Your SlideShare is downloading. ×
Abstract
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Abstract

9,336
views

Published on

Published in: Technology, Business

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
9,336
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
68
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide

Transcript

  • 1. 18th International Pulsed Power Conference Conference Record - Abstracts June 19 -23, 2011 Hyatt Regency, McCormick Place Chicago, IL Sponsored by the Pulsed Power Science and Technology Committee of the IEEE Nuclear and Plasma Science Society PPC 2011 Conference Website: ppc.missouri.edu 2
  • 2. Copyright and Reprint Permission: Abstracting is permitted with credit to source.Libraries are permitted to photocopy beyond the limit of U.S. copyright law forprivate use of patrons those articles in this volume that carry a code at the bottom ofthe first page, provided the per-copy fee indicated in the code is paid throughCopyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923.For other copying, reprint or republication permission, write to IEEE CopyrightsManager, IEEE Operations Center, 445 Hoes Lane, P.O. Box 1331, Piscataway, NJ08855-1331. All rights reserved.IEEE Catalog Number: CFP11PPC-ARTISBN: 978-1-4577-0631-8Additional copies of this Conference Record are available from:IEEE Operations Center445 Hoes LaneP.O. Box 1331Piscataway, NJ 08855-13311-800-678-IEEEAll Rights Reserved. Copyright © 2011 by Institute of Electrical and ElectronicsEngineers, Inc.
  • 3. Table of ContentsWelcome ................................................................................................................................................................. 7PPC2011 Committee Chairs and Staff................................................................................................................ 8PPC2011 Technical Program Committee ......................................................................................................... 10Standing Pulsed Power Science and Technology Committee ......................................................................... 11PPC2011 Student Paper and Travel Grants Awards Chairs.......................................................................... 13PPC2011 Oral Session Chairs ............................................................................................................................ 14PPC2011 Poster Session Chairs ......................................................................................................................... 15PPC2011 Conference Exhibitor Logos.............................................................................................................. 17PPC2011 Conference Sponsor Logos ................................................................................................................ 18PPC2011 Sponsors and Supporters ................................................................................................................... 19PPC2011 Exhibitors ............................................................................................................................................ 1919th International Pulsed Power Conference................................................................................................... 20Previous Pulsed Power Conferences ................................................................................................................. 21About Chicago ..................................................................................................................................................... 22Registration Schedule ......................................................................................................................................... 22Exhibit Times ...................................................................................................................................................... 22Conference Format ............................................................................................................................................. 22Student Travel Grants ........................................................................................................................................ 23IEEE Arthur H. Guenther Pulsed Power Student Award .............................................................................. 23Registration ......................................................................................................................................................... 24Publications ......................................................................................................................................................... 24Companion Program .......................................................................................................................................... 25Social Events ........................................................................................................................................................ 27PPC2011 Schedule-At-A-Glance ....................................................................................................................... 28Plenary Speaker, Monday, June 20 ................................................................................................................... 33 4
  • 4. Plenary Speaker, Tuesday, June 21 .................................................................................................................. 34Marx Award ........................................................................................................................................................ 35Haas Award Winner, Thursday, June 23 ......................................................................................................... 36PPC2011 Abstracts ............................................................................................................................................. 37PL1: Plenary 1 ..................................................................................................................................................... 371A: Explosive and Compact Pulsed Power I .................................................................................................... 371B: Microwaves I: Microwave and RF Sources and Antennae ...................................................................... 391C: Components I: Insulation and Dielectric Breakdown .............................................................................. 411P: Components posters I: Insulation and Breakdown, Transmission Lines and Diagnostics ................... 441P: Microwaves posters I: Sources and Antennae, Slow Wave Devices, Systems ........................................ 541P: Applications posters I: Fusion, EM, Beam & Lasers ................................................................................ 601P: Explosive and Compact Pulsed Power posters .......................................................................................... 612A: Explosive and Compact Pulsed Power II .................................................................................................. 692B: Microwaves II: Microwave and RF Sources, Antennae and Systems..................................................... 702C: Pulsed Power Systems I: Generators and Networks ................................................................................ 73PL2: Plenary 2 ..................................................................................................................................................... 753A: Accelerators and Beams I: LTDs and High Current Accelerators ......................................................... 753B: Microwaves III: High Power Microwave Devices..................................................................................... 783C: Components II: High Energy Density Storage, Transmission Lines and Diagnostics .......................... 812P: Components posters II: High Energy Density Storage, Opening and Closing Switches ....................... 832P: Microwaves posters II: High Power Microwaves ..................................................................................... 902P: Accelerators and Beams posters ................................................................................................................. 932P: Pulsed Power Systems posters I: Electromagnetic Launch, Generators and Networks, and Lasers. 1004A: Radiation Sources I: Z and X-pinches and Lasers ................................................................................. 1064B: Applications I: Fusion, EM, Beam, Laser and Space Applications ...................................................... 1084C: Components III: Arc Discharge Switching ............................................................................................. 111PL3: Marx Award Winner............................................................................................................................... 112 5
  • 5. 5A: Accelerators and Beams II: High Energy Accelerators, Particle Beams and Free Electron Lasers . 1125B: Applications II: General Applications ..................................................................................................... 1155C: Components IV: Solid State Switching .................................................................................................... 1183P: Radiation Sources posters ......................................................................................................................... 1213P: Applications posters II: Medical, Biological, Environmental & General ............................................. 1243P: Pulsed Power Systems posters II: Repetitive and Single Shot Systems ................................................ 1383P: Power Electronics posters.......................................................................................................................... 1416A: Radiation Sources II: High Power Diodes ............................................................................................... 1486B: Power Electronics I: Power Electronics and Prime Power .................................................................... 1516C: Pulsed Power Systems II: Electromagnetic Launch and Lasers ........................................................... 153PL4: Hass Award Winner ................................................................................................................................ 1547A: Pulsed Power Systems III: Repetitive and Single Shot Systems............................................................ 1557B: Applications III: Medical, Biological and Environmental Applications .............................................. 1577C: Power Electronics II: Modulators and Power Supplies ......................................................................... 160 6
  • 6. WelcomeI am pleased to welcome you to the 18th International IEEE Pulsed Power Conference held in Chicago IL,June 19-23, 2011. The conference is organized by the Pulsed Power Science and Technology StandingCommittee of the IEEE Nuclear and Plasma Sciences Society and is held biennially and is the primary forumfor the interchange of information on pulsed power science and technology. The conference proceedings alsoserve as the archival source of technology and science for papers published in the field of pulsed power. Thefield of pulsed power is international in scope. This year approximately 60% of the presentations are from theinternational community illustrating the growing significance of the technology and science in the internationalcommunity.The conference venue this year is the Hyatt Regency McCormick Place, one of the largest conference centers inthe Midwest. The conference is also co-located with the International Conference on Plasma Science (ICOPS)and the Symposium on Fusion Engineering (SOFE). The conference center is centrally located near downtownChicago. Chicago is one of the cultural centers of the Midwest with centrally located museums, restaurants, andentertainment. It is famous for the architecture and skylines, which I invite you to view at sunset. I also wouldsuggest that you explore the museums and architecture, as well as the Frank Lloyd Wright museum whilevisiting Chicago. Chicago is also home to two national laboratories and we have tours of these laboratoriesscheduled for many of the attendees this year.The conference will begin on June 19th, with a reception at the Hyatt Regency McCormick. On Tuesday June21st, a cruise on Lake Michigan is scheduled with 4 hours to view the impressive architecture of Chicago. OnWednesday evening, the Marx and Haas Awardees will receive their awards at the banquet which will be heldthis year at the Field Museum. In addition to the great venue this year, the conference has received a recordnumber of scheduled presentations. We received 480 abstracts this year, with 180 oral talks scheduled and 300poster presentations planned. These presentations cover a wide range of topic areas in the pulsed powercommunity including dielectrics, energy storage, charged particle beam sources, high voltage switches,components, and much more. These technologies are the support base for pulsed power and its applications.The conference will begin on Sunday June 19th, with an early reception and check-in. The technical sessionswill follow on Monday morning with a plenary session. Two plenary sessions are planned for the conference inaddition to the Marx and Haas award talks. Following the plenary talks, the oral sessions are scheduled for themorning sessions, with both poster and oral presentations scheduled for the afternoon sessions.In large part, the Pulsed Power Conference is supported by government, national laboratories, and corporatesponsors, as well as the participation of exhibitors. These sponsorships and exhibitors provide financial supportfor the conference. In addition to their financial support, state of the art components and technology, as well asservices, are on display which support the field of pulsed power. The attendees are encouraged to speak with theexhibitors and express their appreciation for their support of the conference sponsorships.It has been a privilege to serve as Chair of the 18th International IEEE Pulsed Power Conference. Thank you foryour attendance and technical contributions this year. I hope that you find the conference professionallyrewarding and would welcome your feedback on the conference. For more information on the conference Iencourage you to visit the website at http://ppc.missouri.edu.Randy Curry, PhDPPC 2011 7
  • 7. PPC2011 Committee Chairs and StaffGeneral Conference Chair Randy Curry University of Missouri – ColumbiaTechnical Program Chair Bryan Oliver Sandia National LaboratoriesConference Treasurer Mark Rader U.S. Army Space and Missile Defense CommandConference Coordinator Keisha Carr IEEE Meeting and Conference ManagementSponsor Chair Thomas Hussey Air Force Office of Scientific ResearchPublication Chair Robert Druce University of Missouri – ColumbiaAwards Chair Ian Smith L-3 Communications, Pulse SciencesPPS&T Student Advocate Andreas Neuber Texas Tech UniversityStudent Travel Grants Chair Roger White L-3 Communications, Pulse SciencesTour Director and National Lab Liaison Craig Burkhart SLAC National Accelerator LaboratoryCompanion Program Co-Chairs Noreen Curry/Janice AshbyVisa Assistance Ken Struve Sandia National Laboratories 8
  • 8. Webmaster Tanys Nelson University of Missouri – ColumbiaAbstract and Paper Submission Bo Yu Brookhaven National LaboratoryGraphic Artist Linda MaconStaff Assistants Janice Ashby, University of Missouri – Columbia Bill Carter, University of Missouri – Columbia Vicki Edwards, University of Missouri – Columbia 9
  • 9. PPC2011 Technical Program CommitteeBryan Oliver (Technical Program Chair)Sandia National LaboratoriesRaymond AllenNaval Research LaboratoryMichael GiesselmannTexas Tech UniversityVictor KantsyrevUniversity of Nevada – RenoDavid ReismanLawrence Livermore National LaboratoryDon ShifflerAir Force Research LaboratoryMark SinclairAWE AldermastonKen StruveSandia National LaboratoriesRoger WhiteL-3 Communications, Pulse Sciences 10
  • 10. Standing Pulsed Power Science and Technology CommitteeJane Lehr (Chair)Sandia National LaboratoriesRaymond AllenNaval Research LaboratoryLarry AltgilbersU.S. Army Space and Missile Defense CommandStephen BayneTexas Tech UniversityGerald CoopersteinNaval Research Laboratory (Retired)Pat CorcoranL-3 Communications, Pulsed SciencesMark CrawfordThe University of Texas at AustinRandy CurryUniversity of Missouri – ColumbiaSteven GitomerLos Alamos National Laboratory (Retired)Susan HeidgerAir Force Research LaboratoryMark HendersonNaval Air Warfare Center – China LakeThomas HusseyAir Force Office of Scientific ResearchWeihua JiangNagaoka University of TechnologyRavi JoshiOld Dominion UniversityJuergen KolbOld Dominion UniversityJohn MaenchenNational Nuclear Security Administration – DOE 11
  • 11. Andreas NeuberTexas Tech UniversityBryan OliverSandia National LaboratoriesFrank PeterkinNaval Surface Warfare Center – DahlgrenMark RaderU.S. Army Space and Missile Defense CommandLuís RedondoLisbon UniversityRobert ReinovskyLos Alamos National LaboratoryEdl SchamilogluUniversity of New MexicoMark SinclairAWE AldermastonIan SmithL-3 Communications, Pulse SciencesLaura TullyLawrence Livermore National LaboratoryPeter TurchiLos Alamos National LaboratoryRoger WhiteL-3 Communications, Pulse Sciences 12
  • 12. PPC2011 Student Paper and Travel Grants Awards ChairsStudent Travel Grants Chair Roger White L-3 Communications, Pulse SciencesArthur Guenther Pulsed Power Student Award Chair Ian Smith L-3 Communications, Pulse SciencesNPSS Student Paper Awards Chair Bryan Oliver Sandia National LaboratoriesNPSS Student Paper Coordinator Andreas Neuber Texas Tech University 13
  • 13. PPC2011 Oral Session ChairsMatthew DomonkosAir Force Research LaboratoryDon ShifflerAir Force Research LaboratorySteven GloverSandia National LaboratoriesGerald KiuttuVariTech ServicesStephen BayneTexas Tech UniversityBrett HuhmanU.S. Naval Research LaboratoryNeal GraneauAWEDale ColemanSandia National LaboratoriesBruce FreemanKtech CorporationVictor KantsyrevUniversity of Nevada – RenoWeihua JiangNagaoka University of TechnologyMark SavageSandia National LaboratoriesFrédéric BayolITHPPBill ReassLos Alamos National LaboratoryDarryl DroemerNational Security TechnologiesRobert CommissoNaval Research Laboratory 14
  • 14. Ivor SmithLoughborough UniversityJess NeriNaval Research LaboratoryFrank HegelerCommonwealth Technologies, Inc.Richard NessNess Engineering, Inc.Ravi JoshiOld Dominion UniversityBucur NovacLoughborough University PPC2011 Poster Session ChairsKen StruveSandia National LaboratoriesSusan HeidgerAir Force Research LaboratoryRoger WhiteL-3 Communications, Pulse SciencesDavid ReismanLawrence Livermore National LaboratoryRobert DruceUniversity of Missouri – ColumbiaMichael MazarakisSandia National LaboratoriesPeter MardahlAir Force Research LaboratoryColin WhyteUniversity of StrathclydeMark SinclairAWE 15
  • 15. Minh NguyenSLAC National Accelerator LaboratoryRonnie ShepherdLawrence Livermore National LaboratoryJennifer ZirnheldUniversity of BuffaloAllen StultsAviation and Missile Research Development and Engineering LaboratoryMichael GiesselmannTexas Tech University 16
  • 16. PPC2011 Conference Exhibitor Logos 17
  • 17. PPC2011 Conference Sponsor Logos Pulsed Power Conference, Inc. 18
  • 18. PPC2011 Sponsors and SupportersAir Force Office of Scientific ResearchDefense Threat Reduction AgencyL-3 Communications, Pulse SciencesLawrence Livermore National LaboratoryNational Nuclear Security AdministrationNational Security Technologies, LLCNaval Research LaboratoryOffice of Naval ResearchPulsed Power Conference, Inc.Sandia National LaboratoriesScience Applications International CorporationStangenes Industries, Inc.Tektronix, Inc.University of Missouri – Columbia PPC2011 ExhibitorsABB Semiconductors, Ltd.Applied Energetics, Inc.CST of America, Inc.Dawonsys Company, Ltd.Dean Technology, Inc.Diversified Technologies, Inc.General Atomics Electronic Systems, Inc.HVR Advanced Power Components, Inc.Ktech CorporationKumamoto UniversityLawrence Livermore National LaboratoryMagnetic Metals CorporationMajor Tool & Machine, Inc.NWLPowerex Inc.SBE, Inc.ScandiNova Systems ABStangenes Industries, Inc.TDK-Lambda Americas Inc.Tech-X CorporationTRS TechnologiesU.S. Army Space and Missile Defense Command/Army Forces Strategic CommandUltraVolt, Inc.University of Missouri – Columbia, College of Engineering 19
  • 19. 19th International Pulsed Power Conference 20
  • 20. Previous Pulsed Power Conferences1976 Lubbock, TX T.R. Burkes, M. Kristiansen1979 Lubbock, TX A.H. Guenther, M. Kristiansen1981 Albuquerque, NM T.H. Martin, A.H. Guenther1983 Albuquerque, NM T.H. Martin, M.F. Rose1985 Arlington, VA M.F. Rose, P.J. Turchi1987 Arlington, VA P.J. Turchi, B.H. Bernstein1989 Monterey, CA B.H. Bernstein, J.P. Shannon1991 San Diego, CA R. White, K. Prestwich1993 Albuquerque, NM K. Prestwich, W. Baker1995 Albuquerque, NM W. Baker, G. Cooperstein1997 Baltimore, MD G. Cooperstein, I. Vitkovitsky1999 Monterey, CA C. Stallings, H. Kirbie2001 Las Vegas, NV R.E. Reinovsky, M.A. Newton2003 Dallas, TX M. Glesselmann, A. Neuber2005 Monterey, CA J.E. Maenchen, E. Schamiloglu2007 Albuquerque, NM E. Schamiloglu, F. Peterkin2009 Washington, DC F. Peterkin, R. Curry 21
  • 21. About ChicagoChicago, the largest city in Illinois, sits on the southwestern shore of Lake Michigan. Jean Baptiste Point duSable became Chicagos first permanent resident in the late 1770s. Incorporated as a city in 1837, Chicagosposition on Lake Michigan naturally became a trading center. Chicago was the birthplace of mail order retail(Sears and Montgomery Ward), the car radio (Motorola), and the TV remote control (Zenith).With more than 86 million visitors a year, Chicago is a tourist hub that includes cultural, entertainment, andsports attractions. Home to the Chicago Cubs and White Sox baseball teams, Bears football team, Blackhawksice hockey team, and Fire soccer team, Chicago is a mecca for sports enthusiasts. The Art Institute, CulturalCenter, numerous examples of Frank Lloyd Wrights architecture, the Lincoln Park Zoo, and over 40 museumsprovide cultural and educational opportunities for everyone. At an elevation of 1,350 feet above ground, theSkydeck Chicagos The Ledge gives a breathtaking view of the city and surrounding area.Registration Schedule Day Date Start Time End TimeSunday 6/19/2011 1:00 pm 7:00 pmMonday 6/20/2011 7:00 am 5:00 pmTuesday 6/21/2011 8:00 am 5:00 pmWednesday 6/22/2011 8:00 am 5:00 pmThursday 6/23/2011 8:00 am 12:00 pmExhibit TimesSunday, June 19, 2011 Exhibit Hall Opens 5:30 pm Reception in Regency Ballroom 5:30 pm to 8:30 pm Exhibit Hall Closes 8:30 pmMonday, June 20, 2011 Exhibit Hall Hours 7:00 am to 5:30 pmTuesday, June 21, 2011 Exhibit Hall Hours 7:00 am to 5:30 pmWednesday, June 22, 2011 Exhibit Hall Hours 7:00 am to 5:30 pmThursday, June 23, 2011 Exhibit Hall Hours 7:00 am to 12:00 pm Conference FormatThe conference will include plenary, oral, and poster sessions. One plenary presentation will be an address bythe 2011 IEEE NPSS Erwin Marx Award recipient; another address will be given by the 2011 IEEE NPSS PeterHaas Pulsed Power Award recipient. 22
  • 22. Oral presentations will include both invited and contributed papers. Invited talks will be 30 minutes andcontributed talks 15 minutes including five minutes for questions. Oral presentations will be delivered using acomputer and LCD projector. The expected applications are Microsoft Powerpoint and Adobe Acrobat (pdffiles). Presentations are to be submitted on a CD or flash memory and will be transferred to the database atregistration. Student Travel GrantsA limited number of travel grants, typically $750, were available to encourage graduate IEEE members toattend PPC-2009. Applicants were required to submit the following information by 15 April 2011.• Copy of submitted abstract• IEEE membership number• A brief summary of the students research contributions, with most recent references if published (limited to 5 conference or journal papers and not more than 1/2 page in length).• A summary of the students educational accomplishments, including awards, recognition, GPA, etc (no more than 1 page cv or resume).• Proposed travel budget to the conference (cost sharing with other students is encouraged) • One letter of recommendation, preferably from the student’s advisor, stating research to be presented. Chair of the Student Travel Grant Committee: Roger White roger.x.white@L-3com.com IEEE Arthur H. Guenther Pulsed Power Student AwardThe IEEE NPSS Pulsed Power Science and Technology Committees Outstanding Pulsed Power Student Awardwas established in 1997. In 2007 this award was renamed the Arthur H. Guenther Pulsed Power Student Awardfollowing the passing of the former Peter Haas award recipient. It is offered annually, but presented bienniallyat the Pulsed Power Conference. This award is designed to encourage student contributions and participation asprincipal or sole authors of papers and to recognize outstanding student contributions in pulsed powerengineering, science or technology.In order to be eligible for an award, the student must fulfill the following three eligibility criteria: 1) The studentmust be the first author of the paper or poster and must have performed the majority of the work. 2) The workmust be original. And 3) The student must receive endorsement from her/his graduate advisor.The evaluation committee will review the eligible abstracts and down-select up to 8 abstracts for presentationbefore the committee at the conference. Complete instructions will be provided to the finalists once theirabstracts have been selected.Selection criteria of presentations will be based on the quality of the work as well as the student’s grasp of thesubject matter and his/her ability to communicate clearly. A $1,000 and a certificate will be granted to thewinner. For information contact: Ian Smith ian.d.smith@L-3com.com 23
  • 23. RegistrationAdvance RegistrationAdvance registration is highly recommended. Register in full (including payment of the registration fee) by1 June 2011 to qualify for the lower advance registration fee.Advance registration can be carried out online at the conference website. The online registration facility wasactivated by January 2011.Registration on-siteThere will be a Registration Desk at the Conference for attendees who have not registered in advance. TheRegistration Desk will be open on Sunday 19 June from 1 pm-7 pm. On Monday 20 June it will open at 7:00 am, andTuesday 21 June through Thursday 23 June, it will open at 8:00 am. PublicationsAll PPC2011 abstracts appear in this Conference booklet only. Authors of papers presented at the conference areexpected to submit manuscripts for inclusion in the published IEEE Proceedings of the 18th International PulsedPower Conference. Manuscripts from oral and poster presentations may be up to six pages in length. Invited authorsmay contribute papers up to eight pages in length, and plenary speakers may submit manuscripts up to ten pages. Inaddition, all authors from PPC2011 are encouraged to consider contributing to a Special Pulsed Power Issue of theIEEE Transactions on Plasma Science (TPS) planned to be published in October 2012. The submission deadline forthis issue is expected to be 30 September 2011. Please note that authors of papers presented at PPC2011 which mightnot be appropriate for the Special Issue are encouraged to submit their manuscripts to a regular issue of theTransactions on Plasma Science. Information about the Transactions and links to the Manuscript Central manuscriptsubmission portal may be found on the TPS website at http://www.ieeetps.org. Any questions about the Transactions,submissions, etc. should be directed to the Editor, Dr. Steven J. Gitomer (sgitomer@aol.com). 24
  • 24. Companion ProgramCompanion Tour - Day 1Chicago is a city of surprises, from the glimmering lakefront to its fantastic architecture. Highlights of the tourinclude: • 4-hour professionally guided tour • Towering buildings of downtown and upscale residences of the Gold Coast • Museum campus; the Shedd Aquarium, Adler Planetarium, Field Museum, and Soldier Field • Michigan Avenue, the "Magnificent Mile" • Harpo Studios (where Oprah Winfrey tapes her show) • The site where the Great Chicago Fire started • Buckingham Fountain • Take-your-breath-away views from the Willis Tower Skydeck (the former Sears Tower) • Lunch in Greektown • Various stops for photosCost$45 per person (includes roundtrip transportation from hotel, admission to Skydeck, and 3-course lunch)Number of Guests: Minimum 35 / Maximum 50Companion Tour - Day 2Journey to the nearby suburban neighborhood of Oak Park, where Frank Lloyd Wright’s legend began.Highlights of this tour include: • Private tour through his home and studio with an expert docent 25
  • 25. • Learn about Wright’s famous “Prairie style” architecture and the meaning behind it • A relaxed guided stroll through the neighborhood’s Historic District which boasts 25 other Wright buildings • Lunch at a favorite neighborhood Oak Park restaurantCost$45 per person (includes roundtrip transportation from hotel, admission to home and studio, and 3-course lunch)Number of Guests: Minimum 35 / Maximum 50 26
  • 26. Social Events Welcome and Exhibitor Reception Sunday, June 19 Regency Ballroom 5:30 – 8:30 pm Odyssey Cruise Tuesday, June 21 Lake Michigan 6:30 – 10:15 pm Conference Banquet Wednesday, June 22 Field Museum 6:30 – 10:15 pm Awards PresentationErwin Marx Award – Patrick A. Corcoran, L-3 Communications, Pulse Sciences Peter Haas Award – Roger White, L-3 Communications, Pulse Sciences Arthur H. Guenther Pulsed Power Student Awards 2010 – Sarita Prasad, University of New Mexico 2011 – Jonathan Foster, Texas Tech University NPSS Student Paper Awards (2) Lab Tours Argonne National Laboratory Thursday, June 23 1:00 – 4:30 pm Fermi National Accelerator Laboratory Thursday, June 23 1:00 – 4:30 pm Breakfast (Conference Registrants Only) 6:45 – 8:00 am Regency Ballroom Breaks (Conference Registrants Only) 8:50 – 9:30 am 3:00 – 3:30 pm Regency Ballroom 27
  • 27. PPC2011 Schedule-At-A-Glance Sunday, June 19, 20111:00 – 7:00 pm Registration5:30 – 8:30 pm Exhibits5:30 – 8:30 pm Welcome Reception Monday, June 20, 20116:45 – 8:00 am Breakfast (registrants only)7:00 – 5:00 pm Conference Registration and Office Hours7:00 – 5:30 pm Exhibits8:00 – 8:50 am Plenary Session 1 M. Lavan, Space and Missile Defense Technical Center “U.S. Army Directed Energy Weapons Technology Programs” Conference Center 12A-D8:50 – 9:30 am Break (registrants only)9:30 – 12:00 pm Oral Session 1A Explosive and Compact Pulsed Power I Conference Center 10A-B9:30 – 12:00 pm Oral Session 1B Microwaves I: Microwave and RF Sources and Antennae Conference Center 10C-D9:30 – 12:00 pm Oral Session 1C Components I: Insulation and Dielectric Breakdown Conference Center 11A-B12:00 – 1:30 pm Lunch Break1:30 – 3:30 pm Poster Session 1P Components I: Insulation and Breakdown, Transmission Lines and Diagnostics Regency Ballroom1:30 – 3:30 pm Poster Session 1P Microwaves I: Sources and Antennae, Slow Wave Devices, Systems Regency Ballroom1:30 – 3:30 pm Poster Session 1P Applications I: Fusion, EM, Beam, and Lasers 28
  • 28. Regency Ballroom1:30 – 3:30 pm Poster Session 1P Explosive and Compact Pulsed Power Regency Ballroom3:00 – 3:30 pm Break (registrants only)3:30 – 5:30 pm Oral Session 2A Explosive and Compact Pulsed Power II Conference Center 10A-B3:30 – 5:30 pm Oral Session 2B Microwaves II: Microwave and RF Sources, Antennae, and Systems Conference Center 10C-D3:30 – 5:30 pm Oral Session 2C Pulsed Power Systems I: Generators and Networks Conference Center 11A-B Tuesday, June 21, 20116:45 – 8:00 am Breakfast (registrants only)8:00 – 5:00 pm Conference Registration and Office Hours7:00 – 5:00 pm Exhibits8:00 – 8:50 am Plenary Session 2 G. Mueller, Karlsruhe Institute of Technology “Status and Recent Progress in Pulsed Power Applications at Karlsruhe Institute of Technology” Conference Center 12A-D8:50 – 9:30 am Break (registrants only)9:30 – 12:00 pm Oral Session 3A Accelerators and Beams I: LTDs and High Current Accelerators Conference Center 10A-B9:30 – 12:00 pm Oral Session 3B Microwaves III: High Power Microwave Devices Conference Center 10C-D9:30 – 12:00 pm Oral Session 3C Components II: High Energy Density Storage, Transmission Lines, and Diagnostics Conference Center 11A-B12:00 – 1:30 pm Lunch Break 29
  • 29. 1:30 – 3:30 pm Poster Session 2P Components II: High Energy Density Storage, Opening and Closing Switches Regency Ballroom1:30 – 3:30 pm Poster Session 2P Microwaves II: High Power Microwaves Regency Ballroom1:30 – 3:30 pm Poster Session 2P Accelerators and Beams Regency Ballroom1:30 – 3:30 pm Poster Session 2P Pulsed Power Systems I: Electromagnetic Launch, Generators and Networks, and Lasers Regency Ballroom3:00 – 3:30 pm Break (registrants only)3:30 – 5:30 pm Oral Session 4A Radiation Sources I: Z and X-Pinches and Lasers Conference Center 10A-B3:30 – 5:30 pm Oral Session 4B Applications I: Fusion, EM, Beam, Laser, and Space Applications Conference Center 10C-D3:30 – 5:30 pm Oral Session 4C Components III: Arc Discharge Switching Conference Center 11A-B Wednesday, June 22, 20116:45 – 8:00 am Breakfast (registrants only)8:00 – 5:00 pm Conference Registration and Office Hours7:00 – 5:00 pm Exhibits8:00 – 8:50 am Plenary Session 3 Marx Award Speaker P. A. Corcoran, L-3Communications, Pulse Sciences “Practical Circuit Models and Simulations using Transmission Lines” Conference Center 12A-D8:50 – 9:30 am Break (registrants only)9:30 – 12:00 pm Oral Session 5A Accelerators and Beams II: High Energy Accelerators, Particle Beams, and Free Electron Lasers 30
  • 30. Conference Center 10A-B9:30 – 12:00 pm Oral Session 5B Applications II: General Applications Conference Center 10C-D9:30 – 12:00 pm Oral Session 5C Components IV: Solid State Switching Conference Center 11A-B12:00 – 1:30 pm Lunch Break1:30 – 3:30 pm Poster Session 3P Radiation Sources Regency Ballroom1:30 – 3:30 pm Poster Session 3P Applications II: Medical, Biological, Environmental, and General Regency Ballroom1:30 – 3:30 pm Poster Session 3P Pulsed Power Systems II: Repetitive and Single Shot Systems Regency Ballroom1:30 – 3:30 pm Poster Session 3P Power Electronics Regency Ballroom3:00 – 3:30 Break (registrants only)3:30 – 5:30 pm Oral Session 6A Radiation Sources II: High Power Diodes Conference Center 10A-B3:30 – 5:30 pm Oral Session 6B Power Electronics I: Power Electronics and Prime Power Conference Center 10C-D3:30 – 5:30 pm Oral Session 6C Pulsed Power Systems II: Electromagnetic Launch and Lasers Conference Center 11A-B Thursday, June 23, 20116:45 – 8:00 am Breakfast (registrants only)8:00 – 12:00 pm Conference Registration and Office Hours7:00 – 12:00 pm Exhibits 31
