D. M. Barrett University of California Lawrence Livermore National Laboratory Livermore, CA (b) PFL Voltage at Point A. (c) Voltage Across Load

Similar documents
FLASH X-RAY (FXR) ACCELERATOR OPTIMIZATION BEAM-INDUCED VOLTAGE SIMULATION AND TDR MEASUREMENTS *

FLASH X-RAY (FXR) ACCELERATOR OPTIMIZATION INJECTOR VOLTAGE-VARIATION COMPENSATION VIA BEAM-INDUCED GAP VOLTAGE *

A NEW BROADBAND PULSED HIGH VOLTAGE MONITOR *

ANALYSIS OF A PULSED CORONA CIRCUIT

RAVEN, A 5 kj, 1.5 MV REPETITIVE PULSER* G. J. Rohwein Sandia National Laboratories Albuquerque, New Mexico 87185

TRANSMISSION LINE AND ELECTROMAGNETIC MODELS OF THE MYKONOS-2 ACCELERATOR*

ANALYSIS OF SWITCH PERFORMANCE ON THE MERCURY PULSED- POWER GENERATOR *

An experimental system was constructed in which

DEVELOPMENT OF AN ULTRA-COMPACT EXPLOSIVELY DRIVEN MAGNETIC FLUX COMPRESSION GENERATOR SYSTEM

INVESTIGATION OF A HIGH VOLTAGE, HIGH FREQUENCY POWER CONDITIONING SYSTEM FOR USE WITH FLUX COMPRESSION GENERATORS

IB2-1 HIGH AVERAGE POWER TESTS OF A CROSSED-FIELD CLOSING SWITCH>:< Robin J. Harvey and Robert W. Holly

PULSED POWER SWITCHING OF 4H-SIC VERTICAL D-MOSFET AND DEVICE CHARACTERIZATION

Modeling Antennas on Automobiles in the VHF and UHF Frequency Bands, Comparisons of Predictions and Measurements

W. L. Bird University of Texas, Austin, Center for Electro-Mechanics, Taylor Hall 167, Austin, TX 78712

ULTRA FAST, HIGH REP RATE, HIGH VOLTAGE SPARK GAP PULSER

9.4 A HIGH CURRENT PULSER FOR EXPERIMENT 11225, "NEUTRINO ELECTRON ELASTIC SCATTERING" C. Dalton, G. Krausse, and J. Sarjeant

PHASING CAPABILITY. Abstract ARRAY. level. up to. to 12 GW. device s outpu antenna array. Electric Mode. same physical dimensions.

8.2. Washington, D. C delivered 65 kj into a matched load with 63 ns FWHM. Peak power was about 1 TW.

PULSED BREAKDOWN CHARACTERISTICS OF HELIUM IN PARTIAL VACUUM IN KHZ RANGE

David L. Lockwood. Ralph I. McNall Jr., Richard F. Whitbeck Thermal Technology Laboratory, Inc., Buffalo, N.Y.

PERFORMANCE OF A 10 KV, 625 KA, 85 KJ ENERGY DISCHARGE MODULE UTILIZING A SOLID DIELECTRIC SWITCH.*

L. B. Gordon Space Power Institute 231 Leach Center Auburn University, Alabama 36849

CHARACTERIZATION OF PASCHEN CURVE ANOMOLIES AT HIGH P*D VALUES

Strategic Technical Baselines for UK Nuclear Clean-up Programmes. Presented by Brian Ensor Strategy and Engineering Manager NDA

UPGRADES TO THE DARHT SECOND AXIS INDUCTION CELLS

OPTICAL EMISSION CHARACTERISTICS OF HELIUM BREAKDOWN AT PARTIAL VACUUM FOR POINT TO PLANE GEOMETRY

EFFECT OF TRANSFORMER LEAKAGE INDUCTANCE ON THE THREE PHASE CAPACITIVE INPUT RECTIFIER

Robotics and Artificial Intelligence. Rodney Brooks Director, MIT Computer Science and Artificial Intelligence Laboratory CTO, irobot Corp

