV while providing a peak current of 58 kA. The simulation confirms the design should perform within these constraints.

Authors: T. Klein; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009641

Abstract: A pulse charger module was designed and tested for use in a larger system Each pulse charger module is powered with a 12 V Lithium-ion battery and set to charge a nF sized capacitor up to 100 kV in less than 10 μs. This is achieved by initially charging a μF sized capacitor to 3 kV, then switching a thyristor to discharge this capacitor into a step-up pulse transformer to charge the load capacitor. A PIC 18F26K80 8-bit microcontroller in each pulse charger module will be used to control the module, communicate with other modules and a computer, and monitor voltages. The modules are programmed to automatically detect the total number of modules well as communication delays between each module at startup, allowing for synchronous triggering and induvial identification and control. Each module is kept in a low power mode when not in use, and fiber optic communication is used throughout such that electrical isolation between modules and the master computer is ensured.

Authors: X. Chen; J. Zhao; M. He

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8662198

Abstract: This paper present a new conceptual framework of PV power generation credit sharing by leveraging social tie in residential community to maximize the financial benefit of community solar programs. Social tie-driven credit sharing schemes for residential community is designed to manage the dynamic allocation of solar PV power production ratio and credit among community members, so as to avoid unnecessary devaluation of solar PV power production by electric utilities. Along this avenue, a community solar management system that incorporates and integrates social tie network and credit sharing schemes is then developed.

Authors: J. Culpepper; A. Miller; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009753

Abstract: It is desired to integrate a photoconductive semiconductor switch (PCSS) capable of holding off and switching 100 kV into a package with small parasitic inductance such that sub-nanosecond rise time is still achievable at current amplitudes of hundreds of amperes. A GaAs based PCSS is utilized, which makes it necessary to address the filamentary nature of the current, which may lead to a shortening of device lifetime. In order to design a practical package, COMSOL based 2D electric field simulations have been utilized to aid in shaping the field between the PCSS semiconductor, the electrodes, and the high voltage encapsulant. To deal with the unavoidable high field stresses in the small package, the switch is brought to voltage within a few microseconds only, and then closed. Thus, keeping the duration of voltage stress very short, and the risk of self-triggering due to leakage current low.

Authors: J. A. Rodriguez; M. Kim; S. B. Bayne; H. O'Brien; A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009755

Abstract: Gallium nitride (GaN) high-electron-mobility transistors (HEMT) are of great interest for pulsed power applications due to their proven capabilities in RF applications. With further advances in GaN power semiconductors, there's an interest in the evaluation of their performance under repetitive overcurrent operation in power electronics applications beyond the manufacturer's prescribed operating parameters. A GaN HEMT from two different vendors were evaluated in a pulsed ring down testbed at 475 V with a peak current above 80 A over a repetition rate of 138 Hz. The testbed employed a temperature chamber to adjust the case temperature of the device during testing. The devices' electrical characteristics, such as transconductance, forward I-V curve and reverse blocking voltage were measured throughout testing and have not shown significant degradation. The collected data from these measurements allowed a comparison of the devices' performance and shows their ability to handle transient overcurrent conditions commonly found in power semiconductor device applications.

Authors: R. P. Joshi; T. Shibata; K. Ogata; Y. Matsumoto

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8956308

Abstract: The recent demographic trend across developed nations shows a dramatic increase in the aging population, fallen fertility rates and a shortage of caregivers. Robotic solutions to clothing assistance can significantly improve the Activity of Daily Living (ADL) for the elderly and disabled. We have developed a clothing assistance robot using dual arms and conducted many successful demonstrations with healthy people. It was, however, impossible to systematically evaluate its performance because human arms are not visible due to occlusion from a shirt and robot during dressing. To address this problem, we propose to use another robot, Whole-Body Robotic Simulator of the Elderly that can mimic the posture and movement of the elderly persons during the dressing task. The dressing task is accomplished by utilizing Dynamic Movement Primitives (DMP) wherein the control points of DMP are determined by applying forward kinematics on the robotic simulator. The experimental results show the plausibility of our approach.

Authors: W. Brooks; D. Barnett; J. J. Mankowsi; J. C. Dickens; A. A. Neuber; W. A. Harrison; D. Hattz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9009756

Abstract: A stochastic model of downward lightning propagation and attachment was developed based on existing models of the propagation of a negative leader as a gas diffusion problem. Attachment was modelled using the rolling sphere concept. The resulting simulation enables Monte Carlo methods of assessing the efficacy of geometric lightning protection schemes. Protection afforded to a 100 m by 50 m building was investigated to elucidate the relative impact of selected protection configuration parameters. The results were found to be insensitive to variations in building height for configurations of less than 30 m. Variation in footprint area found protection for large buildings generally exhibits worse performance relative to smaller structures. This trend was examined in detail for a regular, square arrangement of lightning rods. Variation in building aspect ratio was found to have a pronounced impact on protection performance and is attributed to non-preferential striking to the available geometries. A method for arriving at the expected rate of a given building being stuck is outlined which expects the aforementioned structure to be struck at a rate of about four times per 10 000 years.

Authors: S. Dinkhah; C. A. Negri; M. He; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8790615

Abstract: In this paper, we propose a stable electrical grid model, in which a home with a Photo-Voltaic (PV) system and Vehicle to Grid (V2G) capable Electric Vehicle (EV) can operate in both grid-connected and islanded modes. The model is used for studying load transients, power-sharing, and fault analysis. The implemented control system overcomes challenging situations such as load changing and transient conditions by managing the power of the battery and PV and regulating the voltage and frequency in the islanded mode. The Maximum Power Point Tracking (MPPT) is modified to include a feature for limiting the power in case of islanded mode and fully charged EV battery. Furthermore, the droop control and virtual inertia is utilized in a unified control manner. The model is implemented in MATLAB/Simulink and deployed to a real-time simulation by using an OPAL-RT simulator to validate the feasibility of the proposed model. The results for the real-time simulations are presented, showing the capabilities for voltage and frequency regulation of the controller, in load variations and fault condition.

Authors: B. Esser; Z. Shaw; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575563

Abstract: A pulsed RF source for multipactor research capable of providing 2 kW at 2.85 GHz to a multipactor test cell is described utilizing state-of-the-art GaN HEMTs from Cree/Wolfspeed and integrated GaN amplifier from Qorvo/Tri-Quint. A Mini-Circuits ZX95-2920CA+ VCO is used as the signal generator with low phase noise and modulation capable with a bandwidth of 55 MHz. An inline attenuator provides the operator power control in the range of approximately 42 to 64 dBm (16 to 2800 W). A high-speed RF switch creates a 100-microsecond pulse with a rise time of 25 ns and a typical switching time of 35 ns. A single Tri-Quint TGA2585-SM provides the majority of the gain in the system, 32 dB, and the necessary power for the final output stage (6 W). Four CGHV31500F amplifiers operating in parallel comprise the output stage providing the bulk of the power needed, 500 W each for a total of 2 kW with 12.5 dB gain each, with a single unit providing the necessary input power before splitting to the four finals (50.5 dBm, ~ 112 W). A custom PCB was designed to properly bias the GaN stages and prevent device failure due to improper bias sequencing. Custom power splitters and combiners had to be used due to the high-power levels being considered with directional couplers on the output to monitor output (forward) power and reflected (reverse) power during operation. This will provide valuable insight into characteristics of the plasma cloud generated during the multipactor event.

Authors: V. Roy; S. S. Noureen; S. Bayne; A. Bilbao; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8416176

Abstract: In this present study a hybrid renewable energy system is proposed to optimize the load demand for Center Pivot Irrigation System for remote cultivation areas in Lubbock, Texas. The energy resources included in the study are solar photovoltaic and wind energy. For the ease of study, a 120-acre cultivable land area is considered as model. Simulation and analysis of the load is developed using HOMER (Hybrid Optimization Model for Electric Renewable) software and the optimum use of renewable resources is determined. A RT-LAB based model is also developed to determine real time analysis of center Pivot load. As the impact of wind is remarkable and solar irradiance is significant in Lubbock area, the wind turbine and solar photovoltaic system is given priority for effective optimization. Using HOMER software and RT-LAB based analysis of different combinations and multiple components are considered for simulation study. An optimal solution is proposed in the work by considering the renewable energy resources as prime sources in Lubbock area for this optimization method.

Authors: Matthew Kim; Jose A. Rodriguez; William B. Ray; Stephen B. Bayne; Heather O’Brien; Aderinto Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8936768

Abstract: Gallium nitride (GaN) high electron mobility transistors (HEMTs) are an ideal option in applications of power electronics due to the wide-bandgap properties of the material. High electron mobility is gained through the device's unique channel structure. This research investigates whether state-of-the-art GaN HEMT semiconductors are reliable in a long-term operation in high power switch-mode conditions. Information on overcurrent capability about GaN HEMTs is not well established, thus a demand to investigate the devices exists. The GaN HEMT GS66508P from GaN Systems, was tested in pulsed overcurrent operations to establish the performance and to observe any operational changes after the experiment. The device is rated at 650 V and 30 A continuous. The goal of this research is to see if the device characteristics change after overcurrent pulsing and to analyze the device degradation that occurs in higher energy density.

Authors: Q. Hu; A. R. Chowdhury; R. P. Joshi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575663

Abstract: Electric pulse driven membrane poration finds applications in the fields of biomedical engineering and drug/gene delivery. Shock waves are known to permeabilize cell membranes as well. Here we focus on the synergistic effects of both inputs in concert based on molecular dynamics simulations. Our results show that shockwaves could be used for pretreating cell membranes in the electroporation process. The dual strategy would either reduce the external voltage requirements (leading to more compact external circuitry) or help create larger pores. Furthermore, shockwaves could form pores at any desired membrane site location, and suitable combinations of nanojets and electric pulses would help control the aspect ratio and size as desired.

Authors: N. Shamim; A. Bilbao; D. Reale; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8373603

Abstract: This Paper analyses the integration of a super capacitor with fuel cell in grid connected mode. A super capacitor is a fast charging and discharging device. In a fault condition super capacitor will discharge and maintain the voltage of the DC bus. In this paper a mathematical model of the fuel cell power system is designed. A super capacitor and a grid are modeled in PSCAD simulation environment. The fuel cell power system is integrated with the grid using model predictive control technique. A fault is applied at the DC bus to study the impact of integrating a super capacitor with the fuel cell. The simulation results show the super capacitor can maintain the voltage at DC bus at fault condition.

Authors: Hieu K. A. Nguyen; John Mankowski; James C. Dickens; Andreas A. Neuber; Ravi P. Joshi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9575287

Abstract: Multipactor in a rectangular waveguide is studied using numerical simulations. Particular attention is given to the secondary electron emission characteristics including their energy spectrum (hence velocity spread) and angular distribution. Elastically scattered, rediffused and true secondary electrons are all comprehensively included based on the Furman-Pivi model [1] for the TE10 mode. The focus is on small waveguides and lowest order resonance conditions.

Authors: D. H. Barnett; A. A. Neuber; J. C. Dickens; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8936727

Abstract: This study serves to describe three-dimensional particle-in-cell (PIC) simulations of a tunable reflex-triode virtual cathode oscillator (vircator). Experimental data from the compact hard-tube reflex-triode vircator developed at Texas Tech University (TTU) is used to validate simulated results. The vircator developed at TTU is capable of burst-mode operation at pulse repetition rates (PRFs) up to 100 Hz for a period of one second. A pulse energy of 158 J drives the vircator, and 600 kV (open circuit) pulse forming network (PFN) based Marx generator. The vircator is comprised of a bimodal, carbon fiber cathode and a pyrolytic graphite anode, with the ability to quickly change the distance between the anode-cathode (A-K) gap, back wall distance, and bottom plate distance between experiments. The PIC simulations have been performed using CST PIC Solver, by Dassault Systemes. The models detail virtual cathode formation and the subsequent extraction of radiated microwave power for a variety of cavity geometries. A working three-dimensional, relativistic, electromagnetic, particle-in-cell model of a vircator allows for quick, predictive results relative to building an experimental setup. The model is used to determine the necessary driving voltages, A-K gap distances, and cathode current densities to extract microwave radiation at a desired. Simulated results aid in identifying mode contributions. Voltage, current, and microwave data are presented and compared against experimental results at different operating conditions.

Authors: M. J. Kumar; S. Hahmady; R. Gale; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8936701

Abstract: In this paper, we investigate the new approach on designing a high voltage PIN diode without introducing any doping into the silicon. The concept of charge plasma is basically using different metals with appropriate work-function as anode and cathode contacts, which cause the formation of "p" and "n" plasma regions in silicon respectively. We have used the Silvaco Atlas simulation to compare the forward and reverse I-V characteristics of the proposed device with the conventional PIN diode. Also, we looked at the temperature dependency of their reverse saturation current. We demonstrate that by using the charge plasma concept, the proposed CP-PIN diode and conventional PIN diode have identical characteristics.

Authors: J. C. Stephens; J. C. Tucek; M. A. Basten; K. E. Kreischer; M. A. Shapiro; R. J. Temkin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8391585

Abstract: This report summarizes the progress of experimental efforts for the development a 94 GHz extended interaction klystron (EIK) with a photonic bandgap (PBG) based interaction structure. Input return loss measurements of the interaction structure are presented, demonstrating the successful fabrication of the input and output cavities of the EIK. Additionally, input return loss and insertion loss measurements of microwave-vacuum feedthrough windows are also presented. Finally, a review of the complete W-band experimental apparatus is given. Hot test experimental efforts for this study are expected to be completed by the 2018IVEC meeting.

Authors: B. Esser; J. C. Dickens; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575799

Abstract: Ionospheric Heating, traditionally performed utilizing large fixed location arrays, while effective, limits research efforts to those fixed locations. As one may envision, a more compact transportable array will provide additional research opportunities at latitudes previously unexplored. For instance, closer to the equator where the Earth's magnetic field is primarily parallel to the ground. An electrically small inductively coupled antenna is considered for its suitability in such an array. A Small or Semi Loop Antenna (SLA) inductively couples to a Capacitively Loaded Loop (CLL) providing high efficiency and natural 50 $\Omega$ port matching. The CLL, of hinged petal design, consists of a large parallel plate capacitor, and half cylinder inductive sections which are hinged at the base to allow for tuning in the range of ionospheric heating (3 - 10 MHz) with a plate included angle of 0 - 16°. A prototype antenna was designed and evaluated at 1/10^th scale - for ease of research efforts - with tuning range of 30 - 100 MHz including capability of tuning the coupling between SLA and CLL to achieve good source matching across the tuning range, particularly at the low and high ends. This tuning method may also be used to compensate for mutual impedance effects in array operation, improving array performance. A gain of approximately 5 dBi is observed when placed upon a sufficiently large ground plane. Tuning of the antenna prototype is achieved via stepper motor driven system to remotely adjust both the CLL and SLA angles continuously.

Authors: V. Roy; S. S. Noureen; S. B. Bayne; A. Bilbao; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8671342

Abstract: Recent advancement in Power System Analysis shows that implementation of PMU (Phasor Measurement Unit) in Smart Grid playing a significant role over SCADA. The main reasoning for that is more sampling data than traditional SCADA system. Every PMU data like voltage, current and Phase angle gives more samples in every second which is helpful for event detection. The enormous data send by each PMU in every second energies the big data issue. To find out and predict the transient situation and even small disturbances or anomalies from big data analysis within the specified short period of time is a challenge for near future. Because introduction of new smart electrical devices will boost up the big data issue. Processing of big data for post disturbance analysis is also an important task. This paper gives a scenario of PMU measurements received to PDC (Phasor Data Concentrator) from PMUs placed in distinct locations and detection of transient events for post disturbance analysis. In this analysis, the disturbances are evaluated with the R programming analysis and compare findings of chronological data from separate locations and also shows the relation between disturbances in a grid. For this analysis, the impacts of frequency and voltage data are also considered.

Authors: N. Shamim; G. M. Munoz; D. Reale; M. Harral; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8612317

Abstract: This Paper evaluates the integration of fuel cell power system with the grid using finite control set model predictive controller. A mathematical model of the phosphoric acid fuel cell power system is designed. The grid and the fuel cell are modeled in PSCAD simulation environment. Direct power control technique for model predictive controller is used to control the converter. The steady state results are shown in the paper, Also, a single phase fault and three phase fault is applied at the grid side to study the effect of integrating the fuel cell with the grid. The simulation results show model predictive controller can effectively integrate fuel cell with the grid.

Authors: I. A. Aponte; B. Esser; Z. Shaw; J. C. Dickens; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575875

Abstract: RF breakdown at a frequency of particular relevance to ionospheric heating - 3.3 MHz, close to the low end of the applicable frequency range - is studied at centimeter sized gap distances and compared to literature for small gaps. Paschen's famous DC breakdown study utilizing two brass spheres of radius 1 cm was recreated following the original procedure, from which the data was used to compare to a study of RF breakdown. Through testing it was found that brass as an electrode material exhibits a large standard deviation in breakdown voltage and as such stainless-steel Bruce profile uniform electric field electrodes were substituted. Steel's resistance to surface ablation provided for extremely low standard deviation of measurements and hence good repeatability. Additionally, unlike brass, steel is not contaminated through reaction with gaseous elements such as carbon, oxygen and nitrogen. Electrodes were polished to remove any field enhancements and cleaned of polishing compound such that dielectric inclusions were largely avoided.

Authors: X. Lu; J. C. Stephens; I. Mastovsky; M. A. Shapiro; R. J. Temkin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8391535

Abstract: We present the experimental results of a high power microwave source with a metamaterial structure designed at 2.4 GHz. The structure is a waveguide loaded with two metamaterial plates consisting of complementary split ring resonators placed in the waveguide with reverse symmetry. Two backward wave modes can propagate in the waveguide and interact with an electron beam of up to 490 kV, 84 A. Full microsecond long output microwave pulses up to 2.9 MW were measured with an efficiency of 9% from the Cherenkov-cyclotron instability.

Authors: M. Powell; Z. Shaw; J. C. Dickens; J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575375

Abstract: The design and implementation of an experimental setup are demonstrated for the breakdown testing of pressurized $\text{SF}_{6}$ (1–3 atm) at 2.85 GHz. Sulphur hexafluoride $(\text{SF}_{6})$ is a common insulating gas used to mitigate breakdown in high power microwave systems. While this gas is known for its high dielectric strength, available data for the breakdown of pressurized SF 6 and $\text{SF}_{6}$ mixtures at S- Band frequencies is limited. The obtained breakdown characteristics are compared to mixtures of $\text{SF}_{6}$ and Nitrogen $(\mathrm{N}_{2})$ for similarities in breakdown fields in a pressurized environment.

Authors: Nicholas A. Wilson; Jared A. Culpepper; Vladimir Kuryatkov; Matthew Gaddy; James C. Dickens; Sergey Nikishin; Richard Ness; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9575438

Abstract: The suitability of commercially available wide bandgap GaN material for the fabrication of photoconductive semiconductor switches, PCSS, was investigated. A variety of PCSSs were fabricated utilizing diverse GaN samples, which were shown to exhibit significantly diverse physical properties. That is, sample characterization techniques such as cathodoluminescence (CL), photoluminescence (PL), secondary ion mass spectrometry (SIMS), Current-Voltage behavior, and scanning electron microscopy (SEM) were applied to characterize the samples prior to processing.

Authors: G. M. Quintero; Y. Reddy Challapuram; A. Bilbao; S. B. Bayne; A. S. Subburaj; M. A. Harral

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8612417

Abstract: The integration of Microgrids (MGs) into the mains must be done with consideration of control techniques that ensure the appropriate synchronization and power balance between distributed generators (DGs) and the grid. This paper presents the development of a PQ-control model for the grid connected single-phase and three-phase inverters present in the Distributed Asset Research Testing (DART) facility in Lubbock, Texas. In a grid-tied configuration, the inverters will operate as a current source that inject current into the grid based on the established reference setting of active and reactive power. To achieve this operation, the inverter current must be monitored, and it will be subjected to abc to dq0 transformation (Clarke and Park transformation). In the case of a single-phase inverter, two orthogonal phase variables are required in order to perform the Park transformation (αβ-dq). A phase shift of 90° with respect to the real phase variable is introduced to get the beta component required to complete this transformation. By showing the power characteristics in simulations, the proposed control strategy will be illustrated. The entire work is performed in MATLAB/SIMULINK environment.

Authors: Giesselmann, M; Bayne, S; Shamin, N; Kelley, M; Reale, D; Cingoz, F

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8936837

Abstract: We are reporting on models for a group of inverters that can feed real and reactive power into a utility grid in Grid-Tied mode and is able to smoothly transition to islanded mode. In grid tied mode, the inverters are operating in P/Q mode and inject controllable amounts of real and reactive power into the grid. In islanded mode the inverters are grid forming and share power using droop control. We are presenting MatLAB-Simulink models and results of the simulations including the transitions.

Authors: M. Giesselmann; V. Roy

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8936816

Abstract: LTspice is a powerful simulation language that is specifically optimized for modeling Switch Mode Power conversion. It is not limited to small numbers of nodes and freely available [1]. We are presenting several examples of simulations for popular electronic power factor correction circuits that improve the input power factor of AC Power Supplies by active wave-shaping of the AC input current and the associated avoidance of harmonics. The simulations are performed using a cycle-by cycle switching approach as well as using a time averaged PWM model. A fast inner current loop is controlling the shape of the input current such that it matches the sinusoidal shape of the AC input voltage. A time averaged PWM model is useful for fast simulations covering many cycles to design and study secondary feedback control that regulates the output voltage of the converter.

Authors: S. B. Negrete; R. Prakash Joshi; R. Labuguen; J. Matsumoto; T. Shibata

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8641000

Abstract: In this paper, we propose a visualization framework for mouse anatomical cardinal planes and axes by extending an open-source platform called "3DTracker-FAB" and demonstrate its capability towards augmentation. Previously, the 3DTracker-FAB was only able to determine the mouse anatomical model, showing its head, neck, trunk, hip, and nose. We enhance the software to include body axes and planes of the subject in relation to its anatomical model. This work will help scientist working with animals since anatomical axis and planes are used for describing motion, and anatomical location.

Authors: J. P. Verboncoeur; N. Behdad; J. H. Booske; J. C. Dickens; R. M. Gilgenbach; M. Gilmore; N. M. Jordan; R. P. Joshi; Y. Y. Lau; J. Mankowski; D. Morgan; A. A. Neuber; S. Portillo; E. Schamiloglu; P. Zhang

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575295

Abstract: Multipactor onset, growth, associated space charge effects, and transition to ionization breakdown due to ambient or desorbed gases represent key stages of single and multifrequency RF -driven phenomena that inhibit performance in space-based and terrestrial vacuum electronics devices. Performance degradation through space charge detuning and interference with gain is expected for medium duration pulses, and ion generation and damage for longer pulses. In this research, combined theoretical, computational, and experimental approaches are applied.

Authors: A. R. Chowdhury; R. P. Joshi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8936791

Abstract: Time-dependent photocurrent response in semi-insulating GaN is simulated with a focus on the Lock-On phenomenon. A one-dimensional, time-dependent model based on the drift-diffusion theory is used. The model is tested for GaAs and shown to yield good agreement with data. The GaN simulations are then performed. The main findings are that deeper traps nearer the valence band at higher densities, and materials with larger high-field drift velocity would all aid in attaining Lock-On. The threshold field for Lock-on in GaN is predicted to be around 150 kV/cm, though this is strongly dependent on the trap parameters.

Authors: Z. Shaw; B. Esser; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575386

Abstract: An experimental setup was designed and implemented for the study of the multipactor effect in a rectangular waveguide under high vacuum. The experiment and test piece were constructed using WR 284 waveguide sections operating at S- Band frequencies. Multipactor effects are to be detected via an Electron Multiplier Tube (EMT). The test setup includes sections of standard WR 284 waveguide which will be in line with a custom test piece. Design of the test setup incorporates a standard piece of waveguide with a custom steel collar which can be used to insert different tapers, and even periodic structures into the hollowed out broadside wall. A solid state RF transmitter with 2.8 kW peak power and a center frequency of 2.85 GHz is used to inject an RF signal into the test section.

Authors: Z. Shaw; A. Hewitt; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9575943

Abstract: A multi -channel phased array is used to synthesize signals of varying frequencies through apertures in EM shielded structures. Bipolar impulses with a GHz frequency spectrum are emitted from multiple antennas and used as basis functions to pass through the aperture and undergo constructive and destructive interference at a specific point in space. Experiments based on a small antenna array demonstrate that this technique enables synthesizing a below GHz signal (~ 400 to 800 MHz) in a shielded volume with an open aperture size that would heavily attenuate a direct sub-GHz electromagnetic wave. While the experimental results demonstrate that this technique is effective at UHF frequencies, it is certainly not limited to this frequency band.

Authors: M. Czerniak; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8373136

Abstract: As the semiconductor industry makes devices and integrated circuits (ICs) that are increasingly complex, a consequence has been that the number of processing steps is increasing, from 400 at 90nm to > 1000 in state-of-the-art designs [1], shown in Figure 1. Furthermore, despite many process steps becoming more utility (and especially electrical power) - efficient, the increasing number of times a wafer visits process chambers has resulted in the energy use per cm2 wafer area increasing, reversing the trend of previous years. Figure 2 illustrates this phenomenon, which is also exacerbated by the use of double and quadrupole patterning, 3D device stacking and the use of EUV in HVM (which reduces the number of process steps but is utility-intensive).

Authors: Jonathan Forbes; Fernando Salcedo; Cedrick Tchoupe-Nono; Richard Gale; Stephen Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8936769

Abstract: As an increasing number of silicon carbide (SiC) power semiconductor devices become commercially available, and as silicon devices have reached their theoretical power density limits, SiC devices are being utilized in an increasing number of power electronics and pulsed power applications. A few examples of these applications include high-power DC-DC converters, inverters, motor drives, or high-voltage pulse generators such as a solid-state Marx generator. To encourage further adoption of SiC devices in these and other applications, further reliability testing and analysis must be conducted. One parameter that is important to study is the surge current capability of both SiC diodes and the body diodes of SiC MOSFETs. In this research, a surge current testbed was designed and built to test commercially available 1200 V / 10 A SiC JBS diodes from 3 different manufacturers, and the body diodes of 1200 V / 10 A SiC MOSFETs from 3 different manufacturers. The purpose of this work is to independently verify manufacturer datasheet claims regarding the surge current capabilities of their diodes. In addition, surge current ratings on the body diodes of the SiC MOSFETs are determined and published.

Authors: D. H. Barnett; K. Rainwater; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8291279

Abstract: This paper presents the electrical and mechanical hardware considerations of a compact, 160 J modular pulse forming network (PFN) Marx generator used to drive a high-power microwave (HPM) source that is a time variant load at a PRF of 100 Hz. The modular Marx generator is designed to produce an open-circuit output voltage of 600 kV from a 50 kV source using twelve stages. Each stage of the Marx was constructed from a PFN fashioned from five, 2.1 nF, high voltage capacitors in parallel. Each Marx module was machined out of acetyl copolymer or Delrin© to provide optimal strength, rigidity, and a dielectric constant that closely matches transformer oil. These Marx modules include air supply lines that are machined directly into each block of Delrin© allowing airlines to connect to each module chamber rather than every spark gap. Each module has two electrode inserts placed into the sealed pressure vessel contained within the module. After the Marx erects, the energy is directed into the virtual cathode oscillator (vircator) where subsequent frequency generation is manipulated through a rectangular waveguide contained within a new resonator cavity design. The new design allows the bottom wall, back wall, and anode cathode gap to be moved by two linear actuators, a linear bellows, and another linear actuator, respectively. The cavity is contained within a 10-inch circular vacuum chamber with a round stainless steel sleeve running from the back wall to the linear bellows. Contained within the round sleeve is a rectangular waveguide where the bottom wall and the cathode are housed. The anode is connected to the Marx generator via a nickel shaft that feeds through the back wall into the circular sleeve and into the rectangular waveguide. The anode made from pyrolytic graphite, remains stationary while the bottom wall, and carbon fiber velvet cathode move relative to its position. The benefit of this design is the height and depth of the cavity resonator can be controlled independently of each other while still allowing the A-K gap to be manipulated on its own.

Authors: K. Saxena; R. Labuguen; R. P. Joshi; N. Koganti; T. Shibata

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8324415

Abstract: As the robot technology is advancing, it is possible to use robots for basic day-to-day chores, so that the burden can be taken off from humans. To make the robots perform such tasks, it is necessary for them to handle different types of objects. Manipulation of deformable objects such as cloth is a challenging task for a robot because of high dimensionality and large number of possible configurations of cloth. Previous studies have covered large number of simple manipulations of cloth articles. In this paper, we are focusing on table setting task that requires putting on a sheet of cloth on the table. This paper proposes human-robot collaboration for table-setting task based on visual assessment. We used Baxter robot to hold two corners of rectangular tablecloth and other two corners are held by human. A head-mounted Kinect sensor is used to get the state of cloth and Robot arms are used for controlling the position of cloth corners. We use features from Kinect sensor to assess whether the placement of the cloth is successful or not. We demonstrate an initial study of the system that can achieve promising results towards table setting task through human-robot interaction.

Authors: S. S. Noureen; V. Roy; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8283453

Abstract: This paper summarizes the features and advantages of the renowned Real-Time Digital Simulator (RTDS) called OPAL-RT Simulator. Alongside with its applications in the electric power system for the system design, analysis and testing. This comprehensive study also includes a brief description of its software, hardware, I/O system, modeling, Hardware-in-the-Loop and other prospects. Tremendous advancement in the storage capacity and computational speed of the modern computer system has done for the last few decades. Assimilating this advancement, this simulation tool has designed with unique features like real-time model-in-the-loop, Hardware-in-the-loop. Also for moderate operational and maintenance cost, OPAL-RT Simulator has become more prevalent to the researcher. There are immense scopes of experimental research in power systems analysis. This is the prime reasoning for describing the necessity and significance of OPAL-RT simulator in this work. In addition, a semi-simulation experiment of a PV system is presented using RT-LAB software. Two of the most salient tools of OPAL-RT are 1) RT-LAB and 2) ARTEMIS, which were applied in power electronics are also summarized.

Authors: R. Rodriguez-Molina; D. Mauch; V. Meyers; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496332

Abstract: Results detailing the current handling capability of Gold bond wires (1 mil diameter), and Aluminum ribbon (1mil x 10 mil) under pulsed conditions are presented. Gold wire bonds were formed through the ball-bonding technique, and the Aluminum wire was bonded through the wedge bonding technique. Both wires were bonded to gold plated bond pads. Varying peak current densities ranging from $2 \mathrm {x}10 ^{4}\mathrm {A}$ cm $^{-2}$ to $2 \mathrm {x}10 ^{7}\mathrm {A}$ cm $^{-2}$ were applied to bond wires at pulse-widths ranging from 50 ns to $500 \mu \mathrm {s}$. In addition, the effect of surrounding medium (SF6, HV epoxy, transformer oil) on the current capability was investigated. Failure modes observed were investigated with SEM imaging, and the evolution of these failure modes during current pulsing was captured with high-speeding imaging. Finally, the obtained results are compared with the experimental and theoretical results obtained from previous investigations conducted on exploding wires.

Authors: J. C. Stephens; G. Rosenzweig; M. A. Shapiro; R. J. Temkin; J. C. Tucek; M. A. Basten; K. E. Kreischer

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8289729

Abstract: A complete design of a 94 GHz extended interaction klystron (EIK) amplifier has been developed at the MIT Plasma Science and Fusion Center. The device utilizes a novel, mode selective photonic bandgap (PBG) structure which enables the use of a modified PBG coupling scheme. Using particle-in-cell simulation, the EIK is predicted to generate 42 dB of gain and 130 W saturated power.

Authors: G. Rosenzweig; J. C. Stephens; M. A. Shapiro; R. J. Temkin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496291

Abstract: The challenge in manufacturing Traveling Wave Tubes (TWTs) at high frequencies is that the sizes of the structures scale with, and are much smaller than, the wavelength. This requires advanced nano-machining techniques or sheet-beam devices and puts strict limits on the peak output power that can be safely handled by the devices. Furthermore, the diameter of the electron-beam tunnel limits the amount of beam current, requires high magnetic fields for beam compression and creates difficulties in alignment.

Authors: N. A. Wilson; D. L. Mauch; V. E. Meyers; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496006

Abstract: Summary form only given. A very compact system integrating a high voltage pulser (up to 250 V) and high power UV LED (365 nm wavelength) was developed for triggering SiC photoconductive semiconductor switches (PCSS). The relationships between LED drive current and forward voltage to optical power were established for currents ranging from 0 A to 100 A at pulse widths ranging from 100 ns to 5 μs. The maximum optical power observed was 25 W. An optical system composed of a parabolic reflector and short focal length lenses was used to focus the emitted light onto a lateral geometry PCSS. The observed delivery efficiency was 63% onto a target 11 mm in diameter. Using the developed system, a SiC PCSS could be successfully triggered under high impedance load conditions.

Authors: H. K. Nguyen; A. Chowdhury; J. C. Dickens; R. P. Joshi; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496212

Abstract: Breakdown of air at atmospheric pressure in response to AC fields in gaps larger than 1 cm was simulated. Most previous literature concerning breakdown in this regime has focused on much smaller gaps1.

Authors: Zhang, SM; Li, HW; Dai, YS; He, M; Lu, RX

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8254988

Abstract: In cloud computing, outsourcing computation, as an economically promising paradigm, enables clients with limited computing resources to delegate intensive tasks to the powerful cloud server. Nevertheless, outsourcing the involved data without encryption to public brings a high risk of revealing the confidential information of the client, and no matter the returned results from server are valid or not. Due to the low trustiness of the server, it is crucial to guarantee both the security and efficiency in delegation. Motivated by those challenges, a rapidly-growing body of research mostly focuses on how to delegate mostly utilizing a feasible mechanism-verifiable computation, where the server returns results along with corresponding verification proofs. In this paper, we propose an efficient and privacy-protected delegation scheme for matrix multiplication (EPP-DMM) of amortized model. EPP-DMM can realize efficient public verification together with sensitive protected data. Security analysis proves the reliability and security of our scheme, while performance evaluation presents the efficiency compared with other schemes.

Authors: Flack, T; Parson, J; Bittner, K; Driver, B; Zameroski, N; Bayne, S; Hettler, C

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291289

Abstract: Field Effect Transistor (FET) controlled devices, such as MOSFETs and IGBTs, exhibit several desirable characteristics over other solid-state devices for pulsed power applications; benefits include high frequency switching, voltage control, and compact control circuitry. This paper details a testbed used for evaluation of the switching performance and characterization of FET controlled devices used in pulsed power systems as well as the diagnostics used to characterize the devices under test (DuT). The testbed presented in this research operates in two modes: (1) High energy pulsed mode, with charge voltages up to 300 V, pulse width of 3 seconds and up to 18 kJ total stored energy (2) Continuous pulse-train mode, with charge voltage up to 300 V, up to 18 kJ total stored energy and average current output up to 40 A. Both of these modes utilize a 396 mF capacitor bank to store energy. A fast, custom, gigabyte-memory-depth data acquisition oscilloscope records voltage and current waveforms at a 60 Mega-Sample/second rate. Due to the frequency regime (1-50 kHz), high current levels (up to 300 A peak) and wide voltage range of these experiments, making these measurements, reliably, is a non-trivial effort. Several methods of making each measurement were examined. Calibrated voltage, current, energy, and power waveforms quantify the DuT's turn-on / conduction / turn-off characteristics. Measurements of interest in these experiments are device current and device voltage; energy dissipated in the DuT is determined from these measurements.

Authors: Buntin, T; Collier, L; Dickens, J; Mankowski, J; Walter, J; Neuber, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291285

Abstract: Transient magnetic diffusion through conductors of thickness comparable to the skin depth is investigated. Since an analytical solution is unavailable in this case, such magnetic diffusion results must be determined via simulation or experimentation. In the experimental approach, a sinusoidal current with peak values in the range of 20-30 kA (approx. 7 kHz ringing frequency) is passed through a two turn coil generating a sinusoidally varying magnetic field. A hollow structure with metallic walls of controlled thickness is placed roughly 10 cm away from the exciting coil. The focus of this investigation is on the transient skin depth, which occurs during the first half-wave of the signal, as that is most relevant for pulsed power applications. A calibrated B-dot probe placed inside the structure facilitates measurement of the diffused field. As expected, experimental data shows that magnetic field diffusion through the wall is not instantaneous, causing a delay before the diffused field is measured inside the test structure. The impact of cracks and holes in the conductor on the speed and magnitude of the magnetic field diffusion is elucidated. Results for materials of different conductivities are compared and analyzed for the transition between thin and medium walled cases. FEM simulations are validated alongside these experimental results and used to access a larger parameter space.

Authors: N. Shamim; A. Subburaj; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7923935

Abstract: This paper describes the Model Predictive Control technique for three phase bi-directional converter to integrate a battery system with the grid. The paper presents an overview of different predictive control technologies. The paper describes the basic concept, operating principle, governing equations and control algorithm of model predictive control for the power converter. The control technique is analyzed to integrate a 1MWh battery system model with the grid. The analysis is done in PSCAD simulation environment for both steady state and fault scenarios. The simulation results are presented to show the effectiveness of Model Predictive Control technology for battery integration.

Authors: Hirsch, EA; Schrock, JA; Bayne, SB; O'Brien, H; Ogunniyi, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291248

Abstract: With the progression of silicon carbide (SiC) technologies, single semiconductor switches with higher voltage and current capabilities are emerging. Evaluating the pulsed current capability of SiC semiconductor devices for pulsed power and power electronics applications is required to understand their performance and reliability. This paper presents the narrow pulse evaluation of 15 kV SiC MOSFETs (0.25 cm(2) active area) and IGBTs (0.32 cm(2) active area) with pulse widths in the range of 500 ns to 2 mu s. Testing results are presented with an 8 kV charge voltage and 50 A and 330 A peak conduction current for the MOSFET and IGBT, respectively. A fairly low inductance (< 500 nH) RLC circuit was used to generate the pulses and the device under test (DUT) was switched off during the pulse to create a trapezoidal type current waveform through the device. Transient characteristics, such as turn-on and turnoff times and energies, were measured to benchmark the devices(narrow pulse characteristics. The results presented in this paper demonstrate the characteristics of these devices for over-current narrow pulse applications.

Authors: B. Esser; J. C. Dickens; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496347

Abstract: An electrically small antenna (ESA) is evaluated for its potential future use in a Transportable Ionospheric Heating (TIH) array. Consisting of a Small Loop Antenna (SLA) which inductively couples to a Capacitively Loaded Loop (CLL) the antenna provides a high-Q natural match to a $50 \Omega $ source. The capacitance of the CLL may be adjusted to tune the antenna in the range of ionospheric heating of approximately 3 – 10 MHz. Several methods are evaluated to achieve this tuning including a horizontal sliding plate design, and a hinged petal design.

Authors: V. Meyers; D. Mauch; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291268

Abstract: Intensity-dependent nonlinear light absorption in bulk 4H-SiC at the above-bandgap energy of 3.49 eV (λ = 355 nm) is studied. Characterization and understanding of such nonlinear optical behavior in 4H-SiC forms the basis efficiency improvements and design of optoelectronic SiC devices, including photoconductive semiconductor switches. It is noted that previous research performed elsewhere had focused primarily on nonlinearities at below-bandgap energies, while little had been explored above-bandgap. In this study, absorption of short laser pulses with fluences ranging from 1 mJ/cm2 to 30 mJ/cm2 incident on 160 μm-thick high purity semiinsulating 4H-SiC samples of varying recombination lifetimes is addressed. Sample bulk recombination lifetimes vary from 0.5 ns to 100 ns displaying the range of effects from growth, electron irradiation, and annealing. The effective absorption coefficient varies significantly within this range as an apparent function of bulk recombination lifetime. A four-level time-and space-dependent finite difference time domain (FDTD) model taking into account electron trapping, interband absorption, and free-carrier absorption was constructed that yielded further insight into the absorption dynamics. For instance, the importance of free carrier absorption and deep-level trapping in the nonlinear absorption behavior is elucidated.

Authors: S. S. Noureen; V. Roy; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8288967

Abstract: The Phasor Measurement Unit (PMU) is an imperative part for monitoring, protecting and controlling the current power system. For the current power systems, PMUs not only offer the synchronized measurements of real-time data of voltage but also the current and frequency. Placement of PMUs in each bus for monitoring the system is not feasible from economic point of view and also for big data handling. So it is an obligatory matter to reduce the number of PMUs in the bus system with the aim of attaining the maximum power system observation. Different techniques are being applied from past to present to solve this optimum PMU placement (OPP) problem such as heuristic method, mathematical programing. Heuristic method is a quick experience-based technique for solving the optimization problems. Various optimization methods for solving the OPP problems are being reviewed in this paper.

Authors: M. Kim; J. J. Forbes; A. V. Bilbao; J. A. Schrock; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291207

Abstract: The evaluation of pulsed power systems and their constituent components requires unconventional loads with exceptional voltage, current, impulse energy, and continuous power dissipation capability. This paper presents the design and construction of a reconfigurable resistive load with active temperature monitoring for the evaluation of ultra-high voltage pulsed power modulators and semiconductor devices. The load consists of a network of 15 ceramic resistors (outer diameter of 2.54 cm and length of 30.48 cm) mounted vertically in an oil filled aluminum tank. To enable exceptionally high-power dissipation, the oil is pumped through the tank and through a radiator. A microcontroller based module activates a fan on the radiator if a preset oil temperature is surpassed. Experimental results gathered demonstrate that the load withstood 10 kW at 10 kV for 30 minutes, and that the temperature of the oil reached 80 °C.

Authors: Barnett, DH; Rainwater, K; Dickens, JC; Neuber, AA; Mankowski, JJ

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291276

Abstract: This study focuses on achieving wide tunability of a compact reflex triode virtual cathode oscillator (vircator). The cathode is of a bimodal carbon fiber (CF) material paired with a pyrolytic graphite anode. These materials display ideal operating characteristics which include but are not limited to, long lifetime > 10(6) shots, high operating temperatures > 1000 K, and large current densities similar to 200 A/cm(2). A 12 stage, 158 J pulse forming network (PFN) based Marx generator serves to drive the vircator at 350 kV, 4 kA with similar to 100 ns pulsewidth. The operating frequency of interest is in the range of 1-6 GHz, where tunability is achieved by varying the length of the anode-cathode (A-K) gap, the length from the back wall to the A-K gap, and/or the distance from the bottom of the cavity to the A-K gap. The primary focus in this research was to increase the achievable frequencies by placing a square waveguide within a sealed vacuum tube. This allows the bottom part of the waveguide to be easily adjusted while still maintaining the waveguide integrity. The resulting microwave frequencies are shown along with the system performance.

Authors: Joshi, RP; Koganti, N; Shibata, T

PDF: https://dl.acm.org/doi/pdf/10.1145/3132446.3134878

Abstract: The need of robotic clothing assistance in the field of assistive robotics is growing, as it is one of the most basic and essential assistance activities in daily life of elderly and disabled people. In this study, we are investigating the applicability of using Dynamic Movement Primitives (DMP) as a task parameterization model for performing clothing assistance task. Robotic cloth manipulation task deals with putting a clothing article on both the arms. Robot trajectory varies significantly for various postures and also there can be various failure scenarios while doing cooperative manipulation with nonrigid and highly deformable clothing article. We have performed experiments on soft mannequin instead of human. Result shows that DMPs are able to generalize movement trajectory for modified posture.

Authors: Pushpakaran, B; Bayne, S; Ogunniyi, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291214

Abstract: The application of silicon carbide technology in p-i-n diode has facilitated the development of p-i-n rectifiers up to several kV blocking voltage with a much thinner drift region thickness as compared to its silicon counterpart. This research focuses on the 2D electrothermal simulation of a 10 kV 4H-SiC p-i-n diode model developed using Silvaco ATLAS software. The p-i-n diode structure was designed for 100 A/cm(2) forward current density with a cell pitch of 16 lam and an active area of 10 mu m(2). Physics based models were included to account for low-field mobility, carrier-carrier scattering, carrier generation recombination, avalanche breakdown, and lattice heating. The device model was simulated under steady state and transient conditions. Pulsed simulation of the p-i-n diode was carried out using an RLC ring down circuit to generate a 5 mu s wide pulse with peak current densities up to 5000 A/cm(2). The reverse recovery characteristics of the diode was analyzed for a forward current density of 100 A/cm(2) and varying turn-OFF dJ/dt to assess the limitation on usable switching frequency. Lattice temperature profile of the p-i-n diode was generated by including heat generation models during transient simulation to identify thermal hot spot formation and areas of possible failure during pulsed operation.

Authors: Pushpakaran, B; Bayne, S; Ogunniyi, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291218

Abstract: Key requirements for a solid state switch in a fast switching pulsed power circuit include high blocking voltage, high current conduction and fast switching capability. Typical pulsed power applications like plasma initiation and high-energy LASER require operating voltages in the order of several kilovolts. The development of a multi-kilovolt SiC MOSFET for fast switching pulsed power application would require detailed analysis of the device switching characteristics. Since the switching speed of a MOSFET is primarily dependent on the inter-electrode capacitances, it becomes critical to have a comprehensive understanding of the device capacitance and its effect on the gate driver requirements for narrow-pulse switching. In this research, 2D model of a 10 kV 4H-SiC MOSFET was developed using Silvaco ATLAS TCAD software and simulated for its steady state, AC, and transient characteristics. The device cell was designed for an active area of 5 mu m(2) and 100 A/cm(2) drain current density. The capacitance-voltage and gate charge curve for the SiC MOSFET were obtained via AC and transient simulation respectively. This data was used to estimate the gate drive requirements for the device under fast switching conditions.

Authors: A. Subburaj; A. R. Arra; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7923936

Abstract: In recent years the significant improvementin micro grid technology has led to the utilization of thelocal distributed sources. The microgrids also have agreater role in the minimization of transmission losses. During power blackouts they serve as a main source ofpower for few emergency loads. It is common that usuallythe AC grids are preferred over DC grids to transmitpower over long distances in spite of various issuesassociated with AC grids such as frequency dip, voltagedrop due to reactance, charging currents, leakagecurrents, low power factor problems, skin effect andFerranti effect. One of the major reasons for suchpreference is that voltage can be easily stepped up orstepped down using transformers. Whereas in a DC gridsystem voltage conversion accounts for switching lossesand equipment costs. But at the micro grid level the supplyvoltages are low, and most of them are DC sources. Theenormous growth in the power semiconductor technologymakes it possible to convert voltages in DC with low costand reduced power losses. The AC and DC microgridmodels were developed in MATLAB/SIMULINKenvironment. The stability and transient analysis areperformed during faults and sudden load variations onboth the AC and DC microgrids in real time using OPALRTreal time digital simulator. The effects of stability on aweak grid scenario have been analyzed for both AC andDC microgrids.

Authors: Z. Shaw; W. Feilner; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8496174

Abstract: A low power prototype antenna array with multi-channel transmission capabilities is incorporated to synthesize sinusoidal signals in free space from non-periodic source signals. Synthesis of a specified continuous signal is suitable in applications for spatially targeted plasma generation at high power. The system operates with an external 2.4 GHz clock source which allows for a 0.417 ns step resolution. Each antenna transmits an amplitude scaled set of non-periodic bipolar impulses $( \sim 2$ ns wide) which enables synthesizing an arbitrary signal at a specific point in free space only limited by the available number of antennas. A specific bipolar impulse set is found through the particle swarm optimization, PSO, technique with appropriate goal function selection. While the optimization is simply carried out in the time domain, the respective delay between antennas is matched in the experiment through electronic delays and/or physical antenna positioning.

Authors: Vincent Meyers; Daniel Mauch; Vladimir Kuryatkov; Sergey Nikishin; James Dickens; Andreas Neuber; Richard Ness

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8291269

Abstract: Photoconductive semiconductor switches (PCSS) made from bulk, semi-insulating GaN have been fabricated and tested under pulse-charged conditions. Switching response and photocurrent efficiency of GaN PCSSs triggered by sub-10 ns, 355 nm laser pulses is reported. It is shown that fast rise time (<;300 ns) voltage pulses can be used to charge a GaN PCSS to fields well beyond the DC breakdown field strength of GaN and improve switching performance. GaN's wide band gap, breakdown field strength, and electron mobility make it a material superior to SiC and far superior to GaAs for PCSS applications, though historically these materials have dominated PCSS research due to their relative ease of fabrication. Recent improvements to crystal quality and wafer size have allowed GaN and more recently semi-insulating GaN to play an increasing role in high-power and high-voltage solid state devices.

Authors: S. Lacouture; J. Schrock; E. Hirsch; S. Bayne; H. O'Brien; A. A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8291191

Abstract: To continually increase the voltage and current capabilities of power semiconductor devices, whether pushing older materials such as Si to its' intrinsic electrical limits or by employing newer substances like SiC or GaN, a thorough understanding of the entire device is required, from the basic physics of the material and its interactions with defects and passivation, up to the complete device structure, including terminal performance and device - level limitations. Of the fundamental parameters that affect device performance, the most complex and malleable is the carrier lifetime. Carrier lifetime has a profound effect on power devices designed for high voltage applications and power devices relying on conductivity modulation. This parameter cannot be given as a ball - park figure unlike mobility (and hence diffusion coefficients) as it is affected by nearly every processing step a device undergoes: a final device can have carrier lifetimes that differ drastically from the starting bulk material. The work herein utilizes a relatively new set of techniques collectively known as Lifetime Spectroscopy (LS) methods to extract fundamental material parameters relating to recombination activity: τη₀, τρ₀ and AEt. These LS methods directly measure recombination activity of defects and hence acquire characteristic data of defects and dopants that is complimentary in nature to the information gleaned about them from more orthodox methods such as Deep - Level Transient Spectroscopy (DLTS). The Open Circuit Voltage Decay (OCVD) method is used along with improved data manipulation algorithms to extract the effective carrier lifetime as injection and temperature are swept. A complete stand - alone system has been constructed that allows a very wide range of current injection (~1mA to > 200A) and built - in OCVD waveform acquisition. The first complete Temperature - Injection Dependent Lifetime Spectroscopy (T-IDLS) studies are carried out on a small signal PiN commercial diode.

Authors: K. Rainwater; D. Barnett; C. Lynn; J. Dickens; A. Neuber; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8012819

Abstract: This paper presents the electrical and mechanical design considerations of a compact, 160 J modular pulse forming network (PFN) based Marx generator used to drive a high-power microwave (HPM) source with a time variant load at a PRF of 100 Hz. The modular Marx generator is designed to produce an open circuit output voltage of 600 kV from a 50 kV source using twelve stages. Each stage of the Marx was constructed from a PFN created with five, 2.1 nF, high voltage capacitors in parallel. Each Marx module was machined out of acetyl copolymer or Delrin to provide optimal strength, rigidity, and a dielectric constant that closely matches transformer oil. These Marx modules include air supply lines that are machined directly into each block of Delrin allowing airlines to connect to each module chamber rather than every spark gap. The spark gaps are comprised of two electrode inserts placed into the sealed pressure vessel contained within the Marx modules. The electrode inserts are of a sleeve-electrode design, which allows the user to insert the electrode into the sleeve, then the sleeve into the Marx module. The benefit of this design is the ability to adjust the electrode gap spacing without compromising the high pressure seal. Two continuous charging inductors run between each PFN and underneath the Marx modules. Due to high voltages generated by the Marx, outer field shaping rings are used to reduce the field stress across the induction coils, resulting in longer lifetime. The inductors are also of modular design allowing for individual coils to be replaced in the event of failure. Output voltage and current waveforms from a 60 Ohm water load are presented.

Authors: William B. Ray; Matthew Kim; Argenis Bilbao; James A. Schrock; Stephen B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7799967

Abstract: Gallium Nitride (GaN) transistors are of great interest for pulsed power and high power applications due to the proven capability of Silicon Carbide (SiC) transistors. Due to recent advances in GaN power semiconductors, lateral GaN transistors need to be evaluated for their performance under repetitive pulsed overcurrent operation that can occur in power electronics or pulsed power applications. A normally-off GaN Systems GS61008P-E03-TY was evaluated in a pulsed ring down circuit at peak currents of up to 230 A over frequencies ranging from 0.5 to 20 Hz. Measurement of switching transient energy dissipation showed minimal difference over cumulative pulse history and pulse frequency. In addition, the device's electrical characteristics, including forward IV and transconductance, were measured throughout testing and revealed no significant degradation. These results demonstrate the GaN FET's robust ability to handle transient pulsed overcurrent conditions common for commercial power semiconductor device applications.

Authors: A. S. Subburaja; N. Shamim; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7462526

Abstract: The paper helps to understand the performance of the grid with battery and doubly-fed induction generator (DFIG) wind turbines when operating in a weak grid scenario with a low short circuit ratio (SCR). The concept will analyze the performance of the grid with battery connected DFIG both in steady state and transient scenario. The grid can be represented as a Thevenin equivalent circuit and is rated at 12.5 kV. A battery system and a wind system are connected at the point of common coupling (PCC). The battery system being modeled in PSCAD software is rated at 1 MW/1 MWh and the wind system model is rated at 2 MW. The grid has been characterized and analyzed based on two categories: strong grid and weak grid. A grid is considered to be weak when the flow of active and reactive power in the network causes a significant amount of voltage fluctuation at the point of common coupling. A grid is considered to be strong when the grid is stable with allowable nominal deviations with voltage and frequency. The strength of the grid can be measured by taking the ratio of grid's short circuit power with grid nominal power.

Authors: Takeshi Ihara; Daniel Mauch; James Dickens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7534406

Abstract: Summary form only given. Development of a switch capable of operation at high voltages, high repetition frequency, with long lifetime is essential for furthering pulsed power applications. Photoconductive semiconductor switches (PCSSs) possess inherent optical isolation and extremely low switching jitter (~10 ps), and have also been experimentally shown to be capable of switching high voltages (up to 50 kV) and currents with very fast rise and fall times (<; 1 ns)[1-2]. In this paper, we report the breakdown characteristics of a SiC PCSS triggered spark gap obtained via measurement of the voltage and current, and simulation of the electric field distribution. The triggered spark gap is composed of sphere-to-sphere electrodes and a field distortion electrode, which is kept at mid-potential in the center plane between the two spherical main electrodes. A normally open PCSS is connected between one of the main electrodes and the mid-plane (trigger) electrode, whose center bore diameter is varied from 5 to 15 mm. With application of the optical pulse to the PCSS, the trigger electrode is temporarily connected to the main electrode, effectively doubling the electric field between the trigger and opposite electrode, leading to main gap closure. In essence, while bulk SiC PCSS switching currents demand very high optical power input, the synergy of bulk SiC PCSS and traditional spark gaps enables the triggering of large current flows at very modest optical powers with low jitter. Overall, the obtained results reveal that incident laser energy and mid-plane electrode geometry heavily influence the breakdown characteristics of the spark gap including jitter time, and breakdown voltage.

Authors: S. Feathers; J. Stephens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7534367

Abstract: Summary form only given. A pulsed UV arc lamp discharge is driven by short pulses (100 ns FWHM) with peak current levels up to 550 A, with a repetition frequency up to 1 kHz. The arc lamp incorporates an ellipsoidal reflector which efficiently focusses > 60% of the total light output onto an approximately 100 mm2 area, appropriate to trigger a large area photoconductive semiconductor switch, PCSS. Experiments using xenonchloride (XeCl*, 308 nm) and xenon-fluoride (XeF*, 351 nm) were conducted. Using nanosecond electrical excitation pulses, excimer emission in both gases was observed at differing absolute output power levels. With XeCl* (HCl 0.06%, Xe 1.5%, Ne 98.44%) and XeF* (NF3 0.12%, Xe 0.36%, Ne 99.52%), XeF* exhibited both higher power and efficiency (~0.01%) compared to XeCl*, which, however, carried comparatively low levels of HCl in the available gas mixture. The optical output energy, temporal shape of the optical power, and efficiency over varying pulse-width, current, and gas pressure were measured. Overall, the efficiency was observed to increase with decreasing pulse-width, and the peak optical power was found to increase with increasing pressure. Using XeF*, the experiments yielded an average peak power just above 300 W with the maximum peak power observed being 400 W. To demonstrate possible future application, the UV output of the 1 kHz repetition rate lamp was utilized to successfully modulate a SiC photoconductive semiconductor switch.

Authors: A. V. Bilbao; M. G. Giesselmann; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012811

Abstract: Rapid capacitor chargers are typically used to charge a bank of capacitors with the purpose of discharging it into a pulsed power load [1,3]. Previous research shows that the charging voltage of the load can be accurately calculated in real-time using microcontroller software algorithms [1,5]. The objective of this paper is to report a hardware based approach to measure the charge transfer into the load capacitor and implicitly the capacitor charging voltage. The proposed circuit uses operational amplifiers in order to integrate the input charge. A microcontroller receives the integrated signal to compute the output voltage and stop the charging process when the target voltage has been reached. Failure to accurately detect the end of charge time could lead to an excessively large capacitor bank voltage. For this reason, the proposed method can be utilized as a primary means of end-of-charge detection in conjunction with a traditional voltage sensing scheme.

Authors: Barnett, DH; Rainwater, K; Lynn, CF; Dickens, JC; Neuber, AA; Mankowski, JJ

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7534308

Abstract: This study focuses on the achieving wide tunabiltiy of a compact reflex triode virtual cathode oscillator (vircator). The cathode is of a bimodal carbon fiber (CF) material paired with a pyrolytic graphite anode. These materials where chosen for their operating characteristics which including but not limited to, long lifetime > 106 shots, high operating temperatures > 1000 K, and large current densities ~200 A/cm3 . A 12 stage, 158 J pulse forming network (PFN) based Marx generator serves to drive the Vircator at 350 kV, 4 kA with ~100 ns pulsewidth. The operating frequency of interest is in the range of 1-6 GHz, where the tunability is achieved by varying the size of the anode-cathode (A-K) gap, the length from the back wall to the A-K gap, or/and the distance from the bottom of the cavity to the A-K gap. The primary focus in this experiment was to increase the achievable frequencies by placing a square waveguide within a sealed vacuum tube. This allows the bottom part of the waveguide to be easily adjusted while still maintaining the waveguide integrity. The resulting microwave frequencies are shown along with the systems performance.

Authors: A. V. Bilbao; J. A. Schrock; M. D. Kelley; E. Hirsch; W. B. Ray; S. B. Bayne; M. G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012882

Abstract: Silicon carbide power semiconductor devices are capable of increasing the power density of power electronics systems [1, 2]. In recent years, devices rated to block voltages up to 20 kV have been demonstrated [3]. These research grade devices must be fully characterized to determine operating characteristics as well as failure mechanisms. The purpose of this paper is to demonstrate the continuous switching performance of ultra-high voltage metal oxide semiconductor field effect transistors (MOSFET) rated for 15 kV / 10 A. A high voltage boost converter was developed to evaluate the continuous switching performance where the high-voltage MOSFET is utilized as the main switching element. During operation, the on-state voltage, gate leakage current, and dc characteristics are monitored to determine device degradation. Measured device degradation is presented as a comparison of initial and final dc characterization.

Authors: Hirsch, EA; Schrock, JA; Lacouture, S; Bilbao, A; Bayne, S; Giesselmann, M; O'Brien, H; Ogunniyi, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8012879

Abstract: Silicon Carbide (SiC) is a leading wide bandgap semiconductor for increasing the power density of high power applications. This paper overviews the long term reliability and safe operating area of 15 kV SiC PiN diodes during pulsed current conditions. An automated system is used to stress these devices with ultra-high current pulses and monitor degradation with in-system characterization. The system is capable of a 100 mu s full-width half maximum pulse width up to 15 kA, with a repetition rate of 0.5 Hz. Periodic in-system characterization measures device forward conduction and reverse breakdown. The devices in this paper are pulsed at current levels from 1.5 kA to 2.5 kA. Over 100,000 pulses at 1.5 kA have been performed with no degradation. The long term reliability and failure mode results for the 15 kV PiN diodes will be reviewed.

Authors: P. M. Kelly; C. F. Lynn; J. M. Parson; J. Dickens; A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534045

Abstract: Summary form only given. The results from the development of a three-dimensional particle-in-cell (PIC) model investigating frequency tunability of a reflex-triode virtual cathode oscillator (vircator) are presented. These efforts have focused upon achieving stable frequency output at many different frequencies in the S, L, and C-bands, from a single tube, using the ICEPIC (Improved Concurrent Electromagnetic Particle-In-Cell) code from the Air Force Research Laboratory (AFRL). Experimental data from the hard-tube vircator at Texas Tech University (TTU), which operates at background pressures less than 10-9 Torr and utilizes a bimodal carbon fiber cathode and pyrolytic graphite anode, is used to validate simulation results. Additionally, the vircator at TTU is capable of changing the accelerating voltage, the anode-cathode (A-K) gap distance, and the distance of the cavity backwall relative to the position of the A-K gap on the fly without breaking tube vacuum, all in an effort to achieve greater frequency tunability and output power. However, this creates a very large, time-intensive experimental parameter space and makes a simulation model attractive for exploring additional output capabilities. ICEPIC results from a large combination of A-K gap distances, backwall distances, driving voltages, and cavity diameters are presented and compared to experimental results, highlighting frequency tunability of the system from a single tube.

Authors: S. Abedi; M. He; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7747960

Abstract: Real-time actuation of scheduled reserve capacity in power system operations with high penetration of wind power is prone to failure on account of unexpected shortcomings in network transfer capability. In this paper, a graph partitioning-based reserve zoning method is incorporated into the security-constrained unit commitment to improve the deliverability of operating reserves in a reserve zone and mitigate possible congestions caused by uncertain wind power. A graph representation of power system is proposed in which the edge weights are quantified by the likelihood of secure transmission utilization for each line. The probability distribution of line flows are characterized by the uncertainty of multiple correlated wind farm output forecasts as well as credible line outage contingencies reflected on the line flows using distribution factors. The minimum k-cut problem using the Gomory-Hu equivalent tree is addressed as a simple and efficient method to solve the NP-complete partitioning problem. The resultant zones can assure reduced risk of congested operating conditions and thus, provide a new approach to efficient management of intra-zonal congestions.

Authors: A. R. Chowdhury; H. K. Nguyen; R. P. Joshi; J. C. Dickens; J. J. Mankowski; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534285

Abstract: Summary form only given. Breakdown of air at atmospheric pressure in high frequency uniform electric fields and large gaps is discussed. In the high frequency band of a few MHz to few tens of MHz, the breakdown threshold voltage is lowered from its DC value due to enhanced space charge from ions that become trapped in the gap.1 While there is some literature concerning breakdown in this frequency range, it does not consider gaps larger than 1 cm.²A fluid model is developed to simulate plasma development in a baseline 6 cm gap primarily to explore power limitations for high power, electrically small antennas, which are operated cw at MHz frequencies. The ion densities are obtained from a drift-diffusion model, though data for the ionization, electron collision, and attachment parameters were obtained from Monte Carlo simulations, while ion diffusion and drift velocities were taken from the literature. As expected, the Monte Carlo simulations reveal that the EEDF follows any change in the electric field on the picosecond timescale at atmospheric pressures, much faster than any variation due to the externally applied electric field. Results from the simulation for gap lengths varying from the 6 cm baseline and air pressures are obtained, analyzed, and also compared with available reports.3

Authors: D. L. Mauch; V. E. Meyers; R. P. Joshi; A. A. Neuber; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534404

Abstract: A comprehensive picture of the relationship between optical fluence, optical wavelength, system load, and photocurrent efficiency (PE) in SiC photoconductive semiconductor switches (PCSSs) is presented. Variation of the optical wavelength (300-380 nm) and optical fluence (0.2-200 J m-2) was accomplished with a Nd:YAG pumped optical parametric oscillator (7 ns FWHM) and a broadband variable attenuator. The PE was found to typically be in the range of 1-2 %, depending on wavelength, and the bulk PCSS on-state voltage driven by external circuit parameters. Features of the high electric field stress behavior (> 200 kV/cm) of the bulk PCSS were captured with high fidelity in a 1D drift-diffusion model with a self-consistent Poisson solver including trap assisted tunneling, Poole-Frenkel, and barrier lowering with enhanced tunneling effects. In addition, trap to band impact ionization as well as Coulombic and repulsive trapping potentials were included.