  • 31. 8:00 – 8:50 am Plenary Session 4 Haas Award Speaker R. White, L-3 Communications, Pulse Sciences “From Coalminer’s Grandson to Peter Haas Award” Conference Center 12A-D8:50 – 9:30 am Break (registrants only)9:30 – 12:00 pm Oral Session 7A Pulsed Power Systems III: Repetitive and Single Shot Systems Conference Center 10A-B9:30 – 12:00 pm Oral Session 7B Applications III: Medical, Biological, and Environmental Applications Conference Center 10C-D9:30 – 12:00 pm Oral Session 7C Power Electronics II: Modulators and Power Supplies Conference Center 11A-B 32
  • 32. Plenary Speaker, Monday, June 20 Dr. Michael J. Lavan is the Director of the Directed Energy & Missile Defense Technology Directorate of the Space and Missile Defense Technical Center (SMDTC) in Huntsville, AL. SMDTC is the research and development element of the U.S. Army Space and Missile Defense Command/Army Forces Strategic Command. He has held this position since November of 2004. He previously served as the Associate Director of the SMDTC for Missile Defense from June 2001 to November 2004 and as Director, Advanced Technology Directorate of the SMDTC from June 1994 to June 2001. From 1986 to 1994 he was the Director of the Directed Energy Weapons (DEW) Directorate, where he was responsible for the commands programs in neutral particle beams, free electron lasers, beamcontrol and propagation, and space demonstrations of beam weapon technology. Before his selection asDirector of DEW, Dr. Lavan was chief of the Ground-Based Laser Division where he was responsible fordefining and initiating the free electron laser and beam control programs which grew into the Ground-BasedLaser Project at White Sands Missile Range, N.M.Dr. Lavan has also directed missile defense technology programs in multiple kill vehicle interceptors,distributed radars, wide bandwidth distributed computing, missile materials and directed energy missile defenseconcept studies.His current responsibilities are focused primarily on Army efforts in high energy solid state laser devices andsystems and on compact high power microwave devices.Dr. Lavan, who received his Ph.D. in physics from the University of Iowa, is a member of the Association ofthe United States Army and the Optical Society of America. He has been a member of the Senior ExecutiveService since 1986. Dr. Lavan has received numerous letters of commendation and awards including theCommander’s Award for Civilian Service, the Achievement Medal for Civilian Service and the rank ofMeritorious Executive in the Senior Executive Service. 33
  • 33. Plenary Speaker, Tuesday, June 21 Georg Mueller was born in 1961. He received the Diploma degree in physics and Ph. D. degree from the University of Karlsruhe, Germany, in 1990 and 1999 respectively. Since 1990, he was with the Research Centre Karlsruhe, where he worked in different research fields including thin film deposition by channel spark, plasma and electron beam diagnostics, development of multi point explosive emission cathodes, transport of large area powerful pulsed electron beams and surface modification by pulsed electron beams (GESA-process). He is author and co-author of more than 200 publications in peer reviewed journals,conference proceedings and two book chapters. He is member of the European Pulse Power Society, theInternational Advisory Committees of the HLMC-, BEAMS- and EAPPC- conferences and member of thecontact expert group of the European Commission on transmutation and the OECD/NEA expert group on HLM.Since 2006 he is Deputy Director and Head of the Pulsed Power Department of the Institute for Pulse Powerand Microwave Technology at Karlsruhe Institute of Technology (KIT). He is responsible for the developmentof pulsed power applications in the field of: • electrodynamic fragmentation of solid dielectric materials, • material surface modification by pulsed electron beams, • electroporation of biological cell membranes by pulsed electric fields and • basic research in bioelectrics. 34
  • 34. Marx Award Winner, Wednesday, June 22 Patrick A. Corcoran is a Senior Scientist at L-3 Pulse Sciences where he has roles in both technical leadership and project management. He was first introduced to pulsed power as a summer hire at Pulse Sciences, Inc. (PSI) in 1983 after an internship as a mechanical engineer at the NASA Ames Research Center a year earlier led him to change his major emphasis to Physics. Mr. Corcoran received his B.A. in Physics from San Francisco State University in 1984 and thereafter commenced full time employment at PSI where he has remained employed through its acquisition by the Titan Corporation and later by the L-3 Communications Corporation.Mr. Corcoran has distinguished himself in the design of large, high power, pulsed machines where his expertisein, and development of, pulse power circuit modeling and simulation have been an essential component. He hasauthored and coauthored over 50 publications and is the coauthor of a patent. He is a member of the IEEE andis presently serving on the NPSS Pulse Power Science and Technology Committee.Mr. Corcoran’s work in pulse power technology development includes system design, pulse compression andpulse forming line design, vacuum power flow and diode design, high voltage and high current componentdesign, empirical characterization and prototyping, design validation, and circuit model and code developmentto support design efforts. He has made a variety of notable and important contributions to the progression ofsuperpower accelerators at Sandia National Laboratories in Albuquerque including Proto 2, Saturn, Z, and ZRefurbishment (ZR) as well as to next generation conceptual designs. He has also made notable contributionsto high voltage radiography and is an authority on Inductive Voltage Adder (IVA) design and operation throughhis work on a progression of machines which have included Hermes 3 and RITS at Sandia, Cygnus at theNevada National Security Site (NNSS), and Hydrus for AWE in the UK. And he has made notablecontributions to other machines including the DARHT 1 and the DARHT 2 injectors at the Los AlamosNational Laboratory (LANL), the AIRIX injector for the CEA in France, the NIKE laser at NRL, and FXR atLawrence Livermore National Laboratory (LLNL).Mr. Corcoran credits his association with his colleagues at PSI for the opportunity to participate in a wide rangeof prominent projects and is particularly grateful to Ian Smith, Phil Spence, Lee Schlitt, and Jim Fockler forpatiently teaching him the art and science of pulse power. 35
  • 35. Haas Award Winner, Thursday, June 23 Roger White was born on 11th January 1939 in Llwynypia, in the mining valleys of South Wales, UK. At 16, he entered a five year Student Apprenticeship program at the Atomic Energy Research Establishment (AERE) at Harwell, England. There he was trained as an electrical engineer while simultaneously attending Oxford Polytechnic where he was awarded Higher National Certificates in both Electrical and Mechanical Engineering. AERE employed him in the Plasma Physics Division after he completed his apprenticeship. For AERE he worked on high voltage switching, first at Harwell and at Culham when it was opened in 1962. In 1964 Roger immigrated to Canada, and spent a year working on satellite systemsfor RCA in Montreal. He then returned to high voltage engineering at Ion Physics in Boston, where he was firstintroduced to nuclear weapons simulators in the form of flash X-ray and electromagnetic pulse (EMP) systems.Roger joined Maxwell Laboratories in San Diego, California, in 1967 and began a 35 year relationship with thatcompany. Roger had the honor to work with many of the original thinkers in the field of Pulsed Power. Thelong list includes Alan Kolb, Richard Fitch, Richard Miller, John Shannon, John Harrison, Bob Hunter and JorgJansen. He made contributions to the Blackjack series of simulators for the Defense Nuclear Agency, and EMPgenerators for the US Department of Defense and foreign governments. This lead to field installation andcommissioning of systems such as Casino at NSWC White Oak, Empress II at Little Creek, Virginia, andsystems in France, and in Germany.At the same time Roger managed up to forty people in the Maxwell Engineering Department. This matrixorganization prompted Roger to market and manage programs within the group, as well as to support theengineering needs of the entire company. His last major assignment before Maxwell sold its pulsed powersystems business was to manage its group in Albuquerque and win a large contract at the Air Force ResearchLaboratory.Roger chaired the 1991 IEEE Pulse Power Conference and was co-chairman of the 1994 BEAMS conference,both in San Diego. He has served on the IEEE Pulse Power committee for twenty years.Since the purchase by Titan Corporation in 2001 and Titan’s purchase by L-3 Communications in 2005, Rogerhas directed the operation of the L-3 Pulse Sciences group in San Diego, originally Maxwell’s pulsed powergroup. 36
  • 36. PPC2011 AbstractsPL1: Plenary 1 available previously. This paper will discuss the design and performance of the pulser.Monday, June 20 08:00-08:50 CC 12A-DU.S. Army Directed Energy Weapons Technology Programs 1A-3: 1MJ Compact Pulsed Current Source B. E. Fridman1, B. Baoming Li2, V. A. Belyakov1, R. S. Enikeev1,M. Lavan N. A. Kovrizhnykh1, Y. L. Kryukov1, K. M. Lobanov1, A. G. Roshal1,U.S. Army SMDTC, Huntsville, AL, United States R. A. Serebrov1 1 STC, D.V. Efremov Scientific Research Institute of Electrophysical1A: Explosive and Compact Pulsed Power I Apparatus, St.Petersburg, Russian Federation 2Monday, June 20 09:30-12:00 CC 10A-B National Key Laboratory of Transient Physics, Nanjing University of Science &Technology, Nanjing, P. R. China1A-1:(INVITED) Reducing PFN Marx Generator Size UsingNested Solid Insulation The pulsed current source is intended for research of high-currentR. J. Adler, J. A. Gilbrech, D. New electrical discharges in dense media in laboratory conditions. TheApplied Energetics, Tucson, AZ, United States facility includes two modules of capacitive energy storage, control panel unit and cable collector which connects facility with the load.. Each module is a functionally completed remote-operated capacitiveCurrent state-of-the-art, high voltage, pulsed power systems above 0.5 energy storage 0.5 MJ, 18 kV, built on eight capacitor cells withGW output power generally use Marx or PFN Marx generator designs. reverse-switching dynistor (RSD) switches. The module has volumeThe high voltage insulation of these devices has been traditionally 1.3 m3 and comprises a charger with a high-frequency inverter, abased on either high dielectric strength oil or sulfur hexafluoride (SF6). protection contactor with normally closed high-voltage contacts toThe dielectric strength of the insulating medium determines the neutralize the cell charge, trigger units for triggering of RSD switches,minimum size of the tank enclosure of the system since it determines control and diagnostics apparatus with a programmable logic controller.the maximum voltage standoff between the fully erected Marx output The semiconductor switches in the capacitor cells are triggered by lightvoltage and the tank wall. We have applied our "Nested High Voltage" pulses transmitted from a control panel unit via fiber-optical cables.insulation technology to the problem of PFN pulse generator design. Information exchange between the modules control and diagnosticsThis technique uses solid insulation with field grading foils that system and the control panel unit is also realized through fiber-opticalminimized the distance from the fully erected voltage and the tank wall. cables. The facility is designed for operation in the programmableThis report describes work performed in adapting this technology to discharge mode, when semiconductor switches in the capacitor cells arebuilding a PFN pulse generator. We believe that this technology will switched on in the specified time sequence according to the presetallow size reductions relative to other insulation technologies. The program. A maximal current pulse amplitude of 800 kA at the moduletechnology also allows integration of the PFN charge power supply into output is attained during simultaneous discharge of all cells in thethe PFN. Data will be presented on the performance of a 130 kV short-circuit mode. The wave-front of the current pulse is 150 mcs.version of this type of device capable of up to 5 kA output. Wedemonstrated that making a "flat" pulse is feasible in this geometry,and we determined that there is an optimal grading technique for this 1A-4:Development of High Power Long Longevity GaAstype of design. A nominal 130 kV, 3-stage pulse was successfully Photoconductive Semiconductor Switches for Compactproduced using our design techniques. Both high gradient insulation Pulsed Powerand pulse generation have been demonstrated. The systems will be J. Yuan1, W. Xie1, H. Li1, H. Liu1, J. Liu1, X. Wang2, W. Jiang2useful in the 50 kV - 2 MV range with impedances in the 20 - 200 ohm 1 Institute of Fluid Physics, China Academy of Engineering Physics,range. This work was supported by AFRL under contract FA9451-10- Mianyang, ChinaM-0093 2 Department of Electrical Engineering, Tsinghua University, Beijing, China1A-2: Pulser for High Altitude Jet Engine Re-IgnitionI. S. Roth, M. A. Kempkes, M. P. J. Gaudreau, P. VerPlanck High power photoconductive semiconductor switches (PCSSs) areDiversified Technologies, Inc., Bedford, MA, United States considered a promising device for compact, repetitive pulsed power generation due to their advantages over other switches, such as fast response time, negligible time jitter, precise synchronization, highThe maximum altitude of a jet aircraft is limited by ignition stability. repetition rate, and optical electrical isolation. Therefore, in the lastAt high altitude the engine may flame out because existing igniters several years, high power long longevity PCSSs have been investigatedcannot couple enough energy into the low-pressure air-fuel mixture. To at Institute of Fluid Physics, CAEP. Photoconductivity tests have beenreignite the engine, the aircraft must descend to a lower altitude where performed at different bias voltages, laser wavelengths and energies.the air pressure is higher. To address this issue Diversified Several packaging methods for lateral and bulk PCSSs wereTechnologies, Inc. designed and built a fast ignition pulser under an investigated and maximum voltage of 60 kV, maximum current of 2.4SBIR grant. It was delivered to the Air Force Research Laboratory for kA and maximum electric field of 127 kV/cm have been achieved.testing to determine the optimum parameters for high-altitude jet Lifetime of the GaAs PCSS is critical for repetitive operation. By usingignition. The pulser design is based on the following considerations: • laser spots to trigger the opposite side of electrodes for lateral PCSS,To best produce ignition, the pulse rate should be fast, with low energy damages to ohmic contact were alleviated. By using laser spots withper pulse. • The risetime should be also be fast to maximize the energy different profile to trigger the switch, proper illumination pattern wascoupled into the fuel mixture. • The pulser should be capable of chosen. A lifetime of more than 100,000 shots for GaAs PCSSs hasoperating into a short circuit. Based on magnetic pulse compression, been achieved under the experimental conditions of 20 kV in the biasthe pulser produces a 27 kV, 37 mJ pulse with a risetime of 17 ns and a voltage, 1 kA in the photocurrent and 20 Hz in the repetition rate. Thepulse rate of 8 kHz. A pulser with these characteristics had not been 37
  • 37. capability to repetitively operate at 1 kHz will be tested with the proper requirements of these applications are repetitive operation and fasttrigger laser. risetime voltage. Recently, semiconductor power device technology has improved the performance of fast high voltage switching and low switching loss. In particular, insulated gate bipolar transistor (IGBT) is1A-5: Development of Solid-State Pulse Forming Lines highly efficient semiconductor switching device. However, the IGBTJ. Liu, H. Li, H. Liu, J. Yuan, W. Xie switch is still not sufficient to drive the pulse laser and the pulseInstitute of Fluid Physics, China Academy of Engineering Physics, ozonizer itself. In practical systems, semiconductor switches are usedMianyang, China with the assistance of magnetic switches. We have studied and developed high repetition rate small size pulsed power generator forThree types of solid-state pulse forming lines (PFL), ferroelectric generation of discharge plasma. This generator consists of an IGBTceramic PFL, glass-ceramic PFL, and multiferroics PFL, have been switch circuit, a step-up pulse transformer and multi-stage magneticdeveloped. Because of high permittivity of ferroelectric, longer pulse compression circuit (MPC) at the last magnetic pulseduration square pulse can be realized by ferroelectric ceramic PFL compression stage, has a coaxial configuration. This fast-rising pulsedwithout increasing PFL size. With gas spark as closing switch, as power generator using a coaxial MPC has been developed for fast riseferroelectric ceramic PFL be charged up to 30 kV, square waveform time output current. We use two kinds of magnetic core materials, apulses with load peak current up to 7 kA, pulse duration about 140 ns Co-based amorphous alloy and a nanocrystalline Fe-based softwas obtained. Because of high permittivity and high breakdown electric magnetic alloy. Two kinds of magnetic cores were investigated in thestrength, glass-ceramic PFL is one of the potential candidates for coaxial MPC to evaluate loss of the magnetic cores, leakage current ofconstructing compact pulsed power equipment. Output performances of saturable inductors, and circuit inductance. The pulsed power generatorglass-ceramic PFL with photoconductive semiconductor switches produced a pulsed-high-current of 3.7 kA with a rise time of 7 ns at a(PCSSs) as closing switch were investigated. The insulation properties repetition rate of 1000 pulses per second (pps). This generator is able toof ferroelectric ceramic PFL and glass-ceramic PFL were explored and generate an output voltage of about 20kV with voltage rise time of lesscompared. Novel PFL based on not only ferroelectric but also than 10 ns. We did the operation test and generate the streamerferromagnetic, viz. multiferroic, was proposed. Multiferroic PFL with discharge with 1000 pps.relative permittivity coefficient εr approaching 2000 and effectiverelative permeability coefficient μr>1.4 was demonstrated. It was also 1A-8:Miniature Pulsed Power Generator Using a Magneticfound that the output performances of the multiferroic PFL are differentbetween different peripheral magnetic circuits. Pulse Compression Circuit Y. Ito, K. Kouno, T. Sakamoto, T. Sakugawa, H. Akiyama Graduate School of Science and Technology, Kumamoto University,1A-6:Study of Nanosecond Pulsed Power Generator Based Kumamoto City, Japanon Epi-Si Drift-Step Recovery DiodeL. M. Merensky1, A. F. Kardo-Sysoev2, D. Shmilovitz3, Y. Sharabani1, Pulsed power technology aiming industrial applications, such as theI. Shafir1, A. Sher1, A. S. Kesar11 ozone production and the exhaust gas treatment using atmospheric Soreq NRC, Yavne, Israel discharge plasmas, has been progressed, and a compact and portable2 Ioffe PTI, St. Petersburg, Russia pulsed power generator (PPG) has been developed in recent years. In3 Tel Aviv University, Tel Aviv, Israel terms of energy transfer efficiency, pulse compression rate and reliability, a magnetic pulse compression circuit (MPC) is suitable as aEpi-silicon powerful drift-step recovery diodes (DSRDs) are developed circuit system of the PPG for many of industrial applications. However,at Soreq NRC. A circuit which enables efficiency measurement is a low pulse energy and extremely compact pulsed power generatorpresented. The circuit contains two compression stages. The first stage using the MPC has not been reported. This paper reports onis based on a MOSFET in which, when closed, it charges a coil while miniaturization of a pulsed power generator (PPG) with the MPC. Inpumping a DSRD through a resonant circuit. When the transistor opens the point of view of volume and performance of the miniature PPG,the coil energy and the resonant circuit jointly reverse the DSRD discrete ceramic capacitors were used and the core materials of Fe-current. Special bias circuit provides easy adjustment of balance of based nanocrystalline with a high magnetic flux swing and Co-basedDSRD pumped and removed charges. The DSRD opening results in a amorphous with a high permeability were compared. Losses ofnanosecond scale pulse into the load. The adjustment of the balance is components are sensitive to increase the energy transfer efficiency. Thevery important for high efficiency of the circuit. The circuit is developed PPG is a double staged MPC with 0.2 J/pulse. The Fe-basedembedded with current and voltage probes in order to study the losses nanocrystalline at the first stage of the MPC and the Co-basedand calculate the efficiency. Temporal resolution of the current and amorphous cores at the second stage were selected. The parallelvoltage probes are investigated as well. Experimental results will be connection of low resistance capacitors was valid to increase the energypresented. transfer efficiency. As the result, approximately 13 kV pulse with a risetime less than 30 ns was obtained. The size of PPG is a B5 (260×200×70 mm) size.1A-7:Fast Rise Time Pulsed Power Generator Using IGBTsand Coaxial MPCT. Sakugawa1, S. Ueda1, H. Akiyama1, K. Suematsu2, A. Kouda2, Design and Optimization Techniques for the 1A-9:M. Watanabe2 Generation of Intense, Ultrafast Pulses with Nonlinear1 Graduate School of Science and Technology, Kumamoto University, Transmission LinesKumamoto, Japan J. M. Sanders, A. Kuthi, M. A. Gundersen2 Suematsu Electronics Co. Ltd., Yatsushiro, Kumamoto, Japan Ming Hsieh Department of Electrical Engineering, VSoE, University of Southern California, Los Angeles, CA, United StatesAll solid-state pulsed power generators, which are operated with longlifetime, compactness and high reliability, have been developed to be Solid-state systems that produce kW or MW pulses on nanosecondused for industrial applications, such as excimer laser, high energy timescales typically consist of multiple stages composed of nonlineardensity plasma (EUV sources) and pulse ozonizer. Moreover, components that compress and switch energy. The transient and 38
  • 38. nonlinear nature of these stages complicate the analysis of system [1] A.M. Belyantsev, et. al. Tech. Phys., 40, 820-6 (1995).dynamics to an extent that closed form expressions describing metrics [2] A.M. Belyantsev and A.B. Kozyrev, Tech. Phys., 43, 80-85, (1998).such as output impedance, energy compression ratio, and efficiency in [3] N. Seddon, et. al. 2007 IEEE Pulsed Power Conf. Proc., p. 678.terms of circuit components or constitutive material parameters are [4] A.M. Belyantsev and A.B. Kozyrev, Tech. Phys., 47, 1477-1480,difficult if not impossible to obtain. To address these challenges, a (2002).number of techniques, some closed-form and some numerical, have [5] N. Seddon and T. Bearpark, Science, 302, 1537-40, (2003).been developed to elucidate system dynamics and provide design rules. [6] A.B. Kozyrev and D.W. van der Weide, Phys. Rev. Lett., 94,These techniques will be discussed in relation to a solid-state, 203902-4, (2005).multistage pulse generator that uses magnetic pulse compression anddiode opening switches to generate high voltage, nanosecond pulses. Pspice Simulations of Nonlinear Transmission Lines 1B-2:The discussion will include methods to optimize diode pumping, tooptimize energy transfer through a saturable core, and to compensate Based on Ferroelectric Dielectricsfor parasitic reactances. This pulse generator has most recently been P. Norgard, R. D. Curryused for research into the design of nonlinear transmission lines for Dept. of Electrical & Computer Engineering, University of Missouri,risetime sharpening and soliton generation. Matching the system’s Columbia, MO, United Statesoutput impedance to nonlinear transmission lines, both dispersive andnondispersive, will be discussed, and measured results from the The search for efficient, compact, high-power conversion of electricalnonlinear transmission lines will be reported. impulses into high power microwave energy is driving a modern rebirth of nonlinear transmission line studies. A nonlinear transmission line1B: Microwaves I: Microwave and RF Sources and Antennae utilizes the nonlinear permittivities and permeabilities, εr and µr for dielectrics and magnetics, respectively, to transform a fast-rising pulseMonday, June 20 09:30-12:00 CC 10C-D front into microwave energy. The amplitude-dependent phase velocity of nonlinear materials enables an input pulse of relatively slow rise-1B-1:(INVITED) Non-Resonant Parametric Amplification time to be converted into a pulse exhibiting an extremely fast rise-timeand Higher Harmonic Generation of High Power through the formation of an electromagnetic shockwave. A majority ofMicrowave Signals in Nonlinear Transmission Lines research today is conducted on nonlinear magnetic materials, namelyA. B. Kozyrev ferrites, with an impressive degree of success. By contrast, theDepartment of Electrical and Computer Engineering, University of University of Missouri – Coumbia (UMC) has focused upon nonlinearWisconsin-Madison, Madison, WI, United States transmission lines comprised of nonlinear dielectric materials in order to take advantage of some unique materials and techniques recently developed at UMC. Pulsed power generated from the nonlinearRecently, the possibility of the direct and efficient conversion of a transmission lines indicate 100 MW-gigawatt generation andvideo pulse into a RF pulse during its propagation along a nonlinear simultaneous compression of pulses.transmission line (NLTL) with special dispersion has been A simulation study of nonlinear transmission line behavior underdemonstrated in several publications [1-3]. different parametric constraints is reported. The simulations wereThis publication presents a novel technique for high power microwave conducted in PSpice and utilize a custom-developed nonlinearsignal generation, frequency up-conversion and amplification which are dielectric model based upon laboratory data provided by commercialenabled by non-resonant parametric interactions in NLTLs. Frequency manufacturers of ferroelectric material. The model implemented in theup-conversion, harmonic generation and amplification occur when a study is discussed in detail and the compromises made to balancehigh power RF wave catches up to a receding electromagnetic shock model accuracy with numerical convergence are discussed. A one-wave front that is propagating along a NLTL. Under certain conditions, dimensional model of a coaxial transmission line filled with athis may give rise to several reflected waves having frequencies higher ferroelectric material was implemented and the results are discussed.then the incident wave. Results of simulations have shown that thereflected pulse has a higher carrier frequency (inverse Doppler up-conversion) and increased power, gained from the co-propagating 1B-3:Pulsed High Power RF Generation from Nonlinearshock wave (non-resonant parametric amplification). In the considered Dielectric Ladder Networks - Performance Limitscase, the ratio of the carrier frequency of the output RF pulse to the P. W. Smithcarrier frequency of the input pulse is approximately 1.9 and the power Pembroke College, Oxford University, Oxford, United Kingdomof the output pulse is 2.3 times larger than the power in the incidentwave. It has become apparent from an extensive study of the generation ofFurthermore, in addition to the Doppler shifter carrier, the Doppler bursts of soliton-like pulse trains on nonlinear dielectric ladders thatshifter higher harmonics can be generated at the reflection from the there are fundamental physical problems that will limit the applicationshock wave front as well. Moreover, the power present in the higher of this technology. The purpose of this paper is to review these issuesharmonics can be larger than the power in the Doppler shifted and suggest areas, in particular material and/or component development,fundamental (carrier) wave. where research may lead to significant performance improvements.It worth be mentioned also that the inverse Doppler effect recently High power operation has been confined to ladder networks or lumpedpredicted, observed and explained in [4-6] is a particular case of element transmission lines which use ferroelectric ceramics dielectricsparametric interaction under investigation. In contrast to [4-6], in our such as barium or strontium titanate ceramics as the nonlinear dielectriccase both the incident and the reflected waves are forward waves. elements. These dielectrics are commonly used in high voltage ceramicFurthermore, although the systems described in [4-6] provide frequency capacitors and an obvious radio frequency power limitation resultsup-conversion, they don’t allow for RF power amplification. from the dielectric breakdown strength of these materials. SuggestionsThe up-converted signal generated by one NLTL can be applied to the by which this issue may be circumvented will be discussed.input of another NLTL, and the signal from this NLTL to another, and The difficulty of coupling the RF pulse from the nonlinear ladders orso on. As a result, the RF carrier frequency and the power are boosted lines efficiently into a linear resistive load is well known and is causedat each stage. This novel approach allows for boosting the power primarily, by the voltage dependency of the ladder/line impedance. A(energy) of the RF pulse from multiple relatively low-power sources. closer examination of the way that the voltage and current waveforms 39
  • 39. on the line are propagated and also the reflection process at a linear methods and performance of the system. We also compare the results toresistive load explains the structure of the voltage waveforms, seen at predictions made from theoretical models of gyromagnetic interactions.various stages on the line, and also possible solutions to the couplingproblem. Results from experimental lines and new computer modelling 1B-6:Temperature Dependence of Ferrimagnetic Basedtechniques will be presented.Finally attempts to generate radio frequency bursts at frequencies into Nonlinear Transmission Linethe microwave bands have yielded disappointing results. In order to do J.-W. B. Bragg, J. Dickens, A. Neuberthis, lumped element lines have been built which usually comprise a Department of Electrical and Computer Engineering, Texas Techparallel plate transmission line that is periodically loaded with University, Lubbock, TX, United Statesnonlinear dielectric tiles or slabs. This approach has resulted in thegeneration of poorly modulated radio frequency bursts and Ferrite loaded nonlinear transmission lines (NLTLs) are able to act as aconsequently this technology is limited to operating frequencies below high power microwave source, utilizing the nonlinearities present ina few hundred MHz. An explanation of this phenomenon will be ferrimagnetic materials and the excitation of damped gyromagneticpresented that shows that loss in the nonlinear line elements is the most precession at high incident power levels. Ferrimagnetic propertieslikely cause of the problem. This will backed up with results from depend greatly on operating temperatures; therefore, there exists a needexperiments and modelling of ladders with simulated variable loss. 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 Characterization of a Synchronous Wave Non Linear1B-4: to allow internal ferrite temperature uniformity. Experimental temperatures ranged from approximately -50 °C up to 110 °C, which isTransmission Line slightly under the Curie temperature of the loaded ferrites. ThisP. D. Coleman, J. J. Borchardt, J. A. Alexander, J. T. Williams, temperature range allows observation of precession dependence onT. Peters temperature while still maintaining ferrimagnetic properties. Above theDept. 5443, Sandia National Laboratories, Albuquerque, NM, United Curie temperature the loaded ferrites become paramagnetic and loseStates ferrimagnetic properties. The design, testing, and results are detailed for an NLTL measuring 0.3 m length and ferrite inner and outerMany aspects of Non Linear Transmission Lines (NLTL’s) make them diameters of 3 mm and 6 mm respectively. Figures comparing outputattractive sources for High Power Microwave (HPM) systems. This waveforms at different temperatures, output power versus temperature,paper describes a test bed that is being used to investigate synchronous and output frequency versus temperature are shown.wave NLTL operations. In the synchronous wave mode, a highamplitude pulse drives the nonlinear material (ferrites in this case) into Development of High Power Gyromagnetic Nonlinear 1B-7:saturation forming a shock front. Energy from this shock front thencouples to RF waves that have phase velocities matching the shock Transmission Linesvelocity (i.e. synchronous wave). I. V. Romanchenko, V. V. Rostov, A. I. Klimov, I. K. Kurkan,Analytical models and numerical simulations are used in design of the A. V. Guninline and interpretation of results. Shock wave propagation is Institute of High Current Electronics SB RAS, Tomsk, Russiancharacterized and compared to theoretical predictions. Adjustment of Federationthe shock velocity is used to demonstrate the NLTL’s tunability. Thesaturated line left behind the shock front provides a linear dispersive Gyromagnetic nonlinear transmission lines (NLTLs) based on saturatedline for RF wave propagation. Both forward and backward wave modes NiZn ferrites have been shown to be promising candidate forare supported by this dispersion relation, and results from operating in subgigawatt RF sources [1]. This technology transforms energy ofboth modes are presented. Special emphasis is placed on studying the pumping high voltage pulse directly into RF oscillations. According tobackward wave mode which potentially has longer pulse and higher experimental investigation the length of the RF pulse is limited byfrequency capabilities. several pseudo periods with duration of 4-5 ns. The RF peak power is constricted by 0.6-0.7 GW for 3 GW power of incident pulse. The NiZn1B-5: A Novel Solid-State HPM Source Based on a gyromagnetic NLTLs can be used for RF generation in a frequency band from 0.4 to 2.0 GHz. The lower limit is caused by magnetic lossesGyromagnetic NLTL and SOS-Based Pulse Generator in ferrite, upper limit is caused by electrical breakdown of the NLTL.S. J. F. Chadwick1, N. Seddon1, S. Rukin21 The length of the NLTL at which the RF power experiences saturation MBDA ltd., Bristol, United Kingdom is close to 1 m. The first subgigawatt RF source with NiZn NLTL has2 Institute of Electrophysics, Russian Academy of Sciences, been developed to produce ~4 ns pulses at central frequency of ~1 GHz.Ekaterinburg, Russia It irradiates pulses with Gaussian distribution and peak power of about 0.2 GW. The high repetition rate of 200 pulses per second in bursts ofHigh power RF pulses of up to 1 GHz have been produced using a fully 1000 pulses has been achieved by pressurizing of the NLTL. Furthersolid state system that consists of a gyromagnetic NLTL and a solid steps of development of such RF sources needs optimization of RFstate pulse forming circuit. The novel solid state SOS diode-based extraction equipment. Another area of research deals with possibilitiesdriver produces sub-10ns pulses of the order of 100 kV. The pulse of elongation of the RF pulse which needs new approaches in thegenerator is capable of running at pulse repetition frequencies from 100 NLTLs design.Hz continuous up to 2 kHz in burst mode, with a rise-time of a few 1. V.V. Rostov, N.M. Bykov, D.N. Bykov, A.I. Klimov, O.B.nanoseconds. The gyromagnetic NLTL is a solid state RF source that Kovalchuk, I.V. Romanchenko. "Generation of Subgigawatt RF Pulsesconsists of an oil-insulated coaxial transmission line, which is loaded in Nonlinear Transmission Lines" // IEEE Trans. Plasma Sci., 2010, v.with ferrite cores and housed within a solenoid electromagnet coil. The 38, N 10, pp. 2681-2685.solenoid is used to apply a DC bias to the ferrite material and thus The work was partially supported by RFBR projects No. 10-08-00420-control the level of RF generation. The system has been operated at a, 10-08-00277-a.peak RF power output levels of around 90 MW. We present details ofthe pulse generator, the gyromagnetic NLTL, diagnostics, calibration 40