ACCELERATOR FAST KICKER R&D WITH ULTRA COMPACT 50MVA NANO-SECOND FID PULSE GENERATOR

EVALUATION OF RESISTORS FOR TRANSIENT HIGH-VOLTAGE APPLICATIONS

DEVELOPMENT OF STITCH SUPER-GTOS FOR PULSED POWER

Report Documentation Page

COMPACT FLASH X-RAY UNITS. Abstract

Sea Surface Backscatter Distortions of Scanning Radar Altimeter Ocean Wave Measurements

DIELECTRIC ROTMAN LENS ALTERNATIVES FOR BROADBAND MULTIPLE BEAM ANTENNAS IN MULTI-FUNCTION RF APPLICATIONS. O. Kilic U.S. Army Research Laboratory

Development of a charged-particle accumulator using an RF confinement method FA

Signal Processing Architectures for Ultra-Wideband Wide-Angle Synthetic Aperture Radar Applications

Investigation of a Forward Looking Conformal Broadband Antenna for Airborne Wide Area Surveillance

A Comparison of Two Computational Technologies for Digital Pulse Compression

Experimental Observation of RF Radiation Generated by an Explosively Driven Voltage Generator

"OPTIMAL SIMULATION TECHNIQUES FOR DISTRIBUTED ENERGY STORE RAILGUNS WITH SOLID STATE SWITCHES"

J, 1. lj, f J_ Switch DESIGN OF A PULSED-CURRENT SOURCE FOR THE INJECTION-KICKER MAGNET AT THE LOS ALAMOS NEUTRON SCATTERING CENTER ABSTRACT

Durable Aircraft. February 7, 2011

INDUCTIVE VOLTAGE ADDER NETWORK ANALYSIS AND MODEL SIMPLIFICATION

Solar Radar Experiments

Loop-Dipole Antenna Modeling using the FEKO code

MEASUREMENTS OF THE RADIATED FIELDS AND CONDUCTED CURRENT LEAKAGE FROM THE PULSED POWER SYSTEMS IN THE NATIONAL IGNITION FACILITY AT LLNL

FAA Research and Development Efforts in SHM

STABILITY AND ACCURACY OF THE REALIZATION OF TIME SCALE IN SINGAPORE

MINIATURIZED ANTENNAS FOR COMPACT SOLDIER COMBAT SYSTEMS

U.S. Army Training and Doctrine Command (TRADOC) Virtual World Project

A COMPACT, 1-MV, 6-kA RADIOGRAPHY SOURCE WITH A ONE- METER EXTENSION AND RIGHT-ANGLE BEND

EFFECTS OF ELECTROMAGNETIC PULSES ON A MULTILAYERED SYSTEM

Lattice Spacing Effect on Scan Loss for Bat-Wing Phased Array Antennas

9.2. FWG As An RF Source

REGULATED CAPACITOR CHARGING CIRCUIT USING A HIGH REACTANCE TRANSFORMER 1

D.V.Giri, Pr<r Tech, 1630 North Main Street, #377 Walnut Creek, California and L A REALISTIC ANALYTICAL MODEL FOR THE PULSER

Modeling of Ionospheric Refraction of UHF Radar Signals at High Latitudes

INFRASOUND SENSOR MODELS AND EVALUATION. Richard P. Kromer and Timothy S. McDonald Sandia National Laboratories

Non-Data Aided Doppler Shift Estimation for Underwater Acoustic Communication

23.3 PRELIMINARY DESIGN OF A FAST RISE, FULL THREAT LIGHTNING SIMULATOR. F. Swartz Air Force Institute of Technology Wright-Patterson AFB OH 45433

SILICON CARBIDE FOR NEXT GENERATION VEHICULAR POWER CONVERTERS. John Kajs SAIC August UNCLASSIFIED: Dist A. Approved for public release

August 9, Attached please find the progress report for ONR Contract N C-0230 for the period of January 20, 2015 to April 19, 2015.