Authors: Z. Shaw; W. Feilner; C. Lynn; A. A. Neuber; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012905

Abstract: A two antennae, software controllable phased array was designed and fabricated to study the generation and transmission of short, nanosecond, nonperiodic pulses. This system allows transmitting a train of appropriately scaled and time-shifted bipolar signals to generate received signals with major frequency components adjustable from approximately 600 MHz to 1.5 GHz. The main components in the system include two digital to analog converters (DACs), two data pattern generators (DPG), and two power amplifiers which drive two TEM Horn antennas. The horn antennas are based on a Chebyshev taper design and a Microstrip-type balun is utilized for the transition from the coaxial feed. This approach yielded a reasonably flat frequency response in a wide range from 0.2 to 3 GHz. Thus far, 100 ps synchronization between channels was achieved, and signals of varying shape and amplitude have been received via the shifting and inverting of Gaussian input pulses defined by the user generated data input vectors. This paper presents an experimental evaluation of the hardware used to generate the multichannel array, the ability to steer the signals and generate signals of varied frequency via superposition in free space.

Authors: B. Esser; J. Dickens; J. Mańkowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534197

Abstract: An electrically small, three dimensional metamaterial inspired antenna at MHz frequencies is evaluated for its potential future use as the principle radiating element in a Mobile Ionospheric Heating (MIH) installation. The design consists of a small loop antenna (SLA) which inductively couples to a capacitively loaded loop (CLL) providing a high Q with a natural match to a 50 Ω source. At approximately 2.5 × 3 × 1.2 m in size the antenna is significantly smaller than the element, 21 × 21 × 16 m, used in the High-frequency Active Auroral Research Program (HAARP) array designed for ionospheric modification. With a gain upwards of 7 dBi when mounted above a sufficiently large ground plane, and up to 90 % efficiency, this electrically small antenna, ESA, provides a compact, efficient radiating element as a standalone radiator or as part of an array. It is demonstrated that tuning from approximately 3 to 10 MHz is possible via adjustment of the capacitance in the CLL and the mutual inductance between the SLA and CLL. A full-scale prototype antenna was fabricated and evaluated to obtain experimental radiation efficiencies and patterns at 500 W power levels, which compare favorably to simulations. Scaling for array operation revealed that it is conceivable to generate 3.6 GW ERP power in equatorial zones from a transportable platform based on the ESA. Such power levels suffice effecting ionospheric modification and, for instance, ELF wave generation in the range of 1-70 Hz.

Authors: S. Ramabhotla; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7462516

Abstract: A microgrid integrated with Distributed Energy Resources (DERs), Energy Storage, and Controllable Loads along with critical and non-critical loads is considered. The operation and maintenance cost (O&M) optimization is performed by the Economic Dispatch using the Reduced Gradient Method in the grid connected mode of microgrid. The minimized cost function of the system must be obtained for the optimization of the O&M cost of microgrid while meeting the load demand. The O&M cost includes the operation and maintenance cost of generated energy by each source and also the energy purchased from the utility. For obtaining the minimum cost of the system, the reduced gradient algorithm is implemented. To improve the reliability and to enhance the economic dispatch operation, a diesel generator and a battery energy storage are included in the microgrid. Different scenarios of the energy sources are compared along with the change in wind and battery profiles of microgrid to obtain the minimum O&M cost of the system. Various profiles of battery and wind energy are considered and minimal O&M cost of each profile is obtained and compared. As a result, the optimal cost of the system is obtained by considering the change in wind and battery profiles and hence provides the optimal solution while meeting the critical and non-critical loads demand.

Authors: V. Meyers; D. Mauch; J. Mańkowski; J. Dickens; R. Joshi; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534405

Abstract: Nonlinearity of optical absorption in semi-insulating bulk 4H-SiC has been investigated. Of interest was the optical bleaching behavior of 4H-SiC at and just above the band edge in the range 3.11-3.33 eV (wavelength 380-355 nm). Results of experiments on 200 μm and 490 μm thickness samples indicate partial bleaching in the optical fluence range from 70 W/cm2 to 1.8 kW/cm2, and the absorption coefficient was found to vary by approximately 10% within this range. These experimental findings are supported by simulation results obtained from a first order semi-empirical rate based model linking excitation-induced change in density of states with the absorption coefficient over the range of tested power densities. As expected, this effect scales with photon energy. Characterization of 4H-SiC absorption behavior under varying fluence will aid in design optimization of a Photoconductive Semiconductor Switch (PCSS).

Authors: P. M. Kelly; J. Dickens; A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534046

Abstract: This study describes a three-dimensional particle-in-cell (PIC) simulation for a reflex-triode virtual cathode oscillator (vircator). In particular, these efforts have focused upon the development of a robust, consistent model using the ICEPIC (Improved Concurrent Electromagnetic Particle-In-Cell) code from the Air Force Research Laboratory (AFRL). The vircator operates at background pressures less than 10^-9 Torr and utilizes a bimodal carbon fiber cathode and pyrolytic graphite anode. Experimental data from the hard-tube vircator at Texas Tech University (TTU) is used to validate simulation results. A working, three-dimensional model of a reflex-triode vircator allows for better understanding of the physical processes responsible for microwave generation and thus enables the development of a more efficient and more customizable system. Simulation results detail the virtual cathode formation and the subsequent extraction of radiated microwave energy. Rather than relying on a non-directional isotropic radiation pattern for the radiated power, the true effective radiated power (ERP) from a three-dimensional, frequency-dependent radiation pattern has been extracted from the ICEPIC model. Furthermore, contributions from higher-order modes, particularly in the upper C-band regime, lead to frequency hopping and decreased microwave output power. Simulated results aid in identifying mode contributions and developing schemes to minimize contributions from undesirable modes. ICEPIC results are presented and compared against experimental results at several different operating conditions.

Authors: Nicholas A. Wilson; Daniel L. Mauch; James C. Dickens; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7534040

Abstract: Summary form only given. The electrical and optical characteristics of a high power UV LED (365 nm wavelength) were evaluated under pulsed operating conditions for pulsed power applications. Measurements were made over varying pulse width (30 ns-100 μs), current (0 A-250 A), repetition rate (single shot -1 MHz), and temperature (23° C-80° C). Diagnostics used included a calibrated photodiode operating in the linear regime for transient optical power measurements, a grating / high speed ICCD based spectrograph for transient spectral analysis, and multiple 10:1 standard oscilloscope probes configured differentially for electrical measurements. A red shift was observed in the output spectrum of the LED with increasing temperature and increasing pulse-width. LED forward voltage was observed to increase linearly with increasing current (≈ 3.5 V-5.2 V) and decrease with increasing pulse-width. The peak optical power observed was > 13 W and a maximum efficiency of 22 % was observed. The evaluated LED and auxiliary hardware were successfully used as the optical trigger source for a SiC photoconductive semiconductor switch (PCSS) under high impedance conditions.

Authors: A. Ogunniyi; H. O'Brien; M. Hinojosa; J. Schrock; S. Lacouture; E. Hirsch; S. Bayne; Sei-Hyung Ryu

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7578933

Abstract: Future Army pulsed power applications semiconductor devices that will meet requirements for high-power, low weight and volume, and fast switching speed. The following paper presents the pulsed power evaluation of high voltage silicon carbide (SiC) super gate turn-off (SGTO) thyristors. These devices are well suited for high voltage, high temperature pulsed power and continuous power electronic systems. A pulse-forming network (PFN) circuit and a low inductance, series resistor-capacitor (LRC) circuit were developed to evaluate both the fast dI/dt capability and the pulse safe operating area (SOA) of the SiC SGTO. Transient simulations of the high voltage SiC SGTOs were also performed on a narrow pulse LRC circuit to investigate the device's switching behavior under extreme pulsed conditions.

Authors: C. Lynn; D. Barnett; K. Rainwater; A. Neuber; J. Dickens; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534307

Abstract: A stand-alone vircator based high power microwave system has been designed as an effects test source. The goal of this system is to cover the frequencies from 4.0 GHz to 6.0 GHz. To date a working prototype has been manufactured and field tested. The current system is capable of producing microwaves at frequencies of 4.06, 4.27, 4.45, 5.83, 5.95 and 6.06 GHz at field levels in excess of 100 kV/m, measured at 3 meters. Additionally, the source and all subsystems are capable of burst mode operation for a duration of 2 s at 500 Hz pulse repetition frequency (PRF). The wide range of tunability was achieved by varying the A-K gap as well as the location of the A-K gap inside the cavity. The A-K gap is varied by a bellows sealed linear actuator which is attached to the cathode. The position of the A-K gap within the cavity is altered by moving a liner and microwave reflector (which form a cavity) within the main vacuum tube. However, the experimentally observed frequencies obtained with the current system leaves a gap from 4.45 to 5.6 GHz. In order for the test system to produce microwaves within this frequency range, a new tube is under development. The cavity of the new tube has a smaller cavity which should push the resonant frequencies into the desired range (from 4.45 GHz to 5.6 GHz). This presentation discusses the experimental results obtained with the new tube and compares that with the previously built and tested source.

Authors: S. Ramabhotla; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7462530

Abstract: In a microgrid, PTC Wind Solutions is used toimplement the optimization of reliability with the help of FaultTree Analysis (FTA). The reliability of each energy source alongwith the non-critical load is calculated. To interpret the faulttree results, the quantitative and qualitative analysis arecalculated. Then the importance measures like RiskAchievement Worth, Risk Reduction Worth, CriticalityImportance and Fussel -- Vesely Importance are used tocalculate the sensitivity and uncertainty of fault tree results. Thecomponents which are sensitive and at high risk are calculatedfrom the results. Using the logic gates in the PTC WindchillSolutions, the entire fault tree for a non-critical load outage isbuilt and studied. From the results, the unreliability andunavailability of the fault tree are used to calculate thereliability and availability of non-critical load outage. From theFault Tree Analysis, the unavailability and unreliability of noncriticalload outage are calculated which illustrates the values ofavailability and reliability. From the fault tree analysis, theunavailability and unreliability of non-critical load outage arecalculated as 0.01228, which illustrates that the availability andreliability as 98.77%. Minimal cut sets of circuit breaker acrossthe non-critical load, Point of Common Coupling at the maingrid, and a transformer are calculated from the QualitativeAnalysis. The top event probability evaluation of a non -- criticalload is performed using the Quantitative analysis whichindicates the system failure probability. The calculation ofImportance measures -- Risk Achievement Worth, RiskReduction Worth, Criticality Importance, Fussel -- VeselyImportance is performed. Thus, the reliability and availabilityof non-critical load is obtained using the PTC WindchillSolutions. The top event occurrence is caused by the basic andintermediate events of a fault trees. The components at high riskare calculated using the importance measures. Therefore, fromthe Qualitative and Quantitative analysis the components whichare at high risk and sensitive are obtained and maintained wellto optimize the reliability.

Authors: B. N. Pushpakaran; S. B. Bayne; A. A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012789

Abstract: A 2D model of a 1200 V normally-ON 4H-SİC Trenched and Implanted Vertical Junction Field Effect Transistor (TIV-JFET) cell structure was designed and simulated using Silvaco ATLAS TCAD software to investigate and understand the effects of extremely high current density pulsed switching on the device characteristics. The JFET cell was designed for an active area of 2 μm2 and a threshold voltage of -7 V. Physics-based models were included to account for impact ionization, recombination effects, band gap narrowing, mobility and lattice heating. The electro-thermal simulation was performed using a resistive switching circuit at an ambient lattice temperature of 300 K. The circuit was designed for an ON-state drain current density of 5000 A/cm2. The device was simulated using a 100 kHz 50% duty cycle gate signal consisting of four switching cycles considering the simulation duration bottleneck. The analysis of lattice temperature profile revealed the formation of thermal hot spot in the channel area close to the gate P+ regions in the JFET structure. Further analysis showed an increase in the minority carrier concentration in the vicinity of the gate implants which affected the switching characteristics of the JFET at extremely high current density.

Authors: A. Ogunniyi; J. Schröck; M. Hinojosa; H. O'Brien; A. Lelis; S. Bayne; S. Ryu

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7921334

Abstract: The silicon carbide (SiC) "Super" gate turn-off thyristor (SGTO) is a viable device for high voltage and fast dI/dt switching applications. These devices are well suited for various pulsed power applications requiring high peak currents in the kilo-amp regime. The turn-on transition speed is determined by the spreading velocity, which depends on applied gate current, applied anode current density, minority carrier lifetime, and both the gate base-width and the drift region of the thyristor. The impact of device parameters on switching performance is discussed in this work.

Authors: J. A. Schrock; E. A. Hirsch; A. Bilbao; S. Lacouture; W. Ray; S. Bayne; M. Giesselmann; A. Ogunniyi; H. O'Brien

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012878

Abstract: Silicon carbide Super Gate Turn-Off (SGTO) thyristors are an advanced technology for increasing the power density of high voltage pulsed power or power electronic systems. However, the transient characteristics and failure modes of these devices have to be further understood. This paper presents the Atlas TCAD simulation of a 15 kV SiC SGTO thyristor during extreme pulsed overcurrent conditions. The simulated device is first validated against dc measurements of a physical device. The device is then simulated at various pulse current amplitudes using a 10 stage 100 μs PFN. In addition, a tradeoff study for the drift region and anode mesa width is performed.

Authors: A. A. Neuber; D. L. Mauch; V. E. Meyers; B. Esser; R. P. Joshi; J. C. Dickens; J. J. Mankowski; T. M. Antonsen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7534401

Abstract: A transportable ionospheric heater, TIH, research design is presented that will enable plasma studies of the ionosphere in latitudes that are presently inaccessible by fixed installations such as HAARP (High Frequency Active Auroral Research Program). The equatorial latitude with close to zero vertical magnetic fields is especially of interest for basic plasma physics studies as well as rf communication enhancement. To achieve a power level in the ionosphere of at least 70 dBW ERP in a footprint significantly smaller than HAARP the radiated power needs to be substantially increased. This minimum ERP is achievable in a 4 × 4 antenna array with 370 kW input power per element with about 25 m by 25 m footprint vs. HAARP's equivalent 365 m by 365 m (360 antenna elements total, 10 kW maximum per antenna). Maximum ERP, up to 95 dBW, may be achieved with the TIH on a 115 m by 70 m platform, a factor 17 reduced size from HAARP. Tunable, Electrically Small Antennas, ESAs are employed to overcome the maximum power limitations of the HAARP dipole based antennas. This demands a step-up from 10 kW to several 100 kW cw power in the 3 to 10 MHz band, which is required to effectively heat the ionosphere. Driving the ESAs necessitates a tunable rf source in the same power and frequency regime, where a more traditional rf tube or all solid state approach may be pursued. The focus of the driver related research has been on photoconductive solid state switching, PCSS, in a direct drive mode that incorporates the driver into the antenna itself. A full size ESA operating at 9.5 to 10 MHz has been demonstrated at 500 W cw power levels and ~ 90% efficiency, driven by a single SiC switch mimicking the full power PCSS operation. The challenges and physics limitations of scaling the switch, the tunable ESA antenna design, as well as their coupling are presented. The significant progress made towards a transportable ionospheric heater as it relates to the physics of the PCSS switching efficiency, electrical breakdown in the MHz regime in large gaps, lower power experiments, and numerical simulations is discussed.

Authors: W. Feilner; Z. Shaw; C. Lynn; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8012903

Abstract: Arbitrary transient electric field shapes are generated in free space utilizing a set of transient signals with proper shape, amplitude, and time shift. Akin to wavelets in signal processing, brief, non-periodic oscillations are superimposed at a pre-selected location in space to effect destructive and constructive interference. With a properly chosen signal set, an entirely different frequency or shape is generated. Two methods have been employed to find optimum signal sets, the Discrete Wavelet Transform (DWT) and Particle Swarm Optimization (PSO). While the DWT approach dictates constant time step and rectangular matching between wavelets, PSO is not restricted in this manner, allowing for more flexibility in choosing amplitudes and signal delays.

Authors: J. Johnson; D. Reale; D. Barnett; R. Garcia; W. Cravey; J. Parson; A. Neuber; J. Dickens; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296894

Abstract: This paper details the design, fabrication and performance of a coaxial ferrimagnetic nonlinear transmission line (NLTL), four element array, high power microwave (HPM) system operated at a 1 kHz repetition rate. Prime power is delivered from an 802L TDK Lambda power supply which charges a 5.2 nF capacitor bank up to -40 kV. The capacitors are discharged through a center pin trigatron spark gap. The trigger generator is optically isolated and battery powered for noise immunity and portability. It produces a 20 kV positive polarity pulse with a 20 ns risetime. The high dV/dt (1 kV/ns) is desirable to reduce jitter inherent to spark-gap switching. After the spark-gap switch has closed, the pulse is split four ways. The four pulses propagate through four adjustable delay lines for synchronization of the individual outputs. The four delay lines connect directly into four 76 cm NLTLs with NiZn ferrites where SF6 is the insulating dielectric. Each NLTL is terminated into a custom fabricated, Rexolite-filled, TEM horn antenna via a zipper balun. Lastly, a LabVIEW based control system automates the whole system using a National Instruments cRIO controller. Experimental observations will include in-line D-dot measurements of voltage waveforms and radiated D-dot field measurements.

Authors: J. Shaver; D. Mauch; R. Joshi; J. Mankowski; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296867

Abstract: 4H-SiC Photoconductive Semiconductor Switches (PCSSs) have shown significant promise for use in pulsed power related switch applications. This simulation uses the finite difference method, parallelized using a NVIDIA graphical processing unit and the CUDA framework, to solve the system of partial differential equations that model the semiconductor physics involved in the high voltage blocking state of the photoconductive switch. By taking into consideration material properties (mid-band gap trap energy level and concentration, etc.), we are able to gain an understanding of how changes in these parameters affect the space-charge-limited (SCL) currents observed in the high voltage blocking state. This subsequently allows for a fundamental understanding of the parameters controlling the high voltage switching capability of photoconductive switches. Results of the simulation are presented.

Authors: D. H. Barnett; J. M. Parson; C. F. Lynn; P. M. Kelly; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296790

Abstract: The paper presents the design and operational characteristics of an optically isolated, compact, modular pulse generator for spark-gap triggering applications. The pulse trigger generator is capable of operating at pulse repetition frequencies (PRFs) > 1 kHz for short bursts with variable pulse magnitudes and risetimes. The trigger generator utilizes a transformer, magnetic switch and an IGBT primary switch. It has been successfully used to trigger a trigatron-driven 10-stage, Marx generator driving a high power load. For portability and noise immunity, the trigger generator is optically isolated from its low voltage control and powered via a lithium ion polymer battery pack. A constant current dc-dc power supply charges the high voltage circuitry of the trigger generator and enables continuous operation with two modules. For operation, a large capacitor is initially charged and used as a buffer energy source. The intermediate charge storage in conjunction with a command-triggered MOSFET, provides the ability to quickly re-charge a trigger capacitor between each pulse of the burst. Circuit topology, experimental data, including voltage and current waveforms, and jitter of the overall system are discussed at various PRFs.

Authors: Shu Lin; Sterling Beeson; Yongdong Li; Chunliang Liu; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7179681

Abstract: Summary form only given. The mechanism of plasma formation under high power pulse (HPM) excitation in gases (nitrogen and argon) at atmospheric pressures is studied utilizing a fluid model verified against experimental data. A 2-D approximation was introduced to model the cylindrically shaped gas volume and the associated electric fields. Thus, the fluid equations and Poisson equation for space charge effects are updated in x and y directions only (setting the gradient in the z-direction to 0). The goal of this numerical simulation was to accurately predict the plasma formation delay time under different gas types and pressures based on the calculated evolution of the plasma conductivity.An S-band TE111 resonator with a built-in quartz gas isolation tube in the center was fabricated1. A 4 MW, 4 μs pulse in the dominant TE10 mode provided by a 2.85 GHz magnetron propagates through it. In the fluid model, the effect of HPM on plasma formation is modelled with particle heating as well as elastic and inelastic collisions driven by the incident electric field in the vertical (normal) coordinate. The amplitude distribution of the HPM electric field is obtained from numerical simulation of the resonator using commercial EM software. Since the focus is on the onset of plasma formation, simulation is stopped before a large plasma density develops up to the moment when the transmitted power is reduced by 10% (-0.5 dB), which corresponds to an average plasma conductivity of 0.02 S/m. Both measurement and simulation cover a gas pressure range from 25 to 700 torr with delay times for N2 and Ar from 22 to 204 nanoseconds. The development of the particle densities and temperatures during the simulation is presented to reveal the dominant mechanism of plasma formation in atmospheric gases. The delay times for different gas types and pressures are in good agreement for low gas pressures (<; 200 torr). Deviations between the model and experiments at higher pressure are found to be primarily due to the onset of plasma filamentation.

Authors: M. He; V. Vittal; J. Zhang

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7285972

Abstract: Short-term wind farm power forecasting is studied by exploiting the spatio-temporal correlation between individual turbine's power output. A multivariate time series model for wind farm power generation is developed by using vector autoregression (VAR). In order to avoid the possible over-fitting issues caused by a large number of autoregressive coefficients and the impact on the forecasting performance of VAR models, a sparsified autoregressive coefficient matrix is constructed by utilizing the information on wind direction, wind speed and wind farm's layout. Then, the VAR model parameters are obtained through maximum likelihood estimation of real-time measurement data, by taking into account the sparse structure of the autoregressive coefficient matrix. The proposed approach is compared with univariate autoregressive models through numerical experiments, resulting in significant improvement, which is attributed to the turbine-level correlation captured by the developed VAR model.

Authors: K. Eldridge-Looker; A. Fierro; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296929

Abstract: Large antenna sizes present a severe limitation in the generation of low frequency signals. To alleviate this issue, a series of high frequency sources (and thus, smaller antenna sizes) may be combined to produce a lower frequency output in the far-field. This reconstruction technique applies to signals of arbitrary shapes in addition to frequency lowering. Due to the large parameter space, wavelet decomposition theory is coupled with particle swarm optimization to define appropriate time shifts and amplitude adjustments to the high frequency sources (wavelets) in order to synthesize an output signal with the desired frequency or shape. Fifteen individual high frequency wavelets (f = 1 GHz) are utilized in this application to produce a final output signal in the far-field of 600 MHz. The radiation of a single wavelet is simulated using a broadband Chebyshev TEM horn antenna in order to perform far-field constructive and destructive interference analysis of all 15 wavelet signals. Simulation of the TEM horn antenna shows reasonable wavelet signal fidelity with minimal reflections from the antenna aperture as the pulse is propagated to the far-field. Examination of the far-field electric field enables accurate temporal depiction of the reconstructed signal from the 15 pulsed sources at any given point in space. By adjusting the source array geometry, the location of the desired signal (determined by the particle swarm optimization) can be narrowed down to a single location. Overall, combining particle swarm optimization, wavelet decomposition theory, and electromagnetic wave propagation enables the accurate reconstruction of far-field temporal electric fields from the combination of 15 wavelet sources and verification of the desired signal location.

Authors: Esser, B; Beeson, S; Dickens, J; Mankowski, J; Neuber, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7296897

Abstract: A tunable, metamaterial-inspired, electrically small antenna topology is evaluated for a possible future use as the principle radiating element in a mobile Ionospheric Heating (MIH) system. The RF source signal is fed via a 50 Omega coaxial cable into a small semi-loop antenna (SLA). This inductively couples to a capacitively loaded loop (CLL) providing a natural 50 Omega match to the source. The resonant frequency of the antenna can be adjusted by varying the capacitance of the CLL via inserting a large permittivity dielectric. A simplified circuit model is used to show that the resonant frequency can be tuned between 40 - 100 MHz. Also, the maximum power handling capabilities achievable with this antenna topology at frequencies relevant to ionospheric heating (similar to 10 MHz and below) are estimated.

Authors: A. A. Ogunniyi; H. K. O'Brien; M. Hinojosa; L. Cheng; C. J. Scozzie; B. N. Pushpakaran; S. Lacouture; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296946

Abstract: Future Army power systems will require utilizing high-power and high-voltage SiC devices in order to meet size, weight, volume, and high power density for fast switching requirements at both component and system levels. This paper presents the modeling and simulation of a high voltage (>12kV) silicon carbide PiN diode for high action pulsed power applications. A model of a high power PiN diode was developed in the Silvaco Atlas software to better understand the extreme electrical stresses in the power diode when subjected to a high-current pulse. The impact of carrier lifetime on pulsed switching performance of silicon carbide (SiC) PiN diode was investigated.

Authors: W. B. Ray; J. A. Schrock; A. V. Bilbao; M. Kelley; S. Lacouture; E. Hirsch; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7297037

Abstract: The advancement of wide bandgap semiconductor materials has led to the development of Gallium Nitride (GaN) power semiconductor devices, specifically GaN Power MOSFETs. GaN devices have improved characteristics in carrier mobility and on-state resistance compared to Silicon solid state switches. With the development of these new power semiconductor devices a need was established to understand the behavior of the devices switching performance under stress, with regards to situations in pulsing circuits. Through the examination of the switching characteristics of GaN devices, the results can be used for the improvement of advanced pulsing circuit design with GaN solid state switches. In this paper the authors develop a test bed to expose the GaN Power MOSFETs to single and repetitive pulsed overcurrents. The test bed was developed using a Pulse Ring Down board in a radially symmetric configuration to minimize the total equivalent inductance and resistance. The test bed switches the GaN MOSFET with low impedance between the DC bus and ground to induce the stress the MOSFET experiences during pulsed overcurrents. The DC characteristics were measured between switching sets to reveal characteristic signs of potential degradation and failure modes due to pulsed overcurrents. The single and repetitive pulse switching characteristics are captured, analyzed, and shown.

Authors: A. V. Bilbao; J. A. Schrock; W. B. Ray; M. D. Kelley; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296953

Abstract: The power density of pulsed power systems can be increased with the utilization of silicon carbide power devices1. With the latest developments in manufacturing techniques, the fabrication of insulated gate bipolar transistor (IGBT) devices with blocking voltages as high as 20 kV are now possible2. A complete practical understanding of ultra-high voltage silicon carbide device switching parameters is not yet known. The purpose of this research is to show switching parameters extracted from inductive and resistive switching tests performed on state of the art 20 kV silicon carbide IGBTs. Resistive switching tests were used to extract device rise time, fall time, turn-on delay, turn-off delay and conduction losses. Double pulsed inductive switching tests were used to extract turn-on and turn-off switching energies and peak power dissipation. The data was obtained at case temperatures from 25 C to 150 C.

Authors: A. Majzoobi; R. P. Joshi; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7297005

Abstract: Intense electron emission from cathodes that provide very high current densities (several kA/cm2) is necessary for a various pulsed power applications. This contribution presents a quantitative analyses of the following processes and inherent physics: (a) Local field enhancements at micro-protrusions, (b) role of ion/ions near the emitting surface in lowering and thinning the potential barrier which increases emission. (c) localized heating at cathode tips that could produce hot-electrons and hot-phonons, ultimately leading to localized melting. Temperatures are predicted to possibly reach the cathode melting point on the nanosecond time scales. This is in keeping with the explosive emission phenomenon that is well known.

Authors: A. S. Subburaj; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7356938

Abstract: The paper provides an analysis of the performance of the battery (1MW/1MWh) and wind system (1.7MW) when connected to weak or strong grid. Understanding the performance of the battery and analysis of the battery system tied to the distribution grid with heavy penetration of wind as a function of the strength of the grid is discussed with the aim to understand the optimized energy flow between the battery, the wind farm, and the utility grid, energy management for storage, battery performance, energy dispatch and reliability. The advanced battery model utilizes the dual polarization electrical equivalent model for the analysis. The grid-connected battery and wind system simulation results are obtained using PSCAD software.

Authors: S. L. Holt; C. F. Lynn; J. M. Parson; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7297029

Abstract: Fast capacitor charging is a power electronics application with unique challenges requiring both low voltage, high current operation and high voltage, low current operation from the same system at different points in the charge cycle. Such dynamically differing operating points create distinct challenges in obtaining high efficiency throughout a charge cycle. In addition, the power supply must be protected from negative voltage swings when attached to a high rep-rate power modulator. This paper presents the design and testing of a rapid capacitor charger designed for high rep-rate command charging of a Marx generator. The output stage uses a large inductor to delay discontinuous conduction during the early portion of the charge cycle and greatly improve efficiency. The design goals and tradeoffs will be discussed and simulation results will be compared to experimental data.

Authors: Meyers, V; Mauch, D; Mankowski, J; Dickens, J; Neuber, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7296862

Abstract: The optical properties of bulk semi-insulating GaN:Fe are obtained to assess its future suitability as a high voltage photoconductive semiconductor switch (PCSS). The material properties of GaN: Fe hold significant promise to improve devices for pulsed power and other applications. Growth techniques of bulk GaN: Fe, which have hitherto been largely insufficient for commercial applications, are nearing the point that anticipatory characterization research is warranted. In this paper, the optical constants of bulk GaN: Fe (refractive index, absorption coefficient, and off-state dielectric function) were determined by optical reflection/transmission analysis. The results of this analysis are compared with a similar treatment of bulk 4H-SiC as well as possible elements of PCSS housing: Sylgard 184 elastomer, and EFI 20003/50013 electrical potting epoxy. The data presented provide foundational material characterization to enable assessment of the feasibility of GaN: Fe as a practical high voltage PCSS material. Beyond basic materials research, these properties inform design optimization in PCSS construction and implementation.

Authors: D. Thomas; D. Mauch; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296950

Abstract: A method of characterizing mid-bandgap defect states in high purity semi-insulating 4H-SiC through leakage current analysis for optimization of SiC photoconductive switches is presented. The method utilizes two custom IV curve tracer systems to measure leakage currents through the material under various voltage/current conditions. The first system is used under low current conditions and is capable of measurements from 0 to 45 kV at currents ranging from 0 to 3 mA with pA resolution. A second system measures the transient discharge of a charged capacitor bank through the material. Due to power dissipation concerns, the second system is used for currents higher than 0.1 mA. Voltage/current measurements in this region (>0.1 mA) are of interest due to the information concerning defect states near the conduction band. These shallow defect states are detrimental to switching performance while offering little benefit to voltage hold-off. From the combined data of these two systems, characteristics of the defect states are extracted and presented. We further elucidate the effect of contact annealing temperature on shallow trap levels in electron-beam irradiated material (2*1018 1/cm2).

Authors: Rocha, E; Parson, JM; Lynn, C; Dickens, JC; Neuber, A; Mankowski, J; Queller, T; Gleizer, JZ; Krasik, YE

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7179763

Abstract: This study focuses on the performance of a bimodal carbon fiber (CF) cathode and a carbon-epoxy multicapillary (CEM) cathode with microwaves at 50kV/m at 2m lasting 100 ns in a compact reflex triode virtual cathode oscillator (vircator). It was previously revealed that the CEM cathode was able to produce uniform emission distribution for a long duration1 . To further uncover the location(s) where plasma is forming, an intensified CCD camera is used to image both cathodes with high spatial and temporal resolution. A titanium grade 1 (TiG1) anode that is 70% transparent is used to minimize outgassing in the system resulting in the cathode being the primary outgassing constituent2 . One data set for each cathode was taken. Each data set contains diode current, voltage, and microwave fields over the course of 10,000 shots. A performance baseline is shown by comparing the evolution of these data over the course of the experiments. An 8 stage, 168 J pulse forming network (PFN) based Marx generator serves to drive both cathodes at 250 kV, 4 kA with ~175 ns pulsewidth, achieving current densities of ~200 A/cm3 . The operating frequency of interest is set in the range of 1-2 GHz, where the tunability is achieved by varying the size of the anode-cathode (A-K) gap. In all experiments the gap is fired at a 10 Hz pulse repetition rate. The characteristics of one type of cathode over the other, depending on the operating conditions, are discussed in detail.

Authors: A. D. Udai; R. P. Joshi; S. K. Saha

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7353871

Abstract: Peg-in-tube assembly stands ahead of a more common benchmark task for industrial assembly, i.e., `pegin-hole'. The robot can easily be deceived to detect the actual hole while performing a `peg-in-tube' task as the tube has a surrounding pocket that cannot support the peg. The paper presents a thorough geometrical analysis of the `peg-in-tube' assembly process, and proposes a novel algorithm based on depth measurements of peg center to perform `peg-in-tube' task. The results are demonstrated on a KUKA KR5 Arc industrial robot with a chamferless cylindrical peg and a tube having a minimum clearance of 0.10 mm.

Authors: J. A. Schrock; W. B. Ray; A. V. Bilbao; M. D. Kelley; E. A. Hirsch; S. L. Holt; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7296921

Abstract: Silicon carbide (4H-SiC) is a leading option for increasing the power density of pulsed power and power electronic systems1, 2. SiC devices used in high voltage switching applications experience high dV/dt due to fast switching transients. Under high dV/dt conditions the devices can exhibit spurious turn-ON. For SiC devices to achieve widespread acceptance the dV/dt limit must be established. To measure the dV/dt limit, a circuit comprised of four silicon avalanche BJTs operating in secondary breakdown was constructed. This circuit is capable of generating dV/dts well in excess of what SiC unipolar and bipolar devices might be exposed to in typical applications. Two SiC diodes in an “OR” configuration are used to perform a comprehensive dV/dt analysis as a function of dc bias. Using this experimental setup dV/dts up to 200 V/ns were applied to SiC MOSFETs, and the induced gate to source voltage was measured. Preliminary dV/dt results achieved with the secondary breakdown circuit are shown for a range of dc biases.

Authors: T. Flack; C. Hettler; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296985

Abstract: Pulsed power systems which utilize solid state switching devices, rather than spark/gas-type devices, could potentially exhibit increased energy density, repetition rate, operational lifetime, and ruggedness. However, further evaluation of advanced solid state devices, such as thyristor type devices, is required to better understand their operation for pulsed power applications. This paper details experimental evaluation of the dI/dt capabilities of a silicon (Si) n-type, asymmetric-blocking gate turn-off thyristor (GTO) manufactured by Silicon Power. The device under test (DUT) is rated to block up to 4 kV with rated dI/dt of 30 kA/μs. The DUT was designed as a solid state replacement for spark/gas-type switching devices in pulsed power applications. A low inductance test circuit was designed and built to evaluate the dI/dt capabilities of the DUT. Specific care was taken to minimize the parasitic inductance and thereby exploit the achievable dI/dt ratings by the test devices. An external fiber-driven gate driver is used to trigger the device with gate current (I_G) of approximately 1.3 A and rate of current change (dI_G/dt) of approximately 25.5 A/μs. Experimental dI/dt values greatly exceeded rated values; specifically dI/dt values of approximately 77 kA/μs were readily achieved at 4 kV. No device degradation was observed over the course of evaluation.

Authors: W. B. Ray; A. S. Subburaj; J. A. Schrock; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7356781

Abstract: Grid-connected battery energy storage systems (BESS) are essential for improving the transient dynamics of the power grid. There is ongoing research about how BESS integration with renewable energy sources can improve renewable energy deployment in the grid. However, the economic feasibility of BESS is a practical limitation of their integration into power systems. BESS costs include both the start-up costs of building the system and the operating costs. Optimizing the operation of the BESS to maximize operating profit would make the BESS more economically feasible to power system operators, and lead to smoother integration of BESS. This manuscript overviews a program tool that analyses grid connected BESS in real world situations and optimizes the operation of the battery system. Through the use of this tool, a better understanding of the economic feasibility of BESS is achieved.

Authors: S. Abedi; M. He; S. M. Fatemi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7131779

Abstract: Penetration of Plug-in Electric Vehicles (PEVs) parking lots into distribution systems and microgrids offers useful decision tools for improving energy management and operations of power systems. In this paper, with an objective to minimize the operation cost while subject to the system, unit and security constraints, a microgrid energy management scheme is proposed, by taking into account PEVs throughout the grid as distributed storage responsive to the nodal prices. The presented method comprises the ex-ante optimal dispatch to minimize the anticipated operation cost, followed by a near real-time dispatch to manage the violation of security constraints regarding any congestion and voltage limit in the microgrid. The second dispatch, as recourse to the ex-ante dispatch, aims to minimize the operation cost and the deviation from the ex-ante dispatch decisions, while any possible binding constraint occurred in the ex-ante dispatch is mitigated. Simulation results on a test microgrid system demonstrate that the presented scheme can effectively ensure secure and economic operations of the microgrid as well as the risk reduction of the ex-ante decisions.

Authors: N. R. Joseph; M. E. Savage; J. C. Stephens; J. A. Lott; B. A. Lewis; R. D. Thomas; M. A. Torres; E. G. Holman

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296969

Abstract: The Sandia National Laboratories SPHINX accelerator is used to study the response of electronics to pulsed x-ray and electron environments. The system consists of a Marx generator and an oil-insulated pulse-forming line with self-closing oil switches. SPHINX has a peak load voltage of 2 MV and an adjustable pulse width ranging from 3 to 10 ns. The previous pulsed-power system had reliability and triggering issues with the Marx generator and subsequent undesired variations in voltage output. SPHINX was upgraded to a new Marx-generator system that has solved many of the voltage-output fluctuation and timing issues. The new Marx generator uses recently developed low-inductance 100-kV capacitors and 200-kV spark-gap switches. This paper provides an overview of SPHINX while capturing in detail the design, characterization, and comparative performance of the new Marx generator.

Authors: M. D. Kelley; A. V. Bilbao; W. B. Ray; J. A. Schrock; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296939

Abstract: Silicon Carbide (4H-SiC) is a state-of-the-art solution for increasing the energy density of pulsed power and power electronics. High power SiC MOSFET modules have only recently become commercially available; for widespread acceptance further device characterization and reliability testing is necessary. The purpose of this work is to establish and compare device characteristics for two SiC power modules. Of the two modules tested, one contained Cree die and the other Rohm die. The device characteristics presented for the two modules are switching losses (EON & EOFF) and on-state resistance (RDS(ON)). EON, EOFF, and RDS(ON) were measured at 25°C and 125°C. The RDS(ON) of the two modules was determined to be approximately equal; however, the SiC module containing the Cree die yielded significantly lower turn-on and turn-off switching losses. The measurements presented in this work demonstrate SiC power modules are a leading solution for high energy density applications.

Authors: D. Mauch; J. Dickens; V. Kuryatkov; V. Meyers; R. Ness; S. Nikishin; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296989

Abstract: Semi-insulating Gallium Nitride is evaluated as a candidate material for use as a high voltage photoconductive semiconductor switch (PCSS) for pulsed power applications. The GaN:Fe samples used for this investigation were commercially available, bulk, semi-insulating samples measuring 10 mm × 10 mm × 475 μm. Their optical and crystallographic properties were determined utilizing cathodoluminesence, photoluminescence, RHEED, as well as microwave reflection techniques for carrier lifetime studies. Experimental results are presented elucidating the potential of GaN:Fe sustaining high potential differences in both lateral and vertical geometry devices. For instance, electric field hold-off exceeding 100 kV/cm was observed in lateral geometry with mm sized gaps. In addition, a process for the homo-epitaxial growth of GaN:Si was developed in order to facilitate the fabrication of high quality ohmic contacts. Lastly, experimental results evaluating the on-state performance and photo-current efficiency of a GaN:Fe based PCSS are presented.

Authors: Lacouture, S; Schrock, JA; Ray, WB; Hirsch, EA; Bayne, S; Giesselmann, M; O'Brien, H; Ogunniyi, A; Scozzie, C

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7296932

Abstract: While Silicon Carbide (SiC) based power switching elements are starting to appear that are able to perform better than their Si counterparts in terms of voltage hold off, current density and operating temperature, the material is still relatively new in the semiconductor arena, and although new device designs are simulated extensively before being committed to fabrication, there is often a large discrepancy between actual device performance and simulated results. Manufacturers certainly carry out some electrical testing of these quasi experimental components, but there is a dearth of information pertaining to Safe Operating Area (SOA) and device longevity. Texas Tech University's Center for Pulsed Power and Power Electronics, in cooperation with Army Research Lab, has carried out extensive long term, high - energy testing of SiC Super Gate Turn Off Thyristors (SGTOs) produced by Cree Inc. To conduct this extremely high volume testing at high energy levels, an automated test bed was designed that pulses the devices for an arbitrary number of cycles and alternately switches the device to a low energy characterization system, with all waveforms and current - voltage characteristics recorded. Approximately 350,000 high energy cycles on various SGTOs have been recorded. From this large database of results, actual SOA at high cycle count (>> 10,000 pulses) has been extracted for the devices. With each cycle's waveforms recorded, and the devices' characteristics traced at chosen intervals, several distinct changes in these parameters have been found to inevitably herald the imminent failure of a device. The most common change is in the gate - anode junction, where curve traces show a leaking, almost resistive behavior immediately before the junction becomes forward biased. As the system is completely automated, and limits can be set to halt a test sequence upon being broached, several devices have been brought to the brink of failure - an event that is usually catastrophic, physically destroying the device - to be examined by the manufacturer.

Authors: B. N. Pushpakaran; S. B. Bayne; G. Wang; J. Mookken

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296918

Abstract: The superior electro-thermal properties of Silicon Carbide (SiC) as compared to silicon make them a viable candidate for high voltage and high frequency applications. Due to the relatively recent surge in commercially available SiC power MOSFETs, there is an immediate demand for accurate simulations models to predict device behavior and aid circuit design process. This paper discusses the development of an accurate SPICE based model for a commercially available 1200V, 20A SiC power MOSFET manufactured by CREE Inc. based on the Enz - Krummenacher - Vittoz (EKV) MOSFET model. The advantage of using EKV model over the simplified quadratic model is the ability to characterize MOSFET behavior over weak, moderate and strong inversion regions with a single equation. The model was developed using parameters extracted through experimental data conducted at wide temperature range. Package parasitic components have been incorporated into the model to predict device behavior in high frequency switching applications. The model was simulated for its static and transient behavior and compared with actual device results to determine accuracy over a wide operating range.

Authors: S. Feathers; A. Fierro; S. Beeson; J. Stephens; J. Dickens; A. Neuber; K. M. Williamson

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7799673

Abstract: A 3 microsecond risetime, 300 μs fall time, 25 kV high voltage pulse, chosen to emulate lightning discharges is used to investigate breakdown in close proximity to a high permittivity rutile dielectric (Er ~ 100) cylinder with 1 cm diameter. A needle-plane electrode structure is utilized with 1.05 cm electrode separation. The dielectric location is varied from immediately under the needle, to more than a 1 cm away from the needle axis. Basic system design and simulation are presented, along with simulations of the electric field distribution for various dielectric locations. The dependence of breakdown voltage on dielectric proximity is presented in detail. For instance, a minimum in breakdown voltage is observed when the dielectric cylinder is placed slightly off-center from the needle's center axis. Using high speed imaging, discharge inception and growth are presented with ~10 ns resolution.

Authors: W. H. Cravey; D. V. Reale; R. S. Garcia; J. M. Johnson; A. A. Neuber; J. C. Dickens; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296987

Abstract: This paper discusses the design of a gas flow system for a 1 kHz repetition rate trigatron based spark gap. The system requires a flow rate high enough to clear the gas from the gap in under 1 ms while also maintaining the required operating pressure. The gap is a trigatron based design with an alumina insulated tungsten trigger pin. A 20 kV positive polarity pulse, with a 20 ns risetime, is applied to trigger the main gap which is charged to −40 kV. A brass guard ring is employed to shield the walls of the containment structure from contaminants in order to achieve a longer operational lifetime. Dry air and Nitrogen gasses are tested at various pressures and flow rates. Design considerations, such as inlet and outlet sizing and gap geometry, are taken into account to ensure a high air flow is achieved between the electrodes.

Authors: E. Cordero; S. Holt; J. Dickens; A. Neuber; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296920

Abstract: High power pulsed systems often require high current 3-phase electrical service or large generators to meet prime power requirements, limiting the portability of the systems. A high power battery system offers a useful alternative for mobile applications. Of all battery chemistries, Lithium-ion polymer (LiPo) has become popular in consumer electronics due to its high energy density, low self-discharge rate and lack of memory. Unfortunately, the chemistry still has limitations. Overcharging or overheating of a LiPo cell may result in ignition and over-discharge can destroy the cell. To prevent these hazards, overcharge and undercharge conditions must be monitored at the cell level rather than the battery level because the charge and discharge efficiency varies from cell to cell causing cell voltages within a battery to diverge during normal operation. A battery management system designed to monitor and maintain a large battery designed for pulsed power applications is presented in this poster. The design of this battery management system is presented and its implementation in a multi-cell, high voltage battery capable of high current pulsed operation. A hazard analysis of high power batteries and the implemented safety system is also provided.

Authors: C. F. Lynn; J. M. Parson; P. Kelly; D. H. Barnett; A. A. Neuber; J. C. Dickens; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296907

Abstract: Virtual cathode oscillators (vircators) can be easily tuned by altering the current density of the electron beam. This can be accomplished by changing the voltage applied to the a-k gap, or by physically changing the gap separation. Previous results with an adjustable a-k gap vircator achieved discrete tuning at frequencies of 1.5 GHz, 2.1 GHz, 4.1 GHz, 4.6 GHz, and 5.9 GHz. To achieve better tunability an adjustable cavity reflector was incorporated into the vircator design. Initial testing of the new vircator design has shown increased output power, as well as the ability to greatly increase the number of frequencies that can be tuned. Sweeping the A-K gap with the cavity reflector in two positions yielded 10 different frequencies of operation. Also sweeping the cavity reflector with the a-k gap set at 8.0mm yielded another unique frequency. Currently 11 unique frequencies have been identified, and the peak field measured from this vircator increased from ~40 kV/m to ~69 kV/m (> 70% increase in field). This paper illustrates the design of the vircator and presents some of the microwave data obtained. Additionally, a summary of all of the frequencies obtained and the peak field measured are included.

Authors: He, M; Abedi, S; Boker, A

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8352818

Abstract: Security-constrained unit commitment is studied by taking into account multiple correlated wind sources and the deliverability of operating reserves in a reserve zone. A power transfer margin for each transmission line is incorporated into the unit commitment problem, so as to mitigate possible congestions caused by uncertain and non-dispatchable wind power. Specifically, the power transfer margins of transmission lines are determined by utilizing the probabilistic information of multi-locational wind power generations, and by considering credible line-outage contingencies. The power transfer margins collectively guarantee that congested operating conditions occur with a reduced probability, and thus provide a new approach to the efficient management of intra-zonal congestions. The effectiveness of the proposed approach is verified through case studies on IEEE RTS-96, with comparisons to unit commitment with conventional reserve zone management.

Authors: J. Stephens; D. Mauch; S. Feathers; J. Mankowski; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7179942

Abstract: Summary form only given. Using voltage pulses, 10s-100s of nanoseconds in length, microdischarges (MDs) are driven with up to 1 kV, and current levels from 30-150 A. Time-averaged input power levels from 10 W-1+ kW are achieved depending on the selection of pulse width and pulse repetition rate (1 kHz-1 MHz). Using an argon-hydrogen mixture, intense VUV radiation is generated at 121.6 nm (10.2 eV, Lyman-alpha). With this source, instantaneous power levels in excess of 60 watts have been achieved with several watts time-averaged power in the VUV. Additional work utilizing XeCl* (308 nm) and XeF* (351 nm) excimer sources is also reported. Achieved power levels, instantaneous and time-averaged, efficiency, and impurity content are reported. As an application, the MD-UV source is utilized as an alternative to a high power laser for the triggering of high voltage photoconductive SiC switches.

Authors: J. Stephens; A. Fierro; S. Beeson; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7179992

Abstract: HBSummary form only given. Using a self-matched transmission line pulse generator, high voltage pulses with magnitude 10-25 kV and ~25 ns pulse width are produced to excite low temperature plasma (LTP) formation. In using a short pulse, plasma formation is initiated, but the external excitation is removed before spark formation is completed, thus enabling a study of the LTP during its developmental phase. Using imaging in the visible spectrum and electrical diagnostics, this has been confirmed.To observe the 120 nm - 150 nm range, discharges are guided along the surface of an MgF2 observation window. On the opposing side of the window an evacuated spectrograph with intensified CCD enables the observation of VUV emission. Since the LTP is intentionally limited to its developmental phase, very low total light emission is becomes a diagnostic issue. Hence, it was necessary to accumulate light over several 10s of formation events before discrete lines in the spectra became evident. With a sufficient number of pulses, an experimental spectrum is generated in the 120+ nm range which was found to be in excellent agreement with theoretical atomic spectra of nitrogen and oxygen, with an assumed electron temperature of 1.6 eV. Using a gas puff system combined with an otherwise evacuated spectrograph enables the observation of extreme UV (EUV) to VUV light (80nm - 120 nm). In the photoionization relevant region of the spectra (below 103 nm), the observable emission consists of atomic neutral and singly ionized oxygen as well as molecular nitrogen. Data suggests that additional ground state transitions are present, but unobservable due to self-absorption and low overall light intensity. Most notably, the presence of photoionization capable emission in developing low temperature plasmas in air is verified.

Authors: B. N. Pushpakaran; S. B. Bayne; A. A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296926

Abstract: The unique design of Silicon Carbide (SiC) Junction Barrier Schottky (JBS) diode has proved its superiority over silicon in the field of high energy density pulsed power applications. JBS diode design enables the development of high blocking voltage silicon carbide rectifiers with low ON-state voltage drop, low leakage and negligible reverse recovery. In pulsed power applications, devices get driven above their rated current carrying capacity for a transient duration. Under this scenario, it becomes critical to have a thorough understanding of the electro-thermal behavior of the device under pulsed condition. This research focuses on the design and simulation of a 4H-SiC JBS diode structure in Silvaco ATLAS software under steady state and pulsed conditions. Physics based models were incorporated to account for drift diffusion process, mobility, impact ionization and lattice heating. The JBS diode was designed for a blocking voltage of 3.3 kV and an ON-state current density of 100 A/cm². A schottky barrier height of 1.1 eV was selected for the device. An array of interdigitated P+ regions with optimized separation was designed to shield the schottky interface from the high blocking electric field without affecting the ON state characteristics. The simulation results were used to analyze breakdown electric field distribution, forward current conduction path, switching performance and areas of localized lattice heating. The diode structure was simulated under pulsed condition pertaining to 500 A/cm² current density and the lattice temperature profile was analyzed to identify the formation of thermal hot spots in the device lattice and possible failure mechanism. The JBS diode structure was simulated for its reverse recovery at varying magnitudes of turn OFF di/dt for an ON-state current density of 100 A/cm².

Authors: M. G. Giesselmann; A. Bilbao

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7297019

Abstract: We are reporting on a comprehensive study on protective de-coupling networks for High Voltage (HV) pulsed power charging supplies. Typically HV power supplies charge large capacitor banks [1, page 3], [2, page 4], which are rapidly discharged into a pulsed power load. Even during a normal discharge, this can put severe stress on the power supply if it is not properly decoupled from the load. A fault at the load capacitor such as a flashover resulting in a ringing discharge with voltage reversal would put even more stress on the power supply, since the load capacitor could discharge through the rectifier diodes in forward direction. In such a case the output rectifier of the power supply could be instantaneously destroyed. Protective networks between the power supply and the load can prevent such damage but may limit the efficiency as well as the available power output and rep-rate of the HV power supply. We are reporting on a number of protective networks including combinations of resistors, inductors, and diodes that can be placed between the output of the power supply and the load. We are also considering the effects of parasitics and the surge I²t action integral [3, Page 20] of the output rectifiers of the power supply to arrive at guidelines for optimal system protection.

Authors: M. He; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7131804

Abstract: Microgrids, as individual controllable entities that can operate either islanded from or interconnected to main power grid, have emerged as a promising solution to improving energy efficiency and resilience to disturbance. When linked together in a self-organized manner, a cluster of microgrids can significantly enhance the reliability and power quality for critical load. With this insight, we study the self-organization and decentralized energy management of a microgrid cluster islanded from main grid after a disruptive event. In the self-organization stage, depending on the available generation resources, each microgrid decides on whether to connect to the cluster; and the microgrid energy management systems then “negotiate” on the optimal power exchange with each other in the cluster. Once the power exchange is determined, the generation and storage resources of each microgrid are managed to guarantee the energy reliability of critical loads and overall energy efficiency, through a scheduling procedure followed by a dispatch procedure. The effectiveness of the proposed method is revealed via case studies.

Authors: J. M. Parson; C. F. Lynn; D. H. Barnett; S. L. Holt; P. M. Kelly; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296909

Abstract: Repetitive operation of high-power microwave (HPM) sources is primarily restricted by thermal properties of anode and cathode materials. Note that even in single shot operation, vircators generally inject undesired plasma from the anode/cathode surfaces under high current densities, leading to excessive heating of electrode materials, vacuum degradation and impedance collapse that may ultimately lead to cut-off of microwave power. Heating of the anode and cathode materials during repetitive operation intensifies these problems and will cause accelerated erosion and/or permanent damage of the anode and cathode surfaces, especially when metallic electrodes are used. Hence, the vircator herein utilizes a carbon fiber cathode and a pyrolytic graphite anode. This paper elucidates the heating of the anode and cathode, and their spectrally resolved black-body radiation properties during repetitive operation.

Authors: L. Collier; M. B. Walls; J. Dickens; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296884

Abstract: The Marx pulse generator topology has been widely used in pulsed power applications1,2. Another pulse generator topology, the linear transformer driver (LTD), has been developed3 that may serve as a viable alternative to the Marx generator. LTDs utilize inductively added stages to achieve high voltages and currents. Unlike a Marx generator, each stage in an LTD features multiple bricks, all of which are ground referenced and allow current to be distributed amongst an arbitrary number of switches. This allows for LTDs that utilize solid-state switches, potentially resulting in more compact and reliable pulse generators. A solid-state, >10 kA peak current, multiple-stage LTD is developed. The generator's performance will be analyzed for viability as a replacement for driving a high power microwave generator.

Authors: M. He; S. Nimmagadda; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7285973

Abstract: In this paper, a novel scheme for subsynchronous oscillation detection and modal parameter estimation is proposed, by leveraging the rich information contained in high-rate phasor measurements as well as the effectiveness of synchrosqueezing transform for multimodal signal analysis. Specifically, an instantaneous time-frequency representation of a voltage/current signal is first obtained by applying synchrosqueezing transform to the real-time data collected by a phasor measurement unit. The non-zero synchrosqueezing transform coefficients quantify the undamped frequency components of the original voltage/current signal at each time instant. For an unknown number of undamped frequency components, unsupervised clustering is applied to the non-zero synchrosqueezing transform coefficients in the frequency domain, so as to determine how many modes comprise the signal, as well as which mode each non-zero synchrosqueezing transform coefficient belongs to. Then, for each detected mode, the corresponding non-zero synchrosqueezing transform coefficients are utilized to reconstruct a component of the original voltage/current signal. Finally, the magnitude, damping factor and phase angle of each mode are estimated by applying a least square estimation algorithm to the reconstructed component signal. The effectiveness of the proposed approach is revealed through several case studies using IEEE benchmark models. Further, practical issues involving missing data, measurement noise and transform basis functions are also systematically addressed in this study.

Authors: Veliadis, V; Steiner, B; Lawson, K; Bayne, SB; Urciuoli, D; Ha, HC

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7369297

Abstract: A requirement for the commercialization of power SiC transistors is their long term reliable operation under the hard switching conditions and high temperatures encountered in the field. Normally-ON 1200 V vertical-channel implanted-gate SiC JFETs, designed for high-power bidirectional (four quadrant) solid-state-circuit-breaker applications, were repetitively pulse hard switched at 150 degrees C from a 1200 V blocking state to an on-state current of 115 A, which is in excess of 13 times the JFET's 250-W/cm(2) rated current at 150 degrees C. The JFETs were fabricated in seven photolithographic levels with a single masked ion-implantation forming the p+ gates and guard rings, and with no epitaxial regrowth. The pulsed testing was performed using a low inductance RLC circuit. In this circuit, energy initially stored in a capacitor is discharged in a load resistor through the JFET under test. The JFET hard-switch stressing included over 2.4 million 1200-W115-A hard-switch events at 150 degrees C and at a repetition rate of 10 Hz. The peak energies and powers dissipated by the JFET at each hard-switch event were 73.2 mJ and 68.2 kW, respectively. The current rise rate was 166 Alps and the pulse FWHM was 1.8 mu s. After over 2.4 million hard-switch events at 150 degrees C, the JFET blocking voltage characteristics remained unchanged while the on-state conduction slightly improved, which indicate reliable operation. An optically triggered solid-state-circuit-breaker, based on these rugged JFET, is proposed.

Authors: R. S. Garcia; D. V. Reale; J. M. Johnson; W. H. Cravey; A. A. Neuber; J. C. Dickens; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7296893

Abstract: This paper describes the system integration of a four element, nonlinear transmission (NLTL) array. The HPM system components include a negative polarity 40 kV capacitor charging supply, a trigatron spark gap, a solid-state trigger generator, four delay lines with biasing coils, four coaxial ferrimagnetic NLTLs with biasing coils, eight dc current supplies, and four TEM horn antennas. A National Instruments cRIO FPGA based controller is used to interface the capacitor charging supply, the biasing coil power supplies and the trigger generator to a laptop based GUI. In order to minimize impact from EMI, all lines were properly shielded and the data acquisition equipment was located a sufficient distance away from the NLTL array. The GUI consists of options for controlling the current output on all 8 power supplies for biasing purposes, as well as pulse options that include repetitive pulsing based on the number of pulses or for a specific length of time. Pulse duration for the capacitor charger, the command charge, and the trigger can also be adjusted. An emergency stop button is included for safe shutdown of the trigger generator and all power supplies. A diagnostics setup is outlined describing the placement of inline coaxial D-Dot probes and a high voltage probe for laboratory testing.

Authors: A. Mishra; S. B. Bayne; Changzhi Li

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7443270

Abstract: Technology used for high performance DC-DC voltage regulator based Integrated Circuits is evolving. Zero voltage switching buck converter is one such DC-DC regulation schemes integrated in the power conversion circuitry and is designed such that power density is maximized and the switching losses are reduced. This paper discusses the design of AMI06 technology based zero voltage switching buck converter operating at 1 MHz switching frequency. A design example accommodating input voltage range from 5V–8V DC at a constant DC output voltage of 3V was considered. The simulations were performed on Cadence tool to verify the feasibility of the proposed converter. Finally this zero voltage switching buck converter is compared to the conventional buck converter in terms of switching losses and efficiency.

Authors: J. M. Johnson; J. M. Parson; D. V. Reale; A. A. Neuber; J. J. Mankowski; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287219

Abstract: This paper presents a design and implementation of a 1 kHz, 10 kV, positively pulsed trigger generator with an average risetime of 44 ns for use in a coaxial ferrimagnetic Nonlinear Transmission Line (NLTL) High Power Microwave (HPM) system. The pulse trigger generator is based on two stacked 4 kV current-controlled n-type thyristors in a low inductance package capable of driving inductive loads with risetimes less than 200 ns. However, by implementing a magnetic switch this relatively is slow risetime is sharpened to tens of nanoseconds. The stacked boards allow for voltage addition at the output while maintaining the fast riestime.

Authors: E. Cordero; S. Holt; J. Dickens; A. Neuber; J. Mankowski; S. Calico; M. Scott

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287370

Abstract: This paper presents the design of a scalable, high power battery system for pulsed power operations. The battery system is modular in design, with each module containing four Lithium Ion Polymer (LiPo) cells and a custom designed cell management board that actively monitors the voltage and temperature of each cell and also provides cell balancing functionality. The system is designed to be scalable by adding up to 25 modules in a series configuration. While the battery management system should be compatible with any lithium ion cells, this implementation uses 8 Ah capacity dual-core LiPo cells, rated for a 150 C discharge rate; allowing for a peak current output of 1,200 A. With 25 modules (96 LiPo cells) the system would have an open circuit voltage of 385 V and be capable of providing up to 1,200 A at 355 V for a peak output power of 420 kW. Special attention has been placed on safety features including overvoltage, undervoltage and temperature monitoring of every cell in the system. The charging/balancing system is capable of automatically shutting down if any of the voltages or temperatures exceeds established limits. The management circuitry is designed to have a low off-state power draw in order to maximize battery life when the system is not in use.

Authors: Eldridge, K; Fierro, A; Dickens, J; Neuber, A

PDF: https://ieeexplore.ieee.org/document/7287262

Abstract: Wavelet decomposition and reconstruction are utilized to synthesize a high power microwave (HPM) signal at a frequency below the frequency of the employed radiating sources. Employing a larger number (on the order of ten) of smaller sources that produce short radiating pulses combined with appropriate amplitude scaling and shifting of the individual pulses enables the generation of a single waveform of longer duration. We describe the mathematical approach to the wavelet synthesis and give examples. For instance, an array of 10 sources, each producing a 0.5 ns pulse can be adjusted to generate a sinusoidal wave with a period of approximately 2 ns. The results of low power experiments are discussed in detail to demonstrate the practical feasibility of the wavelet approach.

Authors: S. Holt; E. Cordero; J. Mankowski; J. Dickens; A. Neuber; M. Scott; S. Calico

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287211

Abstract: The design of a battery powered, rapid capacitor charger is discussed. The charger design is capable of an average output power of 80 kVA, for a burst time of roughly 5 seconds. Cool down time is required between bursts. A Lithium Ion Polymer (LiPo) battery pack with a nominal voltage of 355 V and fully charged open circuit DC voltage of 385 V will be used to power the rapid capacitor charger. The general topology of the charger is as follows. An IGBT H-bridge inverter uses pulse width modulation to create an AC waveform that is stepped up to a maximum of 58kV by a 1:170 ratio transformer. This high voltage output of the transformer is rectified and used to charge the capacitor bank.

Authors: P. Gatewood; A. Neuber; J. Dickens; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287248

Abstract: A metamaterial-inspired, efficient, electrically small antenna (ESA) is designed for the 2 to 20 MHz range, intended for use in a mobile Ionospheric Heating (MIH) system. The ESA consists of two main parts: a small semiloop antenna (SLA), and an outer capacitively loaded loop (CLL). To increase the capacitance of the CLL and reduce the size of the ESA, a high permittivity dielectric is utilized. All designs were simulated in Ansoft HFSS with particular attention on radiation efficiency and bandwidth. The design aims for these ESAs are a high overall efficiency at dimensions much smaller than a wavelength.

Authors: Kaili Eldridge; Andrew Fierro; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7287262

Abstract: Wavelet decomposition and reconstruction are utilized to synthesize a high power microwave (HPM) signal at a frequency below the frequency of the employed radiating sources. Employing a larger number (on the order of ten) of smaller sources that produce short radiating pulses combined with appropriate amplitude scaling and shifting of the individual pulses enables the generation of a single waveform of longer duration. We describe the mathematical approach to the wavelet synthesis and give examples. For instance, an array of 10 sources, each producing a 0.5 ns pulse can be adjusted to generate a sinusoidal wave with a period of approximately 2 ns. The results of low power experiments are discussed in detail to demonstrate the practical feasibility of the wavelet approach.

Authors: A. S. Subburaj; P. Kondur; S. B. Bayne; M. G. Giesselmann; M. A. Harral

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6824615

Abstract: This paper provides an overview of battery technologies and the grid-connected battery projects that integrate wind. It also discusses the deployment of a battery system at Reese Technology Center (RTC) in Texas. The research at RTC involves deploying energy storage devices (i.e. Batteries with grid-tie inverters) for application in wind farms to understand the interaction between wind energy, the grid and the grid loads. The research work focus on the modeling of a battery system and wind turbines when it is connected to the grid, with the aim to understand the optimized energy flow between the battery, the wind farm, and the utility grid, energy management for storage, battery performance, energy dispatch and reliability. With the view of modeling the test bed of the grid-connected battery project at Reese, this paper provides the preliminary simulation results using PSCAD on discharge characteristics of a single cell of a battery at various C rates.

Authors: A. S. Subburaj; S. B. Bayne; M. G. Giesselmann; M. A. Harral

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6839166

Abstract: This paper provides an analysis of the equivalent circuit of the 1 MW battery tied to the grid. It also discusses the installation of a 1 MWh battery system at Reese Technology Center (RTC) in Lubbock, Texas. The research involves deploying energy storage devices for application in wind farms to understand the interaction between wind energy, the grid and the grid loads. A 1 MW/1 MWh battery storage system at the RTC is connected to the South Plains Electric Cooperative (SPEC) grid. The batteries are used for energy storage and for mitigation of transient conditions grid dynamics. In this paper the 1 MW battery is modeled in PSCAD and analyzed for its discharge characteristics when it is tied to the grid.