  • 40. Circuit Modeling of Nonlinear Lumped Element1B-8: valuable information for modeling of the flashover physics. The visibleTransmission Lines emission is imaged through an externally focused ICCD, while VUVN. S. Kuek1, A. C. Liew1, E. Schamiloglu2, J. Rossi3 emission is imaged through a vacuum spectrograph. The variable1 Department of Electrical & Computer Engineering, National length gap was excited with a pulser designed for a 100 ns rise time andUniversity of Singapore, Singapore, Singapore 50 kV peak output. Captured images of visible light emission from2 Department of Electrical & Computer Engineering, University of New streamers produced in oxygen are diffuse whereas nitrogen producesMexico, Albuquerque, USA streamers that are segmented. To allow VUV transmission to the3 Associated Plasma Laboratory, INPE, Sao Jose dos Campos, Brazil detector down to 115 nm, light emitted from dielectric surface flashover on the atmospheric side of an MgF₂ entrance window was focused using custom-modified off-axis parabolic mirrors into aA nonlinear lumped element transmission line (NLETL) that consists of Princeton Instruments VM-504 spectrograph that is held under vacuum.an LC-ladder network can be used to convert a rectangular input pump The use of mirrors (rather than lenses) provided better spectralpulse to a series of RF oscillations at the output. The discreteness of the resolution in the VUV range while also allowing imaging of lightLC-sections in the network contributes to the line dispersion while the emission throughout the visible and VUV wavelength regime withoutnonlinearity of the LC-elements affects the nonlinear characteristics of any chromatic aberration. For high speed imaging of VUV emission, athe line. Both of these properties combine to produce wave trains of VUV sensitive ICCD camera with 3 ns exposure was used. VUVhigh frequency. This paper describes a NLETL circuit model that is spatial images were taken by opening the entrance slit as wide asused to simulate RF generation for a given input pump pulse. The possible (3 mm), thus allowing the entire width of the discharge to becircuit model is used to study a NLETL with nonlinear capacitance viewed at the exit slit of the spectrograph. For instance, the recorded(nonlinear C but linear L), and a NLETL with nonlinear inductance images of VUV emission from oxygen exhibit higher emission closer(nonlinear L but linear C). For the prospect of better matching to a to the cathode region, while nitrogen produces a more distributedresistive load, a hybrid line that comprises both nonlinear elements L intensity profile across the gap. Also using this new method it wasand C was also investigated using the circuit model. Parametric studies possible to image the electrode region, where strong emission fromwere carried out for the various NLETLs to understand the behavior metallic atom transitions could contribute to ionization in the high fieldand characteristics of these lines. region. While MgF₂ enabled transmission and measurement of VUV, streamer characteristics of surface flashover on quartz and BK7 Experimental Demonstration of Nonlinear Lumped1B-9: dielectric windows were also investigated. In this paper, the observedElement Transmission Lines Using COTS Components streamer images in both visible and VUV wavelength range will beN. S. Kuek1, A. C. Liew1, E. Schamiloglu2 discussed as it relates to surface flashover at atmospheric pressure.1 * Work supported by the U.S. Air Force Office of Scientific Research Department of Electrical & Computer Engineering, National (AFOSR)University of Singapore, Singapore, Singapore2 Department of Electrical & Computer Engineering, University of NewMexico, Albuquerque, USA 1C-2:A Finite-Difference Time-Domain Simulation of Formative Delay Times of Plasma at High RF Electric FieldsIn recent years, there has been great interest in using nonlinear lumped in Gaseselement transmission line (NLETL) to generate RF pulses. The NLETL P. J. Ford, H. Krompholz, A. Neuberconsists of an LC-ladder network where either the capacitive element Center for Pulsed Power and Power Electronics, Departments ofor the inductive element is nonlinear. There has been the suggestion of Electrical and Computer Engineering, Texas Tech University, Lubbock,using a hybrid line where both the elements are nonlinear. This paper TX, United Statesfocuses on the implementation of a low-voltage nonlinear capacitiveline (with nonlinear capacitors but linear inductors) using COTS A Finite Difference (FD) algorithm was developed to calculate thecomponents. Experiment results that show good match to the formative delay time that is observed between the initial application ofsimulation results obtained from the circuit model of the NLETL will an RF field to a dielectric surface and the formation of a field-inducedbe presented. plasma or breakdown interrupting the RF power flow. The analysis is focused on the surface being exposed to a background gas pressure1C: Components I: Insulation and Dielectric Breakdown above 50 torr. The Finite-Difference algorithm provides numerical solutions to partial differential equations with high resolution in theMonday, June 20 09:30-12:00 CC 11A-B time domain, making it suitable for simulating the time evolving interaction of microwaves with plasma; this approach is chosen over1C-1:Advanced Imaging of Pulsed Atmospheric Surface particle-in-cell methods due to its higher computational speed and itsFlashover ease of being ported to commercial electromagnetics solvers. EmpiricalA. S. Fierro, G. R. Laity, L. L. Hatfield, J. C. Dickens, A. A. Neuber and theoretical equations allow the effective ionization and collisionCenter for Pulsed Power and Power Electronics, Texas Tech frequencies at given pressures to be found for standard FD outputUniversity, Lubbock, TX, United States variables such as the instantaneous electric field; for example, a flashover plasma induced by a 11.4 kV/cm peak field RF pulse in N2 at 60 Torr has a momentum transfer collision frequency of 300 GHz andVacuum ultraviolet (VUV) radiation is commonly believed to enhancestreamer formation, as it is energetic enough to cause photoionization an ionization frequency of 3.7 GHz. The dynamic frequency-dependent permittivity of the plasma medium is mapped to the sampled timein the gaseous volume. Light with wavelengths below 180 nm, i.e. domain of the FD algorithm using the Z transform formulation as anVUV, is highly absorbed in the atmosphere which increases the auxiliary differential equation. Therefore, together with the electrondifficulty of measuring any VUV emission from gaseous breakdown atatmospheric pressure. Nevertheless, VUV emission below 180 nm from density, the effect of the developing plasma on the instantaneous microwave field is calculated. For instance, up to 60 % absorption,pulsed surface flashover in atmospheric conditions was previously rather than reflection, of the incident RF pulse at GHz frequencies hasrecorded at Texas Tech. However, those measurements lacked detailed been simulated. This high value of absorption is primarily a result ofspatial resolution. It was desired to image VUV and visible emission the momentum transfer collision frequencies in the developing plasmadirectly while also preserving the spatial profile, which provides 41
  • 41. being much larger than the microwave frequency, and is consistent with to an estimate of the early electron temperature of 10 eV usingexperimental data under similar conditions. As a result, the electron SpectraPlot software developed at TTU, which uses theoreticaldensity increases to values well beyond the density simply calculated emission line information from the NIST atomic spectra database. Infrom setting a plasma frequency equal to the microwave frequency. addition, high sensitivity photomultiplier measurements of VUVMore specifically, a critical electron density of approximately 1E13 emission demonstrate that most of the emission is released in therather than 1E11 electrons/cm3 is adopted to explain the observed nanoseconds before voltage collapse, while emission with wavelengthsexperimental behavior. In the experiment, flashover is induced across a greater than 300 nm occurs primarily after breakdown. SinceLucite window by a 4 MW S-band magnetron operating at 2.85 GHz to significant VUV emission is released during surface flashover atproduce a 3 µs pulse propagating in the TE10 mode with ~ 50 ns rise atmospheric pressure, the relevant contribution this emission has on thetime after conditioning. The results of the FD simulation are compared photo-ionization physics of atmospheric discharges is discussed alongwith experimental data obtained from flashover with background gases with future studies in the deep VUV regime (11 – 14 eV).such as nitrogen, air, and argon all at pressures exceeding 50 Torr. * Work supported by the U.S. Air Force Office of Scientific Research (AFOSR), with additional student fellowship support from NASA / Texas Space Grant Consortium, the Directed Energy Professional1C-3:Study of Pulsed (nanosecond) Electric Breakdown of Society, and the IEEE Dielectrics and Electrical Insulation Society.Pressurized GasV. Vekselman, S. Yatom, J. Gleizer, Y. KrasikPhysics, Technion, Haifa, Israel High Dielectric Constant Composites for High Power 1C-5: Antennas K. A. OConnor, R. D. CurryThe phenomenon of ultra-fast electrical gas breakdown wasinvestigated. Nanosecond high-voltage pulses with durations of 1 ns Dept. of Electrical & Computer Engineering, University of Missouri,and 5 ns and amplitudes of 100 kV and 200 kV, respectively, were used Columbia, MO, United Statesto study the parameters of the discharge in a pressured (1-7)x10^5 Paair-filled gap. The development of the discharge and the plasma The volume and weight of high power antennas can be a limiting factorpropagation velocity was examined using optical fast frame imaging. for compact pulsed power systems. Options for antenna minimizationThe generation of runaway electrons in the breakdown process was are limited due to the relationship between an antenna’s physicalconfirmed by electron imaging and time-resolved X-ray diagnostics. dimensions and the frequencies which can be transmitted. Thus, lowRunaway electron beam energy distribution was obtained for a 1 ns and frequency antennas often require dimensions on the order of meters or5 ns duration high-voltage pulse. The origin and the role of runaway less. An effort was undertaken at the University of Missouri-Columbiaelectrons in the discharge initiation are also discussed. to circumvent these fundamental constraints on antenna size by developing and integrating high dielectric constant composite materials that can reduce the physical size of high power antennas. Traditional Nanosecond-Scale Spectroscopy of Vacuum Ultraviolet1C-4: high dielectric constant materials, including sintered perovskiteEmission from Pulsed Atmospheric Discharges ceramics, are not well suited for integration into high power antennasG. R. Laity, A. A. Neuber, A. S. Fierro, J. C. Dickens, L. L. Hatfield due to their low dielectric strength, poor mechanical properties, andCenter for Pulsed Power and Power Electronics, Texas Tech limitations in forming complex shapes. Therefore, a composite materialUniversity, Lubbock, TX, United States is necessary for improved dielectric strength and mechanical properties over sintered ceramics while maintaining a high dielectric constant.This paper describes a 2nd-generation system for directly studying the Previous research on high dielectric constant composite materials hasemission of vacuum ultraviolet (VUV) light from pulsed dielectric primarily been focused on low power applications for chip-basedsurface flashover at atmospheric pressure. The role of self-produced component minimization. The development of composites for highVUV emission (i.e. energies greater than 7 eV) on photo-ionization power systems presents additional challenges, including large thicknessprocesses during the early nanoseconds of pulsed discharges is virtually and high dielectric strength requirements for high voltage use and theunexplored, and yet could be a significant factor in the physics of fast capability to be formed or machined into complex antenna geometries.breakdown of directed energy systems (such as MW-class high power Two distinct classes of high dielectric constant composites have beenmicrowave devices) in the aerospace community. First generation developed through this effort. The two classes of composites areexperiments at Texas Tech University have shown that VUV emission generally characterized by the range of dielectric constant valuescorresponding to nitrogen and oxygen excitation in the energy range 8 measured at frequencies from 200 MHz to 4.5 GHz. The first– 10 eV is easily produced, but the use of MgF2 optics inhibited future composite class can be manufactured with a dielectric constant of 30-work with existing hardware due to the transmission cutoff of this 50 and exhibits the best machining properties of the three classes. Thedielectric material and chromatic aberration if used as a lensing second composite class has a dielectric constant of 80-100, is easilymedium. In an effort to enhance the detection capabilities of our machined, and has a loss tangent of less than 0.1 up to 4.5 GHz. Thehardware in the wavelength range from 115 – 135 nm, the current three most crucial dielectric measurements for high dielectric constantsystem utilizes a custom designed set of off-axis parabolic MgF2- composites for high power components are reported for each of theAluminium coated mirrors as the primary focusing element. For our composite classes. First, the complex permittivity is given in thestudies, a vacuum monochromator (F-number = 5.4, 3600 g/mm, frequency ranges from 100 kHz to 30 MHz and from 200 MHz to 4.5blazed at 130 nm) was used for analysis of VUV emission, which is GHz. Second, the pulsed dielectric strength is reported statistically asobserved by either VUV sensitive intensified CCD’s or the cumulative probability of breakdown measured from a set ofphotomultipliers. Additional diagnostics include fast rise-time samples of each composite class. Lastly, the electric field dependenceRogowski type current monitors, optically-isolated differential high of each composite’s dielectric constant is reported through polarizationvoltage probes, and externally focused intensified CCD cameras for measurements taken on a custom high voltage test stand.imaging with nanosecond resolution. High resolution spectroscopy with This work was supported by the Office of Naval Research underthe upgraded system resulted in observation of an additional oxygen contract N00014-08-1-0267.doublet at 130.4 nm and a nitrogen doublet at 149.5 nm, while thereduction of chromatic aberration led to an increased accuracy in lineshape calculation in the range 130 – 135 nm. Spectral calculations led 42
  • 42. 1C-6: Weibull Statistical Analysis of Impulse-Driven Surface 1C-7:Dielectric Surface Effects on Transient Arc FormationBreakdown Data in Lightning Arrestor Connector (LAC) DevicesM. P. Wilson1, M. J. Given1, I. V. Timoshkin1, S. J. MacGregor1, H. P. Hjalmarson1, A. C. Pineda2, M. F. Pasik1, R. E. Jorgenson1M. A. Sinclair2, K. J. Thomas2, J. M. Lehr3 1 Sandia National Laboratories, Albuquerque, NM, United States1 2 Dept. Electronic & Electrical Engineering, University of Strathclyde, Space Electronics Branch, Air Force Research Laboratory,Glasgow, United Kingdom Albuquerque, NM, United States2 Hydrodynamics Division, AWE Aldermaston, Reading, UnitedKingdom Lightning arrestor connector (LAC) devices protect electronic3 Exploratory Pulsed Power, Sandia National Laboratories, components by providing a conductive path to ground for electricalAlbuquerque, NM, United States power surges caused by lightning. Such devices consist of an insulating material between electrodes. This insulating region is composed of anThe introduction of solid dielectrics into liquid insulation located air gap and a high permittivity dielectric such as rutile. In thisbetween two electrical conductors is often necessary to provide presentation, the physics of the phenomena active in the early stages ofmechanical support in sub-systems of high-voltage, pulsed-power the flow of transient electrical current will be described. The results ofmachines, and it is necessary to understand the insulation behaviour of calculations that focus on an idealized structure with simple geometrysuch liquid-solid gaps. Information on the variation in breakdown will be described. In these calculations, kinetic equations are solved forvoltage/time data with an increasing number of breakdown events for the temporal evolution of the densities of species such as electrons andliquid-solid gaps subjected to over-voltages, and for breakdown ions. The key physics at early times is the release of electrons in the airinitiation measurements are reported in [1]. Five different solid by impact ionization caused by hot electrons that were heated by thematerials were tested, namely polypropylene (PP); low-density large electric fields within the air-filled regions. For these calculations,polyethylene (LDPE); ultra-high molecular weight polyethylene the initial electrons are released by ionizing radiation. As time(UHMWPE); Rexolite (cross-linked polystyrene); and Torlon progresses, some of the charged species cause charging of the solid(polyamide-imide). The applied voltage wave-shape was 100/600 ns, dielectric. The resultant phenomena will be discussed in terms of theirand the impulses were of negative polarity and magnitude up to 450 kV. relation to the behavior of LAC devices and Dielectric Barrier Discharge (DBD) devices.A detailed statistical analysis has now been conducted on the Sandia National Laboratories is a multi-program laboratory managedbreakdown data generated under both uniform and non-uniform field and operated by Sandia Corporation, a wholly owned subsidiary ofconditions, for breakdown events on or around the solid surfaces when Lockheed Martin Corporation, for the U.S. Department of Energysimmersed in mineral oil. The present paper is focused on the statistical National Nuclear Security Administration under contract DE-AC04-analysis of the breakdown initiation results reported in [1], using a 94AL85000.three-parameter Weibull distribution. Values of the location parametergamma provide an estimate of applied voltage below which breakdown Effect of Electrode Surface Roughness on the 1C-8:will not occur, and under uniform-field conditions, gamma varied from192 kV (corresponding to average applied field 480 kV/cm) for PP to Breakdown Jitter of a Nanoparticle-Infused Dielectric Oilzero for UHMWPE (i.e. the data for UHMWPE were better described Spark Gap Switchby a two-parameter distribution). Longer times to breakdown were C. A. Yeckel, R. D. Currymeasured for UHMWPE in [1] when compared with the other materials. Dept. of Electrical & Computer Engineering, University of Missouri,However, high values of the shape parameter beta reported in the Columbia, MO, United Statespresent paper suggest greater sensitivity to an increase in appliedvoltage – that is, the probability of breakdown increases more sharply Electrode surface roughness is generated over the lifetime of an oilwith increasing applied voltage for UHMWPE compared to the other spark gap by high-temperature discharge during normal circuitmaterials. Only PP consistently reflected a low value of beta across the operation. These electrode pits and protrusions create macroscopicdifferent sets of test conditions. electric field enhancements allowing electrons to initiate breakdown at lower switch voltages. Nanoparticles with a high dielectric constant areThe average streamer propagation velocities implied from time-to- thought to polarize and interact with these enhancements to smooth thebreakdown data are also presented and discussed. According to the electric field profile on the electrode surface, thus reducing the spuriousclassification presented in [2], streamers were found to propagate in the low-voltage breakdowns generated by local field maxima.third (10-20 km/s) and fourth (100 km/s) modes in non-uniform fields, Experimental testing is underway to isolate the effect that electrodeand in the second (1-5 km/s) and third modes in uniform fields. The roughness has on the breakdown jitter of a single-shot oil spark gapupper streamer velocity in the present study was over 100 km/s. switch using a pressurized oil dielectric. The dielectric oil utilized is a synthetic hydrocarbon-based liquid, designated Nycodiel, which[1] M.P. Wilson, S.J. MacGregor, M.J. Given, I.V. Timoshkin, M.A. conforms to military specification MIL-PRF-87252C. Nycodiel hasSinclair, K.J. Thomas, and J.M. Lehr, “Surface flashover of oil- experimentally demonstrated significantly lower switch self-breakimmersed dielectric materials in uniform and non-uniform fields,” voltage standard deviation (jitter) when infused with a suspension ofIEEE Transactions on Dielectrics and Electrical Insulation, vol. 16 (4), high-dielectric constant BST nanoparticles with polished electrodepp. 1028-1036, 2009. surfaces. Further tests will examine the effect that nanoparticle-infused[2] R. Liu, C. Törnkvist, V. Chandramouli, O. Girlanda, and L.A.A. oils have on the breakdown jitter of switches whose electrodes havePettersson, “Ester fluids as alternative for mineral oil: The difference in roughened surfaces. The results of these breakdown tests willstreamer velocity and LI breakdown voltage,” 2009 Annual Report: determine if the benefits of the particles scale with the magnitude of theConference on Electrical Insulation and Dielectric Phenomena, pp. electrode field enhancement at various pressures. Experiments are543-548, 2009. being performed on the HVADTS system at UMC. The HVADTS is a single shot oil spark gap switch test stand capable of applying a 250 kV, 21 J pulse to a planar 0.16 cm gap with a ‘1-cos’ rise-time of 1.6 µs. A recirculation pump and filter allows the oil to be reconditioned between pulses. Both oil pressure and electrode surface roughness are controlled 43
  • 43. variables in these experiments, as pressure has been shown to be a alignment BN nanosheets in a viscous matrix. BN nanosheets havesignificant factor in jitter behavior. Simulations of the effect of surface unique physical and thermal properties and exhibit semiconductingroughness on the electric fields in the breakdown gap with a behavior with a wide band gap (5.5~6.4 eV) depending on itsnanoparticle suspension are being performed concurrently with this polymorphs. The thermal conductivity of hexagonal BN increases withexperiment. increasing anisotropy: when BN nanosheets are aligned perpendicular to the c-axis their thermal conductivity is almost 20 times greater than that when they are aligned parallel to the c-axis. Therefore combined1C-9: Pulsed Pre-breakdown Phenomena in High with the merits of electric insulation and workability of polymer cores,Pressurized Carbon Dioxide including Supercritical State the polymer/BN composite films are potential for applications forT. Ihara, T. Furusato, S. Kameda, T. Kiyan, S. Katsuki, M. Hara, thermal interface materials in semiconductor industry. Well-alignedH. Akiyama nanosheets with a one-dimensional orientation in a polymer matrixGraduate School of Science and Technology, Kumamoto University, exhibit a conductivity percolation transition at a remarkably lowKumamoto, Japan volume fraction compared with other powders. Controlling the reorientation of nanotubes in polymer matrix using electric field isIn recently years, electrical discharge in high pressurized carbon advantageous to accomplish the linear structure induced by coulombdioxide including supercritical state has been studied as a basic of attraction without modifying the surface. However, the researches overpromising technology to create new chemical reaction fields. the one dimensional arrangement of BN nanosheets under electric fieldSupercritical fluids (SCFs) have unique characteristics of high are limited. Because the electric field should be regulated lower thandiffusivity, high solubility and low viscosity, and discharge plasma the breakdown voltage of polymer/BN composite film, and which isproduces chemical species such as radicals, excited particles and ion [1, disadvantageous to give electrophoresis to BN; the higher voltage of2]. Especially pulsed streamers might be very effective discharge to electric field is needed to allow electric permittivity to BN in polymerobtain non-thermal plasma which gives the dense chemical species matrix due to its wide band gap property. In this study, we attempted towithout temperature rise of fluids. However researches of streamer fabricate anisotropic aligned BN nanosheets in a viscous polymerprocesses are not well understood yet due to the difficulty of matrix without surface modification of BN nanosheets using nanoobservation of the phenomena in a high-pressure and high-voltage pulse-width electricity. The nano pulse-width electricity was used tovessel. overcome the limitation of breakdown voltage of the hybrid film atIn the present study, pulsed streamers in high pressurized carbon high electric field. The hexagonal BN nanosheets were dispersed bydioxide including supercritical state were observed using Schlieren sonication in a prepolymer mixture of polysiloxane followed by a highmethod. The discharges were produced in a high pressure vessel with a speed mixing. The suspension was cast on a spacer of microscalevolume of 50 ml and two sapphire windows for observation. Fluid thickness and applied to electric field before it became cross-linked.temperature was set to T= 305 K and pressure was changed to P= 8.0, The electric responses of various nano-dimensional BN nanoparticles5.0, 2.0 MPa. The gap is a needle-to-plane: the gap length and the tip were compared by zeta potential. Two kinds of polymers, polysiloxaneradius were d=1 ~ 10 mm and 5 um, respectively. In a series of and epoxy resin, were prepared to compare the viscosity effects of theexperiments, positive/negative pulse voltage having rising rate of 0.8 polymer matrix. X-ray diffraction, scanning electron microscopy, UV-kV/ns and pulse width of 400 ns generated by magnetic pulsed visible and thermal conductivity measurements were used tocompression generator with a step-up transformer was applied to needle characterize the composites.electrode. Nd:YAG laser (532 nm) with pulse width of 5 ns has The polymer/BN nanosheet composite film was prepared under nanosynchronized with the discharge images which were taken by a digital pulse-width electric field at 40 kV, 100~120 ns of electric pulse andCCD image sensor (Nikon, total pixels: 10.75 million). 20Hz of frequency. The breakdown voltage, which was 2.5 kV underThe obtained main results can be summarized as follows. DC electric field, could be highly increased to 40 kV by using nano1) Shape of positive streamers in supercritical state is a branched tree pulse-with electricity without electric currency. The anisotropicassociated with shock waves of spherical and cylindrical shapes. The alignment of none-modified BN nanosheets inside polysiloxane couldnumber of positive streamer branches decreases with decreasing the be controlled with variation of the shape and dimension of BNfluid density. nanosheets, electrodes, DC electric fields, and by application of nano2) The negative corona shape in supercritical state differed quite from pulse-width electricity. The anisotropic alignment of BN nanosheetsthe positive one, and was a spherical shape without branch at the initial attributed to the thermal conductivity enhancement of the composite.stage. The spherical shock wave started from the needle tip at the The mechanism of heat conduction through before and aftercorona onset. The structure of shock front at the negative polarity orientations of BN nanosheets are discussed.changed with decreasing the fluid density. 1P: Components posters I: Insulation and Breakdown,References Transmission Lines and Diagnostics[1] T. Kiyan et al., Plasma Processes and Polymers, Vol. 6, No. 11, pp.778-785 (2009) Monday, June 20 13:30-15:30 Regency Ballroom[2] T. Tomai et al., J. Supercritical Fluids, Vol. 41, pp.404-411(2007) 1P-1: A Repetitive Operated High-Current Beam Collector Highly Oriented BN Nanosheets in polymer/BN1C-10: T. Xun, H.-W. Yang, J.-D. Zhang, J. Zhang, Z.-C. ZhangNanosheet Composite Film with Increased Thermal College of Opto-electric Science and Engineering, National UniversityConductivity Using Nano Pulse-Width Electric Field of Defense Technology, Changsha, Hunan, ChinaH.-B. Cho, T. Nakayama, S. Tanaka, W. Jiang, H. Suematsu, T. Suzuki,N. C. Tung, K. Niihara With the increase in power and miniaturization in size of a repetitiveExtreme Energy-Density Research Institute, Nagaoka University of operated high-power microwave (HPM) source, the thermalTechnology, Nagaoka, Niigata, 940-2188, Japan management and cooling methods for a beam collector have great effects on efficient operations of the devices. In this paper, based on a traditional coaxial beam collector of Torch-01, the effecting area due toPolymer/boron nitride (BN) nanosheet composite films were fabricated bombardment by energetic electrons was experimentally quantified,under nano pulse-width electric field to enhance the anisotropic and the power per unit area as pulsed thermal loading for a 700 keV 44
  • 44. and 7 kA beam current was calculated. By using an ANSYS package, operating conditions that lead to thermal failure. This model is used tothe 10-100 Hz results of temperature history and distribution of the simulate the peak temperatures generated by a series of maximum ratedcollector were typically presented. The dependence of peak energy impulses, periodically delivered as to meet the maximum ratedtemperatures on forced cooling condition at different pulse repetition average power of the device. To experimentally validate the model, therate was also obtained. Comparison results show that the cooling resistor was subjected to similar voltage stresses and the surfaceprocess is influenced by collector thermo-physical parameters, interface temperatures are compared.thickness, and convective factor of the forced water cooling. Finally,the hydraulic system necessary to achieve appropriate cooling fluid 1P-5: Comparison of Dielectric Strength of Transformer Oilflow was discussed. These results may supply an efficient reference tothe optimizing design and reasonable use of a repetitive operated beam at DC and Multimillisecond Pulsescollector in HPM tubes. A. Pokryvailo, C. Carp Spellman High Voltage Electronics Corporation, Hauppauge, United States1P-2: Design of Compact Feed Through for 500 kV HighVoltage Cable The goal of this work was determining dielectric strength ofL. Veron1, R. Rosol2, J.-C. Brion31 transformer oil at multimillisecond pulses as compared to DC voltage. CEA, Bruyères-le-Châtel, 91297 Arpajon Cedex, France The test sphere-to-plane gap was varied from zero to 15mm to mimic2 CERN, TE/ABT/FPS, 01631 CERN Cedex, France field nonuniformity from unity to about four as shown by detailed field3 Europulse, Route de Gignac, 46600 Cressensac, France analyses. Both clean and contaminated oils were tested. Standard ASTM D 1816-84a breakdown tests preceded DC and pulsed testing.When we use high voltage cables as transmission lines, it can be The latter was conducted at up to 160kV at repetition rate (RR) up tonecessary to plug two cables together to increase the length or to pass 30Hz, with 1 ms risetime and much longer load-dependent fall time.through an interface (to work in a closed vessel for instance). This The main results are as follows. In conformance with published data,paper presents the design and the tests of a high voltage connector the breakdown voltage, Vbr, is higher at negative polarity for both freshworking at 500 kV. Operational needs and technical considerations and contaminated oil, and the spread is smaller. However, there is ahave been taken into account to design this component. The high marked difference between pulsed (RR 30Hz, 20% duty cycle) and DCpulsed voltage (500 kV - 50 ns), the limited size (maximum diameter : breakdown at both polarities for soot-contaminated oil: pulsed Vbr is100 mm), the possibility to work into a primary vacuum (10-2 mbar) higher than its DC counterpart by 15% and 10%, on the average,and our given cable (Z = 56 ohm) have been the input parameters. The respectively. Most probably, the bridge mechanism governs breakdownimpedance matching of the connector has been specially studied to for contaminated oil.avoid voltage reflection and drastic deformation of the voltage andcurrent shapes. After the design of the connector, many tests into DC 1P-6: Streamer Initiation and Propagation in Transformervoltage (60 kV) and pulsed voltage (500 kV) have been performed tooptimize it. For some applications into a primary vacuum, it can be Oil under Positive and Negative Impulse Voltagespossible to pressurize the connector to avoid surface flashover. J. Jadidian1, J. G. Hwang1, M. Zahn1, L. A. A. Pettersson2 1 Dept. of Electrical Eng. and Computer Sci. , Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA,1P-3: Isolation Concepts for a HVPS-System with <5µs Pulse United StatesGeneration 2 ABB Corporate Research, Västerås, SwedenM. HohmannTranstechnik GmbH & Co.KG, Munich/Holzkirchen, Germany Liquid dielectrics are widely used in compact pulsed power technologies, high voltage insulation and power apparatus cooling.The HV power supplies for klystrons or kicker magnets needs to Particularly, transformer mineral oil is of high technical interest sincegenerate pulses of 100 kV and higher for rise times of 5 µs. This its electrical breakdown strength and thermal conductivity are greatergenerates displacement charges, which generate ionisations. This leads than gaseous insulators, while its ability to conform to complexto flashover, even for distances of significantly more than 1 kV/mm geometries means that they are often of more practical use than solidand the housing must be made of metal according to relevant standards insulators. The phenomenology of liquid breakdown has been theand be designed with double walls for personal protection. The subject of theoretical and experimental studies for many years but stillmodules must also be completely kept away from the housing walls. is not fully understood. Electrical breakdown in liquid dielectrics isAn isolation concept for this reqirements is discussed in the following. carried out by streamers which are low-density, conductive structures that form in regions of the liquid that are electrically over-stressed. The extent of a streamer depends upon the nature of the pulsed electrical1P-4: Thermodynamic Modeling with Experimental excitation such as polarity, magnitude, duration, rise time, etc.Validation of the Pulsed and Periodic Operation of a High Sustained over-excitation can result in development of a plasmaPower Resistor channel through the liquid gap between electrodes, known as anD. P. Muffoletto, T. M. DiSanto, K. M. Burke, J. L. Zirnheld electrical arc which can dissipate huge amount of energy. In this paper,Electrical Engineering, University at Buffalo, Buffalo, NY, United the breakdown physics of transformer oil is investigated using electro-States thermal hydrodynamic modeling of streamers. Results of the model are compared for streamers formed by positive and negative standardThe size and performance demands placed on high power resistors are impulse voltages in a needle-sphere electrode geometry 25 mm apart.increasing, in particular for their use in the growing electric vehicle The model demonstrates that the formation of streamers in transformerindustry. Ceramic composite resistors, which demonstrate high peak oil is due to the field dependent molecular ionization of differenttemperature limits and high power handling capabilities in low families of hydrocarbon molecules at varying electric field levels. Afterinductance packages, are attractive choices for these applications. A inception of streamers, migration of the electric charge carriers enablesthermodynamic model of one such resistor is presented to better streamers to develop from the tip of the needle electrode that eventuallyunderstand their use under repetitive high power stresses and the reach the sphere electrode. The migration of each charge carrier is 45