Technology Maturation Planning for the Autonomous Approach and Landing Capability (AALC) Program

Best Practices for Technology Transition. Technology Maturity Conference September 12, 2007

Radar Detection of Marine Mammals

COM DEV AIS Initiative. TEXAS II Meeting September 03, 2008 Ian D Souza

14. Model Based Systems Engineering: Issues of application to Soft Systems

EXPERIMENTS ON A HIGH-VACUUM, HIGH-ELECTRIC FIELD STRESS PULSED POWER INTERFACE

Student Independent Research Project : Evaluation of Thermal Voltage Converters Low-Frequency Errors

Fuse and Load Testing With Mid-Sized, High Energy Density Flux Compression Generators

Thermal Simulation of a Silicon Carbide (SiC) Insulated-Gate Bipolar Transistor (IGBT) in Continuous Switching Mode

REPORT DOCUMENTATION PAGE

EVALUATION OF COMMERCIALLY AVAILABLE IGNITRONS AS HIGH-CURRENT, HIGH-COULOMB TRANSFER SWITCHES*

HIGH TEMPERATURE (250 C) SIC POWER MODULE FOR MILITARY HYBRID ELECTRICAL VEHICLE APPLICATIONS

A R Miller Maxwell Laboratories, Inc Balboa Ave., San Diego, CA

A HIGH CURRENT, HIGH VOLTAGE SOLID-STATE PULSE GENERATOR FOR THE NIF PLASMA ELECTRODE POCKELS CELL*

Innovative 3D Visualization of Electro-optic Data for MCM

Remote Sediment Property From Chirp Data Collected During ASIAEX

Coherent distributed radar for highresolution

Modeling an HF NVIS Towel-Bar Antenna on a Coast Guard Patrol Boat A Comparison of WIPL-D and the Numerical Electromagnetics Code (NEC)

0.18 μm CMOS Fully Differential CTIA for a 32x16 ROIC for 3D Ladar Imaging Systems

REPORT DOCUMENTATION PAGE. A peer-to-peer non-line-of-sight localization system scheme in GPS-denied scenarios. Dr.

Hybrid QR Factorization Algorithm for High Performance Computing Architectures. Peter Vouras Naval Research Laboratory Radar Division

Underwater Intelligent Sensor Protection System

CFDTD Solution For Large Waveguide Slot Arrays

UNCLASSIFIED UNCLASSIFIED 1

REPORT DOCUMENTATION PAGE. 1. REPORT DATE (DD-MM-YYYY) 2. REPORT TYPE 3. DATES COVERED (From - To) Monthly IMay-Jun 2008

Effects of Radar Absorbing Material (RAM) on the Radiated Power of Monopoles with Finite Ground Plane

ULTRASTABLE OSCILLATORS FOR SPACE APPLICATIONS

SYSTEMATIC EFFECTS IN GPS AND WAAS TIME TRANSFERS

Adaptation of ASTERIX to Positive Polarity for 2 to 4-MV Rod-Pinch Diode Experiments and Diode Electrical Analysis *

3. Faster, Better, Cheaper The Fallacy of MBSE?

Drexel Object Occlusion Repository (DOOR) Trip Denton, John Novatnack and Ali Shokoufandeh

A 1.1 MV REP-RATE IN-LINE OUTPUT SWITCH AND TRIGGERING SYSTEM

Frequency Stabilization Using Matched Fabry-Perots as References

Reduced Power Laser Designation Systems

Rump Session: Advanced Silicon Technology Foundry Access Options for DoD Research. Prof. Ken Shepard. Columbia University

Marine Mammal Acoustic Tracking from Adapting HARP Technologies

Design of Synchronization Sequences in a MIMO Demonstration System 1

Transcription:

DESIGN CRITERIA FOR A MAGNETIC SWITCH WHEN USED TO DISCHARGE A PULSE FORMING UNE* D. M. Barrett University of California Lawrence Livermore National Laboratory Livermore, CA 94550 Abstract Much has been written concerning the design of magnetic switches in Melville line pulse compression networks. In these networks, magnetic switches are used to discharge one pacitor into another. Pulse compression is achieved by discharging each pacitor at a faster rate than it was charged. The relationship between the circuit parameters of the Melville line the individual magnetic switches is fairly well understood. However, magnetic switches are sometimes st in roles which depart from their traditional use in a Melville line network. One such applition is that of a magnetic switch being used to discharge a pulse forming line (PFL) into a load. In this paper, the requirements of such an applition are discussed. Relationships between the PFL discharge parameters (i.e., pulse energy, risetime, pulsewidth) the magnetic switch parameters are derived. A design example of a magnetic switch for use in a PFL discharge applition is presented the design trade-offs are also discussed. Introduction Some devices, such as linear induction accelerators, klystrons, kicker magnets certain types of lasers, must be driven with precisely shaped rectangular pulses. In most ses, it is desired to minimize the rise fall times of the pulse while maintaining a peak amplitude variation of a few percent or less. For pulsewidths of less than a few hundred nanoseconds, a pulseforming-line (PFL) is often used to generate the desired pulseshape. When this technique is used, the PFL is typilly charged on a time sle which is signifintly longer than the output pulsewidth. Once the PFL is fully charged, an output switch discharges the PFL into the load. Traditionally, gas discharge switches such as spark gaps thyratrons have been used for this applition. However, the reliability of such switches limits the performance of these systems at high average power levels. More recently, saturable reactors have been used as PFL discharge switches [1,2,3]. Magnetic switches do not suffer from many of the limitations associated with discharge switches such as relatively long recovery times serious electrode erosion problems. Unfortunately, the performance of a magnetilly switched PFL circuit is limited by the losses associated with the magnetic switch. The principal parameters which determine the loss of a magnetic switch are the loss characteristics the volume of the magnetic material which is being used in the reactor core. The purpose of this paper is to develop relationships between the desired PFL circuit performance the required magnetic switch parameters. 0 Pulse Forming Line!.e._ td= 2 'Zo= R (a) Diagram of a magnetilly switched PFL circuit. _.! tp~ ~ I (b) PFL Voltage at Point A. (c) Voltage Across Load Fig. 1. Circuit diagram typil voltage waveforms for a magnetilly switched PFL circuit. The charging waveform departs slightly from the traditional (1-cosine) voltage waveform due to reflections which propagate back forth on the PFL as it is being charged. As the PFL charge time decreases, the reflection hence the distortion of the charging waveform becomes increasingly pronounced. However, the integral of the voltage waveform is modified only slightly. When the PFL is fully charged, the output reactor L 1 saturates discharging the PFL into the load. Since R1 = Z 0, The maximum load voltage is half of the PFL load voltage as shown. Therefore, the energy content of the output pulse is where V 0 Rl 2 maximum PFL charge voltage load impedance width of the output pulse. I 1 I Operation of Magnetilly Swi)Ched PFL Circuits A magnetilly switched PFL network is shown in Fig. 1. Here, a transmission line having a characteristic impedance a one-way transist time of Z 0 tpf2. respectively, is used as the PFL. For this discussion, it is assumed that the load impedance R 1 is matched to Z 0 Upon closure of switch S1, energy initially stored in c 1 is transferred to the PFL charging it with a voltage waveform as shown. * This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48. 735 The risetime of the load voltage pulse is determined by the UR time constant of the PFL discharge path is given by where Lsat = saturated inductance of the output reactor. [2) In many applitions, the pulse risetime is defined as the duration over which the leading edge of the pulse rises from 1 0% to 90% of its peak value. Under this constraint tr (1 0%-90%) = 2.2 UR. However, in applitions such as linear induction accelerator drive systems [4), where the useable portion of the pulse begins within a few percent of the flat-top, this definition of

Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering maintaining the data needed, completing reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithsting any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE JUN 1991 2. REPORT TYPE N/A 3. DATES COVERED - 4. TITLE AND SUBTITLE Design Criteria For A Magnetic Switch When Used To Discharge A Pulse Forming Une 5a. CONTRACT NUMBER 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) University of California Lawrence Livermore National Laboratory Livermore, CA 94550 8. PERFORMING ORGANIZATION REPORT NUMBER 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR S ACRONYM(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release, distribution unlimited 11. SPONSOR/MONITOR S REPORT NUMBER(S) 13. SUPPLEMENTARY NOTES See also ADM002371. 2013 IEEE Pulsed Power Conference, Digest of Technil Papers 1976-2013, Abstracts of the 2013 IEEE International Conference on Plasma Science. Held in San Francisco, CA on 16-21 June 2013. U.S. Government or Federal Purpose Rights License. 14. ABSTRACT Much has been written concerning the design of magnetic switches in Melville line pulse compression networks. In these networks, magnetic switches are used to discharge one pacitor into another. Pulse compression is achieved by discharging each pacitor at a faster rate than it was charged. The relationship between the circuit parameters of the Melville line the individual magnetic switches is fairly well understood. However, magnetic switches are sometimes st in roles which depart from their traditional use in a Melville line network. One such applition is that of a magnetic switch being used to discharge a pulse forming line (PFL) into a load. In this paper, the requirements of such an applition are discussed. Relationships between the PFL discharge parameters (i.e., pulse energy, risetime, pulsewidth) the magnetic switch parameters are derived. A design example of a magnetic switch for use in a PFL discharge applition is presented the design trade-offs are also discussed. 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF ABSTRACT SAR a. REPORT b. ABSTRACT c. THIS PAGE 18. NUMBER OF PAGES 4 19a. NAME OF RESPONSIBLE PERSON Stard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18

risetime is inappropriate. In such applitions, the risetime must be defined as the duration over which the pulse rises from 0% to 100% of its peak value. As a result, tr in these applitions is given by 2.5 lsat tr =SluR=-- RI Derjyation of Mjnjmum Magnetic Material Volume In order to achieve the most efficient magnetic switch design, it is necessary to minimize the volume of the given magnetic material. The minimum magnetic material volume is determined by the saturated inductance required to yield the specified pulse risetime as well as the cross-section area required to yield a specified volt-second product. Typilly, a magnetic switch is constructed on a toroidal core having an inner radius, outer radius a height of ri, r 0 h, respectively, as shown in Fig. 2. The core is surrounded by a winding which is displaced a distance x from the surface of the core. The magnetic cross-section area Am is the product of the core cross-sectional area Ac the core stacking factor f 5. The winding area Aw is defined as the crosssectional area enclosed by the winding the packing factor fp is defined as the ratio of Am to Aw. [3] Rearranging Eq. [6] substituting into Eq. [5], Volm n be rewritten as [7] Using Eq. [3], Lsat n be expressed in terms of the required pulse risetime substituted into Eq. [7] yielding 2.5 <Jlsat> Volm =...;;;:;;;_ tr Using Eq. [1], Eq. [8] n be rewritten as [8] <Jlsat> represents the saturated permeability of the magnetic core averaged over the cross-sectional area of the winding therefore n be written as [ 9] The saturated inductance of the magnetic switch is given by [4] where Jlsat = saturated permeability of the magnetic material. Therefore, Eq. [9] n be rewritten as Winding Magnetic Core where 2.5 Jlo E 0 2 Volm = fv (gam)!182 fv = magnetic volume factor [ 1 1] fp(jlsat - 1) + 1 fp where Aw. Fig. 2. Typil geometry of a magnetic switch. N = number of turns on the core <I> = mean magnetic path length <Jlsat> = saturated permeability of the core averaged over Since the magnetic volume Volm = Am<l>, Volm n be expressed as [5] Eq. [11] indites that the magnetic volume is directly proportional to the energy which is being switched as well as the square of the gain of the magnetic switch. The volume is inversely proportional to the square of the flux swing in the core. In addition, Volm is proportional to the volume factor fv which is determined by the saturated permeability of the magnetic core as well as the packing factor. The magnetic volume factor as a function of the packing factor for two values of Jlsat is shown in Fig. 3. As n be seen, the required volume of magnetic material increases rapidly as the packing factor decreases. In addition, volume factor increases as the value of Jlsat increases. Fortunately, the value of Jlsat is usually about 1.1 for a magnetic switch having a drive field in excess of 5 ka-t/m [5]. The value of fp is determined by the core winding dimensions. In order to achieve the most efficient magnetic switch design, the value of fp must be maximized. The packing factor of the magnetic switch shown in Fig. 2 is given by [ 1 2] The volt-second product of the magnetic switch is defined as where!18 = flux swing in the magnetic core. [6] 736 where x = winding margin h = core height. In a perfect magnetic switch, the stacking factor is equal to one the windings rest directly on the core. As a result, x=o, f 5 = fp = 1 fv = 1. In the less ideal se, the core has a stacking factor of less than one, but the winding continues to rest directly