Authors: A. S. Subburaj; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6972132

Abstract: With worldwide growth of battery technologies in renewable energy production, the selection of battery model has become an important factor for a Battery Energy Storage System (BESS) to be deployed in a grid environment. It is essential to understand the battery type and the model, to achieve a coordinated control of the performance of BESS, Renewable Energy Sources (RES) and load management, including battery applications such as ramp control, frequency response, voltage response, emergency backup, transmission and distribution capacity utilization and peak load leveling, when connected to the grid. Recent studies showed that the battery capacity utilization is based on the battery discharge characteristics. In this paper, two types of battery models such as `stacked cell model' and `single cell model' of a 1 MWh Lithium Manganese Oxide (LMO) battery are taken into consideration to study the discharge characteristics. The approach is based on the use of the Dual Polarization Model (one of the Electrical Equivalent Circuit Models also called as Two Time Constant Model) in building the above mentioned battery models. The purpose of this paper is to build and simulate the battery models in Power Systems Computer Aided Design (PSCAD). Also the battery models are compared and analyzed on their performance during faults when connected to the resistive load at 1C discharge rate. The `single cell model' is constructed using the electrical equivalent circuit to build the 1 MWh battery module. In an attempt to simulate the `stacked cell model', initially one cell of the battery rated at 4.12 V, 60 Ah is modeled in PSCAD and then the cells are connected in series and parallel to build the model of 1 MWh battery module. The key results include the modeling of single cell and stacked cell models of the 1 MWh battery. The discharge characteristics of the battery models rated at 960V, 1200 Ah will be analyzed for the faults applied near the resistive load. The models will be tested for node compatibility, compilation issues and the response effectiveness when connected to the resistive load. Thus, the research paper will perform a comparative analysis of the simulation results of the two battery models in terms of discharge characteristics, speed, performance, stability and compatibility.

Authors: C. White; D. Mauch; D. Thomas; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287328

Abstract: A system for measuring the recombination lifetime of high purity, semi-insulating (HPSI) 4H-SiC through transient free carrier absorption (FCA) for optimization of SiC photoconductive semiconductor switches (PCSS) is presented. The system measures the transient absorption of a continuous, low-power (<;5mW) 1550 nm infrared probe laser. Free carriers were generated with both above and below bandgap illumination from the harmonics of a Nd: YAG laser (532 nm, 355 nm, and 266 nm-10ns FWHM), and the carrier lifetime was numerically calculated from the absorption transient. High spatial resolution (~10 um) was attained through the use of a high-precision, three-axis stage. The carrier lifetime measurements of various regions of several SiC PCSSs over varying levels of photo-excitation are presented.

Authors: Thomas, D; Mauch, D; White, C; Neuber, A; Dickens, J

PDF: https://ieeexplore.ieee.org/document/7287329

Abstract: A method of characterizing mid-bandgap defect states in high purity semi-insulating 4H-SiC through leakage current analysis for optimization of SiC photoconductive semiconductor switches, PCSS, is presented. The method utilizes two custom IV curve tracer systems to measure leakage currents through the material under various voltage/current conditions. The first system is used under low current conditions and is capable of measurements from 0 to 45 kV at currents ranging from 0 to 3 mA with pA resolution. While voltage/current measurements in the region >0.1 mA are of primary interest for quantifying defect states near the conduction band, standard IV measurements become difficult due to excessive power dissipation in the PCSS. Hence, a second system operating in transient mode is used for currents higher than 0.1 mA. This system measures the transient discharge of a charged capacitor through the PCSS, allowing for high current measurements while subjecting the material to high power dissipation for only a short period of time (milliseconds). It is the goal to extract from the combined data of these two systems characteristics of the defect states (concentration, energy level).

Authors: A. S. Subburaj; W. B. Ray; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7351778

Abstract: This paper performs a comparison study on the performance of operation of a battery connected grid system with bi-directional grid-connected inverter and conventional PQ controller in steady state conditions. The deployment of a 1 MWh battery energy storage system (BESS) at Reese Technology Center (RTC) in Lubbock, Texas is described. The battery energy storage device was installed to further understand the interaction and performance of the battery within the power grid with the existing wind turbines at RTC as well as renewable energy sources and grid loads. The paper shows the simulation results obtained from PSCAD with the design of the different controllers. Compared with conventional PQ controllers, the bi-directional converter controllers greatly enhance the grid-connected converter's performance in the aspects of reduced power oscillations and increased operating stability.

Authors: He, M; Zhang, JS

PDF: https://ieeexplore.ieee.org/document/7007662

Abstract: Deadline-aware concentration of synchrophasor data from spatially dispersed phasor measurement units (PMUs) is studied, with an objective to find the optimal policy for the wait time of a phasor data concentrator (PDC), such that the deadline-constrained synchrophasor data delivery from PMUs via PDC to control center achieves maximal expected throughput. When the statistical information on the communication latency from PMUs to PDC and from PDC to control center is known, the wait time problem is cast as a continuous-time optimal stopping problem. The optimal stopping policy is then shown to exist, and then obtained by utilizing the infinitesimal look-ahead rule under mild conditions. Further, the scenarios with substation PDCs or super PDCs are also considered. The dynamic wait time policy obtained by using the proposed optimal stopping approach, in comparison with two fixed wait time policies, results in significant improvement, as revealed through simulation studies.

Authors: D. H. Barnett; J. Parson; C. Lynn; P. Kelly; M. Taylor; J. Dickens; A. Neuber; J. Mankowski; S. Calico; M. Scott

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287287

Abstract: This paper presents the design and performance characteristics of an optically isolated, trigger generator for spark gap applications that is capable of operating at over 500 Hz rep rate. The pulsed trigger generator is designed to achieve fast risetimes on the order of 1kV/ns to cause breakdown between the trigatron trigger pin and the opposite electrode. The pulsed trigger generator is designed to trigger a 10-stage, 500 kV, 42 J Marx generator. The system is capable of delivering a 4"“25 kV pulse with a 10/90 risetime of 20 ns. A board was fabricated to control a capacitor charger and high voltage thyristor via fiber optic inputs. The capacitor charger, a 60 W constant current dc-dc power supply with variable voltage control, is capable of charging 1 nF in less than 0.3 ms. The system is battery-powered using a 33 V lithium polymer battery pack for isolated operation in noisy environments. For burst mode operation, a larger 250 nF buffer capacitor, which is initially charged, is connected through a 100 kΩ resistor to the trigger capacitor, which is varied between 1 and 40 nF. The value of the trigger capacitor is adjusted to provide optimum performance for a given transformer and magnetic switch. System modeling results along with experimental data are discussed including the reliability and performance of the overall system including jitter at 500 Hz operation.

Authors: M. G. Giesselmann; A. Bilbao

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287357

Abstract: We are reporting on a software implemented HV-sensor for a Digitally Controlled Rapid Capacitor Charger. The new contribution of this paper is to replace a physical voltage sensor with an algorithm that executes in real time in the digital controller for the charger. This avoids the cost, bandwidth limitations and insulation challenges of a conventional HV-sensor. The capacitor charger is controlled by a digital signal controller using a peak current mode control algorithm with adaptive slope compensation as reported in [1]. The peak current mode algorithm uses the analog comparator to detect the peak current and the duty cycle of the inverter at the current peak. Through knowledge of the current waveshape, we can deduct the charge that was transferred during each cycle. This information can be used to predict the voltage of the load capacitor during the charging process. To evaluate the feasibility of this approach, we compare the results from the capacitor voltage prediction algorithm to data obtained with a laboratory grade HV probe.

Authors: S. Ramabhotla; S. Bayne; M. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7061838

Abstract: Microgrid combines with Distributed Energy Resources (DERs), energy storage, and controllable loads. It is connected to the maingrid through the point of common coupling (PCC). Economic dispatch using reduced gradient method is implemented for the optimization of energy in the microgrid using MATLAB. The optimization is obtained by minimizing the cost function of the system while meeting the load demand. The operation & maintenance cost, and investment costs are considered in the cost functions of the microsources. The minimum total cost of the system is obtained by comparing different scenarios of the microsources in the microgrid.

Authors: M. Hinojosa; A. Ogunniyi; H. O'Brien; S. B. Bayne; C. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287217

Abstract: This paper presents preliminary results on the static and dynamic characterization of 12 kV and 20 kV N-channel silicon carbide Insulated-Gate Bipolar Transistors (IGBTs). These state-of-the-art devices were evaluated for their possible use in pulsed-power and energy conversion applications. The 12 kV IGBTs had a chip area of 0.7 cm2 and were rated for 10 A. Their active area was 0.32 cm2, with a drift region of 140 μm and two different field-stop buffers of 5 μm and 2 μm. The 20 kV IGBTs had a chip area of 1 cm2 and were rated for 12 A. Their active area was 0.37 cm2, with a drift region of 180 μm, and their FSB was 2 μm. The switching and conduction losses were calculated for both devices with short pulses and low-inductance resistive loads. Both types of IGBTs displayed promising results for possible replacement of gas switches and Si IGBTs in high voltage applications.

Authors: Pol, S; Taylor, A; Bilbao, A; Doostalab, A; Novoa, S; Westergaard, C; Hussain, F; Sheng, J; Ren, BB; Giesselmann, M; Glauser, M; Castillo, L

PDF: https://iopscience.iop.org/article/10.1088/1742-6596/524/1/012175

Abstract: As a first step to study the dynamics of a wind farm, we experimentally explored the flow field behind a single wind turbine of diameter 1.17 m at a hub height of 6.25 m. A 10 m tower upstream of the wind farm characterizes the atmospheric conditions and its influence on the wake evolution. A vertical rake of sonic anemometers is clustered around the hub height on a second tower, 6D downstream of the turbine. We present preliminary observations from a 1-hour block of data recorded in near-neutral atmospheric conditions. The ratio of the standard deviation of power to the inflow velocity is greater than three, revealing adverse effects of inflow turbulence on the power and load fluctuations. Furthermore, the wake defect and Reynolds stress and its gradient are pronounced at 6D. The flux of energy due to Reynolds stresses is similar to that reported in wind tunnel studies. The swirl and mixing produces a constant temperature wake which results in a density jump across the wake interface. Further field measurements will explore the dynamics of a model wind farm, including the effects of atmospheric variability.

Authors: C. F. Lynn; J. Parson; P. Kelly; M. Taylor; D. Barnett; A. Neuber; J. Dickens; J. Mankowski; S. Calico; M. Scott

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287245

Abstract: Vircators (virtual cathode oscillators) are easily frequency tunable by simply altering the electron beam current density through either adjusting the accelerating voltage or varying the anode cathode gap (A-K gap). It is noted that vircators typically exhibit frequency hopping, an often undesired feature. Here we address the frequency hopping problem by introducing a rectangular shaped carbon fiber cathode with the width of the cathode less than one quarter of a wavelength in the direction of microwave propagation. Operating the vircator with the rectangular cathode revealed a match with a resonant condition from back wall reflections, which explains the observed dominant microwave frequency selection. That is, the vircator emitted stably during 50 ns operation at 3.89 GHz or 2.07 GHz with a 3 dB bandwidth of 37.5 MHz or 93.7 MHz, at A-K gap spacing of 8 mm, and 12 mm respectively. To further investigate frequency tuning, the A-K gap was swept from 3 mm to 17 mm while keeping a constant 5.1 cm diameter circular cathode. The resulting frequencies of operation were 1.5 GHz, 2.1 GHz, 4.1 GHz, 4.6 GHz, and 5.9 GHz at A-K gap spacing of 17mm, 14 mm, 7 mm, 5.75 mm, and 3 mm respectively. This demonstrates that a tuning range of 4.4 GHz should be achievable for a practical vircator system.

Authors: H. K. O'Brien; W. Shaheen; A. Ogunniyi; C. Scozzie; L. Cheng; M. Hinojosa; K. Lawson; S. Lacouture; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287331

Abstract: Asymmetric thyristors require protection from voltage and current reversals in high-inductance capacitor discharge systems. Silicon carbide (SiC) PiN diodes capable of blocking up to 16 kV were demonstrated to have the high-current capability to transmit forward pulse current in a series configuration with a thyristor, and to clamp reverse current in an anti-parallel configuration. In series with a thyristor, diodes were switched 1000 pulses at a single-shot rate at 2000 A peak current (3.8 kA/cm² over anode area and 2100 A²s per pulse) without any notable increases in forward voltage or reverse leakage current. In the reverse clamp configuration, a parallel pair of PiN diodes was demonstrated to block 12 kV charge on the capacitor bank, then clamp a total of 4200 A current reversal with good parallel current sharing. These evaluations demonstrate that for high current density pulsing above 10 kV, individual 16 kV PiN diodes yield lower on-state voltage loss (16 V at 2000 A) than series-stacked assemblies of 9 kV SiC PiN diodes or 6 kV Si diodes.

Authors: Andrew Fierro; James Dickens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7287318

Abstract: The development of plasma in a medium pressure (50 torr) environment in nitrogen was studied by simulation and measurement under the influence of non-uniform, pulsed electric fields. A GPU-accelerated, 3-dimensional particle-in-cell (PIC)/Monte Carlo Collision (MCC) simulation code was written utilizing the CUDA platform to simulate pulsed plasma development in a nitrogen environment and uncover the transient plasma characteristics in detail. The simulation provides significant speed-up over the CPU equivalent implementations. Experimentally, a needle-protrusion (1.5 mm in length, 200 μm tip radius) opposite a brass ground plane with the distance between needle-tip and wall held at 1.5 mm provided a non-homogeneous field. Excitation of the needle-plane gap was achieved with a ~100 ns rise-time high-voltage pulser with a peak voltage of 30 kV. Diagnostics included time-resolved nanosecond gated imaging for light intensity measurements and high speed electrical probes for timing. A time series of the plasma formation captured with a 5 ns camera gate revealed a mostly uniform expanding plasma cloud from the needle tip.

Authors: W. B. Ray; J. A. Schrock; K. Lawson; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287371

Abstract: A solid-state modular Marx Generator was designed for the purpose of testing the dV/dt capability of Power Semiconductor devices. The Marx Generator is capable of producing up to 1 kV voltage transients with rise times on the order of 10-nanoseconds. This capability to do variable voltage amplitude leads to customizable dV/dt tests. The solid-state modular design will be covered in detail within the paper. The solid-state construction allows for this adjustable dV/dt rating through the utilization of multiple modular stages of Power MOSFETs. These Power MOSFETs have a lower voltage blocking ability, but faster switch closing times. The different modules are controlled through the use of fiber-optic links. These links signal the floating Gate Driver circuitry to synchronize their switch closing time. The floating Gate Driver utilizes isolated switch-mode DC-DC converters to supply the power needed to charge the MOSFET gates, from one ground-referenced power source. The Marx Generator uses high voltage Silicon Carbide Schottky Diodes for the voltage blocking elements when the individual modular stages trigger. The lack of reverse recovery for these high voltage SiC Diodes enables the fast voltage transients that are requisite for dV/dt ranges needed to test Power Semiconductor devices.

Authors: M. B. Taylor; P. M. Kelly; J. M. Parson; C. Lynn; J. C. Dickens; A. A. Neuber; J. J. Mankowski; J. . -W. B. Bragg; S. Calico; M. Scott

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287288

Abstract: A 42 J, 10-stage pulse forming network (PFN) Marx generator capable of producing a 500 kV, 50 ns full-width-half-max (FWHM), ~5 ns rise time pulse into an open load at a rep-rate of 500 Hz has been designed for use as a pulsed power source for a reflex triode virtual cathode oscillator (vircator). Rayleigh PFNs are used in place of discrete capacitors for each stage of the 10-stage Marx generator. Effort was taken to minimize parasitic inductance such that the quality of the pulse shape is maintained as much as possible. In order to rep-rate the Marx generator, a trigatron-based triggering scheme is used to initiate erection of the Marx generator. A 20 ns risetime, 24 kV solid-state pulse trigger generator capable of operating at high repetition rates is used to drive the trigatron. The required charge rate for a 500 Hz pulse repetition frequency (PRF) for the Marx generator is 24 kW. Repetitive operation requires additional design considerations that would be irrelevant to single pulse firing. Pressurized air is jetted across the spark gaps by means of built-in gas manifolds to remove remaining ionized gas between each pulse and prevent premature erection during the subsequent charging cycle. The built-in gas manifolds were designed using a hydrodynamic simulation to ensure equal flow rate across each of the spark gaps and equal pressure along the length of the tube chamber.

Authors: J. Stephens; D. Ryberg; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012397

Abstract: Summary form only given. A 400 μF capacitor bank, charged as high as 3.5 kV (2.45 J) is used to drive an exploding wire in air with up to 10 s kA over a 100 s of microsecond timescale. Utilizing a short wire with 286 μm diameter enables generating an intense shockwave with overpressures on the order of 1 psi (6.9 kPa). Investigation of the energy dependence of the overpressure magnitude was completed by both varying the initial capacitor voltage, as well as parasitically introducing additional resistive dampening into the circuit. Additionally, variation in both capacitance and inductance were made in order to observe the effect of varying the driving source timescale on generated overpressure. Optimization of the exploding wire load involved exploring various materials, radii, length, as well as the number of exploding wires. Preliminary theoretical predictions are made using the Sandia National Laboratories ALEGRA-MHD software. Theoretical and experimental results are benchmarked against one another to assess the accuracy of the theoretical predictions.

Authors: D. Mauch; C. White; D. Thomas; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287198

Abstract: Recent advances and the current state-of-the-art for high speed 4H-SiC photoconductive semiconductor switches (PCSS) developed at Texas Tech University are summarized. A performance comparison of multiple generations of switch designs is also presented. These devices have experimentally demonstrated the capability of blocking DC electric fields up to 705 kV/cm (<;0.1 mA leakage current), rise times of 0.63 ns (20/80), and switching 20 kV at 250 A with a di/dt of 75 kA/us at a burst repetition frequency of 65 MHz. Findings and optimizations pertaining to device geometry, sub-contact doping, contact thickness, triggering wavelength, and electron irradiation are presented. Device modeling and experimental results investigating current issues with device lifetime are presented as well.

Authors: P. Kelly; J. M. Parson; C. Lynn; M. Taylor; J. C. Dickens; A. Neuber; J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012655

Abstract: Summary form only given. This study focuses on the use of St707 non-evaporable getter (NEG) material in a high power microwave (HPM) sealed-tube virtual cathode oscillator (vircator) operated at 500 Hz repetition rate. High-current pulsed operation releases gases trapped within the bulk materials and gas monolayers on material surfaces, leading to localized plasma production in the A-K gap. This can lead to gap closure, shorten the duration of microwave emission, and spoil vacuum. A single current pulse increases the chamber pressure to the low 10-6 Torr range from an initial background pressure in the low 10-9 Torr range, desorbing approximately 1014 particles. At 500 Hz operation, a sufficiently large pumping speed (~2,500 L/s) is necessary to evacuate desorbed particles from the vircator volume during consecutive shots. Previous work has identified hydrogen (H, H2) as the main outgassing species during vircator operation, with contributions from CH4, N2, CO, CO2, and Ar as the other primary gas constituents[1]. The St707 NEG pumps shows an affinity for pumping hydrogen, making it a suitable choice to adsorb vircator outgassing species, of which hydrogen is an order of magnitude greater than any other gas species. Previously, without the presence of NEG material, degradation of microwave output power from the vircator has been observed during subsequent shots at 1 Hz, 10 second burst mode operation. Subsequently, an increase in chamber pressure from 1.25×10-6 Torr on the first shot to 10-5 Torr on the tenth shot has been experimentally observed. This paper details the performance of the St707 NEG material (70% Zr, 24.6% V, 5.4% Fe) for maintaining UHV conditions during rep-rated vircator operation. Pumping characteristics of multiple St707 NEG pumps in the presence of rep-rated high-current pulses are presented. Diagnostic results obtained with a residual gas analyzer to observe individual gas constituents and two inverted magnetron cold cathode gauges for absolute pressure are utilized to analyze vircator and getter performance in detail.

Authors: M. Walls; J. Dickens; J. Mankowsi; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012301

Abstract: Summary form only given. An eleven stage flat top type output Marx utilizing sparkgap switches is operated to produce a desired 500 kV output at 10 Hz rep-rate with > 500 J pulse energy to drive a ~ 20 Ohm load. While this Marx generator utilizes custom-designed spark gaps, the significant progress in semi-conductor pulsed switching technology is evaluated through PSICE modeling of the semiconductor switches integrated into the Marx electrical model. The switches are experimentally evaluated at conditions that a single switch (stack) would experience in a full Marx generator. Alternatively to the Marx topology, the potential benefits of choosing a linear transformer driver (LTD) as a HPM pulsed power driver are evaluated. We note that the main advantage for the LTD is the savings in insulation material and size; the same semiconductor switches are utilized. The potential benefits of such a generator include significant reduction of electromagnetic noise, increased lifetime, improved reliability, and reduced maintenance.

Authors: J. Stephens; A. Fierro; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012344

Abstract: Summary form only given. A micro hollow cathode discharge VUV source, with time averaged power above 3 watts, and peak power above 40 watts is presented1. Such operation is achieved by driving the microdischarge (MD) in the arc mode where high electron densities and low discharge impedances exist. Improved theoretical estimations regarding the timescales of arc formation are presented. With typical electrical sources, the arc regime is associated with a collapse of the potential and poor coupling of energy. However, the pulsed driver utilized in this study features a low output impedance, thus sustaining a reasonable voltage across the MD while simultaneously delivering 10s of amperes when the arc mode is achieved. This allows for time averaged input power on the order of 550 watts. Gas mixtures such as Ar-H2 (99.7%/0.3% by volume) have demonstrated strong VUV emission at Lyman-α (121.6 nm), with measured power levels exceeding 40 watts peak.This study details the influence of the MD performance under various drive modes. Utilizing a pulse forming line based driver the behavior of the MD under 10 ns excitation and various repetition frequencies, with and without DC bias are presented. The MOSFET based low impedance driver, enables investigating the operation of the MD under high average input power. We will demonstrate the feasibility of utilizing the developed MD to achieve volume pre-ionization in appropriate target gases.

Authors: J. M. Parson; J. -W. B. Bragg; M. Taylor; D. Barnett; P. Kelly; C. F. Lynn; S. Holt; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012435

Abstract: Thermal limitations of anode and cathode materials have shown to negatively impact operation of cold-cathode high-power microwave (HPM) sources. High pulse-repetition-frequency (PRF) operation of these devices exacerbates the problems already experienced in single shot mode where cold-cathode devices, specifically carbon fiber cathodes, are plagued by plasma formation on the anode and cathode followed by plasma/gas expansion that causes impedance collapse of the anode-cathode (A-K) gap. Hence, for frequency stable, repetitive operation, cold-cathode HPM devices require the use of thermally robust electrode materials and ultra-clean surfaces, leading to repeatable tube operation. This study focuses on burst-mode operation of an HPM sealed tube reflex-triode virtual-cathode-oscillator (vircator) for PRFs greater than 100 Hz. The vircator is driven by a 54 J, ∼200 kV Marx generator with an approximate pulse width of 50 ns FWHM, and the vircator chamber has an empty volume of approximately 5 L with background pressures in the low 10−9 Torr. The anode materials studied include grade-1 titanium (TiG1), nickel 201L (Ni201L), and stainless steel 316L (SS316L); all in combination with a carbon fiber cathode. Empirically observed outgassing characteristics in conjunction with anode thermal modeling are presented under single-shot and rep-rate conditions. In addition, scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) techniques were employed to investigate anode and cathode surface integrities before and after vircator operation.

Authors: D. V. Reale; D. Mauch; J. M. Johnson; A. A. Neuber; J. C. Dickens; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287223

Abstract: An all solid-state high power microwave (HPM) source is constructed using a photoconductive semiconductor switch (PCSS) based HV pulse generator to drive a sulfur hexaflouride (SF6) insulated coaxial ferrimagnetic nonlinear transmission line (NLTL) which feeds a TEM horn antenna. The PCSS was fabricated from high purity semi-insulating (HPSI) 4H-SiC and is illuminated with 2 mJ from a frequency tripled Nd:YAG laser at 355 nm with a 7 ns FWHM. Fixed fiber optic delay lines are utilized to generate a burst of four optical pulses from a single solid-state laser source for rep-rate operation. The input to the NLTL is an adjustable pulse from 3 kV to 6 kV with sub-ns rise time and the resulting output of the NLTL is radiation in the L-band to S-band regime with RF power from 100–200 kW depending on charge voltage. A Rexolite® insulated zipper transition, in which the inner conductor of the coax is gradually exposed, is used at the output of the NLTL to transition from a coaxial structure to parallel plate structure in order to feed a free space TEM horn. The zipper transition and feed section of the TEM horn are potted using Sylgard® 184 silicone elastomer to prevent breakdown at the antenna feed and maintain a homogenous dielectric constant for the transition region. Radiated waveforms are presented for several charge voltages and bias conditions.

Authors: S. Beeson; J. Dickens; A. Neuber; S. Lin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012670

Abstract: Summary form only given. The delay time between the rising edge of a high power microwave (HPM) pulse and plasma formation in a gas at close to atmospheric pressure is studied for a multi-pass geometry. A fundamental lower limit of the breakdown delay time is the formative delay time, defined as the time it takes for the exponentially rising electron density to reach the critical electron density for which the EM wave begins to reflect and attenuate; which is on the order of 10's to 100's of nanoseconds for the conditions of interest. The goal of this research was to surpass this limit by making two passes through the generated plasma with the HPM pulse. That is, an external structure is utilized to time delay the HPM pulse on itself for a second pass through a gas cell.

Authors: J. M. Parson; J. -W. B. Bragg; M. Taylor; D. Barnett; P. Kelly; C. F. Lynn; S. Holt; J. C. Dickens; A. A. Neuber; J. J. Mankowski

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012435

Abstract: Thermal limitations of anode and cathode materials have shown to negatively impact operation of cold-cathode high-power microwave (HPM) sources. High pulse-repetition-frequency (PRF) operation of these devices exacerbates the problems already experienced in single shot mode where cold-cathode devices, specifically carbon fiber cathodes, are plagued by plasma formation on the anode and cathode followed by plasma/gas expansion that causes impedance collapse of the anode-cathode (A-K) gap. Hence, for frequency stable, repetitive operation, cold-cathode HPM devices require the use of thermally robust electrode materials and ultra-clean surfaces, leading to repeatable tube operation. This study focuses on burst-mode operation of an HPM sealed tube reflex-triode virtual-cathode-oscillator (vircator) for PRFs greater than 100 Hz. The vircator is driven by a 54 J, ~200 kV Marx generator with an approximate pulse width of 50 ns FWHM, and the vircator chamber has an empty volume of approximately 5 L with background pressures in the low 10^-9 Torr. The anode materials studied include grade-1 titanium (TiG1), nickel 201L (Ni201L), and stainless steel 316L (SS316L); all in combination with a carbon fiber cathode. Empirically observed outgassing characteristics in conjunction with anode thermal modeling are presented under single-shot and rep-rate conditions. In addition, scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) techniques were employed to investigate anode and cathode surface integrities before and after vircator operation.

Authors: J. M. Parson; C. F. Lynn; J. -W. B. Bragg; P. Kelly; M. Taylor; D. Barnett; S. Holt; J. C. Dickens; A. A. Neuber; J. J. Mankowski; M. C. Scott; S. E. Calico

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287274

Abstract: Thermal limitations of anode materials are known to impose limits on rep-rate operation of cold-cathode high-power microwave (HPM) sources. This study focuses on performance of pyrolytic graphite (PG) anodes at a 500 Hz burst-mode operation in a reflex-triode virtual-cathode-oscillator (vircator). In most experiments, a 42 J, 300 kV pulse forming network (PFN) based Marx generator with an approximate pulse width of 50 ns full-width-half-max (FWHM), was utilized to drive the vircator. Rep-rated operation of the vircator exacerbates the problems already experienced in single-pulsed mode where vircators are plagued by plasma formation on the anode and cathode followed by plasma/gas expansion that causes degradation of anode materials. Hence, for frequency-stable, repetitive operation, vircators require the use of thermally robust electrode materials and ultra-clean vacuum surfaces, leading to repeatable diode operation. This contribution presents thermal modeling of anode heating and experimental electrical behavior of vircator rep-rate operation.

Authors: Lawson, K; Schrock, J; Ray, W; Bayne, S; Cheng, L; Palmour, J; Allen, S

PDF: https://ieeexplore.ieee.org/document/6856036

Abstract: Modern power electronics systems try to maximize power density and efficiency. As such, the active switch is required to safely handle very stressful transient conditions. A 56 mm(2), 180 A, SiC DMOSFET manufactured by Cree Inc. is evaluated by electrically stressing the device in a RLC ringdown test system capable of producing peak current in excess of 600 A (> 3X rated current) and di/dt's as high as 860 A/mu s. The device was hard-switched 5,000 times at repetition rates of 1, 2, 5, and 10 Hz for a total of 20,000 switching events. The device characteristics were monitored every 1,000 shots on a high power curve tracer to determine device degradation. The devices showed no changes in blocking characteristics and minimal changes in on-state characteristics due to shifts in the threshold voltage after 20,000 hard switching events. The threshold voltage shifts over the test period are minimal with a +/- 93 mV deviation from the average of 4.39 V. With the stability of the threshold voltage, on-state characteristics, and blocking characteristics; this shows that this device would perform reliably within commercial applications that include stressful switching conditions.

Authors: A. Neuber; J. Dickens; J. Mankowski; L. Hatfield; H. Krompholz; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7012489

Abstract: We review the past 20 years of progress in pulsed power science and technology at Texas Tech University, a US academic institution, specifically focusing on efforts that came to fruition with the support of Dr. Robert Barker, AFOSR.

Authors: J. D. Carlson; M. Mittek; S. A. Parkison; P. Sathler; D. Bayne; E. T. Psota; L. C. Pérez; S. J. Bonasera

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6944048

Abstract: As a first step toward building a smart home behavioral monitoring system capable of classifying a wide variety of human behavior, a wireless sensor network (WSN) system is presented for RSSI localization. The low-cost, non-intrusive system uses a smart watch worn by the user to broadcast data to the WSN, where the strength of the radio signal is evaluated at each WSN node to localize the user. A method is presented that uses simultaneous localization and mapping (SLAM) for system calibration, providing automated fingerprinting associating the radio signal strength patterns to the user's location within the living space. To improve the accuracy of localization, a novel refinement technique is introduced that takes into account typical movement patterns of people within their homes. Experimental results demonstrate that the system is capable of providing accurate localization results in a typical living space.

Authors: Yang, L; He, M; Vittal, V; Zhang, JS

PDF: https://ieeexplore.ieee.org/document/7039381

Abstract: We study stochastic optimization of economic dispatch (ED) and interruptible load management using short-term distributional forecast of wind farm generation. Specifically, we develop a Markov-chain-based distributional forecast model for wind farm generation based on spatial and temporal characteristics of the wind turbine power output in a wind farm. Built on the distributional forecast model, the joint optimization of ED and interruptible load management is cast as a stochastic optimization problem. The proposed stochastic ED problem is compared with the deterministic ED problem using the persistence wind generation forecast model, and also with the scenario-based ED formulation which uses all possible wind generation states. Numerical studies, using a modified IEEE RTS 24-bus system and realistic wind measurement data from an actual wind farm, demonstrate the significant benefits obtained by leveraging the Markov-chain-based distributional forecast and the interruptible load management.

Authors: B. N. Pushpakaran; S. B. Bayne; A. A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7287327

Abstract: This research investigates the electro-thermal switching characteristics and lattice temperature profile of a two dimensional (2D) silicon carbide (4H-SiC polytype) Metal Oxide Semiconductor Field Effect Transistor (MOSFET) cell under resistive switching using Silvaco ATLAS Technology Computer Aided Design (TCAD) physics based simulation software. Physics based models were included to account for recombination effects, bandgap narrowing, low field and high field mobility and lattice heating. The electro-thermal simulation was performed at an ambient lattice temperature of 300K. The device was simulated for 100 A/cm2 and 1000 A/cm2 drain current densities using a 1 kHz 50% duty cycle gate signal consisting of two cycles and 1.6 kHz 80% duty cycle signal consisting of three cycles. The analysis of lattice temperature profile revealed the formation of thermal hot spots in the vicinity of the (Junction Field Effect Transistor) JFET region in the DMOSFET structure during the switching phase and at the edge of the channel during the conduction phase. The magnitude of temperature rise was dependent on the drain current density used during the simulation.

Authors: Dallas, T; Karp, T; Nutter, BS; Lie, YCD; Gale, RO; Cox, R; Bayne, SB

PDF: https://peer.asee.org/university-industry-partnerships-in-semiconductor-engineering

Abstract: We describe a long-standing and successful university-industry partnership in semiconductor device engineering with a primary focus on product and test engineering. The partnership, now (2013) in its 15(th) year, relies on a symbiotic relationship that has evolved over the years to reflect semiconductor industry trends and advancing university capabilities. The success of the partnership is due to a multifaceted approach with an emphasis on frequent interactions between company personnel and university faculty and students. These interactions feed the core component of the program, student internships. These internships, for which students can obtain course credit, are done at both the undergraduate and graduate level and provide a nearly seamless pathway from school to full-time employment.

Authors: M. Giesselmann; T. Vollmer; W. Carey

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6633266

Abstract: A 100kV high frequency transformer / rectifier package was designed and tested which is capable of dual polarity operation. The primary of the transformer is driven by an H-Bridge at a frequency of 20 kHz. The transformer has two HV-coax output cables with a grounded shield on each cable. One of the HV output cables is designated as the positive output terminal whereas the other output cable is designated as the negative output terminal. The shields of both output cables are always grounded. In addition, the core of the transformer is grounded. To obtain either positive or negative output voltage, either of the HV center conductors of the output coax cables is grounded. If the center conductor of the coax cable designated as negative output is grounded, positive voltage is obtained from the coax cable designated as positive output is obtained and vice versa.

Authors: N. R. Gonsalves; D. V. Reale; J.-W. B. Bragg; W. W. Sullivan; A. A. Neuber; S. L. Holt; J. J. Mankowski; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6627665

Abstract: Recent work on Coaxial Ferrimagnetic Nonlinear Transmissions Lines (NLTL) has been focused on developing an array of NLTLs for use as a solid-state High Power Microwave (HPM) source. The pulsed output of an NLTL requires a combiner that can combine transient signals at voltage levels up to 50 kV. Existing combiner designs found in literature require resonant structures to achieve efficient power combination. The presented coaxial combiner is an in-plane structure designed to combine two 50 Ω transmission lines into a single 25 Ω coaxial line output which then uses a logarithmic taper to transition back to 50 Ω. The combiner design was simulated using a transient Finite Element Method (FEM) model in COMSOL Multiphysics® and experimental results are compared with simulation. A 4-way combiner based on an in-plane 2-way design is simulated and the field stresses are examined to determine maximum electric field levels encountered in the structure. This enabled estimating the associated maximum voltage level that the structure is able to support.

Authors: G. Laity; A. Fierro; L. Hatfield; J. Dickens; A. Neuber; K. Frank

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635196

Abstract: Summary form only given. There has been a continued interest in utilizing streamer and spark discharges for new technologies that require low temperature plasma generation at atmospheric pressure, but diagnostics of these plasmas typically require external sources of probing radiation such as UV lamps or laser systems. Specifically, it is desired to measure the dissociated gas density from pulsed surface flashover plasmas, without the use of potentially invasive optical techniques such as two-photon absorption laser induced fluorescence (TALIF) spectroscopy, which may artificially increase the dissociation degree. We demonstrate a method for determining the dissociated gas density of N and H atoms in an N2/H2 surface flashover plasma by passively monitoring the self-absorption of intrinsic radiation produced by the 2s2 2p2 3s→2s2 2p3 NI transition(s) at 120.0 nm, and the 2p→ls HI Lyman-α transition at 121.57 nm. This radiation is partially trapped by the spark plasma, assumed to be of Gaussian cylindrical shape with 500 μm diameter. The resulting emission line shapes can be calculated by inferring the plasma temperature, gas mixture, and the estimated dissociated atom density of each species in the plasma volume of measurement. For example, 80%/20% N2/H2 discharges with a measured electron temperature of ~3.0 eV produce peak dissociated concentrations of 2% and 9% for atomic N and H, respectively, during the spark phase ~100 ns after voltage collapse. By assuming the quasi-contiguous approximation of the Holtsmark micro-field due to local electron perturbation of the HI radiators, the Stark line width of Lyman-α radiation yields electron densities on the order of 1018 cm3 during the spark phase. This self-absorption method has been extended to provide density information of surface flashover plasmas in air environments by passively monitoring the 2s2 2p3 3s→2s2 2p4 OI transitions) at 130.2 nm / 130.5 nm / 130.6 nm, which yield peak dissociated concentrations of 20% and 7% for atomic O and N, respectively.

Authors: D. Ryberg; G. Laity; A. Fierro; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635150

Abstract: Summary form only given. Limited experimental literature exists for measuring radiation with wavelengths shorter than 115 nm primarily due to the difficulty associated with measuring VUV radiation due to the high absorption in molecular oxygen and optical materials. Furthermore, there is an increasing interest in quantifying the role of this radiation in the generation of low temperature plasmas. This paper describes a system for studying the early mechanisms of vacuum UV generation below 115 nm by inducing breakdown in an air environment near atmospheric pressure. However to avoid the attenuating effects of O2, the experiment is set up such that the pressure drops to deep vacuum (i.e. ~10-6 torr) within a distance of several millimeters from the location of breakdown without the use of a solid barrier. In order to perform this measurement, the experiment consists of an enclosure at high vacuum, a pair of electrodes, and a micro puff valve. The puff value releases an above atmospheric pressure wave of air in the vicinity of the electrodes, whose differential electrical potential is then raised to about 20kV with a rise-time on the order of 100 nanoseconds. With application of the pulsed electric field, the radiation emitted from the excited species escapes the very localized puff of air and propagates through the high vacuum largely unattenuated where it is resolved by vacuum UV-sensitive optics and detectors. With this setup, the generation of vacuum UV radiation below 115 nm can be studied in detail and in the future the data recorded with this system will be used in our modeling approaches for further understanding of low temperature plasma generation.

Authors: David Reale; J.-W. Braxton Bragg; Shad Holt; Andreas Neuber; John Mankowski; James Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6633209

Abstract: Coaxial Ferrimagnetic Nonlinear Transmission Lines (NLTL) are utilized as all solid state High Power Microwave (HPM) sources with virtually fixed phase relationship between input voltage pulse and output rf wave. Since the output power of a single NLTL at a given frequency, and of a given size, is limited by the effect of line dimensions as well as charge voltage on operating frequency, we employ an NLTL array to achieve higher power levels. To operate in microwave generation mode the ferrite domains are aligned along the axial direction of the NLTL using a biasing magnetic field generated by a dc current driven solenoid. This biasing method also provides a means of controlling the line delay by adjusting the magnetization of the ferrites using an auxiliary delay coil. By utilizing the delay coil and the primary biasing coil together on an NLTL, the operating frequency can be locked at a given frequency while the delay of the line can be independently adjusted. The relationship between the delay coil field and the corresponding temporal delay is investigated, and two NLTLs fed by identical voltage pulse shapes are combined using the phase control method described.

Authors: D. V. Reale; J.-W. B. Bragg; S. L. Holt; A. A. Neuber; J. J. Mankowski; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6627412

Abstract: Coaxial Ferrimagnetic Nonlinear Transmission Lines (NLTL) can be utilized as High Power Microwave (HPM) sources. The output power of a single NLTL at a given frequency, and of a given size, is limited by the effect of line dimensions and charge voltage on operating frequency. Therefore, to achieve higher power levels, an array is required. To operate in microwave generation mode the ferrite domains are aligned along the axial direction of the NLTL using a biasing magnetic field generated by a dc current driven solenoid. This biasing method also provides a means of controlling the line delay by adjusting the magnetization of the ferrites using an auxiliary delay coil. By utilizing the delay coil and the primary biasing coil together on an NLTL, the operating frequency can be locked at a given frequency while the delay of the line can be independently adjusted. The relationship between the delay coil field and the corresponding temporal delay is investigated and two NLTLs are combined using the phase control method described.

Authors: C. F. Lynn; J. Parson; P. Kelly; M. Taylor; J. Mankowski; J. Dickens; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635085

Abstract: Summary form only given. Vircators are high power microwave sources which operate at frequencies approximately proportional to the plasma frequency of the electron beam as it passes through the anode. As a result, stable frequency operation requires flat top voltage pulses, and diodes which exhibit little variation in impedance over a given pulse duration. However, plasma formation on the anode and cathode followed by plasma expansion causes the effective A-K gap and thus the impedance of the diode to collapse with time. As the impedance collapses current density increases, possibly causing a shift to higher frequency operation. The vircator presently discussed exhibits frequency chirping from 4 to 5.5 GHz during a pulse length of 700 ns when operated with an 8 stage, 1.5 kJ Marx generator that has a pulse shape of an overdamped RLC discharge (assuming arc formation does not occur). However, stable frequency operation is observed with an 8 stage 80 J Marx, each stage constructed from pulse forming networks rather than discrete capacitors. The PFN Marx generator produces a flat top voltage for a 75 ns FWHM pulse; however the energy radiated from the vircator is much lower due to the reduced energy storage of the PFN Marx. Burst mode operation has the potential to improve frequency stability and energy radiated from the vircator. This contribution reports the experimental results of a compact sealed tube vircator operated in burst mode with a repetition rate of up to 100 Hz. The data show voltage and current waveforms during burst cycles as well as microwave waveforms with temporally resolved frequency spectra.

Authors: Patrick Kelly; Curtis Lynn; Jonathan Parson; Mark Taylor; John Mankowski; Andreas Neuber; James Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6634857

Abstract: The electrical characteristics and current capabilities of a low inductance Marx generator used to drive a high power microwave (HPM) source are described. The operating frequency of many HPM sources is dependent upon various characteristics of the input voltage pulse from the pulsed power generator. Texas Tech University has developed an 8 stage, 80 J Marx generator used to drive HPM sources at repetition rates up to 500 Hz. The Marx generator is constructed using pulse forming networks rather than discrete capacitors. Experimental efforts have shown stable frequency operation of a virtual cathode oscillator (Vircator) source using a low energy, PFN Marx. Efforts were taken to reduce parasitic inductance of the PFN Marx resulting in a voltage pulse with a risetime less than 25ns with 50ns FWHM. Additionally, efforts were taken to improve impedance matching between the PFN Marx and the vircator load. This article presents experimental results of an 80 J Marx at voltages ranging from 150 kV to 250 kV operating in burst mode. Voltage and current waveforms from the Marx are shown as well as some typical results of compact sealed tube vircator operation.

Authors: Shelby Lacouture; Kevin Lawson; Stephen Bayne; Michael Giesselmann; Heather O'Brien; Aderinto Ogunniyi; Charles J. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6627575

Abstract: An advanced evaluation system for experimental high power Super Gate Turn Off Thyristors (SGTOs) with built - in custom data acquisition and characterization electronics was designed and built in a cooperative agreement between engineers at Texas Tech University's Center for Pulsed Power and Power electronics (P3 E) laboratory and research scientists at Army Research Lab (ARL). The system consist of a Pulse Forming Network (PFN) energized by a rapid capacitor charger, a data acquisition system which records chosen waveforms for each test cycle and a curve tracing module which the test devices are mechanically switched into to record current and voltage characteristics at arbitrary intervals between high power cycles. Testing is completely automated, with all test parameters including charge level, repetition rate, volume, etc. set within a windows based GUI. The evaluation system has successfully recorded changing I - V characteristics before actual physical failure in several devices. Extremely high volume testing has also been carried out with one device having been cycled over 42,000 times at moderate (2.5 kA) conduction levels.

Authors: K. Lawson; S. Lacouture; S. Bayne; M. Giesselmann; H. O'Brien; A. Ogunniyi; C. J. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6634860

Abstract: An advanced evaluation system for experimental high power silicon (Si) and silicon carbide (SiC) Super Gate Turn Off Thyristors (SGTOs) with custom data acquisition and characterization electronics was designed and built in a cooperative agreement between engineers at Texas Tech University's (TTU) Center for Pulsed Power and Power Electronics (P3E) laboratory and research scientists at the U.S. Army Research Lab (ARL). The system consists of a 4.4 kJ Pulse Forming Network (PFN) energized by a 10 kW rapid capacitor charger developed at TTU, a data acquisition system which records chosen waveforms for each test cycle and a curve tracing module which the test devices are mechanically switched into to record current and voltage characteristics at arbitrary intervals between high power cycles. Testing is completely automated, with all test parameters including charge level, repetition rate, volume, etc. set within a custom windows based GUI. The evaluation system has successfully recorded changing I-V characteristics before actual physical failure in several devices. Extremely high volume testing has also been carried out with one device having been cycled over 30,000 times at moderate (2.5 kA) conduction levels. This paper expands on work presented at the 2012 Power Modulator Conference and presents newly acquired data and modifications.

Authors: J. Stephens; B. Loya; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6634980

Abstract: Summary form only given. A high side, high voltage pulser is used to excite a micro hollow cathode discharge (MHCD) geometry with short pulsed voltages adjustable on the order of 100-500 ns, at repetition rates from 10-200 kHz. The goal of this research is to develop an intense, efficient UV/VUV radiation source capable of generating photons of sufficient energy to affect direct photo-ionization in a gaseous volume rather than photoemission from surfaces. The pulser used in this study has experimentally demonstrated to be capable of efficiencies on the order of 80% when delivering power to a 1 KΩ resistive load at 200 kHz repetition rate. The final design is to be compact, and robust, capable of long term rep-rated operation with minimal electrode erosion and dielectric degradation. Present designs involve laser machining a direct bonded copper (DBC) substrate consisting of Cu-Al2O3-Cu. Transient ignition of the MHCD, voltage and current waveforms, and spectral emission are all experimentally studied. A transient, 2-dimensional, cylindrical coordinate (r-z) fluid model is used for the theoretical treatment of the microdischarge. Using known rate coefficients for argon, time and space resolved behavior of discharge under pulsed conditions is presented.

Authors: B. N. Pushpakaran; S. B. Bayne; A. A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6627596

Abstract: This research illustrates the transient performance of N-channel silicon carbide (4H-SiC) power MOSFET rated for a blocking voltage of 1200V and drain current density of 100A/cm2. The simulation of vertical D-MOSFET half cell structure was performed at room temperature of 300K. The 2D device model was created and simulated using Silvaco© ATLAS Technology Computer-Aided Design (TCAD) physics based simulation software. Physics based models were used to accurately model electrical device parameters including carrier mobility, recombination effects, bandgap narrowing, impact ionization and lattice heating.

Authors: Mark Taylor; Patrick Kelly; Curtis Lynn; Jonathan Parson; John Mankowski; Andreas Neuber; James Dickens; Magne Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6635087

Abstract: Summary form only given. A virtual cathode oscillator (vircator) is a high power microwave device that exhibits frequency tunability which is inherent to its principle of operation. Two types of electron motion within the vircator generate microwaves (virtual cathode oscillation and reflexing electron motion). Although it is difficult to precisely predict the dominant microwave frequency of a vircator design prior to experimental observation, the oscillation frequency of the virtual cathode is approximately proportional to the plasma frequency of the electron beam as it enters the virtual cathode. Additionally, the reflexing electrons oscillate at a frequency which is inversely proportional to approximately four times the transit time of an electron through the anode-cathode gap (A-K gap). As a result, assuming space charge limited diode operation, the virtual cathode and reflexing electron oscillation frequencies, though different, are proportional to V1/2/d, where d is the gap separation of the A-K gap and V is the accelerating voltage applied to the gap. Thus vircators are tunable via adjusting the A-K gap and varying the applied accelerating voltage. Texas Tech University has developed a sealed tube vircator which radiates approximately 100 MW with an operational frequency of 4GHz. Operating at 4GHz the diode has an A-K gap of 8mm, an accelerating voltage of ~200 k V, and electron beam current of ~3.5kA. Here we present an experimental investigation of the tunability of a reflex triode vircator by varying the A-K gap from 4 mm to 10 mm and accelerating voltages ranging from 150 kV to 250 kV.

Authors: L. Cheng; A. K. Agarwal; M. Schupbach; D. A. Gajewski; D. J. Lichtenwalner; V. Pala; S. -H. Ryu; J. Richmond; J. W. Palmour; W. Ray; J. Schrock; A. Bilbao; S. Bayne; A. Lelis; C. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6694395

Abstract: In this paper, we report our recently developed 2^nd Generation, large-area (56 mm² with an active conducting area of 40 mm²) 4H-SiC DMOSFET, which can reliably block 1600 V with very low leakage current under a gate-bias (V_G) of 0 V at temperatures up to 200°C. The device also exhibits a low on-resistance (R_ON) of 12.4 mΩ at 150 A and V_G of 20 V. DC and dynamic switching characteristics of the SiC DMOSFET have also been compared with a commercially available 1200 V/ 200 A rated Si trench gate IGBT. The switching energy of the SiC DMOSFET at 600 V input voltage bus is > 4X lower than that of the Si IGBT at room-temperature and > 7X lower at 150°C. A comprehensive study on intrinsic reliability of this 2^nd generation SiC MOSFET has been performed to build consumer confidence and to achieve broad market adoption of this disruptive power switch technology.

Authors: David Reale; J.-W. Braxton Bragg; Richard Solarski; Andreas Neuber; Shad Holt; John Mankowski; James Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6633210

Abstract: Previously developed coaxial ferromagnetic Nonlinear Transmission Lines (NLTL) relied solely on pressurized Sulfur Hexaflouride (SF6) as high-voltage insulating dielectric medium [1]. While the use of SF6 provides the necessary electric insulation, there are drawbacks including gas storage and pressure fittings that increase system size and add to the design complexity of the NLTLs themselves. We evaluate solid dielectric materials as an alternative. Initial attempts used a standard high voltage (HV) epoxy to pot the NLTL assembly. This method was effective at producing magnetic precession in the NLTL; however, there was a reduction in output power due to the high loss tangent of the epoxy. Sylgard 184, commonly used in solar cells, has been widely used as an HV potting material and has an order of magnitude lower loss tangent than standard HV epoxy at 1 kHz. Samples of HV epoxy and Sylgard 184 are tested in a microwave cavity resonator at 3 GHz to determine their loss tangents. The performance of an NLTL potted with Sylgard 184 is compared to that of the HV epoxy NLTL and the current SF6 insulated NLTL.

Authors: R. C. Solarski; D. V. Reale; J.-W. B. Bragg; A. A. Neuber; S. L. Holt; J. J. Mankowski; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6627413

Abstract: At the Center for Pulsed Power and Power Electronics, previous coaxial ferrimagnetic Nonlinear Transmission Lines (NLTL) relied solely on pressurized Sulfur Hexafluoride (SF6) as high-voltage insulating dielectric medium [1]. While the use of SF6 provides the necessary electric insulation, there are drawbacks including gas storage and pressure fittings that increase system size and add to the design complexity of the NLTLs themselves. Hence it was deemed necessary to evaluate solid dielectric materials as an alternative. Initial attempts utilized a standard high voltage (HV) epoxy to pot the NLTL assembly. This method was effective at producing magnetic precession in the NLTL; however, there was a reduction in output power due to the high loss tangent of the epoxy. Sylgard 184, commonly used in solar cells, has also found use as an HV potting material. Per datasheet, its loss tangent is an order of magnitude lower compared to standard HV epoxy at 1 kHz. Samples of HV epoxy and Sylgard 184 were tested in a microwave cavity resonator at 3GHz, which yielded their respective loss tangents. The performance of an NLTL potted with Sylgard 184 is compared to that of the HV epoxy NLTL and the earlier SF6 insulated NLTL.

Authors: J. Stephens; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635217

Abstract: Exploding wires (EWs) are subjected to high current densities of the order of 107 A/cm2 inducing metal vaporization and plasma formation on a microsecond timescale. Single strand EWs, silver and copper, are tested in gaseous media from atmospheric pressure to 790 kPa. To theoretically predict EW behavior, one-dimensional, radially directed cylindrical, Lagrangian coordinate hydrodynamic (HD) and magnetohydrodynamic (MHD) models are applied. Such models require accurate material equation-of-state (EOS) and electrical conductivity data throughout the temperature density range experimentally achieved (ρ=0.1-10 gm/cm3 and T=300-20,000 K). In this study, the Lee-More-Desjarlais (LMD) conductivity, and its quantum molecular dynamic modification (QLMD) are used. The Los Alamos National Laboratory SESAME database is used to provide the EOS parameters. When utilized as an opening switch the metal plasma is exposed to higher electric fields, atypical to traditional exploding wire experiments. Recent studies have shown that the behavior of the strongly coupled plasma in such conditions is reasonably well modeled assuming a semi-empirical electron impact ionization process. The HD and MHD based models are benchmarked against experimental data to confirm their accuracy for predicting the behavior of EWs in an opening switch type operation.

Authors: J.-W. Braxton Bragg; William W. Sullivan; David V. Reale; Daniel L. Mauch; Shad Holt; Andreas A. Neuber; John Mankowski; James C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6633211

Abstract: An all solid-state high power microwave system comprised of a photoconductive semiconductor switch (PCSS) and nonlinear transmission line (NLTL) is presented. A single 50 kV 4H-Silicon Carbide (SiC) PCSS switches 7 ns (FWHM) pulses with 2 ns risetime into the gyromagnetic NLTL. The PCSS achieves sub-ohm resistance when illuminated by approximately 3 mJ of laser energy emitted from a tripled Nd:YAG laser (355 nm). Utilizing a fiber optic based optical triggering system enabled generating a train of laser pulses, and burst-mode operation with 65 MHz pulse repetition frequency was achieved. The NLTL sharpens the fast rising pulse from the PCSS and produces MW-class rf power levels at 2-4 GHz. Microwave generation is achieved through stimulation of damped gyromagnetic precession when the static and pulsed external magnetic fields interact with the magnetic moments of the nonlinear ferromagnetic material. The acting nonlinear material is a NiZn composite with dimensions 3 mm × 6 mm (ID × OD). The system is terminated with a wideband antenna consisting of a coaxial to parallel-plate conversion balun which is then immediately flared to provide a linearly-flared TEM horn antenna. The entire length of the compact HPM system measures 1.15 m. Radiation results for single shot and burst-mode operation are presented as well as demonstration of operational frequency control.

Authors: T. J. Rosson; S. Bayne; R. Gale

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6663374

Abstract: Traditional multi-cell battery packs use a fixed configuration to connect multiple individual battery cells in fixed configurations to achieve pre-determined voltage and current. Even with modern advances in battery chemistry and greater power density, this fixed configuration results in low reliability, low fault tolerance, and non-optimal energy conversion efficiency. This system of manufacture is hindering advances in battery performance. This paper proposes a novel scheme to manufacture batteries by individualizing cells. By changing the packaging and adding low cost circuitry, a single cell in a battery pack will be able to control its orientation with other cells in a system. The idea has potential to revolutionize the way in which battery systems are developed and controlled. By using Texas Instruments value line MSP430 microcontrollers, the cost of the system can be minimized. The use of these low cost microcontrollers makes it possible to monitor voltage and temperature levels on individual cells. Also, by utilizing the same microcontroller, the individual cell is capable of shifting individual battery cells from off state, to a positive or negative polarity. The ability to switch single cells will make it possible to create a multi-level DC output. This independent switching scheme will also make it possible to create a modified sine wave AC output. The possibilities of switching to negative polarity will double the peak-to-peak voltage of the AC waveform. The battery will also be capable of identifying and implementing only the cells that are in optimal condition. Using cells at optimal temperature and state of charge will prolong the life of each cell. A battery system that individualizes each cell into a replaceable package makes replacement of "dead cells" possible. These are the cells that currently cause traditional battery packs to be replaced. With such systems available, not only will efficiency of large multi-cell battery systems increase, but with replaceable cells the cost of ownership of a complete battery pack will decrease over the lifetime of the system.

Authors: D. L. Mauch; W. W. Sullivan; A. B. Bullick; A. A. Neuber; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635034

Abstract: A 20 kV, lateral geometry, contact face illuminated, silicon carbide (SiC) photoconductive semiconductor switch (PCSS) is presented. The SiC PCSS was fabricated from high purity semi-insulating, bulk 4H-SiC (12.7 mm × 12.7 mm × 0.35 mm), in a lateral geometry, (0.6 mm gap) with both the anode and cathode contacts located on the same face of the device. The device was illuminated with light from a tripled Nd:YAG laser (355 nm-7 ns FWHM) entering from the contact face. The device demonstrated sub-ohm on-state resistance for laser pulse energies in the mJ range, and micro-amp leakage currents at 20 kVdc in the off-state. Voltage hold-off and low leakage currents in the off state were achieved through high energy electron beam irradiation of the bulk material. The switch's geometry and packaging are discussed, along with experimental switching and blocking characteristics.

Authors: P. Kelly; J. M. Parson; C. F. Lynn; J. J. Mankowski; J. C. Dickens; A. A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635086

Abstract: Summary form only given. This study focuses on the use of non-evaporable getter (NEG) materials in a high power microwave (HPM) virtual-cathode oscillator (vircator). The vircator has a background pressure in the low 10-9 torr range with a vacuum volume of 5 L backed by a small 20 L/s ion pump. Presently, the source is driven by an 80 J, 300 kV Marx generator with a current density of 225 A/cm2 and pulse width of 100 ns. Initial current pulses increases the chamber pressure to the low 10-6 torr range desorbing approximately 1014 particles1,2. Reduction in the number of particles desorbed decreases an order of magnitude after approximately 2,000 current pulses, with no further noticeable decrease after 10,000 pulses3. The HPM system will eventually be modified to operate at 500 Hz with a pulse width between 10-20 ns. In first order approximations, a pumping speed of 2500 L/s will be needed to pump 1014 particles within 2 ms from a 5 L volume. This corresponds to a surface area of approximately 2400 cm2 for a zirconium-vanadium-iron (ZrVFe) getter material. An introduction of NEG materials into the vacuum chamber will aid in achieving repetitive operation; and preliminary results on the use of the NEG material in the vacuum environment will be presented. A residual gas analyzer (RGA) and an inverted-magnetron cold cathode vacuum sensor are used for the gas diagnostics and are rated to survive the 300°C, 72 hour vircator bake out process.

Authors: Sterling R. Beeson; James C. Dickens; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6635055

Abstract: Summary form only given. We study the recovery of the low temperature plasma generated by a microsecond high power microwave, HPM, pulse in different gases at pressures approaching one atmosphere. The experiment mimics the electrical breakdown at the interface between the vacuum HPM-source environment and the radiating environment (generally held at atmospheric pressure). This low temperature plasma is generated along the surface on the atmospheric side where the high Eeff/p, or reduced effective electric field with pressure, causes high levels of ionization leading to the formation of a highly conductive plasma within the duration of the microsecond pulse. With the maximum HPM pulse repetition rate being highly dependent on the relaxation of this plasma, it is the goal of this study to determine the kinematics and chemistry of this low temperature plasma as it pertains to ion, electron, and excited species densities. For this study, a 2.85 GHz magnetron operating in the TE10 mode of a WR-284 standard waveguide is used to generate a 3 MW, 3 microsecond pulse with a rise time of less than 50 ns that is incident on the dielectric window separating the two environments. The atmospheric side is enclosed in a structure that mimics an open radiation pattern similar to that of current HPM systems. To understand the post-pulse features, two custom multi-standard waveguide couplers were designed to implement and extract a low power 10 GHz CW source into the main waveguide structure while keeping a low insertion loss for the HPM pulse. The results of this power signal (max attenuation values range -10 to -40 dB) along with a 1D plane wave excited plasma model is used to infer the temporal average electron density (specifically the longitudinal integral of the surface plasma density) at a range of pressures and different gases, typically 10-400 torr for air, N2, Ar, and He. The peak electron density and loss rates are then correlated with diffusion lengths, recombination and attachment rates given in literature to ascertain the dominant plasma relaxation path and species along with a means to extrapolate the time required to relax to a nominal background electron density. For instance, the dominant electron loss process in 90 torr air is attachment with a frequency of 121 kHz and peak electron density of ~1013 cm-3 resulting in relaxation times of a few hundred microseconds while in N2, the dominant process soon after the pulse is determined to be 2-body dissociative recombination.

Authors: A. Bilbao; W. B. Ray; J. A. Schrock; K. Lawson; S. B. Bayne; L. Cheng; A. K. Agarwal; C. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6627636

Abstract: The purpose of this research is to characterize and compare CREE's new N-Channel Silicon Carbide (4H-SiC) vertical power D-MOSFET with CREE's previous generation of N-Channel Silicon Carbide (4H-SiC) vertical power D-MOSFET. Changes made to the newest MOSFET design lead to a 400% increase in pulsed current handling capability over the previous generation device with the same active area.

Authors: S. L. Holt; J. C. Dickens; J. J. Mankowski; G. Rodriquez

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6635187

Abstract: The design of a scalable, burst-mode rapid capacitor charger and a high power battery system will be discussed. The design is scalable up to 40 kW output power and operates at 50 k V. The system is powered by a Lithium Ion Polymer (LiPo) battery pack and includes a battery management system. The topology is derived from a 5 kV, 3 kJ/s battery power capacitor charger that was previously demonstrated. This smaller charger occupied a volume of 1.5 L including batteries and demonstrated a 2 kW/L average power density over a 1 s burst. This power density is five to ten times higher than commercially available systems. The performance of this previous design was limited by strict form factor restrictions (1.5 L, 6" diameter). The new design is intended to match or exceed the power density of the previous design due to reduced form factor restrictions. Electrical and mechanical design, component selection, and circuit simulations will be presented.

Authors: Steiner, B; Bayne, SB; Veliadis, V; Ha, HC; Urciuoli, D; El-Hinnawy, N; Borodulin, P; Scozzie, C

PDF: https://www.scientific.net/MSF.740-742.921

Abstract: A necessity for the successful commercialization of SiC power devices is their long term reliability under the switching conditions encountered in the field. Normally-ON 1200 V SiC JFETs were stressed in repetitive hard-switching conditions to determine their fault handling capabilities. The switching pulses were generated from an RLC circuit, where energy initially stored in capacitors discharges through the JFET into a resistive load. The hard-switching included one million repetitive pulsed hard-switching events at 25 degrees C from a drain blocking-voltage of 600 V to an on-state current of 67 A, and an additional one million 600-V/63-A pulsed hard-switching events at 150 degrees C. The JFET conduction and blocking-voltage characteristics are virtually unchanged after over two million hard switching events proving the devices are reliable for handling high surge-current faults like those encountered in bidirectional circuit breaker applications.

Authors: A. Bullick; D. Mauch; W. Sullivan; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6633401

Abstract: High energy electron beam irradiation of bulk SiC creates crystal defects within the material which act as deep traps. These deep traps capture charge carriers and in turn increase the resistivity/blocking voltage of SiC photoconductive semiconductor switches (PCSS). Irradiation has been reported to increase the blocking field of a vertical SiC PCSS to 705 kV/cm when the entire sample region is irradiated with a 1×1018 cm-2 dose from a 1 MeV electron beam1. This paper investigates selectively irradiating specific regions of a SiC PCSS gap to characterize the effects on DC operation. Lateral switches were fabricated on a high-purity semi-insulating 4H-SiC sample and subsequently irradiated to form three sample types of differing irradiation regions. The selectively irradiated regions are: 1) mid-gap, 2) both contact-gap interfaces, and 3) a single gap-contact interface used for both anode and cathode testing configurations depending upon the polarity of the applied voltage.

Authors: He, M; Yang, L; Zhang, JS; Vittal, V

PDF: https://ieeexplore.ieee.org/document/6504247

Abstract: In this paper, we propose a spatio-temporal analysis approach for short-term forecasting of wind farm generation. Specifically, using extensive measurement data from an actual wind farm, the probability distribution and the level crossing rate (LCR) of wind farm generation are characterized by using tools from graphical learning and time-series analysis. Based on these spatial and temporal characterizations, finite state Markov chain models for wind farm generation are developed. Point-forecast of wind farm generation is derived using the Markov chains and integrated into power system economic dispatch. Numerical study on economic dispatch using the IEEE 30-bus test system demonstrates the significant improvement compared with conventional wind-speed-based forecasting methods.