  • 45. described by its own mobility. These mobilities which principally 1P-8: Field Enhancement Simulation of a Nanoparticle-determine the shape and speed of the streamers are imported to the Infused Dielectric Oil with Roughened Electrodesmodel as functions of temperature, particle density and electric field. C. A. Yeckel, R. D. CurryThe results indicate that the volume over which the space charge exists Dept. of Electrical & Computer Engineering, University of Missouri,is much greater for negative streamers as the high-mobility electrons Columbia, MO, United Statesare repelled by negative needle electrode and migrate into the liquidbulk resulting in a lower electric field enhancement, which reducesionization in the oil. Consequently, negative streamers formation The Center for Physical and Power Electronics at UMC hasrequires a greater applied voltage magnitude than their positive significantly reduced the self-break jitter of a single-shot pulsed oilcounterparts. When a positive impulse voltage is applied, the electrons switch with various nanoparticle-infused oil dielectrics. In support ofexit the ionization zone back to the needle electrode which creates a this effort, electromagnetic simulations have been completed to betterpositive space charge region that reduces the electric field near the understand the phenomena observed. The simulations focus onneedle electrode and enhance the electric field ahead of the positive modeling the electric fields associated with the spatially randomspace charge region. The field enhancement makes the positive placement of both electrode enhancements and nanoparticles in an oilstreamer initiation possible at lower applied voltages. According to the gap.model, for a given applied voltage, negative streamers travel slower A field enhancement on the surface of an electrode allows electrons tothan positive streamers and occupy more radial space, which is in initiate breakdown at a lower voltage and are thought to negativelyagreement with experimental studies found in the literature. The results influence switch jitter. The nanoparticles may positively influence jitteralso reveal that the shape of positive streamers are approximately by mitigating the effects of these field enhancements. By performinginsensitive to electron mobility as long as it is about two orders of simulations using CST we have concluded that nanoparticles createmagnitude greater than the ion mobilities. However, the shape and the non-uniform field enhancements on the surface of a smooth planarvelocity of negative streamers is quite sensitive to the electron mobility electrode. A single high-k nanoparticle in close proximity to thevalue. electrode surface can locally increase the electric field by several percent. The current simulations vary the magnitude of the field enhancement on the electrode surface and an analysis is performed on1P-7: The Influence of Magnetite Nano Particles on the their field interactions with the suspended nanoparticles.Behaviour of Insulating Oils for Pulse Power Applications. These simulations are designed to supplement experimental work beingM. J. Given1, M. P. Wilson1, I. V. Timoshkin1, T. Wang1, done on oil spark gap switches. Experimental testing has indicated thatS. J. MacGregor1, J. M. Lehr2 some nanoparticle-infused dielectric oils reduce the jitter in a pulsed1 University of Strathclyde, Glasgow, United Kingdom single-shot switch with planar polished electrodes. The nanoparticles2 Sandia National Laboratories, Albuquerque, USA are thought to interact with the field enhancements to smoothen the electric field profile on the electrode surface. By correlating theTo allow the compact design of pulse power systems, it is often experimental and simulation data, further insight may be gained into anecessary to employ liquid insulating systems such as mineral oil. The nanoparticle’s role in the breakdown process. The results of thedielectric and breakdown behaviour of existing and candidate oil simulations are discussed with the final results of the study.formulations such as Diala D, MIDEL 7131 and THESO have beenreported by the authors [1,2]. Work has also been performed to assess 1P-9: Glass Ceramic Breakdown Characteristics underthe behaviour of flashovers at the boundary between liquid and solid Repetitively Pulsed Conditioninsulation systems [3]. There is evidence in the literature that the S. Wang, J. Zhang, H. Yang, T. Shuaddition of ferroferric oxide (Fe3O4 ) nano particles to liquid insulation College of Optoelectronic Science and Engineering, Nationalcan lead to significant increases in breakdown strength [4]. This effect University of Defense Technology, Changsha, Chinawill be of interest to engineers designing pulse power supplies. Thispaper will report on some preliminary findings as to the changes inbreakdown behaviour under dc and impulse conditions resulting from On the road to the portable pulsed power generators, application thethe addition of Fe3O4 nano particles to Diala D, MIDEL 7131 and ceramic with high energy density to energy storage for pulse formingTHESO insulating oils. In addition the effects of the addition of Fe3O4 line has aroused much enthusiasm in the research group of pulsednano particles on the behaviour of breakdowns at a solid liquid power technology. The glass ceramic investigated in this paper is ainterface will be reported . kind of promising ceramic to be applied in the parallel-plate stacked[1]Dielectric Properties of Diala D, MIDEL 7131 and THESO Blumleins, which is a nice alternative to achieving compactness, thusInsulating Liquids Timoshkin, I.V.; Fouracre, R.A.; Given, M.J.; being portable. Experiments for testing its breakdown characteristicsMacGregor, S.J.; Mason, P.; Clephan, R.; Annual Report Conference have been performed on the platform named ARC-01 with the ability toon Electrical Insulation and Dielectric Phenomena, 2008. CEIDP 2008. output repetitive pulses. Before the test, a circuit simulation with the26-29 Oct. 2008 Page(s):622 - 625 same parameters of the platform has been presented to predict the[2]Mixtures of Insulating Liquids For Pulsed Power Applications output pulse of the test. Test of Breakdown characteristics underTimoshkin I.V., MacGregor S. J., Given M. J., Wilson M. P. Accepted repetitively pulsed condition with various frequencies and pulsefor publication IEEE Transactions on Dielectrics and Electrical numbers has been carried out. The results indicate that the profileinsulation design of electrode has a significant effect on the performance of the[3]Surface flashover of oil-immersed dielectric materials in uniform ceramic strength. To avoid the field enhancement points and find anand non-uniform fields Wilson, M.P.; Macgregor, S.J.; Given, M.J.; optimized profile of electrode, some different electrodes have beenTimoshkin, I.V.; Sinclair, M.A.; Thomas, K.J.; Lehr, J.M.; IEEE studied. The influence of testing environment has also beenTransactions on Dielectrics and Electrical Insulation, Volume 16, Issue investigated via varying some sorts of dielectrics as water, glycerin and4, August 2009 Page(s):1028 - 1036 capacitor oil.[4]Electrical properties of nano modified insulating vegetable oil ZhangZ Li J Zou P and Grzybowski S Ann Report Conference on Electrical 1P-10: High Voltage Breakdown of Alumina InsulatorsInsulation and Dielectric Properties 2010 CEIDP 2010 17-20 Oct 2010 T. P. Hughes, L. I. EspadaPage(s) 34-37 Sandia National Laboratories, Albuquerque, United States 46
  • 46. We are carrying out a computational study of the high-voltage of the media electrical insulation property and technology under thebreakdown properties of alumina insulators using an electrostatic, high repitition rate (from hundreds of Hz to thousands of Hz) is a mainfinite-element, particle-in-cell code Aleph being developed at Sandia. development direction. This paper analyzes two features ofSecondary-electron emission is modeled using an expression for the withstanding characteristic about the gas switch in the control of highyield as a function of primary-electron energy and incident angle due to repitition rate, which is obvious distinctive from low pulse repetitionVaughan (1993) and Jonker (1957). The free parameters in the function frequency: first, breakdown field strength(voltage) of the gas gap hasare fit to data obtained by Dawson (1966). We have modeled the dropped dramatically when increase the repetition frequency, but whenbuildup of surface charge on the insulator surface and its effect on the frequency increases to a certain, the decrease of electric breakdownelectric potential distribution. Results will be compared to data from a field will reduce after further increasing repetition frequency; Secondly,parallel experimental effort. Work is in progress to include the effects electric breakdown field will appear abnormal increase or reductionof neutral desorption and ionization. under a certain high repitition rate. Generally consider that whenSandia National Laboratories is a multi-program laboratory managed repetition frequency is increased, breakdown voltage of the gas gap hasand operated by Sandia Corporation, a wholly owned subsidiary of dropped dramatically is due to charge accumulation effect, this point ofLockheed Martin Corporation, for the U.S. Department of Energys view may account for the phenomenon that breakdown field strengthNational Nuclear Security Administration under contract DE-AC04- will reduce under low repetition frequency((from hundreds of Hz to94AL85000. thousands of Hz), however it is difficult to explain that breakdown electric field will appear abnormal increase or reduction in control of a certain frequency. The article set up a control model of electric1P-11: Theoretical and Experimental Investigation of Electro potential automatic control in the switches gap, through the electronDischarge Destruction of Non-Conducting Materials avalanche of relaxation oscillation and the development process ofN. S. Kuznetsova, V. V. Lopatin, V. V. Burkin, D. V. Zhgun, streamer, explore gaseous dielectric insulating abnormal breakdownN. A. Ivanov mechanism under high repitition rate.Institute of High-Technology Physics, National Research TomskPolytechnic University, Tomsk, Russian Federation Three-Dimensional Electromagnetic Modeling of 1P-13: Composite Dielectric MaterialsThe subject of investigation is the phenomenon of electro burst in the K. A. OConnor, C. A. Eastman, N. J. Grove, R. D. Currycondensed dielectrics, which is a basis of electro dischargetechnologies of destruction and processing of solids. The task of Dept. of Electrical & Computer Engineering, University of Missouri,investigation is theoretical and experimental research of the electrical Columbia, MO, United Statesdischarge in non-conducting materials. To accomplish the task theunifying physical and mathematical model is developed. Operation of When manufacturing composite dielectric materials with varyingthe discharge circuit, discharge channel expansion, generation and shapes, sizes, dielectric properties, and amounts of each constituentpropagation of the mechanical stress waves in dielectric materials, element, it is beneficial to predict the effective dielectric constant andwave reflection, generation of compressive, shear and tensile the distribution of electric fields within the composite. A three-deformations are self-consistently considered in the model. Power dimensional electromagnetic model has recently been developed at thecharacteristics of wave disturbances generated by the expanding University of Missouri-Columbia to aid in this analysis. CST EMchannel in rock depending on the circuit parameters are analyzed for Studio, which specializes in three-dimensional electromagnetic modelthe choice of the effective modes of energy-release in the channel. simulation, was chosen for the simulation environment. A custom-Results of physical experiments of electro burst in concrete blocks are developed program automates the construction and electromagneticpresented at initiation of the discharge by the electrical explosion of a analysis of the composite material within CST EM Studio. Thecopper wire. The electro explosive cartridge for the solid destruction program virtually constructs a composite with several hundred tohas been developed. Effectiveness of application of materials with thousands of distinct composite elements placed in a quasi-randomhigher acoustic stiffness in comparison with water as the transfer arrangement according to user-defined input parameters. In this way,medium for increasing the efficiency of electrical energy conversion to the program provides a means of improving the accuracy in modelingthe wave energy, which destruct the material has been shown. composite structures without the hundreds of man hours required toApplication of polythene allows increasing the wave energy by ~20– draw the many individual composite elements.25 % in comparison with the electrical discharge in water. Thus the The program enables user specification over the simulation parameterspressure amplitude of generated wave increases for ~25–30 %. The and automated analysis of the simulation results through a userchoice of polythene as the material surrounding the wire is caused also interface. The dielectric constant of each composite element, theby absence of crush zone in a vicinity of the channel that allows to loading factors, and multiple physical particle parameters are user-avoid considerable expenditure of energy in this zone. Effect of the defined. These parameters include the unique arrangement, shape, andlength and diameter of the exploding wire in an electro explosive size of the composite elements. The effective dielectric constant of thecartridge on the efficiency of electro discharge destruction of solid is composite is determined by analyzing the capacitance of a parallel plateexperimentally investigated. Verification of simulation and capacitor made with the virtual composite material and accounting forexperimental results is carried out. the geometry and dimensions. Additionally, the distribution of electric fields through the composite elements and at potential triple points can be analyzed for even distribution and potential sites of failure. A1P-12: The Study of Gas Abnormal Breakdown detailed description of the modeling method and program is provided.Characteristic under High Repetition Rate Pulsed Power An analysis of the change in the effective dielectric constant for variousC. Yu composite elements and loading fractions is given, and examples ofBeijing Institute of Special Electromechanical Technology, Beijing, how the electric fields are distributed through the composite structureChina are included. This work was supported by the Office of Naval Research underIn defence and industrial applications along with the development of contract N00014-08-1-0267.pulsed power technology, electrical insulation property of the dielectricwhich under extreme conditions have aroused broad attention, the study 47
  • 47. 1P-14: Numerical Generation of the Random Variable shock wave is located on the propagating streamer channel; a sphericalImpulse Breakdown Voltage of Gases shock wave is formed on the branching points. 3) The bubble formationE. Dolicanin1, K. Stankovic1, R. Maric2, B. Iricanin1, G. Ilic2, in the streamer initiation process is not able to be recognized in SCCO2R. Radosavljevic1 [1, 2].1 Faculty of Electrical Engineering, University of Belgrade, Belgrade, ReferencesSerbia [1] JF Kolb, R P Joshi, S Xiao and H Schoenbach: J. Phys. D:Appl.,2 Electric Power Industry of Serbia, Belgrade, Serbia Phys., 41, 234007 (2008) [2] W. An, K. Baumung and H. Bluhm: J.Appl.Phys.101, 053302 (2007)Electrical breakdown of gases is self-sustained avalanche process basedon the positive feedback of successive avalanches initiated with the 1P-16: Study on Contact Resistance of Electric Circuit Modelfirst, initial electron. Mathematical model of electrical breakdown, for Electromagnetic Railgunbased on the geometrical progression of avalanche processes, has been P. Liu1,2 1well elaborated and verified in practice. This model gives possibility Department of Electrical Engineering, Tsinghua University, Beijing,for calculating the minimal value of breakdown voltage, i.e. dc China 2breakdown voltage value of the observed insulating system, by Beijing Institute of Special Electromechanical Technology, Beijing,knowing the Townsend avalanche coefficients, which depend on the Chinaratio of the electrical field value to the pressure value within theinterelectrode gap. Such obtained theoretical result is in very good The interface characteristics between rails and armature in a railgunagreement with experimental results, independently of weather the system are complex and essential to built a model for electromagneticfeedback of self-sustained discharge arised due to processes active on railgun. In this paper, a novel electric circuit simulation modelelectrodes (Townsend mechanism) or due to processes active in gas considering the contact resistance for electromagnetic railgun is(streamer mechanism). However, this agreement of theoretical and suggested The railgun system is moduled and implemented withexperimental results exists only in case when the voltage rise rate is Simplorer. This paper also presents the experimental results with themuch lower then the rate of change of the elementary processes of gas 6MJ railgun system of BISIT and simulation results of a sample railgundischarge (dc voltages). But, when the voltage rise rate is comparable system with and without contact resistance.with the rate of change of the elementary processes (impulse voltages),the breakdown voltage becomes stochastical quantity so its theoreticaldetermination is not possible any more. By using the dynamic 1P-17: Development of Small Dimension High-Voltagemodeling of impulse gas breakdown, the algorithm which allows Electronic Vacuum Devicesgeneration of statistical sample of random variable impulse breakdown V. D. Bochkov1, D. V. Bochkov1, V. N. Nikolaev1, V. I. Teryoshin1,voltage has been presented within this paper. Test and verification have P. V. Panov1, A. V. Batrakov2, K. V. Karlik2, G. E. Ozur2,been performed by the comparison of the parameters obtained from D. I. Proskurovsky2stochastical samples of random variable impulse breakdown voltage 1 Pulsed Technologies Ltd., Ryazan, Russian Federation 2generated by the suggested algorithm with the corresponding statistical Institute of High Current Electronics RAS, Tomsk, Russian Federationsample parameters obtained by the experiment within the wide range ofpressure values and interelectrode distances for electronegative, Reliability of high-voltage electron vacuum and gas-discharge deviceselectropositive and noble gases. – X-Ray tubes, neutron tubes, gyrotrons, thyratrons, accelerators, is mostly limited by hold-off voltage capability of the tubes. However the Visualization of Positive Pulsed Streamer in1P-15: known manufacturing technologies do not allow to achieve a high hold-Supercritical Carbon Dioxide by Schlieren Method off voltage in compact design. In order to cope with the problem it isT. Furusato, T. Ihara, S. Kameda, T. Kiyan, S. Katsuki, M. Hara, essential to consider the whole lot of issues emerging when designingH. Akiyama and manufacturing the devices and in particular interaction of processesGraduate School of Science and Technology, Kumamoto University, on surface of high voltage electrode system, on surface and in theKumamoto, Japan volume of dielectric envelope as well as processes outside the envelope. X-Ray tubes are typical representatives of high-voltage vacuum and gas-discharge devices. Therefore the work on creation of reliableDischarge phenomena in supercritical fluids have attracted rising compact X-Ray tube completely reflects basic trends of ourattention in the electrical and chemical engineering because both of investigations: • Reduction of intensity of charged particles parasiticdischarge plasma and supercritical fluids have remarkable potential in emission and improvement of vacuum insulation dielectric strength.the reaction fields. However the discharge phenomena in such high For this purpose we use technology of modification of electrode surfacepressure fluids have still remained on several studies such as discharge structure and condition by thermo-processing with high-current (up tomechanism and discharge behavior. In this work, we concentrate on the 30 kA) pulse (duration ~5 µsec) low-energy ( up to 35 keV) electronicvisualization of positive pulsed streamer in supercritical carbon dioxide beam down to the depth from 30 nm to 30 µm [1]. • Improvement of(SCCO2) by means of Schlieren Method. In a series of experiments, a hold-off voltage capability owing to influence of dielectric elements, inmedium condition is fixed at the pressure range of 0.1 to 10 MPa and particular, tube envelope [2]. We use a coating technology of envelopetemperature range of 273 to 310 K. A magnetic pulse compression internal surface by multi-layers consisting of metal oxides crystals,module with a step-up transformer was employed as pulse generator. when each particle of the layer has on the surface a nano-layer of inertAn electrode system has adopted a needle-to-plate configuration, which non-organic material – silicon oxide. • Development of impregnatedthe curvature radius of electropolished needle tip and gap length are and carbon-graphite cathodes based on nano-structures. • Application5µm and 10 mm, respectively. Schlieren images are visualized by Nd- of methods for tube oil cooling without mechanical drive owing toYAG laser (532nm) with time resolution of 5 ns; it is stored in a digital electro-hydrodynamic effects. • Improvement of reliability of externalCCD sensor synchronized to the discharge in SCCO2. The medium in which the device is operated, in particular dielectric strengthexperimental results are as follows. 1) The streamer formation in of transformer oil. In the report results of investigations of X-Ray tubesSCCO2 is different from that in gas and liquid states. 2) There are two for anode voltages of 160 kV and 250 kV with length of 150 mm andtypes of shock wave surrounding propagating streamer: a cylindrical diameter of 55 mm are presented. The proposed solutions reduce 48
  • 48. intensity of charged particle parasitic emission and prevent high density their parallel capacitance after the dynamic dielectric recover prccess.charge accumulation inside the tube, thus providing sound reduction of Ther efore, the application of low dispersion operation mechanism anddielectric envelope through-breakdowns possibility and high hold-off the selection of interrupter garding capacitance are the key factors ofvoltage capability. successful breaking of vacuum switch with double-breaks and multi-[1] Ozur G.E., Proskurovsky D.I., Rotshtein V. P., Markov A.B.. breaks.Production and application of low-energy, high-current electron beams.Laser & Particle Beams, 2003, vol.21, No.2, pp.157-174. 1P-20: Research on Breaking Capacity of Hybrid Circuit[2] Bochkov V.D. and Pogorelskii M.M., "Study of the chargedistribution over insulating envelope in a high-voltage vacuum device", Breaker Base on Vacuum Interrupter and SF6 InterrupterJournal of Technical Physics (Russian), Vol.69, No.6, 1999, pp.30-35. in Series X. Cheng, M. Liao, X. Duan, J. Zou School of Electrical Engineering, Dalian University of Technology,1P-18: Research on Breaking Capacity of Hybrid Circuit Dalian, ChinaBreaker Base on Vacuum Interrupter and SF6 Interrupterin Series Hybrid circuit breaker, consisting of a vacuum interrupter and an SF6X. Cheng interrupter in series, which take the place of SF6 circuit breaker, is aSchool of Electrical Engineering, Dalian University of Technology, research area to Low-carbon technology for high-voltage big–capacityDalian, China circuit breaker. The experimental model of fiber-controlled hybrid circuit breaker with a vacuum circuit breaker and an SF6 circuit breakerHybrid circuit breaker, consisting of a vacuum interrupter and an SF6 in series was set up, and the fault current breaking experiment wasinterrupter in series, which take the place of SF6 circuit breaker, is a conducted on two circuit breakers in different collaborative actionresearch area to Low-carbon technology for high-voltage big–capacity moment. The voltage distribution across the vacuum circuit breaker andcircuit breaker. The experimental model of fiber-controlled hybrid SF6 circuit breaker was analysed, the characteristics and theircircuit breaker with a vacuum circuit breaker and an SF6 circuit breaker interactions of vacuum arc and SF6 arc during the interruption processin series was set up, and the fault current breaking experiment was were studied;The hybrid circuit breaker simulation model was builtconducted on two circuit breakers in different collaborative action up by using Alternative Transients Program base on the vacuum arcmoment. The voltage distribution across the vacuum circuit breaker and and SF6 arc mathematical model, and the breaking test of hybrid circuitSF6 circuit breaker was analysed, the characteristics and their breaker model was simulated. The results of the experiment andinteractions of vacuum arc and SF6 arc during the interruption process simulation indicate that the vacuum circuit breaker covers the shortagewere studied;The hybrid circuit breaker simulation model was built of low dielectric recovery speed of SF6 interrupter in the initial periodup by using Alternative Transients Program base on the vacuum arc of dielectric strength recovery and withstand larger parts of very steepand SF6 arc mathematical model, and the breaking test of hybrid circuit rising transient recovery voltages, whereas the SF6 circuit breaker takesbreaker model was simulated. The results of the experiment and the advantage of high voltage endurances of SF6 and withstands largersimulation indicate that the vacuum circuit breaker covers the shortage parts of the transient recovery voltages in the peak period of dielectricof low dielectric recovery speed of SF6 interrupter in the initial period recovery voltage, and prove the validity of simulation model very well.of dielectric strength recovery and withstand larger parts of very steeprising transient recovery voltages, whereas the SF6 circuit breaker takes 1P-21: The Impact of the Shunt Compensation on Effectivethe advantage of high voltage endurances of SF6 and withstands largerparts of the transient recovery voltages in the peak period of dielectric and Reliable Power Transmissionrecovery voltage, and prove the validity of simulation model very well. Y. Dvorkin1, D. Rimorov2, S. Tuzhilov2, A. Mamishev1 1 University of Washington, Seattle, United States 2 Moscow Power Engineering Institute, Moscow, Russia1P-19: Simulation and Experimental Research on DynamicDielectric Recovery Characteirstics for Vacuum Switch with The reactive shunt compensation is a critical component of the powerDouble-Breaks system control; therefore, versatile efforts have been attempted toX. Cheng, M. Liao, X. Duan, J. Zou optimize its utilization in order to get maximal technological,School of Electrical Engineering, Dalian University of Technology, economical, and ecological benefits. Since, some objective practicalDalian, China concerns confine distribution of these devices; additional theoretical investigations, modeling, and computer simulations are still required.Based on the equivalent model of vacuum switch with double-breaks, This paper summarizes general methodologies to allocate reactivethe dynamic dielectric recovery characteirstics of the vacuum switch shunt compensators and provide the new approach to regulate reactivewith double-breaks were analysed; The simulation test platform of power output to maintain secure and stable modes of the power system.vacuum switch with double-breaks was build up by Alternative The proposed regulation rule provides ability to control shuntTransients Program, and the transient arc model of vacuum interrupter compensator mode to retain the optimal power flow, voltage andwas realized using transient analysis of control systems. The breaking current constraints of the power network equipment, and optimal-costfault current simulaton were operated under the conditions of two local generators output. Finally, authors sketch the principle algorithminterrupters contact separation at same moment or different moment for combined shunt compensator and generator control loop and showand two interrupters parallel same capacitance or different capacitance; topics for further development.The experimental model of vacuum switch with double-breaks in serieswas set up, and the dynamic dielectric recovery characteristics of 1P-22: Simulation Analysis of Transmission-Line Impedanceexperimental model were studied. The experimental results and Transformers with Gaussian, Exponential, and Linearsimulated data show that the successful breaking of vacuum switchwith double-breaks mostly due to the voltage average distribution of Impedance Profile for Pulsed Power Acceleratorstwo interrupters in the dielectric recover process. The voltage Y. Hu, A. Qiu, T. Huangdistribution between two interrupters is decided by their arc resistances Northwest Institute of Nuclear Technology, xian, Chinain the dynamic dielectric recover process, and they were decided by 49