3.0 2.8... 0 2.6 u 2.4 u. 2.2 G) E 2.0 ::s 0 1.8 > 1.6 () ;; G) 1.4 1: C) 1.2 :!l 1.0 0.4 0.5 0.6 0.7 0.8 0.9 Packing Factor Fig. 3. Magnetic volume factor graphed as a function of the switch packing factor for two values of saturated permeability. of the core (f 5 < 1 x = 0) so that f 5 = fp < 1 fv > 1. Finally, for the se when the winding is spaced off the core (f 5 <1 x>o), fp < f 5 resulting in a further increase of fv. Design of a Practjl PFL Discharge Magnetic Switch Unfortunately, practil considerations such as a limited choice of core heights, electric field stress limits, or a preferred core aspect ratio may use the design of the output switch to deviate from a minimum magnetic volume geometry as specified by Eq. [11) [12). For example, when it is required to maintain spacing between the winding the core (x>o), fp is maximized by making Ac as large as possible. This implies a single-turn switch design. However, when a single-turn switch is designed for low pulse energies, the ratio of the core height to mean diameter of the core n be very large. Construction of such a switch may be impractil due to electric field stress limitations. A multiple turn switch, on the other h, may provide a more favorable aspect ratio at the cost of increasing the amount of magnetic volume in the core. As a result, the efficiency of the output switch is decreased. When a switch design is initiated, the required values of Lsat <VI> are known. The selection of a magnetic material for the switch core also specifies ~B f 5. Finally, physil constraints may determine other parameters such as the height h inside radius ri of the core. If the effect of the winding margin is initially neglected, then the six parameters Lsat <VI>, ~B. f 5, h ri specify a core design completely. The parameters of this initial core design n then be iterated to yield a practil magnetic core design. The value a Lsat of the magnetic switch n be rewritten using Eq. [4) as 2 Lsat = llo [f 8 %at - 1) + 1] N 2 Am 7t fs(ri + 2 ~h) 1.0 [ 1 3) Eq. [13) n be rewritten to yield a quadratic expression for N as follows llo <YI> [fs4lsat - 1) + 1] 2...;;.._..,..-..;.,...;;,=;...-=--- N - r 1 N 2 7tfs lsat ~ <Vt> 0 [14) 2 ~B h f 8 From Eq. [14), N therefore Am n be lculated for various values of ri h. As a result, fv n be expressed in terms of either ri or h as well as the winding margin. The design of a PFL discharge switch n be best illustrated by solving Eq. [11) Eq. [14) for a specific applition. For example, it is desired to drive a load with a 80 J, 125 kv, 100 ns FWHM pulse. The desired risetime of the drive pulse is 15 ns. The design requirements for the PFL discharge magnetic switch are listed in Table 1. IabW...l. PFL Discharge Magnetic Switch Design Requirements for the 80 J, 125 kv, 100 ns FWHM Applition Electril Reguirements PFL Charge Voltage PFL Energy PFL Charge Time Output Pulsewidth (FWHM) Output Pulse Risetime Load Impedance Physil ReQuirements Approximate Inside Radius of Core Winding Margin 250 kv 80 J 300 ns 100 ns 15 ns 19.5 ohm 10 em 0.40 em For this design, Metglas 2605 SC amorphous material has been selected for the core. This material has a ~B of about 3 T. For a material thickness of 25.4 J.Lm, the stacking factor is approximately 0.80. The lculated gain of the desired discharge switch as defined in Eq. [11) is 7.74 the required value of Lsat is 117 nh. For an ideal switch design (J.lsat = 1.0, f 5 =1 x = 0), the required Volm as lculated by Eq. [11) is 1.67 x 1o-3m 3. However when f 5 = 0.8, the required magnetic volume is 2.25 x 1 o 3 m3, an increase in magnetic volume of 35%. This is the minimal magnetic volume n only be achieved when the winding margin is zero. For non-zero values of x, additional magnetic material is required to achieve the necessary saturated inductance. The least amount of additional material is required when N = 1. Under these conditions, Am=1.25 x 10-2M2, Ac=1.56 x 10 2 m2 <r>=2.8 x 10-2m. As a result, the core must be long have a thin build. It may be desired to decrease the length of the core by increasing the number of turns on the core. Keeping with the desire to maintain an inside radius of about 10 em as listed in Table 1, the required number of turns n be lculated for various values of h using Eq. [14). Rounded values of N c.an then be used to lculate the required magnetic crosssectional area of the core. Since the stacking factor is specified, the volume factor n be graphed as a function of x for various values of N as shown in Fig. 4. When the winding margin is zero, fv for each value of N is 1.35. This indites that 35% more magnetic material is required beuse the core has a stacking factor of 0.80 as opposed to an ideal core which has a stacking factor of unity. Displacing the winding from the surface of the core is accompanied by an increase in the magnetic volume factor. The increased magnetic volume that accompanies an increase in x is a result of additional magnetic path length which is required to compensate for the increase in winding area. For the switch requirements as listed in Table 1, it is observed from Fig. 4 that a four-turn switch design minimizes the magnetic volume of the core at large values of x. However, at large values of x (x = 0.40 em), the volume factor increases 45% resulting in a fv of about 1.6. 737