Authors: He, M

PDF: https://ieeexplore.ieee.org/document/6493922

Abstract: With the increase of people's income and the improvement of their living standards, the need for sports and fitness, recreation and entertainment is also undergoing substantial increase. It is of great importance to meet these growing needs, which provides a broad market for the sports industry. It is an inevitable trend to develop the sports industry market on a large scale. It has positive effect on sports quality improving, sports level promoting and sports development accelerating so as to promote the development of sports economy. From the perspective of economy environment, this paper describes the necessity of develop sports industry on a scale and problems need to be concerned. From the development perspective, the economic environment analysis and research of sports industry market scale offer managers and relevant decision makers meaningful reference. By using the methods of documentation, questionnaire, interview and investigation, mathematical statistics and logical analysis, the author,in two aspects of theories and demonstration, studied the relationships between development in scale of sports industry and economic environment.

Authors: Andrew S. Fierro; George R. Laity; Sterling R. Beeson; James C. Dickens; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6634918

Abstract: Summary form only given. A GPU-accelerated 3-dimensional PIC/MCC simulation code was developed using the CUDA environment to study the physical processes involved in the development of a low-temperature plasma. The simulation results aid in quantifying transient plasma development as it is often inaccessible experimentally in detail even with modern noninvasive techniques such as non-linear laser spectroscopy or high-speed electrical diagnostics. Hence, computational methods, such as Particle-in-Cell (PIC) and Monte Carlo Collision (MCC), provide a complementary approach to determining the mechanisms leading to plasma development. However, fully modeling the physics of the plasma development is made difficult by the number of plasma processes that must be tracked simultaneously, and only recently have computing resources provided the capability to track tens of millions of particle interactions. Furthermore, the introduction of graphics processing unit (GPU) computing provides an attractive means for economical and efficient parallelization of scientific codes through a framework such as NVIDIA CUDA. As such, a GPU-accelerated 3-dimensional PIC/MCC simulation was developed using the CUDA environment to provide characteristics during the initial stage of plasma development in atmospheric pressure nitrogen. The simulation was run on a NVIDIA GTX 580 with 3 GB of memory and 512 CUDA cores. The simulated geometry consists of two paraboloid electrodes with a gap distance of 5 millimeters with Dirichlet boundary conditions, and 22 unique electron interactions with molecular nitrogen are considered. The electrodes are excited with a step voltage pulse of several thousand volts also assuming a uniformly distributed initial electron density of 104 cm-3 in the volume. For instance, results from a 5 nanosecond simulation reveal the development of positive ion space charge channels near the anode and cathode regions. These channels appear consistent with high-speed streamer photographs captured during plasma formation. The electron energy distribution function (EEDF) indicates a non-Maxwellian velocity distribution during the application of the high electric field. Furthermore, a typical electron density near the cathode is on the order of 7 × 108 cm-3. The results from numerical simulation will be compared in detail to experimentally accessible parameters such as electron temperature and dissociation degree.

Authors: P. J. Ford; J. Krile; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518732

Abstract: Surface flashover induced by a High Power Microwave fast rise-time pulse causes a significant drop in transmitted power, along with reflections that can damage the source. Momentum transfer collision rates in the range of 100s of GHz (for pressures exceeding 5 kPa) lead to low plasma conductivity, corresponding to absorption levels of up to 60 % of the incident power. A simulation algorithm was developed using the finite-difference time-domain (FDTD) method in order to model the growth and transport of the electron density near a dielectric surface, and the resulting interaction with the microwave pulse. The time-dependent plasma parameters are governed by empirical and simulated scaling laws for ionization and collision rates, along with diffusion coefficients; the resulting frequency-dependent plasma permittivity is transformed to a discrete algorithm to describe the spatially resolved plasma in the FDTD algorithm. A plasma thickness of up to 2 mm is simulated that compares with side-on ICCD imaging of surface flashover. Breakdown parameters, such as delay times and breakdown electric fields, in nitrogen, air and argon, are compared with experimental data on surface flashover across a polycarbonate window at atmospheric pressures; the simulated results correlate well with measured, and the model exhibits low computational complexity when simulating a pulse on the order of microseconds, making it a good alternative to standard particle-in-cell codes. The source is a S-band magnetron that produces a 2.5 MW peak power, 50 ns rise-time pulse with 3 μs duration at 2.85 GHz center frequency.

Authors: Litz, MS; Fan, ZY; Carroll, JJ; Bayne, S

PDF: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/8377/1/Alpha-Schottky-junction-energy-source/10.1117/12.918588.short?SSO=1

Abstract: Isotope batteries offer solutions for long-lived low-power sensor requirements. Alpha emitting isotopes have energy per decay 103 times that of beta emitters. Alpha particles are absorbed within 20 mu m of most materials reducing shielding mitigation. However, damage to materials from the alphas limits their practical use. A Schottky Barrier Diode (SBD) geometry is considered with an alpha emitting contact-layer on a diamond-like crystal semiconductor region. The radiation tolerance of diamond, the safety of alpha particles, combined with the internal field of the SBD is expected to generate current useful for low-power electronic devices over decades. Device design parameters and calculations of the expected current are described.

Authors: S. R. Pappu; S. Nimmagadda; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6316396

Abstract: This paper compares the characteristics of two wind farms of same power capacity having two different size generators. The research compares the performance of mid-size turbines to large turbines. The systems were modeled in MATLAB-Simulink. With rapid development in wind power generation it's necessary to study the different characteristics of wind farms having different generator configurations. A 9MW wind farm consisting of eighteen Fixed Speed Induction Generators (FSIG) with STATCOM at the terminals of the wind farm is compared to a 9MW wind farm consisting of six Doubly Fed Induction Generators (DFIG) interfaced with Power Electronics on the rotor. The models of FSIG wind power generation system and DFIG wind power generation system connected to the grid are built. An analysis on total power available for transmission, reactive power consumption at different wind speeds (cut in, rated, cut off), power loss due to loss of turbines.

Authors: J. M. Parson; J. C. Dickens; A. Neuber; J. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6383332

Abstract: Summary form only given. This presentation describes a study on gas evolution of plasma expansion in a reflex-triode virtual cathode oscillator (vircator) at ultra-high vacuum (UHV). Research has shown that explosive electron emission (EEE) processes at the cathode and ion / electron bombardment at the anode cause material erosion that produces regions of localized plasma. This plasma expansion has shown to lower gap impedance, cut off microwave performance and spoil low vacuum levels over time1. The goal of the study is to identify the gas species and their sources to better understand and limit the negative effects of plasma expansion in sealed tubes. The vacuum tube under investigation is a triode-geometry vircator with 20 cm2 cathode surface, driven by a 80 J Marx Generator with an approximate peak voltage and current, and pulse width of 200 kV and 5 kA, and 200 ns, respectively.

Authors: Hinojosa, M; Bayne, S; Veliadis, V; Urciuoli, D

PDF: https://www.scientific.net/MSF.717-720.1025

Abstract: The energy dissipation capabilities of a 4H-SiC, 1200 V, 0.1 cm(2) JFET operating in blocking mode were investigated. These devices, which are used in bidirectional solid-state circuit breaker applications, can conduct a current of 13 A in forward-conduction mode, and typically block a voltage up to 1200 V in reverse bias mode. In this document, the blocking limits of the device were pushed slightly to the point where avalanche breakdown occurs. A high voltage, short-pulse generator was designed and constructed to drive the JFET into this state and to monitor the dissipated energy. The devices were able to handle up to 18.14 mJ.

Authors: Hettler, C; Sullivan, WW; Dickens, J

PDF: https://www.scientific.net/MSF.717-720.301

Abstract: Annealing of high purity semi-insulating (HPSI) 4H-SiC is investigated as a method to improve bulk photoconductive semiconductor switches through recombination lifetime modification. Five samples of HPSI 4H-SiC were annealed at 1810 degrees C for lengths of time ranging from 3 to 300 minutes. The recombination lifetime of the unannealed and annealed samples was measured using a contactless microwave photoconductivity decay (MPCD) system. The MPCD system consists of a 35 GHz continuous microwave probe and a tripled Nd:YAG pulsed laser. The recombination lifetime was increased from 6 ns, as received, up to 185 ns by annealing for 300 minutes. To experimentally verify switch improvements, identical switches from unannealed and annealed material were fabricated and tested at low voltage. The unannealed device generated a 15 ns pulse with a 2 ns rise-time. The annealed device conducted for upwards of 300 ns with a comparable 2 ns rise-time. The increased recombination lifetime resulted in lower on-state resistance and increased energy transfer.

Authors: C. Lynn; J. Walter; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518859

Abstract: Many high power microwave devices use explosive or flashover electron emission cathodes in order to generate the electron beam and thus drive the device. These diodes are simple to operate, requiring no heater or other external systems, and are capable of generating beam currents of several kA at accelerating voltages on the order of 100s of kV. However, they generally suffer from non-uniform emission, anode heating, out-gassing, and pulse shortening due to anode and cathode plasma expansion. The ability to rep-rate such a diode is generally limited by anode heating and out-gassing which are both affected by beam uniformity. Two diodes are compared in this work. One uses a machined aluminum cathode, which is made from solid aluminum with grooves milled onto the surface. The other diode utilizes a carbon velvet cathode which is CVD coated with CsI. Time integrated scintillator images of the electron beam at the anode were taken for both the carbon velvet and aluminum cathodes. Additionally, time resolved images of the emission centers were taken for the carbon velvet cathodes. Data sets of over 1000 shots were taken with each cathode and shot to shot variation in the peak “turn-on” electric field are compared. The lifetime of the aluminum cathode was exceeded before 1000 shots, whereas the carbon velvet cathode showed no degradation in operation.

Authors: K. Lawson; S. Lacouture; S. B. Bayne; M. Giesselmann; T. Vollmer; H. O'Brien; C. Scozzie; A. Ogunniyi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518751

Abstract: In a collaborative effort between Army Research Lab (ARL) and Texas Tech University's center for Pulsed Power and Power Electronics (P3E) lab, a high power, high energy test bed meant to characterize experimental Si and SiC Super Gate Turn Off (SGTO) devices was designed and built. The system was engineered to run the devices through an arbitrary number of test cycles while recording all pertinent data automatically. Test parameters are set through a windows GUI which communicates with a microprocessor - based control system that orchestrates timing and settings of each subsystem as well as acquiring voltage and current waveforms with high speed ADCs operating simultaneously in parallel. The test waveform itself is generated by a Pulse Forming Network (PFN) which accurately controls rise time, fall time and pulse width. The PFN is charged by a Rapid Capacitor Charger (RCC) system designed at the P3E lab that is capable of 10 kW and allows precise charge voltage levels to be set. Waveforms are acquired through isolated probes specifically designed to capture desired signals even in the presence of a large bias voltages.

Authors: S. Beeson; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6383529

Abstract: Plasma generation along the dielectric interface between the vacuum medium of the source and the atmospheric environment of the antenna is one of the limiting factors in power thresholds of high power microwave, HPM, systems. The maximum repetition rate for HPM platforms are ultimately determined by the relaxation times of this low-temperature surface plasma. While the microwave scattering parameters can be determined during the HPM pulse from the transmitted and reflected pulse itself, additional diagnostics effort has to be expended to capture post-pulse plasma relaxation and microwave scattering parameters. For this, a Bethe hole-type multi-standard waveguide coupler was designed and fabricated to inject a continuous low power signal into the main waveguide structure that carries the main 3 MW, 3 µs pulse at 2.85 GHz in an S-band waveguide structure. To facilitate flashover the waveguide is terminated by a Lexan window into a controlled atmospheric chamber with absorbing walls that mimic radiation into free space. The coupler injects 1W of continuous power at 10 GHz to measure the scattering parameters many microseconds after the pulse extinguishes. Using a model developed from a 1D plane wave excitation, the plasma's electron density can be inferred from these power measurements. Then, from the temporal analysis of the electron density, the kinematic and chemical behavior of the plasma in relation to the recombination and attachment properties of the gas are inferred. Tested conditions include three gas types: N2, air, Ar, ranging in pressure from 10 to 155 torr. The transmission coefficients range from −40 to −10 dB corresponding to an electron density of ∼1014 to ∼1011 cm−3 for these pressures, respectively. The decay of this plasma has been experimentally determined to be on the order of tens of microseconds with the density falling off proportional to t−1, t−2, exponential, or a linear combination thereof, depending on the dominant electron loss path. Reported here, are the design parameters of the waveguide coupler along with the post-pulse evolution of the electron density and recombination physics that it entails.

Authors: Lacouture, S; Lawson, K; Bayne, S; Giesselmann, M; O'Brien, H; Scozzie, CJ

PDF: https://www.scientific.net/MSF.717-720.1183

Abstract: The development of new semiconductor designs requires that extensive testing be completed in order to fully understand the device's characteristics and performance capabilities. This paper describes the evaluation of experimental Silicon Carbide high power Super Gate Turn Off Thyristors (SiC SGTOs) in a unique test bed that is capable of stressing the devices with very high energy/power levels while at the same time mimicking a realistic, real world application for such devices.

Authors: S. R. Pappu; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6184582

Abstract: This paper simulates the transient response of three wind farms interconnected, rated 9MW each. Each wind farm consists of six Doubly Fed Induction Generator (DFIG) turbines rated at 1.5MW each, interfaced with power electronics on the rotor side. The three wind farms connected to the grid are built and modelled in MATLAB-Simulink. With the rapid development in Generation Interconnect Requests and development of the Hub concept, it is necessary to model the interconnection of wind farms to study the transient characteristics. The three wind farms are interconnected at a single point on the transmission line to increase efficiency and reliability. Analysis on voltage stability of wind farms due to faults on the neighbouring wind farm was done. A STATCOM was installed at the terminals of each wind farm to provide additional voltage stability and reactive power capabilities.

Authors: Sterling Beeson; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6518710

Abstract: The relaxation time of pulsed rf-generated plasma is investigated. A 3 MW, 3 μs width, 50 ns risetime HPM pulse is transmitted through a dielectric window that terminates a WR-284 waveguide filled with insulating gas. The investigated plasma is formed across the dielectric window on the atmospheric side. This produces electron densities on the order of 1013 to 1012 cm-3 for 60 to 145 torr in air, respectively. In the same pressure range, initial attenuation (~ 0.5 dB) of the microwaves is observed after tens to hundreds of nanoseconds with final attenuation values approaching -40 to -10 dB, respectively. To determine plasma relaxation times after the HPM pulse terminates a multi-standard waveguide coupler (X/S-band) was designed to inject a low power 10 GHz signal used for probing the surface plasma. The coupler was designed to have high coupling coefficients (> -5 dB) for the specific narrowband around 10 GHz (BW ~ 10 MHz) along with negligible insertion loss of the HPM propagation. From the measured attenuation and reflection of the 10 GHz probe signal, the evolution of the electron density is inferred. Furthermore, in an air environment, the density falls exponentially implying attachment as the dominate electron loss mechanism.

Authors: J. C. Stephens; A. A. Neuber; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518739

Abstract: The primary objective of the research discussed in this paper is to develop a compact electro-explosive fuse (EEF) for a flux compression generator (FCG) power conditioning system, capable of rapidly obtaining and maintaining high impedance. It was observed that significant gains in EEF performance are introduced with the application of an insulating coating to the surface of the EEF wire. A 2 kA small scale test bed has been designed to provide a single wire EEF with similar current density (~107 A/cm2), voltage gradient (~7 kV/cm), and timescale (~8 μs) as to what is seen by and EEF utilized in a HPM generating FCG system. With the small scale test bed EEF performance data was rapidly obtained at a significantly lower cost than equivalent full scale FCG experiments. A one-dimensional finite difference model coupled with the Los Alamos National Laboratory SESAME Equation-of-State database was utilized to simulate the resistive behavior of the single wire EEFs. Further, a large scale test bed, designed to provide a similar current action as to what is provided by a FCG is used to test 18 wire EEF arrays at the 40 kA level.

Authors: G. Laity; A. Fierro; L. Hatfield; A. Neuber; K. Frank

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6383426

Abstract: There has been a continued interest in utilizing streamer and spark discharges for new technologies which require low temperature plasma generation at atmospheric pressure. One key area of study is the role of photon dominated processes, such as ionization and dissociation, as a result of emission and re-absorption of vacuum UV radiation (e.g. photons with energies greater than 8 eV). This experiment consists of a triggered surface flashover event along a dielectric surface between two point-point electrodes, where spectral measurements are performed in the vacuum UV regime (i.e. 115 – 175 nm). Previous studies of air breakdown alluded to a cross-species photon absorption process between N2 and O2 molecules, and a later study revealed an impurity in the form of HI Lyman-α radiation in spark discharges in air. In an effort to understand the self-absorption of this radiation at atmospheric pressure, a basic study was performed in various N2/H2 mixtures in a controlled environment. Spectral measurements were taken via intensified CCD devices in the VUV range, photomultiplier electronics with accompanying electrical diagnostics, and external imaging with nano-second resolution. It was concluded from the experiments that significant self-absorption of HI radiation is occurring during plasma formation at electron temperatures of ∼10 eV, and detailed spectral line-fitting showed that the parameters which define absorption vary as a function of distance from the anode (i.e. where the streamers originate due to the geometry in this experiment). Therefore, rough estimates can be made of parameters such as H2 dissociation percentage (as much as 10−3 in the plasma channel) and electron density (upwards of 1019 cm−3 in the spark phase) as a function of position between the electrodes. These studies elucidate the plasma kinetics (specifically absorption) which occurs during the transition from streamer to spark discharge at atmospheric pressure.

Authors: D. Mauch; C. Hettler; W. W. Sullivan; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518752

Abstract: A fiber optic system was constructed to demonstrate high frequency operation of a silicon carbide (SiC) photoconductive semiconductor switch (PCSS). The goal was to transform a single high-energy laser pulse into a train of pulses by adding static delays into a multimode fiber bundle. The individual optical fibers comprising the fiber bundle incrementally add an additional amount of flight time to the light pulse. The end result is a train of pulses with a fixed delay between each pulse. A frequency-tripled Nd:YAG laser (10 ns FWHM) generating up to 300 mJ of light energy at 355 nm is coupled from free space into the optical fibers comprising the fiber light guide. Experimental results examining the collection and transmission efficiency, and the temporal output are presented.

Authors: Durukan, I; Ekwaro-Osire, S; Bayne, SB

PDF: https://asmedigitalcollection.asme.org/IMECE/proceedings-abstract/IMECE2011/54907/1227/356097

Abstract: Most recent grid codes require wind turbines to contribute to the recovery of frequency drops in the grid. More of the recently build wind turbines use variable speed generators. Unlike fixed speed generators, these generators do not naturally contribute to the recovery of the frequency drop since the rotor rpm is decoupled from the grid frequency. This decoupling is achieved by controller and power conditioning units. The studies reviewed in this paper focused on the design of such a controller so that the wind turbine could react to frequency drops. Another approach to responding to frequency drops is to connect an energy storage system to the DC bus of variable speed generator. Flywheels have been used as energy storage systems to fill energy gaps in several applications and can be used for frequency recovery application for wind turbines as well. The objective of this study was to demonstrate the improvement of frequency stability of wind turbines connected to electrical grids in the presence of flywheel energy storage systems (FESS). Studies reviewed show that FESS can enhance the power quality and frequency stability of wind turbines connected to an electrical grid.

Authors: Christopher Simmons; J.-W. Braxton Bragg; James Dickens; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6518853

Abstract: A nonlinear transmission line (NLTL) provides a solidstate means of generating high power, microwave pulses. The NLTLs in this study are coaxial transmission lines whose center conductor is encapsulated by ferrite beads. Operational frequency can be controlled by varying the dimensions of the ferromagnetic material, which affects azimuthal magnetic fields and material losses, or by varying the biasing field strength. This research demonstrates frequency agility of an NLTL by documenting the design, construction, and testing of three NLTLs. The NLTL will be one meter long, and to test for frequency agility, three different sizes of ferrites will be loaded onto NLTL with various biasing fields applied. Azimuthal field strengths due to an incident high voltage pulse range from 10-36 kA/m with magnetic biasing fields between 0 kA/m and 50 kA/m.

Authors: J. Parson; J. Dickens; J. Walter; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518724

Abstract: This paper presents a study on gas evolution of three different anode materials in vacuum sealed tubes under UHV conditions. The experimental apparatus consists of a high-power microwave (HPM) virtual-cathode oscillator (vircator) driven by a 200 ns, 80 J, 225 kV low-impedance Marx Generator. Plasma expansion due to explosive electron field emission has shown to lower gap impedance, spoil consistent low vacuum levels, and cut-off microwave radiation. The anode materials compared are nickel 201L (Ni201L), stainless steel 316L (SS316L) and grade-1 titanium (TiG1); with the cathode material being aluminum. The anodes were cleaned by the following method: rough polishing followed by electro-polishing, a ten minute microwave argon / 10% oxygen plasma cleaning process (ArO₂) and finally, a 72 hour in situ bake-out at 300°C. Outgassing characteristics of each anode material are presented and compared.

Authors: G. Laity; A. Neuber; A. Fierro; L. Hatfield; J. Dickens; K. Frank

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518678

Abstract: This paper describes recent advances in the study of self-generated emission of vacuum ultraviolet (VUV) radiation which is produced during the early time period leading to high voltage breakdown at atmospheric pressure. Previous studies of air breakdown showed the presence of 121.5 nm radiation which is spontaneously emitted by excited hydrogen atoms, HI. Since this Lyman-α line is self-absorbed, it enabled inferring various plasma parameters from recording emission spectra from 115-135 nm for species of HI and NI. For instance, measurements in H2/N2 mixtures have revealed that the highest amount of absorption via HI atoms occurs in the high field region near the anode, implying that significant H2 dissociation for radiation-trapping is occurring in this zone. Selective spatial measurements further showed that the apparent VUV emission centers (i.e. streamer heads) move away from the anode and the mechanisms leading to line broadening (i.e. Stark effect from space charge) are a function of streamer position. The presented self-absorption mechanisms are essential in quantitatively understanding the role of VUV radiation transport via absorption for photo-ionization during streamer breakdown, in which re-absorption of high energy photons is inherently a requirement.

Authors: A. Fierro; G. Laity; A. Neuber; L. Hatfield

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6383314

Abstract: High energy photons released during the early stages of pulsed surface flashover are believed to be a contributing factor to the overall plasma formation. It is clear that only radiation with wavelength in the Ultraviolet (UV) and Vacuum Ultraviolet (VUV) regimes is capable of ionization and excitation processes of atmospheric gases, thus potentially providing seed electrons for secondary electron avalanches. To investigate the role of VUV photons in the breakdown process, an experiment was designed enabling detection of photons with energies greater than 7 eV with high temporal and spatial resolution. A transient low temperature plasma is generated on an MgF2 surface between two point-to-point stainless steel electrodes using a 200 ns rise-time high voltage pulser. Light emitted during the plasma formation is transmitted through the VUV transmissive MgF2 window into vacuum avoiding the strong absorption of VUV radiation in the atmosphere. High resolution measurements of the flashover event are acquired using a VUV-sensitive ICCD camera and photomultiplier tube (PMT) mounted to the exit ports of the VUV spectrograph. Detailed analysis of the two electrode regions clearly shows the presence of VUV emission lines prior to voltage collapse in atmospheric gases. Although a symmetric field geometry was chosen, it was seen in all cases that initial VUV emission originates from the anode. Furthermore, measurements of the 120 nm nitrogen ground transition taken at locations away from the anode demonstrated peak emission that coincided with streamer head locations. The observed behavior is consistent with a build-up of positive ion space charge in the gap attracting electrons from the surrounding medium causing excitations and ionizations in this high field region. Further time-resolved spectroscopy of the developing plasma in a high purity nitrogen environment captured the presence of the second positive system from molecular nitrogen as well as atomic emission lines. It was seen that the lifetime of the second positive system is much shorter (nanoseconds) than that of the atomic nitrogen emission lines (microseconds). The experimentally observed behavior will be discussed is it relates to the transient formation of a low temperature plasma along a dielectric surface.

Authors: C. Hettler; W. W. Sullivan; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518682

Abstract: A 50 kV silicon carbide photoconductive semiconductor switch (PCSS) is presented. The SiC PCSS device is fabricated from semi-insulating 4H-SiC in a newly-proposed rear-illuminated, radial switch structure. The improved structure reduces the peak electric field within the switch, extending the blocking voltage to over 50 kVdc. Electrostatic field simulations of the PCSS are presented along with experimental blocking curves. The PCSS demonstrated low on-state resistance, delivering over 27 MW of peak power into a 31 Ω load. Device modeling was performed to further optimize the switch for peak efficiency when illuminated with 355 nm light, a common laser wavelength. The switch structure was modified for peak operation at 355 nm and the experimental and theoretical results are compared.

Authors: S. Pappu; A. Rahnama; M. Tovar; S. Bayne; B. Little; S. Friend; M. Borhani

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6200953

Abstract: This paper discusses the set up and analyses of Power Quality using a Phasor Measurement Unit (PMU) located at a Semiconductor Foundry Fabrication Facility (sensitive load). The paper demonstrates the setting up of an SEL 421 PMU at the Facility (software and hardware), data archiving and real time waveform capture of voltages, currents and frequency. A code was developed for filtering the data collected from the PMU in MATLAB. A GUI was developed in MATLAB. The GUI analyses the data from the PMU and highlights aberrations in the values of Voltages and Frequency from the Power Quality limits for sensitive loads. The paper evaluates a unique method for analysis of Power Quality before the installation of a wind turbine directly feeding the Facility so as to set a standard after the wind turbine is installed.

Authors: A. Bilbao; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518860

Abstract: The goal of this research is to develop device models for Silicon Carbide (SiC) MOSFETs. Parameters are extracted and used to create PSPICE models that can be utilized for circuit simulation. Two silicon carbide power MOSFETs made available by CREE Semiconductor are considered. The first silicon carbide power MOSFET tested is the CMF20120A64410. This MOSFET features a 1200V drain-to-source breakdown voltage and 30A continuous current capacity. The second device tested is an experimental MOSFET that is still not available in the market as of the date of this paper. The experimental MOSFET features a 1200V drain-to-source breakdown voltage and 80A continuous current capability. Custom made circuits are developed for extracting some of the parameters. In some cases where the tests only require low drain current, a HP B1505A curve tracer is used to aid the development of the model. The effect of temperature over the gate threshold voltage is also investigated. By externally increasing and monitoring the die temperature of the SiC MOSFETs, new device parameters can be extracted and modeled. Once the parameters are extracted they are converted into a PSPICE model. The model is tested and compared to the real device to verify accuracy. This is achieved using custom switching circuits with both inductive and resistive loads and software suites like MATLAB.

Authors: T. T. Vollmer; M. G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518827

Abstract: The RCC (rapid capacitor charger) previously developed at the P3E Center [1] has been adapted to charge an 1800 J PFN (pulse forming network) for rep-rated operation. The entire automated system to test and evaluate SGTOs (Super Gate turn-off Thyristors) runs at a 1 Hertz repetition rate; thus requiring a power supply to charge the PFN within 500 ms and have a 3.6 kJ/s average power capability to allow for data acquisition and storage between shots. The hard-switching H-bridge topology with 10 kW burst mode handling capability is very well suited for this compact table top system design. The control of the RCC has been shifted to a PIC controller responsible for PFN charging. Charging parameters include: an adjustable charging time from 50 to 500 ms, high voltage monitoring with adjustable voltage level, and RCC Go/shut-off. All charging parameters are determined by the main CPU handling the automation process and are sent to the PIC controller before each PFN charging event. With the addition of forced air cooled heat-sink for the IGBT modules, enough heat can be removed to allow continuous automated operation.

Authors: Lawson, K; Alvarez, G; Bayne, SB; Veliadis, V; Ha, HC; Urciuoli, D; Scozzie, C

PDF: https://www.scientific.net/MSF.717-720.1021

Abstract: A necessity for the successful commercialization of SiC power devices is their long term reliability under the switching conditions encountered in application. Normally-ON 1200 V SiC JFETs were stressed in hard-switching conditions to determine their fault handling capabilities. The hard-switching included single shot tests ranging from drain voltages of 100 V to 500 V and repetition rate tests at 1 Hz, 5 Hz, 10 Hz, and 100 Hz with peak currents exceeding 100 A (8 times the rated current at 250 W/cm(2)). The JFET conduction and blocking-voltage characteristics are unchanged after 16,000 pulsed and numerous single shot hard switching events proving the devices are reliable for handling high surge-current faults.

Authors: J.-W. Braxton Bragg; Christopher Simmons; James C. Dickens; Andreas A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6518721

Abstract: Nonlinear transmission lines (NLTLs) utilizing ferrimagnetic materials for microwave generation have been realized as a possible solid-state replacement to traditional high power microwave (HPM) sources. The nonlinearities present in the material, along with interaction between pulsed, azimuthal magnetic fields and static, axial-biasing magnetic fields provide microwave (mesoband) generation with peak powers exceeding 30 MW at 2-4 GHz center frequency with 25 kV incident pulse magnitude. Additionally, an incident pulse of several nanoseconds is sharpened to hundreds of picoseconds. This study focuses on a serial arrangement of two NLTLs with 5 ns electrical length separation. Tests with 25 kV incident voltage are performed with varying bias schemes for each NLTL structure. The lines are terminated into a 50 Ω matched load. Measurements taken before and after each NLTL provide insight to the behavior of the traveling pulse. Results regarding peak power, frequency of operation, and system delay are discussed.

Authors: Sullivan, W; Hettler, C; Dickens, J

PDF: https://www.scientific.net/MSF.717-720.813

Abstract: This paper investigates n-type doping of point-defect compensated high purity semi-insulating (HPSI) 4H-SiC using a pulsed laser (10 ns FWHM @ 260 nm) for the introduction of nitrogen to shallow depths. A thermal model is presented using COMSOL Multiphysics featuring nonlinear temperature dependent material properties and a volumetric heat source term that takes into account the laser absorption depth for common ultraviolet irradiating wavelengths. The temperature distribution in the material and the amount of time that the surface and near-surface regions are at high temperature determines how many laser pulses are required to dope to the desired depth, and simulation results are presented and fit to measured data. The simulations and measured data show that for shallow doping a short wavelength ultraviolet laser should be used to localize the heat at the surface so the dopant can't diffuse deep into the material. The laser enhanced diffusion process has been used to incorporate nitrogen into HPSI 4H-SiC with a measured surface concentration greater than 10(20) cm(-3) and a nonlinear thermal model was built.

Authors: J. C. Stephens; A. A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6781418

Abstract: An exploding wire model that accounts for the electric field enhanced conductivity of dense metal plasma is applied to simulate an exploding wire opening switch. In contrast to many z-pinch experiments, operated in vacuum, the experiments here discuss wires vaporized in a high pressure gas environment. In addition to this, these experiments are primarily concerned with sub-eV temperatures, with a specific emphasis on the liquid-vapor phase transition, where significant decreases in conductivity provide the opening switch behavior. It is common that fuses operating within this regime are analyzed using 0-dimensional models, where the resistance is taken to be an experimentally determined function of energy or action. A more accurate 1-dimensional model with added field enhanced conductivity has been developed to better model the fuse dynamics throughout a significantly larger parameter range. The model applies the LANL SESAME database for the equation-of-state, and the conductivity data developed with the Lee-More-Desjarlais (LMD) algorithm. Using conductivity based on conditions of thermal equilibrium accurately predicts fuse opening as well as current re-emergence after a few microseconds dwell time for the case of small electric fields, however, this simple approach fails to capture early fuse restrike if the differential voltage across the wire becomes too large (~few kV/cm for the investigated conditions). It is demonstrated that adding an electric field driven conductivity term to the model will accurately capture the fuse dynamics for the low field as well as the high field case.

Authors: Andrew Fierro; George Laity; Andreas Neuber; Lynn Hatfield; James Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6518677

Abstract: The development of streamers during the initial stage of a pulsed atmospheric discharge, which carries a high content of vacuum UV (VUV) emission, is investigated. Due to the high spectral absorptivity of atmospheric air in the VUV regime, few experiments have been conducted that observe the wavelength range shorter than 180 nm. However, direct photoionization is believed to play an important role in streamer formation during this phase of breakdown. VUV radiation (hν > 7 eV) is energetic enough to promote step-ionization and directly ionize background gas mixtures. Utilizing a VUV-sensitive experimental apparatus, spatially-resolved photomultiplier tube (PMT) measurements were recorded showing that initial VUV emission is dependent on the inhomogeneous field distribution near the electrodes. It was revealed that further into the developing streamer the position of instantaneous VUV emission is propagating from anode to cathode away from the initial VUV emissions. From these observations, the low temperature plasma is capable of producing high energy photons during the streamer phase as well as after voltage collapse.

Authors: A. Neuber; J. Stephens; C. Lynn; J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6383930

Abstract: Summary form only given. The challenges in developing and designing a 5 GW stand-alone pulsed power generator for driving a >; 100 MW high power microwave, HPM, source are discussed. Both, applied physics aspects of operating principles and limits of the individual sub-systems will be addressed. The energy, initially provided through a set of lithium-ion batteries internal to the generator, is boosted by an explosively driven flux compressor, FC, and inductively stored at a several kJ level. While the energy is stored on a microsecond timescale, it is, however, released into the load on a nanosecond timescale via an electro-explosive fuse, EEF, thus delivering GWs of electrical power to the HPM source for a duration of about 100 ns.

Authors: D. V. Reale; J. Mankowski; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6518716

Abstract: Commercially available Electro-Explosive Devices (EEDs), such as blasting caps, use electrical current to initiate a primary charge. Various detonators including bridge wire, match-type, exploding bridge wire, and slapper. The basic operating principle of the match-type device is to heat the ignition element to the ignition temperature of the primary explosive. The normal operation current profiles, both constant current and pulsed excitation, are well known, as is the ignition temperature. However, as safety and reliability are of great concern, both in the operation and storage of EEDs, the susceptibility of these devices to transient or spurious fields is of interest. The susceptibility of match-type EEDs to high pulsed electric fields is examined. A Finite Element Method (FEM) simulation is performed using COMSOL to determine the induced current in the bridgewire due to applied electric fields and the resulting Joule heating of the wire. Several situations are investigated including the EED in conductive and non-conductive media and leads open or terminated representing operational and storage conditions.

Authors: Sullivan, W; Hettler, C; Dickens, J

PDF: https://ieeexplore.ieee.org/document/6518684

Abstract: Forming non-rectifying (ohmic) contacts to wide band gap semiconductors such as silicon carbide (SiC) requires a heavily doped subsurface layer to reduce the Schottky barrier height and allow efficient electron injection. Nitrogen, a common n-type dopant in SiC, was incorporated into a SiC sample using a laser enhanced diffusion process in which an impurity is incorporated into the semiconductor to very high surface concentrations (> 10(20) cm(-3)) and very shallow depths (< 200 nm) with the use of a pulsed 266 nm laser. This paper evaluates the effects of nitrogen introduced through laser enhanced diffusion on the contact formation and the efficiency of silicon carbide photoconductive switches at low and high injection levels under different biasing conditions. Nine lateral switches were fabricated on a high-purity semi-insulating 4H-SiC sample; three with no sub-contact doping, three with sub-contact doping on only one contact, and three with sub-contact doping on both contacts. Results are presented for tests under pulsed laser illumination with sub-contact doping on only the anode, only the cathode, neither, and on both of the contacts.

Authors: R. Karhi; D. Wetz; J. Mankowski; M. Giesselmann; I. El-Dana

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191671

Abstract: The design and experimental results of a 40-stage distributed energy store (DES) plasma arc railgun are presented. The railgun drives a free running hypervelocity plasma arc, one that is not pushing a payload, to velocities in excess of 10 km/s. These high velocities are of interest as they are required to successfully launch payloads into low earth orbit (LEO). The ability to launch payloads into LEO using a hypervelocity electromagnetic launcher has many financial benefits over the more conventional chemical combustion launchers. In collaboration on an Air Force Office of Scientific Research (AFOSR) funded Multidisciplinary University Research Initiative (MURI) project, the Center for Pulsed Power and Power Electronics (P3E) at Texas Tech University in Lubbock. Texas has been responsible for developing and investigating a functional scale model of a multi-stage DES railgun to determine its effectiveness to suppress restrike phenomenon and increase plasma armature railgun performance. The distributed energy scheme is theorized to suppress restrike arc formation because the back emf voltage is localized to active stage regions. B-dot sensors positioned along the length of the launcher provide data to measure the plasma arc velocity and detect restrike, arc splitting, or additional secondary arc formation phenomena.

Authors: P. Ford; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191415

Abstract: A Finite Difference (FD) algorithm was developed to calculate the formative delay time between the application of an RF field to a dielectric surface and the formation of a field-induced plasma interrupting the RF power flow. The analysis is focused on the surface being exposed to a background gas pressure above 50 torr. The FD-algorithm is chosen over particle-in-cell methods due to its higher computational speed and its ease of being ported to commercial electromagnetics solvers. The dynamic frequency-dependent permittivity of the plasma is mapped to the time domain of the FD algorithm using the Z transform. Therefore, together with the electron density, the effect of the developing plasma on the instantaneous microwave field is calculated. The high observed value of absorption, up to 60 %, is a result of the momentum transfer collision frequencies in the developing plasma being much larger than the microwave frequency. As a result, the electron density increases to values well beyond the density calculated from setting a plasma frequency equal to the microwave frequency. In the experiment, flashover is induced across a Lucite window by a 4 MW S-band magnetron operating at 2.85 GHz with ~50 ns rise time. The results of the FD simulation are compared with experimental data obtained from flashover with background gases such as nitrogen, air, and argon all at pressures exceeding 50 Torr.

Authors: D. Reale; J. Mankowski; S. Holt; J. Walter; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191489

Abstract: Current work at Texas Tech University is focused on enabling technologies for a mobile Pulsed Ring Down Array. A Monte Carlo Simulation was developed to evaluate the potential performance of an array using current capabilities and to determine the future requirements to achieve the desired system performance. Error sources included are the position error of each element, the jitter of the closing switch, and the synchronization error between elements. Error distributions for the simulation are based on device testing. The array is tested by adding an artificial delay that represents an error source. The simulation is verified by running each error source independently and comparing the results to the experimental data. A system of low power radiators based on stacked MOSFET switches were built to simulate the Pulsed Ring Down Source (PRDS) elements for testing purposes. The testing methodology and results are presented.

Authors: P. Kelly; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191453

Abstract: The development of a high power in-line limiter utilizing varactor-loaded metamaterial structures is presented. A metamaterial structure is an artificial structure engineered to provide electromagnetic properties not available in nature, more explicitly defined as a material having simultaneously negative permittivity and negative permeability. A singly-negative material (SNG) structure, the split-ring resonator (SRR), is a negative permeability material which acts as a notch filter with resonant frequency f0. The resonant frequency of the SRR filter yields itself to tuning since the capacitance between the SRR and transmission lines is easily changeable through the use of varactors. At nominal power levels, f0 is significantly offset from the receiving frequency such that the receiving frequency is unattenuated. When an in-band high power microwave (HPM) is incident upon the filter, a DC bias is applied to several varactors and shifts the resonant frequency of the filter to that of the receiving frequency due to the change in capacitance of the varactors. This effectively attenuates the incident HPM. The filter uses a microwave rectifying circuit to extract a DC voltage from the in-band HPM, which serves as the DC bias voltage across the varactors. Ansoft's HFSS was used to accurately model and design the SRR structure to minimize the E-field and maximize resonant effects. Both high and low power continuous wave testing verified minimal insertion loss as well as verification that the use of varactors in conjunction with a split ring would effectively shift the resonant frequency of the notch filter.

Authors: A. Fierro; G. Laity; L. Hatfield; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191414

Abstract: Vacuum Ultraviolet (VUV) radiation is commonly thought to enhance streamer formation, as it is energetic enough to cause photoionization in the gaseous volume. Light with wavelengths below 180 nm, i.e. VUV, is highly absorbed in the atmosphere which increases the difficulty of measuring any VUV emission from gaseous breakdown at atmospheric pressure. Nevertheless, VUV emission from pulsed surface flashover at atmospheric conditions was previously recorded at Texas Tech. A second generation system was designed to image VUV and visible emission directly while also preserving the spatial profile. The visible emission is imaged through an air-side focused ICCD, while VUV emission is imaged through a vacuum spectrograph. The variable length gap was excited with a pulser designed for a 100 ns rise time and 50 kV peak output. Captured images of visible light emission from streamers produced in oxygen are diffuse whereas nitrogen produces streamers that are segmented. VUV spatial images taken in oxygen reveal stronger emission closer to the cathode region, while nitrogen produces a more distributed intensity profile across the gap. While MgF2 enabled transmission and measurement of VUV, streamer characteristics recorded in the visible light spectrum of surface flashover on BK7 dielectric windows were also investigated. In this paper, the observed streamer images in both visible and VUV wavelength range will be discussed as it relates to surface flashover at atmospheric pressure.

Authors: J. Walterx; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191558

Abstract: A common figure of merit utilized for HPM sources is the instantaneous power efficiency at the time of peak output power. This conversion efficiency from the electron beam power to radiated microwave power disregards the overall energy efficiency of the system, which is vitally important in systems intended for practical application. Optimizing the overall system energy efficiency allows a reduction in volume, weight, and prime power requirements that makes for a much more practical and fieldable system. Texas Tech University has developed a vircator-based laboratory HPM system that has a high end-to-end energy efficiency relative to other HPM systems. The system utilizes a sealed-tube vircator source that requires no external vacuum pumping subsystem. For rep-rate use, the tube contains an integrated low-power sputter ion pump. The lack of need for an externally applied magnetic field saves complexity, weight, and potentially power. The vircator tube is simple and robust. The source is driven by a Marx generator utilizing pulse forming networks (PFNs) instead of lumped capacitors. Utilizing PFNs allows the driver to apply a more ideal pulse shape to the source than a traditional Marx, enhancing the performance of the source and reducing the amount of energy wasted in the rise and fall of the pulse. The system has demonstrated greater than 1% energy efficiency from the energy stored in the Marx to the radiated HPM, with potential for improvement. The peak total radiated output power of the system is between 50 and 100 MW.

Authors: K. Lawson; G. Alvarez; S. Bayne; V. Veliadis; D. Urciuoli

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191651

Abstract: The purpose of this research is to investigate the performance of Silicon Carbide JFET for us in solid state circuit breakers. The device under test for these results is a research grade JFET with a rated blocking voltage of 1200V and a rated forward current of 10A with a power density of 200W/cm2. In order to drive the JFET device, a unique gate driver had to be designed and built to provide switching between two independent rail voltages. The gate driver had to be able to provide adjustable rail voltages with one rail ranging between 0V and -40V and the other rail going between 0V and 2.5V. The gate driver is completely isolated to operate on a high-side switch. In order to test these devices in pulsed switching applications a pulse ring down circuit was designed and built to provide a current pulse of 100A (10 times the rated current) with a charging voltage range between 100V and 500V. Special consideration had to be given to the design of this pulse ring down circuit in order to achieve a high di/dt, and therefore reach the target peak current levels.

Authors: K. Lawson; S. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191513

Abstract: Testing was conducted to determine the performance of new Silicon Carbide (SiC) Super Junction Transistors (SJTs) in pulsed operation. These devices, developed at GeniSiC Semiconductor, are quasi-majority carrier devices. They were first characterized using an Agilent B1505A Power Device Analyzer to determine their operational characteristics. The devices were then pulsed using a high current test apparatus developed at Texas Tech University. After single shot measurements were taken the devices were pulsed at rep-rates of 1 Hz, 5 Hz, and 10 Hz for a total of 3000 pulses. These devices were then placed back on the curve tracer to determine if the pulse testing had any permanent effects on the device.

Authors: Mohamed A. Elsayed; Andreas A. Neuber; John W. Walter; Andrew J. Young; Charlie S. Anderson; Shad L. Holt; James Dickens; Magne Kristiansen; Larry L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191480

Abstract: Continued efforts at the Center for Pulsed Power and Power Electronics at Texas Tech University have led to improvements to the design, testing, and performance of a high power microwave (HPM) system, which is sourced by Lithium-ion Polymer batteries, a polypropylene capacitor, and high energetics. An indirectly seeded two-stage helical flux compression generator (HFCG) produces electrical energies in the kilo-Joule regime into a low impedance inductive load, varying from 2 μH to 3 μH. This high current output of the explosively driven generator is conditioned with a pure silver-wire-based electro-explosive opening switch, which reaches voltage levels in excess of 300 kV into a 18 Ohm load. Upon reaching levels high enough to close an integrated peaking switch, this high voltage is sufficient to drive a reflex triode virtual cathode oscillator, also known as a Vircator, into radiation. The Vircator employed in the system has reached microwave radiation levels well over 100 Megawatts from a cavity volume of less than 5 Liters at a microwave frequency of a few GHz. The complete system is governed through a microcontroller that regulates seed and detonator charging levels as well as discharge times using built-in feedback diagnostics. The complete system aforementioned is constrained to 15 centimeter diameter and measures 183 centimeter in length with an overall volume of less than 34 Liters. No external power or vacuum pumping for the HPM tube is required. This report will discuss recent design advancements and improvements on the HPM system and its sub-components that include the compact seed source, HFCG, and the power conditioning system. Improved safety features implemented as well as novel diagnostic integration will be discussed as well.

Authors: J. C. Stephens; A. A. Neuber; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191477

Abstract: This paper presents the optimization of a compact electro-explosive fuse designed for the power conditioning system to be driven by a helical flux compression generator (HFCG). An electro-explosive fuse interrupts the current flow from the HFCG through a storage inductor on a 50 to 100 ns timescale inducing a voltage large enough to close a peaking gap, which commutates the energy in the storage inductor into a 20 Ohm load at voltage levels above 200 kV. Experimental data has revealed that electro-explosive fuses with wires in closer proximity to one another have consistently produced lower pulsed voltages than fuses with larger wire spacings. This paper addresses possible factors that might contribute to this drop in performance. An electro-magnetic field solver is used to model the current redistribution in the fuse wires. The experimentally observed performance of compact fuses with varying wire spacings is presented.

Authors: J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191487

Abstract: Breakdown phenomena in a high power microwave (HPM) system present unique obstacles for the further development of HPM technology. The non-uniformity of a high frequency electric field and the statistics associated with breakdown in general along with the stochastic nature of naturally occurring electron generating mechanisms introduce significant challenges for predicting and preventing breakdown occurrences within a HPM system. An experiment consisting of an S-band multi-megawatt HPM pulse is used for observing an alternating field induced plasma sheath across a dielectric surface. In order to minimize experimental deviations, a continuous UV lamp is used to provide a constant source of initiatory electrons through the process of photoemission. This reduces the waiting time for flashover initiating electrons to appear, however, primarily due to avalanche statistics, variations are still observed. A statistical model that uses an exponential distribution sampling procedure was developed to predict the surface flashover delay times for a variety of conditions. A supporting experiment that uses a continuous UV lamp and a DC electric field is used for measuring low current due to photoemission from the dielectric window. An explanation of the model describing these phenomena is presented along with a comparison of current measurements from the supporting experiment.

Authors: A. Young; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191479

Abstract: An investigation aimed at optimizing the integration between a capacitor based prime power source and flux-trapping helical flux compression generator (FT-HFCG) is presented. An FT-HFCG simulation code, previously benchmarked with single and multi-pitch generators, was employed to study the kilo-joule class explosive system for this purpose. The details of this effort, which include the optimization of the field coil and stator coupling, as well as an examination of the effect of field coil parameters on the system performance, will be described in this document. For the simulated parameter space, the choice of field coil configuration caused the system energy gain to vary by 300%, and the optimum field coil configuration was found to be a single Litz wire conductor that had an axial length which was approximately 60% of the stator axial length.

Authors: J. Parsony; J. Dickens; A. Neuber; J. Walter; J. Krile; J. Vara

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191565

Abstract: This study focuses on the design of an experimental apparatus for optical and vacuum ultraviolet (VUV) spectroscopy of explosive emission processes at the cathode in a high power vacuum sealed tube. The system consists of a high power pulsed Nd:YAG laser with a highly tunable optical parametric oscillator (OPO, 200 to 2600 nm), a 300 kV, 80 J Marx Generator, and a triode-geometry vircator that is 6 inches in diameter and 11 inches in length. It has been observed that the explosive emission occurring at the cathode forms a plasma front propagating across the A-K gap negatively affecting the impedance of the gap thus shortening pulse length and spoiling the desired consistent low vacuum within the sealed tube. The scope of this project is to probe the plasma formed between the A-K gap to determine the species emitted off the cathode. The timing integration of each sub-system is very critical as the window for measurement is approximately 200 ns. A compact, fiber optically coupled, battery-operated, low jitter (500 ps), fast risetime (20 ns) pulse trigger generator has been designed and incorporated as the trigger source in a trigatron triggering scheme for the Marx generator. Preliminary jitter measurements of 20 to 30 ns have been seen on the fully erected Marx Generator. The intent of this paper is to discuss the details of the various sub-systems and the timing between them, enabling optical / VUV spectroscopic measurements of the explosive emission process.

Authors: C. Lynnx; J. Walter; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191570

Abstract: Vacuum diodes utilizing explosive emission cathodes generally suffer from non-uniform emission, gap closure by anode and cathode plasma expansion, and outgassing. Also, the Child-Langmuir relation does not apply at the edge of the emission area. This results in a high current density sheath at the edge of the emission area. Each of these phenomena presents its own technical challenge in HPM source design and optimization. Diagnostic techniques and particle in cell simulations for a vacuum diode, to be operated as a greater than 100 MW class vircator, have been used in order to compare the characteristics of the electron beam produced by various cathode materials and geometries. Uniform current density plays a key role in vircator performance and efficiency, as well as in the lifetime of the diode. The diode under investigation has an emission area of 20.3 cm2, and operates at a current density on the order of 300 A/cm2 at 200 kV. The background vacuum level and the associated adsorbed and absorbed gases at the surface play a major role in the behavior of an explosive emission cathode. This vircator source is operated as a sealed tube, requiring no vacuum pumping until the device is repetitively operated. The small amount of gas generated during operation is pumped down in between shots, if needed, utilizing an integrated small sputter-ion pump. The background pressure in the diode is in the ultra-high vacuum (UHV) range, on the order of 10−9 Torr to 10−8 Torr. Particle in cell simulations were performed to investigate current density distributions at the surface of both the anode and cathode due to various field shaping profiles. Additionally, scintillator images were taken to compare the beam profile of a machined aluminum cathode with a CsI coated carbon fiber cathode.

Authors: J. Krile; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191562

Abstract: High Power Microwave (HPM) system efficiency has typically been calculated by dividing the peak output power by the peak input power or the input power at the time of the peak output power. This method is typically chosen over energy based calculations because of the difficulties in obtaining an accurate power envelope for certain HPM sources. Power based efficiency calculations have also traditionally excluded any sub-systems which can be highly inefficient. As HPM systems begin moving from lab to field operation the power based efficiencies calculations fall well short of accurately characterizing the complete system. Efficiency calculations should instead be based on energy, typically in the form of fuel or batteries, as they are the major limiting components of any mobile system.

Authors: Shelby Lacouture; Stephen B. Bayne; Michael G. Giesselmann; Kevin Lawson; H O'Brien; C. J. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191511

Abstract: The development of new semiconductor devices requires that extensive testing be completed in order to fully understand the device characteristics and performance capabilities. This paper describes the evaluation of experimental Silicon high power Super Gate Turn Off Thyristors (Si SGTOs) in a unique testing environment. The SGTOs are capable of blocking in the forward direction up to 5kV and are also capable of handling several kA when pulsed. The device structure is asymmetric so the reverse blocking of these devices is only a couple hundred volts. Since these devices are SGTOs special consideration had to be given to the gate trigger circuit so that noise would be minimized on the gate therefore preventing false triggering of the devices.

Authors: D. Bolyard; A. Neuber; J. Krile; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191538

Abstract: A linear transformer driver (LTD) is being constructed at Texas Tech University's Center for Pulsed Power and Power Electronics with the goal to achieve roughly 100 ns pulse width into an 18 Ohm load and energy densities similar to or greater than a previously designed and built 500 J compact Marx generator. A single experimental LTD stage, previously constructed, tested, and reported on by TTU, has undergone several improvements as well as the inclusion of additional diagnostics. Testing at a charging voltage of 14 kV into a 1.7 Ω resistive load has experimentally achieved a 10% to 90% risetime of ~60 ns, peak voltage of 9.2 kV, peak current of 5.5 kA, and instantaneous power of 50.6 MW. Simulations of the LTD stage correlate closely with the experimental results. Extrapolating the simulations to the final desired charging voltage indicate peak voltages and peak currents exceeding 15 kV and 10 kA respectively, with instantaneous power greater than 150 MW from a single LTD stage. The design for a multi-stage LTD using magnetic cores with a flux swing of 6 kV-μs will be presented. Improvements in diagnostics and modeling of the LTD will also be discussed as well.

Authors: J. Walter; J. Vara; C. Lynn; J. Dickens; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191517

Abstract: During the development and optimization of a compact sealed tube virtual cathode oscillator (vircator) at Texas Tech University, it has become apparent that processes at the anode have a significant impact on tube performance. The impact of the high energy, high current density (100-200 A/cm2 or higher) beam on the anode will cause outgassing, plasma production, and anode melting and material ejection. The emitted material expands, eventually impacting the anode transparency and (combined with the plasma formed at the cathode) shorting out the anode-cathode gap. This expansion limits the maximum radiated pulse width, and can also limit the peak output power. The residual evolved gas also negatively impacts the maximum repetition rate of the tube. An effort is underway to study the thermal behavior, gases evolved, and transparency versus time for different vircator anode materials and material treatments. Several different anode materials are under investigation, including stainless steel, copper tungsten, tantalum, nickel, and molybdenum. The effect of different treatments on the anodes before tube assembly is also being studied. The gases that are evolved during operation have been characterized utilizing pressure and residual gas analyzer measurements. The pre-shot background pressure in the tube is in the ultra-high vacuum range (10-8 to 10-9 Torr), and the vircator is not pumped on during firing. The data collected for the different materials is presented.

Authors: S. Beeson; J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191488

Abstract: When dealing with the propagation of High Power Microwaves (HPM), special precautions must be used to prevent the onset of plasma generation. In this paper, we investigate the plasma located on the high pressure side of the dielectric boundary separating the vacuum environment of the microwave source from the high pressure environment of the transmitting medium, e.g., atmosphere. Because the collisional ionization rates are a monotonously increasing function of Eeff/p in the range of interest, the effective electric field normalized with pressure, implementation of HPM in high altitude (low pressure) environments are subject to dielectric breakdown due to this generated plasma, more than at sea-level altitudes. Dielectric breakdown causes the interruption in transmission of electromagnetic radiation due to the reflection and absorption properties of the plasma generated on the dielectric surface. In this paper, transmission, reflection, and absorption data is presented for plasma generated under various pressures ranging from 5 to 155 torr in N2 and air environments. In addition, seed electrons from UV illumination of the dielectric surface and physical vapor deposited metallic points are implemented and their implications to the overall transmission properties are discussed.

Authors: W. Sullivan; C. Hettler; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191650

Abstract: Non-rectifying (ohmic) contacts are essential for efficient photoconductive semiconductor switch performance and maximizing breakdown voltage. Fabricating ohmic contacts requires a very heavily doped surface layer (>; 1018 cm-3), and in silicon carbide (SiC) is typically done by ion implantation. The high energy ions from this process often cause surface and bulk damage, and a high temperature anneal is required to repair the crystal structure and activate the impurities. This paper investigates the use of a gas immersion laser enhanced diffusion system to selectively dope the SiC as an attractive, low cost alternative to ion implantation. A pulsed 260 nm laser with a peak irradiance of 69.9 MW/cm2 was used to dope a high purity semi-insulating (HPSI) 4H-SiC sample with nitrogen to a depth of 150 nm, with measured a surface concentration greater than 1020 cm-3. Using a one dimensional thermal model, the experimental data was fit to diffusion coefficients that are orders of magnitude greater than typically seen in SiC. The gas immersion laser doping technique has been demonstrated as a viable alternative to ion implantation for selective area doping of SiC bulk photoconductive switches.

Authors: M. Chamana; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6025192

Abstract: There has been a keen interest on Distributed Generation (DG) due to their restricted goals of meeting local loads and improving reliability of the overall system. Microgrids (MGs) are connected to the main grid through a Point of Common Coupling which separates the former from the latter. At the time of an intentional islanding or fault at the grid level, a microgrid is able to disconnect itself from the rest of the grid and operate by itself. A microgrid may contain both directly connected and inverter interfaced sources with different control configurations. When disconnected or islanded from the main grid there are various approaches to share the load, one of them being master-slave control where a storage device may become the reference DG to set the nominal voltage and frequency. When the main grid is brought back to normal operation, the microgrid is able to resynchronize itself to the main grid only when it meets certain conditions so as to avoid transients. All the microsources, power electronics and their control with power management were developed in Matlab/Simulink.

Authors: M. Chamana; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6099766

Abstract: Microgrid is a combination of Distributed Energy Resources (DERs) and loads that are connected within a locality. Microgrids are connected to the main grid through a Point of Common Coupling (PCC) which separates the former from the latter. In grid connected mode all the Distributed Generators (DGs) are PQ controlled with pre-defined set-points and the power imbalance is imported from or exported to the main grid. At the time of an intentional islanding or fault at the main grid level, a microgrid is able to disconnect itself from the rest of the grid and operate by itself. A microgrid may contain both directly connected and inverter interfaced sources with different control configurations. When disconnected or islanded from the main grid there are various approaches to share the load, one of them being the master-slave control where a storage device (battery) becomes the reference DG to set the nominal voltage and frequency for the entire microgrid. The State Of Charge of the battery is considered for the battery dynamics when in islanded mode for bi-directional flow. All the microsources, power electronics and their control with power management were developed in Matlab/Simulink.

Authors: G. Laity; A. Neuber; A. Fierro; J. Dickens; L. Hatfield

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191416

Abstract: This paper describes a 2nd-generation system for directly studying the emission of vacuum ultraviolet (VUV) light from pulsed dielectric surface flashover at atmospheric pressure. The role of self-produced VUV emission (i.e. energies greater than 7 eV) on photo-ionization processes during the early nanoseconds of pulsed discharges is virtually unexplored, and yet could be a significant factor in the physics of fast breakdown of directed energy systems (such as MW-class high power microwave devices) in the aerospace community. First generation experiments at Texas Tech University have shown that VUV emission corresponding to nitrogen and oxygen excitation in the energy range 8-10 eV is easily produced, but the use of MgF2 optics inhibited future work with existing hardware due to the transmission cutoff of this dielectric material and chromatic aberration if used as a lensing medium. In an effort to enhance the detection capabilities of our hardware in the wavelength range from 115-135 nm, the current system utilizes a custom designed set of off-axis parabolic MgF2-Aluminium coated mirrors as the primary focusing element. High resolution spectroscopy with the upgraded system resulted in the observation of the nitrogen doublet at 149.5 nm, leading to a better fit for the appropriate line broadening parameters for an approximate 10 eV Boltzmann electronic temperature. Evidence of self-absorption for HI (121.5 nm) provides new insight into the generation of space charge in these plasma structures, which has been investigated quantitatively in both SF6-H2 and N2-H2 mixtures.

Authors: D. Reddy; N. S. Beniwal; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6099816

Abstract: This paper presents the design of an ultra-low power energy scavenging system capable of collecting and managing energy from ambient vibrations and RF electromagnetic waves. Firstly, low power and broadband equiangular spiral antenna which is able to receive and rectify ambient RF radiation is designed and simulated. The energy transducers, commercial piezoelectric generators with a wide frequency range of 26 "“ 205Hz and the designed spiral antenna are evaluated and characterized to maximize the efficiency. Secondly, the power electronic circuits involved in the energy harvesting are designed in 0.6um CMOS technology and the simulation results are presented. Charge pumps, rectifier and Low drop-out regulator (LDO) were optimized to operate with low voltage ranges since the energy produced by the piezoelectric generator and the equiangular spiral antenna is found to be in microwatts and less from the test results. The AC output from the piezoelectric generator is rectified and boosted to required output level using an AC-DC charge pump. Rectifier and DC-DC charge pump are adopted for the efficient conversion of voltage from the broadband antenna. A back-up battery is provided for the start-up of DC-DC charge pump at low input conditions. An LDO with a drop-out voltage of less than 400mV is designed to provide regulated output of 4.1V to the battery. Finally, the collected energy is stored in a 50uAh capacity thin film battery which is intended for low-voltage and low-power applications.

Authors: G. Laity; A. Fierro; A. Neuber; L. Hatfield; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5993029

Abstract: Summary form only given. The role of self-produced VUV emission (i.e. energies greater than 7 eV) on photo-ionization processes during the early nanoseconds of pulsed discharges is a new area of interest in the aerospace community, and could play a significant role in the theoretical understanding of plasma generation at short (nanosecond) timescales. Our previously reported experiments have shown VUV emission from atomic species of oxygen and nitrogen, which were excited during breakdown, but detailed analysis of emission in the range 115 - 135 nm was difficult due to chromatic aberration of VUV transparent optics in this regime. These limitations were recently alleviated by fitting the spectral apparatus with a custom set of MgF2-Aluminum coated off-axis parabolic mirrors used in conjunction with a high resolution vacuum monochromator. VUV emission is observed by either VUV sensitive intensified CCD or photomultiplier sensors, with additional current/voltage monitors and externally focused fast-frame (nanosecond capable) imaging complimenting the diagnostic setup. High resolution spectroscopy has been achieved in the excitation range of interest (8 - 10 eV), where species of atmospheric gases and electrode metal have been identified during the early nanoseconds of plasma generation. Temporal studies have shown that most VUV emission occurs during the time before voltage collapse and subsequent power flow, while the majority of visible emission is released after breakdown when the electron energy distribution has shifted to lower energies. This experiment also allowed for direct emission imaging of VUV radiation, where the spatial profile relative to plasma position is still intact. While the observed metal species are only emitted in the regions close to the electrodes as expected, significant differences were observed for species of NI (which is released throughout the plasma volume) and OI (which is released strongly in regions of high electric field) atoms during breakdown. This is most likely due to the accumulation of ionic space charge from photo-ionization in the gas volume, coupled with the high absorption cross section of molecular oxygen in the VUV range. These measurements will be discussed in context to current physics models of electric breakdown and plasma generation at atmospheric pressure.

Authors: C. Hettler; W. Sullivan; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191652

Abstract: A series of high temperature annealing experiments were performed to characterize the processing parameters that alter the recombination lifetime in high purity, semi-insulating (HPSI) silicon carbide (SiC). All annealed samples were diced from a single 4H-SiC wafer with a measured resistivity of greater than 109 Ω-cm. The samples were annealed for various lengths of time in a PID-controlled high temperature induction furnace at 1810 °C. A 35 GHz microwave photoconductivity decay (MPCD) system was used to measure the transient photoconductivity of the as-grown and processed samples. Through numerical processing of the temporal characteristics of the illuminating laser pulse, the photoconductivity transients were simulated with various recombination lifetimes to fit the experimental MPCD data. The results show that the as-grown material has an average recombination lifetime of 6 ns. However, samples annealed for more than 100 minutes demonstrated recombination lifetimes in excess of 100 ns. The annealing process reduces the concentration of shallow point defects (Z1/Z2) in the bulk material which serve as recombination centers in HPSI 4H-SiC, extending the carrier lifetime. Finally, the impacts of increased recombination lifetime in photoconductive switch operation and performance are presented and discussed.

Authors: T. T. Vollmer; M. G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191687

Abstract: We are reporting results on a compact high voltage capacitor charger with modular topology for rep-rated operation. This modular topology utilizes synchronized hard-switching H-Bridge inverters. Each current-mode controlled inverter is fed into the highly coupled primaries of a high frequency transformer with a nano-crystalline core. This charging system with two synchronized H-Bridges has reached voltages of over 20 kV with power outputs in excess of 10 kW. Rep-rated operation along with system efficiency has been explored with this modular topology system. Burst mode operation has achieved rep rates close to 20 Hz. Overall system efficiencies have been measured at nearly 80%.