  • 49. Based on transmission line code, 1-D circuit model for transmission- the key factors which effect the power and energy efficiency areline impedance transformer has been developed and the validities are discussed.proved with the existing literatures. Using this model, the power-transport efficiencies and pulse compression ratio of such transformers 1P-25: Development of the 1/4:7 165kV Fractional Turn(with Gaussian, Exponential, and Linear impedance profile) arequantified as functions of Q (the ratio of the output impedance to the Ratio Pulse Transformerinput impedance of the transformers) and P(the ratio of the pulse width H. Hu, K. Gan, Z. Tan, T. Li, H. Zhangto the one-way transit time of the transformers). For the cases Institutes of the Applied Electronics, China Academy of Engineeringconsidered, it is found that with the continuous increase of P (from zero Physic, Mianyang, Chinato infinity), the power-transport efficiency first decreases and thenalmost remains constant , and the pulse compression ratio has a Rise time of the fractional turn ratio pulse transformer is one of themaximum value. Moreover, simulation results indicate that very important parameters, we found its rise time is proportional to thetransformers with exponential impedance profile have the maximum square of the numbers of the second windings and voltage second timespower-transport efficiencies and pulse compression ratio. But with of the fractional turn ratio, and inverse to the square to the magnetich<0.05 (where h is the Gaussian coefficient), transformers with flux and equal stacking factor of the magnetic core, and then weGaussian profiles almost have the same power-transport efficiencies designed and fabricated the 1/4: 7 165kV pulse transformer to drive theand pulse compression ratio of those with exponential profiles. TH2163 Klystron. The inductance and stray capacitance are about 632uH and 187.6pF respectively, which were measured with the1P-23: Parallel Plate Transmission Line Transformer impedance analyzer. The low voltage experiments show that the riseS. J. Voeten1, S. Brussaard1, G. Pemen2 time of the 1/4:7 165kV pulse transformer is about 1us when the load is1 about 1500 Ohms. Applied Physics, Eindhoven University of Technology, Eindhoven,Netherlands2 Electrical Engineering, Eindhoven University of Technology, 1P-26: High-Voltage Picosecond Reflectometry inEindhoven, Netherlands Investigations of Dynamic Characteristics of Discharge GapsA Transmission Line transformer, TLT, can be used to transform high M. I. Yalandin, K. A. Sharypov, V. G. Shpak, S. A. Shunailov,voltage nanosecond pulses. These transformers rely on the fact that the A. G. Reutova, M. R. Ul’masculovlength of the pulse is shorter than the transmission lines used. This Laboratory of electron accelerators, Institute of Electrophysics, Uralgives the freedom to connect the transmission lines in parallel at the Branch of Russian Academy of Sciences, Ekaterinburg, Russianinput and in series at the output. In the ideal case such structures Federationachieve a voltage gain which equals the number of transmission linesused. To achieve maximum efficiency, mismatch and secondary modes Time intervals of breakdown initiation in highly overvolted gas-filledmust be suppressed. Here we describe a TLT based on parallel plate electrode gaps and transition of field emission into explosive electrontransmission lines. The chosen geometry results in a very high emission in vacuum e-beam diodes are referred to subnanosecond scale.efficiency, due to good matching and a minimized secondary mode. A Typical amplitudes of voltage pulses at the gaps (hereinafter, load) andsecond advantage of this design is that the electric field strength in discharge currents attain hundreds of kilovolts and kiloamperes,between the conductors is the same throughout the entire TLT. This respectively. Investigation of dynamic current-voltage characteristic ofmakes the design very suitable for high voltages applications. To the load requires synchronous, picosecond resolution registration ofinvestigate the concept of this TLT design, measurements are done with voltage at the gap and total discharge current or e-beam current.2 different TLTs. One TLT has 4 transmission lines and the other 8. Aiming above requirements, independent probes of current and voltageBoth are made of DiBond™. This material consists of a flat appear not applicable, while time-domain reflectometry could be used.polyethylene inner core with an aluminum sheet on both sides. Both A single voltage probe is sufficient to do this in a limited number ofTLT’s have an input impedance of 3.125 ohm. The output impedances cases. It should be placed in homogeneous transmitting line where aof the 4 and 8-line TLT are 50 and 200 ohm, respectively. The short, fast rise time voltage pulse runs to the load in travelling wavemeasurements show that, on a matched load, such a structure can regime. It is ideal, desirable requirement that reflected pulse to beachieve a voltage gain factor of 3.8 when using 4 transmission lines and delayed with respect to incidental one, and transmitting duct between7.6 when using 8 lines. the probe and the load should be free of reflecting inhomogeneities. Report presents examples of practical reflectometry measurements,1P-24: Analysis of Transmission Performance of the Radial when recorded reflection from static or dynamic load is added toImpedance Transformers chaotically variable, extended tail of incident pulse. We suggest theR. Zhang, X. Zou, X. Wang way to distinct (in some approaches) non-disturbed reflected signalDepartment of Electrical Engineering, Tsinghua University, Beijing, from the load by using two identical, specially designed voltage probes.China By compare such reflection with incident signal one can judge about dynamic changes in parameters of discharge gap. This task is close to the problem of short-pulsed radiolocation on recognizing of usefulThe radial impedance transformers used in high pulsed power device signal when presence of reflections from underlying surface. Acan efficiently transmit electric power, and transform voltage and principal distinct feature is that the getting of desirable informationcurrent between the input and output terminals. Based on analytic should be done at a single pulse (i), for a high voltage of both incidentmethod and numerical method, we analyse the relations of power and and reflected pulse (ii), and for chaotic nature of accompanyenergy efficiency with the frequency spectrum of the forward-going- background (iii). This work is supported by RFBR, Grant 11-08-00189voltage pulse being transformed, the impedance of the pulse generator and Grant 10-08-00081, and also within framework of the RASat the transformer input, and the impedance of the load at the Presidium project “The problems of physical electronics, beams oftransformer output. The results of three kinds of impedance profiles, charged particles, and generation of EM radiation in the powerfullinear mode, saturated mode and exponential mode, are compared, and systems”. 50
  • 50. 31P-27:Coaxial Capacitive Voltage Divider for High Voltage Bioelectrics Research Center, Kumamoto University, Kumamoto,Pulses with a Very Fast Rise Times JapanT. Hobejogi, J. BielaDept. of Electrical Engineering, Laboratory for High Power Electronic In recent years, the environmental reform by pulsed discharge isSystems, Zurich, Switzerland commanded attention. Pulsed discharge has been widely used for various fields, such as ozone generation, exhaust gas treatment,In many pulsed power applications, an accurate and fast voltage removal of volatile organic compounds. Pulsed streamer dischargemeasurement system is crucial for designing and investigating the pulse plasmas which are one of the non-thermal plasma have been used tomodulator so that it meets the required pulse parameters. With very fast generate ozone and to treat pollutant.rise/fall times of pulses (>1MV/µs), the measurement becomes more Recently, it is reported that a pulse width of applied voltage todifficult to make since, the parasitic capacitance/inductance discharge reactor has a strong influence on the energy efficiencies ofsignificantly affects the quality of the measurement and could result in ozone generation and pollutant removal, therefore, the development ofoscillations in the measured signal. Furthermore, the parasitic a short pulse generator is of paramount importance for energy efficientcapacitance can significantly influence the current/voltage distribution improvement. In our previous studies, a nano-seconds pulsed generatorin the pulsed power system at high dv/dt. which has 5 ns of pulse duration in output pulsed voltage was alreadyFor measuring such fast voltage pulses, often coaxial voltage dividers developed and its performances as non-thermal plasma wereare applied. These are simple, flexible as well as cost efficient and demonstrated. However, some physical characteristics such as electronenable measuring very fast rise transients. The dividers published in energy, electron temperature, discharge propagation and so on of nano-literature often use the capacitance of a coaxial cable and a second seconds pulsed streamer discharge are still unclear.dielectric layer plus copper foil around the coaxial cable to realize the In this work, the axial propagation of nano-seconds pulsed voltage andcapacitive voltage divider [1-3]. This result in a relatively simple setup, nano-seconds pulsed discharge in coaxial electrode would be observed.however, defining the capacitance value this way is not straightforward The results would give useful information for nano-seconds pulsedand may not result in the required division ratio. discharge plasma process.In this paper, a capacitive voltage divider based on a 200kV HVDCLight sea cable [4] with evenly distributed discrete capacitors mounted A System for Pulsed Measurements Based on 1P-29:around the cable is investigated. The cable has conductor cross-section LabVIEWarea of 1650mm2 and isolation thickness of 16mm. With this method S. Korenev, C. Dewdetermining the value of the second, low voltage capacitor is straight Caterpillar Inc, Mossville, IL, United Statesforward and the voltage divider is mechanically robust, having a largecross sectional area for conducting large pulse currents, as required inpulse modulators for plasma channel drilling [5]. Furthermore, no oil is This paper describes a simple and reliable system for the measurementrequired for isolation even at relatively high pulse voltages. and treatment of pulsed voltage and pulsed current signals. The systemFor the presented voltage divider a numerical model is presented, includes high voltage dividers, current transformers, an oscilloscope,which is used for the effects resulting in an upper frequency limit and and a laptop to run software. The high voltage divider based on seriespossible oscillations of the measurement signal. Additionally, based on high and low resistivity resistors from de-ionized water. The currentthis model the influence of a feed-through or a T-connection with transformers used represent standard devices. The process to measurematched impedance as well as of the interconnection between the pulsed parameters consists of timed linear gating the signals,voltage divider and the pulse modulator on the upper frequency limit is calculating average signal amplitude, graphically displaying data, andinvestigated. The presented results are validated by measurements storing data for later analysis. The acquisition and processing softwareusing a network analyzer and by comparisons with HF standard voltage is built in the LabVIEW environment. The results of evaluation of thisprobes. system are presented in this report.1. Kumar, R., Novac, B.M., Smith, I.R. & Larour, J. (2007), "Simple,non-invasive and wide-band current and voltage sensors for use with 1P-30: Multivariate Analysis of Pulsed Power Diagnostics oncoaxial cables", In Proc. 16th IEEE Int. Pulsed Power Conf. Volume 1, the 2.4MV, 1MA Zebra Z-Pinch Generatorpp. 486-489 V. Nalajala1, B. Le Galloudec1,2, R. Presura1, V. Ivanov1, V. Kantsyrev1,2. Lorusso, A., Nassisi, V. & Siciliano, M.V. (2008), "Fast capacitive N. Le Galloudec1, A. Astanovitskiy1, S. Batie1, A. Covington1probe for electromagnetic pulse diagnostic", Review of Scientific 1 Nevada Terawatt Facility/Dept of Physics, University of Nevada, Reno,Instruments. Vol. 79(6) Reno, NV, United States3. Edson, W.A. & Oetzel, G.N. (1981), "Capacitance voltage divider 2 National Ignition Facility, Lawrence Livermore National Laboratory,for high-voltage pulse measurement", Review of Scientific Instruments. Livermore, CA, United StatesVol. 52 Issue 4, pp. 604 - 606.4. http://www.abb.com, "Products & services > Product Guide > PowerCables and Cable Accessories > Underground Cables > HVDC-Light" Recently, an effort has been initiated to more thoroughly analyze and5. Lisitsyn, I.V., Inoue, H., Nishizawa, I., Katsuki, S. & Akiyama, H. track temporal changes in the electrical characteristics of the Zebra Z-(1998), "Breakdown and destruction of heterogeneous solid dielectrics pinch accelerator at the Nevada Terawatt Facility. These data will beby high voltage pulses", Journal of Applied Physics. Vol. 84(11), pp. processed using a multivariate analysis program that will interface with6262-6267 the existing data acquisition and control system. The results of these efforts will help to provide the technical and experimental teams with real time information about changes in machine characteristics as a1P-28: Axial Propagation of Nano-Seconds Pulsed Discharge function of time and changes in the target load. Moreover, long-termin Coaxial Reactor changes in the behavior of the machine will also be monitored to helpT. Hirota1, S. Okada1, D. Wang2, T. Namihira3, H. Akiyama1 determine more efficient maintenance schedules. At present, Zebra1 Graduate School of Science and Technology, Kumamoto University, electrical measurements along the transmission line include a lowKumamoto, Japan inductance current viewing resistor, voltage monitors for Marx and2 Priority Organization for Innovation and Excellence, Kumamoto Intermediate storage capacitors, and v-dot and b-dot devices throughoutUniversity, Kumamoto, Japan the system. In addition, two TG70, high voltage fast rise time, trigger 51
  • 51. generators are used to trigger the Marx and the intermediate storage 1P-33: X-Ray Diode Preparationswitch to discharge energy to the load. Precision, reproducibility, D. J. Henderson1, C. V. Mitton1, D. E. Good1, K. W. Hogge1,efficiency and reliability of these measurements are vital in assessing I. Molina1, R. A. Howe1, P. A. Flores1, K. D. McGillivray1,the validity of scientific experiments performed on these systems. W. M. Skarda1, S. S. Lutz1, D. S. Nelson2, E. C. Ormond2,Trigger generators play a major role in the performance and efficiency S. R. Cordova2, J. R. Smith3, T. J. Haines3, W. M. Wood3of pulsed power system as they affect the timing and magnitude of total 1 NSTec, Las Vegas, NV, United Statesoutput. Monitoring these parameters for every shot and evaluation and 2 Sandia, Albuquerque, NM, United Statescalibration of pulsed power measurements periodically is necessary to 3 LANL, Los Alamos, NM, United Statesassess triggering shift, jitter, power losses and reliability ofmeasurements over period of time. A rod pinch x-ray diode assembly culminates in a coaxialWork supported by DOE/NNSA under UNR grant DE-FC52- anode/cathode arrangement where a small anode rod extends through06NA27616 the aperture of a cathode plate. Shot-to-shot variation in rod placement, and thus x-ray source spot position, has potential to negatively affect radiographic image quality. Thus, how to both control and measure1P-31: Electro-Optic Kerr Effect Measurements of Intense anode rod tip displacement off the beam line-of-sight retical and alsoPulsed Electric Fields in Water anode rod tip extension (position) along the line-of-sight center lineF. Banakhr1, B. M. Novac1, I. R. Smith1, L. Pecastaing2, R. Ruscassie2, become salient issues relative to radiographic image set utility. ToA. de Ferron2, P. Pignolet2 address these issues, both hardware fabrication and x-ray diode1 assembly method were reviewed and additional control was introduced. Electronic and Electrical Engineering, Loughborough University,Loughborough, United Kingdom A photogrammetric system was developed to quantify anode rod tip2 position in situ. Computer models and mock-up assemblies with Laboratoire de Genie Electrique, Pay University, Pau Cedex 9,France precision fiducials were produced to validate this system. Results were that both anode rod tip displacement and anode rod tip extension were adequately controlled. Rod position was measured and met the requiredLoughborough University (UK) and Pau University (France) are specifications: (1) radial displacement < 0.25 mm, (2) axial placementcurrently developing innovative non–invasive pulsed electric field of +/- 0.25 mm. Implications of this work are that precision control andtechnologies for industrial food processing. An essential requirement of measurement of large scale components is achievable in a pulse powerthis research is the accurate measurement of intense fast transient system. Correlations with diode performance and radiography areelectric fields inside a warm liquid environment. For this purpose, it discussed.was decided to adopt the electro-optic Kerr effect in water, for which ________________________________published results unfortunately show extreme variability. The paper * This work was done by National Security Technologies, LLC, underwill therefore present results provided by reliable and accurate Kerr Contract No. DE-AC52-06NA25946 with the U.S. Department ofmeasurements, and obtained for a transient electric field strength Energy.exceeding 300 kV/cm and a water temperature up to 50°C. Relatedwork on the novel non-invasive technology will also be included. 1P-34: Cygnus X-Ray Pinhole Camera Measurements D. S. Nelson1, E. C. Ormond1, S. R. Cordova1, J. R. Smith2,1P-32: Preliminary Experiment on Electro-Optical M. J. Berninger3, D. E. Good3, M. D. Hansen3, D. J. Henderson3,Measurement of Electric Field on Insulator Surface K. W. Hogge3, S. S. Lutz3, C. V. Mitton3, I. Molina4W. Liu, H. Zhu, X. Zou, X. Wang 1 Sandia National Laboratories, Albuquerque,NM, United StatesDept. of Electrical Engineering, Tsinghua University, Beijing, China 2 Los Alamos National Laboratory, Los Alamos, NM, United States 3 National Security Technologies LLC, Las Vegas, NV, United States 4It is well known that the flashover voltage of an insulator between two Great Basin Technology, Inc., Albuquerque,NM, United Stateselectrodes is much lower than the breakdown voltage of the same gapwithout the insulator. The reduced value of the breakdown voltage is The Cygnus Dual Beam Radiographic Facility consists of two identicalgenerally attributed to the surface charging of the insulator and radiographic sources, each with a dose rating of 4 Rads at 1 m, and a 1resulting inter-electrode field modification. The electric field along an mm diameter spot size. The development of the rod pinch diode isinsulator surface in vacuum was measured using electro-optical responsible for the ability to meet these criteria. The rod pinch diode intechnique based on Kerr effect. The tested insulator is hollow with its a Cygnus machine uses a small diameter anode rod, which extendsouter surface immersed in nitrobenzene liquid and its inner surface in through a cathode aperture. Electrons born off the aperture edge canvacuum. Since the liquid is separated from the vacuum by a thin self-insulate and pinch onto the tip of the rod, creating an intense, smallinsulator wall of about 2mm, the electric field derived from the x-ray source. The Cygnus sources are utilized as the primary diagnosticmeasured Kerr effect in the electric stressed liquid close to the outer on numerous experiments which include high-value, single-shot events.surface is assumed to be the field in vacuum close to the inner surface. In such applications there is an emphasis on machine reliability, x-rayThe hollow insulator between two plane-parallel electrodes spaced by 1 reproducibility, and x-ray quality. We have observed that an additionalcm was stressed by a high voltage pulse, 100 ns in pulse width and 200 pinch occurs at the interface near the anode rod and the rod holder. This~ 300 kV in amplitude, that was produced with a pulse power generator suggests there are stray electrons emitted from the surfaces of theconsisting of a Marx bank, a pulsed forming line, a laser triggered surrounding area. In this paper we present results of x-rayswitch. A Q-switched YAG laser, 532nm in wavelength and 7 ns in measurements using pinhole cameras. Two camera geometries are used:pulse width, was used as light source to detect the Kerr effect in the (1) a side on view (i.e. 90° with respect to the diode centerline), and (2)stressed nitrobenzene liquid. After passing through the liquid, the laser an upstream view (i.e. 37° with respect to the diode centerline). Thesebeam was analyzed with a Mach-Zehnder interferometer for diagnostics will be employed to: (1) diagnose x-ray reproducibility anddetermining the electric field. quality, and (2) investigate the effect of different diode configurations. ________________________________ * Sandia National Laboratories is a multi-program laboratory managed and operated by Sandia Corporation, a wholly owned subsidiary of 52
  • 52. Lockheed Martin Corporation, for the U.S. Department of Energys dose due to impedance collapse. Understanding the nature of theseNational Nuclear Security Administration under contract DE-AC04- plasmas is important in modeling their behavior and providing insight94AL85000. into their mitigation. In this paper we describe a streak camera–based optical diagnostic that is capable of observing and measuring plasma evolution within the A-K gap. By imaging a region of interest onto the1P-35: Virtual Plasma Diagnostic Tool input slit of streak camera, we are able to produce a time-resolved one-A. Eroglu dimensional image of the evolving plasma. Typical data are presented.Engineering Department, Indiana University - Purdue University FortWayne, Fort Wayne, IN, United States 1P-37: Infrared Imaging Diagnostics for Parameters ofInnovative diagnostic techniques are gateways to breakthrough in Powerful Ion Beams Formed by a Diode in a Double-Pulseplasma systems. Accurate plasma diagnostic tools are required to assess Modeplasma dynamics and provide necessary control mechanism to attain Y. Isakovathe desired performance. However, plasma diagnostic is a very Tomsk Polytechnic University, Tomsk, Russian Federationchallenging and vivid discipline. The biggest challenge for havingaccurate diagnostic tools is due to narrowband analysis of critical Infrared imaging diagnostics on the parameters of pulsed ion beams ofparameters using existing diagnostic methods. gigawatt power [1] is an effective method for rapid control. It allowsDiagnostic tools can be used to measure the plasma critical parameters for the measuring of energy density distribution at the target,for limited conditions. The limited number of data points collected for optimization of an ion diode operation and control of the mode of targetthe conditions defined lead to erroneous results or results that are not irradiation. A standard thermal imager provides registration of auseful for application. The diagnostic tools are time consuming and not thermal imprint per pulse at energy densities above 0.1 J/cm2. The ioncost effective when specifically the measurement conditions and diode with a passive anode operates in a double-pulse mode. During theparameters are not well defined. The results then cannot be used for first (negative) pulse an explosive emission plasma forms on thesolution of any challenging problems. Some of the plasma critical potential electrode surface. During the second (positive) pulse aparameters that are measured for diagnostic purposes using different powerful ion beam is generated and accelerated from the plasma. For adiagnostic methods are: ion density, electron density, ion temperature, correct application of thermal imaging diagnostics for a diode operatedelectron temperature, current density, space potential. The existing in a double pulse mode it is necessary to evaluate the contribution ofdiagnostic methods that are used to study the plasma critical parameters the accelerated electrons (in the first pulse) which heat the target. Thecan be categorized as passive diagnostics, or active diagnostics study was performed at the TEMP-4M accelerator with the followingmethods. None of the existing diagnostic methods, regardless of what parameters: the first pulse is negative (80 ns, 100-150 kV), followed bycritical parameter is being measured, study complete plasma system positive pulse (80 ns, 250-300 kV) with an adjustable pause (300-600and its interaction with plasma interfacing components using modeling ns) between pulses. The ion current density is 30-300 cm2 (for differentand simulation. types of diodes), the ion energy is 250-300 keV, and the beam isIn this paper, virtual plasma diagnostic tool to study thoroughly the comprised of protons and carbon ions [2]. Ion current density wascomplete system and its interaction with plasma interfacing measured by a Faraday cup with magnetic cutoff, the total current wascomponents is developed using modeling and simulation in time- measured by a Rogowski coil. The electron current in the ion beamdomain. The complete model includes plasma load, RF generator, transport region was recorded by a Faraday cup without magneticmatching networks, transmission lines, high power non-directional and cutoff. A target (brass, 220 µm) was placed at different distances fromdirectional couplers for signal pick up for diagnostics. The models at the diode and, by measuring the distribution of the temperature field,the component level gives ability to understand interaction between we estimated a contribution of the electrons to the heat of the target.plasma interfacing components when they are integrated to obtain Similar measurements of non-magnetized electrons left to the region ofcomplete plasma system. The modeling and simulation have been the beam transport were performed using a Faraday cup. This paperperformed using time-domain simulator. The complete system with its also presents the results of homogeneity over the cross section of an ioninterfacing components is simulated for various conditions to measure beam formed by a planar diode with a passive anode. The reasons fordifferent critical plasma parameters. One of the biggest advantages of increasing the ion current energy density at the end of the diode (near athe developed virtual plasma environment is the ability to study the grounding point) have been researched. Studies have shown that theplasma dynamics against several different conditions in a single change of potential along the length of the grounded electrode stronglycomplete model, which can be simulated to get the desired information. depends on an induction of the grounded electrode, related to a rate of the current change dI/dt and a constructive inductance of grounded1P-36: An Optical Streak Camera Plasma Diagnostic for electrode U=L•dI/dt. The grounded electrode potential reverses itsRadiographic Source Development polarity when passing through a maximum current of the diode. On theM. D. Crain1, S. L. Payne1, D. W. Droemer1, M. D. Johnston2 first negative voltage pulse, which is plasma-forming, the potential of1 the grounded electrode does not introduce significant changes to the Sandia Operations, National Security Technologies, LLC, operation of the diode. In the second (positive) pulse the acceleratingAlbuquerque, NM, United States2 voltage is largely determined by the nature of potential change at the Dept 1656, Sandia National Laboratories, Albuquerque, NM, United grounded electrode.States References: 1. H.A. Davis et al. J. Appl. Phys., Vol. 82, No. 7, 3223 (1997)National Security Technologies, LLC is collaborating with Sandia 2. A. Pushkarev, Y. Isakova and D. Vahrushev Physics of Plasmas 17,National Laboratories to develop pulsed power–driven flash x-ray 123112 (2010).radiographic sources that utilize high-intensity electron beam diodes onthe RITS 6 (Radiographic Integrated Test Stand) accelerator. The highelectric fields and current densities associated with these short A-K gappinch beam diodes present many challenges in diode development.Plasmas generated both at the anode and cathode can affect diodeperformance and manifest in varying spot (source) sizes and decreasing 53
  • 53. 1P: Microwaves posters I: Sources and Antennae, Slow model, as well as allowing for user defined secondary electron yieldWave Devices, Systems curves. We compare VORPAL results using some of these techniques with accepted analytical models for multipacting in a coaxial cable.Monday, June 20 13:30-15:30 Regency Ballroom 1P-41: Electric Field Distributions in High Power Nonlinear Dielectric-Based NLTL Modeling Using1P-38:ICEPIC Microwaves Confined by Plasma ColumnB. W. Hoff1, A. D. Greenwood1, S. L. Heidger1, D. M. French1, S. S. M. ChungJ. Watrous2 Electronics Engineering, Southern Taiwan University of Technology,1 Air Force Research Laboratory, Albuquerque, NM, United States Tainan, Taiwan2 NumerEx, Albuquerque, NM, United States High Power Microwave (HPM) sources has advanced to GW levelResearch is underway to develop nonlinear dielectric models for the today, however the delivery of sufficient energy to desired target is stillfinite difference time domain (FDTD) particle in cell (PIC) code, limited by pulse shortening, and large antenna is needed to achieveICEPIC. This modeling effort is being pursued in support of nonlinear small beam width for long distance transmission. Confinement ofdielectric-based nonlinear transmission line (NLTL) work currently in electromagnetic wave via laser created plasma columns represent a newprogress at the Air Force Research Laboratory. The initial modeling thinking, in which air is ionized by powerful lasers into plasma column,efforts are focused both on shock lines and NLTLs utilizing two and electromagnetic wave travels within them much like light in acommon nonlinear dielectrics, Barium Strontium Titanate (BST) and photonic crystal fiber, thus relieves the system from large reflectorLead Manganese Niobate (PMN). Simulation and experimental results antenna used to concentrate the beam. We try to simulate this scenariowill be presented. with FDTD method, and explore the relationships between plasma column numbers, plasma density, to confinement of HPM electric field. Higher number of plasma column and plasma density increases1P-39: Dielectric Nonlinear Transmission Line confinement efficiency, at the trade off of more energy input to theD. M. French, B. W. Hoff, S. Heidger, D. Shiffler laser generation.High Power Microwave Division, Directed Energy Directorate, AirForce Research Laboratory, Albuquerque, NM, United States 1P-42: Modular, High-Power, Wideband Transmitters for Electromagnetic Environmental Effects (E3) TestingA parallel plate nonlinear transmission line (NLTL) was constructed. T. A. Holt, M. B. Lara, C. Nunnally, C. W. Hatfield, J. R. MayesPeriodic loading of PMN-38 [1] nonlinear dielectric slabs provides the APELC, Austin, TX, United Statesa nonlinear capacitance and the gaps between provide linear inductiveinterconnects, this is essentially the same design used by Ikezi [2],[3].The NLTL was modeled in a circuit simulation code using the Applied Physical Electronics, L. C., (APELC) has developed a suite ofexperimentally measured form of the nonlinear capacitance. Dielectric high-power, wideband, dipole antennas targeted for use by the test andloss included in the model as an equivalent series resistance derived evaluation and the directed energy communities. Four dipoles withfrom the measured loss tangent data affects the formation of RF center frequencies of 60, 100, 250, and 400 MHz have been developed,oscillations. The diagnostics used on the experimental system are Bdots manufactured, and used to support testing for various customers. Thealong the line and a current viewing resistor at the load. The diagnostics suite of dipoles was developed to address the new MIL-STD 464 C andprovide a measurement of the pulse evolution as it travels down the line. all of the dipoles can be sourced by the APELC MG15-3C-940PF MarxThe waveforms from the experimental line qualitatively agree with the generator. The dipoles feature corner reflectors to increase theircircuit model, showing no strong RF formation as a result of the loss. directivity, which act to reduce side lobe levels. Additionally, a[1] TRS Technologies, Inc., http://www.trstechnologies.com/. common, proprietary connector is implemented into the design of each[2] H. Ikezi, J.S. DeGrassie, and J. Drake, “Soliton generation at 10 dipole to facilitate the interchange of dipoles during deployment orMW level in the very high frequency band,” Applied Physics Letters, testing scenarios. The dipoles manufactured to date vary in size fromvol. 58, 1991, p. 986. 20 cm in diameter and 245 cm in height to 17 cm in diameter and 37[3] H. Ikezi, S.S. Wojtowicz, R.E. Waltz, and D.R. Baker, “Temporal cm in height for the 60-MHz and 400-MHz dipoles, respectively. Thecontraction of solitons in a nonuniform transmission line,” Journal of average amplitude of the peak electric field measured 110 kV/m andApplied Physics, vol. 64, Dec. 1988, pp. 6836-6838. 230 kV/m for the 60-MHz and 400-MHz dipoles, respectively (electric field strengths normalized to 1 meter from source). Temporal and spectral data will be presented for each dipole offered and methods for1P-40: Multipacting Simulations in a Coaxial Transmission obtaining higher peak electric field amplitudes will be discussed.Line with VORPALC. M. Roark, C. Nieter, P. H. Stoltz 1P-43: A Multiple Burst, Variable Frequency, High PowerTech-X Corporation, Boulder, CO, United States Driver for Antenna Characterization K. A. OConnor, R. D. CurryMultipacting occurs when an oscillating electromagetic field in a cavity Dept. of Electrical & Computer Engineering, University of Missouri,or waveguide resonates with the periodic motion of electrons traveling Columbia, MO, United Statesbetween and colliding with the walls, leading to an exponential increasein secondary electrons. Multipacting is often the factor that limits theperformance of high-power RF devices. Numerical simulations can Compact dielectric loaded, high power antennas that are substantiallyprovide insight into the effectiveness of potential mitigation techniques. smaller than conventional antennas are actively being developed at theIn order to test these techniques, numerous simulations are required University of Missouri-Columbia. The antennas are characterized usingover an extended parameter space. VORPAL has a new method to scan network analysis techniques at low power. As such, high powerover multiple powers during one simulation by scaling the value of the evaluation is also necessary to test operation at high voltage. Thus, afields felt by each particle. The secondary emission models in high power driver was required to evaluate antennas in two ways thatVORPAL include a simple one-in-one-out model, the Furman-Pivi cannot be accomplished at low power. First, high power evaluation 54