Cl) E.2 0... > 0.!:! 0.. CI)I.L c 01 :E 2.6 2.4 2.2 2.0 1.8 1.6 1.4 1.2 0.0 0.10 0.20 0.30 0.40 Winding Margin (em) Fig. 4. Magnetic Volume factor graphed as a function of winding margin (x) for various values of h N. Summary 0.50 The requirements of a PFL magnetic discharge switch have been discussed. An expression for the minimum magnetic volume of such a switch has been derived. In addition, factors which influence the required magnetic volume hence the efficiency of a practil PFL discharge switch have been discussed. It was shown that the required magnetic volume is signifintly influenced by the stacking factor of the magnetic core as well as the packing factor of the switch. Specifilly, a non-zero winding margin n signifintly increase the amount of magnetic material required for a switch resulting in a less efficient discharge switch. However, the required increase in magnetic volume n be minimized by the design of a core with the appropriate dimensions. References [ 1 ] D. L. Birx, E. Cook, S. Hawkins, S. Poor, L. Reginato, J. Schmidt M. Smith, "Magnetic Switching," Conference Record of the Fourth IEEE Pulsed power Conference, Albuquerque, N.M., June 1983, pp. 231-235. [ 2] B. J. Briggs, "Induction Accelerators Free-Electron Lasers at LLNL,'' Lawrence Livermore National Laboratory Report No. UCID 21639, February 1989. [ 3] D. L. Birx, E. Cook, S. Hawkins, S. Poor, L. Reginato, J. Schmidt M. Smith, "Magnetic Switching, Final Chapter, Book One: The ATA Upgrade Prototype,'' Lawrence Livermore National Laboratory Report No. UCRL-89128, March 1983. [ 4] W. C. Turner, D. M. Barrett S. E. Sampayan, "Critil Systems Issues Modeling Requirements - The Problem of Beam Energy Sweep in an Electron Linear Accelerator, "Workshop Proceedings of the International Maanetic Pulse Compression Workshop, Tahoe City, CA, February 1990. [ 5] D. L. Birx, E. J. Lauer, L. L. Reginato, J. Schmidt M. W. Smith, "Basic Principles Governing the Design of Magnetic Switches," Lawrence Livermore National Laboratory Report No. UCID-18831, November 1980. 738