Authors: J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5993097

Abstract: Summary form only given. The development of high power microwave (HPM) systems and technologies has been inhibited by breakdown phenomena which limit their transmitting capabilities. For a system in which there is no obvious source of breakdown initiating electrons, HPM breakdown can be even less predictable. A system in which microwaves are generated in a vacuum environment for the purpose of radiating into atmosphere typically uses a dielectric window to separate the vacuum and atmospheric sides of the system. At sufficient field levels, surface flashover can occur across this dielectric window resulting in a severe drop in transmitted power. The time between the application of the HPM and the sharp drop in transmitted power is described as the delay time, which consists of a statistical waiting time for initiatory electrons combined with an electron amplification time, or formative time. The experimental setup for this project consists of a 4 MW HPM source operating a 2.85 GHz attached to a traveling wave structure and a dielectric window mounted on the output side of the system. Dielectric surface flashover has been observed in air and nitrogen with pressures ranging from 60 to 155 torr. To provide a constant source of seed electrons, a UV lamp is used to illuminate the window resulting in photo-emitted electrons appearing at the surface. Another way to provide seed electrons is the inclusion of metallic points on the window which provide a source of field emitted electrons, and also results in a field enhancement at the dielectric surface. In the absence of a constant source of seed electrons, it is expected that field detachment from ion clusters is the primary mechanism for providing the high field region with flashover initiating electrons. A statistical model has been developed for predicting surface flashover that takes into account relevant parameters such as field level, ionization rate, gas type, and pressure. This model has shown good agreement with experimental data in nitrogen with UV illumination providing a constant electron seed rate. Presented here is an adaptation of this statistical model to an environment consisting of field enhancing metallic points as well as a comparison of results for UV illumination and stochastic seeding through field detachment from ion clusters.

Authors: J. -. Braggy; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191411

Abstract: Ferrite loaded nonlinear transmission lines (NLTLs) are able to act as high power microwave sources, utilizing the nonlinearities present in ferrimagnetic materials and the excitation of damped gyromagnetic precession at high incident power levels. Ferrimagnetic properties depend greatly on operating temperatures; therefore, there exists a need to know the ideal temperature at which to operate ferrite loaded NLTLs. Ferrites are chilled or heated to a certain temperature for a time suitable to allow internal ferrite temperature uniformity. Experimental temperatures ranged from approximately -20 °C up to 150 °C, which is slightly above the Curie temperature of the loaded ferrites. This temperature range allows observation of precession dependence on temperature while maintaining ferrimagnetic properties and a single look at the behavior outside the ferrimagnetic regime. Above the Curie temperature the loaded ferrites become paramagnetic and lose ferrimagnetic properties. The design, testing, and results are detailed for an NLTL measuring 0.3 m length and ferrite inner and outer diameters of 3 mm and 6 mm respectively. Figures comparing output waveforms at different temperatures, output power versus temperature, and output frequency versus temperature are shown.

Authors: D. Realey; J. Mankowski; S. Holt; J. Walter; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191656

Abstract: An inexpensive and mobile array of Pulsed Ring Down Sources (PRDS) were required to verify previous simulation results. Initial attempts to use a stacked MosFET circuit as a closing switch were unsuccessful due to the additional series inductance of the MosFET stack lowering the frequency of the oscillation on the coaxial radiator. Experimental results showed that by reducing the parasitic inductance due to the geometry of the MosFET stack, the frequency of the oscillation could be increased. Increased series resistance due to the stacked MosFETs was also a concern. In order to minimize the parasitic inductance of the stack and allow for multiple stacks to be connected in parallel, a printed circuit board was designed. Results from testing of the original switch stacks and board assembly are presented and compared with PSPICE simulations. Using the simulation results, the reduction in parasitic inductance can be estimated.

Authors: C. S. Anderson; A. A. Neuber; A. J. Young; J. T. Krile; M. A. Elsayed; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6191476

Abstract: Helical Flux Compression Generators, HFCGs, are powerful high current sources for pulsed power applications. Due to the single shot nature of HFCGs, electrical output reproducibility is of great importance. One factor known to contribute to unpredictable performance is mechanical inconsistencies introduced during manufacturing of the stator. In an attempt to minimize these deviations during productions, two different winding forms for stator coils, designed to ensure repeatable generator dimensions, turn and coil pitch, were investigated. The differences between the methods were quantified by comparison of measurements made of the physical parameters of the coil (i.e. radius, inductance, etc.), as well as analysis of experiments conducted with the HFCGs fired into a 3 μH load inductor. With any particular fabrication method, the stator insulation material has a distinct impact on generator operation. Quad-built Polyimide coated magnet wire as stator insulation material and Teflon Fluorinated Ethylene Propylene (FEP) as field coil insulation material were investigate to improve HFCG performance. Insulation testing was carried out by firing HFCGs into the inductive load mentioned above. Experimental data and analysis, as well as conclusions on insulation material, will be presented along with a brief discussion of the optimum fabrication method.

Authors: Shelby Lacouture; Stephen B. Bayne; Michael G. Giesselmann; Kevin Lawson; H O'Brien; C. J. Scozzie

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6191666

Abstract: This paper details the design of a test platform for experimental silicon and silicon carbide Super Gate Turn Off devices (SGTOs) capable of stressing the devices with very high energy/power levels while at the same time mimicking a realistic, real world application. To this end an aircraft ground power Motor - Generator set was acquired consisting of a high frequency synchronous generator, a D.C. powered brushless exciter machine, and a 100 HP induction motor. The Si SGTO devices were then placed in a three phase controlled rectifier circuit connected between the generator output and a low impedance high power purely resistive load.

Authors: C. Lynn; A. Neuber; E. Matthews; J. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958383

Abstract: A low impedance Marx generator was developed as part of a test bed for vacuum diodes of various electrode materials and geometries. The generator supplies sufficient energy to initiate and sustain the typically unwanted plasma formation within the diode; which facilitates the observation of the plasma, current uniformity, and electron current densities of various diode structures. The generator consists of ten stages; each stage utilizes a 220 nF 50 kV capacitor, with a series inductance of ~20 nH. When charged to the rated voltage of the capacitors the energy density of the complete generator with case, spark gaps, insulation, etc., is 19.2 mJ/cm3; this is roughly the energy density of a typical ceramic doorknob capacitor without any supporting structure or isolation. The energy density of the capacitors utilized in the Marx generator by themselves is 104 mJ/cm3. Fired into a low inductance short, the ringing frequency was measured to be 1.4 MHz resulting in an output impedance of 5.2 Ω. Erection of the Marx required adding forward feeding capacitors as the stray capacitance to ground is smaller than the capacitance (~60 pF) of the low inductance, low profile spark gap switches. The design and construction of this generator are discussed as well as selected experimental results obtained with the generator.

Authors: J. Foster; M. Thomas; J. Krile; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5534225

Abstract: The production of high power microwaves (HPM) in a vacuum environment for the purpose of radiating into atmosphere requires the use of a dielectric interface to separate the vacuum and atmospheric sides of the radiating structure. For high power to pressure ratios the interface will exhibit surface flashover on the atmospheric side, thus limiting the transmission of microwave power. An experimental setup that utilizes a magnetron operating at 2.85 GHz to produce a 4.5 MW, 3μs pulse propagating in the TE10 mode along with an atmospheric test chamber enables investigating HPM surface flashover phenomena in the presence of various atmospheric conditions. One of the principle parameters measured is the delay time between application of the microwave pulse (50 ns rise time) and the sharp drop in transmitted power due to the flashover plasma formation. Several methods of delay time reduction have been employed to gain a better understanding of the source of breakdown initiatory electrons. For an environment composed of air at, for instance, 155 torr a delay time of 600 ns is observed. Illuminating the dielectric surface with continuous UV radiation reduces the average delay to about 380 ns. An even more distinct reduction in delay time was observed when electric field enhancement was introduced to the window surface via vapor deposition of sub-mm metallic points on the dielectric. These metallic points have proven to reduce the delay time to ~150 ns while increasing the global effective electric field by a factor of ~1.5. This presentation will include an overview of a variety of methods for investigating flashover initiation, including UV radiation and the application of an external DC electric field, as well as the introduction of field enhancing metallic points on the dielectric surface. An analysis of flashover behavior at atmospheric pressures (60-155 torr) in air, argon, and nitrogen will also be given along with an estimation of field enhancement factors for various geometries.

Authors: G. Rogers; A. Neuber; L. Hatfield; G. Laity; K. Frank; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958293

Abstract: In an attempt to identify the mechanisms leading to pulsed dielectric surface flashover in atmospheric conditions, surface flashover across a magnesium fluoride (MgF2) window was studied. The electrode configuration and the applied pulsed voltage level were chosen such that the generated electric field was symmetric with respect to the centerline between the electrodes. That is, neither electrode was favored with respect to flashover/breakdown initiation. A semiconductor-switched 32 kV pulse with 140 ns rise time was applied to the 8 mm wide flashover gap in air, nitrogen, and oxygen at atmospheric pressure. Fast voltage and current measurements along with nanosecond imaging revealed four stages of flashover development: (1) Onset of a cathode directed streamer with a charge on the order of 100 pC and traveling with a speed of ~1 mm/ns at a macroscopic field level of ~10 kV/cm associated with a slow current rise (on the order of 10-3 A/ns) temporarily augmented by (2) a 5 ns wide current spike at the moment when the streamer reaches the cathode followed by (3) a cathode directed streamer focused toward the center of the flashover gap with a slow rising current leading to (4) a sharp current rise (on the order of 10 A/ns) reaching roughly a circuit limited 45 A about 20 ns after the return strike meets an anode directed streamer. Although present in all tested gases, the current spike at the end of stage (2) is most different for all three gases and having the greatest impact in air.

Authors: A. Young; A. Neuber; M. Elsayed; J. Korn; J. Walter; S. Holt; J. Dickens; M. Kristiansen; L. L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958309

Abstract: Recent progress in the development of a compact, portable, explosively-driven high power microwave source is presented. The envelope to which the system must fit has a 15 cm diameter, which means each sub-system fits within this dimension, with an optimized overall length. The system includes an autonomous prime energy source, which provides the initial energy for a two-stage, flux-trapping helical flux compression generator (FCG). Typical output from the FCG is several kilojoules into a 3 μH inductor. The amplified energy from the generator, after pulse conditioning, is used to drive a virtual cathode oscillator (vircator). Recorded voltages at the vircator with this arrangement were greater than 200 kV in experiments, where radiated output powers of greater than 100 MW have been measured. Voltages of at least 300 kV, with an electrical output power of 4 GW or greater, were generated by the FCG driven pulsed power source into a water resistor load with an impedance similar to the operating impedance of the vircator. A description of each component of the compact microwave source will be given, along with waveforms from tests performed with the components independent of the rest of the system. Data from experiments with the fully integrated microwave system will be shown, and analysis will be offered to detail the performance of the system in its present state.

Authors: C. Hettler; C. James; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958289

Abstract: A contactless microwave photoconductivity decay (MPCD) method has been used to measure recombination lifetime and relative conductivity of semi-insulating (SI) silicon carbide (SiC) wafers. A pulsed laser, tunable from 210 nm to 2 μm, has been used to probe above and below band gap photoconductive responses of four SI SiC wafers. The carrier lifetimes were calculated by comparing the reflected microwave signal to the photo response of a fast (<; 300 ps) photodiode. Three vanadium-doped 6H-SiC wafers, with bulk resistivities ranging from 105 Ω-cm to 1011 Ω-cm, and one high purity semi-insulating (HPSI) 4H-SiC wafer (>; 109 Ω-cm) were studied. The photoconductive response of each wafer set is presented. The HPSI wafer demonstrated longer carrier lifetime and improved above band gap photoconductivity compared to the vanadium-doped wafers. The difference in carrier lifetimes are attributed to higher densities of recombination centers (vanadium acceptors) in the 6H-SiC substrates.

Authors: David Matia; Hermann Krompholz; Travis Vollmer; Andreas Neuber; Michael Giesselmann; Magne Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958372

Abstract: A detailed characterization of a 50 J linear transformer driver (LTD) stage is presented. The specific goal of the design is to achieve energy densities superior to typical Marx generators, such as a 500 J compact Marx generator previously designed and built at Texas Tech's Pulsed Power lab. Experimental and analytical techniques for determining circuit elements and especially parasitic elements were used, yielding the magnetizing, primary and secondary leakage inductances associated with the transformer, core saturation effects, parasitic capacitances, the inductance of the pulse discharge circuit, and losses in both copper and the deltamax core. The investigations into these characteristics were carried out using both sinusoidal excitation from 1 kHz to 20 Mhz, and pulsed excitation with rise times down to 5 ns. Pulse amplitudes were varied to cover both the linear and saturation regimes of the core. Distributed parasitic capacitances and the inductance of the pulse discharge circuit were estimated analytically and compared with experimental results. This work was carried out to seek an ideal arrangement of the capacitors and switches on the LTD stage and gain a better basic understanding of fast rise time pulse transformers. Adjustments to the 50 J stage are proposed based on this characterization in order to optimize a future ten stage, 500 J assembly.

Authors: J. Foster; M. Thomas; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958313

Abstract: High power microwave (HPM) surface flashover can be rapidly induced by introducing metallic points on to the dielectric surface with negligible effect on the transmission properties. An experimental setup comprised of a magnetron operating at 2.85 GHz to produce a 4.5 MW, 3 μs pulse is used for observing surface flashover in various atmospheric conditions. An active pulse sharpening mechanism is used to reduce the pulse rise time in order to apply the electric field in tens of nanoseconds. For a system in which HPM transmission must be quickly suppressed, field enhancing geometries can provide a way for flashover to develop rapidly while keeping insertion loss at a minimum (<;0.01 dB). Initial experiments utilizing 0.2 mm2 aluminum points with a spatial density of 25/cm2 have increased the global effective electric field by a factor of ~1.5. This increase in electric field has sharply reduced delay times for surface flashover (i.e. the time between the application of the HPM pulse and a sharp drop in transmitted power). For an environment consisting of air at 155 torr, for instance, the delay time is reduced from 455 ns to 101 ns. Presented in this paper is a comparison of various field enhancing geometries and how they relate to flashover development. Also, an analysis of time resolved images will be given along with an estimation of field enhancement factors.

Authors: Y. Chen; J. Dickens; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958315

Abstract: Recent research efforts at Texas Tech University on impulse antenna phased array have shown that an ideal jitter of a small fraction of the rise-time is required to accurately synchronize the array to steer and preserve the rise-time of the radiated pulse. This has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with very low jitter. This manuscript presents the impact gases and gas mixtures have on switch performance which includes recovery rate and in particular jitter. A 50 Ω, 1 nF pulse forming line is charged to 50 kV and provides the low inductance voltage source to test the different gases. Gases tested include N2, dry air, H2, and SF6, as well as N2-H2, N2-SF6, N2-Ar, and gas mixtures containing Kr85. This manuscript will discuss in detail 50 kV, 100 Hz triggered switch operations of such gases. Switch jitter as a result of triggering conditions is also discussed, also including an evaluation of jitter as a function of formative delay in various gases. An evaluation of switch jitter as a function of operation time and gas temperature is also included. Triggering is provided by a solid state opening switch voltage source that supplies ~150 kV, 10 ns rise-time pulses at a rep-rate up to 100 Hz in burst mode. A hermetically sealed spark gap with a Kel-F -PCTFE (PolyCholoroTriFluoroEthylene) lining is used to house the switch and high pressure gas. It is shown that jitter is strongly dependent on the triggering technique, as well as the trigger magnitude, with ionization rates playing an important role. Sub-ns jitter is seen with a variety of gases and gas mixtures with H2 producing the best results. Varying the gas temperature and addition of radioactive sources are seen to improve the switch jitter.

Authors: J. Vara; J. Walter; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958432

Abstract: The gases evolved during operation of high power microwave devices, such as virtual cathode oscillators (vircators), can be detrimental to their power output, output pulse width, and maximum pulse repetition rate. Gases are evolved both from processes at the cathode (such as explosive electron emission) and at the anode (due to heating from the electron beam). In this project, a residual gas analyzer and pressure measurements have been utilized to characterize the gases generated during operation of a sealed vircator tube. The background pressure in the tube before firing is in the ultra-high vacuum range (~10-8 Torr). The vircator is not pumped on during firing. After firing, the measurements are made, and then the tube is ion pumped back down to the background pressure before the next shot. Multiple anode and cathode materials will be tested, with measurements made of both the quantity and types of gases evolved during firing. Example test materials include stainless steel, OFHC copper, copper tungsten, and tantalum. For all tests, the vircator is driven directly from a low impedance Marx generator, with no intermediate pulse forming. The system design and diagnostic systems are described, and gas quantity and composition data is presented.

Authors: T. Maxwell; K. Patil; S. Bayne; R. Gale

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5562427

Abstract: This paper presents Hardware-in loop (HIL) testing for GM two-mode hybrid. First Modeling and simulation of GM two-mode hybrid using model based design (MBD) process is performed. HIL testing is performed using dSpace MicroAutoBox (MABX) and National Instruments PXI. The control strategy compiled code is uploaded to MABX using ControlDesk software and the corresponding plant model is transferred to PXI using NI Veristand software. This work is part of EcoCAR competition by the Texas Tech University. EcoCAR: The Next Challenge is a vehicle engineering competition organized by the US Department of Energy (DOE) and General Motors (GM).

Authors: M. Thomas; J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958314

Abstract: Surface flashover imposes a fundamental limitation to the magnitude of high power microwaves which can be radiated from the vacuum environment of the source into atmospheric conditions. Providing seed electrons through various methods allows for initiatory conditions to be more closely controlled and the delay time variations to be reduced so that developmental mechanisms can be more closely examined. The experiment uses a coaxial magnetron capable of producing a ~4.5 MW, 3 μs pulse, at 2.85 GHz propagating in the TE10 mode. The pulse rise time measured at the window is reduced using a spark gap pulse steepening technique. The fast rise time pulse propagates through the dielectric into an atmospheric test chamber where various conditions such as gas pressure, type of gas, UV illumination, and charged particle creation by radioactive sources can be controlled. Previous research has shown the significant impacts of UV radiation on the delay time averages and statistical distributions. Surface distributed seeding sources and volume distributed sources will be discussed while the primary focus of this paper will address use of alpha radiation as an ionizing agent. Thus far, a reduction in average delay time by as much as 60% has been achieved at sub-microsecond time scales, which also significantly affected the width of the statistical distributions of the delay time. Alpha particles have a short penetration distance in air which makes them a good candidate for study since the number of electron-ion pars created along the path is large. Analysis of the alpha particles influences will be discussed along with a statistical analysis of breakdown delay in the presence of ionization.

Authors: C. James; C. Hettler; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958348

Abstract: A photoconductive semiconductor switch (PCSS) fabricated from high-purity semi-insulating (HPSI) 4H-SiC is presented. This switch shows improvement over similar designs based on 6H-SiC due to higher carrier mobility and longer carrier lifetimes. Using a novel micropipe identification method, vertical PCSS devices have been fabricated from c-plane wafers that lead to higher hold-off voltage and decreased on-state resistance as compared to a lateral geometry. Operation of the photo switch in both on- and off-state is demonstrated and material characterization via microwave photoconductance decay (MPCD) is given.

Authors: J. Bragg; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958430

Abstract: Ferrite filled coaxial lines have been shown to sharpen input pulses due to their nonlinear magnetic properties. This effect can be amplified by applying an axial magnetic biasing field. In addition to a steeper leading edge, oscillations in the microwave bands have been observed. These oscillations arise from damped gyromagnetic precession occurring in the ferrite. The magnetic bias necessary to generate damped oscillations and steepened pulse sharpening effects is produced by two separate means; a current carrying solenoid or, the less investigated case, permanent magnets. Multiple NLTLs are tested at lengths of 0.3 meters and 1 meter with an outer diameter of 13 mm. Permanent magnets with varying strengths were placed in different locations for multiple tests. Experiments and setups for these designs are detailed and their results are discussed.

Authors: D. Bolyard; A. Neuber; J. Krile; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958310

Abstract: Hydrodynamic pressure simulations combined with an empirical algorithm are used to model the open-circuit voltage output of several explosively compressed ferroelectric materials. The empirical algorithm was initially developed using detonating cord containing PETN and a metal driver element to compress the ferroelectric materials while the open-circuit voltage is recorded. A hydrodynamic code suite, CTH from Sandia National Labs, enables calculating Shockwave propagation and localized pressures. The resulting pressure profile in the ferroelectric material is then used as input for an empirically derived algorithm to calculate the predicted open-circuit voltage of the ferroelectric material. This previously developed empirical algorithm exhibited reasonable correlation between experimental and calculated open-circuit output voltages, but began to deviate when more powerful explosives were used. Hence, the amount of explosive material and geometry of the metal drive was varied to produce a wide range of peak pressures, including pressures higher then the maximum of 3.1 GPa previously modeled by the empirical algorithm. This data serves as the base to further develop the empirical algorithm for various ferroelectric materials and to more accurately model the open-circuit output voltage (experimentally observed range, normalized for thickness, of 1.3 to 3.8 kV/mm) over the wide range of applied pressures.

Authors: M. G. Giesselmann; T. T. Vollmer; L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958381

Abstract: We are reporting on a new compact high power capacitor charger with modular topology for rep. rated High Power Microwave generators and other Pulsed Power Applications. The charger is capable of using a number of synchronized H-Bridge inverters feeding into a common transformer. The common transformer uses litz wire windings and a nano-crystalline core to reduce losses and AC impedance. The H-Bridge inverter modules are individually controlled using peak current mode control which assures proper current sharing and protects the H-Bridge modules from overload while achieving the maximum current handling capacity. To assure stability of the current loops for peak current mode control, slope compensation is used. Major advancements over previous designs are the ability to use multiple H-Bridge inverters with proper synchronization, improved efficiency through advanced transformer design and improved current mode control.

Authors: J. Krile; J. Foster; M. Thomas; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958294

Abstract: High Power Microwave (HPM) induced surface flashover is currently being investigated in order to gain a better understanding of the underlying processes involved and reduce the limitations it places on transmittable pulse lengths. The present experimental setup is designed to produce a flashover on the high pressure side of a transmission window without the influence of a triple point. A 2.5 MW magnetron produces a 900 ns pulse at 2.85 GHz with a 50 ns rise time. The experimental setup allows for the control of several parameters including gas pressure, gas composition, and external UV illumination of the window. Diagnostic equipment enables the analysis of incident, reflected, and transmitted power levels with sub-nanosecond resolution. A previously developed Monte Carlo simulation is used to model the processes involved in the flashover discharge formation. This Monte Carlo code is upgraded to account for the occurrence of field induced electron detachment from negative ion clusters within the high field region, >; 10 kV/cm, near the window. The code has also been expanded to include the occurrence of photoelectrons, emitted from the window while under UV illumination. Such illumination of the transmission window was experimentally shown to reduce the time to flashover by over 100 ns in air at 155 torr, and thereby the total pulse energy that can be transmitted. In addition, UV illumination also reduces the variation in flashover delay times from shot to shot, up to 67% in air at 155 torr. The simulation will determine if the observed reductions in delay time and variation can be explained by the addition of initiatory electrons via UV illumination of the surface.

Authors: J. Korn; A. Young; A. Neuber; C. Davis; M. Elsayed; M. Kristiansen; L. L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5958378

Abstract: The results from an investigation into the performance of a fuse opening switch, to be used as a component of a compact power conditioning unit, PCU, in an explosively driven high power microwave system, is presented. A pulse forming network designed to mimic the current action of a flux compression generator is utilized for these experiments. The investigation focused on the effect of altering fuse parameters, such as the diameter of fuse conductors and conductor winding geometry, on voltage and power levels delivered to a resistive load. Also, experiments were conducted with a thin layer of Semicosil, a commercial silicon material used in slower fusing opening switches, applied to fuse conductors, to investigate the possible advantages of using such a coating. Experiments showed that an increased number of conductor wires (with an approximately constant total cross-sectional area) resulted in similar voltages delivered to the resistive load, and that coating fuse conductors with Semicosil had a negligible effect on power delivery. A detailed description of these experiments will be given, along with data and waveforms illustrating the effects of these parameter variations on power levels delivered to the load.

Authors: C. Davis; A. Neuber; J. Stephens; A. Young; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958416

Abstract: Exploding wire arrays as fast switches are of interest for explosive pulsed power applications utilizing magnetic flux compression. This type of opening switch has proven effective in producing pulses of several hundred kilovolts into vacuum diode loads. The research presented here discusses an exploding wire array capable of producing single digit kilo-joules, 100 nanosecond pulses when driven by a 45 kilo-Amp current with a waveform closely resembling typical magnetic flux compression output. For this specific parameter range, the optimal fuse design was developed based on the experimental behavior of the fuse under variation of parameters such as wire spacing, shielding, and quenching medium. Each fuse is composed of several silver wires arranged in a straight wire cylindrical array and is typically pressurized in a chamber filled with about 0.6 MPa of SF6. The tradeoff between wire spacing and voltage output was addressed by designing four different fuse termination pairs each with a diameter that increased wire spacing from 5 to 20 mm in 5 mm increments. A wire shield test was also conducted as an extension to the wire spacing experiment to uncover any mutual radiative effects between wires on fuse opening behavior. The optimum fuse design, including the optimum fuse wire diameter, will be discussed with a 20 Ohm resistive load as well as a vacuum diode load with similar impedance.

Authors: M. Elsayed; A. Neuber; C. Lynn; J. Korn; C. Anderson; A. Young; J. Dickens; M. Kristiansen; L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958300

Abstract: Recent efforts at the Center for Pulsed Power and Power Electronics at Texas Tech University have been focused on the development of a compact and explosively driven High Power Microwave, HPM, system. The primary energy source (other than the seed energy source) driving the microwave load in this system is a mid-sized, dual-stage helical flux compression generator, HFCG. The HFCG has a constant stator inner diameter of 7.6 cm, a length of 26 cm, with a working volume of 890 cm3. Testing at the Center has revealed energy gains in the 30's and 40's with output energy levels in the kilo-joules regime into loads of several micro-Henries. Over the last few years, close to one hundred shots have been taken with these generators into various loads consisting of dummy inductive loads, power conditioning systems, and HPM sources. Throughout these tests, the working volume of the HFCG, i.e. the volume in between the wire stator and the explosive-filled aluminum armature, was filled with SF6 at atmospheric pressure. This was primarily done do avoid electrical breakdown in the generator volume during operation, resulting in flux loss. Recent design updates enable pressurizing the generator volume to pressures up to 0.5 MPa, which is needed, for instance, to replace the SF6 with other gases such as air or nitrogen. The performance of the dual-stage HFCG with pressurized working volume (SF6 and N2) is presented in this paper along with an analysis of the maximum electric field amplitude held off in the volume during operation. The design technique to seal the HFCG will also be briefly discussed.

Authors: A. Myers; J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958326

Abstract: A pulsed ring-down source array (PRDS), also called a Resonant Antenna-Source System, is a robust high-power transient RF transmitter that provides substantial power in the far field region. In a typical PRDS, a high potential is stored in the device structure and discharged through a switch, generating a damped ringing radiated waveform. Using an array of PRDS antennas, the radiated far field power density can be increased far beyond that possible with a single source. The array's performance is dependent on the individual antenna elements, along with the main switch triggering jitter and the jitter of the trigger source. The array being tested is composed of four antennas spaced a quarter wavelength apart from one another in a straight line. The PRDS element is a half coaxial monopole antenna tuned for ~100MHz. The switch used is a custom made trigatron spark gap pressurized up to 100 PSI of N2. The results presented from a working example of a PRDS array will enable future development of the system.

Authors: S. L. Holt; M. A. Elsayed; B. Gaston; J. C. Dickens; A. A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958312

Abstract: The design and testing of an integrated front-end power and control system for helical flux compression generators (HFCG) is presented. A current up to 12 kiloamps needs to be pushed into the 5.8 microhenry field coil of the HFCG to establish the necessary seed flux for generator operation. This current is created with the discharge of a 5 kilovolt, 50 microfarad metalized polypropylene film capacitor using a single-use semiconductor closing switch. Once peak current/flux is obtained in the seed coil an exploding bridge wire (EBW) detonator is initiated with a discharge from a 1 kilovolt, 500 millijoule capacitor array contained in the compact fire set. Both capacitances, seed and fire set, are charged using a rapid capacitor charger system. The rapid capacitor charger is a solid state step up converter supplied by lithium-ion polymer (LiPo) batteries. It provides the 5 kilovolts and 1 kilovolt dual output voltages required for the compact seed source and compact fire set, respectively. The rapid capacitor charger operates at an average output power of 3 kilowatts and charges both capacitances simultaneously in under 250 milliseconds. The rapid capacitor charger is reusable if protected from the explosive detonation.

Authors: P. Hawkins; A. Neuber; K. Vepa

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5551471

Abstract: Applied Materials is working with customers to assess new ways to identify opportunities to reduce resources usage (energy, process chemicals/gasses) by existing semiconductor process tools. Through six customer pilot projects we have identified potential average savings of $22.5K per year, for the CMP and CVD process tool examined, by using a consultative service approach to business and financial analysis that pinpoints opportunities for process resource reduction without diverting critical engineering personnel from their core responsibilities.

Authors: G. Laity; A. Neuber; G. Rogers; K. Frank; L. Hatfield; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5534416

Abstract: Summary form only given. It is commonly accepted that vacuum ultraviolet (VUV) radiation, corresponding to emission from 180 nm to 115 nm and below, is responsible for photoionization contributing to streamer propagation during the initial stages of atmospheric discharges. An experimental setup was constructed to observe the VUV emission of pulsed surface flashover along a dielectric surface between atmosphere and vacuum. However, VUV radiation is highly attenuated in the atmosphere, which makes observation of detailed spectra difficult. For VUV transmission down to 115 nm the light emitted by surface flashover across an MgF2 window (front side of window in air, backside in vacuum) was focused by an MgF2 lens onto the entrance slit of the spectrograph. The high speed detection scheme consists of a VUV sensitive ICCD camera and a photomultiplier, both with nanosecond temporal resolution. Spectra were measured in various gas mixtures at atmospheric pressure with a flashover spark length of about 8 mm with a 35 kV pulsed excitation, and spectral calibration was done utilizing a VUV calibration lamp with a known emission spectrum. Virtually all lines from 115 to 180 nm can be identified as atomic oxygen and nitrogen transitions during flashover in dry air, with most VUV emission occurring during the initial breakdown stage (current rise). The extremely fast decay of VUV emission intensity following this initial stage is evidence of radiationless quenching of the excited energy levels associated with the observed spectral lines. Flashover studies were also performed in pure oxygen and nitrogen environments to reinforce the observed emission trends. Spectroscopy must be carefully detailed, for instance, the Oxygen-I line at 130.2 nm (which corresponds to a ground level transition) is shown to be strongly self absorbed in the atmospheric spark when compared to a similar oxygen emission line at 130.4 nm. Full spectra were simulated using SpectraPlot, a temperature dependent spectral software suite developed at Texas Tech. It has been concluded from the comparison of simulated and measured Nitrogen spectra between 140 and 150 nm that the electronic temperature is about 4.5 eV, assuming that the electronic nitrogen energy level population density is Boltzmann distributed. The measured spectra will be discussed in relation to the physics of surface flashover and volume breakdown at atmospheric pressure.

Authors: Y. Chen; J. Dickens; S. Holt; D. Reale; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958436

Abstract: A collaborative effort at Texas Tech University on high power RF transmitters has directly translated to the development of phased array pulsed ring down sources (PRDS). By operating an array of PRDS, peak radiating power on target can theoretically be multiplied by the number of sources. The primary limitation on the application of the array concept is the jitter with which the individual sources can be fired. An ideal jitter of a small fraction of the risetime is required to accurately synchronize the array to steer and preserve the risetime of the radiated pulse. This paper describes in detail the implementation of a GPS based timing system that will synchronize the individual antennas to operate at different geo-locations to function in a coordinated fashion to deliver the peak power of each element to a single position. Theoretical array performance is shown through Monte Carlo simulations, accounting for switch jitter and a range of GPS timing jitter. Each module will include a control unit, low jitter pulser, low jitter spark gap, antenna element, as well as a GPS receiver. The location of each module is transmitted to a central controller, which calculates and dictates when each element is fired. Low jitter in the timing of the GPS reference signal is essential in synchronizing each element to deliver the maxim power. Testing using a preliminary setup using GPS technology is conducted with both 1 pps and 100 pps outputs. Jitter results between modules are recorded to ~10 ns without any correction factors. With the timing and geospatial errors taken into account, the proposed concept will show usable gains of up to several hundred MHz.

Authors: D. Reale; J. Mankowski; S. Holt; Y. Chen; J. Walter; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958437

Abstract: Current research at Texas Tech University is focused on the development of a High-Power Pulsed Ring-Down Source (PRDS) Antenna Array. Previously, a Monte Carlo based analysis was conducted in order to predict the array performance based upon the estimated switching jitter between elements. This analysis showed good performance for jitter times between 0 to 2 periods of the ringing frequency. Therefore, for ringing frequencies up to 500 MHz, jitter times up to 4 nanoseconds can be tolerated. Subsequently, we have shown practical switching solutions capable of the sub-nanosecond switching performance needed for the frequencies of interest. Taking the analysis a step further, we introduce the uncertainty of the absolute position of each antenna element. To implement a randomly distributed array, where the position of elements is not fixed, a method of accurately resolving element positions relative to each other and the target location is required. The use of a variety of GPS technologies and techniques is explored as a method for position and timing resolution. The relative accuracy between elements and the absolute accuracy of each element is discussed. A Monte Carlo based analysis is conducted to predict array performance based upon GPS positional error, GPS timing error, and switch jitter.

Authors: K. Lawson; S. B. Bayne

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5958288

Abstract: This research was conducted to determine the transient performance of Silicon Carbide MOSFET devices. The device under test for these results is a CREE CMF20120D D-MOSFET rated for a blocking voltage of 1200 V and a forward conduction current of 20 A. The first test involves testing the limit of voltage rise time, or the dV/dt of these devices to determine when the device turns itself on. The second test studies the effects of large current pulses, 10x the rated current, on these devices to determine how well these devices are able to handle over current situations. For both of these tests a test bed had to be designed and built.

Authors: K. Patil; T. Maxwell; S. Bayne; R. Gale

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5729177

Abstract: Present work investigates development of the vehicle design in the Texas Tech University EcoCAR program with industry standard vehicle development process (VDP). The three different architectures considered are fuel cell, GM two-mode hybrid and belt alternator/starter system (BAS+). The design process started from the architecture selection with the use of PSAT software. Based on results, best choice of vehicle architecture among the three considered is two-mode hybrid. With Matlab/Simulink environment vehicle model is developed. The plant model and controller model is tested with software-in-loop (SIL) and hardware-in-loop (HIL) simulations. The testing of this stage is done with NI PXI and dSPACE MicroAutoBox.

Authors: D. Matia; H. Krompholz; M. Giesselmann; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5386263

Abstract: The design of a 15 kA linear transformer driver (LTD) is presented. The specific goal of this LTD was improved energy density over the 500 J compact Marx generator previously designed and built at Texas Tech's Pulsed Power lab. The design of an individual 50 joule, 30 kV stage is discussed. For successful operation of the LTD, multiple spark gaps have to be fired with low jitter. Possible approaches for the design of a compact, low jitter triggering circuit will be presented as well.

Authors: S. L. Holt; J. C. Dickens; J. L. McKinney; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5386260

Abstract: Many pulsed power applications have demanding system requirements. Power systems for these applications are expected to provide high energy, high pulsed power and long standby times without recharge, all in a very compact package. The different properties of batteries and capacitors make them most suitable for different uses. When selecting a prime power source for compact pulsed power systems a hybrid system often provides the optimal solution, utilizing a battery for prime energy storage during standby and a capacitor for intermediate energy storage before and between operations. This system takes advantage of the best characteristics of both sources to fulfill the system requirements. The design and testing of such a compact system is discussed. The system utilizes a solid-state converter to charge a 50 ¿F polypropylene capacitor to 5 kV in under 500 ms from lithium-ion polymer (LiPo) batteries. Battery selection and testing is also covered. The battery and charger assembly occupies 1.25 L while the capacitor occupies an additional 1.4 L.

Authors: J. Korn; A. Neuber; A. Young; C. Davis; M. Kristiansen; L.L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386099

Abstract: The generation of high power microwaves using explosively driven pulsed power is of particular interest to the defense community. The high energy density of explosives provides the opportunity to design pulsed power systems which occupy significantly less volume, yet provide the same output power, as traditional methods of high power microwave (HPM) production. Utilizing a flux compression generator (FCG) as explosive driver necessitates introducing an intermediate power conditioning system (PCS) that addresses the typical impedance mismatch between FCG and HPM source. The presented PCS is composed of an energy storage inductor, an opening fuse switch and a self-break peaking gap all of which needed to fit within an envelope of 15 cm diameter. Currents in the tens of kilo-amperes and voltages in the hundreds of kilo-volts have to be handled by the PCS. The design of the system, which takes up less than 11 liters of volume, as well its performance into a 20 ¿ resistive load (used to approximate the operating impedance of certain HPM sources) is presented. Approximately 6 GW of electrical peak power was delivered to the load.

Authors: Ryan Karhi; Michael Giesselmann; David Wetz; Jeff Diehl

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386257

Abstract: The development process pertaining to the design, fabrication, and testing of a 40-stage free-running arc synchronous distributed energy railgun is presented. Investigation of this type of system will determine the effectiveness of a distributed energy scheme to suppress the plasma restrike phenomenon and increase plasma armature railgun performance. Determined by a computer simulation, the proposed system will have a 1.0 cm ? 1.0 cm square bore cross section and a stage length of 15.24 cm producing a total rail length of 6 meters for 40 stages. A free-arc is utilized to relieve the financial burden of a large stored energy facility. A velocity of 8 km/s is desired to emulate conditions during a high altitude microsatellite launch. To achieve this velocity, pulsed power in conjunction with a low pressure (~ 10 Torr) air environment is required. The pulsed power supplies 15 kJ of energy to provide an armature current (~ 50 kA) for 1 millisecond. A real-time feedback control system will accurately release the stage energy upon arc arrival. Experimental data collected from a 7-stage prototype distributed system is discussed which will mimic the design and operation of the first 7 stages associated with the 40-stage railgun. The copper rail length is 1.2 m long with a 1 cm ? 1 cm square bore cross section and a 15.24 cm stage length. Each distributed energy stage contains a 750 ?F capacitor bank, a thyristor with an anti-parallel diode, and a driver board for triggering. The armature is formed using a plasma injector that is powered by a 40 kV Marx generator. Diagnostics for this examination include rail B-dot probes as well as independent Rogowski coils for each stage. Data collected from the rail B-dot probes will be used to measure the armature position and velocity as a function of time. There is no target velocity for this prototype; repeatable energy module operation, accurate stage triggering, and arc propagation toward the muzzle are the main areas focus. Outcomes of these initial experimental results will aid the development of the 40-stage system.

Authors: C. Lynn; A. Neuber; J. Krile; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386199

Abstract: It is known that polymers become conductive under shock loading, which can be critical to the operation of explosive driven high current/voltage devices. Hence, the propensity of several select polymers to conduct under shock loading was investigated. Four polymers, Nylon, Teflon, Polypropylene, and High Density Polyethylene, were tested under shock pressures up to ~22 GPa. Shock waves were generated with high explosives, and CTH, a hydrodynamic code developed at Sandia National Laboratories, was utilized to calculate pressure and temporal resolution of the shock waves. Time of arrival measurements of the shock waves were taken to correlate the hydrodynamic calculations with experimental results. A notable delay between shock front arrival and the onset of conduction is exhibited by each polymer. The delay tends to decrease with increasing pressure down to approximately 500 ns for HDPE at ~22 GPa under electric field strength of ~6.3 kV/cm. The data shows that some polymers exhibit more delay than others, thereby indicating better insulating properties under shock loading. Additionally, experiments revealed that the polymers conducted for a finite time on the microsecond time scale before recovering back to an insulating state. This recovery from a shock wave induced conducting state back to insulating state was investigated for a possible opening switch application.

Authors: J. Parson; J. Dickens; J. Walter; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386262

Abstract: This paper assesses critical activation limits of electroexplosive devices (EED), such as blasting caps, which have recently found more usage in pulsed power environments with high EMI background. These devices, EEDs, can be very sensitive to low levels of energy (7-8 mJ) which make them dangerous to unintended radiation produced by compact pulsed generators. Safe operation and use of these devices are paramount when in use near devices that produce pulsed electromagnetic interference. The scope of this paper is to provide an evaluation of activation characteristics for EEDs that include energy sensitivity tests, thermodynamic modeling, and electromagnetic compatibility from pulsed electromagnetic interference. Two methods of energy deposition into the bridgewire of the EED are used in the sensitivity tests. These methods include single and periodic pulses of current that represent the adiabatic and non-adiabatic heating of the bridgewire. The heating of the bridgewire is modeled by a solution to the heat equation using COMSOL¿ with physical geometries of the EED provided by the manufacturer.

Authors: C. Hettler; C. James; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386380

Abstract: Photoconductive semiconductor switches (PCSS) made from semi-insulating (SI) silicon carbide (SiC) are promising candidates for high frequency, high voltage, and low jitter switching. However, existing switches fail at electric fields considerably lower than the intrinsic dielectric strength of SiC (3 MV/cm) because of the field enhancements near the electrode-semiconductor interfaces. Various geometries were identified which could reduce the electric field near the contact regions. The switches were simulated with various parameters and compared. In all cases, it was determined that a high dielectric constant (high-k) encapsulant is a crucial requirement that reduces high fields within the bulk material while inhibiting surface flashover. Assorted high-k encapsulants were evaluated and a portion was subsequently tested in the lab. The observed dielectric strength and relative permittivity of the encapsulants are presented. Pseudo switches, employing sapphire substrates, were constructed and biased to electrical breakdown. The dielectric strength of the interface between the semiconductor and the encapsulant was tested and improvements were discussed.

Authors: C. James; C. Hettler; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386303

Abstract: SiC manufacturers are continually improving the purity of their wafers, however, interband impurities, while detrimental in many applications, can be useful in the operation of photoconductive switches. Compact, high-voltage photoconductive switches were fabricated using c-plane; vanadium doped 6H-SiC obtained from II-VI, Inc. This material incorporates a large amount of interband impurities that are compensated by the vanadium amphoteric, but at present is only available as c-plane wafers. In order to avoid micropipe defects, lateral switches were fabricated to allow validation of material simulations. Low resistivity contacts were formed on the semi-insulating material and a high-voltage encapsulant increases the surface flashover potential of the switch. Material characteristics were determined and switch parameters were simulated with comparisons made to experimental data.

Authors: Michael G. Giesselmann; Travis T. Vollmer

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386226

Abstract: We are reporting on a new design for a compact high power Capacitor charger with a power output that far exceeds the peak power of previously reported designs. For this purpose we are evaluating parallel modular designs with separate inverters, transformers, & rectifiers and compare them with designs with a larger module size. For larger power implementations with a single inverter, transformer, & rectifier, the main challenge is the design of the inverter using parallel connected IGBT transistors and their current sharing. We are demonstrating an implementation were we are combining 2 H-Bridge inverters in parallel to feed the primary of a 100 kVA compact step-up transformer. Each H-Bridge is using current mode control to enforce current sharing. The oscillators in the current mode control IC's for each H-Bridge are synchronized for this parallel mode of operation. This topology could be expanded to more H-Bridges. In order to reduce the high-frequency impedance and the windings losses of the step-up transformer, litz wire is used for both the primary winding and the secondary windings.

Authors: M.A. Thomas; J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5227632

Abstract: Summary form only given. The introduction of a high voltage DC electrode into an experimental setup used for observing high power microwave surface flashover has shown to significantly vary the total delay time for this type of breakdown. The experiment utilizes an S-band magnetron operating at 2.85 GHz to produce a 4 MW, 3 mus pulse. A plasma switch mounted in a WR-284 waveguide reduces the 10-90% rise time of the pulse to ~50 ns, and it reflects the pulse towards a dielectric window to induce surface flashover. A wire electrode charged to plusmn20 kV is inserted into the center of the dielectric window and oriented perpendicularly to the major electric field component of the TE10 microwave mode. The DC field from the electrode influences charge carriers in the flashover region, forcing potential breakdown initiating charged particles away from or towards the surface, dependent on polarity and particle charge sign. Initial tests were conducted in pure N2 at a pressure of 125 torr. The low probability of negative ions (stable negative N2 ions do not exist) appearing in the volume simplifies the interpretation of the experimental results by allowing for the existence of primarily electrons and positively charged ions. A significant increase (~50%) in the average total delay time for the case of a positively charged electrode has been observed. An increase in the average statistical delay was also observed as well as a decrease in the presumed formative delay for both voltage polarities. The apparent electron production rate in N2 was estimated to be 2 e/mus and 8 e/mus under HPM pulse application in the case of positive and negative DC voltages, respectively. Results of further tests conducted in Argon and Krypton-85 in Argon balance are presented along with a statistical analysis of measured delay times.

Authors: M. A. Elsayed; A. A. Neuber; M. Kristiansen; A. Stults; L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5386300

Abstract: Two integral components that accompany an FCG in an explosively driven system is the prime power source and the trigger set. The objective of the prime power source or seed source is to provide the initial seed current/energy into the primary stage of an FCG. Another integral component in an FCG based pulsed power system is the trigger set. The trigger set is used to detonate an exploding bridge wire (EBW) which triggers the high explosives (HE's) in an FCG. This paper will discuss a recent design of a stand-alone apparatus that implements a self-contained (battery powered with full charge time less than 40 sec), single-use Compact Seed Source (CSS) using solid state components for the switching scheme along with a single-use Compact Trigger Set (CTS) that also implements a similar switching technique. The CSS and CTS stand-alone apparatus developed is a system (0.005-m3 volume and weighing 3.9 kg) capable of delivering over 360-J (~12 kA) into a 5.20-¿H FCG load and approximately 2-mJ (~600 A) into the EBW. Both the CSS and CTS have trigger energies of micro-Joules at the TTL triggering level.

Authors: H. Rahaman; S.H. Nam; Seung H. Kim; S.S. Park; S.H. Kim; H. Heo; O.R. Choi; S.C. Kim; K. Frank

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386426

Abstract: It is our interest to design and develop a high power pulser system employing microplasma discharge in a spark gap system. The type of discharge generates short electric pulses capable of both the characteristics of high repetition rate and fast rise time. Such a switch pulser has wide applications in the industry. An efficient electrical discharge in the spark gap is very important for the switching operation. The regime of operation utilizes the residual plasma in the inter pulse recovery of the spark gap to increase the repetition rate. Therefore, the mechanisms of the discharge in a controlled manner, such as the plasma discharge by the arc channel, the dielectric recovery process as well as residual plasma in the post arc period are of great importance to ensure the high repetition rate. The aforementioned mechanisms are established through optimization of the circuit element parameters such as the electrode geometry, the gap distance, the gas pressure, the gas type as well as the applied voltage and current ratings.

Authors: Y. Chen; J. C. Dickens; J. W. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5227620

Abstract: Summary form only given. A recent research effort at Texas Tech University on impulse antenna phased arrays has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with ultra low jitter. An ideal jitter of a small fraction of the risetime is required to accurately synchronize the array to steer and preserve the risetime of the radiated pulse. A 50 Omega, 1 nF pulse forming line is charged to 50 kV and provides the low inductance voltage source to test the different gases. Triggering is provided by a solid state opening switch voltage source that supplies ~80 kV, 10 ns risetime pulses at a rep rate up to 100 Hz in burst mode. A hermetically sealed spark gap with a Kel-F lining is used to house the switch and high pressure gas. Previously, the system was successfully tested with 50 kV, 100 Hz switch operations. Gases tested include, dry air, H2, N2, and SF6, as well as H2-N2, and N2-SF6 gas mixtures. A discussion on switch operation time and switch gas temperature vs. jitter will be included in this paper. This paper will also discuss in detail the effects on switch jitter when different concentrations of Kr85 gas are introduced.

Authors: Lynn L. Hatfield; Bryan Schilder

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386414

Abstract: A simple system for measuring the attenuation of microwaves in the frequency range of 700 MHz to 13 GHz is described. It has been used to test a large number of commercially available microwave shielding materials. The standard system for such measurements would require IEEE STD 299 2006. This standard requires a number of different sources and receivers depending on the frequency range and, therefore, requires a number of different physical arrangements. The simple system described here uses two microwave horns and a network analyzer to compare the signal strength for an open path between the two horns with the same path with a microwave shielding material inserted. This ratio, expressed in db, can be obtained quickly and easily for any material that can be made into a flat rectangle larger than the receiving horn. The horns used here are A. H. Systems SAS-571 with a usable range of 700 MHz to 18 GHz. The network analyzer is an hp 1397C with a high frequency limit of 13 GHz. The materials tested include conducting paints on cloth such as denim, conducting woven fabrics, and metal meshes. The typical measurements presented here to illustrate the use of the method show that conducting paints and conducting fabrics mostly show large attenuation over the quoted frequency range although almost never as high as stated by the manufacturer. When the manufacturers quote the standard used for their measurements, it is sometimes an obsolete standard that has been withdrawn.

Authors: I. Petzenhauser; K. Frank; U. Blell; B.-J. Lee; J. Jacoby

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386346

Abstract: At the GSI Helmholtzzentrum fuer Schwerionenforschung GmbH a new accelerator complex, called Facility for Antiproton and Ion Research (FAIR), is under construction. Its main components are the SIS 100 and SIS300 heavy ion synchrotrons. To operate their injection/extraction kicker magnet systems, modulators with pulse-forming networks (PFNs) are necessary. The PFNs will be charged to a high voltage up to 70 kV and discharged via a high-voltage switch. The switch has to handle currents up to 6 kA, pulse durations up to 7 microseconds with an overall lifetime exceeding 108 shots. The repetition rate is about 4 Hz and a current rise rate of at least 4*1010 A/s is required. The only commercially available switch in this parameter range is actually a multi-gap thyratron. As an alternative, a three-gap pseudospark switch is under development at GSI. It combines the major advantages of the thyratron with its low stand-by power as a cold-cathode device, as well as its insensitivity to large current reversal. Like for the thyratron, the maximum hold-off voltage of a single gap pseudospark switch is limited to about 35 kV. For a reliable hold-off voltage of 70 kV, a three-gap system was designed. Test results with a first prototype switch of this design are reported. The prototype has demonstrated a voltage hold-off capability of more than 80 kV. The circuit of capacitive and resistive voltage dividers was optimized to improve the switch control, the delay and the jitter values. As trigger unit, a conventional high-dielectric trigger is used. With such a trigger unit a crucial issue for minimum delay breakdown still remains the plasma coupling between the different gaps by drift spaces. Those drift spaces have to be designed carefully in order to minimize the internal delay of breakdown. An additional major issue is that the switch suffers from losses, which principally limit the lifetime of low-pressure gas discharge switches. A common way to minimize losses by anode dissipation is to integrate a so-called anode inductor. To see whether this technique can be used with a cold-cathode switch at low repetition rates and relatively large pulse lengths, preliminary tests with an anode inductor were performed.

Authors: C. Davis; A. Neuber; A. Young; J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386110

Abstract: This paper discusses a non-explosive pulsed power device used to reproduce the output waveforms of a Flux Compression Generator (FCG) driving a High Power Microwave (HPM) source. This system optimizes the power conditioning components of a HPM source while reducing the time and resources inherent to explosively driven FCG schemes. An energy storage inductor, fuse opening switch, and a peaking gap make up the power conditioning system. This system couples large voltage pulses (~300 kV), suited for HPM sources, to the load by disrupting the energy storage inductor current (~40 kA). This paper will show that an optimal fuse length was experimentally searched for by varying the calculated fuse wire base length by ±5, 10, and 15%. Various geometric fuse designs were examined to achieve a 45% reduction in the physical fuse length at constant wire length with acceptable performance losses. This paper will also show that the distance between the electrodes of a peaking gap can be optimized to more efficiently switch in the load. Finally results will be shown that depict the amount of microwave power produced by a vircator before fuse and peaking gap optimization.

Authors: J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386184

Abstract: Vircator high power microwave sources are simple, robust, and require no external magnetic field, making them desirable for use in practical compact high power microwave systems. A vircator can be driven directly from the output of a low-impedance Marx generator, eliminating the need for bulky intermediate energy storage components. A compact high power microwave system has been constructed and tested at Texas Tech University utilizing a triode geometry vircator and a compact Marx Generator. The size and performance of this system is compared to a similar system previously developed at Texas Tech. The current triode vircator is housed within a six inch diameter tube which is eleven inches in length. The Marx is contained in an oil tank that is 36 inches long × 12 inches wide × 18 inches tall. Diode voltage and current, and radiated microwave waveforms are presented.

Authors: D. Bolyard; A. Neuber; J. Krile; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5386196

Abstract: Explosively-driven ferroelectric generators are attractive as potential prime energy sources for one-time use pulsed power systems. While the output voltages of small ferroelectric discs have been shown to be on the order of the theoretical maximum values, scaling the ferroelectric to larger thicknesses has proven less successful. The primary limiting factor is how much of the ferroelectric material is compressed simultaneously. This is difficult to control for thicker ferroelectric discs or stacks of discs due to pressure pulse attenuation in the material and rarefaction waves shortening the pressure pulse. A hydrodynamic code system is utilized to calculate the temporally and spatially resolved pressure. The calculated pressure values are converted into voltage produced by the ferroelectric through an algorithm based on an empirical polarization-pressure hysteresis curve. The validity of the algorithm has been verified for PZT EC-64 with experimental data from a flyer-plate experiment reported in literature and our own experiments with the shock wave from the explosives more directly applied to the ferroelectric. Both calculations and experiments produced normalized output voltages, ranging from 1.4 to 3.4 kV/mm for 2.54 cm diameter discs. We will discuss how this pressure to voltage algorithm along with pressure simulations aided in the scaling of the amount of ferroelectric material in a generator, as well as in the design of new driver elements with the goal to increase the peak output voltage of a generator while keeping the generator compact. The calculated voltage output results are compared with experimental data of explosively-driven ferroelectric generators.

Authors: A. Young; A. Neuber; M. Elsayed; J. Walter; J. Dickens; M. Kristiansen; L.L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386301

Abstract: An explosively driven High Power Microwave (HPM) source has been developed that is based on the use of a Flux Compression Generator (FCG) as the primary driver. Four main components comprise the HPM system, and include a capacitor-based seed energy source, a dual-staged FCG, a power conditioning unit and an HPM diode (reflex-triode vircator). Volume constraints dictate that the entire system must fit within a tube having a 15 cm diameter, and a length no longer than 1.5 m. Additional design restrictions call for the entire system to be stand-alone (free from any external power sources). Presented here are the details of HPM system, with a description of each subcomponent and its role in the generation of HPM Waveforms will be shown which illustrate the development of power as it commutates through each stage of the system, as well as power radiated from the diode. Analysis and comparisons will be offered that will demonstrate the advantages of an explosively driven HPM system over more conventional pulsed power devices.

Authors: J. Foster; M. Thomas; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386323

Abstract: Observed delay times for high power microwave surface flashover are influenced significantly by the presence of a DC electric field. The experimental setup to investigate theses influences is comprised of an S-band magnetron operating at 2.85 GHz with a pulse rise time shortening switch assembly that produces a 50 ns rise time at a ~ 2.5 MW power level. A wire electrode charged to ±20 kV is inserted into the dielectric interface perpendicular to the electric field of the TE10 mode to provide a DC electric field in the flashover region. Tests have been conducted in pure N2 at 125 torr in order to provide an environment composed of primarily electrons and positive ions. The average measured delay of window flashover with a DC field pointing into the dielectric has been observed to increase by ~50%. Additionally, effective emission rates of seed electrons initiating breakdown have shown a decrease from 14 e/¿s to 2 e/¿s, indicating the removal of charged species from the high microwave field region due to charge drift in the applied DC field. An overview of the experimental setup is given along with a statistical analysis of delay times measured in Air as well as N2. The open question of where seed electrons originate from and the quantification of the primary processes involved will be addressed.

Authors: M. Thomas; J. Foster; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386180

Abstract: Delay times of high power microwave surface flashover are affected by radiation illuminating the dielectric. A controlled environment of pure Argon at a range of low pressures as compared to normal atmospheric pressure was used with 2 mW/cm2 UV-radiation illuminating the test window. Argon was chosen due to its relatively small number of processes involved such as inelastic electron collisions and due to the well-known cross-sections for these processes. Delay times in the presence of UV are significantly shorter than without UV illumination. The initial electron density contribution due the UV source is very roughly estimated to be ~106 cm-3. A small admixture of radioactive krypton-85 showed only marginal changes in the observed delay times, likely due to an insufficient concentration of Kr-85 producing ionization events only every few microseconds and the high energy distribution associated with the emitted electrons. A detailed discussion of experimental breakdown delay data, along with theoretical expectations and discussion of the statistically dependent mechanisms and analysis, will be given. The ultimate goal is to develop a model for HPM window breakdown in a UV environment, to describe the role of discharge initiating electrons, and to quantify breakdown at high altitudes.

Authors: G. Rogers; A. Neuber; G. Laity; J. Dickens; K. Frank; T. Schramm

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5386373

Abstract: Spectroscopic measurements in the vacuum ultraviolet (VUV) regime are difficult to make due to extremely large absorption of VUV radiation in most materials. This paper describes an experimental setup designed for studying the optical emission during pulsed surface flashover for the wavelength range between 115 nm to 180 nm at atmospheric pressures with a focus on the scheme used to excite the spark gap. The surface flashover of interest occurred on an MgF2 window (front side of window in air, backside in vacuum) imaged onto the entrance slit of a 1 m vacuum spectrograph. Emission spectra were recorded with an Andor DH740 series ICCD camera. All data was taken at atmospheric pressure with a flashover spark length of about 8 mm created by a pulser designed for a 500 ns pulse, max 50 kV output. The centerpiece of this pulser is the CCSTA14N40 thyristor by Solidtron/Silicon Power which features a rate of change current of maximum 30 kA/¿s and a hold-off voltage of up to 4 kV. A pulse transformer with Metglas® core was used to elevate the voltage to 50 kV with a rise time of 180 ns and a peak current of 500 A. The pulser was designed for a repetition rate of 10 Hz and is triggered by TTL pulses. Discussed in this paper, along with the measured spectra and their relation to the physics of surface flashover at atmospheric pressure, will be the design of the pulser.

Authors: G. Laity; K. Frank; G. Rogers; M. Kristiansen; J. Dickens; A. Neuber; T. Schramm

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5227565

Abstract: Summary form only given. Spectroscopic measurements in the vacuum ultraviolet (VUV) regime are difficult to make due to extremely large absorption of VUV radiation in most materials. This paper describes an experimental setup designed for studying the optical emission during pulsed surface flashover for the wavelength range between 115 nm to 300 nm at atmospheric pressures. A vacuum monochromator VM 505 from Acton Research Corporation was used as the spectrograph. For VUV transmission down to 115 nm the light emitted by surface flashover across an MgF2 window (front side of window in air, backside in vacuum) was focused by an MgF2 lens onto the entrance slit of the spectrograph. A quartz window with sodium salicylate coating exposed to the spectrograph's vacuum was placed in the exit focal plane of the collimating mirror of the spectrograph. This fluorescent coating down-converts the VUV light to longer wavelengths that were recorded with an Andor DH520 series ICCD camera in combination with a Nikon 105 mm lens. Spectra were measured at atmospheric pressure with a flashover spark length of about 9 mm and DC excitation with a capacitance of 4.1 nF. Emission spectra were measured from 300 nm down to 130 nm. In parallel, theoretical spectra were calculated primarily for the identification of radiating species and their temperature. Utilizing the NIST Atomic Spectra Database (ASD) data a library of temperature dependent optical emission spectra was generated with SpectraPlot, a spectral software suite developed at TTU. VUV spectral lines of nitrogen, carbon, magnesium and silicon were identified. In pure nitrogen, for instance, the nitrogen I double line at 174.3 nm and 174.5 nm is clearly visible in the spectrum along with a strong double line at 279.6 nm and 280.4 nm, which is emitted by Magnesium II, eroded from the surface exposed to flashover. Spectra were measured in ambient air, pure nitrogen, and argon. An experiment upgrade is currently underway, increasing the VUV sensitivity of the setup. The measured spectra will be discussed in relation to the physics of surface flashover and volume breakdown at atmospheric pressure.

Authors: Y. Chen; J. J. Mankowski; J. C. Dickens; J. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743614

Abstract: Recent research efforts at Texas Tech University on impulse antenna phased array has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with ultra low jitter. An ideal jitter of a small fraction of the risetime is required to accurately synchronize the array to steer and preserve the risetime of the radiated pulse. In (Y. Chen et al., 2007), we showed the initial test system with sub-ns results for operations in different gases and gas mixtures. This paper presents the impact gases and gas mixtures have on switch performance which includes recovery rate and in particular jitter. A 50 Omega, 1 nF pulse forming line is charged to 50 kV and provides the low inductance voltage source to test the different gases. Triggering is provided by a solid state opening switch voltage source that supplies ~150 kV, 10 ns risetime pulses at a rep rate up to 100 Hz in burst mode. A hermetically sealed spark gap with a Kel-F lining is used to house the switch and high pressure gas. This paper discusses in detail 50 k, 100 Hz switch operations with different gases. Gases tested include, dry Air, H2, N2, and SF6, as well as H2-N2, and N2-SF6 gas mixtures. Switch jitter as a result of triggering conditions is discussed, also including a comprehensive evaluation of jitter as a function of formative delay in the various gases.

Authors: A. Young; T. Holt; M. Elsayed; J. Walter; J. Dickens; A. Neuber; M. Kristiansen; L.L Altgilbers; A.H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743600

Abstract: Single-shot high power microwave (HPM) systems are of particular interest in the defense industry for applications such as electronic warfare. Virtual cathode oscillators (vircators) are manufactured from relatively simple and inexpensive components, which make them ideal candidates in single-shot systems. The flux compression generator (FCG) is an attractive driver for these systems due to its potential for high energy amplification and inherent single-shot nature. A self-contained (battery operated prime power), compact (0.038 m3), FCG-based power delivery system has been developed that is capable of delivering gigawatts of power to a vircator. Experiments were conducted with the delivery system connected to a resistive dummy load and then to a reflex-triode vircator. In order to optimize the performance of the vircator when driven by the power delivery system, a second experimental setup was constructed using a Marx-generator based system operating at similar voltages and rise-times. Performance measures of the delivery system when discharged into a resistive load will be presented, as well as vircator output power levels and waveforms from both experimental setups.

Authors: Ryan W. Karhi; John J. Mankowski; Magne Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4657655

Abstract: An investigation into arc splitting at distributed energy feed locations is presented. Distributed energy scheme experiments conducted at Texas Tech University reveal secondary arc formation by arc splitting at distributed current feeds. At these locations, dynamic magnetic pressure regions are believed to perturb the plasma and lead to its division into two separate current carrying bodies. Continuing research into this hypothesis and viable methods of prevention are discussed. A two stage distributed energy railgun is utilized for this analysis. Diagnostics including armature B-dots and a photodiode array facilitate an understanding of the plasma dynamics in the complex multi-stage railgun environment. The length of the free- running plasma arc allegedly is believed to be a contributing parameter into the arc splitting phenomenon. These luminosity data collected from the photodiodes provides an arc length measurement of the light emitting particles within the plasma body. This length is observed to expand and contract in accordance with the corresponding magnetic pressure present within the railgun bore.

Authors: D. Belt; J. Mankowski; J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743602

Abstract: In recent years, the pulsed ring-down antenna has become of great interest due to its compact size and high power on target potential. Since these systems are fairly new in study, it is often difficult to predict the overall performance without experimental evaluation. A pulsed ring-down antenna operates by charging the single element antenna with a high potential source and then closing a switch to develop transient wave reflections on the antenna, typical CW case analysis does not apply. For this reason, we have constructed a simulation model that allows us to predict the transient behavior of the structure. By utilizing the Comsol RF module transient analysis functions, we are able to characterize various parameters of different antennas, beginning with a dipole pulsed ring-down antenna operating around the 100 MHz range. After examining the simulated results against the experimental results for accuracy, we then moved to more complicated mesoband antenna structures. The simulation model developed within the COMSOL RF module allows us to examine various influential factors such as material losses, transient switching effects, structure capacitance, switch capacitance, and initial charging losses. With this, we are able to examine methods to improve the results in the far field such as capacitive spark gap loading and other capacitive storage methods. Utilizing the pulsed ring-down antenna model, we are able to give a better characterization of mesoband pulsed ring-down structures for implementation into a specific or multi-purpose phased array system.