  • 54. enables diagnosis of potential points of insulation failure due to solid impulse excited helical antennas. Those efforts have been continued todielectric breakdown, flash over, or corona formation in the antenna result in a family of interchangeable antennas sourced by a commonstructure. Second, the peak radiated fields can be directly measured compact pulse power source. This paper describes the system,when the antenna is driven at 100’s of MW or high power. including the pulse power and the helical antenna loads, supported byThe high power antenna driver was designed for flexibility in the peak simulations and experimental results.output voltage, the driving frequency, and operation as a single ormultiple burst source. An inductive energy storage system, with an 1P-46: Compact Relativistic Magnetron with Gaussianexploding wire fuse, is utilized to achieve the high peak power levelsrequired for testing. The current profile of the capacitive discharge Radiation Patternfeeding the inductive energy storage system can be modified from that S. Prasad, M. I. Fuks, E. Schamilogluof a simple capacitive discharge to that typical of an explosive Electrical and Computer Engineering, University of New Mexico,generator. Upon vaporization of the exploding wire fuse, the inductive Albuquerque, NM, United Statesenergy storage system charges a high voltage capacitor, which isshorted at peak voltage to form an oscillator in the VHF to UHF bands. We present a compact relativistic magnetron with a simple modeThe antenna under test is driven in parallel with the high voltage converter that radiates TE_11 mode when the operating mode of thecapacitor and low inductance shunt. The peak charging voltage and magnetron is the pi-mode. The magnetron used in this study has theoutput power can be modified by adjustment of the inductive energy standard A6 magnetron [1] dimensions except that the microwaves arestore, the exploding wire fuse action integral, and the energy delivered extracted axially from a cylindrical waveguide whose radius is theto the components. The frequencies at which the antenna is driven can same as the anode vane. The proposed mode converter implements anbe tuned by modifying the parameters of the high voltage capacitor and anode endcap that electrically opens those cavities whose electric fieldlow inductance shunt. Lastly, the driver can operate in single or distribution corresponds to the polarization of the TE_11 mode andmultiple bursts by modifying the impedance profile of the exploding closes the rest. This configuration has several advantages: 1) Thewire fuse. A detailed description of the high power antenna driver is TE_11 output mode can be easily converted into a Gaussian-like waveprovided, and descriptions of the critical component designs are given. beam by means of a conical horn antenna, 2) Axial extraction ofExperimental results of the driver’s performance are presented and microwaves in this manner allows one to use a compact solenoidcompared with a theoretical model. mounted directly onto the magnetron tube, thereby requiring a smallThis work was supported by the Office of Naval Research under volume for the applied axial magnetic field and reducing the energycontract N00014-08-1-0267. requirement for magnetic field-producing system, 3) The rapidly diverging magnetic field lines from the solenoid enables a quick1P-44: Pulsed Ring down Source Array Steering deposition of the downstream leakage electrons onto the waveguideA. W. Myers, S. Holt, J. Dickens wall, thereby protecting the dielectric window of the radiating horn antenna from electron bombardment [2]. Simulation results using theCenter for Pulsed Power and Power Electronics, Texas Tech MAGIC Particle-In-Cell (PIC) code will be presented.University, Lubbock, Texas, United States 1. G. Bekefi and T.J. Orzechowski, “Giant Microwave-Bursts Emitted from a Field-Emission, Relativistic Electron-Beam Magnetron,” Phys.A pulsed ring-down source (PRDS) is a robust, high-power, mesoband Rev. Lett., vol. 37, pp. 379-382, 1976.RF transmitter that can provide substantial power in the far field region. 2. C. Leach, M. Fuks, S. Prasad, and E. Schamiloglu, “Suppression ofIn a typical PRDS, energy is stored in the structure of the device and Leakage Current in a Relativistic Magnetron Using Various Endcapdischarged through the main switch, resulting in damped oscillations Designs,” (this conference).which are radiated from the structure. When multiple PRDS antennasare implemented in a phased array, the radiated far field power density 1P-47: Study of a Novel Compact P-Band Magneticallyideally increases with the square of the number of elements in the array.The arrays performance is dependent on the physical structure of Insulated Transmission Line Oscillatorindividual antenna elements, and their switching characteristics. The X. P. Zhang, C. W. Yuan, T. Wang, H. M. Ren, L. R. Xu, Z. Q. Hongability to control steering and the radiation pattern emitted from the College of Optoelectric Science and Engineering, National Universityarray is dependent on the precision of the system timing, i.e. the jitter of Defense Technology, Changsha, Hunan, Chinaof the spark gap switch and the jitter of the trigger sources. The arrayused in testing is composed of four antennas spaced one meter apart A novel compact P-band magnetically insulated transmission linefrom one another in a straight line. Each PRDS element is a half oscillator (MILO) has been designed, manufactured and tested. Acoaxial monopole antenna with the switch located at the base of the special inclined configuration is designed in the slow wave structurestructure. The system is tuned to radiate at ~100MHz. The switch used vanes of the novel P-band MILO, and thus makes the MILO moreis a custom designed trigatron spark gap. The results presented are compact. In recent experiment, the novel compact P-band MILO isfocused on the controllability of the array and the ability to steer the typically operated in single-shot mode with the beam parameters offocal point of the radiated waveform. 650kV and 45kA, generating about 2.6GW high power microwave pulses at 660MHz and yielding a conversion efficiency of about 8.9%.1P-45: Modular Interchangeable High Power Helical The output powers are radiated through the extractor coax, where theyAntennas propagate in the TEM mode. A short dipole antenna operated at P-bandM. B. Lara, M. G. Mayes, W. C. Nunnally, T. A. Holt, J. R. Mayes is designed for measuring the P-band high power microwave.Applied Physical Electronics, L.C., Austin, Texas, United States 1P-48: RF Input for Sectioned Relativistic AmplifiersHelical antennas are very appealing for their conformal geometry and M. Liu1, M. I. Fuks1, E. Schamiloglu1, C.-L. Liu2 1relatively small geometries. Moreover, helical antennas can be impulse Department of Electrical and Computer Engineering, University ofexcited, producing several cycles of resonant energy at its designed New Mexico, Albuquerque, NM, United States 2frequency, or resonantly-driven by a frequency matched resonator. Key Laboratory of Physical Electronics and Devices of the Ministry ofApplied Physical Electronics, L.C. has previous reported results from Education, Xi’an Jiaotong University, Xian, China 55
  • 55. As a rule, the injection of the initial RF signal into Cherenkov sectioned Designed permanent magnets take into consideration a specificity ofamplifiers driven by relativistic electron beams is achieved through the options where solenoidal profile of B-field is required in the BWOsmode converters that transform the TE_10 mode of rectangular working zone, which includes the cathode electrode, slow-wave system,waveguide (RW) to the rotating TE_m1 mode of cylindrical waveguide and e-beam collector. The problem of B-field reverse shift out of this(CW), which is the operating mode of the first section [1, 2]. The zone has been solved. Electric field at the shield of the cathodedesign of such mode converters is very simple: a RW is perpendicularly screening electrode was minimized. The probability of injection ofconnected to a CW so that its upper wide wall is tangential to the backward e-beam current was decreased by special shaping of B-fieldgeneratrix of the CW and the radiating aperture of the RW in the line starting from emissive edge of the cathode. Magnetic fieldgeneratrix (the RW does not enter into the CW). The cross section of interpolation of the permanent magnets was used in numerical modelsCW coinciding with one narrow wall of the RW is a reflector to the of microwave devices (code KARAT). Particle-in-cell simulations haveTE_m1 mode whose radius is less than the radius for cut-off. The idea shown that in the frequency ranges of 8 GHz, and 37 GHz, and for theof such converters is to illuminate a caustic surface of the desired mode B-field below cyclotron resonance value, the same parameters of HPMof the CW. In order to find the basic dimensions of the converter an generation are available; it validate prototype models with solenoidsasymptotic procedure, namely the method of geometrical optics [3], is fed by DC or pulsed current. In the prospect, developed constructionsused. Within the framework of this method the TE_m1 mode is of BWOs equipped with highly-coercivity permanent magnets canrepresented as rays successfully being reflected from the wall of the enhance essentially autonomy of HPM generators built around of high-CW (the Brillouin-Keller concept [4]) filling the space between the current electron accelerators type SINUS, RADAN, and ones based oncylindrical wall and the caustic surface, where the phase velocity of the SOS modulators. Besides, a “stable source” of focusing B-field couldrays is equal to the speed of light. It is assumed that the field is absent be an important component of multichannel BWO’s providing coherentwithin the caustic radius. The higher the TE_m1 mode, the more summation of the radiation.appropriate is the method of geometrical optics as occurs, for example, This work was supported by the French MOD (DGA) and CEA Gramatin describing TE_mn whispering gallery modes (with large azimuthal through a research contract involving ITHPP (Thegra, France) andindex m) that often are used as operating modes in gyrotrons, in which HCEI SB RAS (Tomsk, Russia), by integrated research projectsa 99% conversion coefficient is achieved for a fixed frequency [3]. between HCEI SB RAS, IEP UB RAS, and IMP UB RAS, and by RFBR Grant 11-02-00097.In computer simulations using the HFSS code we show that such asimple geometrical consideration for mode converters leads to similar 1P-50: A Novel Compact P-Band Coaxial Relativistichigh conversion coefficients even for small azimuthal indices m > 1 for Backward Wave Oscillator with Only Three Periods Slowfixed frequencies. However, such a consideration was also applied to Wave Structureconverters that are used in amplifiers for m = 2 [1] and m = 4 [2] when B.-L. Qian, L. Gao, X.-J. Geit is required to provide high conversion coefficient over a wide College of Optoelectric Science and Engineering, National Universityfrequency region. Significant corrections to the dimensions of the mode of Defense Technology, Changsha, Hunan 410073, Chinaconverter have been found experimentally [1, 2]. In this present workwe found regular corrections to optimal parameters for such convertersthat provide conversion coefficients ranging from 70–90% for modes A novel P-band coaxial relativistic backward wave oscillator (BWO)with different m. with only three periods slow wave structure is investigated both1. A.B. Volkov, N.I. Zaitsev, E.V. Ilyakov, et al., Sov. Tech. Phys. theoretically and numerically. The characteristics of the coaxial slowLett., vol. 18, pp. 6-10 (1992). wave structure (SWS) are analyzed when the SWS is changed from the2. E.B. Abubakirov, A.N. Denisenko, M.I. Fuks, et al., IEEE Trans. structure with only outer conductor ripple to the structure with bothPlasma Sci., vol. 30, pp. 1041-1052 (2002). inner and outer conductor ripples. It is found that the existence of the3. S.N. Vlasov, L.I. Zagryadskaya, and M.I. Petelin, Radio Engr. and inner conductor ripple can reduce the period length of coaxial SWS toElectronic Phys., vol. 20, pp. 14-17 (1975). maintain the same operating frequency of the BWO and can largely4. L.A. Vainstein, "Open Resonators and Open Waveguides" (Golem, increase the temporal growth rate and the spatial growth rate of theDenver, Colorado, 1969). device. Using this conclusion, a novel compact P-band relativistic1P-49: Simulated Parameters of Subgigawatt Relativistic BWO is optimized and designed by employing the 2.5D fullyBWOs with Permanent Magnetic Systems electromagnetic particle-in-cell code, KARAT (V. P. Tarakanov,V. V. Rostov1, A. V. Gunin1, E. M. Tot’meninov1, K. A. Sharypov2, Berkeley Research Associates, Inc., 1992). Typical simulation resultsV. G. Shpak2, M. I. Yalandin2, A. E. Yermakov3, S. V. Zhakov3, show that, with a 584 kV, 7.85 kA electron beam guided by a 0.8TG. Demol4, R. Vezinet5 solenoidal field, the microwave of 1.65 GW is generated at the1 frequency of 900 MHz, and the interaction efficiency is about 36%. SB RAS, High Current Electronics Institute, Tomsk, Russia2 Compared with the conventional P-band coaxial RBWO with five UB RAS, Institute of Electrophysics, Ekaterinburg, Russia3 periods SWS, the axial length of the SWS is reduced by about one half, UB RAS, Institute of Metal Physics, Ekaterinburg, Russia4 which is only 38.4cm, and the saturation time of the microwave signal ITHPP F-46500, Thegra, France5 is reduced by about 10ns CEA/DAM GRAMAT F-46500, Gramat, France 1P-51: High-Power Surface Field W-Band CherenkovReport presents the results of calculations and numerical modeling ofquasistationary, nanosecond relativistic Ka-band and X-band BWOs Oscillatorwith peak power exceeding 100 MW where magnetic field guided I. V. Konoplev, A. Phipps, A. W. Cross, C. W. Robertson, A. R. Young,electron beam (0.5-1.5 T) could be created by permanent magnets made C. G. Whyte, A. D. R. Phelpsof highly-coercivity materials. The weight of such permanent magnets Department of Physics, University of Strathclyde, Glasgow, United(~100 kg) exceeds somewhat one for equivalent dc solenoids. However, Kingdomdue to the absence of capacitive energy storages and/or rectifiers andpower switches, as well as systems of forced oil cooling of solenoid, High power (HP) W-band oscillators are very attractive for differentthe total weight and sized characteristics of the permanent magnet applications including: active remote sensing and detection inlooks preferable in the case of long time or continuous operation. atmosphere and space, plasma diagnostics and energy transfer. The 56
  • 56. oscillators that are capable of producing the required output power at carrier signal and its harmonics. The multi-drive frequency feature isthis frequency range are presently not broadly available. One of the under development. The main purpose of this code is to aid in thereasons is the necessity to use a high-voltage (above 300kV), high- preliminary design steps of TWT for communications and be used as acurrent electron beam inside an interaction region of limited teaching/learning tool for basic TWT physics as well.dimensions. However, an accessible and compact oscillator able toproduce output pulse-power above 1MW in this frequency range could 1P-53:A Four-Stage Depressed Collector Biasing Voltagesbe beneficial for many applications. One of the ways to achieve thisgoal and to observe oscillators able to produce such a high power is to Study Using the XMGUN Codeuse artificial materials, which have tailored properties optimised for C. C. Xavier, C. C. Mottahigh power operation. The use of the conventional metamaterials in HP University of Sao Paulo - USP, Sao Paulo, SP, Brazilapplications meet a number of challenges such as: coherence of thescattered radiation from the metamaterial; induced surface charge Traveling-Wave Tubes (TWTs), high efficiency vacuum microwavedissipation; metamaterial overheating and field enhancement at a devices, are widely used in radars, ground station transmitters and“single element” – “surrounding material” interface. To overcome satellite communications. Multi-stage Depressed Collectors (MDC)some of the challenges the use of two-dimensional (2D) cylindrical play an important role in overall efficiency in a TWT. When thesurface periodic structures is suggested [1,2], in contrast to the use of collector stages operate at negative potentials, relative to the interactionone dimensional (1D) periodic structures [3,4]. In this work we present structure, a continuous deceleration of the electron beam occurs, andthe results of the studies of a W-band HP Surface Field Cherenkov part of their energy is recovered by the circuit. Whenever an electronoscillator [1,2,5-7]. We discuss the design of the oscillator and its hits the collector surface, a secondary electron emission occurs. Thisparameters. We show that single mode, steady-state operation is emission may cause two major problems: (a) electron backstream to thepossible to achieve if a Cherenkov oscillator based on a Surface interaction structure, heating the circuit and producing noise, and (b)Periodic Lattice (SPL) is driven by a relatively low (250kV) voltage yields a current flow between the collector electrodes, also heatingand current (up to 200A) electron beam. We demonstrate that such an them up. Computational tools are extensively used to design andoscillator is capable of producing MW-level output power. The cavity optimize MDC, because they aid one to understand the primary andbased on a 2D SPL (manufactured using an electroforming method by secondary electron emission phenomena. In order to improve MDCgrowing copper on the surface of an aluminium mandrel and dissolving efficiency, the device designer may modify either the shape or thethe aluminium former in an alkaline solution [6]) is also presented and biasing voltage of the collector electrodes. XMGUN Suite, an axis-its property discussed. We will also discuss the scalability of the device symmetric, non-relativistic particle-path FEM code, developed toconsidered from 80GHz to 200GHz. design electron guns with grids and shadow-grids, was enhanced in1. L. Fisher, I.V. Konoplev, A.W. Cross, A.D.R. Phelps, 2007 IEEE order to incorporate secondary emission. In this work, a four-stagePulsed Power Conference, Albuquerque, USA, (2007). MDC behavior, with specified geometry, was investigated using the2. I.V. Konoplev, L. Fisher, A.W. Cross, A.D.R. Phelps, K. Ronald, XMGUN. Biasing electrodes to 9.45 kV, 8.65 kV, 6.45 kV, and 3.45IET Conf. Pub. 2009, O22, London, UK, (2009). kV, respectively, the simulations were run considering a 10% of the3. I.V. Konoplev, P. McGrane, A.D.R. Phelps, A.W. Cross, K. Ronald, biasing variation. A total of 30 keV, 264 primary particles and threeAppl. Phys. Lett., 87, 121104, (2005). generations of secondary emissions were observed and the best trade-4. I.V. Konoplev, P. McGrane, A.W. Cross, K. Ronald, A.D.R. Phelps, off is presented.J. Appl. Phys., 97, 073101, (2005).5. I.V. Konoplev, L. Fisher, A.W. Cross, A.D.R. Phelps, K. Ronald, 1P-54: High Power Microwave Generation from KALI 5000C.W. Robertson, Appl. Phys. Lett., 96, 261101, (2010).6. I.V. Konoplev, L. Fisher, K. Ronald, A.W. Cross, A.D.R. Phelps, Pulse Power Systemand C.W. Robertson, M. Thumm, Appl. Phys. Lett., 96, 231111, (2010). A. Roy, R. K. Menon, S. Mitra, S. Kumar, V. Sharma, A. Sharma,7. I. V. Konoplev, L. Fisher, A. W. Cross, A. D. R. Phelps, K. Ronald, K. V. Nagesh, D. P. ChakravorthyM. Thumm, Appl. Phys. Lett., 97, 261102, (2010). Accelerator and Pulse Power Division, Bhabha Atomic Research Centre, Mumbai, India1P-52: A 1D Large Signal Time-Domain Code for TWTsD. T. Lopes1, C. C. Motta2 Experiments were carried out to generate High Power Microwaves1 Instituto de Pesquisas Energeticas e Nucleares, Sao Paulo, SP, Brazil (HPM) from KALI 5000 pulse power system using an axial and a2 University of Sao Paulo - USP, Sao Paulo, SP, Brazil coaxial virtual cathode oscillator (vircator). The KALI 5000 pulse power system is a Marx generator and Blumlein line based system capable of operation at 1 MV, 50 kA, 100 ns. The typical electron beamIn previous PPC, we presented a 1D frequency domain code for TWTs parameters were 400 kV, 20 kA, and 100 ns, with a current density of aand used it to verify the feasibility of a high-power (~1kW) device for few hundreds of amperes per square centimeter. It was found that theKu band. That code, as others 1D frequency domain codes for TWTs, generated HPM power is almost negligible in an axial vircatorwas based on some simplification assumptions that now we removed in configuration due to severe impedance collapse problem. Impedancea new time-domain version of the same code. The main removed collapse in a planar diode due to fast cathode and anode plasmasimplification states that the e-beam is surrounded by a neutralizer ion expansion and bipolar flow across the diode gap. However, HPM couldbackground. Therefore, we now consider both the ac and the dc space be generated in a coaxial vircator configuration because of the fact thatcharge forces in the e-beam. Additionally, the early knowledge of the the electrode plasma expansion velocity of the cylindrical diode isspace-charge reduction factor is no longer needed. The input data much smaller as compared with the planar diode for the sameconsist in the cold circuit parameters as a function of the axial position, accelerating gap and diode voltage. Therefore, much higher voltage cannamely, the phase velocity, the interaction impedance, the dielectric be obtained for the cylindrical diodes as compared with the planarand the conductivity losses, and the input VSWR profile. This paper diodes for the same accelerating gap. The measured HPM powerwill describe the development of such code and present its results density and the electric field strength at antenna mouth are (4.3 mcompared to the previous frequency domain code. We intend to get distance from the Vircator window) 104.5 kW/m2 and 6.27 kV/msome experimental data by the conference date in order to validate the respectively. Frequency of the HPM signal was 2.9 GHz and a time-code. Up to date the code is capable to describe the amplifying of a 57
  • 57. dependent frequency analysis shows that the frequency remains around the entire perimeter of the facility. This produced a six footconstant in time for almost entire pulse duration. thick wall of concrete which absorbs any ionizing radiation produced during the utilization of the 4-8 MeV accelerator or the HPM sources. Construction of the facility was complicated by the fact that it was1P-55: Development of the Microwave Test Facility at the being built inside of the current laboratory. The progression ofUniversity of Missouri Center for Physical and Power construction had to be carefully planned so as to avoid disturbing theElectronics integrity of the preexisting structure. The building constraints alsoS. R. Ashby, R. L. Druce, M. B. Young, R. D. Curry required foresight so as to not inhibit the ability to place blocks intoDept. of Electrical & Computer Engineering, University of Missouri, specified locations later on in the construction process. The interior ofColumbia, MO, United States the shielding structure includes an overhead lift capable of hoisting objects in excess of 1 ton and maneuvering the components anywhereIn 2009 the Center for Physical and Power Electronics at the University within the interior of the structure. The radiation facility utilizes a highof Missouri-Columbia began to construct a facility to enable scaled velocity exhaust system with the ability to displace air with a laminartests using both low power and high power, pulsed and continuous flow to prevent ozone buildup. The exhaust intakes were placed at floor,radio frequency, and microwave energy. This facility has been online mid-wall, and ceiling heights in order to remove ozone or byproductssince early 2010 and has been used to perform advanced tests on of the electron beam air interfaces that may be produced. Powerantennas and electronics. We are continuing to expand its capabilities sources available within the room are standard 110 V outlets, as well asboth at low power and high power using HPM sources. 220 V three phase supplies.Sources currently integrated into the facility include CW, modulated,and pulsed sources from 100 MHz to 4.2 GHz at peak power levels 1P-57: Compact High Power Microwave Sourcebetween 100 W and 1000 Watts, and arbitrary waveforms generated in K. H. BaxterMatlab with frequency content up to 6 GHz. Radiated signals can be Center for Engineering and Technological Excellence, Northropcaptured in either the frequency or time domain at corresponding Grumman Corporation, Fairfax, VA, United Statesfrequencies, and analyzed in hardware or software for their modulationcontent. Antennas available include commercial horns, yagis, and logperiodic antennas and a full suite of broadband helical antennas Northrop Grumman has designed, built and tested a compact,designed and built at the center. ruggedized ground mobile HPM system at TRL 6. This compact HPMTesting takes place inside a 4.25 m wide x 7.3 m long x 4.25 m tall source produces HPM effects at tactically useful ranges and providesanechoic chamber with better than 100 dB of isolation. Power on target scalable HPM effects against a variety of electronic systems includingis calibrated by the use of fiber optically coupled time resolved field sensors, communications, computers, and networks. Effects includeprobes from dc to 1 GHz at fields up to 10 kV/meter and fiber optically classic upset, latch-up, burn out as well as nonlinear effects on a varietycoupled three axis field probes up to 6 GHz at 800 Volts/meter. These of components, subsystems and systems. Current military HPMprobes are also used to calibrate antennas. All equipment is under systems are large, cumbersome, heavy (use oil as insulation media),computer control via Ethernet and control software written in LabView. difficult to maintain, and are 2-3 times the size and weight of thisWe are presently building two high power vircators and bringing online system. By using an epoxy formulation as the insulation andthe associated pulsed power to drive them. These will reside in the co- encapsulation media, we reduce the size of the source and canlocated 7.3 m x 7.3 m x 4 m X-ray shielded test facility. This capability demonstrate effects at tactically significant ranges, making HPMis expected to be online in early 2011. suitable for battlefield conditions for the first time.1P-56: Multisource Radiation and Microwave Facility 1P-58: Influences of Coil Current and Gas Pressure onP. T. Heffernan, N. G. Kinsey, W. E. Carter, R. D. Curry Inductively Coupled Plasma ParametersDept. of Electrical & Computer Engineering, University of Missouri, N. Delkash RudsaryColumbia, MO, United States radiation department of shahid behesahti university, tehran, IranA high density containment facility with the ability to shield ionizing One of the usual ion sources for uses in accelerators is RF ion sourcesradiation is a necessity for many facets of high powered that worked with applying high power electro-magnetic waves. Towelectromagnetic research. The radiation facility located at the types of RF plasma sources are CCP and ICP. In this paper, a newUniversity of Missouri-Columbia Center for Physical and Power design of inductively coupled plasma (ICP) that worked with applyingElectronics has been designed with the capability to accommodate a current to coil has been presented. Firstly, influences of coil current andvariety of different high power electromagnetic sources. The design gas pressure of ICP on plasma parameters has been investigated withconstraints originated from the need to contain a 4-8 MeV Electron COMSOL software. Then a new geometry for ion extraction units thatBeam Accelerator and a Virtual Cathode Oscillator (VIRCATOR) simulated with CST software has been suggested. Finally Simulationoperating at 19.5 MW with a frequency of 1 GHz. The 4-8 MeV results of suggested geometry for plasma chamber and extraction unitstunable relativistic electron beam is to be used for material irradiation, have been compared with the made model and get good resultsconditioning, and x-ray production. These devices will be driven by thenewly reconstructed MAG 1-D pulse modulator, which is now capable 1P-59: A Saturable Metamaterial-Based Passive Limiter forof producing 250 kV pulses at 100pps. The ability to support the Protection from HPM and UWB Sourcesionizing radiation shielding of multiple pulsed megawatt sources makes P. Kelly, J. Mankowskithis facility unique. Construction of the facility began with theplacement of 860, 6x2x2 ft. concrete blocks, totaling 3.2 million lbs. of Department of Electrical and Computer Engineering, Center forconcrete, as well as 58,000 lbs. of steel support structure. The finished Pulsed Power and Power Electronics, Texas Tech University, Lubbock,structure is 48x44x23 ft., supplying ample space inside to contain TX, United Statesmultiple sources. To support this weight, a high density concrete floorwas prefabricated during the original construction of the building. This paper describes the development of a high power in-line limiterThree layers of the concrete blocks were placed in a staggered fashion utilizing saturable metamaterial structures. A metamaterial structure is 58
  • 58. an artificial structure engineered to provide electromagnetic properties This work was supported by the Office of Naval Research undernot available in nature, more explicitly defined as a material having contract N00014-10-1-0463.simultaneously negative permittivity and negative permeability. Such amaterial, with both negative µ and negative ε are classified as a double- 1P-61: Gigahertz Sources for Cancer Detectionnegative material (DNG). A material exhibiting only one of the two S. Tantong, J. Baker, Z. Lu, N. E. Islamaforementioned electromagnetic properties is termed a singly-negativematerial (SNG). One such SNG structure, the split-ring resonator Electrical and Computer Engineering, University of Missouri,(SRR), is a negative permeability material which acts as a notch filter Columbia, MO, United Stateswith resonant frequency f0. The SRR structure yields itself to ferriteloading since the inductance of the structure is easily changeable. At Of the various techniques that have been employed for the detectionnominal power levels, f0 is significantly offset from the receiving and reconstruction of embedded objects [1, 2], the non-destructivefrequency such that the receiving frequency is unattenuated. When an methods using electromagnetic radiation are gaining wider acceptance.in-band high power microwave (HPM) is incident upon the filter, the In electromagnetic detection methods, both far-field and near-fieldferrite quickly saturates and shifts the resonant frequency of the filter to analysis can be employed. However, using near-fields have greaterthat of the receiving frequency. This effectively attenuates the incident advantages. Since near-field ranges, specifically in the microwaveHPM. Ansoft’s HFSS was used to accurately model and design the region, are very short, the experimental setup will require much lessSRR structure to minimize the E-field and maximize resonant effects. space for measurements as compared to the large distances required forFurthermore, accurate modeling of custom ferrite materials was far-field imaging.achieved through a combination of HFSS and Ansoft’s Maxwell 3D The work described in this presentation is an alternate technique whichSimulator at both saturated and unsaturated conditions. Both high and we call the Dual-Probe Imaging through Detection and Reconstructionlow power continuous wave testing verified minimal insertion loss as (DPIDR) method. This method uses dual near-field probes to measurewell as verification that a saturated ferrite material would effectively the scattering parameters in a matrix form, which contains reflectionshift the resonant frequency of the notch filter. The saturable SRR and transmission spectral information in the frequency domain. Thelimiter was exposed to an HPM source supplied by a vircator and data analyses using this technique to locate and reconstruct theresults are presented. embedded object will be presented. For a low cost realization, a compact and easy to fabricate Ultra Wideband Gigahertz (UWG) antenna pattern was deposited on a lossy1P-60: Comparison of TDR and FDR Measurements with FR4 substrate and designed to operate over a bandwidth of 3.1 to 10.6Established Models in Sandy Soil Types GHz. The UWG antenna with a compact size was fabricated withC. Umenyiora1, R. L. Druce1, R. D. Curry1, J. J. Bowders2 copper on both sides of a low-profile rectangular FR4 base. The1 Dept. of Electrical & Computer Engineering, University of Missouri, designed antenna can achieve a good impedance matching over a veryColumbia, MO, United States wide frequency bandwidth (2.85 to 12.00 GHz, defined by 2:1 VSWR),2 Dept. of Civil & Environmental Engineering, University of Missouri, and is suitable for both stationary and portable devices.Columbia, MO, United States A Vector Network Analyzer (VNA) coupled with a near-field probe was utilized to provide transmitted and reflected signal measurement inThe Center for Physical and Power Electronics at the University of the frequency domain. All of the data from the PC recorder wasMissouri, Columbia is studying radio frequency wave propagation in processed through an in-house developed code. The magnitude andsilica sand with different moisture content levels. A precision coaxial phase response from transmitted signal were also saved for analysis.cell geometry has been fabricated for the soil tests and will be The final presentation involves a brief introduction to the experimentpresented. The infiltration rate of water in sandy soil materials is high, and our image reconstruction technique. It is expected that the resultsresulting in inhomogeneous moisture distribution through the soil under from this effort not only be applied for breast cancer research [3], butthe influence of gravity. Previous work indicates a relationship between also for the detection of embedded objects in a dielectric.bulk soil density, moisture content, and dielectric constant of different [1] CHEUNG, et al., Detection of objects embedded in a non-uniformsoil types, including sand. Experiments, using time-domain- turbid medium vol. 3459. Bellingham, WA, INTERNATIONAL:reflectometry (TDR) and frequency-domain-reflectometry (FDR) Society of Photo-Optical Instrumentation Engineers, 1998.measurements, performed in a cylindrical test cell, indicate the [2] M. Soliman and Z. Wu, "Buried object location based on frequency-development of significant inhomogeneity in moisture content in the domain UWB measurements," Journal of Geophysics and Engineering,span of a few minutes. The compaction method, dense or loose, has vol. 5, p. 221, 2008.significant impact on the bulk density of the sand and on the migration [3] A. M. Campbell and D. V. Land, "Dielectric properties of femaleof moisture through the sand. TDR and FDR methods are used to human breast tissue measured in vitro at 3.2 GHz," Physics in Medicinemeasure average dielectric constant, which is compared with and Biology, vol. 37, p. 193, 1992.established models of propagation of RF through sand and soil matrices.Simulations were performed to supplement experimental observation of 1P-62: Design of a Damped Sinusoidal Oscillator Systemsoil moisture content in coaxial cell geometry. Experimental testing has J. M. Lee, H. O. Kwon, S. M. Hwang, J. W. Ahnindicated that the average dielectric constant decreases as the Research&Development Department 2, Hanwha Corporation, Gumi,compaction of the soil changes from dense to loose. Simulations using Kyungbuk, South KoreaCST Microwave Studio confirm that the change in dielectric constant,through the soil, is a result of the moisture content distributioninhomogeneity. By correlating the experimental and simulation data, A damped sinusoidal(DS) oscillator system has been designed and willfurther insight may be gained in determining the dielectric constant at be fabricated in our laboratory in the near future. The system consistsany level of the soil based on the impedance and time relationship. of a primary power source, a Marx generator, a DS oscillator withComparisons of silica sand dielectric constant as a function of moisture radial transmission line, and a helical antenna. A 8 stage Marxcontent and density with existing models will be presented. The generator is used to feed 400kV high voltage pulse to the DS oscillator.Microwave CST simulations that address the inhomogeneity of the The designed characteristic impedance and average switching voltagedielectric constant will be presented along with a comparison of the of the DS oscillator is 7.5 ohm and 300kV respectively. The axial moderesults. helical antenna is designed to operate at a frequency around 230MHz. 59