Authors: E. J. Matthews; A. A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743637

Abstract: The size of compact pulsed power generators capable of producing pulses with ~100 ns duration at Gigawatt power levels is primarily determined by the specific energy density of the utilized energy storage medium. Capacitors capable of delivering large pulsed currents at several 10 kV voltage levels have been most frequently used as the energy storage medium for portable pulse generators. To increase the specific energy density of the pulsed power generator, the capacitors are often voltage overstressed at the cost of capacitor life. However, rapid charging (milliseconds) of the capacitor immediately followed by discharging alleviates somewhat of the lifetime problem. For repetitive operation of the pulsed power generator, the charging/discharging energy loss is the more important parameter. The energy, WC, needed to charge a capacitor to a set voltage is measured along with the energy released, WR, by the capacitor under conditions corresponding to a compact Marx generator operating with ~10 Hz rep-rate into a ~20 Ohm load. For the tested capacitors with Mica as dielectric, the capacitor efficiency, eta, i.e. the ratio between WR and WC, is roughly equal to 97% and largely independent of the charging time. Also tested ceramic capacitors revealed an efficiency of ~90% for fast charging and an efficiency of ~94% for slower charging (from ~2 to 35 mus time constant).

Authors: M. Giesselmann; T. Vollmer; M. Lara; J. Mayes

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743625

Abstract: We are reporting on a compact high power charger which is integrated into compact Marx generators for rep. rated high power microwave generators and other pulsed power applications. The charger uses rectified AC mains input voltages of 120 V single phase or 208 V three phase and produces output voltages of up to ten's of kV with HV output power levels of 10 kW. The rep. rate capability is up to 100 Hz. Major advancements over previous designs are current mode control of the main inverter and improved voltage feedback control.

Authors: M. Giesselmann; T. Vollmer; R. Edwards; T. Roettger; M. Walavalkar

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743626

Abstract: We are reporting on a compact high voltage (HV) power supply that can be used in the mobile or airborne generation of high power microwave generators. The charger uses rectified AC mains input voltages of 480 V three phase and produces output voltages of more than 55 kV with HV output power levels of 100's of kW continuously. Major advancements over a previous design are the design of the secondary winding of the HV nano-crystalline transformer, the control of the primary inverter and the thermal management and diagnostic of the HV-DC rectifier.

Authors: C. James; C. Hettler; J. Dickens; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743565

Abstract: Vanadium compensated, 6H silicon carbide (SiC) is investigated as a compact, high-power, linear-mode photoconductive semiconductor switch (PCSS) material. SiC is an attractive material due to its high resistivity, high electrical breakdown strength, and long recombination times compared to other photoconductive materials. The PCSS is designed for fast-rise time, low-jitter (sub-nanosecond) operation in a matched 50 mu test bed. Ohmic contacts were applied by physical vapor deposition and initial test utilized an external Nd:YAG laser trigger source. Analysis of the optical properties of Va-compensated SiC and of switch conduction resistance are presented and performance of contact material is discussed.

Authors: Klaus Frank; Byung-Joon Lee; Isfried Petzenhauser; Udo Blell

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743693

Abstract: This paper presents the current status of the development of a multi-gap pseudospark switch that is designed to control the fast kicker magnet systems of the SIS100/300 heavy ion synchrotron. Finally there is a qualitative explanation presented for the different hold-off voltages for H2 and D2.

Authors: J. Parson; J. Dickens; J. Walter; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743684

Abstract: This paper assesses the critical activation energy required to set off electroexplosive devices (EED) at constant joule heating and rates of joule heating. Safe operation and use of these devices are of great concern in and around pulsed electromagnetic interference. Sensitivity characterization of EEDs include firing sensitivity plots, thermodynamic modeling and electromagnetic interference. Activation energy evaluation of single and periodic rectangular pulses are included to represent adiabatic and non-adiabatic bridge wire heating of the EED. The scope of this paper is to provide a short overview of sensitivity, thermodynamic, and electromagnetic compatibility of EEDs. The results provide crucial information in evaluating energy induced by pulsed electromagnetic fields of compact pulse generators.

Authors: M. Elsayed; T. Holt; A. Young; A. Neuber; J. Dickens; M. Kristiansen; L. L. Altgilbers; A. H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743567

Abstract: Flux Compression Generators (FCGs) are some of the most attractive sources of single-use compact pulsed power available today due to their high energy density output and mobility. Driving FCGs requires some seed energy, which is typically provided by applying a high seed current, usually in the kilo-Ampere range for mid-sized helical FCGs. This initial current is supplied by a high-current seed source that is capable of driving an inductive load. High-current seed sources have typically been comprised of discharging large capacitors using spark-gaps and over-voltage triggering mechanisms to provide the prime power for FCGs. This paper will discuss a recent design of a self-contained (battery powered with full charge time less than 35 sec), single-use Compact Seed Source (CSS) using solid state components for the switching scheme developed at the Center for Pulsed Power and Power Electronics at Texas Tech University. The CSS developed is a system (0.007-m3 volume and weighing 13 lbs) capable of delivering over 250-J (~10 kA) into a 6-muH load with a trigger energy of micro-Joules at the TTL triggering level. The newly designed solid-state switching scheme of the CSS incorporates off-the-shelf high-voltage semiconductor components that minimize system cost and size as necessary for a single-use application. An in-depth and detailed evaluation of the CSS is presented primarily focusing on the switching mechanics and experimental characterization of the solid state components used in the system.

Authors: Yeong-Jer Chen; John J. Mankowski; John W. Walter; James C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4590809

Abstract: Summary form only given.Recent attention to impulse antenna phased array has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with ultra low jitter. An ideal jitter of a small fraction of the rise time is required to accurately synchronize the array to steer and preserve the rise time of the radiated pulse. This paper presents the impact gases and gas mixtures have on switch performance which includes recovery rate and in particular, jitter. A 50 Omega, 1 nF pulse forming line is charged to 50 kV and provides the low inductance voltage source. Triggering is provided by an SOS voltage source that supplies >100 kV, 10 ns rise-time pulses at a rep rate up to 1 kHz in burst mode. A hermetically sealed spark gap with a Kel-F lining is used to house the switch and high pressure gas. The system includes a gas mixing chamber that can mix various gases up to 2000 psi. Gases tested include dry air, H2, N2, and SF6. Initial testing with 30 kV, 10 Hz switch operations have shown reliable sub-ns jitter times with pure gases including dry air, H2, N2, and with H2-N2, and N2-SF6 gas mixtures. The system was then modified for 50 kV, 100 Hz operations with no recovery issues. Jitter data for pure gases, H2-N2, N2-SF6, and various Kr85 gas mixtures at the 50 kV, 100 Hz operation is also documented and compared in this paper.

Authors: C. Lynn; J. Krile; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743696

Abstract: It is known that polymethylmethacrylate, PMMA, becomes conductive under shock loading. To develop an opening switch utilizing shock induced conduction, the reversibility of this process must be studied. It is suggested in literature that changes in electrical properties begin at pressures as low a ~2 GPa. Applying the minimum pressure necessary for conduction is desirable in order to maximize the reversibility by limiting compression heating of the material. CTH, a hydrodynamic code written at Sandia National Laboratory, was used to design various drivers that deliver pressures in the range of ~2 GPa to ~6 GPa to the PMMA. By utilizing the switch to trigger an RC discharge, the resistance and on-time of the switch was characterized. Experiments have shown conduction durations on the order of ~4 mus. The switch was then placed into a capacitive driven inductive energy storage circuit, IES, to determine the polymer's ability to recover. This paper will present experimental data, CTH simulation results, and discuss the attained switching characteristics under varying shock pressure profiles.

Authors: J. Chaparro; H. Krompholz; A. Neuber; L. Hatfield

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743700

Abstract: Previous research at Texas Tech University has been conducted on the physics governing highly over-voltaged gas breakdown resulting from ultrafast applied voltage pulses with risetimes less than 200 ps and durations less than 400 ps. Experimental results have shown that the breakdown characteristics of such events significantly differ from those observed in standard gas breakdown and a complete understanding of the physics behind ultrafast discharges is far from being clear. As a companion to experimental work, a numerical model is an attractive means of discerning more about the underlying physics behind such events. In this paper, a relativistic, Particle in cell model utilizing Monte-Carlo calculations is discussed as a way to directly simulate the experimental conditions, with similar geometry, background gas, and pulse characteristics. Diagnostic output from the simulation includes space-charge development over time, field and particle energy distributions, and particle number growth rates and spatial distributions. An overview of the structure and formulation behind the simulation code is given followed by a comparison of output data to experimental results. Specific points of interest for comparison include formative and statistical delay times, examination of inhomogeneous ionization regions in the discharge, and the behavior of high-energy particles in the runaway state.

Authors: D. Bolyard; A. Neuber; J. Krile; J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4743573

Abstract: Explosively-driven ferroelectric generators are capable of producing single-shot voltage pulses of more than 100 kV, while requiring no seed electrical source, being very compact, and shelf stable. Problems with ferroelectric generators are the low energy output, high dielectric constant of the ferroelectric material, low surface flashover voltage, inconsistent ferroelectric material quality, and uneven or excessive shockwave compression. Initial generator voltage waveforms show that breakdown occurred towards the end of the generator operation time. Several designs and methods have been tested and implemented to prevent surface flashover with varying results. The ferroelectric discs used for the generators were 0.4 inch thick, 1 inch diameter EC-64 PZT ceramic discs. Several six-disc generators were built and tested with resulting open-circuit voltage pulses of 80-140 kV with a FWHM of 2-4 mus. Further improvements to the generators have been designed to prevent surface flashover, improve the explosive driver element and propagating shockwave, as well as increasing the number of discs per generator. Measured output waveforms into varying loads including direct driven antennas will be shown and discussed.

Authors: W. Sullivan; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743706

Abstract: Ferrite-filled coaxial shock lines have recently been used to significantly decrease the rise time of a high voltage pulse. This decrease can be enhanced by initially axially biasing the ferrite material with an applied external magnetic field, allowing for a faster transition from the unsaturated to the saturated state. The simulation of the ferrite material's operation, including saturation, is discussed as well as the simulation of coaxial nonlinear transmission lines. The project explores the rise time changes with variations of magnetic bias, ferrite geometry, input signal characteristics, and transmission line characteristics. Simulated waveforms are discussed for a nickel-zinc ferrite-filled coaxial line. The pulse steepening effect observed in electromagnetic shock lines occurs primarily because of an increase in phase velocity for points higher on the waveform due to the saturation of the ferrite material. An incident pulse of high enough amplitude will drive the ferrite material into saturation, decreasing the relative permeability to one. This saturation front propagates through the ferrite material in the direction of the incident wave until the entire material is saturated, producing a sub-nanosecond rise time pulse. The shock line is designed for a saturated impedance of 50 Ohms to couple easily into existing systems. Pulsed operation of up to low kilohertz repetition is desired and being explored. Applications of electromagnetic shock lines include laser triggering and ultra-wideband radar generation, as well as others.

Authors: Jonathan Foster; Greg Edmiston; John Krile; Herman Krompholz; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743599

Abstract: High power microwave transmission is ultimately limited by window flashover at the vacuum-air dielectric boundary. While surface flashover in the presence of a vacuum has been studied in some detail, the mechanisms associated with flashover in an atmospheric environment need further investigation. For an aircraft based high power microwave system, atmospheric pressures ranging from 760 torr (sea level) to 90 torr (50,000 ft.) are of principal concern. The experimental setup uses a 2.85 GHz, 3 mus microwave pulse with a 10 to 90% rise time of approximately 600 ns from a magnetron capable of producing 5 MW. The slow rise time of the microwave pulse is sharply reduced by using a waveguide spark gap switch used for fast microwave reflection and a high power four port circulator [6]. This reflected pulse has a reduced rise time on the order of 50 ns. The shorter rise time produces a more ideal step waveform that can be more easily compared with theoretical perfect square pulse excitation. Past investigations showed that the delay time for breakdown in air increases with pressure as is expected from the right hand side of the Paschen curve as long as the electron collision frequency is much larger than the microwave frequency. Surface flashover experiments have produced similar results. At a pressure of 155 torr, for instance, the breakdown electric field strength is 6 kV/cm (power density 0.08 MW/cm2) and the overall delay time from HPM pulse application to reaching critical breakdown plasma density is 600 ns. An overview of the experimental setup is given along with a discussion of breakdown delay times as a function of pressure as well as an investigation of surface flashover in the presence of external UV (ultraviolet) illumination.

Authors: Jordan Chaparro; Lynn Hatfield; Hermann Krompholz; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743701

Abstract: For subnanosecond switching, physical phenomena as well as basic breakdown data, such as delay times and breakdown voltages, are of interest. With a RADAN Pulser as source for voltage pulses with up to 180 kV amplitude and risetimes at a test gap of 180 ps, we investigate statistical and formative delays for argon and dry air at sub-atmospheric pressure, for gap widths of 1 and 11 mm. Formative times have minima between 50 and 200 torr, and range from 70 ps at 1.5 MV/cm to 200 ps at 50 kV/cm. For this range of electric fields, this dependence on pressure and applied field can be explained by the behavior of ionization coefficient and electron drift velocity for homogeneous discharges. For higher fields exhibiting a narrow ionization zone in cathode vicinity with pronounced electron runaway conditions, the experimental data agree with results of Monte-Carlo simulations. Statistical delays are about the same as formative delays at fields of 50 kV/cm, and are reduced with increasing field amplitude to less than 50 ps at 1.5 MV/cm. It appears that field emission is the major source for starting electrons, influencing the statistical delay time near the field emission threshold only.

Authors: A. Neuber; G. Edmiston; J. Krile; J. Foster; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4591128

Abstract: Summary form only given as follows. For flashover in air, nitrogen, and SF6, we have previously shown that the delay time between microwave pulse application and breakdown increases with pressure in the regime where the elastic electron collision frequency is larger than the microwave frequency (roughly >10 torr for 2.85 GHz microwave frequency), which also coincides with conditions found at the right hand side of the Paschen curve. The specific window flashover geometry was carefully chosen to avoid local field enhancement. That is, no metallic parts are exposed to high fields. Hence, only the window surface itself, the gas, and interaction processes between surface and volume contribute to flashover. A Monte Carlo based electron motion code developed for the flashover conditions predicts formative flashover delay times reasonably well in the pressure regime between 100 to 600 torr (10,000 Pa to 80,000 Pa). However, the statistical delay time, that is the time interval required for the initiatory electron(s) to appear, is unaccounted for in the code. Further computational efforts investigating seed electron production via collisional detachment from, for instance, negative ions in the gas have shown that while effective at unipolar fields, collisional detachment is unlikely to contribute to the production of seed electrons at higher microwave frequencies above several GHz. Experiments show that illuminating the surface with light/photons (180 nm < lambda < 350 nm) reduces the observed statistical delay considerably indicating the importance of seed electron production from the surface. This paper will discuss the key processes of high power microwave surface flashover and present experimental flashover data along with continued investigation into the statistics of possible seed electron sources, including trace contaminates present in the gas or on the dielectric surface.

Authors: J. Krile; G. Edmiston; J. Dickens; H. Krompholz; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4743719

Abstract: Surface flashover development at the output window of high power microwave (HPM) systems presents a major limitation to the power densities and pulse lengths transmitted through these interfaces. As a result, developing a physical model accurate in predicting surface flashover initiation is of prime interest. A Monte-Carlo type electron motion simulation has been developed to estimate the delay time from initial electron to flashover. Although this approach has shown reasonable agreement with experimental results, the process yielding the initial seed electron(s) was neglected in the model, primarily due to the lack of quantitative and qualitative information on seed electron production. For instance, computational efforts investigating seed electron production via collisional detachment from negative oxygen ions have shown that while effective at DC, the collisional detachment model cannot remain a likely contributor of electrons at high frequencies (Gt ~5 GHz). The key parameters impacting high power microwave surface flashover will be discussed and presented along with continued investigation into the statistics of possible seed electron sources, including trace contaminates present in the gas or on the dielectric surface.

Authors: R. W. Karhi; J. J. Mankowski; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4346322

Abstract: Summary form only given. Experimental results utilizing a distributed energy scheme and free-running arc are presented. Analysis and observations of the issues associated with distributed energy switching of a plasma armature in the railgun will be explored. The use of a free running arc allows experiments to emulate a plasma armature railgun at high speeds (> 5 km/s) without the requirement of a large amount of stored energy. Diagnostics for this examination include rail and plasma current probes as well as independent Rogowski coils for each stage. The distributed system is comprised of four stages spaced uniformly along the rail length. Each stage contains a high voltage capacitor, an inductor, a phase controlled SCR, and a driver board for triggering. The high voltage capacitors used are capable of storing 50 kJ, but are typically charged to store less than 20 kJ. Fiber optic lines are integrated into the system to prevent misfire in the noisy EMI environment. Optimal switch timing is predicted by a computer simulation and tested for accuracy. The assembled railgun is 2.4 m long with a 1.7 cm times 1.7 cm bore cross section. The rails are made of machined UNS C11000 ETP copper. G-10 insulation contains the current probes and separates the Aluminum 6061-T6 outer support structure from the rails. Alumina ceramic inserts are secured by the G-10 and function to reduce in-bore ablation. A PVC chamber encloses the railgun and allows experiments to be conducted within the desired 1-10 torr environment. A plasma source at the breech reliably supplies plasma for each experimental test. The plasma injector is powered by a Marx generator to supply a 40 kV voltage and pulse length of 10 s. The coaxial injector is comprised of a tungsten rod, ceramic insulator, and stainless steel outer casing. Numerous experimental tests were conducted to investigate the dynamics of plasma armatures within a distributed energy source railgun. Variations of switch timing, bore pressure, current amplitude, and current pulse length within each stage have been tested. This data is analyzed to determine the effectiveness of a distributed energy system to suppress the plasma restrike phenomenon and increase plasma armature railgun performance.

Authors: R. Karhi; J. Mankowski; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4652541

Abstract: Experimental results comparing a breech-fed scheme and a distributed energy scheme for a free-running arc are presented. Analysis and observations of the issues associated with distributed energy switching of a plasma armature in the railgun will be explored. The use of a free running arc allows experiments to emulate a plasma armature railgun at high speeds (≫5km/s) without the requirement of a large amount of stored energy. Numerous experimental tests were conducted to investigate the dynamics of plasma armatures within a distributed energy source railgun. Variations of switch timing, bore pressure, current amplitude, and current pulse length within each stage have been tested. These data are analyzed to determine the effectiveness of a distributed energy system to suppress the plasma restrike phenomenon and increase plasma armature railgun performance.

Authors: Ryan C. Edwards; Michael G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346175

Abstract: Summary form only given. Nano-crystalline ferro-magnetic core material has very favorable magnetic properties for high frequency power applications. This paper investigates the properties of a nanocrystalline transformer with multiple secondary windings through experimental evaluation over a range of frequencies. Design considerations are addressed with special attention focused on scaling with respect to frequency, leakage inductance, and effective AC resistance of the windings. The data derived from the experimental evaluation yields scaling factors to help determine a transformer design for optimal power density.

Authors: Ryan C. Edwards; Michael G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652473

Abstract: Nano-crystalline ferro-magnetic core material has very favorable magnetic properties for high frequency power applications. This paper investigates the properties of a nano-crystalline transformer with multiple secondary windings through experimental evaluation over a range of frequencies. Design considerations are addressed with special attention focused on scaling with respect to frequency, cores losses, and effective AC resistance of the windings. The data derived from the experimental evaluation helps determine a transformer optimal design for the best power density.

Authors: M. G. Giesselmann; R. C. Edwards; M. Lara; J. Mayes

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4345949

Abstract: Summary form given only. We are reporting on a compact high power charger which is integrated into compact Marx Generators for rep. rated High Power Microwave generators and other Pulsed Power Applications. The charger uses rectified AC mains input voltages of 120V single Phase or 208V three phase and produces output voltages of up to 50 kV with HV output power levels of 10 kW. The rep. rate capability is up to 100 Hz. Major advancements over previous designs are current mode control of the main inverter, improved efficiency through advanced transformer design and improved feedback control.

Authors: J. T. Krile; S. L. Holt; D. J. Hemmert; J. W. Walter; J. C. Dickens; L. L. Altgilbers; A. H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4346027

Abstract: Summary form only given. Explosively driven magnetic flux compression generators (MFCG) are effective high current, compact, disposable pulsed power supplies. A common type of MFCG in use is a helical generator because it can provide very high current gain in a compact form. Successful implementation of a MFCG as a pulsed power source requires several peripheral systems including a seed current source and a high current switching mechanism. Additionally, for driving higher impedance loads, a power conditioning system is required to convert the high current, low voltage MFCG output into a more suitable voltage. HEM Technologies' currently developed system utilizes an ultra compact, seed source previously developed by HEM Technologies. An explosively driven closing switch provides both the switching action and acts as a delay generator to allow for the current rise in the MFCG. The output of the MFCG is conditioned via an exploding fuse wire and spark gap pair to convert the high current output to high voltage. HEM Technologies has performed extensive modification and testing of the end-to-end system. The current and energy conversion will be presented along with typical output voltages using different fusing techniques.

Authors: John T. Krile; Andreas A. Neuber; Hermann G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346186

Abstract: Summary form only given. Dielectric surface flashover in an atmospheric environment is an important consideration in the design of insulating support structures for pulsed, high voltage applications. Only recently, increased effort has been invested in characterizing and quantifying the physical processes involved in surface flashover occurring at atmospheric conditions. We have previously shown qualitatively that UV illumination of the surface, either externally or possibly generated by the developing discharge itself, affects the distance between flashover path and surface for small gaps with a nonnegligible field component normal to surface. By studying the effects of UV illumination on the flashover behavior, information was gained about the underlying mechanisms of dielectric surface flashover. Utilizing a solid-state UV source with a much faster turn-off time than gas tubes along with the flashover testing apparatus' generally high temporal resolution enabled us to measure applied UV pulse, voltage, current and flashover self luminosity with high temporal precision. For all experiments, the dielectric flashover sample is placed inside an environmentally controlled chamber with a constant gas flow. Using advanced field simulations the electrode/dielectric geometry was designed to produce an optimized field shape for the testing of surface effects. Besides reducing the flashover delay times by up to 50%, the application of a 1 mW/cm2, 20 mus UV pulse, prior to the voltage pulse, forces the flashover discharge in nitrogen to "hug" the surface rather than develop a few millimeters away from the surface along the field lines. Increasing the time delay between UV and voltage pulse application, it was revealed that the impact of the UV pulse on the flashover path becomes weaker with a time constant, tau, of ~ 3.1 mus. That is, after roughly 10 mus (3 times tau), the flashover path develops as without any UV application at all. In addition to experimental data of dielectric surface flashover with varying degrees of UV illumination, we present a simple model that describes the observed behavior as well as a more advanced analysis utilizing a Monte Carlo type code for the electron collision dynamics.

Authors: J. Krile; A. Neuber; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652476

Abstract: Undesired surface flashover of high voltage support structure can severely limit the compactness of open air high voltage systems. Only recently, increased effort has been invested in characterizing and quantifying the physical processes involved in surface flashover occurring under atmospheric conditions and under the influence of UV illumination. Presently, UV flash lamps as well as a solid-state UV source, for their much faster turn-off times, are utilized in conjunction with a high temporal resolution testing apparatus. The UV pulse, voltage, current and flashover self luminosity will be measured with high temporal precision. A simple model that describes the observed behavior as well as a more advanced analysis utilizing a Monte Carlo type code for the electron collision dynamics will be presented.

Authors: J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4346029

Abstract: Summary form only given. Electrical and optical measurements were performed on explosively driven plasma jets. The explosive devices are in the form of a conical shaped charge. The electrical diagnostic consisted of measuring the current passed through a jet when it made a connection between two metal plates. An energy storage capacitor connected to the two plates was charged to between 350 V and 10 kV, and then discharged through the jet. Also, an ICCD camera was used to capture high-speed single frame pictures of the propagating jet. Several shots were fired with variations in the capacitor voltage, the plate geometry, and the jet material. Current waveforms and optical images are presented. All experiments were performed in the explosive and high power RF experimental facility at the Center for Pulsed Power and Power Electronics at Texas Tech University.

Authors: D. McCauley; D. Belt; J. Mankowski; J. Dickens; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4345962

Abstract: Helical Flux Compression Generators (HFCG) of 50 mm form factor have been shown to produce a maximum energy deposit of 3 kJ into a 3 muH inductor from a seed current. A large dl/dt into a coupled load is possible when an electro-explosive fuse is used. Previous work with a non-optimized fuse has produced ~100 kV into a 15Omega load which leads into a regime relevant for High Power Microwave (HPM) systems. It is expected that ~3()0kV can be achieved with the present 2 stage HFCG driving an inductive storage system with an electro-exploding fuse. In order to optimize the electro-explosive fuse design, a non-explosive test bed, which closely simulates the 45 kA HFCG output, is used. To optimize the tiise, effects of fuse material, fuse length, and fuse shape will be examined as well as the effects of various quenching materials. Our previous work has characterized fuse material but we are also looking into the effects of the processes used to create the fuse wire, such as tempered wire versus fully annealed wire. Additionally, to maximize the output voltage and minimize the fuse recovery time, we are optimizing the length of the fuse wire. For shorter fuse lengths, we are optimizing fuse shape as well as fuse length to find the best fuse recovery time. By optimizing the individual parameters of an electro-explosive fuse, the fuse as a whole will be optimized to produce maximum output voltage when used with an HFCG.

Authors: Bell, D; Mankowski, J; Neuber, A; Dickens, J; Kristiansen, M

PDF: https://ieeexplore.ieee.org/document/4530702

Abstract: Helical Flux Compression Generators coupled with an inductive energy storage system have shown promising results as a driving source for High Power Microwave (HPM) loads. The output performance of the inductive energy storage system is contingent upon the opening switch scheme, usually an electro-explosive fuse. Our previous work involving fuse parameter characterization has established a baseline for potential fuse performance. By applying this fuse characterization model to an HFCG powered system, a non-optimized fuse has produced 60 kV into an HPM equivalent load with an HFCG output of 15 kA into a 3 mu H inductor. Utilization of a non-explosive HFCG test-bed has produced 36 kV into an HPM equivalent load with an output of 15 kA into a 1.3 mu H inductor. The use of a non-explosive HFCG test bed will allow the verification of scalability of the fuse parameter model and also allow testing of exotic fuse materials. Prior analysis of fuse parameters has been accomplished with various materials including Silver (Au), Copper (Cu), and Aluminum (Al), but particular interest resides in the use of Gold (Ag) fuse material. We will discuss the a-priori calculated baseline fuse design and compare the experimental results of the gold wire material with the silver wire material baseline design. With the results presented, an accurate Pspice model applicable to our 45 kA HFCG systems will be available and allow the development of accurate modeling for higher current systems.

Authors: D. McCauley; D. Belt; J. Mankowski; J. Dickens; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652362

Abstract: Helical Flux Compression Generators (HFCG) of 50 mm form factor have been shown to produce a maximum energy deposit of 3 kJ into a 3 μH inductor from a seed current. A large dI/dt into a coupled load is possible when an electro-explosive fuse is used. Previous work with a non-optimized fuse has produced ∼100 kV into a 15 Ω load which leads into a regime relevant for High Power Microwave (HPM) systems. It is expected that ∼300 kV can be achieved with the present 2 stage HFCG driving an inductive storage system with an electro-exploding fuse. In order to optimize the electro-explosive fuse design, a non-explosive test bed, which closely simulates the 45 kA HFCG output, is used. To optimize the fuse, effects of fuse material, fuse length, and fuse shape will be examined as well as the effects of various quenching materials. Additionally, to maximize the output voltage and minimize the fuse recovery time, we are optimizing the length of the fuse wire. For shorter fuse lengths, we are optimizing fuse shape as well as fuse length to find the best fuse recovery time. By optimizing the individual parameters of an electro-explosive fuse, the fuse as a whole will be optimized to produce maximum output voltage when used with an HFCG.

Authors: C. James; J. Dickens; S. Holt; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651906

Abstract: The design and jitter performance of a high-pressure, coaxial spark gap, triggered by field distortion by a center plane electrode is presented. Additional comparisons with trigatron style triggering are also presented. The switch was tested up to 100 pulses per second (pps) with a maximum switching voltage of 50 kV in nitrogen. Analysis will focus on jitter measurements taken over the full lifetime of the switch. This paper presents the results of this analysis. Specifically, switch jitter and lifetime will be evaluated as a function of switch geometry as a whole and as a function of trigger electrode geometry.

Authors: C. Lynnx; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651971

Abstract: It is known that PMMA and PVC become conductive under shock compression. However, to develop an opening switch, the material’s recovery from conductor back to insulator is critical. Heating must be minimized in order to optimize recovery. Two sources of heat in this switch include shock induced heating and switch loss heating, which should be balanced for optimal results. Furthermore, it is also important to determine if the observed switching behavior is due to shock unloading or intrinsic material relaxation properties. In the extreme case of recovery, bulk breakdown may become an issue. This paper presents experimental data and discusses initial results as they relate to the development of an opening switch.

Authors: Andrew J. Young; Thomas A. Holt; Mohamed A. Elsayed; Andreas A. Neuber; M. Kristiansen; L.L. Altgilbers; A.H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346025

Abstract: Compact pulsed power systems require power sources that are small in size yet can produce the necessary electrical energy required to drive the system. Helical magnetic flux compression generators (HFCGs) are attractive for single shot applications due to their rapid conversion of chemical energy to electrical energy. The small total volume of a generator coupled with the energy density of the fast-reacting high explosives makes mid-sized HFCGs an appealing option as sources in single shot compact pulsed power systems. Consistent output current and energy gain from shot to shot are key variables in the ability of an HFCG to drive compact pulsed power systems efficiently.

Authors: A. Young; T. Holt; M. Elsayed; A. Neuber; M. Kristiansen; L.L. Altgilbers; A.H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652394

Abstract: Compact Pulsed Power Systems (CPPSs) require power sources that are small in size yet can produce the necessary electrical energy required to drive a given load. Helical Flux Compression Generators (HFCGs) are attractive for single shot applications due to their rapid conversion of chemical energy to electrical energy. Mid-sized generators occupy little total volume (∼4,000-cm3 total with a compressible volume of ∼300-cm3 in the present generator design), while the high explosives used in an HFCG provide an energy density of ∼8,000 MJ/m3. Consistent output current and energy gain from shot to shot are key variables in the ability of an HFCG to drive CPPSs effectively. An investigation into the practicality of using mid-sized HFCGs as the driver for single shot CPPSs is presented. Data and waveforms from generators fired into 3 μH inductive loads are shown, with results measuring the generator’s performance as a driver for an inductive energy storage (IES) system. Results are also shown from adding a power conditioning system to the output of the HFCG, where the measurements demonstrate the ability of an HFCG to drive high impedance loads. The effectiveness of a mid-sized HFCG as drivers for these systems will be evaluated.

Authors: Gregory F. Edmiston; Andreas A. Neuber; John T. Krile; Luke McQuage; Hermann Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4345565

Abstract: Summary form only given. Surface flashover formation at dielectric/air interfaces during pulsed high power microwave (HPM) excitation can severely limit the power densities which can be transmitted into atmospheric medium. Previous studies on HPM surface flashover in the S-band at 5 MW power levels have reported on the contributing factors to flashover development including the effects of gas type, pressure and relative humidity. Furthermore, analysis on optical emission spectra collected from the developing discharge has determined that the vibrational and rotational temperatures of the plasma are approximately 2700 degK and 300 degK, respectively. In addition to experimental efforts, a Monte Carlo-type electron motion simulation code, MC, has been developed to calculate the increasing electron density during flashover formation. Results from this code have exhibited a quantitative agreement with experimental data over a wide range of atmospheric conditions. A critical parameter to flashover development is the stochastic process involving the appearance of initiatory or "seed" electrons, as seen by the reduction in flashover delay time by approximately 10-20% in the presence of external UV illumination. While the current version of the MC code seeds the flashover location with electron densities on the order of background ion densities produced by cosmic radiation, it fails to incorporate the field assisted collisional detachment processes which are the primary origin of these electrons on the time scales of interest. Investigation of these processes and development of more accurate seeding in the MC code is a key step towards predicting HPM flashover over a wide range of parameters, particularly in the presence of highly electronegative gasses such as SF6 or O2, in which there is an absence of free electrons. Theoretical results of HPM surface flashover with the improved seeding model will be benchmarked against previously measured data obtained with HPM pulse excitation. Further, the slow rise-time data (~500-600 ns risetime) that revealed a distinct reduced field vs. pressure delay time product dependence will be supplemented by short rise-time pulse data.

Authors: G. F. Edmiston; A. A. Neuber; J. T. Krile; L. McQuage; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651836

Abstract: Surface flashover formation at dielectric/air interfaces during pulsed high power microwave (HPM) excitation can severely limit the power densities which can be transmitted into atmospheric medium. Previous studies on HPM surface flashover in the S-band at 5 MW power levels have reported on the contributing factors to flashover development including the effects of gas type, pressure and relative humidity. In addition to experimental efforts, a Monte Carlo-type electron motion simulation code, MC, has been developed to calculate the increasing electron density during flashover formation. Results from this code have exhibited a quantitative agreement with experimental data over a wide range of atmospheric conditions. A critical parameter to flashover development is the stochastic process involving the appearance of initiatory or “seed“ electrons, as seen by the reduction in flashover delay time by approximately 10–20% in the presence of external UV illumination. While the current version of the MC code seeds the flashover location with electron densities on the order of background ion densities produced by cosmic radiation, it fails to incorporate the field assisted collisional detachment processes which are often assumed to be the primary origin of these electrons on the time scales of interest. Investigation of these processes and development of more accurate seeding in the MC code is a key step towards predicting HPM flashover over a wide range of parameters, particularly in the presence of highly electronegative gasses such as SF6 or O2, in which there is an absence of free electrons with zero applied field.

Authors: Thomas A. Holt; Andrew J. Young; Mohammed A. Elsayed; Andreas A. Neuber; M. Kristiansen; Kevin A. O'Connor; Randy D. Curry

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346106

Abstract: Summary form only given. A cooperative effort was initiated between Texas Tech University and the University of Missouri-Columbia to develop a single-shot power conditioning system. The purpose of the system is to convert prime power to an output capable of driving a load with an impedance ranging from 15 to 30 Ohms. A helical flux compression generator (HFCG) was chosen as the electrical energy amplification stage due to its portability and high energy density. Certain topologies of IIFCGs are better suited to drive low impedance loads (i.e. short circuits or similar), however, cascaded HFCG systems are capable of driving higher impedance loads, thereby reducing the requirements from subsequent pulse forming stages to match the HFCG output to the load impedance. Therefore, a dual-stage HFCG was chosen to drive a transformer and series fuse in order to step-up the voltage to the 150 k V-300 kV level. A staged fuse has been designed to open in 280 ns or less with 25 kA-40 kA peak currents. The output voltage will be used to drive an RF load or to charge a mesoband oscillator. Both topologies for power conditioning are being considered and tests to date indicate that both types of geometries can be driven by the HFCG and power conditioning system. The results of the experimental tests as well as the energy transfer efficiency will be discussed.

Authors: Thomas A. Holt; Andrew J. Young; Mohammed A. Elsayed; Andreas A. Neuber; M. Kristiansen; Kevin A. O'Connor; Randy D. Curry

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652438

Abstract: A cooperative effort was initiated between Texas Tech University and the University of Missouri-Columbia to develop a single-shot power conditioning system to drive an RF load. The purpose of the system is to convert prime power to an output capable of driving a load with an impedance ranging from 15 to 30 Ohms. A Helical Flux Compression Generator (HFCG) was chosen as the electrical energy amplification stage due to its portability and high energy density. Certain topologies of HFCGs are better suited to drive low impedance loads (i.e. short circuits or similar), however, cascaded HFCG systems are capable of driving higher impedance loads, thereby reducing the requirements from subsequent pulse forming stages to match the HFCG output to the load impedance. Therefore, a dual-stage HFCG was chosen to drive a pulse transformer and series fuse in order to produce voltages on the order of 150 kV to 300 kV across the secondary of the pulse transformer. The fuse has been designed to open in 280 ns or less when a peak current of 25 kA-40 kA is reached. The output voltage will be used to drive an RF load or to charge a mesoband oscillator. Both topologies for power conditioning are being considered and tests to date indicate that both types of geometries can be driven by the HFCG and power conditioning system. The results of the experimental tests as well as the energy transfer efficiency will be discussed.

Authors: Yeong-Jer Chen; John J. Mankowski; John W. Walter; James C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4345561

Abstract: Summary form only given. Recent attention in impulse antenna phased array has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with ultra low jitter. An ideal jitter of a small fraction of the rise time is required to accurately synchronize the array to steer and preserve the rise time of the radiated pulse. This paper presents the impact, gases and gas mixtures have on switch performance which includes recovery rate and in particular, jitter. A 50 Omega, 2.4 nF pulse forming line is charged to 50 kV and provides the low inductance voltage source to test the different gases. Triggering is provided by an SOS voltage source that supplies >100 kV, 10 ns rise-time pulses at a rep rate up to 1 kHz in burst mode. A hermetically sealed spark gap with a Kel-F lining is used to house the switch and high pressure gas. The system will also introduce a high pressure gas mixing chamber that can mix various gases up to 2000 psi. Gases tested include dry air, H2, and SF6.

Authors: Yeong-Jer Chen; John J. Mankowski; John W. Walter; James C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651833

Abstract: Recent attention to impulse antenna phased arrays has necessitated the need to develop a reliable high voltage, high repetition rate switch that will operate with ultra low jitter. The ideal jitter of such a switch would be a small fraction of the risetime in order to synchronize the array to steer and preserve the risetime of the radiated pulse. This paper presents the impact, gases and gas mixtures have on switch performance which includes recovery rate and in particular, jitter. A 50 Ω, 1 nF pulse forming line is charged to 30 kV and provides the low inductance voltage source to test the different gases. Triggering is provided by an SOS voltage source that supplies ≫100 kV, 10 ns risetime pulses at a rep rate up to 1 kHz in burst mode. A hermetically sealed spark gap with a Kel-F lining is used to house the switch and high pressure gas. The system contains a high pressure gas mixing chamber that can mix various gases up to 2000 psi. Gases tested include dry air, H2, N2, and SF6.

Authors: D. Belt; J. Walter; J. Mankowski; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346122

Abstract: Summary form only given. A pulsed ring-down phased array antenna provides substantial energy deposition in the far field region in addition to a broad range main beam with scanning capabilities. This allows remote neutralization of Improvised explosive devices (IEDs) at far field distances and in virtually any direction. The pulsed ring-down antenna operates by charging the single element antenna with a high potential source and closing a switch to develop transient wave reflections on the antenna which then propagate in air. The performance of a pulsed ring-down phased arrav is highly contingent upon the design and performance of the individual antenna elements within the array. Such factors as operating voltage, antenna capacitance, material losses, antenna geometry and closing switch conductance characteristics must be examined for optimal performance to be achieved. By utilizing the Comsol RF module transient analysis functions, we are able to characterize the various parameters beginning with a monopole and a dipole pulsed ring-down antenna operating in the hundreds of MHz range. We have examined and compared the results achieved from the experimental setup to the simulation model in order to better characterize the individual components of the antenna. We have also examined the discrepancies between an ideal closing switch and the experimental setup closing switch, which dramatically affects the far field range of the antenna. We have examined the material properties of the antenna to improve losses and increase system capacitance allowing an increase in the number of RF cycles per antenna discharge. With the results presented, an accurate model of pulsed ring-down antennas is available and will allow future development of more complex geometries that will improve the operation of pulsed ring- down phased array.

Authors: D. Belt; J. Mankowski; J. Walter; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652443

Abstract: A pulsed ring-down phased array antenna provides substantial energy deposition in the far field region in addition to a broad range main beam with scanning capabilities. This allows remote neutralization of Improvised Explosive Devices (IEDs) at far field distances and in virtually any direction. The pulsed ring-down antenna operates by charging the single element antenna with a high potential source and closing a switch to develop transient wave reflections on the antenna which then propagate in air. The performance of a pulsed ring-down phased array is highly contingent upon the design and performance of the individual antenna elements within the array. Such factors as operating voltage, antenna capacitance, material losses, antenna geometry and closing switch conductance characteristics must be examined for optimal performance to be achieved. By utilizing the COMSOL RF module transient analysis functions, we are able to characterize the various parameters beginning with a monopole and a dipole pulsed ring-down antenna operating in the hundreds of MHz range. We have examined and compared the results achieved from the experimental setup to the simulation model in order to better characterize the individual components of the antenna. We have also examined the discrepancies between an ideal closing switch and the experimental setup closing switch, which dramatically affects the far field range of the antenna. We have examined the material properties of the antenna to improve losses and increase system capacitance allowing an increase in the number of RF cycles per antenna discharge. With the results presented, an accurate model of pulsed ring-down antennas is available and will allow future development of more complex geometries that will improve the operation of pulsed ring-down phased array.

Authors: C. Lynn; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4345862

Abstract: Summary form only given. The feasibility of utilizing shock loaded and unloaded dielectrics as a true closing/opening switch as part of an explosive-driven pulse power system is addressed. While it is known that shock wave compressed PVC and PMMA become conductive, the details of the material's recovery from the conducting back to the insulating state are much less known. To be effective as an opening switch, the recovery time has to be minimized, i.e., for instance, heating of the material must be minimized. The two primary sources of heat in the switch are shock induced heating and switch loss heating. These sources must be balanced for optimal results. Furthermore, it is also important to determine if the observed temporal behavior is due to finite shock unloading or intrinsic material relaxation properties. In the extreme case, bulk breakdown may occur during recovery as voltage increases across the switch. Previous work, performed primarily with a C-4 packed compression rod, has achieved a switch on-state resistance of less than 1 ohm, with an average on-state duration of 80 microseconds. It was also shown that PMMA appears to have a much sharper transition time than PVC, between both insulator to conductor, and conductor to insulator. The results presented here use a timed explosion to more carefully control the intensity and duration of the shock wave. This allows for more control over shock induced heating of the sample. We will present recent experimental data and discuss results as they relate to the development of an opening switch.

Authors: Russell Vela; John T. Krile; Andreas A. Neuber; Herman G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346336

Abstract: Summary form only given. Volume and surface discharges in SF6 are investigated to support refurbishment of the Z-machine's laser triggered gas switch, LTGS, at Sandia National Laboratories (SNL). The recently upgraded LTGS has exhibited a failure mode which results in surface flashover on the inside of the dielectric switch envelope. These flashover events degrade the LTGS performance and cause a pre-fire in the successive shot. The principal physical mechanisms involved in surface flashover at various pressures of SF6 are therefore vital to the understanding of this specific switch failure mode. A 340 kV system was designed which replicated conditions found in the LTGS between two adjacent backbone electrodes. Besides measurement of flashover delay times for different dielectric materials (Lexan, acrylic, epoxies, etc.), emphasis was put on the detailed characterization of the flashover phenomenology. Hence, in addition to voltage and current diagnostics, fast optical imaging of the flashover along with the acquisition of optical emission spectra (~200 nm to 700 nm) was employed. One possible reason for the LTGS failure was thought to be ultraviolet (UV) radiation emitted from the volume discharge between the backbone electrodes, which could induce surface flashover on the inside of the envelope. Our present setup is utilized to gather evidence that would confirm this possibility by adding a field-stressed dielectric surface directly exposed to the volume arc produced between the model backbone electrodes. With a current amplitude as low as ~ 2 kA, the occurrence of sulfur and fluorine ions as well as hydrogen and carbon during dielectric surface flashover was revealed with optical emission spectroscopy. From the spectra, the plasma temperature was determined to be between 1.5 and 3 eV, largely depending on the SF6 pressure. The presence of carbon and hydrogen is of significance because it indicates removal of material from the dielectric surfaces during discharge. X-ray fluoroscopy has also confirmed sulfur deposits on the surfaces of the dielectrics tested. Experimental results with higher current amplitudes will be discussed as well as the results showing the UV's influence on inducing surface flashover in an SF6 environment.

Authors: Russell Vela; John T. Krile; Andreas A. Neuber; Hermann G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652547

Abstract: Volume and surface discharges in SF6 are investigated to support refurbishment of the Z-machine’s Laser Triggered Gas Switch, LTGS, at Sandia National Laboratories (SNL). The recently upgraded LTGS has exhibited a failure mode which results in surface flashover on the inside of the dielectric switch envelope. These flashover events degrade the LTGS performance and cause a pre-fire in the successive shot. In the following, experimental results of both volume breakdown and surface flashover across different dielectric materials in SF6 primarily at 20 psig are presented. In addition to fast voltage and current monitoring of the breakdown events, there was an emphasis put on imaging and optical emission spectra (∼200 nm to ∼700 nm). As much as possible, the laboratory experiment was designed to reproduce the conditions found in the 5.5 MV LTGS. The principal physical mechanisms involved in surface flashover at various pressures of SF6 are vital to the understanding of this specific switch failure mode.

Authors: W. Justis; J. Chaparro; H. Krompholz; L. Hatfield; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346256

Abstract: Summary form only given. With a RADAN 303 A pulser (risetime 150 ps, maximum voltage 150 kV into matched load), fast breakdown in argon and air is investigated. An oil-filled coaxial transmission line is coupled with a lens to a biconical section and a radial millimeter size gap operated at sub-atmospheric pressure. On the other side of the gap, the arrangement is symmetrically continued to represent a matched load. Pulse risetime at the gap is increased to about 180 ps. With capacitive dividers the voltage across the transmission line separating incident and reflected pulses is measured, which allows to determine voltage across and current through the gap. Temporal resolution is defined by the digitizer (20 Gs/s, 6 GHz). Breakdown usually happens during the rising part of the applied voltage pulse. Breakdown curves, i.e. breakdown voltage or time-to-breakdown vs. pressure, have been measured for different applied dV/dt's (from 2times1014 V/s to 8times1014 V/s) and they resemble Paschen curves with a steep increase toward low pressure, and a slow increase toward high pressure. The major findings, such as shifts of the minimum formative time toward increasing pressure with increasing dV/dt, are discussed in terms of similarity laws. Discharges for this case are characterized by runaway electrons over much of the pressure range, with a strong excitation and ionization layer at the cathode surface, and "free-fall" conditions with negligible gaseous ionization for the rest of the gap. Monte-Carlo simulations for the initial stage of the discharge are expected to confirm and quantify the experimental findings.

Authors: W. Justis; J. Chaparro; H. Krompholz; L. Hatfield; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652506

Abstract: With a RADAN 303 A pulser (risetime 150 ps, maximum voltage 150 kV into matched load), fast breakdown in argon and air is investigated. An oil-filled coaxial transmission line is coupled with a lens to a biconical section and a radial millimeter size gap operated at sub-atmospheric pressure. On the other side of the gap, the arrangement is symmetrically continued to represent a matched load. Pulse risetime at the gap is increased to about 180 ps. With capacitive dividers the voltage across the transmission line separating incident and reflected pulses is measured, which allows to determine voltage across and current through the gap. Temporal resolution is defined by the digitizer (20 Gs/s, 6 GHz). Breakdown usually happens during the rising part of the applied voltage pulse. Breakdown curves, i.e. breakdown voltage or time-to-breakdown vs. pressure, have been measured for different applied dV/dt’s (from 2x1014V/s to 8x1014 V/s) and they resemble Paschen curves with a steep increase toward low pressure, and a slow increase toward high pressure. The major findings, such as shifts of the minimum formative time toward increasing pressure with increasing dV/dt, are discussed in terms of similarity laws. Discharges for this case are characterized by runaway electrons over much of the pressure range, with a strong excitation and ionization layer at the cathode surface, and “free-fall” conditions with negligible gaseous ionization for the rest of the gap. Monte-Carlo simulations for the initial stage of the discharge are expected to confirm and quantify the experimental findings.

Authors: Luke McQuage; Gregory Edmiston; John Mankowski; Andreas Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4345564

Abstract: Summary form only given. High power microwave (HPM) surface flashover is investigated in order to gain a better understanding of this phenomenon and reduce the limitations it imposes on transmitted power levels. The experimental setup is designed to produce window flashover without the influence of a triple point. The HPM source for this testing is an experimental virtual cathode oscillator (vircator) capable of producing greater than 50 MW for 100 ns with an adjustable frequency from 3 to 5 GHz. This work builds on previous testing using a magnetron producing 5 MW for 4 mus at 2.85 GHz. The dominant modes of the vircator and magnetron are the circular TE11 and rectangular TE10 modes respectively, with the electric field component in both setups normal to the direction of propagation, yielding comparable conditions. Due to the large differences in output power and pulse length, the two setups operate in different regimes and mechanisms take on differing degrees of importance. Additional factors under investigation include gas pressure, composition, temperature, and air speed. Diagnostic equipment permits the analysis of power levels with sub-nanosecond resolution. Experimental results are compared with data from literature, previous testing, and Monte Carlo simulations.

Authors: L. McQuage; G. Edmiston; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651835

Abstract: High Power Microwave (HPM) surface flashover is investigated in order to gain a better understanding of this phenomenon and reduce the limitations it imposes on transmitted power levels. The experimental setup is designed to produce window flashover without the influence of a triple point. The HPM source for this testing is an experimental virtual cathode oscillator (vircator) capable of producing greater than 50 MW for 100 ns with an adjustable frequency from 3 to 5 GHz. This work builds on previous testing using a magnetron producing 5 MW for 4 μs at 2.85 GHz. The dominant modes of the vircator and magnetron are the circular TE11 and rectangular TE10 modes respectively, with the major electric field component in both setups normal to the direction of propagation, yielding comparable conditions. Due to the large differences in output power and pulse length, the two setups operate in different regimes and mechanisms may take on differing degrees of importance. The experimental setup permits study of factors including gas pressure, composition, temperature, and air speed. Diagnostic equipment allows the analysis of power levels with sub-nanosecond resolution. Experimental results are compared with data from literature, previous testing, and Monte Carlo simulations.

Authors: H. Chen; C. Jiang; A. Kuthi; M. A. Gundersen; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4345736

Abstract: Summary form only given. We report studies of mini (6.3 mm diameter electrodes) back-lighted thyratron (BLT) switches, operating with currents up to 10 kA, and hold off voltages in excess of 40 kV for the mini-BLT. The pseudospark is a thyratron with a cathode variant: the cathode emission is provided by an initially hollow cathode mode that transitions to a self-heated super-emissive mode. As a result of this mechanism, the pseudospark does not require externally supplied power for heating the cathode. The BLT is an optically-triggered version of this switch. The switch studies reported here are of (optically triggered) BLTs with helium or hydrogen fill. Hold-off voltage and optical triggering studies of these small switches suggest that they can be optimized for miniaturized high voltage pulse generation. Overvolted operation was also studied for potential applications such as implementation into Marx generators. These studies suggest that the small pseudospark BLT represents progress towards ultra compact high peak power switches for compact repetitive pulsed power applications. Research to develop sealed versions will be reported, and a new plasma source based from the BLT structure will also be discussed.

Authors: S. L. Holt; J. T. Krile; D. J. Hemmert; W. S. Hackenberger; E. F. Alberta; J. W. Walter; J. C. Dickens; L. L. Altgilbers; A. H. Stults

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652397

Abstract: Explosively driven ferroelectric generators (FEGs) are reliable, compact, high voltage sources that utilize high pressures to liberate charge trapped in the crystal structure of ferroelectric materials. For the active ferroelectric element most FEG designs use commercial lead zirconate-titanate (PZT) compositions designed for either precision actuators or naval sonar transducers. However, the material properties that are important in FEG applications are not the same material properties for which these materials have been designed to maximize. FEG designs utilizing these commercial materials are performance limited by high voltage breakdown, mechanical failure and low energy densities. TRS Technologies inc. has produced a new series of ferroelectric elements designed specifically for FEG applications. HEM Technologies has performed dielectric strength and shock compression experiments on these new materials to evaluate their performance in comparison to existing commercially available materials.

Authors: J. Walter; J. Dickens; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4345788

Abstract: Summary form only given. High-power transient RF transmitters are currently being developed for multiple applications. One potential high- power pulsed source being considered is the pulsed ring- down source (PRDS). In a typical PRDS, electrical energy that is stored in the device structure is discharged through a switch, generating a damped ringing radiated waveform. The magnitude of the transmitted field is limited by parameters such as the overall device geometry and switch characteristics. The potential exists to increase the radiated far field power density by utilizing an array of PRDS transmitters. The performance of such an array depends on the triggering jitter of the device switches and the design of the individual radiating elements. The application of the array concept to pulsed ring-down sources using gas spark gap switches is examined. Theoretical array performance is discussed, taking into account practical switch jitter distributions, practical switch triggering schemes, and individual element characteristics.

Authors: J. Walter; J. Dickens; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4651932

Abstract: High-power transient RF transmitters are currently being developed for multiple applications. One potential high-power pulsed source being considered is the pulsed ring-down source (PRDS). In a typical PRDS, electrical energy that is stored in the device structure is discharged through a switch, generating a damped ringing radiated waveform. The magnitude of the transmitted field is limited by parameters such as the overall device geometry and switch characteristics. The potential exists to increase the radiated far field power density by utilizing an array of PRDS transmitters. The performance of such an array depends on the triggering jitter of the device switches and the design of the individual radiating elements. The application of the array concept to pulsed ring-down sources using gas spark gap switches is examined. Theoretical array performance is discussed, taking into account practical switch jitter distributions.

Authors: J. Chaparro; W. Justis; H. Krompholz; L. Hatfield; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4346257

Abstract: Summary form only given. The X-ray emission of highly overvolted spark gaps under electron runaway conditions is investigated. The pulse source, a RADAN 303 A, is connected to a test chamber through an oil-filled coaxial line, a coupling lens, and a biconical transmission line section, with a symmetrical arrangement attached on the opposite side of the chamber with a matching load. The test chamber allows pressure variation from 10-6-670 torr with argon or dry air used as a background gas. Voltage pulses with amplitudes of 40-150 kV, risetimes less than 200 ps, and FWHM less then 300 ps are applied across hemispherical electrodes with 1 mm spacing. A scintillator-photomultiplier combination with a temporal resolution of 2 ns is used as X-ray detector. Metallic absorber foils of different thicknesses are used to obtain a rough energy spectrum of the x-rays and electrons in the range of about 10 to 150 keV. Results show a high electron-energy component (>60 keV) existing up to atmospheric pressure, and an intense soft component (5 to 20 keV) at pressures around 100 torr. The observations are compatible with gaseous ionization and runaway conditions for extremely high E/p.

Authors: J. Chaparro; W. Justis; H. Krompholz; L. Hatfield; A. Neuber; T. Gibson

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4652507

Abstract: The x-ray emission of highly overvolted spark gaps under electron runaway conditions is investigated. The pulse source, a RADAN 303 A, is connected to a test chamber through an oil-filled coaxial line, a coupling lens, and a biconical transmission line section, with a symmetrical arrangement attached on the opposite side of the chamber with a matching load. The test chamber allows pressure variation from 10−6 – 670 torr with argon or dry air used as a background gas. Voltage pulses with amplitudes of 40–150 kV, risetimes less than 200 ps, and FWHM less then 300 ps are applied across hemispherical electrodes with 1 mm spacing. A scintillatorphotomultiplier combination with a temporal resolution of 2 ns is used as X-ray detector. Metallic absorber foils of different thicknesses are used to obtain a rough energy spectrum of the x-rays and electrons in the range of about 10 to 150 keV. Results show a high electron-energy component (≫60 keV) existing up to atmospheric pressure, and an intense soft component (5 to 20 keV) at pressures around 100 torr. The observations are compatible with gaseous ionization and runaway conditions for extremely high E/p.

Authors: B. T. McDaniel; J. J. Mankowski; D. Wetz; B. McHale; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4530677

Abstract: The experimental results of a five stage distributed energy source railgun are presented. The advantages of such a scheme are increased efficiency due to less energy remaining in the railgun inductance and lower rail resistive losses. The design is a bench top 81 cm long solid armature railgun with a 1.27 cm × 1.27 cm bore cross-section separating the 1.27 cm × 2.54 cm copper rails. Multiple capacitive storage banks are connected at different positions along the length of the rails. Each bank is composed of electrolytic capacitors, toroidal inductors, SCRs, and free-wheeling diodes. The primary bank (stage 1) can store 11.6 kJ, and the other stages store 5.8 kJ of energy. The diagnostics include Rogowski coils at each bank, a flux ruler along the entire railgun to monitor the armature's velocity, and B-dots. The switches are controlled by pulsers with fiber optic inputs, which accurately trigger the SCRs. To sequence trigger the storage banks, B-dot sensors provide feedback to the armature's position. Additionally, experimental results are compared to PSPICE simulations.

Authors: Thomas A. Holt; Andrew J. Young; Andreas A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4530689

Abstract: Performance reproducibility is a necessity when considering sources for single-shot, high-voltage applications. Helical Flux Compression Generators (HFCGs) are attractive for a variety of single-shot applications and are capable of high energy amplification that can be used in conjunction with other pulse-shaping techniques such as an exploding wire fuse for achieving high output voltages [1,2]. Small scale HFCGs (with active volumes on the order of ~100-200 cm3), however, are known to perform unreliably from shot to shot [3] and can lose as much as 80% of the flux available in the system based on previous experience with small to mid-sized HFCGs [4]. The performance variation is often attributed to erratic armature expansion behavior and/or fabrication methods and tolerances [3, 4]. As the compressible volume increases, HFCGs are known to conserve more flux and perform more reliably [2]. A fabrication method is presented for a midsized (with active volumes on the order of ~300-400 cm3) dual-stage HFCG that aims to improve the reproducibility in shot to shot performance with the goal of increasing the appeal for use of HFCGs in single-shot pulsed-power applications. Results of experiments with inductive loads of ~3μH are discussed.

Authors: D. Belt; J. Mankowski; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4216234

Abstract: Helical flux compression generators (HFCG) of a 50 mm form factor have been shown to produce output energies on the order of ten times the seeded value and a typical deposited energy of 3 kJ into a 3 muH inductor. Our previous work with a non-optimized fuse has produced-100 kV into a 15 load, which leads into a regime relevant for high power microwave (HPM) systems. It is expected that-300 kV can be achieved with the present 2-stage HFCG driving an inductive storage system with electro-exploding fuse. In order to optimize the electro-explosive wire fuse, we have constructed a non-explosive test bed which simulates the HFCG output with high accuracy. We have designed and implemented a capacitor based, magnetic switching scheme to generate the near exponential rise of the HFCG. The varying inductance approach utilizes 4 stages of inductance change and is based upon a piecewise linear regression model of the HFCG waveform. The non-explosive test bed will provide a more efficient method of component testing and has demonstrated positive initial fuse results

Authors: S. L. Holt; D. J. Hemmert; J. W. Walter; J. J. Mankowski; J. C. Dickens; M. Kristiansen; L. L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4530693

Abstract: Explosively driven ferroelectric generators (FEGs) are reliable, compact, high voltage sources for driving ultra-compact high power devices. When energized by mechanical stress, they release charge from their crystal structure, acting similar to a charged high voltage capacitor. However, due to the relatively low energy of these devices, they must be well matched to their application so that the full potential of the material can be utilized. This requires proper characterization of the material's electrical output as a function of the input shockwave. HEM Technologies and Texas Tech University characterized a ferroelectric ceramic, EDO EC-64, for different pressures to determine possible loss mechanisms and its maximum output power and energy.

Authors: G. Edmiston; A. Neuber; J. Krile; L. McQuage; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4216218

Abstract: One of the major limiting factors for the transmission of high power microwave (HPM) radiation is the interface between dielectric-vacuum, or even more severely, between dielectric-air if HPM is to be radiated into the atmosphere. Surface flashover phenomena which occur at these transitions severely limit the power levels which can be transmitted. It is of major technical importance to predict surface flashover events for a given window geometry, material and power level. When considering an aircraft based high power microwave platform, the effects on flashover formation due to variances in the operational environment corresponding to altitudes from sea level to 50,000 feet (760 Torr to 90 Torr) are of primary interest. The test setup is carefully designed to study the influence of each atmospheric variable without the influence of high field enhancement or electron injecting metallic electrodes.

Authors: D. Belt; J. Mankowski; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4530702

Abstract: Helical Flux Compression Generators coupled with an inductive energy storage system have shown promising results as a driving source for High Power Microwave (HPM) loads. The output performance of the inductive energy storage system is contingent upon the opening switch scheme, usually an electro-explosive fuse. Our previous work involving fuse parameter characterization has established a baseline for potential fuse performance. By applying this fuse characterization model to an HFCG powered system, a non-optimized fuse has produced 60 kV into an HPM equivalent load with an HFCG output of 15 kA into a 3 μH inductor. Utilization of a non-explosive HFCG test-bed has produced 36 kV into an HPM equivalent load with an output of 15 kA into a 1.3 μH inductor. The use of a non-explosive HFCG test bed will allow the verification of scalability of the fuse parameter model and also allow testing of exotic fuse materials. Prior analysis of fuse parameters has been accomplished with various materials including Silver (Au), Copper (Cu), and Aluminum (Al), but particular interest resides in the use of Gold (Ag) fuse material. We will discuss the a-priori calculated baseline fuse design and compare the experimental results of the gold wire material with the silver wire material baseline design. With the results presented, an accurate Pspice model applicable to our 45 kA HFCG systems will be available and allow the development of accurate modeling for higher current systems.

Authors: H. Veselka; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4216173

Abstract: Summary form only given. Investigations of shock induced conductivity of non-crystalline insulators and crystalline semiconductors were performed. Although we measured the insulator-to-metallic state transition time and conductivity, the focus of this investigation was on the recovery phase of the induced conductivity (i.e. metallic-to-insulator state transition). The recovery time and shock conditions were measured with high speed electrical diagnostics. The goal of this research is to determine the feasibility of using shock induced conductivity as a means of producing a high power opening switch. To minimize switch losses, the insulator-to-metallic transition time and conductance is also important, but has been more widely studied. Initial impact studies have shown that certain insulator can be conductive for 100 microseconds and recover under modest voltage less than one microsecond using a ten gram explosive charge. Various shock intensities are used in the study. The shock is produced primarily with conventional commercial explosives. In addition, the impact of sample thickness and compression duration on the induced the conductivity were also studied. The correlation between modest voltage and high voltage recovery time and shut-off current were studied

Authors: K. Morales; J. Krile; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4216162

Abstract: Dielectric surface flashover along insulators in vacuum has been comprehensively researched over the years. However, the primary mechanisms involved in dielectric flashover at atmospheric pressures have yet to be as extensively analyzed with variable parameters such as electrode geometry, background gas, humidity, and temporal characteristics of the applied voltage. Understanding the fundamental physical mechanisms involved in surface flashover at atmospheric pressures is vital to characterizing and modeling the arc behavior. Previous DC and unipolar excitation experiments have shown distinct arc behavior in air and nitrogen environments for an electrode geometry that produces electric field lines that curve above the dielectric surface. Specifically, flashover arcs in an air environment were observed to develop along the dielectric surface. Experiments conducted in nitrogen revealed that the arc developed along the electric field lines, above the surface of the dielectric. It was also of importance to alter the temporal characteristics of the applied voltage to simulate lightning situations and investigate the impact on the arc behavior and voltage delay times. A solid state high voltage pulser with an adjustable pulse width of ~500 ns at FWHM and amplitudes in excess of 30 kV was specifically developed to replicate the temporal characteristics of a voltage pulse observed when a building structure is hit by a lightning strike. Based on these results, the physical mechanisms primarily involved in pulsed unipolar surface flashover will be discussed. Additional studies regarding the effects of humidity and surface roughness on the flashover arc behavior will also be presented

Authors: Giesselmann, M; McHale, B; Neuber, A

PDF: https://ieeexplore.ieee.org/document/4216273

Abstract: We designed and tested several rapid capacitor chargers for rep-rated operation of low-inductance, compact Marx generators with rep-rates ranging from 10 Hz to 100 Hz. All chargers are designed to be packaged in cylindrical volumes with inside diameters in the range of 5 in - 12 in. Our capacitor chargers are based on H-Bridge inverters using ultra fast 600V class IGBTs. The high voltage is obtained by driving step-up transformers with nano-crystalline cores at 30 kHz. These chargers are capable of average DC output power levels of more than 5 kW for short time operation, during which the thermal inertia of the IGBT assembly provides effective cooling (up to seconds). To achieve reliable rep-rated operation of the chargers, we developed HV feedback sensors to monitor the charging process and solid state Marx-style trigger generators to command trigger the discharge of the main Marx.