  • 59. The whole DS oscillator system is simulated using PSpice and CST form pinch electromagnetic volume force. The critical phenomenonMWS. We are expecting the peak radiated electric field of the system at was pointed out which was about the average value of equivalenta distance of 1m is around 130kV/m. Detailed characteristics of surface electromagnetic pressure. With numerical calculation, thefabricated system and tested results will be presented. critical value of relative radius was one which was intimate with the radius value of a shaped charge jet and the skin depth. With the1P: Applications posters I: Fusion, EM, Beam & Lasers simulation of applicable given parameters,the average value of equivalent surface electromagnetic pressure would be steeply increasedMonday, June 20 13:30-15:30 Regency Ballroom in the thin radius location of a shaped charge jet, which would result in the sausage instability. The result of 5kV passive electromagnetic1P-63: Expressions of the Optimal Electromagnetic Force armor experiments was as: when the shaped charge jet was acted by anQ. A. Lv electromagnetic pressure the diameter of perforation was lager than thatmechanical engineering college, shijiazhuang, China of no power.The electromagnetic launcher should work in the most rigorous 1P-66: Analysis of Conductor Impedances Accounting forsituation, where the conductors load the highest current density in the Skin Effect and Nonlinear Permeabilitynear-melting temperature. But the optimizing theory about M. P. Perkins, M. M. Ong, C. G. Brown Jr., R. D. Speerelectromagnetic launcher by now remains ambiguous. In this paper, the Lawrence Livermore National Laboratory, Livermore, Ca, Unitedoptimal distribution of magnetic energy between that of B0 and I in Statesf=B0I was discussed for the least energy consume. When the externaluniform magnetic strength B0 equals to the electromagnetic strength BI It is often necessary to protect sensitive electrical equipment fromat the round current-carrying conductor surface, the optimal condition pulsed electric and magnetic fields. To accomplish this electromagneticof least energy is satisfied. After that, the optimal matching condition shielding structures similar to Faraday Cages are often implemented. Ifwas also expressed with the magnetic pressure theory. The magnetic the equipment is inside a facility that has been reinforced with rebar,field configuration is directly connected with the optimally designed the rebar can be used as part of a lightning protection system.EM launcher. The optimized criterion of the projectile local surface Unfortunately, such shields are not perfect and allow electromagneticreached is that any local magnetic flux angle is half of the local outer fields to be created inside due to discontinuities in the structure,surface angle , and the two angles are all with respect to the moving penetrations, and finite conductivity of the shield. In order to performdirection of the projectile. According to the magnetic pressure theory, an analysis of such a structure it is important to first determine thethe typical magnetic force on a thin long round current-carrying effect of the finite impedance of the conductors used in the shield. Inconductor exerted by a uniform external magnetic field B0 is obtained, this talk we will discuss the impedances of different cylindricalwhich is consistent with the force formula f=B0I. conductors in the time domain. For a time varying pulse the currents created in the conductor will have1P-64: Unsymmetrical Lateral Electromagnetic Action and different spectral components, which will change the current densityOptimization of the Distance Between Plates in the Passive due to skin effects. Many construction materials use iron and differentElectromagnetic Armor types of steels that have a nonlinear permeability. The nonlinearS. H. Chen material can have an effect on the impedance of the conductormechanical engineering college, shijiazhuang, China depending on the BH curve. Although closed form solutions exist for the impedances of cylindrical conductors made of linear materials, computational techniques are needed for nonlinear materials.The unsymmetrical action mechanism of the passive electromagnetic Simulations of such impedances are often technically challenging duearmor was analyzed, in which a shaped charge jet was forced to move to the need for a computational mesh to be able to resolve the skinlaterally by an electromagnetic action. With the virtual origin concept, depths for the different spectral components in the pulse. The results ofthe action time between current and jet element was analyzed and the such simulations in the time domain will be shown and used tooptimization method of the distance between plates was achieved. determine the impedance of cylindrical conductors for pulses that haveBased on an equivalent circuit, the lateral action model of a shaped low frequency content.charge jet was established, and the electromagnetic force distribution vs. *This work performed under the auspices of the U.S. Department oftime and location was concretely discussed. With the virtual origin Energy by Lawrence Livermore National Laboratory under Contractconcept, the lateral velocity of the jet element with different longitude DE-AC52-07NA27344.speed was derived after the jet passed through the two armor plates.When the capacitors voltage increased to 35 kV, the obvious lateralexcursion of the majority jet except the head section was calculated. By 1P-67: Magnetic Forming and Cutting of Flat Thin Al Sheetsthe way, the lateral excursion of the metal jet was observed in M. T. Pereira1, H. Canacsinh2,3, L. M. Redondo2,3 1correlative experiments. The research work is profound on the defense João Bettencourt Teotónio Pereira, Lisbon, Portugal 2mechanics of passive electromagnetic armor. Lisbon Engineering Superior Institute, Lisbon, Portugal 3 Nuclear Physics Center from Lisbon University, Lisbon, Portugal1P-65: Pinch Electromagnetic Action on the Shaped ChargeJet in the Passive Electromagnetic Armor This work studies the optimization of electrical and mechanicalS. H. Chen parameters for magnetic forming and cutting of flat Aluminum sheets, <1 mm, for the food industry. Electromagnetic forming is one of themechanical engineering college, shijiazhuang, China most successful pulsed power applications focused in cutting and molding of Al tubes or thick Al sheets for a number of industries suchThe axisymmetric action mechanism of the passive electromagnetic as the automobile industry. The industrial application of pulsedarmor was analyzed, in which a shaped charge jet was forced into a magnetic fields in applications of thin flat Al sheets, below 1 mm, hassausage by an electromagnetic action. Based on an equivalent circuit, not been so successful due to numerous technological issues that lowerthe surface electromagnetic pressure of a shaped charge jet was derived the efficiency of the process. One way of circumventing these problems 60
  • 60. is increasing the peak current in the range of hundred kA. Considering microseconds for FRCHX. Current lifetime measurements of the FRCsthe latter, the modulator technology has been dominated by hard-tubes formed with FRCHX show lifetimes of only 7 ~ 9 microseconds onceand solid-state SCR devices in low efficiency circuits. The utilization the FRC has entered the capture region.of IGBTs allows the use of more efficient topologies, recovering part of A description of the pulsed power systems that comprise FRCHX willthe energy, but IGBT technology has still constraints in handling the be presented along with an overview of the magnetic and plasmahigh peak currents needed. Hence, several techniques are used to diagnostics fielded on the experiment. Results from recent FRCHXincrease the yield of the magnetic forming process. In the industry, experiments will then be presented, and possible reasons for thebefore the Al sheet is formed, with a convenient shape, it must be cut. lifetime limitations will be discussed along with several approaches forTherefore, results will be shown for the forming and cutting of flat Al overcoming these limitations.sheets, below 1 mm, considering a single 30 kA, SCR based modulator, This work is supported by DOE-OFES.and a parallel stack association for higher currents. In order to increasethe efficiency of the process, optimum shape forming and clean cutting, 1P: Explosive and Compact Pulsed Power postersa number of techniques were used. The reduction of system inductancewas crucial for increasing the ringing frequency, in this way the coil Monday, June 20 13:30-15:30 Regency Ballroomtechnology is crucial. A vacuum camera was used to reduce thelimitations regarding the presence of air with thin sheets both for 1P-70: Electric Breakdown of Longitudinally-Shock-forming and for cutting the Al. In addition, considering the problems Compressed Pb(Zr0.52Ti0.48)O3 Ceramicsassociated with the attraction of the Al thin sheet after the repulsion S. I. Shkuratov1, E. F. Talantsev2, J. Baird1period due to the sinusoidal shape of the electrical current, a technique 1 Loki Incorporated, Rolla, MO, U.S.A.was used to increase the pulse current fall. 2 Pulsed Power LLC, Lubbock, TX, U.S.A.1P-68: Statistical Properties of Modern Fast Photo Detectors Electric breakdown of longitudinally-shock-compressedV. Ivanov, Z. Insepov Pb(Zr0.52Ti0.48)O3 ferroelectric ceramics was experimentallyHigh-Energy Physics Division, Argonne National Laboratory, Argonne, investigated. It was found that a dependence of breakdown fieldUnited States strength of shocked ferroelectrics on the thickness of the element ranging from 0.65 to 6.5 mm is described by the law [1,2] whichThe original mathematical models and algorithms developed within the describes breakdown measurements for dielectrics at ambientcode “Monte Carlo Simulator” (MCS) are presented. These algorithms conditions. It follows from the experimental results that the tunnelare devoted to model the photo-emission and cascades of secondary effect is a dominant mechanism of injection of prime electrons in theelectron emission in micro-channel plate (MCP) amplifiers are widely shocked ferroelectric elements.used as photo detectors in accelerator physics, medical diagnostics,astrophysics etc. A theoretical method for calculating secondary [1] F. Forlani and N. Minnaja, “Thickness Influence in Breakdownelectron emission (SEE) yields has been developed. The method uses Phenomena of Thin Dielectric Films,” Physica Status Solidi, Vol. 4, pp.Monte Carlo simulation, empirical theories and close comparison to 311-324 (1964).experimental data in order to parameterize the SEE yields of highly [2] F. Forlani and N. Minnaja, “Electrical Breakdown in Thinemissive materials for the MCPs. Numerical simulations were used to Dielectric Films,” J. Vacuum Science and Technology, Vol. 6, pp. 518-study the statistical properties of large-area fast photo detectors which 526 (1969).are being developed at Argonne National Laboratory. 1P-71: Manufacturing of Targets and Assemblies for High1P-69:FRC Lifetime Studies for the Field Reversed Explosive and High Energy Pulsed Power Research at LosConfiguration Heating Experiment (FRCHX) Alamos National LaboratoryC. Grabowski1, J. H. Degnan1, D. J. Amdahl1, R. K. Delaney1, F. FierroM. Domonkos1, F. M. Lehr1, R. Magallanes1, P. R. Robinson1, MST, Los Alamos National Laboratory, Los Alamos,NM, United StatesE. L. Ruden1, W. White1, H. Wood1, D. G. Gale2, M. Kostora2,J. McCullough2, W. E. Sommars2, M. H. Frese3, S. D. Frese3, The ability to fabricate precision assemblies and targets for high energyJ. F. Camacho3, S. K. Coffey3, T. P. Intrator4, G. A. Wurden4, J. Sears4, and high explosive pulsed power research can be challenging. PartP. J. Turchi4, W. J. Waganaar4, T. Weber4, R. E. Siemon5, S. Fueling5, surface finish and tight tolerances require that special tooling, fixturesB. S. Bauer5, A. G. Lynn6, N. F. Roderick61 and fabrication methods be used. In this poster I will give examples of Air Force Research Laboratory, Kirtland AFB, NM, United States two assemblies that were made and the methods used to ensure the2 Science Applications International Corporation, Albuquerque, NM, requirements were met. LA-UR 11-00085United States3 NumerEx, Albuquerque, NM, United States4 Los Alamos National Laboratory, Los Alamos, NM, United States 1P-72: Research on Magnetic Field Generated by MFCG5 Driven Solenoid University of Nevada, Reno, Reno, NV, United States6 H. Li University of New Mexico, Albuquerque, NM, United States Mechanical Engineering College, Shijiazhuang, ChinaThe goal of the Field-Reversed Configuration Heating Experiment(FRCHX) is to demonstrate magnetized plasma compression and Strong enough pulse magnetic field action could prevent the shapedthereby provide a low cost approach to high energy density laboratory charge jet from disrupting when it flight through the action region, thus,plasma (HEDLP) studies, which include such topics as magneto-inertial a propriety longitudinal magnetic field could enhance the penetratingfusion (MIF). A requirement for the field-reversed configuration (FRC) power of jet. In this paper, a shaped charge aligned with solenoid whichplasma is that the trapped flux in the FRC must maintain confinement current circulates in it was provided by a MFCG was described, andof the plasma within the capture region long enough for the also the circuit model. Based on the circuit parameters, the formulationcompression process to be completed, which is approximately 20 of current and distribution of magnetic field were analyzed. When the 61
  • 61. magnetic field on the axis of solenoid exceeded a threshold value the systems span wide ranges of permittivity, dielectric strength, andtime of current approaching to peak and the duration of current practical operating voltage, all of which are combined to deliver thecirculating were computed. The diameter of jet determined evidently desired electrical pulse to the load. Recent work has focused onthe frequency of magnetic field and the dimensions of solenoid. These polymer-ceramic nano-composites and TiO2 based ceramics whichresults are significant for the initiation time of both MFCG and shaped have permittivities of approximately 50 and 140, respectively. Thecharge and the solenoid parameters. corresponding propagation velocities in the dielectric are 4.2 and 2.5 cm/ns. The descriptor “compact” is a function of the application, and while a flat transmission line may satisfy some applications, folding the1P-73: An Ancillary Boundary Integral Equation for lines offers some different possibilities, including delivering pulseMagnetostatic Analysis lengths up to several hundred ns with a largest dimension less than 1 m.M. S. Ingber, G. F. Kiuttu, J. A. Ingber, B. T. Smith Marxed capacitor banks are routinely used to scale the voltage fromAccurate Solutions in Applied Physics LLC, Albuquerque, NM, United standard power supplies to 100’s of kV without the use of a pulseStates transformer. Marxed operation is applied to the transmission line development described here. Theory, simulation, and fabricationThe boundary element method (BEM) has been established as an eff consultation were convolved to design flat and folded transmissioneffective means for magnetostatic analysis. Direct BEM formulations lines. The experimental evaluation of the lines progressed from low tofor the magnetic vector potential have been developed over the past 20 high voltage and from single lines to a Marxed configuration of severalyears. There is a less well known direct boundary integral equation lines. This paper presents the recent results of this development effort.(BIE) for the magnetic flux density which can be derived by taking thecurl of the BIE for the magnetic vector potential and applying 1P-76: Pulsed Power Generator Using Solid-State LTDsproperties of the scalar triple product. On first inspection, the ancillary W. Jiang1, A. Tokuchi1,2boundary integral equation for the magnetic flux density appears to be 1 Department of Electrical Engineering, Nagaoka University ofhomogeneous, but it can be shown that the equation is well-posed and Technology, Nagaoka, Niigata, Japannon-homogeneous using appropriate boundary conditions. In the 2 Pulsed Power Japan Laboratory, Ltd., Kusatsu, Shiga, Japancurrent research, the use of the ancillary boundary integral equation forthe magnetic flux density is investigated as a stand-alone equation andin tandem with the direct formulation for the magnetic vector potential. LTD (linear transformer driver) concept has been used to develop compact, repetitive pulsed power generators. As an experimental demonstration, 10 MOSFET based LTD modules have been put1P-74: The Effects of Inductance on the Metallization together to generate 8 kV, 300 A, and 100 ns pulses at repetition rate ofRemoval of Exploding Films 1 kHz, with rise- and fall-times in the range of 30-40 ns. The objectiveT. M. DiSanto, M. T. Muffoletto, D. P. Muffoletto, K. M. Burke, of this study is to prove the practicability of compact pulsed powerJ. L. Zirnheld generators based on solid-state LTDs.University at Buffalo, Buffalo, NY, United States 1P-77: SLEP-150M Compact Supershort Avalanche ElectronDuring the electrical explosion of a thin metallized film, the Beam Acceleratormetallization layer is heated rapidly up to vaporization whereby the V. F. Tarasenko, I. D. Kostyrya, E. K. Baksht, D. V. R. V. Rybkafilm bursts and the metal layer is ejected from the substrate. It has been High Current Electronics Institute, Tomsk, Russian Federationshown that adding inductance in the discharge path changes thecharacteristics of the explosion, most notably the energy transferefficiency. This work sets out to explore the mechanical efficiency as a Recent years have brought much interest in e-beam and X-rayfunction of inductance, namely how much of the metallized surface is generation in gas diodes at increased pressure. The objective of theliberated during the explosion of the film. Using an image processing work was to design a compact electron accelerator with a more than 50-technique is used to quantify the metallization removal and the results A beam current amplitude in atmospheric pressure air. This workof this effort are discussed herein. continues our study reported in [1]. It can be supposed that increasing the current amplitude of runaway electron beams generated in atmospheric pressure air will expand the use of supershort avalanche1P-75: Compact Pulsed Power Using Solid Dielectric electron beam (SAEB) accelerators. The performed studies madeTransmission Lines possible the compact gas-diode accelerator that ensures a considerableM. T. Domonkos1, S. Heidger1, D. Brown2, A. Devoe3, F. Dogan4, increase in SAEB amplitude in atmospheric pressure air. The numberD. Gale2, J. OLoughlin1, J. Parker2, D. Sandoval2, K. Slenes5, of electrons downstream of the foil of thickness 10 mkm with a grid ofW. Sommars2, J. Watrous6 high transparency was ~5*10^10 electrons. At a FWHM of the current1 AFRL/RDHP, Air Force Research Laboratory, Kirtland AFB, NM, pulse of ~100 ps, this corresponds to a SAEB amplitude of ~80 A. TheUnited States increase in beam current in our work owes to the following factors.2 SAIC, Inc., Albuquerque, NM, United States First, we increased the total length of the emitting surface of the3 Presidio Components, San Diego, CA, United States cathode formed by parallel wires. Second, the use of the transmission4 Missouri Institute of Science and Technology, Rolla, MO, United line allowed an increase in the rate of rise of the voltage and itsStates amplitude across the gap. Third, the conical insulator pressed in the5 TPL, Inc., Albuquerque, NM, United States transmission line allowed a better match between the transmission line6 NumerEx, LLC, Albuquerque, NM, United States and the gas diode. In this work, we leant against the mechanism of the SAEB generation proposed in [2] and developed in [3, 4]. According toThis paper documents recent work developing solid dielectric this mechanism, runaway electrons are generated and acceleratedtransmission lines for sub-microsecond, 100 kV class compact pulsed between the dense plasma front moving from the cathode to the anodepower systems. Recent developments in nano-composite and ceramic and the anode. The end of the current is due to bridging of the gap bydielectric materials processing have opened new possibilities for using the dense plasma. As the dense plasma bridges the gap, the electrictransmission lines in compact pulsed power systems. These material field distribution in the gap becomes more uniform, and the voltage 62
  • 62. across the gap decreases. This work was supported by the Federal voltage (V) for two basic applications in pulsed power: linear &Target Program "The scientific and scientific-pedagogical personnel of nonlinear transmission lines (NLTLs) [2]. For instance, ceramicsInnovative Russia", Contract No. 02.740.11. 11.0562. manufacturers normally do not provide the C×V variation for theseREFERENCES materials in their datasheets as well as the corresponding dielectric HV[1] I. D. Kostyrya, E. Kh. Baksht, and V. F. Tarasenko, “An efficient breakdown strength (a key factor for the design of compact devices). Incathode for generating an supershort avalanche electron beams in air at view of that, the scientific challenge of this paper is to characterizeatmospheric pressure,” Instrum. Exp. Tech., Vol. 53. No. 4, pp. 545- commercial ceramics with high dielectric constant, considering the548, 2010. breakdown voltage and C variation with the applied voltage. Herein the[2] V. F. Tarasenko, V. M. Orlovskii, S. A. Shunailov, “Forming of an main idea is to check the breakdown properties of PZT (Lead Zirconateelectron beam and a volume discharge in air at atmospheric pressure”, Titanate) ceramics for use in storage systems because of their highRuss. Physics J., Vol. 46, No. 3, pp. 325-327, 2003. dielectric constant (1000-3000) and excellent performance in HV[3] V. F. Tarasenko, E. H. Baksht, A. G. Burachenko, I. D. Kostyrya, explosive-driven ferroelectric generators. For testing this material weM. I. Lomaev, and D. V. Rybka, “Supershort avalanche electron beam acquired PZT samples that were de-poled to remove their piezoelectricgeneration in gases,” Laser Part. Beams, Vol. 26, No. 4, pp. 605–617, properties as normally PZT are used as transducers. Due to the2008 mechanic vibration in piezoelectric materials, de-poling is necessary for[4] V. F. Tarasenko, E. Kh. Baksht, A. G. Burachenko, I. D. Kostyrya, possible PZT applications in HV capacitors and transmission lines toM. I. Lomaev, and D. V. Rybka, “Supershort avalanche electron beams avoid dielectric damage and losses. For the breakdown tests and C×Vand x-rays in atmospheric-pressure air” IEEE Trans. of Plasma Science. characterization we have built respectively two testing circuits: a) a 60Vol. 38, No. 4, pp. 741-750, 2010. kV impulse circuit using ignition coils and b) a special HV circuit for measuring the sample capacitance on static condition as a function of the applied voltage. In this work, the results obtained so far for PZT1P-78: Autonomous Compact and Repetitive Low-Energy samples under breakdown and C×V tests will be presented, taking intoPulsed Power Generator the prospects for their applications in compact pulsed power systems.M. J. Parker1, B. M. Novac1, I. R. Smith1, P. Senior1, G. Louverdis21 Electronic and Electrical Engineering Department, Loughborough [1] J.O. Rossi, P. Castro, M. Roybal, E. Schamiloglu, S. Sawhill, and E.University, Loughborough, United Kingdom Savrun, “HV Energy Store in Organic Composite Dielectrics for2 Security Services, Dstl fort Halstead, Sevenoaks, United Kingdom Compact Pulsed Power,” in Proc. of the 2008 IEEE International Power Modulators & HV Conference, pp. 548-551.An autonomous, compact and repetitive pulsed power generator has [2] J.O. Rossi & R.H.M. Siqueira, “Study of Capacitance Variation ofrecently been developed for use in low-energy, general-purpose Commercial Ceramic Capacitors,” to be published in Proc. of the 2010experimentation, in either indoor or inclement outdoor environments. IEEE International Power Modulators & HV Conference.Flexibility, reliability and safety were the most important design *This work has been sponsored by the USAF SOARD under contractcriteria, and these were achieved by the use of interchangeable no. FA9550-10-1-001.components, off the shelf units and the development of fibre-opticcontrolled charging and triggering units that allow the complete system 1P-80: Dissipating Screen of Generators Based onto be run from a single battery source. The present arrangement is Transformer Storage and Combination Vacuum Interrupterbased on a 1 µF capacitor type 31158, although the assembly acceptsany of the ‘outline A’ capacitors in the GAEP series S/SS and SE/SSE and Plasma Opening Switchranges. Two series-connected mounting assemblies are designed to O. G. Egorovhouse a range of resistors, inductors or any other components that can Atomic corporation, TRINITI, Moscow reg.,Troitsk, Russianbe adapted to fit within their footprints. Finally, a trigatron type SG- Federation101M75C4 (R E Beverly III & Associates) connects the generator to anexternal load (or another pulsed power system) through a HV coaxial A generator on the basis of a transformer with a changeable coefficientcable, and enables the system to be operated repetitively. Both the of coupling has been suggested in papers [1-3]. In the next paper [4]output current and voltage are internally monitored using a Pearson suggested is one of the alternatives to change a coupling coefficient bycurrent transformer and a Tektronix high-voltage sensor. The means of an electromagnetic screen.completed system is housed in a weatherproof metallic container, with It is known that there exist two physical factors leading to attenuationa volume of less than 0.3 cubic meters on a large number of occasions. of an external magnetic field, namely the processes of reflection and absorption. The first factor is attributed to eddy currents, while the1P-79: Study of HV Dielectric Ceramics for Compact Pulsed second is caused by the absorption of electromagnetic energy in the substance depth and conversion of this energy to heat [5, 6]. It isPower* obvious that the generation of eddy currents on the screen surfaceJ. O. Rossi, L. P. Silva Neto, A. R. Silva Junior prevents the energy transfer from the storage. So the main job of theAssociated Plasma Laboratory, National Institute for Space Research, screen is to absorb the electromagnetic energy in such a measure thatS.J. Campos, SP, Brazil would provide an effective transfer of energy to load, i.e. R1(t)>R2 (t), where R1(t) and R2 (t), are the dependences of the screen and plasmaThe scientific interest of this paper consists of the study of ceramic opening switch resistances vs. time, respectively.dielectrics for applications in compact pulsed power systems. The In the given paper an alternative screen construction is proposed wherereason for this is that ceramic dielectrics with large dielectric constant the surface is greatly developed by way of making a periodic structure.and high breakdown strength have great potential for decreasing the Such structures are widely used in UHF technique [7]. In our case thisvolume required for pulsed power supplies in future envisioned pulsed structure causes not only loss of energy by heat, but, when usingpower-driven systems on mobile platforms [1]. Ceramic dielectric ferrites, generation of UHF oscillations that favors the effect of energyblocks have been used with great success in compact pulse forming dissipation. Some estimations are given concerning the surfacelines; however, there has been a paucity of research on the nonlinear resistance, energy loss by heat and their influence on the energybehavior of these commercial materials as it is necessary to understand transfer to load.the dielectric constant (or capacitance, C) variation with the applied Reference 63
  • 63. 1. O.G. Egorov Development of the conception of high power pulsed 1P-82: Rapid Capacitor Charger for Compact Pulsed Powercompact generator based on inductive storage and combined opening Applicationsswitch. // Proc.2nd Euro-Asian Pulse Power Conference. Vilnius, S. L. Holt, J. C. DickensSeptember, Lithuania, 2008. Center for Pulsed Power and Power Electronics, Texas Tech2. O.G. Egorov The project of pulse power generator based an University, Lubbock, TX, United Statesinductive storage and a combination of vacuum interrupter and plasmaopening switch// Proc. the IET European Pulsed power Conference,CERN, Geneva, Switzerland, 2009. Making systems more compact has been the focus of much work3. O.G.Egorov The design project of a pulse generator based on throughout the pulsed power and directed energy communities.inductive storage and a combination of vacuum interrupter and plasma However, the majority of this effort has focused on the pulsed poweropening switch// Proc.3nd Euro-Asian Pulse Power Conference. Jeju, generators and loads (microwave, RF or laser sources). Even thoughNovember, Korea, 2010. ancillary components, such as prime power sources and high voltage4. O.G.Egorov Generator of nanosecond pulses based on inductive supplies, often comprise a large proportion of the total system volumestorage made of solid conductor // Proc.3nd Euro-Asian Pulse Power and weight they are typically commercially sourced for budgetary andConference. Jeju, November, Korea, 2010. time restrictions. Unfortunately, commercially available high voltage5. Conducting sheaths in the pulse electromagnetic field. / V.V.Vasiliev, power supplies are typically designed for continuous or near continuousL.L.Kolensky, Yu..A.Medvedev and B.M.Stepanov. operation, sacrificing size and weight to provide high operational dutyM.;”Energoatomizdat, 1982.-200 p. cycles. This type of design often focuses on thermal management and6. G.Kaden. Electromagnetic shields. M.; GEI, 1957, p.132. component selection; packaging and circuit board layout are designed7. Instruments and technique UHF/Ed.V.F.Vzjatichev. around the need to maximize heat dissipation. Market demandsM.;”Energoatomizdat 1980.-145p. reinforce this design mentality; a system that can be operated only a few seconds every minute is not normally desirable and no amount of size or weight savings will make it appealing for the majority of1P-81: Self-Contained Source Based on an Innovating consumer, business or industrial applications. However, in directedResonant Transformer and an Oil Peaking Switch energy systems with short burst or intermittent operationalR. Pecquois1, L. Pecastaing1, M. Rivaletto1, A. Silvestre de Ferron1, requirements, the operating time of the device is typically much lowerP. Pignolet1, L. Caramelle2, J.-M. Duband2, R. Vezinet3 than the heat-transfer time constants from the device to the surrounding1 SIAME - UPPA, Pau, France environment. In this adiabatic operating regime, heat cannot be2 Hi Pulse, Pont de Pany, France dissipated from the device in a relevant timeframe, making most3 CEA GRAMAT, Gramat, France thermal management efforts ineffective. If high voltage power supplies are designed with the goal of maximizing the intermittent or burst modeNowadays, a broad range of modern industrial applications brings the power density, rather than the continuous mode power density, thenecessity of compact high-power electromagnetic wave generators. overall size and weight of these systems can be significantly reduced.Conventionally, this kind of generator consists in a primary energy Using this adiabatic design ideology, Texas Tech University (TTU) hassource and an antenna, separated by a power-amplification system that developed a single shot 5 kV, 3 kJ/s charger that occupies less than 1.5forwards the energy from this source to the antenna. A Marx generator L, including Lithium Ion Polymer (LiPo) batteries. This system has aor a Tesla transformer is classically used as a power-amplifier. Our power density exceeding 2 kW/L including batteries, while moststructure uses an innovating very compact resonant transformer. Our commercially available power supplies offer power densities betweencomplete pulsed power source, named MOUNA, is composed of a set 0.2 and 0.5 kW/L excluding batteries, five to ten times lower than theof batteries, a DC/DC (300V/10kV) converter to load four capacitors technology developed at TTU. This paper will discuss recent efforts towith a 1.6A current, four synchronized spark gap switches, a resonant improve the efficiency, power density, and scalability of this system. Atransformer generating a few hundreds kV and an oil peaking switch. modified system has been demonstrated to operate at charging voltagesEach capacitor is connected to a triggered switch. When the triggered up to 50 kV. Additionally, the circuit topology has been adapted toswitches are closed, the energy stored in the capacitors will load an improve system reusability and provide the possibility of rep-rateoutput oil radial line by resonant transfer via the transformer and its operation.leakage inductance. The oil peaking switch is adjusted in order to be Distribution A: Approved for public release, distribution is unlimited.switched when the voltage on the radial oil line reaches a value closedto the maximum of the transfomer one. This line is connected to the 1P-83: Reducing PFN Marx Generator Size Using Nestedpeaking switch at its input and to the antenna at its output. When the Solid Insulationline is crowbared, the circuit is equivalent to a damped oscillator the R. J. Adler, J. A. Gilbrech, D. Newenergy of which is radiated by the dipole antenna. To achieve a such Applied Energetics, Tucson, AZ, United Statescompactness (38 liters), all these parts are integrated within the antennaand immersed in oil. The innovating transformer is made of fourprimary windings, two secondary windings in parallel and a magnetic Current state-of-the-art, high voltage, pulsed power systems above 0.5core. This specific geometry makes it possible to limit the leakage GW output power generally use Marx or PFN Marx generator designs.inductance and optimize the coupling ratio. Output transformer voltage The high voltage insulation of these devices has been traditionallyand oil line oscillations are measured with V-dot probes included in the based on either high dielectric strength oil or sulfur hexafluoride (SF6).generator. This article presents the mechanical design of the complete The dielectric strength of the insulating medium determines thepulsed power system and the first electrical measurements (particularly minimum size of the tank enclosure of the system since it determinesconcerning the resonant transformer output waveform, the line the maximum voltage standoff between the fully erected Marx outputoscillations and the radiated E-field). voltage and the tank wall. We have applied our "Nested High Voltage"This work is driven and financially supported by the French Ministry insulation technology to the problem of PFN pulse generator design.for Defense (Direction Générale de l’Armement – contract N° This technique uses solid insulation with field grading foils that07.34.027). minimized the distance from the fully erected voltage and the tank wall. This report describes work performed in adapting this technology to building a PFN pulse generator. We believe that this technology will allow size reductions relative to other insulation technologies. The 64