Authors: R. W. Karhi; J. J. Mankowski; D. J. Hemmert; S. L. Holt

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4530676

Abstract: The design and implementation of a real time feedback control system for a distributed energy, bench top, electromagnetic launcher is presented. The feedback control system provides optimum pulse shaping by real time control of solid state switches. Advantages of pulse shaping control include increased energy efficiency and control of armature exit velocity. Lab VIEW 8.0 software¹ is used to program a National Instruments CompactRIO programmable automation controller (PAC). This provides real time processing by use of the reconfigurable I/O (RIO) FPGA technology. The program controls switch timing from analog feedback signals supplied by B-dot probes placed along the rail length. Through signal analysis, real time armature position is derived. The program uses this data to control pulse shape and width. A dedicated B-dot probe is placed at the beginning of each stage which is the desired triggering location. A flux ruler sensor along the bore length provides a secondary velocity calculation excluded from the control system. This sensor provides velocity measurements for every centimeter of bore travel. Collected data is used to characterize the system under test for different load conditions.

Authors: J. T. Krile; A. A. Neuber; G. F. Edmiston; H. G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4216123

Abstract: Flashover along insulators or insulating support structures has to be carefully addressed in the design of any DC, AC, or pulsed high voltage device. Although there is a large body of data on unipolar surface flashover in the atmosphere, which has led to empirical design rules primarily for the power distribution industry, the physics of the involved processes is widely unknown. The major limiting factor in the transmission of high power microwaves (HPM) into the atmosphere has been the vacuum-air interface. Both the unipolar and HPM surface flashover cases have been studied under vacuum conditions and have been found to have the same dominant mechanisms. Similarities between HPM window flashover on the air side and unipolar flashover are observed in an atmospheric environment as well

Authors: D. Wetz; J. Mankowski; D. McCauley; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4216144

Abstract: Experimental results are presented on the impact water conductivity, electrode material, and electrode surface roughness have on the dielectric strength of water. A 2 mm water gap was tested under pulsed conditions with maximum electric fields in excess of 1.5 MV/cm. Bruce profiled electrodes manufactured from various materials including molybdenum, aluminum, copper, tungsten, nickel, stainless steel, and zinc oxide, which have work functions ranging from 4.2 eV to 5.3 eV, have been tested. The roughness average of the various surfaces was varied from 0.1 mum to 1.96 mum and the conductivity of the water within the gap was varied from 1muS/cm to 39 muS/cm to determine the impact each has when tested in combination with the various electrode materials and surface roughnesses. Conclusions are made as to the effect each of these factors has on the holdoff strength of a water dielectric system

Authors: H.C. Kim; J.P. Verboncoeur; G.F. Edmiston; A.A. Neuber; Y.Y. Lau; R.M. Gilgenbach

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1666169

Abstract: In high-power microwave systems, we investigate the transition of breakdown from single surface vacuum multipactor discharge to collisional rf plasma in argon. As the gas pressure increases, electron-neutral collisions prevail against secondary electron emissions. In addition, the discharge formation time is obtained as a function of the gas pressure

Authors: H. Krompholz; L. Hatfield; A. Neuber; J. Chaparro; H.-y. Ryu; W. Justis

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4105497

Abstract: Gas breakdown in quasi homogeneous electric fields with amplitudes of up to 3 MV/cm is investigated. The setup consists of a RADAN 303 A pulser and pulse sheer SN 4, an impedance-matched oil-filled coaxial line with a lens-transition to a biconical line in vacuum or gas, and an axial or radial gap with a width on the order of mm, with a symmetrical arrangement on the other side of the gap. Capacitive voltage dividers allow to determine voltage across as well as conduction current through the gap, with a temporal resolution determined by the oscilloscope sampling rate of 20 GS/s and an analog bandwidth of 6 GHz. The gap capacitance charging time and voltage risetime across the gap is less than 250 ps. Previous experiments at TTU with a slightly larger risetime have shown that breakdown is governed by runaway electrons, with multi-channel formation and high ionization and light emission in a thin cathode layer only. In argon and air, time constants for the discharge development have been observed to have a minimum of around 100 ps at several 10 torr. A qualitative understanding of the observed phenomena and their dependence on gas pressure is based on explosive field emission and gaseous ionization for electron runaway conditions.

Authors: Y. Chen; J. Mankowski; J. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4216158

Abstract: Recent research efforts at TTU include a simple aluminum cathode for E-beam generation in a virtual cathode oscillator (vircator). As a continuation of these results, we are fabricating and testing several different types of cathodes for the vircator driven by a Marx generator and a pulse forming line (400 kV, 60 ns, 30 Omega). The cathodes, each with an emitting area of ~32 cm2, are made from aluminum and oxygen free copper fashioned to similar geometries with either a chemical etch or a CNC machine process. The vircator is tested with all of these cathodes and evaluated for beam voltage, current, and HPM generation. In addition to the cathode testing, several stainless steel and tungsten anode meshes with varying transparencies (50%-80%) are evaluated

Authors: A. A. Neuber; Y. J. Chen; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084440

Abstract: The electrical characteristics and design features of a low inductance, compact, 500 kV, 500 J, 10 Hz repetition rate, Marx generator are discussed. While benefiting from the large energy density of mica capacitors, 4 mica capacitors were utilized in parallel per stage, keeping the parasitic inductance per stage low. Including the spark- gap switches, a stage inductance of 55 nH was measured, which translates with 100 nF capacitance per stage to ~18.5 Omega characteristic Marx impedance. Using solely inductors, ~1 mH each, as charging elements instead of resistors enabled charging the Marx within less than 100 ms with little charging losses. The pulse width of the Marx into a matched resistive load is about 200 ns with 50 ns rise-time. Repetitive HPM generation with the Marx directly driving a small Vircator has been verified. The Marx is fitted into a tube with 30 cm diameter and a total length of 0.7 m. We discuss the Marx operation at up to 21 kV charging voltage per stage, with repetition rates of up to 10 Hz in burst mode primarily into resistive loads. A lumped circuit description of the Marx is also given, closely matching the experimental results.

Authors: J. Mankowski; Y. Chen; J. Dickens; A. Neuber; R. Gale

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084154

Abstract: Recent research efforts at TTU include the testing of a rapidly charged, rep-rated Marx generator driving a reflex triode vircator [1]. As expected, the burning of the cathode material (ordinary cloth velvet), was the primary failure mechanism during repetitive operation. In an effort to achieve a repetitive vircator (>10 Hz), we are exploring a low-cost, all-metal cathode. The cathode is made from aluminum with a patterned surface. A typical pattern is composed of peaks and troughs with dimensions on the order of tens of microns. The pattern is achieved with a simple, low-cost chemical etching process. Results include current, voltage, and microwave waveforms from two solid metal cathodes and a cloth velvet cathode.

Authors: D. Wetz; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4198346

Abstract: Summary form only given. In the past there has been considerable research performed using electromagnetic plasma thrusters as the primary ion source in sputtering applications. In the experiments conducted here, thin film coatings of various hard metallic alloys and oxides were applied to stainless steel electrodes using ion beam deposition. An SPT type Hall effect thruster capable of producing an ion beam with energies between 150 eV and 350 eV was used as the deposition ion source. The surface coatings were applied to Bruce profiled stainless steel electrodes which have an effective surface area of 5 cm2 and a roughness average of roughly 70 nm. The electrodes were then examined using an optical profiler to accurately measure the roughness average of the thin film coating as well as the effective layer thickness. A comparison of the results obtained from each surface material is performed

Authors: W. J. Carey; A. J. Wiebe; D. D. Schwindt; L. L. Altgilbers; M. Giesselmann; B. McHale; K. Heinemann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084191

Abstract: The development of an autonomous RF radiation package for various applications is presented. This work is a coordinated effort to develop a tightly integrated unit, including the batteries, power supply, Marx generator, and plug and play antennas for various applications. ARC technology has designed the Marx generator and its associated high voltage antennas for this effort. Previous work by ARC has demonstrated 75 mm diameter, 700 mm length diameter Marx generators capable of delivering 200 kV pulses into 50 Omega coaxial cable with sub-nanosecond risetimes, enabling it to drive an antenna and generate high power microwaves. This technology has been re-designed into a reduced length geometry and augmented by inductive charging to permit pulse repetition rates. The antenna is incorporated directly onto the Marx output for efficient energy transfer and for compactness. This package has demonstrated peak electric field strengths up to 4700 V/m at 10 m. Texas Tech University has worked closely with ARC in developing a rapid charging power supply to meet stringent package constraints and still permit high pulse repetition rates. This system has already demonstrated the ability to charge a 50 nF capacitance up to 40 kV with a repetition frequency of 100 Hz, delivering an average power of 4 kW. This paper details the present status of the project, which will be completed in July, 2005. The cylindrical geometry of the final package has a diameter of 155 mm, a length of approximately 1500 mm without the antenna, and a mass of approximately 35 kg, depending upon the chosen antenna implementation. Results of preliminary tests are included.

Authors: J. Walter; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084153

Abstract: The application of pulsed power to transient RF/microwave generation for warhead/projectile payloads is currently a significant research area. In this application, traditional fixed antenna designs have two major drawbacks: the size required for efficient radiation can be prohibitive for frequencies much less than 1 GHz, and the fixed antenna can make the device vulnerable to electromagnetic counterattack. One frequently proposed solution to both issues is the use of transient antennas. To effectively integrate these devices, the various types of transient antennas must be characterized over a wide frequency band during their transient formation period. A testing method being developed at the Center for Pulsed Power and Power Electronics at Texas Tech University utilizing a broadband transmitter and receiver is described.

Authors: M. Cevallos; M. Butcher; J. Dickens; A. Neuber; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084448

Abstract: The development of cathode initiated low density channel formation and propagation leading to self breakdown in transformer oil is investigated using high speed electrical and optical diagnostics in a coaxial system with a point/plane axial discharge at various hydrostatic pressures. A cathode initiated channel formation and propagation model based upon single bubble dynamics has been presented by Kattan [1]. Experiments based on high-speed shadowgraphy were conducted to decide if the principles that govern single bubble dynamics could be applied to cathode initiated channel formation. These experiments show bubble motion away from the cathode, with separation velocities on the order of 10's m/s. This separation is similar for single bubbles generated at the cathode and for bubble chains developing into low-density channels. Lifetimes of these channels are recorded and show good correlation with the Rayleigh model [2] used to predict lifetimes of single bubbles. Experiments at reduced hydrostatic pressure reveal a critical pressure below which low density channel expansion occurs, further corroborating the presence of a gas phase. Finally, the pressure dependence of the breakdown voltage due to the expansion of the low density channels is examined and a model for this dependence is presented. The experiments conducted confirm the presence of a gas phase channel, its correlation with single bubble dynamics, and its importance to final breakdown.

Authors: M. Cevallos; M. Butcher; J. Dickens; A. Neuber; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084384

Abstract: The physics of cathode initiated discharge formation leading to self breakdown in transformer oil is investigated using high speed electrical and optical diagnostics in a coaxial system with a point/plane axial discharge. Previous research conducted on self breakdown channel formation using high speed shadowgraphy and photography of the emitted light has shown tree-like structures for both cathode and anode initiated discharges, with characteristic differences. Cathode initiated discharges expand faster to a more "bushy" appearance, whereas anode initiated discharges show branching localized channels. So far, the spatial resolution to detect small luminous areas in pre- breakdown discharges and to determine their correlation to low density regions visible in the shadowgraphs was not sufficient in the experiments described in this paper, thus a systematic variation of exposure times, and time delays between luminosity pictures and shadowgraphy pictures has been performed. These experiments confirm that the luminosity emitted during prebreakdown events is generated from the low density regions seen in the shadowgraphy images, indicating charge amplification mechanisms in the gas phase for cathode initiated events. This statement is further supported by the dependence of both the channel dynamics and the light emission at lowered hydrostatic pressure.

Authors: S. Holt; J. Dickens; J. Walter; S. Calico

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084248

Abstract: The design and testing of explosive-driven ferroelectric generators (FEGs) with output voltages exceeding 200 kV is presented. The generator design is aimed at achieving high voltages in a compact device by explosively compressing stacks of ferroelectric ceramic discs. The ferroelectric used in this application is EC -64 lead zirconate- titanate (PZT), a hard ferroelectric ceramic with a high dielectric constant and a good piezoelectric coupling coefficient in the direction of polarization. The pressure impulse is generated by high explosive detonating cord and travels longitudinally along the polarization vector in the PZT. The effect of variations in the rise-time, width, and magnitude of the pressure pulse on the peak voltage has been experimentally studied by modifying the ferroelectric stack and explosive driver geometries. Different dielectric insulations are experimentally evaluated for good compatibility with the ferroelectrics and maximum hold-off voltage.

Authors: G. Edmiston; J. Krile; A. Neuber; J. Dickens; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084224

Abstract: The major limiting factor in the transmission of HPM has been the interface between dielectric-vacuum or even more severely between dielectric-air if HPM is to be radiated into the atmosphere. Extensive studies have identified the physical mechanisms associated with vacuum-dielectric flashover, as opposed to the mechanisms associated with air-dielectric flashover, which are not as well known. Surface flashover tests involving high field enhancement due to the presence of a triple point have shown that volume breakdown threshold (dielectric removed) is approximately 50% higher than the flashover threshold with a dielectric interface over the 90-760 torr range [1]. In order to quantify the role of field enhancement in the flashover process independent of electron injection from metallic surfaces, the effects of the triple point are minimized by carefully choosing the geometry and in some cases the triple point is "removed" from the flashover location. We will present experimental results, including the impact of gas pressure, and discuss possible causes for the difference in the rf-breakdown field with and without the interface/metallic triple point portion.

Authors: M. Giesselmann; B. Palmer; A. Neuber; J. Donlon

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4084329

Abstract: We are reporting on a High-Voltage Impulse Generator, which consists of a step-up transformer, which is driven by new HV-IGBTs (High-Voltage Isolated Gate Bipolar Transistors). The new HV-IGBTs are individually packaged silicon-dies intended for Pulsed-Power Applications. The silicon dies are normally packaged in large modules for locomotive motor drives and similar traction applications. In our work we used the Powerex QIS4506001 discrete IGBT and the QRS4506001 discrete diode, both with a nominal rating of 4500V/60A, derived from continuous- duty applications. Our experiments have shown that the devices are capable of handling currents in excess of 1 kA during pulsed operation.

Authors: M. Butcher; M. Cevallos; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084425

Abstract: With a fast coaxial setup using a needle/plane geometry and a high sensitivity electrometer, conduction mechanisms in transformer oil at varying temperature and hydrostatic pressure are quantified. There are 3 stages in the conduction process prior to breakdown for highly nonuniform field geometries. Stage I is characterized by a resistive current at low fields. Stage II consists of a rapid rise in the injection current associated with increasing field due to a "tunneling" mechanism through the metal/dielectric interface. The transition from the resistive to tunneling stage occurs when the applied field reduces the barrier at the metal/insulator interface to a point where tunneling of charge carriers through the barrier begins. This transition point is polarity dependent. In stage III, at high fields the current reaches space charge saturation at electron mobilities >100 cm2/V*s prior to breakdown. The processes of final breakdown show distinct polarity dependence. Data for the negative needle exhibits strong pressure dependence of the breakdown voltage, which is reduced by 50% if the hydrostatic pressure is lowered from atmospheric pressure to hundreds of mtorr. Such a strong pressure dependence, at reduced hydrostatic pressure, indicates breakdown is gaseous in nature. This is supported by images of bubble/low density regions forming at the current injection point. Positive needle discharges show a reduction of only about 10% in breakdown voltage for the reduced pressure case. A weak pressure dependence indicates the breakdown mechanism does not have a strong gaseous component. We will discuss possible links between conduction current and DC breakdown.

Authors: Xupeng Chen; James C. Dickens; L. L. Hatfield; John Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4084184

Abstract: In virtual cathode oscillator (VIRCATOR) research, multi-frequency and multi-mode microwaves are coupled out and propagate along a waveguide. Identifying the microwave propagating modes has become a priority work for further research on microwave radiation physical mechanisms. Generally, an antenna matrix at the end of an open waveguide, which can shape the radiated microwave power distribution, is used to identify the microwave propagating modes [1-2]. Actual microwave radiation mechanisms in a VIRCATOR are complicated. Multi-propagating modes and multifrequency microwaves coexist simultaneously, which challenges the antenna matrix method. In this paper, based on microwave propagating theory in a waveguide, a new method is proposed to identify multi-frequency and multi-mode microwaves, which is partially justified by the experimental data at Texas Tech University.

Authors: D. Belt; J. Dickens; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084267

Abstract: Helical flux compression generators of small dimensions have been shown to produce energy output around 3 kJ into an inductive load. Adding a fuse opening switch has allowed us to produce 300 kV into a 15 Ohm load. We are investigating inductive energy storage with emphasis on an electro-explosive fuse opening switch in order to improve upon previous results. We have designed and constructed a non-explosive test bed composed of two pulse forming networks (PFN). Each PFN provides a linear approximation during two different time ranges of the exponential rise response of a typical HFCG. This approach will be more cost and time effective than to drive the fuse with an explosive generator. Our initial goal will be to simulate a 15 kA HFCG unit followed by the simulation of a 50 kA HFCG.

Authors: K. P. Morales; J. T. Krile; A. A. Neuber; H. G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084426

Abstract: Dielectric flashover along insulators in vacuum has been sufficiently researched in the past. Less studied, but of similar importance, is surface flashover at atmospheric pressures and the impact of various electrode geometries, humidity, and type of gas present. Previous research has shown distinct arc behavior in air and nitrogen for an electrode geometry in which the electric field lines curve above the dielectric surface. Specifically, flashover experiments in nitrogen have shown that the arc path will follow the electric field lines, not the dielectric surface. As a result, it was concluded that the arc development path, whether along the electric field line or the surface of the dielectric, is related to the oxygen content in the atmospheric background. It is believed that this dependence is due to the arc's production of UV radiation in an oxygen rich environment. Further testing, in a pure nitrogen environment with UV illumination of the surface prior to the pulse application, has shown that UV plays a significant role in the arc development path. There is a near linear relationship between the percentage of liftoffs and the time delay between UV application and flashover. Additional studies have also shown a relationship between the UV intensity and the percentage of liftoffs. Based on these results we will discuss the physical mechanisms primarily involved in unipolar flashover at atmospheric pressure. Additional experimental results regarding the effects of humidity on the liftoff phenomenon will be presented as well.

Authors: M. Giesselmann; B. McHale

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4084435

Abstract: We designed and build a rapid capacitor charger for 10 Hz, 500 J/shot operation of a low-inductance, compact Marx generator. The charger uses a hard-switched IGBT H-Bridge Inverter, which drives a 30 kHz, nano- crystalline step-up transformer. The transformer, in addition to the high-voltage rectifier and a trigger- transformer are contained in a section which is filled with transformer oil. The main circuit board also contains a solid-state Marx generator to trigger the main Marx generator. We also implemented a self-powered HV-feedback sensor to stop the charge process precisely at the target voltage. This new sensor greatly enhanced the rep-rated performance of the Marx by preventing pre-fires, since it enabled us to charge aggressively without overshooting the target voltage and have more time for spark-gap recovery.

Authors: H. Veselka; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084322

Abstract: Investigations of shock induced conductivity of insulators and semiconductors both crystalline and noncrystalline, were performed. Although the insulator-to- metallic state transition time was measured, the focus of this investigation was on the recovery phase of the induced conductivity (i.e. metallic-to-insulator state transition). The recovery time and shock conditions were measured with high speed electrical diagnostic equipment. The goal of this research is to determine the feasibility of using shock induced conductivity as a means of producing a high power opening switch. To minimize switch losses, the insulator-to-metallic transition time and conductance is also important, but has been more widely studied. Various shock profiles and intensities and used in the study. The shock is produced primarily with conventional commercial explosives. Shock reverberation is planed in some tests to minimize sample heating and to produce tailored shock time profiles. In addition, the impacts of sample thickness and compression duration on the induced conductivity were also studied.

Authors: J. Krile; G. Edmiston; A. Neuber; J. Dickens; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084225

Abstract: Mechanisms in vacuum flashover caused by rf (f<10 GHz) or unipolar voltages are virtually identical. Similarities between rf (representing high power microwave window breakdown on the high pressure side) and unipolar flashover are expected in an atmospheric environment as well. Our experimental setups enable studying both unipolar flashover and rf window flashover at atmospheric conditions while controlling excitation, temperature, pressure, humidity, and type of gas present. The local electric field at the flashover initiating points has been numerically calculated in detail for all test geometries. For both rf and unipolar pulsed excitation, the flashover dynamics are changed by the application of UV light to the dielectric surface. A UV pre-pulse has a distinct impact on the arc's path and a tendency to increase the hold-off electric field. The effect of humidity on the hold-off electric field for both pulsed unipolar and rf excitations, along with temporally resolved emission spectroscopy of the flashover event, will be discussed.

Authors: J. Qian; R. P. Joshi; J. Kolb; K. H. Schoenbach; J. Dickens; A. Neuber; M. Cevallos; H. Krompholz; E. Schamiloglu; J. Gaudet

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084323

Abstract: An electrical breakdown model for liquids in response to a sub-microsecond (~ 100 ns) voltage pulse is presented, and quantitative evaluations carried out. It is proposed that breakdown is initiated by field emission at the interface of pre-existing micro-bubbles. Impact ionization within the micro-bubble gas then contributes to plasma development, with cathode injection having a delayed and secondary role. Continuous field emission at the streamer tip contributes to filament growth and propagation. This model can adequately explain almost all of the experimentally observed features, including dendritic structures and fluctuations in the pre- breakdown current. Two-dimensional, time-dependent simulations have been carried out based on a continuum model for water, though the results are quite general. Monte Carlo simulations provide the relevant transport parameters for our model. Our quantitative predictions match the available data quite well, including the breakdown delay times and observed optical emission.

Authors: H. Krompholz; L.L. Hatfield; A. Neuber; D. Hemmert; K. Kohl; J. Chaparro

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4084242

Abstract: Volume breakdown and surface flashover in quasi homogeneous applied fields in 10-5 to 600 torr argon are investigated, using voltage pulses with 150 ps risetime, < 1 ns duration, and up to 150 kV amplitude into a matched load. The test system consists of a transmission line, a transition to a biconical section, and a test gap, with gap distances of one to several mm. The arrangement on the other side of the gap is symmetrical. An improved system, with oil-filled transmission lines and lens between coax and biconical section to minimize pulse distortion, is being constructed. Diagnostics include fast capacitive voltage dividers, which allow to determine voltage waveforms in the gap, and conduction current waveforms through the gap. X-ray diagnostics uses a scintillator- photomultiplier combination with different absorber foils yielding coarse spectral resolution. Optical diagnostics to obtain information about the discharge channel dynamics is in preparation. Breakdown delay times, and e-folding time constants for the conduction current during the initial breakdown phase, are on the order of 100-400 ps, with minima in the range of several 10 torr. X-ray emission extends to pressures > 100 torr, indicating the role of runaway electrons during breakdown. Maximum X-ray emission coincides with fastest current risetimes at several 10 torr, which is probably related to an efficient feedback mechanism from gaseous amplification to field enhanced electron emission from the cathode.

Authors: X. Chen; J. Mankowski; L. L. Hatfield; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084185

Abstract: In virtual cathode oscillator (VIRCATOR) research, the radiated microwave pulse is coupled out and propagates along a waveguide. The radiated microwave power is an important parameter to evaluate the oscillator design and is a prerequisite parameter to calculate the energy transferring efficiency. Because the radiated microwaves are an ultra-short pulse of multi-frequency in multi propagating modes, the power measurement is not easily manipulated. In this paper, a broadband calorimeter design with considerations of pulsed microwaves is presented. The design includes its material or medium selection, structure optimization and calibration.

Authors: D. Wetz; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084430

Abstract: Experimental results are presented on the degree to which electrode surface area and surface roughness impacts the dielectric strength of water. A 2 mm water gap was tested under pulsed conditions with maximum electric fields in excess of 1 MV/cm and maximum currents of more than 5 kA. Six different pairs of stainless steel electrodes, each having a unique Bruce profile and thus a different effective surface area, were used to achieve a uniform electric field across the gap. The differing electrode pair profiles, with effective areas ranging from 0.5 cm2 to 76 cm2, were designed to minimize the change in gap capacitance. Prior to each test, a different roughness average, ranging from 0.26 μm to 1.96 μm, was applied to the electrodes. Conclusions are made as to the effect both electrode surface area and surface roughness has on the holdoff voltage of water dielectric systems. In addition, shadowgraph images of pre-breakdown events are presented.

Authors: D. Wetz; J. Mankowski; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=4084372

Abstract: In the experiments presented here, various electrode materials were tested in an effort to determine the impact each has on increasing the dielectric strength of water. Prior investigations have tested materials such as stainless steel, copper, nickel, gold, silver, and cuprous oxide [1-4]. In our experiments, thin film coatings of various metallic alloys and oxides were applied to Bruce profiled stainless steel electrodes with an effective area of 5 cm2. An ion beam sputtering process was used to apply the coatings with thicknesses of several hundred nm. The electrodes were then tested across a water gap, with pulse lengths in both the microsecond and nanosecond time regimes. Electric fields in excess of 8 MV/cm were applied. Conclusions are made as to the impact electrode material has on the pulsed breakdown strength of water.

Authors: A. Neuber; J. Krile; J. Dickens; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1433496

Abstract: In a wide variety of high voltage applications surface flashover plays a major role in the system's performance, and yet it has not been studied in great detail for atmospheric conditions with modern diagnostic tools. Environmental conditions to be considered include pressure, humidity, and gas present in the volume surrounding the dielectric. In order to gain knowledge into the underlying process involved in dielectric surface flashover, a setup has been created to produce and closely monitor the flashover event. Within the setup, parameters such as geometry, material, and temporal characteristics of the applied voltage can be altered. Current, voltage, luminosity, and optical emission spectra are measured with nanosecond to subnanosecond resolution. Spatially and temporally resolved light emission data is also gathered along the arc channel

Authors: M. Butcher; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1339891

Abstract: Summary form only given. Any attempt to model the complex interaction of hydrodynamic and electronic processes leading to breakdown in transformer oil suffers from the lack of microscopic transport data. Also, interface processes, such as electron emission from metal electrodes immersed in liquid, are poorly understood. As a first step toward the understanding of breakdown phenomenology, the voltage-current characteristics for pre-discharge conditions are measured. An experimental setup was constructed which allows temperature variations between 10/spl deg/C and 50/spl deg/C, at pressures between 0.5 and 3 bar. DC currents ranging from a few nA with a few kV of applied voltage, to a few /spl mu/A prior to full breakdown are measured using an electrometer. Preliminary results at NTP with a tip-plane geometry indicate Ohmic behavior at low voltages, Schottky emission at intermediate voltages, and saturation due to space charge at high voltages, and allow estimates on the physical parameters governing these effects. The indicated temperature range of the measurements is associated with a variation of the viscosity of a factor of 3, where we anticipate similar relative changes for the transport of electrons. The intermediate voltage range where a Schottky emission process is assumed will be emphasized. For instance, the electron mobility, derived from experimental data to about 0.06 cm/sup 2//Vs at NTP in the intermediate voltage range should distinctly vary with changing temperature and pressure.

Authors: M. D. Cevallos; J. C. Dickens; A. A. Neuber; H. G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1340174

Abstract: Summary form only given. The fundamental breakdown physics of transformer oils is investigated with high-speed electrical and optical diagnostics with temporal resolution down to 500 ps. Univolt 63 and Envirotemp FR3 (biodegradable) are used for this study. The system set up employs a cable discharge into a coaxial system with point/plane axial discharge and load line to providing a matched terminating impedance. Overall, the impedance of the system is matched at 50 ohms throughout with the exception of the very narrow gap region and includes a 50 ohm load resistor terminating the load line. Self breakdown is achieved by applying up to 50 kV to the charging line. Pulsed breakdown is achieved by charging a pulse forming line with a two way transient time of 300 ns, up to 100 kV. The pulse forming line is then fed into the discharge line via an oil spark gap. Transmission line type current sensors and a capacitive voltage divider with fast amplifiers/attenuators are used in order to attain a complete range of information from amplitudes of 0.1 mA to 1 kA with temporal resolutions of 300 ps. Optical measurements are performed on low level light emission using fast photo-multiplier tubes (risetime of 800 ps) spatially resolved, supplemented with high speed and spectroscopic investigations on a nanosecond timescale. Breakdown voltages at gap distances of 5 mm for pre, self, and pulsed breakdown voltages are given, where breakdown with negative needle tips show 20% higher breakdown voltages than the ones with positive needle. Velocities of propagating "tree"-structures for the pre and self-breakdown are a few km/s while pulsed breakdown velocities are several 10's of km/s, with higher velocities for a negative needle. The trees have to reach the opposite electrode before full breakdown occurs. Simultaneous optical measurements for a single breakdown event are presented, such as the luminosity in comparison to shadowgraphy images, which is necessary to describe the complex interaction of hydrodynamic phenomena (channel and tree formation and propagation) and charge carrier multiplication.

Authors: John J. Mankowski; Xupeng Chen; James C. Dickens; Magne Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6220572

Abstract: Experimental findings on a high power Reflex Triode Virtual Cathode Oscillator (Vircator) at Texas Tech University are reported. In order to optimize performance, changes to the AK gap distance were made. The AK gap was varied from 0.5 to 1.5 cm. The type of cathode material employed was velvet. Cathode radius was 2.5 cm. The Reflex Triode Vircator is driven by a 350 kV Marx generator with a 60 ns, 20 ohm pulsed forming line. Typical peak input power is 4 GW. The output microwave power is measured in the far-field with a waveguide to coax adapter. Thus far we have observed microwave peak power output as high as 330 MW corresponding to a peak power efficiency of ~ 11 %. In an effort to provide premodulation to the vircator region, reflecting strips were added in the downstream waveguide. This technique has been employed with success by this group with a coaxial Vircator.

Authors: Giesselmann, M; McHale, B; Crawford, A

PDF: https://ieeexplore.ieee.org/document/1398070

Abstract: This paper describes work towards transient energy extraction from high frequency alternators to drive electromagnetic launchers through intelligent triggering of thyristors/Silicon Controlled Rectifiers (SCRs). The work was done using a 3-phase, 75 kVA level motor-generator set with a DC machine acting as the load. This system serves as a scale model to develop control techniques for full scale, multi-megajoule applications. With the use of a microcontroller and advanced sensors the system is able to maintain synchronization of the SCR converter with the generator voltage despite dynamically changing frequency and voltage distortion.

Authors: Giesselmann, M; Edwards, R; Bayne, S; Kaplan, S; Shaffer, E

PDF: https://ieeexplore.ieee.org/document/1433587

Abstract: The charged controlled model is presented as an effective method to simulate junction barrier schottky (JBS) silicon carbide diodes. Proven as a valuable approach for silicon devices, this model can also account for wide bandgap energy semiconductors. The model was implemented in Orcad's SPICE software package using analog behavioral modelling. The simulation combines the efficiency of a traditional static model with an improved charge controlled model representing transient characteristics of the semiconductor due to stored charge in the depletion layer. The result is an enhanced model that simulates in short period of time. Parameter extraction methods are used to help identify saturation currents, capacitance, resistance, voltages, and modelling coefficients. The simulation results in a close fit to the data taken during reverse recovery of a JBS SiC diode.

Authors: D. Wetz; J. Mankowski; K. Truman; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1433626

Abstract: Experimental results are presented on the impact electrode material and surface treatment has on the dielectric strength of water. A 4 mm water gap was tested under pulsed conditions with pulse widths greater than 1 /spl mu/s, peak electric fields over 2 MV/cm and peak currents over 15 kA. 304 stainless steel electrodes machined with a Bruce profile and an effective area of 5 cm/sup 2/ were tested with surface roughnesses ranging from .34 /spl mu/m to 1.41 /spl mu/m. Results comparing electrodes that have additionally been electropolished are compared to those without an additional treatment. These various surface treatments remove microprotrusions from the electrode's surface which reduce field enhancements across the electrode surface. It is believed that this technique will minimize the number of breakdown initiation points thereby increasing the dielectric strength. Results are given on the degree to which each finish improves the dielectric strength of water.

Authors: T. Namihira; D. Wang; A. Neuber; M. Butcher; J. Dickens; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1340266

Abstract: Summary form only given. Considering highly stressed dielectric liquids, the role of mechanisms such as photoionization in the liquid volume or photoeffect at the cathode for the development of dielectric breakdown is investigated. We used a pulsed 300 W Xenon light source (25 mm output window, 5 degree divergence) with a broad spectral range of 200 to 1100 nm to study the impact of the light beam focused either solely on the high field region between the breakdown electrodes or including the electrodes. Typical field strengths in the electrode gap (/spl sim/4 mm gap, 3 mm tip radius,) were 15 to 25 kV/cm resulting in a DC current amplitude (without light) of up to 2 nA (apparatus resolution /spl sim/10 pA). Standard transformer oil, Univolt 61, and a biodegradable oil, Environtemp FR3 (natural ester fluid), were examined in the present work. Both oils exhibit strong optical absorption in the UV. However, Univolt 61 has its cut off wavelength at 450 nm, while bio oil easily transmits down to 350 nm. Below the cutoff wavelength, virtually all radiation is absorbed within a few mm. When pulsing the Xenon lamp at /spl sim/500 microsec no increase in DC current amplitude (increase <10 pA) could be detected for either oil. Increasing the pulse length to several seconds lead to a distinct increase in current amplitude (up to 300 pA), however, only for Univolt 61. Such an increase in current amplitude can also be achieved by raising the temperature of the dielectric liquid by external heating (/spl sim/100 pA/K). The temperature levels leading to similar current amplitudes due to heating by the Xenon lamp or external heating are comparable. Since bio oil absorbs only below 350 nm, the temperature rise due to the light irradiation was comparably smaller than in Univolt 61. Thus, any heating and increase in current were less pronounced in bio oil. For both oils, the observed behavior can be entirely explained by thermal effects. Both, photoionization and photoeffect have seemingly a minor impact on breakdown development. The detailed discussions are given in the present work.

Authors: Giesselmann, M; McHale, B; Heeren, T

PDF: https://ieeexplore.ieee.org/document/1433595

Abstract: This paper presents new advances in High-Voltage power supplies, which are used for rapid capacitor charging. We are developing very compact power supplies with input voltages of around 250 V and output voltages of 30-50 kV. The output power levels of these chargers are above 10 kW. These chargers are using state-of-the-art Isolated Gate Bipolar Transistors (IGBTs) as well as advanced packaging methods and other innovative circuitry. In addition, the size and weight of the step-up transformer has been significantly reduced. Design details as well as experimental results are presented.

Authors: E. Crull; H. Krompholz; A. Neuber; L. Hatfield

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1339921

Abstract: Summary form only given. Fast gas breakdown with formative times in the sub-nanosecond regime is of interest for pulsed power switching and UWB applications. Use of coaxial transmission lines with conical sections connected to a test gap enables to apply fast voltage pulses to the gap, as well as the simultaneous measurement of voltage across and current through the gap. For small pulse amplitudes, with risetimes of 400 ps, a tip-plane geometry is used, with radii of curvature of 0.5 /spl mu/m. At pulse amplitudes of 5 kV, and macroscopic field enhancements on the order of 1000, delay times between current and voltage of less than 200 ps for pressures larger than 100 torr are observed, in both argon and dry air. Corresponding current risetimes I/(dI/dt) are less than 100 ps. Using a high voltage pulser (RADAN 303B with pulse slicer SN4, risetime 150 ps at 150 kV amplitude) enables the comparison of formative times for the tip-plane geometry with those of more homogeneous field distributions in the gap.

Authors: K. Truman; J. Mankowski; D. Wetz; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1433523

Abstract: Experimental results are presented on the degree to which electrode surface area impacts the dielectric strength of water. A water gap of 4 mm was tested under pulsed conditions with a maximum electric field in excess of 1 MV/cm and a maximum current of more than 20 kA. Stainless steel electrodes with a Bruce profile were used to generate a uniform electric field across the water gap. The profile of the electrodes enabled effective areas ranging from 0.5 cm2 to 75 cm2 while minimizing the change in capacitance of the water gap. Conclusions are made as to the effect electrode surface area has on the holdoff voltage and holdoff time of water dielectric systems

Authors: M. Butcher; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1339892

Abstract: Summary form only given. The scattering cross-sections, transport coefficients, and details of the breakdown mechanism needed for describing current conduction in transformer oil at high applied voltages are virtually unknown. This makes it extremely difficult, if not impossible, to properly model electrical breakdown in oil. To address this issue, we have measured the V-I characteristic of transformer oil in the point-plane geometry for /spl sim/3 mm gap widths. Three regimes can be distinguished. (1) For low voltages, V <3 kV, the relationship between voltage and current is linear, exhibiting just resistive behavior within the measurement accuracy. (2) At intermediate voltages, the dependence is linear on a Fowler-Nordheim plot. (3) Between the breakdown voltage V/sub B/ /spl sim/ 30 kV and 10 kV, the current is proportional to V/sup 2/, indicating space charge limited current. Assuming a Schottky-type emission mechanism for electrons injected into the liquid, we derive an electron mobility of /spl sim/6.0/spl times/10/sup -6/ m/sup 2//Vs, which is close to values reported in the literature before. Assuming Fowler-Nordheim emission leads to 3.8/spl times/10/sup -6/ m/sup 2//Vs. The ion mobility is believed to be more than one order of magnitude lower. At this point, it is difficult to gain any more detailed information on charge transport and possible multiplication based on simple analytical methods. Hence, we applied our numerical modeling techniques already proven in analyzing high-field phenomena in polar liquids (e.g., water), to evaluate current conduction and the breakdown process in non-polar oil. By comparing our calculations with the experimental data, we hope to characterize important transport parameters, such as the electron ionization coefficient as a function of the electric field. We also present our attempts to extract details of the field-dependent non-linear processes and electrode effects close to the breakdown regime.

Authors: Xupeng Chen; J. Dickens; Eun Ha Choi; J. Mankowski; L.L. Hatfield; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1278019

Abstract: Traditionally, the radiated microwave frequency in a coaxial vircator is considered to be determined primarily by the virtual cathode oscillation frequency and the electron reflection frequency. However, some experiments at TTU show that the interaction between the virtual cathode oscillation and the cavity is a key in determining the microwave frequency and propagating mode. Particularly, we observe that the E-beam plays an important role in the cavity formation. Some possible explanations, including a virtual cavity concept, are proposed. The cavity resonance effect on a coaxial virtual cathode oscillator has been investigated in detail. The understanding of the E-beam characteristics is critical in optimizing the design of the cylindrical diode to avoid the microwave frequency shifting and mode competition.

Authors: Gundersen, M; Dickens, J; Nunnally, W

PDF: https://ieeexplore.ieee.org/document/1277651

Abstract: We describe an academic research program designed to investigate fundamental issues that will impact future pulsed power technology. The project is intended to enhance the scientific footing for pulsed power, and to address issues in the development of compact, repetitive pulsed power. These include physics issues implicit in switch development, including both solid state and gas-phase, pulse generation at various levels from extremely small for fast high field generation to >100kV, and applications including ignition of jet fuels and biological response to electric fields.

Authors: M. Giesselmann; T. Heeren; T. Helle

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277806

Abstract: We are developing compact, high-power chargers for rapid charging of energy storage capacitors. The main application is presently rapid charging of the capacitors inside of compact Marx generators for reprated operation. Compact Marx generators produce output pulses with amplitudes above 300 kV with ns or subns rise-times. A typical application is the generation of high power microwaves. Initially all energy storage capacitors in a Marx generator are charged in parallel. During the so-called erection cycle, the capacitors are connected in series. The charging voltage in the parallel configuration is around 40-50 kV. The input voltage of our charger is in the range of several hundred volts. Rapid charging of the capacitors in the parallel configuration will enable a high pulse repetition-rate of the compact Marx generator. The high power charger uses state-of-the-art IGBTs (isolated gate bipolar transistors) in an H-bridge topology and a compact, high frequency transformer. The IGBTs and the associated controls are packaged for minimum weight and maximum power density. The packaging and device selection makes use of burst mode operation (thermal inertia) of the charger. The present charger is considerably smaller than the one presented in Giesselmann, M et al., (2001).

Authors: D. Hemmert; J. Mankowski; J. Rasty; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277997

Abstract: Modeling and characterization of a magnetic flux compression generator (MFCG) requires detailed knowledge of the changes in conductivity of the MFCG materials during the shock-loading phase. In the studies reported here, a thin metallic strip is shocked with an explosively generated shock wave produced from a charge of composition C-4. The shock wave is intended to simulate the shock wave and pressures produced in MFCG research currently being conducted at Texas Tech University. These pressures are estimated to be between 1 and 3 GPa. The experimental setup is arranged so that the shapes of the metallic strip and shock front are the same, as confirmed using optical fibers. This was to ensure that the test sample was shocked uniformly. The metallic test strip is pulsed with a 70 A current pulse during application of the shock wave. The current and voltage across the test sample are measured directly to determine the change in conductivity. Pressure measurements are conducted in separate tests under similar conditions using strain gauges. The results are then compared to results determined previously using a split Hopkinson pressure bar apparatus (SHPB).

Authors: J. Rasty; X. Le; J. Dickens; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277998

Abstract: Helical magnetic flux compression generators (MFCGs) are capable of producing ultra-high power electric pulses by trapping and compressing a seed magnetic field into a load coil via an explosive-driven armature. The efficiency of helical MFCGs is generally very low, about 10%, due to large magnetic flux losses. One of the main sources of magnetic flux loss is the "turn-skipping" phenomenon, in which the expanding armature fails to establish contact with every turn of the helical coil, resulting in magnetic flux loss in the skipped turns of the coil. The "turn-skipping" phenomenon is related to non-uniform or asymmetric expansion of the armature, as well as detonation end effects. Equations describing the "turn skipping" phenomenon are developed in terms of the eccentricity of the armature with respect to the helical coil, the armature's wall thickness variations and the length of the detonation end effect. Design criteria for prevention of "turn-skipping" are presented in order to achieve optimum MFCG performance.

Authors: J. Mankowski; X. Cheng; J. Dickens; M. Kritiansen; E. Choi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277749

Abstract: Experimental findings on the coaxial virtual cathode oscillator (vircator) at Texas Tech University are reported. A major modification to the system extended the pulse forming line several meters. The pulsed power system now delivers up to 750 kV and 60 kA for at least 70 nsec. Additional modifications include diagnostics to measure the diode voltage, current, and radiated field pattern of the output microwaves. In this initial phase of experiments copper reflectors were installed within the diode to provide a simple means of microwave feedback to assist in the e-beam modulation. Thus far we have observed microwave peak power output with feedback reflectors as high as 3.5 GW corresponding to an efficiency of /spl sim/9.5%. Measured radiated field patterns show evidence of multimode operation.

Authors: J. J. Mankowski; E. H. Choi; J. C. Dickens; Xupeng Chen; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1228654

Abstract: Summary form only given, as follows. Summary form only given. Experimental findings on the coaxial virtual cathode oscillator (vircator) at Texas Tech University are reported. Recent modifications to the system include extending the pulse forming line that now provides an applied diode voltage of 600 kV for 70 nanoseconds at 60 kA. Other modifications include diode voltage, current, and radiated field diagnostics. In this initial phase of experiments, copper reflectors were installed within the diode to provide a simple means of microwave feedback to assist in the e-beam modulation. Thus far we have observed microwave peak power output with feedback reflectors as high as 2 GW corresponding to an efficiency of -6%. The mode observed at these high efficiencies is almost entirely TE/sub 11/. At lower efficiencies, both TM/sub 01/, and TE/sub 11/ modes are observed. These results suggest mode competition at lower efficiencies.

Authors: Giesselmann, M; McHale, B; Crawford, M

PDF: https://ieeexplore.ieee.org/document/1277807

Abstract: This paper describes work towards controlling the output of fast AC-Alternators to drive transient loads through intelligent triggering of SCR (Thyristor) switches. The project entails the extension of methods and techniques that have been developed for high performance control of relatively small and slow turning industrial machines, to the more demanding regime of AC-machines used for kinetic energy storage and fast AC-servos for military applications. One of these techniques is a rotational transformation, where the (sinusoidal) AC output voltage of an alternator is converted to a DC level representing the instantaneous amplitude. This transformation, also called 'Vector Rotation' can be used for very fast observation of the momentary amplitudes of all electrical machine quantities, since no averaging is necessary. Through the use of vector rotations, advanced sensors and real time signal processing it is possible to demonstrate fast real-time monitoring of the energy extraction from a high frequency AC-Alternator with a full-bridge SCR converter. A low-voltage DC machine was selected as the load for the system, which closely resembles the transient impedance of an electric launcher.

Authors: N. Schoeneberg; J. Walter; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277996

Abstract: An alternative to the use of capacitors as the seed source for flux compression generators (FCG) are permanent magnet based systems. Permanent magnets provide a larger energy-to-volume ratio given that the use of capacitors requires a power source as well as charging and firing circuitry. A recent design developed at Texas Tech University's Center for Pulsed Power and Power Electronics [S.I. Shkuratov, et al., 2002], focused on the demagnetization of a Neodymium-Iron-Boron magnet (Nd/sub 2/Fe/sub 14/B, B/sub r/= 1.23 T) by a shockwave generated from high explosives. The maximum specific energy achieved with this design was only 2.3 J/kg, which may not be sufficient for effectively seeding an FCG (fast capacitors have at least 20 J/kg). The same magnets were used with an alternative design, referred to as a strip FCG, which utilizes opposing magnets to generate initial magnetic field intensity within an air gap. The air gap exists between a central tube, filled with high explosives, and metal strips placed between the magnets a few millimeters off-axis. The detonation of the explosive causes the expansion of the central tube, subsequently compressing the flux into the strips and then into the load. The original strip FCG design [B.A. Bojko, et al., 1994] used oxide-barium magnets (B/sub r/= 0.2T), which produced an estimated specific energy of 5.27 J/kg into a low inductance load. These magnets were replaced with the Nd/sub 2/Fe/sub 14/B (B/sub r/= 1.23 T) magnets in order to achieve better performance, which will be analyzed with respect to the specific energy. In addition, a design utilizing a shocked ferroelectric material, PZT, as a seed current source is discussed. An explosively generated shock wave is passed through a ferroelectric material, generating a current that establishes the initial seed flux for an FCG. Preliminary tests of the ferroelectric sources indicate a possible theoretical specific energy of more than 11 J/kg or 25 J/dm/sup 3/. A comparison of the designs will be discussed with regard to the specific energy produced and the effectiveness of each to seed FCG's.

Authors: M.G. Hoffman; J.C. Dickens; M.G. Giesselmann

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277679

Abstract: This paper presents a study of temperature variations in a Pulse Power Thyristor (PPT) during and after discharge. The PPT studied was the SPT411BHT. The SPT411BHT is a 5000 V, 4600 A, 125 mm thyristor made by Silicon Power Corporation. In order to determine the temperature of the PPT silicon, a 125 mm diode with identical thermal properties is placed in series with the PPT. There is a strict relationship between the forward voltage and the silicon temperature of the diode. Measurement of the forward voltage of the diode before and after discharge will accurately predict the temperature increase of the silicon. Peak discharge currents will be varied from 30 kA to 90 kA. The forward voltage measurement circuit must be able to resolve millivolts in the presence of common mode voltages in excess of 2 kV. Also, the circuit must isolate the measurement equipment from the circuit. This paper will discuss the temperature measurement concept as well as the design details of the circuit used to measure the diode forward voltage.

Authors: B. W. Morris; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1278086

Abstract: A compact high switching frequency rapid capacitor charging supply was designed and implemented. The challenges of topology design, component selection and implementation are discussed. The end goal is a supply capable of achieving 40 kW at 40 kV on the output and 100-300 V input. The gate driver design is discussed with particular emphasis on gate voltage rise-time and propagation delay. Switching frequencies from 100 kHz to 1 MHz are explored. Transformer and output voltage multiplier designs are also described with expected performance parameters based upon computer simulations.

Authors: Krompholz, H; Hatfield, L; Haustein, M; Spears, J; Kristiansen, M

PDF: https://link.springer.com/chapter/10.1007/978-1-4419-9146-1_37

Authors: M. M. McQuage; A. A. Neuber; J. C. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277717

Abstract: The design and experimental testing of a laser-triggered microwave switch with a nanosecond activation time is described. The objectives of the project include, confirming that a nanosecond to subnanosecond risetime is achievable in the X-band waveguide at 9 GHz with the laser-triggered switch and to determine the minimum laser energy necessary to obtain the fastest possible risetime. A 1 kW pulsed X-band source with a 500 ns output pulse provides the microwave power for the system. A variable power Nd:YAG laser with a maximum 450 mJ at 532 nm, 10 ns FWHM output pulse is used in conjunction with an applied high voltage pulse to trigger the microwave switch. The microwave signal is switched with the rapid formation of plasma caused by the breakdown of a gas contained by a quartz tube inserted through a section of waveguide. The centerpiece of the waveguide system is a magic tee, which controls the direction of power flow through the system. Compared to tests in air and N/sub 2/, the best results have been obtained in argon. Risetimes below 2 ns have been obtained using argon at a reduced pressure of 150 Torr and a high voltage pulse of 28 kV from a spark gap. The impact of gas pressure, applied voltage pulse and applied laser pulse on the risetime of the microwave switch are discussed.

Authors: Wook Jeon; Kew Yong Sung; Seung Hun Chun; Ki Baek Song; Yoon Jung; Mung Chul Choi; Eun Ha Choi; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1286230

Abstract: A virtual cathode reflex triode is investigated by numerical simulation with a 3 dimensional particle-in-cell (PIC) code called MAGIC. The reflex triode virtual cathode oscillator has a disk cathode and a disk anode-mesh. Here anode-cathode (A-K) gap distance has been varied from 10 mm to 30 mm. Simulation results show that the output microwave frequency has a narrow bandwidth and its output microwave power depends sensitively AK gap distance. Output microwave mode is mixture of TE mode at A-K gap distance 10 mm, however, TE is to be dominant mode at A-K gap distance 30 mm.

Authors: M. Butcher; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277987

Abstract: An increased need for compact pulsed power systems requiring new switching technologies combined with the benefits of cryogenic properties, such as higher energy density and miniaturization, has lead to increased interest in liquid nitrogen as a switching medium. High hold off voltage, low dielectric constant, and low environmental impact are further advantages of liquid nitrogen. Characterization of breakdown is investigated using high-speed (temporal resolution < 1 ns) optical and electrical diagnostics in a coaxial system with 52 /spl Omega/ impedance. Experiments are done in self-breakdown mode in super-cooled liquid nitrogen with a temperature near 70 K. Discharge current and voltage are determined using transmission line type current sensors and capacitive voltage dividers. Discharge luminosity is measured with photomultiplier tubes (risetime/spl ap/800 ps) that are focused on the negative electrode tips and the center of the channel. Optical investigations of breakdown and pre-breakdown events on a nanosecond time scale will provide a better understanding of the fundamental physics of breakdown formation. Detailed optical and spectroscopic diagnostics combined with high-speed electrical diagnostics are aimed at clarifying the overall breakdown mechanisms, including electronic initiation and bubble formation. The breakdown initiation/development will be discussed.

Authors: D. Hemmert; K. Shiraki; T. Yokoyama; S. Katsuki; H. Bluhm; H. Akiyama

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277699

Abstract: The use of a pulsed streamer discharge in water to kill algae has led to investigations to determine the degree of influence of potentially lethal byproducts of a streamer discharge. The primary byproducts that may influence the lethality of a streamer discharge on algae cells include the localized electric field, shock wave production, OH and O/sub 3/ radical production, and ultraviolet light production. In the studies reported here, the shock waves generated by a pulsed streamer in distilled and tap water are optically imaged to characterize their temporal history. The streamer discharge was produced in a thin sheet of water using a stainless steel needle and wire geometry with a 2 cm gap between electrodes. This allowed a two dimensional view of the generated shock waves propagating away from the streamer discharge region. A 600 ns pulse with amplitudes of up to 60 kV was applied to the electrodes utilizing a 360 /spl Omega/ Blumlein generator and transformer. A laser schlieren imaging setup combined with a high speed CCD camera with 1.5 ns gating capability captured the temporal history of the shock waves over a series of reproducible discharges. Streamer discharges were observed producing a series of shock waves as they propagated from the needle tip. These shock waves would combine by interference producing a distinct shock wave pattern propagating radially outward at 1.45 km/sec. A second shock wave generating event is observed up to 150 /spl mu/sec after the streamer discharge dissipates. Generation of a single uniform shock wave was also observed at voltage levels associated with the minimum amplitude required for streamer production.

Authors: M. Butcher; M. Cevallos; M. Haustein; A. Neuber; J. Dickens; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1254853

Abstract: In a coaxial test apparatus enabling the measurement of voltage and current at the test gap, dc conduction and breakdown in transformer oil caused by the application of dc voltages are investigated. Current measurements cover the range from 10/sup -1/ A to 1 kA, with temporal resolutions of milliseconds at the lowest current amplitudes to sub-nanoseconds for currents larger than 10/sup -4/ A. The dc current/voltage characteristic for sub-breakdown voltage amplitudes point to the injection of charge carriers, allow us to characterize the transport mechanisms, and the influence of space charges. For voltages approaching breakdown thresholds, quasi dc-currents rising from nanoamperes to microamperes are superimposed by current pulses with amplitudes of milliamperes and above, and durations of nanoseconds. The onset of these current pulses occurs up to 10 /spl mu/s before breakdown. One of these current pulses reaches a critical amplitude causing voltage breakdown and current rise to the impedance-limited value within 2 ns. Additional optical diagnostics using photomultipliers and high-speed photography with gated microchannel plates yield information on hydrodynamic processes and charge carrier amplification mechanisms associated with the current pulses and final breakdown, such as bubble formation, as well as on the development of the spark plasma finally bridging the gap.

Authors: A. A. Neuber; J. -. Hernandez; T. A. Holt; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277740

Abstract: Helical magnetic flux compression generators (MFCG) are attractive energy sources with respect to their specific energy output. A variety of one-time use applications would benefit from small inch-sized helical generators with high specific energy output. However, it is widely accepted that the generator performance deteriorates with decreasing size. Previous experimental data have shown that the increase of the ohmic resistance of the MFCG with a reduction in size is the primary cause for the observed behavior when the initial generator inductance is held constant. We will analyze the situation in more depth and quantify how much the efficiency is determined by ohmic losses and intrinsic flux losses (flux that is left behind in the conductors and lost for compression) for different generator sizes and geometries. Our simple constant diameter MFCGs exhibit more intrinsic than ohmic losses (69% compared to 16%), while our MFCGs with tapered armatures display less intrinsic and more ohmic flux losses (13% compared to 66%), however, at increased overall efficiency. We will show experimental and calculated data and discuss the physical efficiency limits and scaling of generator performance at small sizes.

Authors: Krile, J; Neuber, A; Dickens, J; Krompholz, H

PDF: https://ttu-ir.tdl.org/handle/2346/12789

Abstract: The limits of the applicability of dc, ac, or pulsed high voltage are determined by breakdown along insulators or insulating support structures. It is of major technical importance to predict breakdown voltages for given structures, with parameters such as geometry, material, and temporal characteristics of the applied voltage. The impact of atmospheric conditions such as humidity, pressure, temperature, and types of gas present is also important. A setup has been devised to simulate and closely monitor flashover across various gap distances and insulator geometries at atmospheric conditions at different humidities. Current, voltage, luminosity, and optical emission spectra were measured with nanosecond to sub-nanosecond resolution. Spatially and temporally resolved light emission data yielded quantitative information about the charge carrier amplification, the location of this amplification, and its role in the formation of streamers.

Authors: M. Butcher; A. Neuber; H. Krompholz; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277712

Abstract: The phenomenology of prebreakdown events in transformer oil is investigated using high-speed electrical and optical diagnostics. Data collection using a coaxial test setup terminating into a 50/spl Omega/ load line to simulate a matched impedance system allows very fast risetimes. Transmission line type current sensors and capacitive voltage dividers with temporal resolution of 300 ps provide information about the discharge voltage and current. Steady, DC currents ranging from a few nA with less than 10 kV of applied voltage, to a few /spl mu/A prior to full breakdown are measured using an electrometer. Prebreakdown events are measured with positive and negative charging voltages with respect to ground. Light emission from the discharge is measured using a series of fast photomultiplier tubes, (risetimes 800 ps), that observe positive and negative electrode tips and center of the channel. Preliminary results on self-breakdown (breakdown voltage +44 kV) with a 2.35 mm gap show a DC (seed) current of several hundred nA with prebreakdown spikes of a few mA immediately before final breakdown. Periodicity of the current spikes combined with a general increase in magnitude prior to full breakdown has been observed. Data collection using a negative charging line, with respect to ground with enhanced field at the cathode, indicates current spikes that are typically 25 to 50% faster than spikes using a positive charging line with enhanced field at the anode. Detailed optical diagnostics along with high-speed electrical diagnostics of the pre-breakdown phase will address the physical mechanisms initiating volume breakdown.

Authors: M. D. Cevallos; J. C. Dickens; A. A. Neuber; M. A. Haustein; H. G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1228897

Abstract: Summary form only given, as follows. The fundamental breakdown physics of biodegradable oil is investigated with a set up that employs a cable discharge into a coaxial system with axial discharge and load line to simulate a matched terminating impedance. No discontinuities are created in the system lines when entering the discharge chamber with the implementation of a unique feed-through design. The entire impedance of the system is matched at 50 ohms. A novel design for impedance matching transitions from the discharge cable to the coaxial system to the load line allow for a sub-nanosecond response. Final results are measured on pulsed and self breakdown voltages of up to 200 kV. Self breakdown is achieved by charging the discharge cable and load line to +/- 100 kV respectively. Pulsed breakdown is achieved by charging the discharge cable and load line to +100 kV. Shorting the discharge cable generates a reflected negatively polarized pulse causing breakdown. High speed electrical and optical diagnostics have temporal resolution down to several 100 ps A complete range of information from amplitudes of 0.1 mA to 1 kA with temporal resolutions of 300 ps is achieved by using transmission line type current sensors with fast amplifiers. Capacitive voltage dividers with fast attenuators are also used. Optical measurements are performed on low level light emission using spatially resolved, fast photo-multiplier tubes (risetime of 800 ps), supplemented with high speed photography and spectroscopic investigations on a nanosecond timescale Detailed optical and spectroscopic diagnostic along with high speed electrical diagnostics will address the mechanism initiating/assisting biodegradable oil volume breakdown.

Authors: Lara, MB; Mankowski, J; Dickens, J; Kristiansen, M

PDF: https://ieeexplore.ieee.org/document/1278018

Abstract: An eight-stage four-hundred kilovolt Marx bank, in connection with a 60 nanosecond pulse-forming line, is constructed and utilized as a pulsed source to power a planar version of the Virtual Cathode Oscillator (Vircator). Eight 1upsilonF capacitors, charged to 50kV each, are switched in series by dry-air pressurized spark gaps. The energy from the bank charges a 23 ohm oil transmission line, breaking a peaking gap when the maximum voltage is reached, delivering a 60 ns-300kV pulse to the diode. The design of the Planar or Reflex-Triode geometry Vircator is based upon claims of high efficiency by Didenko et al. A previously constructed TTU Vircator includes a unique E-beam source, the brush cathode; in which a circular array of pins is used as an explosive field emission source to produce relatively high beam currents. The anode consists of a round wire mesh through which the E-beam passes, generating a dense cloud of negative charge known as a Virtual Cathode. This initial phase of testing is composed of basic operation of the entire system and baseline output power and efficiency measurements.

Authors: M.D. Cevallos; J.C. Dickens; A.A. Neuber; M.A. Haustein; H.G. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277989

Abstract: The fundamental breakdown physics of biodegradable oil is investigated with high-speed electrical and optical diagnostics with temporal resolution down to several 100 ps. The set up employs a cable discharge into a coaxial system with axial discharge and load line to simulate matched terminating impedance. A unique feed-through design creates no discontinuities in the system lines through the discharge chamber. The impedance of the system is matched at 50 ohms including a novel design for impedance matching transitions from discharge cable to coaxial system to load line allowing for a sub-nanosecond response. This paper presents results on self-breakdown with voltages of up to 60 kV. Self-breakdown is achieved by charging the discharge cable and load line to +/-30 kV respectively. Transmission line type current sensors and a capacitive voltage divider with fast amplifiers/attenuators are used in order to obtain a complete range of information from amplitudes of 0.1 mA to 1 kA with temporal resolutions of 300 ps. Optical measurements include high speed photography and shadowgraphy. Detailed optical diagnostics along with high-speed electrical diagnostics will address the mechanism initiating/assisting biodegradable oil volume breakdown.

Authors: D. Wetz; K. Truman; J. Dickens; J. Mankowski; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1278009

Abstract: In recent years, researchers have investigated methods of liquid sterilization by applying pulsed high electric fields with some degree of success. The mechanism by which microorganisms are damaged has been shown to be a function of the local electric field and exposure time while independent from thermal and electrolytic effects from the applied pulse. Most published experiments have employed electrical pulses of 10's of kV and microsecond long pulse lengths. We are employing electrical pulses in the 100's of kV range with 100 nanosecond long pulse lengths. This type pulse should be more effective at killing microorganisms and minimize energy losses due to thermal processes in the liquid media.

Authors: S. I. Shkuratov; E. F. Talantsev; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277810

Abstract: Five different types of high voltage capacitors have been tested to determine their maximum high voltage overstress coefficient. Chicago condenser, Reynolds industries, Hivolt, and TDL capacitors of capacitance 100 nF and different nominal voltages were tested in the single shot mode. Experiments have shown that the energy stored under overstress conditions in all types of the capacitors tested is 10 to 20 times more than the nominal energy.

Authors: J. -. Hernandez; A. A. Neuber; J. C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277995

Abstract: End-initiated small volume magnetic flux compression generators (MFCG) have at least one order of magnitude higher energy density (by weight or volume) than capacitive energy storage with similar discharge time characteristics. Since the prime energy is built into the MFCG in form of HE, the capacitor looses even more ground if the necessary prime energy source and the charging supply are included in the weight/volume balance. However, simple MFCGs with a single helix produce high output energy only into low inductance loads, thus producing several 100 kA of current at a voltage level of only a few 10 kV. Many pulsed power devices require less current but a considerably higher voltage level. Two approaches for achieving a higher output voltage level, both utilizing two staged MFCGs, have been reported in the open literature. The first employs a more traditional transformer coupling; the second relies on a dynamic transformer or flux-trapping scheme. Although the traditional transformer coupling has theoretically the better efficiency, we chose the latter approach for our generator design, mostly since it requires a smaller number of components. Our generator has a total length of 250 mm, a helix inner diameter of 51 mm, and is wound with Teflon insulated stranded wire of different sizes in the range from AWG 12 to AWG 22. We have presently achieved an energy gain of /spl sim/ 8 and will discuss the generator performance based on experimental current/voltage waveforms.

Authors: X. Chen; K. Dickens; E.H. Choi; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1277752

Abstract: By introducing a physics approximation method into analyzing the nonlinear Poisson's equation, an analytical expression for the space-charge-limited current density for a one-dimensional (1-D) cylindrical diode has been investigated and developed. This expression is different from Child-Langmuir law for the planar diode and is simpler than Langmuir-Blodgett law for the cylindrical diode. This expression builds an explicit connection between the current density and the physical parameters, which is helpful in optimizing the design of the cylindrical vacuum diode. In addition, a comparison between our analytical result and Langmuir-Blodgett law shows that the physics approximation method is valid in nonlinear differential equation analysis and can be used in other similar cases. Applying the approximation method, we get the relativistic theory corrected current for 1-D cylindrical diodes.

Authors: H. Keene; J. Dickens; A. Neuber; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1277716

Abstract: As superconducting technology becomes more viable in the marketplace, especially in high power applications, the need for a well researched high thermal conductivity electrical insulator is needed. The electrical failure mode for these types of insulators is often surface flashover at subatmospheric temperature and pressure. Testing of two such insulators, aluminum nitride and aluminum oxide, for this failure mode is done for two differing electrode geometries. In addition three coats of GE 7031 dielectric varnish are applied to the exposed parts of the insulator for comparison testing with nonvarnished samples. In general the testing shows an increasing breakdown voltage trend with decreasing temperature. These results indicate a temperature related dependence of the secondary electron emission and electron induced outgassing, which is a component in the process of surface flashover. The addition of the varnish results in a lowered breakdown voltage. The research also covers the effect of electrode conditioning, and presents optical diagnostics of the gas species involved during breakdown.

Authors: Andreas A. Neuber; Thomas Holt; Juan-Carlos Hernandez; James C. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6219386

Abstract: While Magneto Cumulative Generators (MCGs) differ widely in size and operating regimes, it is apparent that the helical flux compression generator is the most promising concept with respect to current amplification and compactness. Though the geometry of the helical generator (dynamically expanding armature in the center of a current carrying helix) seems to be basic, it turns out that the understanding of all involved processes is rather difficult. This fact is apparent from the present lack of a computer model that would be solely based on physical principles and manage without heuristic factors. This means, all programs known to the authors utilize an additional parameter that adjusts for the loss in flux that is currently unexplained. Experimental efforts revealed that the unexplained loss in flux becomes smaller the larger the generator volume is. Specifically, for generators with constant diameter helix and armature, the figure of merit, β, for generator performance exhibits a distinct decline with the angular frequency, ω, of the progressing contact between end-initiated armature and helix. Since ω is proportional to the square root of the ratio of initial MCG inductance to compressed volume, it becomes apparent that the generator performance is limited by size. For large generators on a meter scale, a β of approximately 0.95 has been reported (β = 1 means no flux loss at all), and own tests indicate that β drops quickly to about 0.7 to 0.6 for generator on the 10 cm scale. Still, even very small generators with about 30 mm outer diameter can have a β of around 0.6. Tapered generators, with tapered helix or tapered armature might perform differently and can be used to further assess the situation. Preliminary results for MCGs with a tapered helix show a performance that is somewhat superior to the straight helix generator.

Authors: H. Krompholz; A. Neuber; M. Haustein; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1022721

Abstract: The phenomenology of breakdown in liquid nitrogen is investigated with high-speed electrical and optical diagnostics (temporal resolution down to several 100 ps). The discharge apparatus uses a cable discharge into a coaxial system with axial discharge, and a load line to simulate a matched terminating impedance. Main experiments are done in self-breakdown mode in supercooled liquid nitrogen. Transmission line type current sensors and capacitive voltage dividers with fast amplifiers/attenuators cover an amplitude range of 0.1 mA to 1 kA with a time resolution of 300 ps, providing complete information about discharge voltage and current. The light emission is measured with fast photomultiplier tubes (risetime 800 ps), and these optical measurements will be supplemented by high-speed photography and spectroscopic investigations on a nanosecond time scale. First results on self-breakdown with a gap width of 1 mm and electrodes with 5 mm radius of curvature (breakdown voltage 42 kV) show a three-phase development: the current rises from an unknown level to several mA during 2 ns, stays approximately constant for 100 ns with superimposed ns-duration spikes, and shows a final exponential rise to the full impedance limited current amplitude during several nanoseconds. Detailed optical and spectroscopic diagnostics along with the high-speed electrical diagnostics will in particular address the physical mechanisms initiating/assisting the liquid nitrogen volume breakdown, such as bubble formation during the pre-breakdown phase.

Authors: A. Neuber; H. Krompholz; M. Haustein; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1030428

Abstract: Summary form only given, as follows. Miniaturization of electrical components along with growing superconductor technology requires a better understanding of the phenomenology of breakdown in liquid nitrogen. It is known that the time delay between breakdown-onset and final impedance-limited arc current can occur within a few nanoseconds. For a temporal resolution down to several 100 ps, a discharge apparatus was built and tested that uses a cable discharge into a coaxial system with axial discharge, and a load line to simulate a matched terminating impedance. Main experiments are done in self-breakdown mode in supercooled liquid nitrogen, pulsed breakdown at high over-voltages in standard electrode geometry is investigated as well. Transmission line type current sensors and capacitive voltage dividers with fast amplifiers/attenuators cover an amplitude range of 0.1 mA to 1 kA with a time resolution of 300 ps, providing complete information about discharge voltage and current. The light emission is measured with fast photomultiplier tubes (risetime 800 ps), and these optical measurements will be supplemented by high-speed photography and spectroscopic investigations on a nanosecond time scale. Preliminary results on self-breakdown in the surface flashover mode with a gap width of 2 mm and electrodes with 5 mm radius of curvature (breakdown voltage /spl sim/ 60 kV) show a three-phase development: the current rises from an unknown level to several mA during 2 ns, stays approximately constant for 100 ns with superimposed ns-duration spikes, and shows a final exponential rise to the full impedance limited current amplitude during several nanoseconds. The detailed optical and spectroscopic diagnostics along with the high-speed electrical diagnostics will in particular address the physical mechanisms initiating/assisting the liquid nitrogen volume breakdown, such as bubble formation during the pre-breakdown phase.

Authors: A. Neuber; N. Schoeneberg; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1189493

Abstract: The feasibility of utilizing the chemical energy stored in high explosives to generate an antenna capable of radiating for several microseconds is studied. Crucial parameters such as conductivity as a function of time, maximum achievable antenna length for given initial device volume and weight, and material dependence are assessed by sub microsecond optical and electrical diagnostics. By utilizing a rotating framing mirror camera with up to 4 Megaframes per second, which produces a color image sequence consisting of 125 frames, possible premature breakup of the transient antenna rod is measured with adequate spatial and temporal resolution, thus revealing a rod growth velocity of a few millimeters/microsecond. Electrical diagnostic, primarily aimed at the conductivity between selected points along the rod propagation, enables to make a connection with the observed behavior based on the optical diagnostics. Maintaining electrical contact at the base of the device while the rod is fully formed is crucial and will be discussed in detail.

Authors: D. Hemmert; A. Neuber; H. Krompholz; J. Mankowski; D. Saeks

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1030653

Abstract: Summary form only given, as follows. A new type of plasma limiter is being developed capable of turnon in less than 1 nsec. The approach taken is to initiate streamer breakdown via a micron radius needle tip. Studies were conducted on breakdown with a variety of gases at pressures from 10/sup 3/ to 10/sup -2/ torr. Gases tested included dry air, neon, argon, and krypton. Studies were also conducted on dc-voltage biasing the needle and its effect on breakdown. The experimental setup uses an S-band traveling wave resonant ring (TWRR) capable of power levels up to 100 MW when coupled to a 2.85 GHz, 4 MW, magnetron. High speed diagnostics with a response on the order of 1 ns record the microwave power, luminosity, and x-rays. A high speed CCD camera with an adjustable exposure time down to 10 nsec records a snapshot of the breakdown sequence. Preliminary results exhibit a reduction in expected gas breakdown levels by over two orders of magnitude.

Authors: M. Cevallos; J. Dickens; A. Neuber; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6219396

Abstract: Experiments in self-breakdown mode and pulsed breakdown at high over-voltages in standard electrode geometries are performed for liquids to gain a better understanding of their fundamental breakdown physics. Different liquids of interest include liquids such as super-cooled liquid nitrogen, oils, glycerols and water. A typical setup employs a discharge chamber with a cable discharge into a coaxial system with axial discharge, and a load line to simulate a matched terminating impedance, thus providing a sub-nanosecond response. This study is focused on the feed-through design of the coaxial cable into this type of discharge chamber, with the feed-through being the critical element with respect to maximum hold-off voltage. Diverse feedthroughs were designed and simulated using Maxwell 3-D Field Simulator Version 5. Several geometrically shaped feed-through transitions were simulated, including linearly and exponentially tapered, to minimize electrostatic fields, thus ensuring that the discharge occurs in the volume of interest and not between the inner and outer conductor at the transition from the insulation of the coaxial cable to the liquid. All feedthroughs are designed to match the incoming impedance of the coaxial cable. The size of the feedthroughs will vary from liquid to liquid in order to match the coaxial cable impedance of 50Ω. The discharge chamber has two main ports where the feed-through will enter the chamber. Each feed-through is built through a flange that covers the two main ports. This allows the use of the same discharge chamber for various liquids by changing the flanges on the main ports to match the particular liquid. The feedthroughs were designed and built to withstand voltages of up to 200 kV. The feedthroughs are also fitted with transmission line type current sensors and capacitive voltage dividers with fast amplifiers/attenuators in order to attain a complete range of information from amplitudes of 0.1mA to 1 kA with a temporal resolution of 300 ps.

Authors: Dickens, J; Neuber, A; Haustein, M; Krile, J; Krompholz, H

PDF: https://aip.scitation.org/doi/10.1063/1.1530810

Abstract: Compact pulsed power systems require new switching technologies. For high voltages, liquid nitrogen seems to be a suitable switching medium, with high hold-off voltage, low dielectric constant, and no need for pressurized systems as in high pressure gas switches. The discharge behavior in liquid nitrogen, such as breakdown voltages, formative times, current rise as function of voltage, recovery, etc. are virtually unknown, however. The phenomenology of breakdown in liquid nitrogen is investigated with high speed (temporal resolution < I ns) electrical and optical diagnostics, in a coaxial system with 50-Ohm impedance. Discharge current and voltage are determined with transmission line type current sensors and capacitive voltage dividers. The discharge luminosity is measured with photomultiplier tubes. Preliminary results of self-breakdown investigations (gap I nun, breakdown voltage 44 kV, non-boiling supercooled nitrogen) show a fast (2 ns) transition from an unknown current level to several mA, a long-duration (100 ns) phase with constant current superimposed by ns-spikes, and a final fast transition to the impedance limited current during several nanoseconds. The optical measurements will be expanded toward spectroscopy and high speed photography with the aim of clarifying the overall breakdown mechanisms, including electronic initiation, bubble formation, bubble dynamics, and their role in breakdown, for different electrode geometries (different macroscopic field enhancements).

Authors: D. Hemmert; A. Neuber; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1189459

Abstract: Microwave breakdown of various He-N/sub 2/ mixture combinations is investigated utilizing an S-band double window pillbox cavity. The objective is to determine the best combination of the two gases to maximize the mixture's ability to cool a dielectric surface while maintaining high breakdown thresholds. Helium is known to be an excellent source to dissipate heat from a dielectric, but it also has a low microwave breakdown threshold. Nitrogen does not transport heat as well, but has a much higher breakdown threshold. The studies focused on generating a series of E-field vs pressure curves for breakdown to help identify optimum He-N/sub 2/ mixture ratios. The S-band double window pillbox has a 333 cm/sup 3/ cavity with two ports to flow the gas mixture through. The pillbox is placed in a traveling wave resonant ring (TWRR) coupled to a 2.85 GHz, 4 MW, magnetron. This combination of double window pillbox and TWRR allows for testing power levels up to 40 MW. High speed diagnostics are used to measure the incident/reflected power and discharge luminosity. Coupled mass flow controllers maintain the gas mixture ratio and continuous gas flow through the cavity. Investigations can be conducted with single or multiple pulsed operations. Results ranged from a minimum of 5 MW for pure helium at 760 torr, to greater than 40 MW for pure nitrogen at 3040 torr.

Authors: Y. Jung; M.C. Choi; K.B. Song; G.Y. Sung; E.H. Choi

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1030398

Abstract: Summary form only given, as follows. We have numerically simulated the high power microwave generation by virtual cathode oscillation in a coaxial vircator with 3 dimensional PIC code Magic. It is noted that the coaxial vircator has the advantage of an enhancement of power efficiency and gaining of narrow bandwidth frequency. The study is focused on the design of a diode structure for a coaxial vircator suitable for Chundoong IREB pulser (Max. 600 kV, 85 kA, 60 ns). It consists of a center annular cathode, a cylindrical meshed-anode and a reflector. The simulation results showed that the coaxial vircator can indeed be operated at the selected frequency with the narrow bandwidth and the microwave output power strongly depends on the position and geometry of the reflector. From this simulation results we expect that the maximum microwave output power will be obtained by the reflector of 4 cm in width and also 4 cm in distance away from the IREB, where the resonant frequency of microwave from coaxial vircator is simulated to be 6.577 GHz.

Authors: Krompholz, H; Hatfield, L; Short, B; Kristiansen, M

PDF: https://link.springer.com/chapter/10.1007/0-306-47948-6_52

Abstract: Fast gaseous breakdown is of interest for both UWB/short pulse electromagnetics, and for plasma limiters to protect devices from high power microwave radiation. A quantitative investigation of fast breakdown phenomena, especially for relatively low voltages and for special geometries, does not exist to the authors' knowledge. Breakdown in gases is studied in a point-plane geometry with fast high voltage pulsers, covering the parameter range of voltage amplitude 1.7 to 7.5 kV, risetime 400 ps to 1 ns, and pulse duration 1 to 20 ns. The setup consists of a pulser, 50-Ohm transmission line, axial needle-plane gap with outer conductor, and 50-Ohm load line. The needle consists of tungsten and has a radius of curvature below 0.5 mum. The constant system impedance of 50 Ohm (except in the vicinity of the gap) and special transmission-line-type current sensors enables sub-nanosecond current and voltage measurements with a dynamic range covering several orders of magnitude. Digitizing oscilloscopes with sampling rates of 5 ps and 50 ps are used, with analog risetimes of 80 and 240 ps. In addition, the luminosity is measured with a sensitivity of about 10(8) V/W and a risetime of 800 ps. For pulse amplitudes of 1.7 kV (which are doubled at the open gap before breakdown), long pulse duration, and a gap distance of 1 mm, delay times between start of the pulse and start of a measurable current flow (amplitude > several milli-amperes) have a minimum of about 8 ns. The pressure dependence of this delay time was measured, in 10 to 600 torr argon, and a minimum is observed at 50 torr. Voltages of 7.5 kV produce breakdowns with a delay of about 1 ns. Statistical delays could not be found for either pulse amplitude, with the tip positively pulsed. With negative pulses applied to the tip, at an amplitude of 7.5 W, breakdown is always observed during the rising part of the pulse, with breakdown delay times below 800 ps. This delay time does not depend on the pressure, and points to a vacuum type discharge with field-emitted electrons as the dominant discharge mechanism for this case.

Authors: M. M. Dickens; S. S. Gleason; H. Sari-Sarraf

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=999927

Abstract: A volumetric image segmentation algorithm has been developed and implemented by extending a 2D algorithm based on active shape models. The new technique allows segmentation of 3D objects that are embedded within volumetric image data. The extension from 2D involved four components: landmarking, shape modeling, gray-level modeling, and segmentation. Algorithms and software tools have been implemented to allow a user to efficiently landmark a 3D object training set. Additional tools were built that subsequently,, generate models of 3D object shape and gray-level appearance based on this training data. An object segmentation strategy was implemented that optimizes these models to segment a previously unseen instance of the object. Results of this new 3D segmentation algorithm have been generated for a synthetic volumetric data set.

Authors: R. J. Barker

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=961423

Abstract: Summary form only given, as follows. In June of 2001, a new 3-to-5-year program of basic research on "compact, portable pulsed power" begins at six US universities. Under the Department of Defense's Multidisciplinary University Research Initiative (MURI) Program, this large new effort seeks to focus the talents and capabilities of leading scientists and engineers from a multitude of fields onto this crucial defense topic. This Compact Pulsed Power MURI will be funded at a level of $1.5M/year and encompasses two synergistic consortia. The first is headed by Edl Schamiloglu at the University of New Mexico teamed with Karl Schoenbach at Old Dominion University and Robert Vidmar at the University of Nevada (Reno). The second is headed by Martin Gundersen of the University of Southern California teamed with James Dickens of Texas Tech University and William Nunnally of the University of Missouri (Columbia). The full list of researchers at the six institutions are presented along with highlights of their proposed research efforts.

Authors: M. Giesselmann; T. Heeren; A. Neuber; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1002018

Abstract: This paper describes an advanced PSpice/sup (R)/ model, which is used to complement the experimental work on explosive flux compression generators. This work is conducted at Texas Tech University in the framework of a MURI program. The results from the model are compared to actual results from generators that have been tested in the authors' laboratory. To achieve better fidelity of the model, the effect of transient current diffusion into the conductors of the generator is modeled by using a ladder network with many concentric layers of each current carrying conductor. For the optimum accuracy, the spacing of the concentric layers is closest at the surface of the conductor. The paper shows results, comparisons with experimental data and applications of the model for the design of power conditioning systems for MFC-generators.

Authors: Benton, T; Hsieh, KT; Stefani, F; Neuber, A; Kristiansen, M

PDF: https://www.scopus.com/record/display.uri?eid=2-s2.0-0035197234&origin=inward&txGid=88fa19768efac27451f095b295663564

Abstract: This paper analyzes the transient electromagnetic, thermal, and structural behavior of the stator turns in a simple helical magnetic flux compression generator (MFCG). The main objective is to quantify the losses due to Ohmic heating and flux trapping in the conductors, including the effect of armature motion and armature proximity on the current distribution. The electric current excitation used in the modeling is based on experimental data from tests on a research MFCG. The electromagnetic (EM) and thermal analyses were conducted using EMAP3D, a 3D finite element analysis (FEA) code developed at The University of Texas at Austin (UT), which has the capability to model relative motion and sliding between conductors. Structural analyses were conducted using a version of DYNA3D that allows state data from EMAP3D to be used as input. Details of the analyses include temperature dependence on electrical, thermal, and mechanical properties.

Authors: Xiaobin Le; J. Rasty; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001693

Abstract: During the operation of magnetic flux compression generators (MFCG), the gas-plasma, shocked by the rapidly expanding armature, could lead to electrical arcing across the gas between the armature and the stator at locations where physical contact between the armature and stator has not yet occurred. This will result in a loss of magnetic flux and a decrease in the electrical efficiency of the MFCG. Therefore, knowledge of the ensuing gas temperature and pressure histories is necessary for identification of loss mechanisms in an effort to optimize the efficiency of MFCGs. This paper describes the procedure for estimating the air temperature and pressure histories via finite element (FE) simulation of the armature expansion and its ensuing contact with the stator in an MFCG. First, the validity of the FE model was verified by comparing deformation contours obtained from the simulations to those obtained experimentally via high-speed photography. Utilizing the pressure history data obtained from the FE results, the air temperature was theoretically calculated. The results indicate that the air pressure and temperature in an MFCG, having a compression ratio of 1.8, could be as high as 30 MPa and 4000/spl deg/ Kelvin, respectively.

Authors: M. Giesselmann; E. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1002177

Abstract: This paper describes the downsizing potential of a power supply for charging a 7.2 /spl mu/F capacitor to a voltage of 30 kV, which represents 3.24 kJ of energy, in approximately 40 ms. This process should be repeatable 10-20 times in a short burst mode within a time frame of a few minutes. The primary supply is a DC source with approximately 650 V. A previous design achieved all these goals using 4 Semikron half-bridge IGBTs rated at 1200 V/1200 A each. From the operational experience with the previous charger we concluded, that the weight and volume of the inverter, which is the core of the system, can be reduced to half of the previous size. This can be achieved by using advancements in IGBT modules as well as state of the art capacitors and optimum packaging. Due to the increased capabilities of the new IGBT modules, the number of required modules can be reduced from 4 to 2. In addition, the 15 V/10 A auxiliary power supply, which was previously required, is no longer needed, since the new modules derive their auxiliary power from the main DC bus. The new IGBT modules will be connected to form an H-Bridge. A step-up transformer and a rectifier bank will be connected to the output of the H-Bridge. For compactness, the cooling fins on the old modules have been removed, which was permissible due to the fact that the supply is only operated in short term burst mode. There are no cooling fins to remove on the new module since we are using the water-cooled version. This also gives us the option to water cool if needed. With these modifications, the original charger, which took up a standard 19 inch equipment rack, could be reduced to the size of an oversized suitcase.

Authors: S. I. Shkuratov; E. F. Talantsev; M. Kristiansen; J. Dickens; J. C. Hernandez; A. Neuber

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1002017

Abstract: The results are presented of tests with compact, explosively driven shock wave ferromagnetic generators. The shock wave from a high explosive charge is passed along the axis of a cylindrical, hard magnet. Two types of permanent magnets were used in the experiments: rare-earth NdFeB cylinders (D = 2.5 cm, L = 1.9 cm) and hard ferrite BaFe/sub 2/O/sub 3/ cylinders (D = 2.2 cm, L = 2.5 cm). The shock wave demagnetizes the cylinder, reducing the flux from the remnant value to zero. This change in flux generates a voltage in the winding. The current generated in the loads of the generators yielded a peak of 0.75 kA. The operation of the shock wave ferromagnetic generators was analyzed by the Maxwell 3D code. An analysis is given on the specific features of pulse generation in a system like this.

Authors: D. Hemmert; J. Rasty; A. Neuber; J. Dickens; X. Le; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1002043

Abstract: Modeling and characterization of a Magnetic Flux Compression Generator (MFCG) requires detailed knowledge of the changes in conductivity of the MFCG materials during the shock-loading phase. A Split-Hopkinson Pressure Bar apparatus (SHPB) and current source/ differential amplifier setup was used to study shock-loading under controlled conditions while monitoring changes in resistivity in armature material samples. The SHPB apparatus was capable of producing strain rates up to the fracture limit of the samples tested. Actual fracturing of samples would not have allowed detailed analysis of thermal and mechanical effects in sample resistivity changes. Sample strain rate levels of up to lx104 strain sec-1 were achieved with the apparatus on OFHC copper and aluminum samples.

Authors: W. Jiaig; K. Woolverton; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001883

Abstract: The coaxial virtual cathode oscillator (vircator) at Texas Tech University was studied by varying the pulse length of the electron beam. Previous experimental results obtained with electron beam pulse width of 25 ns have indicated that the microwave output power had not saturated before the electron beam power terminated. Therefore, it has been expected that the maximum microwave power can be increased if the electron beam pulse is lengthened. Experiments were carried out with the electron-beam pulse length of 25 ns and 70 ns, respectively, while keeping other parameters the same. The microwave outputs obtained with different electron-beam pulse lengths are compared in order to understand the effect of the pulse length on the microwave efficiency of the coaxial vircator.

Authors: J. Rasty; X. Le; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=960712

Abstract: Summary form only given, as follows.The ability of MFCGs for creating large current pulses is well accepted among the scientific community; however, little is understood regarding the fundamental mechanical and electrical mechanisms, and their interrelationship, governing the operation and efficiency of MFCGs. The main objective of this research was to conduct a series of experimental and numerical studies in an attempt to gain insight into the inner working of MFCGs. As a first step, the expansion characteristic of the exploding armature was selected as one of the major factors affecting the efficiency of MFCGs. Both numerical as well as experimental techniques were employed to capture the explosive-driven expansion behavior of the armature. Numerical results were verified experimentally to determine the expansion-time history, expansion angle, expansion velocity, armature/stator contact velocity and the severity of the end-effect in a typical MFCG. The experimental and numerical results showed excellent agreement paving the way for future simulations using the established FE model. The results indicate that the radial and axial impact velocities between the armature and the stator are almost constant throughout the operation of MFCG. Moreover, the results indicate that the contact velocity between the armature decreases as the post-detonation time increases, reaching a constant value equal to the detonation velocity.

Authors: S. Shkuratov; M. Kristiansen; J. Dickens; A. Neuber; L. L. Altgilbers; P. T. Tracy; Y. Tkach

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=960832

Abstract: Summary form only given. The design of shock wave ferroelectric generators driven by high explosives is presented and experimental data are discussed. The active elements are lead zirconium titanate (PZT) disks with diameter D=25 mm and thickness L=2.5 mm, and PZT cylinders with D=21 mm and L=25 mm. The high explosive charge was varied from 4.2 g to 30 g. Two different ways to initiate shock waves in the active elements were used: explosively driven flyer plates and direct action of high explosives. The data are presented on the maximum power released in the resistive load and on the effect of the load inductance and the load capacitance on the amplitude of the pulse produced. The experimental results are discussed in comparison to numerical simulation with the commercial circuit simulator PSPICE. An analysis is given of the specific features of pulse generation in a system like this.

Authors: Y. Tkach; S. Shkuratov; L. Dickens; V. Kristiansen; L. L. Altgilbers; P. T. Tracy

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001707

Abstract: Explosive driven ferroelectric generators (EDFEGs) are compact power sources that have been considered for use as seed sources for magnetocumulative generators, as well as prime power sources. Shock waves generated by high explosives are used to shock depolarize ferroelectric materials, which results in a voltage pulse being delivered to a load. These generators have been experimentally investigated at Texas Tech University. Data from these experiments was used to benchmark a code developed at the Institute of Electromagnetic Research. In this paper, a description of the simulation and a comparison of the experimental and simulation results are presented.

Authors: Shkuratov, SI; Kristiansen, M; Dickens, J

PDF: https://ieeexplore.ieee.org/document/1001700

Abstract: The results are presented of experimental studies of explosively driven moving magnet generators. The study was performed with the use of high explosive and propellant charge. The projectiles used were NdFeB cylinders with D = 2.5 cm and H = 1.9 cm. Data are given for amplitude of high current pulses for different designs of the devices. The experiments performed have shown that the best option for generators using a moving magnetic projectile is to utilize the energy of the gases formed as a result of burning of the explosive rather than the energy of a flyer plate accelerated under the action of burning of the explosive. The explosive of preference is not C-4, but a military propellent charge which shows a high burning velocity and a high energy of explosive gases.

Authors: H. Krompholz; L. L. Hatfield; M. Kristiansen; D. Hemmert; B. Short; J. Mankowski; M. Brown; L. Altgilbers

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1002139

Abstract: Gaseous breakdown in the sub-nanosecond regime is of interest for fast pulsed power switching, short pulse electromagnetics and for plasma limiters to protect devices from high power microwave radiation. Previous investigations of sub-nanosecond breakdown were mainly limited to high-pressure gases or liquids, with applied voltages in excess of 100 kV. In this paper, the authors investigate possibilities to achieve sub-nanosecond breakdown at applied voltages below 7.5 kV in point-plane geometries. The setup consists of a pulser (risetime between 400 ps to 1 ns), 50-/spl Omega/ transmission line, axial needle-plane gap with outer coaxial conductor, and a 50-/spl Omega/ load line. The needle consists of tungsten and has a radius of curvature below 0.5 /spl mu/m. The constant system impedance of 50 /spl Omega/ (except in the vicinity of the gap) and a special transmission-line-type current sensors enables current and voltage measurements with a dynamic range covering several orders of magnitude, with temporal resolution down to 80 ps. For pulse amplitudes of 1.7 kV (which are doubled at the open gap before breakdown) delay times between start of the pulse and start of a measurable current flow (amplitude > several milliamperes) have a minimum of about 8 ns, at a pressure of 50 torr in argon. Voltages of 7.5 kV produce breakdowns with a delay of about 1 ns. With negative pulses applied to the tip, at an amplitude of 7.5 kV, breakdown is always observed during the rising part of the pulse, with breakdown delay times below 800 ps, at pressures between 1 and 100 torr. At lower pressure, a longer delay time (8 ns at 50 mtorr) is observed. They authors expect the breakdown mechanism to be dominated by electron field emission, but still influenced by gaseous amplification.

Authors: Shkuratov, SI; Kristiansen, M; Dickens, J; Horrocks, E

PDF: https://ieeexplore.ieee.org/document/1001860

Abstract: Different types of commercial batteries were tested in a high-current mode to determine which is the best in portable pulsed power applications. The testing of the batteries was performed in two regimes: short time regime (one second) and long time regime (one hundred seconds). Alkaline, carbon, lithium, lithium ion, lead acid and nickel cadmium batteries of different dimensions have been tested to determine the maximum usable power for short and long time modes. The resistance of the load was varied from 1 Omega to 22 mOmega. Current-voltage characteristics are presented for sixteen different types of batteries. Data are given for maximum power per unit of weight and per unit of volume.Different types of commercial batteries were tested in a high-current mode to determine which is the best in portable pulsed power applications. The testing of the batteries was performed in two regimes: short time regime (one second) and long time regime (one hundred seconds). Alkaline, carbon, lithium, lithium ion, lead acid and nickel cadmium batteries of different dimensions have been tested to determine the maximum usable power for short and long time modes. The resistance of the load was varied from 1 Omega to 22 mOmega. Current-voltage characteristics are presented for sixteen different types of batteries. Data are given for maximum power per unit of weight and per unit of volume.

Authors: M. Giesselmann; T. Heeren; A. Neuber; J. Walter; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1002004

Abstract: This paper presents high-speed optical diagnostics of an exploding wire fuse, which is used in the power conditioning system for an explosive flux compression generator. The images were taken using an IMACON(R) 790 high-speed framing camera utilizing a gated image intensifier tube. For the authors' measurements, the camera was operated in the high-speed multiple frame mode, yielding 8-10 sequential, 2 dimensional pictures with 100 ns between exposures.

Authors: S. I. Shkuratov; E. F. Talantsev; M. Kristiansen; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001859

Abstract: Three different types of capacitor have been tested to determine maximum usable high voltage. The capacitor testing was performed in the dynamic mode. The voltage rise varied from 200 to 400 V/sec. Disc ceramic and thin film capacitors of different value and different nominal voltages were tested. Experiments have shown that the breakdown voltage for all types of the capacitors tested is about ten times more than the nominal voltage of the capacitors. Data are given for the limiting high voltage for each kind of capacitor. Experiments have shown that the mechanisms for the destruction of each type of capacitor have specific features.

Authors: T. Holt; A. Neuber; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001684

Abstract: Explosively driven magnetic flux compression has been the object of research for more than three decades. Recently heightened interest has been focused on the basic physical mechanisms that determine the performance of helical magneto cumulative generators (MCGs). Two single-pitch helical flux compression generators of different sizes have been tested using current-voltage probes and optical diagnostics. The main parameters used to characterize the experimental performance of the flux compression generators were the flux conservation and theoretical current gain of each type of generator. Helices with constant pitch and differing separation between wires as well as wires with different insulation thickness were tested and analyzed with respect to their flux conservation and theoretical current gain. Preliminary results show that the insulation thickness plays only a minor role for a change in flux conservation due to geometry in the range from 0.01 to 0.5 mm provided that no internal breakdown occurred. Additionally, the overall physical dimension of the generator was modified to allow for a substantial increase in initial inductance. The outer diameter of the generator armature was held constant at 1.5 inches and the coil diameter was varied from 2.6 to 3.5 inches (expansion ratio of 1.7 or 2.3, respectively). The results gained from the conservative expansion ratio of 1.7 were used as a base to compare to the generator performance at the more aggressive expansion ratio of 2.3. First results show that an expansion ratio of 2.3 produces viable results for a partially annealed Aluminum armature with a Gurney angle of approximately 15 degrees.

Authors: Y. Tkach; S. Shkuratov; J. Dickens; M. Kristiansen; L. L. Altgilbers; P. T. Tracy

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1001708

Abstract: The objective of this work was to investigate both parametrically and experimentally the properties of the explosive driven ferroelectric generator (EDFEG). The parametric investigaton was conducted using a simulation developed by the Institute of Electromagnetic Research and the experimental investigation using the explosive test facilities at Texas Tech University. Both efforts were conducted synergistically in order to fully understand the characteristics of the EDFEG and to optimize its operation. Results of both studies are presented.

Authors: Shkuratov, SI; Kristiansen, M; Dickens, J; Hernandez, JC

PDF: https://ieeexplore.ieee.org/document/1001701

Abstract: Results are presented of investigations of the generation of high-current pulses in moving magnet generators designed as an open ferromagnetic circuit. The experimental study was performed with the use of a light gas gun. The magnetic projectiles were ferromagnetic cylinders having 2.54-cm diameters. Data are given for the amplitude of high current pulses and power in the load for different designs of the devices, the effect of various windings on the pulse-generating coils, and different types of loads. It has been shown that with velocities of the magnetic projectiles of 200-300 m/s, the peak current of the pulses produced by the generators reach kiloamperes.

Authors: A. Neuber; T. Holt; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1002003

Abstract: The knowledge of the thermodynamic state of the gas trapped in a helical flux compression generator is crucial for the assessment of flux loss due to internal electrical breakdown/arcing. Besides the helix deformation and armature deceleration at extremely high current amplitudes approaching 1 MA, the thermodynamic state of the shocked and compressed gas causes problems in the prediction of the generator output current vs. time towards the end of generator operation. Such a breakdown is experimentally detected as an abrupt change in the time derivative of the current waveform and it is easily distinguished from partial turn skipping by its sharper fall and nonperiodic occurrence. The thermodynamic state of the generator was measured using primarily optical emission spectroscopy. Three main stages of operation are discussed: (1) the initial stage, which can be represented by a freely expanding armature, that shows fairly low gas temperatures; possibly as low as 2000 K; and (2) the intermediate stage during 14 to 4 microseconds before generator burnout that exhibits mainly an atomic copper line transition at about 0.8 eV; and (3) the last few microseconds that reveal a highly compressed gas with temperatures of about 5000 K and pressures of about 1500 bar. Most experiments were conducted in air, initially at STP, some results are given for SF/sub 6/ initially at one atmosphere. In order to link the thermodynamic state to the breakdown sensitivity, additionally, simple conductivity measurements were conducted in current-free flux compression generator models.

Authors: D. Hemmert; A. Neuber; J. Dickens; H. Krompholz; L. L. Hatfield; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6220112

Abstract: Physical mechanisms leading to microwave breakdown on dielectric/gas interfaces are investigated for power density on the order of 10 MW/cm2 at 2.85 GHz and gas pressures on the order of 10−4 torr to 103 torr. The investigation is focused on an alumina/air interface; other gases are considered for reference purposes. A 3 MW magnetron with 3.5 µs pulse width is coupled to an S-band traveling wave resonator with a pressure adjustable test region. The pre-breakdown phase and the breakdown are monitored by recording the traveling and reflected power, and the spatially integrated luminosity. Electric field probes in the vicinity of the interface are included as well to get information about the local field. Furthermore, the light emission was observed with an image intensifier capable of a minimum gate time of 2.5 ns, in temporal correlation to the other phenomena, or with a framing camera having a 20 ns gate time and 100 ns separation between pictures. The pressure dependent breakdown characteristics, such as appearance, breakdown field, and temporal shape of electric signals, are compared to dielectric/vacuum interface breakdown and volume breakdown, all measured utilizing similar setups.

Authors: Bayne, SB; Giesselmann, MG

PDF: https://ieeexplore.ieee.org/abstract/document/870874

Abstract: Renewable energy such as wind and solar is a clean source of energy that can be integrated with conventional ways of producing energy. Power utility companies are looking at ways to integrate renewable energy with conventional methods. The Central and South West (CSW) renewable project was designed to investigate the role of solar and wind energy in a utility system. When connecting a wind farm to a utility grid, several issues must be taken into consideration such as stability, load matching, cost, location and the wind profile in relation to the peak loads on the system. One other parameter that is seldom considered is the effect of power oscillations due to turbine blades passing through their full are of motion and periodically encountering different wind speeds at different vertical positions. In the following, this effect is called 'blade passing' for short. This paper evaluates the effect of blade passing on the output voltage and current for a grid connected wind farm. The effect of blade passing as a function of the number of wind turbines connected to the grid is also investigated. The exact causes of the blade passing effect are also discussed.

Authors: Giesselmann, M; Kristiansen, M; Grinstead, B; Wilson, M

PDF: https://www.scopus.com/record/display.uri?eid=2-s2.0-0034508877&origin=inward&txGid=9189bf197e8c232dbc11eddcc58886a1

Abstract: A pulse generator, constructed by a group of Russian scientists, which is using a solid state opening switch (SOS), was characterized and tested, The diode acts similar to a step recovery diode but has voltage and current ratings that exceed the levels in US manufactured step-recovery diodes by orders of magnitude. To the authors knowledge, there are no US manufactured diodes for the application described here, With the load chosen for this paper, the generator produces pulses of 100 kV amplitude and 10 ns bah width, using a 600V DC supply. The maximum pulse repetition rate is 100 Hx, limited by thermal considerations of the primary charging resistor, This limit could be easily overcome by using a switch-mode power supply to charge the primary capacitors, This pulser can be used to generate partial discharges in a chemical reactor in order to produce Ozone for a variety of industrial uses such as sterilization, oxidation, NOx or SOx removal from exhaust gases, etc. The greatest advantage is the long lifetime of this all solid-state device.

Authors: A. Neuber; J. Dickens; M. Giesselmann; B. Freeman; D. Dorsey; H. Krompholz; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6220176

Abstract: The design of a helical flux compression generator, driven by 200 g of high explosives, for basic studies is presented and experimental results are discussed. Generator current and dI/dt are measured with Rogowski coils placed on the load. Both crowbar closure and contact velocity have been successfully simulated with CTH, a three-dimensional finite element hydrodynamic code. The generator's magnetic field structure is briefly discussed and magnetic field probe measurements are presented.

Authors: A. Neuber; J. Dickens; M. Giesselmann; B. Freeman; J. Rasty; H. Krompholz; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=855148

Abstract: Summary form only given, as follows. The design of a helical flux compression generator, driven by 150 g of high explosives, for basic studies is presented and experimental results are discussed in comparison to numerical simulations. Simulation of the electric current output with the commercial circuit simulator PSPICE shows that this generator conserves the magnetic flux ideally in the low current mode, <30 kA. At current amplitudes in excess of 100 kA heating and melting of the single wound helix wire, AWG 12, limit the current flow. The volume between armature and stator is spectroscopically probed with fiber optics and valuable insight into the state of the shocked and compressed gas is gained. The same fiber optic probes are used to measure the velocity of the armature-stator contact along the generator axis. This contact velocity is largely affected by armature end effects, mainly due to the pressure loss at the detonator end. Both gas temperature and contact velocity have been successfully simulated with LS-DYNA3D, a three dimensional finite element hydrodynamic code. The generator's magnetic field structure is briefly discussed and magnetic field probe measurements are presented.

Authors: Hemmert, D; Neuber, AA; Dickens, JC; Krompholz, H; Hatfield, LL; Kristiansen, M

PDF: https://spie.org/Publications/Proceedings/Paper/10.1117/12.391818?SSO=1

Abstract: Microwave window breakdown is investigated in vacuum and atmospheric conditions. An S-band resonant ring with a frequency of 2.85 GHz and a power of 80 MW with a 4 MW magnetron as a source is used. Window breakdown on the vacuum side is simulated using a dielectric slab partially filling an evacuated waveguide. Various high-speed diagnostic methods yield a complete picture on the breakdown phenomenology, with far reaching similarities to de surface flashover. During the initiation phase, free electrons are present, which can be influenced by magnetic fields, followed by a saturated secondary electron avalanche with electron-induced outgassing. Final breakdown occurs in the desorbed gas layer above the surface. In order to simulate window breakdown on the gas-side, a segment of the resonant ring separated by two windows was filled with gas at variable pressure, and breakdown was initiated by field-enhancement tips on one of the gas-side surfaces. Threshold power densities for breakdown are measured, and first results on the phenomenology of this gas breakdown are compared with the processes of flashover in vacuum.

Authors: W. Jiang; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6220043

Abstract: A coaxial-type virtual cathode oscillator (vircator) was tested in the Pulsed Power Laboratory at Texas Tech University with electron beam parameters of 500 kV, 40 kA, and 30 ns. The major goal of this experiment was to understand the characteristics of the coaxial vircator and the effect of the feedback field on the microwave efficiency. The experimental results have given the output microwave power of ∼ 400 MW, the microwave energy efficiency of ∼ 2 %, the microwave frequency of ∼ 2 GHz, and the microwave mode of TE11 in circular waveguide. By enhancing the field intensity around the vircator using a microwave reflector, the microwave output power was increased to ∼ 900 MW giving microwave energy efficiency of ∼ 5.5 %. This efficiency is calculated as peak microwave power divided by peak beam power, which does not occur at the same time. The instantaneous efficiency, using the beam power at the time of the peak microwave power is 20%.

Authors: S. I. Shkuratov; M. Kristiansen; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=855151

Abstract: Summary form only given. The design of an explosive-to-electric transducer that uses the remnant magnetization in ferromagnetic cylinders is presented. 10-25 g charge of high explosives was used. The active element was NdFeB cylinders with diameter 2.5 cm and length 1.9 cm. The current generated in the low-resistance load of the generator yielded a peak 0.4 kA. Results of simulation of an explosive-to-electric transducer by Maxwell 3D code is presented. A detail analysis is given on the specific features of pulse generation in a system like this.

Authors: J. Kim; J. Zhang; M. Giesselmann; J. Dickens; J. Mankowski; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823802

Abstract: In this paper, the simulation and evaluation of energy efficiency and voltage amplification in an inductive energy storage system with resistive, capacitive and diode loads is presented. A numerical simulation with Mathcad shows that by proper reduction of the storage energy inductance and of the exploding fuse interruption time, the energy efficiency and voltage amplification can be greatly improved.

Authors: J. Mankowski; J. Dickens; M. Kristiansen; J. Lehr; W. Prather; J. Gaudet

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=829276

Abstract: Summary form only as given.

Authors: D. Hemmert; A. Neuber; J. Dickens; H. Krompholz; L. L. Hatfield; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=829539

Abstract: Summary form only given. Effects of the microwave magnetic field on window breakdown are investigated at the upstream and downstream side of a dielectric interface. Simple trajectory calculations of secondary electrons in an RF field show significant forward motion of electrons parallel to the microwave direction of propagation. The Lorentz-force due to the microwave magnetic field on high-energy secondary electrons might substantially influence the standard multipactor mechanism. As a result, the breakdown power level for the downstream side of a window would be higher than for the upstream side. This hypothesis was tested utilizing an S-band traveling wave resonant ring, powered by a 3 MW magnetron at 2.85 GHz, leading to a total power greater than 60 MW. Breakdown was studied on an interface geometry consisting of a thin alumina slab in the waveguide, oriented normal to the microwave propagation direction.

Authors: J. Rasty; Xiaobin Le; A. Neuber; Jiande Zhang; J. Dickens

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823611

Abstract: Characterization of changes in the conductivity during the shock loading process is of paramount importance in estimating the efficiency of magnetic flux compression generators (MFCG). In this study, the main emphasis was the characterization of dynamic conductivity of the armature material during the shock-loading phase. A Split Hopkinson Pressure Bar apparatus was utilized to subject as-received and annealed specimens of OFHC copper to various shock pressures. Experiments conducted to measure the resistivity of Cu specimens indicated that resistivity initially decreases, followed by a sharp increase before decreasing to a steady state value. Depending on the magnitude of the shock pressure, resistivity changes in excess of 200% were recorded.

Authors: W. Jiang; K. Woolverton; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=829727

Abstract: Summary form only as given.

Authors: M. Giesselmann; J. Zhang; T. Heeren; E. Kristiansen; J. Dickens; D. Castro; D. Garcia; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823810

Abstract: One of the key technologies in a high power microwave system is the pulsed power conditioning system (PPCS). For a system driven by an explosive flux compression generator, the PPCS may consist of an energy storage inductor, a fuse type opening switch and a sharpening spark gap. This paper presents the investigation of a PPCS with a pulse transformer. Before the construction of a prototype, the behavior of the PPCS was simulated using the PSpice circuit simulation code. A transformer with a primary inductance of 3.5 /spl mu/H, secondary inductance of 85 /spl mu/H and coupling coefficient of 0.75, was designed and used in the experiments. The transformer was designed with two coaxial windings. Simulation results as well as experimental waveforms are shown.

Authors: S. I. Shkuratov; M. Kristiansen; J. Dickens; L. L. Hatfield; R. Martin

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823613

Abstract: Results are presented of an experimental study of the generation of high-voltage pulses in generators designed as open and closed ferromagnetic circuits. Experiments were carried out using a light gas gun system. The magnetic projectiles were composed of NdFeB discs having 1.27 and 2.54-cm diameters. Data are given for the effects on the amplitude of high voltage pulses by the length and velocity of the ferromagnetic projectiles and the design of the generating unit for both high voltage and high current modes of pulsed power generation. A detailed analysis is presented on the specific features of pulse power generators using open and closed ferromagnetic circuits. Fundamental limitations due to the physical processes occurring in systems like these are determined.

Authors: S. I. Shkuratov; M. Kristiansen; J. Dickens; L. L. Hatfield; E. Horrocks

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823612

Abstract: Three types of resistors have been tested to determine maximum usable power at pulsed high voltage and pulsed high current. Experiments were carried out using high voltage cable generators and spark-gap generators. Pulse durations were varied from 0.7 /spl mu/s to 21 /spl mu/s. The pulse amplitudes were varied from 1 kV to 12 kV. The peak current reached was 3 kA. Metal film, carbon film and carbon composition resistors of four different rated powers (0.25 W, 0.5 W, 1 W, 2 W) have been tested. Data are given for the limiting pulsed energy for each type of resistor in nanosecond and microsecond time ranges. The mechanisms of failure and destruction of resistors under the action of high voltage and high current pulses are discussed.

Authors: M. Giesselmann; T. Heeren; E. Kristiansen; J. Dickens; D. Castro; D. Garcia; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823799

Abstract: This paper describes a metal oxide varistor (MOV) pulse shaping device, which is to be used for generating a flat-top voltage pulse for high power microwave devices. The MOV pulse-shaping device is attached to a coaxial pulse compression system. The pulse compression system consists of an inductive energy storage section, an exploding wire fuse and an output spark gap. The paper contains a detailed discussion of the pulse compression system followed by a description of the MOV pulse-shaping device. The MOV elements that are used for this pulse shaping application are the SIOV-E32VR602 type, made by Siemens.

Authors: J. Zhang; J. Dickens; M. Giesselmann; J. Kim; E. Kristiansen; J. Mankowski; D. Garcia; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=823610

Abstract: To improve the pulse shape and to obtain /spl mu/s order pulse duration on a diode load in an inductive energy storage system, an oil-submerged compact pulse transformer with diameter of 20 cm and length of 70 cm has been designed and tested. From theoretical calculations, the parameters with primary inductance L/sub p/=3.4 /spl mu/H, secondary inductance L/sub s/=90 /spl mu/H, and coupling coefficient K=0.772 are obtained. To prevent breakdown, a conical secondary design is adopted. Under the condition of 500 kV operation voltage, the maximum electric field in the transformer is 205 kV/cm (in oil). The axial voltage distribution on the secondary is linear. Experiments have been done to test the insulation, the mechanical force and the coupling coefficient. The test results are consistent with the design parameters.

Authors: G. Zamora; M. Dickens; S. Mitra

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=666865

Abstract: This paper describes an improved technique to register multi-sensor images by segmenting the images by adaptive clustering prior to performing preprocessing and cepstrum operation to determine the translational displacement. The difficulty in registering multi sensor images lies in the fact that the images of the same scene acquired by different sensors often appear different in detailed structures. Therefore the common features existing in such images need to be identified by suitable preprocessing operations for the success of the cepstral registration technique. Experimental results demonstrate the feasibility of successful cepstral registration of SAR and electro-optic images of the same scene despite apparent noticeable differences in some embedded structures thus providing a potential powerful tool for automated registration.

Authors: S. Mitra; M. Dickens; S. Pemmaraju

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=686367

Abstract: Automatic detection or recognition of specific objects from a sequence of images acquired under varying conditions and from different modalities requires careful segmentation. We present here the use of an adaptive neuro-fuzzy clustering technique for image segmentation and boundary detection for better extraction of the dominant features in images of the same scene acquired by synthetic aperture radar (SAR) and electro-optic (EO) sensors. The advantage of such adaptive segmentation is clearer identification of objects that appear differently due to different sensor characteristics.

Authors: A. Neuber; J. Dickens; D. Hemmert; H. Krompholz; L. L. Hatfield; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=677696

Abstract: Summary form only given. Physical mechanisms leading to microwave breakdown on dielectric/vacuum interfaces are investigated for power levels on the order of 100 MW at 2.85 GHz. A 3 MW magnetron with 3.5 /spl mu/s pulse width, is coupled to an S-band traveling wave resonator which is kept at a pressure of 10/sup -8/ Torr. The investigation is focused on an interface geometry comprising a thin dielectric polymer slab in the waveguide, oriented vertical to the direction of wave propagation, and two field enhancement tips placed in the middle of each waveguide broad wall. This ensures an almost purely tangential field at the interface surface and a localized breakdown.

Authors: Giesselmann, MG; Haider, MR

PDF: https://ieeexplore.ieee.org/document/741236

Abstract: Prototypes of three-phase synchronous rectifiers using power MOSFETs including the necessary control circuitry have been developed and tested. Experimental results show a significant reduction of conduction losses at room temperature and large additional improvements if the MOSFETs are cooled with liquid nitrogen. The synchronous rectifier is proposed to be used as part of a power system for a next generation ground based radar (GBR) system.

Authors: J. Mankowski; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=677842

Abstract: Summary form only given, as follows. Present day ultra-wideband (UWB) radiation sources a produce Megavolt pulses at 100's of picosecond (ps) risetimes. Empirical data on the breakdown characteristics for dielectric media at these short time lengths and high voltages are either extremely limited or non-existent. In support of the design of these UWB sources, we are investigating the breakdown characteristics, at these voltages and time lengths, of several liquids and gases. These include air, N/sub 2/, H/sub 2/, He, SF/sub 6/, and transformer oil. The two voltage sources used in the experiments are capable of delivering 400 and 700 kV with a 400 ps risetime into an open load. These pulses are applied to the test gap area, capable of housing various gases and liquids at pressures from less than 1 to 150 atm. An empirical relationship of E-field versus breakdown time for the observed dielectrics is presented. Several other breakdown phenomena at these fast risetimes are observed. Dielectric breakdown strength dependence on polarity is investigated. Streak camera images of arc formation are captured, providing information on gap closure velocity. Also observed is the effect of ultraviolet radiation on the statistical lag time of breakdown for gas dielectrics at various pressures.

Authors: E. O'Hair; J. Dickens; J. Fralick; L. Farrar

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=677653

Abstract: Summary form only given, as follows. Thermal plasmas are being evaluated for use in the destruction of deadly chemical and biological warfare agents. The technology approach consists of employing an electric arc to establish a plasma forming gas plume used to kill the harmful agents. The approach has been demonstrated on surrogate chemical and biological warfare agents. The harmful agents are rapidly destroyed, e.g., within a few seconds. Three different plasmas have been evaluated for their effectiveness to destroy harmful agents. The approaches used include nitrogen plasma, air plasma and a novel water plasma arcjet. The work incorporates an evaluation of the durability of the plasma arcjet technology for this application, e.g., cyclic life and steady-state durability. The follow-on phase of this project will result in a prototype system development suitable for field application. Anticipated benefits are the development of an effective system that will allow rapid and f-effective decontamination of surfaces that have been contaminated with chemical and biological warfare agents. No harmful byproducts will be produced. Additionally, the methods developed have application in the destruction of medical wastes, hazardous industrial wastes and in the destruction of hazardous materials associated with worldwide demilitarization activities.

Authors: M. M. Dickens; D. J. Bornhop; S. Mitra

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=701364

Abstract: Flexible micro-endoscopes produce images that are obscured by a 'honeycomb' pattern due to the negative space between the individual optical fibers contained in the imaging conduit. This pattern is found to exhibit a spatial frequency that is visibly distinct from that of the imaged object. By applying a frequency filter, it was possible to remove the honeycomb pattern without significant degradation to the visual quality of the image. This process greatly increases the perceived quality of the information being obtained by the endoscopes and aids in their effective use for medical diagnosis. The technique described employs Fourier spectral analysis to determine the 'noise' component in the original image. A discrete band-reject frequency filter was designed by visually examining the spectral information and creating the necessary filter to block out the undesired frequency bond. The honeycomb pattern was no longer distinguishable after applying this filter to a test gray-level image of an Air Force calibration target. Next, the filter was applied to the individual color planes of a sample color image. The color planes were recombined to produce a full-color image that was free from interference. A full description of the methods involved is presented.

Authors: J. Mankowski; J. Dickens; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=679394

Abstract: A high voltage, sub-nanosecond pulser is designed and built for the purposes of investigating dielectric breakdown. The requirement for the pulser is a voltage pulse of several hundred kilovolts, pulsewidth less than a few nanoseconds, and a risetime <400 psec. This is achieved by using pulse sharpening techniques on the output of a 500 kV Marx bank. Originally designed to stimulate a lightning strike, the voltage waveform from the Marx generator has a slow decay time of tens of /spl mu/sec. In order to obtain a more desirable pulse, the Marx bank is modified. By removing a lumped-element resistor a higher peak output voltage with a faster risetime can be obtained. Circuit simulations have shown the capability of achieving an 800 kV output in less than 40 nsec to a charging (pulse forming) line. The 50 /spl Omega/ impedance, oil-filled, pulse forming line consists of a peaking gap and pulse slicing gap. The peaking gap decreases the risetime of the applied pulse down to about 300 psec. The pulse slicing gap is included to short the voltage applied to the delay line and test chamber.

Authors: Giesselmann, MG

PDF: https://ieeexplore.ieee.org/document/659210

Abstract: This paper shows computer models for common switch-mode converters with averaged and conventional cycle by cycle switching operation. Models for the Buck, Boost, Buck-Boost, and Cuk converter are presented. Also shown are Buck and Boost converters with integrated magnetic components. All time averaged models are using an identical averaged model for the PWM switch. The models are implemented using PSpice(R) for Windows(R). The advantage of averaged models are much faster execution time as well as the ability to perform frequency domain (bode plots) analysis in addition to time domain analysis.

Authors: Woolverton, K; Kristiansen, M; Hatfield, LL

PDF: https://www.spie.org/Publications/Proceedings/Paper/10.1117/12.284009?SSO=1

Abstract: A study of the efficiency of a coaxial virtual cathode oscillator is presented. The coaxial geometry has many physical parameters that can be changed to alter performance. The parameters of interest include the placement of a ring cut in the anode base and the polarity of the system. The ring creates a decelerating field for the electrons and tends to keep them in the right-phased region of the virtual cathode. The ring is varied in width and in position from the center line with the results normalized to the no ring geometry. The results for a positively and negatively pulsed system are also given. Comparisons of frequency, efficiency, and particle dynamics of the positively and negatively pulsed systems are given. MAGIC, a 2-1/2 dimensional particle-in-cell code, and SOS, a 3 dimensional particle-in-cell code, are used to simulate the different geometries.

Authors: Woolverton, K; Kristiansen, M; Hatfield, LL

PDF: https://ieeexplore.ieee.org/document/679455

Abstract: A study on the interaction dynamics of a positively and negatively pulsed coaxial vircator is being performed at Texas Tech University. MAGIC, a 2.5D particle-in-cell code, is used to simulate the different geometries. The simulations performed indicate an increase in efficiency by approximately a factor of 2 for the negatively pulsed system compared to the positively pulsed system. Simulations are also performed to better understand the influencing factors of the systems. This paper describes the experimental results which are performed on a coaxial vircator with diode voltages from 400-500 kV at diode currents of 40-50 kA with pulse durations of similar to 50 nsec. Results that are given include microwave power, efficiency, spectral content and a possible explanation for the results.

Authors: J. Dickens; M. Kristiansen; M. Giesselmann; J. G. Kim

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=679365

Abstract: A Russian power modulator, which is based on a diode that has been touted as a high-power solid-state pulsed power opening switch (SOS) was characterized and tested. The diode has characteristics similar to those of a step-recovery diode, except that the reverse current density is 10-100 times larger than in US manufactured diodes. The modulator has a DC input voltage of 600 V and produces a pulse of 150 kV amplitude and 25 ns half width into a 350 /spl Omega/ load. The maximum pulse repetition rate is 100 Hz. The limit is given by thermal limitations and can be increased through the use of more effective cooling methods.

Authors: Li, SH; Wunsch, DC; OHair, E; Giesselmann, MG

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=611648

Abstract: The power generated by electric wind turbines changes rapidly because of the continuous fluctuation of wind speed and direction. It is important for the power industry to have the capability to estimate this changing power. In this paper, the characteristics of wind power generation are studied and a neural network is used to estimate it. We use real windfarm data to demonstrate a neural network solution for this problem, and show that the network can estimate power even in changing wind conditions.

Authors: Akiyama, H; Nishihashi, Y; Tsukamoto, S; Sueda, T; Katsuki, S; Hagler, M; Dickens, JC; Inoue, N

PDF: https://ieeexplore.ieee.org/document/679286

Abstract: Streamer corona discharges produced by pulsed power make enough high energy electrons to dissociate gases directly. Consequently, pulsed streamer discharges have been proposed for the removal of NOx and SO2 from flue gases, the production of ozone and the treatment of poisonous gases. It is proposed here to produce the pulsed streamer discharges on a spiral transmission line. The characteristics of the discharges are studied, comparing with results using a PSpice code. Then, the effectiveness and advantages of the spiral transmission line are discussed in view of real applications.

Authors: Neuber, A; Dickens, J; Hemmert, D; Krompholz, H; Hatfield, LL; Kristiansen, M

PDF: https://ieeexplore.ieee.org/document/604787

Abstract: Physical mechanisms leading to microwave breakdown on windows and in cavities are investigated for power levels on the order of 100 MW at 2.85 GHz. The test stand uses a 3 MW magnetron coupled to an S-band traveling wave resonator. Various configurations of dielectric windows are investigated. In a standard pillbox geometry with a pressure of less than 10(-8) torr, surface discharges on an alumina window and multipactor-like discharges starting at the waveguide edges occur simultaneously. To clarify physical mechanisms, window breakdown with purely tangential electrical microwave fields is investigated for special geometries. Other configurations, such as air filled two window setups, relevant for vacuum-air interfaces, can be investigated as well. Diagnostics include the measurement of incident/reflected power, measurement of local microwave fields, discharge luminosity, and x-ray emission. All quantities are recorded with 0.2 to 1 ns resolution. In addition, a framing camera with gating times of 5 ns is used. Based on the experimental results, methods to increase the power density which can be transmitted through windows, such as surface coatings and window profiles, will be investigated as well.

Authors: Agee, FJ; Scholfield, DW; Copeland, RP; Martin, TH; Carroll, JJ; Mankowski, JJ; Kristiansen, M; Hatfield, LL

PDF: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/2843/1/Review-of-catastrophic-electromagnetic-breakdown-for-short-pulse-widths/10.1117/12.255410.short?SSO=1

Authors: J. C. Dickens; J. Elliott; L. L. Hatfield; M. Kristiansen; H. Krompholz

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=551438

Abstract: Summary form only given. Investigations of breakdown phenomena at window interfaces from high vacuum (<10/sup -7/ torr) to atmospheric pressure are investigated for microwave power levels of up to 100 MW. The test stand utilizes a 3 MW magnetron operating at 2.8 GHz, coupled to an S-band traveling wave resonant ring. Various configurations of dielectric windows (i.e. vacuum-air, or vacuum-vacuum), in various geometries (standard pillbox geometry, or windows filling the S-band waveguide cross section) are investigated. Diagnostics include the measurement of transmitted/reflected microwave power, luminosity from the discharge plasma, X-ray emission from initially free electrons, and electric field probes. All these quantities are measured with high amplitude and high temporal (0.2 to 1 ns) resolution. Goals are to determine the physical mechanisms-such as the dominant electron multiplication process-leading to flashover. The knowledge gained from these experiments is used to investigate and design methods to increase the power density which can be transmitted through windows. In addition, parametric studies are conducted, in which window material, profile, and surface coatings are varied. The basic system and the diagnostics methods are expanded for the investigation of microwave cavity breakdown as well.

Authors: J. C. Dickens; J. Bridges; M. Kristiansen

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=564449

Abstract: A compact modulator system using inductive energy storage and a diode as an opening switch was investigated. The system was designed around a Russian diode that has characteristics similar to those of a step-recovery diode except that the reverse current density is 10-100 times larger than in US manufactured diodes. The main goal of this investigation was to characterize the Russian diode and develop an understanding of its operation for possible improvement and integration into nanosecond pulse generators. The basic modulator circuit designed uses IGBTs in the forward pumping circuit and is capable of delivering 5 kV at 400 A into the diode. Using this forward pumping circuit, diode voltage and current measurements were made at several input power levels. In addition, data obtained using this test stand was used to produce a computer simulation model of the diode for a more thorough circuit analysis.

Authors: Giesselmann, M; Crittenden, B

PDF: https://ieeexplore.ieee.org/document/564494

Abstract: The design and construction of a Neutral Point Clamped Inverter along with its modulation strategy and circuit simulation using PSPICE 6.2 for Windows is presented. This inverter is a subscale (100 kW) prototype for a inverter that is intended to be used in a 1MW level universal power converter system to produce 3 phase AC power at 60 Hz from variable input sources.

Authors: Giesselmann, MG

PDF: https://ieeexplore.ieee.org/document/552888

Abstract: Different models for switched reluctance machines (SRMs) are presented. All models are developed for the newest version (6.3) of PSpice for Windows. The models demonstrate the software's capabilities to define custom symbols and construct multilevel hierarchical models. PSpice also has strong analog behavioral modeling capabilities that are fully exploited. For this paper, 6-4 motors are simulated, but the models can be extended to other designs. The first model is based on an equivalent circuit representation with linear, voltage controlled inductors. The second model is based on a set of analytical equations which include saturation effects. In both cases, the electrical and mechanical system is modeled. Coupling between the windings is neglected. The models yield information about electrical as well as mechanical parameters such as induced phase voltage and torque. Both models produce comparable results in the unsaturated regime for the phase current. At higher current levels the model based on the analytical equations gives more accurate results.

Authors: Giesselmann, M; Mahmud, Z; Carson, S

PDF: https://ieeexplore.ieee.org/document/564446

Abstract: Operation of power MOSFETs (Metal Oxide Semiconductor Field Effect Transistor) at cryogenic temperatures significantly reduces conduction losses and increases power handling capability. High voltage (1000V+) devices exhibit the largest reduction of conduction losses. The breakdown voltage of the devices is reduced by about 20% when cooled from 75 degrees F down to -319 degrees F.

Authors: J. Mankowski; M. Kristiansen; L. Hatfield

PDF: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6308398

Abstract: Breakdown characteristics of various liquids are investigated. Electric field strengths recorded are over 4 MV/cm and at breakdown times ranging from 1 to 4 nanoseconds. Liquids examined are filtered and unfiltered transformer oil, castor oil, and freon 12.

Authors: Hegeler, F; Krompholz, H; Hatfield, LL; Kristiansen, M

PDF: https://ieeexplore.ieee.org/document/483603

Authors: Miedzinski, B; Okraszewski, Z; Szymanski, A; Kristiansen, M

PDF: https://ieeexplore.ieee.org/document/530542

Authors: CHOWDHURY, G; GIESSELMANN, M

PDF: https://arc.aiaa.org/doi/10.2514/6.1994-3810

Authors: ENGEL, TG; WESTER, SL; KRISTIANSEN, M; HATFIELD, LL

PDF: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/2259/1/Mass-erosion-and-surface-voltage-holdoff-recovery-of-insulators-used/10.1117/12.174659.short?SSO=1

Authors: HEGELER, F; KROMPHOLZ, H; HATFIELD, LL; KRISTIANSEN, M

PDF: https://ieeexplore.ieee.org/document/592035

Authors: GIESSELMANN, M; ELDRED, C; SEITZ, D

PDF: https://www.scopus.com/record/display.uri?eid=2-s2.0-0027885749&origin=inward&txGid=7c4a2bb1b0de4789c46ed68da174b31e

Authors: DICKENS, JC; ENGEL, TG; KRISTIANSEN, M

PDF: https://ieeexplore.ieee.org/document/513376

Abstract: Electrode erosion rates and voltage holdoff !ability are presented for various electrode materials sed in a triggered spark gap switch. The switch tilized an electrical trigger with the triggered electrode ositioned perpendicular to the main gap. Results show significant reduction in electrode erosion rates over Lose for non-triggered gaps. The lowered erosion rates re attributed in part to the lower voltage and current roduct during switch turn-on. The tests are conducted xoss a broad range of peak currents (i.e. 10 kA - 400 A) and materials, includmg copper-tungsten, coppermgsten+LaBg, and tungste

Authors: GIESSELMANN, M; LOREE, DL

PDF: https://www.scopus.com/record/display.uri?eid=2-s2.0-0027904205&origin=inward&txGid=10cf4925e71d4129011e5dd2736705d2

Authors: LOREE, D; GIESSELMANN, M; KRISTIANSEN, M

PDF: https://ttu-ir.tdl.org/bitstream/handle/2346/17351/31295006979529.pdf;sequence=1

Authors: PALSULE, CP; GANGOPADHYAY, S; YOUNG, C; TROST, T; KRISTIANSEN, M

PDF: https://aip.scitation.org/doi/10.1063/1.41032

Authors: GANGOPADHYAY, S; TROST, T; KRISTIANSEN, M; YOUNG, C; ZHENG, P; PALSULE, C; PLEIL, M

PDF: https://link.springer.com/article/10.1557/PROC-192-627

Authors: ENGEL, TG; KRISTIANSEN, M; BAKER, M; HATFIELD, LL

PDF: https://ieeexplore.ieee.org/document/201000

Authors: DONALDSON, AL; KRISTIANSEN, M

PDF: https://ieeexplore.ieee.org/document/201003