  • 64. technology also allows integration of the PFN charge power supply into and uses inexpensive, off-the-shelf components. In a typicalthe PFN. Data will be presented on the performance of a 130 kV configuration, the Marx has an erected voltage of 10 kV, a storedversion of this type of device capable of up to 5 kA output. We energy of 5 mJ, and a risetime of 400 ps. The Marx has been operateddemonstrated that making a “flat” pulse is feasible in this geometry, at a pulse repetition frequency of 250 Hz. Other potential uses includeand we determined that there is an optimal grading technique for this sourcing compact antennas and biological and environmentaltype of design. A nominal 130 kV, 3-stage pulse was successfully applications.produced using our design techniques. Both high gradient insulationand pulse generation have been demonstrated. The systems will be 1P-87: Development of a High Repetition Rate and Highuseful in the 50 kV - 2 MV range with impedances in the 20 - 200 ohmrange. This work was supported by AFRL under contract FA9451-10- Voltage Switching Power Supply with a SiC-JFET for anM-0093 Induction Synchrotron K. Ise1, K. Takaki1, K. Okamura2, M. Wake2, K. Takayama2, Y. Oosawa3, W. Jiang41P-84: Electrical Analysis of Piezoelectric Transformers and 1 Iwate University, Morioka, Iwate, JapanAssociated High-Voltage Output Circuits 2 High Energy Accelerator Research Organization, Tsukuba, Ibaragi,J. A. VanGordon, B. B. Gall, S. D. Kovaleski, E. A. Baxter, B. H. Kim, JapanJ. W. Kwon 3 SUN-A Corporation, Miyoshi, Hiroshima, JapanElectrical and Computer Engineering, University of Missouri, 4 Nagaoka University of Technology, Nagaoka, Niigata, JapanColumbia, Missouri, United States To aim for developing a next-generation switching power supply (SPS)Piezoelectric transformers can be useful as compact, high-voltage for an induction synchrotron, the switching characteristics of a newsupplies. At the University of Missouri, the effect of adding output power device was investigated. The device is a silicon carbide (SiC)circuits to bipolar piezoelectric transformers is being studied. These junction field-effect transistor which was the most excellentpiezoelectric transformers produce output voltages in excess of 25 kV performance in the existing SiC power devices. It was developed byfrom medium-voltage, radio frequency inputs. However, the high KEK and SUN-A Corporation using a SiC die (4.16 mm x 4.16 mm)output voltage and low output current of these devices can make it which was manufactured by SiCED. In this paper, it was demonstrateddifficult to acquire an accurate electrical measurement of the output that the device was operated with 1 MHz, 1 kV and 27 A in continuousvoltage without affecting the transformer ratio or resonance of the mode operation and its max junction temperature was analyzed to bedevice. This paper will analyze capacitive voltage dividers as a means 186 °C. Its thermal resistance from the junction to the case Rj−c wasof diagnostic measurement and Cockcroft-Walton type circuits to evaluated to be 0.14 °C/W. The results show that it is hopeful candidateincrease the voltage multiplication beyond that of the piezoelectric as the semiconductor switch of the next-generation SPS.transformer alone.________________________________ 1P-88: Comparison of Computations and Experiments for* Work supported by Nuclear Regulatory Commission, Qynergy, andLos Alamos National Laboratory Tests of Ranchero Flux Compression Generators above 50 MA J. H. Goforth, R. G. Watt, W. L. Atchison, D. H. Herrera, R. K. Meyer,1P-85: Solid State Impulse Marx Generator H. Oona, R. E. Reinovsky, C. L. Rousculp, L. J. Tabaka, D. T. TorresJ. R. Mayes, W. C. Nunnally, W. J. Carey Los Alamos National Laboratory, Los Alamos,NM, United StatesApplied Physical Electronics, L.C., Austin, Texas, United States The Ranchero simultaneous coaxial flux compression generator wasExtremely compact pulsed power sources designed to deliver sub- initially configured to operate at currents approaching 1 MA/cm ofnanosecond wide pulses are finding applications in impulse radar output circumference, ~98 MA. Los Alamos has never pursued thissystems and RF weapons. Impulse radar systems offer the realtime high current limit, and an experiment during early development thatadvantage of keeping and processing all the data in the time domain, minimally exceeded 50 MA is the highest current recorded on ainstead of transforming the data between the time and frequency Ranchero test. We are now in position to test at higher current levels. Adomains. RF weapons designed for close-proximity applications will variety of presently available computational tools is being used torequire high peak electric fields delivered from compact handheld predict performance of high current tests, and experiments are expecteddevices. Impulse solid state solutions are ideal, with peak electric fields before the conference. Tests using 43 cm and 1 m long Rancheroof 10’s kV/m and repetition rates in the kHz. Applied Physical modules are planned, and the focus prior to conference time will be onElectronics, L.C. has been developing a circuit card, solid state solution the shorter modules. Two load designs are under consideration. One ofwell suited for compact Ultra Wide Band (UWB) radiation. This paper these is a minimum inductance static load, and the other is andescribes the technology and presents experimental results. imploding liner load. The static load would have a total inductance of 0.32 nH, and the initial current would be supplied by a 2.4 MJ capacitor1P-86: Low Cost 400-Ps Rise Time Circuit-Board Marx bank at our explosives firing site. With an estimated 4.5 MA initialGenerator current from the capacitor bank, a peak of 85.1 MA is expected. ThereC. Nunnally, M. B. Lara, T. R. Smith, J. R. Mayes are different liner loads that we are considering, and depending largelyApplied Physical Electronics LC, Austin, TX, United States on how much the liner moves prior to peak current, the current may reach anywhere from 58 to 74 MA. The Ranchero stator will be tapered and will have insulation less that 1 mm thick. The OD of the RancheroSome modern pulsed-power applications benefit from a fast-rising armature is 152.4 mm, and the stator ID is 304.8 mm and 310.8 mm attrigger pulse which can minimize temporal jitter or ensure a desired the input and output respectively. Good agreement for currentmode (e.g. multi-channel spark gap) of breakdown ensues. Applied amplification is currently achieved between computations and dataPhysical Electronics, L. C., (APELC) has developed a low-cost 400-ps from an old test with a 5 nH load. The current peaked at ~40 MA onrise time Marx generator for low-energy triggering applications. The this test. Computations do not predict the current decay rate accurately,low-cost Marx is designed into a printed circuit board (PCB) geometry however. It appears that our codes do not correctly predict shorting of 65
  • 65. the ~1 mm stator insulator. As a result, in the computation the armature were 25 mm long by 25 mm diameter cylindrical solids with thebounces after contacting the stator insulation, which increases the total magnetic direction along the symmetry axis. Most of the experimentscircuit inductance and brings the current down faster than observed in were with the shock front vector aligned along the axis as well. Thisexperiments. Results of the upcoming high current tests are compared type of shot has been done and well published by others1. The shotto computations and projections for 1 m module performance are given. described by this paper aligned the shock wave vector perpendicular to the axis and by shaping, entered the magnet in a nearly radial inward fashion. Because the radial velocity of the shock was so much larger1P-89: Mini-G: the Development of an Optimized FCG than the relief wave from the two ends, the behavior had strong 2DDevice* character. This allowed an approximate modeling with the 2D hydro-D. B. Reisman code CALE in the x-y mode. CALE has a magneto-hydrodynamicLawrence Livermore National Laboratory, Livermore, CA, United modeling capability, but these were not used in the describedStates simulations. The authors were interested in the suggested temperature in the magnetic material behind the shock and how that compared toAt Lawrence Livermore National Laboratory we have developed a the magnetic phase diagram. It seems likely that the flux is released duecoupled helical-coaxial FCG device called the Full Function Test (FFT). to the material being heated beyond the magnetization boundary butThis device was used to deliver 98 MA of current and 66 MJ of energy probably below the Curie temperature. These observations areto an inductive load. The successful testing of the FFT represented the discussed. 1. S. I. Shkuratov, Et Al, See several papers in theculmination of an effort to establish a high-energy pulsed power proceedings of the 11th Megagauss Conference, September 2006.program that would greatly exceed the performance of capacitor bankfacilities. Using the modeling, design, and experimental capabilities 1P-92: Possible Mechanisms of Electric Field-Free Gasdeveloped for the FFT, we have developed a new generator, the Mini-G. BreakdownBased upon a half-scaling of the FFT device, the Mini-G is a coupled S. I. Shkuratov1, J. Baird1, E. F. Talantsev2, L. L. Altgilbers3helical-coaxial FCG capable of delivering up to 60 MA of current and 110 MJ of energy. We will describe the design of this generator which Loki Incorporated, Rolla, MO, U.S.A. 2involved the use of modern simulation codes as well as innovative Pulsed Power LLC, Lubbock, TX, U.S.A. 3pulsed power techniques to obtain a compact, highly optimized device. U.S. Army Space and Missile Defense Command, Huntsville, AL,*This work performed under the auspices of the U.S. Department of U.S.A.Energy by Lawrence Livermore National Laboratory under ContractDE-AC52-07NA27344. The phenomenon of electric field-free gas breakdown in explosively driven generators was recently experimentally documented [1,2]. In this paper we discuss possible mechanisms of this gas breakdown that1P-90: Measuring FCG Voltage Using an Electric Field occurs in the absence of electric fields in the systems. The firstAntenna mechanism is related to the ionization of the gas due to a sequence ofA. D. White, R. A. Anderson, J. B. Javedani, D. B. Reisman, shock processes caused by explosively expanding metallic partsD. A. Goerz (armatures, flyer plates, etc.) within the generators. The secondLawrence Livermore National Laboratory, Livermore, CA, United mechanism is related to the ejection of high-speed particles byStates explosively shocked metallic surfaces.A method of measuring the voltage produced by a helical explosive [1] S. I. Shkuratov, J. Baird, E. F. Talantsev, and L. L. Altgilbers,flux compression generator using a remote electric field antenna is "Electric Discharge Caused by Expanding Armatures in Fluxdescribed in detail. The diagnostic has been successfully implemented Compression Generators," Applied Physics Letters, Vol. 94, Article No.on several experiments. Measured data from the diagnostic compare 171502 (2009).favorably with voltages that were predicted using the code CAGEN. [2] S. I. Shkuratov, J. Baird, E. F. Talantsev, and L. L. Altgilbers,This method provides data that is important to understanding generator "Electric Field-Free Gas Breakdown in Explosively Drivenperformance, and does so using a low-risk, minimally intrusive Generators," Physics of Plasmas, vol. 17, Article No. 074504 (2010).approach.* This work performed under the auspices of the U. S. Department of 1P-93: Fabrication Process for Producing the Dual CavityEnergy by Lawrence Livermore National Laboratory under Contract Liner-Glide Plane Assembly Used on the MS-2 (Ranchero)DE-AC52-07NA27344 High Explosive Pulse Power Experiment Fired at Los Alamos National Laboratory on 2-18-20101P-91: A Simple, Nearly 2D Explosively Shocked NdFeB(52) R. B. RandolphPermanent Magnet and a Comparison to a CALE MST-7 / Materials Science and Technology, Los Alamos NationalCalculation Suggesting the Mechanism for Magnetic Flux Laboratory, Los Alamos, NM, United StatesRelease and Subsequent EMF PulseJ. B. Chase1, S. Ault2, D. Reisman3 This poster describes the fabrication techniques used for producing the1 Caren Co., Tracy, CA, United States dual cavity liner-glide plane portion of the MS-2 (Ranchero) high2 Hyperspectral Sciences, Inc., Cinebar, WA, United States explosive experimental system. Having a dual cavity design allowed3 Lawrence Livermore Laboratories, Livermore, CA, United States for a liner velocity measurement in the lower cavity and the experimental package to be located in the upper cavity. This newSeveral groups have investigated the phenomenon of the generation of fabrication method required a special glide plane alignment fixture thatpulsed electrical energy realized from the shocking of modern strong ensured precise axial alignment, concentricity and parallelism betweenpermanent Ferro magnets. The Care’n Co. together with Hyperspectral critical liner and glide plane features. The special alignment fixtureSciences, Inc. have performed several experiments, one of which this attached to the sub-spindle of a Hardinge T42 super-precision CNCpaper discusses in detail. The shot design and the diagnostics are lathe and was designed to hold the upper and middle glide planes atdescribed in the paper. The subject magnets in these investigations precise locations inside the liner bore. Utilizing the CNC lathes live 66
  • 66. turret tooling the glide planes were drilled and pinned before removing winding are given. Every one of six secondary windings operated onthe alignment fixture to perform final finish machining operations on inductive energy storage with electroexplosive opening switch andthe dual cavity assembly. The specially designed fixture, along with the capacitive load. In result, six small electric capacitors based on non-super-precision CNC lathe, allowed us to produce a very precise robust polar fluid (water) were charged up to about 300 kV for approximatelytarget assembly without damaging critical surface features necessary to 100 ns with time spacing about 10 us.meet the stringent physics requirements of this experiment. LA-UR 11-00086 1P-97:Study of a Possibility to Get Spherical Symmetry of a Quasi-Spherical Liner Implosion under the Effect of Axial1P-94: Analysis of the Impact of Mutual Inductance on the Magnetic FieldMFCG Primary Pulse Current Magnification B. T. Egorychev, P. V. Duday, A. V. Ivanovsky, A. I. Kraev,C. Yu V. B. Kudelkin, A. N. SkobelevBeijing Institute of Special Electromechanical Technology, Beijing, RFNC-VNIIEF, Sarov, Nizhny Novgorod Region, Russian FederationChina One of the known methods to study plasma is to use a linerThis article builds a circuit model of Magnetic Flux Compression ponderomotive unit as the energy releasing device. The ponderomotiveGenerator (MFCG) with mutual inductance. The effect of mutual unit is intended for a transformation of the magnetic field energy toinductance on the primary current amplification is analyzed, and the kinetic energy of the axially-symmetric shell that realizes compressionrelationship of the current magnification and the coupling coefficient of and heating of plasma. The so-called energy liner is a common membermutual inductance is obtained. Two kinds of MFCG were designed and in different types of ponderomotive units. In the simplest case itcompared with each other. The correctness of theoretical analysis is represents a cylindrical conducting shell imploding towards the axisvalidated by experiments, and the result shows that the current under the effect of current flowing along the axis in this shell (metal Z-magnification can be improved by increasing the mutual inductance pinch). The calculations demonstrate that a changeover from cylindricalcoil. to spherical (or quasi-spherical) geometry considerably mitigates conditions of high temperature and plasma density achievement.1P-95: Peculiarities of Formation of a Conical Piston at the Plasma compression with the use of the quasi-spherical liner allows preserving positive properties of spherical compression and at this toHelical-Conical EMG Input have bigger methodical possibilities of experiment diagnostics used inB. T. Egorychev, P. V. Duday, A. V. Ivanovsky, V. B. Kudelkin, the cylindrical liner studies. This paper will describe the experimentalN. I. Sitnikova, A. N. Skobelev data based on the results of studies of the process of quasi-sphericalRFNC-VNIIEF, Sarov, Nizhny Novgorod Region, Russian Federation liner compression. At the present time, when the axisymmetrical magnetic field is used to drive the quasi-spherical liner, the achievedIt is well known that the operation of the helical EMG of megampere symmetry of the liner approach is about ~ 1%. However, in the processlevel requires its powering with initial current of tens of kiloamperes. of liner acceleration its thickness becomes nonuniform due to Raleigh-High-power stationary capacitor facilities are used for this purpose. In Taylor instabilities. Such plasma compression system requires furthersome cases there are limitations on the amplitude of the powering more detailed study. Such studies can be realized with the use of thecurrent, for example, when using a small capacitor or accumulator bank. diagnostic tools (presented in this paper) created by RFNC-VNIIEF.Under these conditions it is expedient to increase its initial inductancein order to preserve the output parameters of the helical EMG. One of 1P-98: The Effects of Stator Insulation Material and Methodsthe possible ways of solution of this problem is the application of aconical helical coil in the input part of a helical EMG; the base of this of Fabrication on the Performance of Compact Helical Fluxcoil is connected with a cylindrical part of the helix and the apex is Compression Generatorsfacing the helical EMG input. The helical conical coil is installed C. S. Anderson, A. A. Neuber, M. A. Elsayed, A. J. Younginstead of an authorized commutator in the prior dimensions of the Center for Pulsed Power and Power Electronics, Texas Techhelical EMG. The initial inductance of the helical EMG increases University, Lubbock, Texas, United Statesapproximately by a factor of two. In the new design the part of the HEcharge not used earlier will be used almost completely to realize Helical Flux Compression Generators, HFCGs, are powerful highmagnetic cumulation. The paper will present the calculated data that current sources for pulsed power applications. Due to the single shotclearly demonstrate the peculiarities of the central armature expansion nature of HFCGs, electrical output reproducibility is of greatat the initial section of the helical EMG with the installed conical importance. One factor known to contribute to unpredictablehelical coil. The scope of the HE explosion products pressure unloading performance is mechanical inconsistencies introduced duringat the armature end will be shown. The visual data on the profile of the manufacturing of the stator. In an attempt to minimize these deviationsconical piston formed at the initial stage of magnetic cumulation will be during productions, two different winding forms for stator coils,presented. The calculated values will be compared with the designed to ensure repeatable generator dimensions, turn and coil pitch,experimental data. The conclusion will be made on workability of the were investigated. The first mandrel considered was assembled fromconsidered design with high efficiency. several pieces of steel, which once the coil had been manufactured, were collapsed and removed from inside of the coil, and then1P-96: Power Pulser for Generation of a Series of High reassembled, enabling the mandrel to be used for multiple generators.Voltage Pulses Based on Multi-Winding Mc-Generator The second winding form considered was a single use mandrel madeK. Gorbachev, E. Nesterov, V. Stroganov, E. Chernykh from hard machinable wax, which, once the stator was fabricated, was either melted or machined out from inside the coil. The differencesJoint Institute for High Temperatures of RAS (JIHT RAS), Moscow, between the methods were quantified by comparison of measurementsRussian Federation made of the physical parameters of the coil (i.e. radius, inductance, etc.), as well as analysis of experiments conducted with the HFCGsExperimental results on generation of a series of six high-current pulses fired into a 3 μH load inductor.by use of multi-winding dynamic transformer with a single primary Choosing any particular fabrication method, the stator insulation 67
  • 67. material has a distinct impact on generator operation. Based on addresses possible factors that might contribute to this negativeprevious experiments, HFCGs that utilized Teflon performance. For instance, we will quantify and discuss thepolytetrafluoroethylene (PTFE) and PVC for wire insulation exhibited redistribution of current in a wire as a result of the magnetic fieldsfluctuations in output energy as much as 50% which was partly produced by other wires within the fuse array typically consisting of 18attributed to electrical breakdown in the stator during the run-time. wires for a current action integral of 4500 kA2 µs and an equivalentSeveral stator insulation materials were investigated to improve the action timescale of 2.65 μs. An electro-magnetic field solver is used toHFCG output repeatability and performance, including Quad-built model the current redistribution in the fuse wires. One compact fusePolyimide coated magnet wire, Teflon Fluorinated Ethylene Propylene with smaller wire spacings (from 6.35 mm to 16.95 mm distance(FEP) and High-density Polyethylene (HDPE). Preliminary results have between adjacent wires) as well as a trifold fuse are compared and theyielded significant decreases in generator failure due to breakdown, data from the parasitic analysis is compared with the experimental data.enabling seed energies of greater than 250 J to produce output energiesin the 4 kJ range. All stator insulation testing was carried out by firing 1P-101: A Novel Type of MFCG with Mutual Inductancethe HFCGs into the inductive load mentioned above. Experimental dataand analysis, as well as conclusions on insulation material, will be Coilspresented along with a brief discussion of the optimum fabrication C. Yumethod. Beijing Institute of Special Electromechanical Technology, Beijing, China1P-99: Explosive Current Opening Switch with Variable The article introduces a novel type of Magnetic Flux CompressionTopology of Current Path Generator(MFCG) with mutual inductance coils . The novel MFCG sI. V. Morozov, V. I. Dudin structure is based on column cone coaxial type. The design of mutualRFNC-VNIIEF, Sarov, Russian Federation inductance coils for the nove MFCG is discussed, and the related parameters are analyzed. Experiments are carried out to verify theIt is proposed to recede from a traditional scheme of constructing performance of the new MFCG . The results of experiments show thatexplosive opening switches (EOS)/ i.e. cutting the conductor (foil) with it can improve the primary current magnification by increasing thea big number of plane shaped cumulative jets or ribbed barriers across mutual inductance coil.the current. It is proposed not to cut the foil completely, but to changeits geometry. A narrow and proportionally longer conductor will be 1P-102: Newton & Einstein and Others for Pulsed Powercarved from a wide and short one. It is necessary that in the process ofheating with current the resistance of the new conductor due to smaller Inside Hydrogens Atomcross section would increase so fast that the current could be transferred N. T. Elfikkyto the load for the time of interest to us. In this way, we radically Extra High Voltage Div, Saudi Electricity Company-Central Regionchange the nature of release of inductive energy that the EOS should Branch, Riyadh 11411, Saudi Arabiadissipate. If in the conventional EOS the energy is released exactlyunder the blade of a dielectric knife and it is necessary to cut plasma Preface. The basic data for Hydrogen’s 1H1 atom like space distanceformation with the temperature of several eV and pressure of tens of between Electron and Proton and linear speed of whirling for bothkilobars, then in the new design the energy is released in the whole proton and electron. To identify the parameters controlling the pulsedmass of the conductor and cutting will be applied to the relatively cold power inside Hydrogen atom in its steady state as a function of atomconductor. The operation of the new EOS potentially provides an basic data, we will employ many physical laws which are the windowsopportunity to reduce the opening time, while reducing the length and of quantum mechanics. - Brief. 1-As known, the simplest stable atomamount of HE in the EOS. The results of the experimental testing are known to date is the hydrogen 1H1 element since it has a simplediscussed. nucleus with one proton balanced with one electron whirling on the orbit around. The orbits are imaginary spherical loci surrounding the1P-100: Compact Electro-Explosive Fuse Optimization for a nucleus in each atom. Each orbit is quantized and equi-energy locus i.e.Helical Flux Compression Generator the energy level is constant or symmetrical along the circumference ofJ. C. Stephens, A. A. Neuber, J. C. Dickens, M. Kristiansen the orbit and the electrons spinning in the same orbit will have the same angular momentum (energy level) irrespective of the number ofCenter for Pulsed Power and Power Electronics, Texas Tech electrons whirling on the same orbit. For the nucleus, we will considerUniversity, Lubbock, TX, United States the hypothesis of the physicist Rutherford, where he proposed a dynamic planetary model in which nucleus plays the role of the sun andThis paper presents the optimization of a compact electro-explosive electrons correspond to the individual planets of the solar system. 2-fuse designed for the power conditioning system to be driven by a Why is the electron in Hydrogen atom restricted orbiting the nucleus athelical flux compression generator (HFCG). For fuse optimization, a certain fixed distance???. As known, electrons are rotating on the orbitsnon-explosive test bed is utilized that is designed to emulate the quasi- circumferences, so it will gain centrifugal forces and kinetic energy dueexponential high current (~40 kA), fast rise time (~10 µs) output of an to its rotational speed, If the speed of the electron will be at least equalHFCG. The non-explosive test bed enables testing and development of to one half of light speed, the electron mass will be effective and willelectro-explosive fuses without the cost or time investment associated have an effective mass. But according to the mass of Hydrogen 1H1with an explosively driven HFCG, a true one-time use device. A few atom, it is clear that the proton 1p1 mass inside 1H1 atom is equal to 99%micro-Henry inductor is used to store the energy from the HFCG or test from the atom’s mass. . Also,the Hydrogen element 1H1 is a fluidbed. An electro-explosive fuse interrupts the current flow through the element and it is always diffusing at the ambient temperature to occupyinductor on a 50 to 100 ns timescale inducing a voltage large enough to all the available space inside a vessel, even, also diffusing betweenclose the peaking gap, which commutates the energy in the storage another gases without any change of its intrinisic physicalinductor into a 20 Ohm load at voltage levels above 200 kV. characteristics i.e. the Hydrogen 1H1 atom at steady state will never beExperimental data has revealed that electro-explosive fuses with wires separated to proton and electron and the stable atom will never bein closer proximity to one another have consistently performed to a transformed to positive ion or negative ion in the normal steady state atlesser standard than fuses with larger wire spacings. This paper ambient temperature conditions. i.e. there is no separate translational 68
  • 68. motion for the proton or the electron inside 1H1 atom which is 2A-2:COMSED 2 - Recent Advances to an Explosivelyconfirming that there is a balanced compacted pulsating forces Driven High Power Microwave Pulsed Power Systeminteracting between both particles inside 1H1 atom in the steady state M. A. Elsayed, A. A. Neuber, A. J. Young, J. W. Walter,although the electron is continuously rotating on its orbit and attracted C. S. Anderson, S. L. Holt, J. R. Korn, M. Kristiansenwith the “motionless” proton due to their opposite equal charges. But, Center for Pulsed Power and Power Electronics, Texas Techaccording to Newton’s third law “The mutual forces of action and University, Lubbock, Texas, United Statesreaction between two particles are equal and opposite and collinear.”and the GUT’s which states that the proton can’t stay or remain in standstill. In fact, there is a reasonable explanation for the advanced question Continued efforts at the Center for Pulsed Power and Power Electronicsand I am looking to introduce. 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 polypropylene2A: Explosive and Compact Pulsed Power II capacitor, and high energetics. An indirectly seeded two-stage helicalMonday, June 20 15:30-17:30 CC 10A-B 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 driven2A-1: (INVITED) Design Considerations for Flux-Trapping generator is conditioned with a pure silver-wire-based electro-explosiveHelical-Flux Compression Generators Energized by opening switch, which reaches voltage levels in excess of 300 kV into aCapacitive Discharge 18 Ohm load. Upon reaching levels high enough to close an integratedA. Young, A. Neuber, M. Kristiansen peaking switch, this high voltage is sufficient to drive a reflex triodeCenter for Pulsed Power and Power Electronics, Texas Tech virtual cathode oscillator, also known as a Vircator, into radiation. TheUniversity, Lubbock, Texas, United States Vircator employed in the system has reached microwave radiation levels well over 100 Megawatts from a cavity volume of less than 5Helical flux compression generators (HFCGs) are not stand-alone Liters at a microwave frequency of a few GHz. The complete system isdevices, and require initial electrical energy for compression. Thus, a governed through a microcontroller that regulates seed and detonatorprime power source must always be used to energize the generator prior charging levels as well as discharge times using built-in feedbackto detonation of the explosives. The most common energy sources for diagnostics. The complete system aforementioned is constrained to 15HFCGs are capacitor based, and in these systems the choice of source centimeter diameter and measures 183 centimeter in length with anparameters are constrained by the initial inductance and energy overall volume of less than 34 Liters. No external power or vacuumrequirements of the HFCG. The use of flux-trapping HFCGs (FT- pumping for the HPM tube is required. This report will discuss recentHFCGs) enables greater design freedom in the system by indirectly design advancements and improvements on the HPM system and itscoupling the prime power source to the HFCG through a field coil. The sub-components that include the compact seed source, HFCG, and theaddition of the field coil, however, also adds a higher degree of power conditioning system. Improved safety features implemented ascomplication to the overall circuit. Choosing values for the source well as novel diagnostic integration will be discussed as well.(capacitor), field coil and HFCG that give the most efficient systemperformance is extremely difficult, given the nearly infinite number of Numerical Simulation of Electromagnetic Flux 2A-3:possible combinations between the three. Such an optimization study is Compression for Super-Intense Magnetic Field Generationtoo costly to be carried out experimentally, and therefore is most W. Jiang1, H. Sugiyama1, A. Miyata2, H. Sawabe2, Y. Matsuda2,effectively performed using theory or simulation. An FT-HFCG S. Takeyama2simulation code, previously benchmarked with single and multi-pitch 1 Department of Electrical Engineering, Nagaoka University ofgenerators, was employed to study a kilo-joule class FT-HFCG system Technology, Nagaoka, Japanfor this purpose. The details of this effort, which includes the 2 International MegaGauss Science Laboratory, Institute for Solid Stateoptimization of the field coil, stator, the coupling between them, the Physics, The University of Tokyo, Kashiwa, Japaneffect of varying source and field coil parameters on systemperformance, and crowbar timing in FT-HFCGs, will be described inthis document. The study revealed that for multi-pitch FT-HFCGs, the An electromagnetic flux compression system targeting magnetic fieldhighest coupling may be achieved when the field coil covers only a of 1000 T is being developed at the Institute for Solid State Physics ofcertain length of the stator, and while the coupling is more dependent The University of Tokyo. In order to assist the experimental systemon the length of the field coil rather than the inductance, the energy design, numerical simulations have been carried out to study thegain of the system is highly dependent on the field coil and prime process of liner implosion driven by pulsed current on the order of MA.power source parameters. As will be seen, variation of the field coil The two-dimensional model included the effects of liner dynamics,inductance, while keeping the HFCG coupling and source parameters driving coil configuration, and external circuit characteristics. Theconstant, resulted in a 300% variation in the system energy gain over simulation results have been compared with previous experimentalthe simulated parameter space. The simulations also revealed that, results in order to confirm its validity. It is then used to predict thecontrary to conventional HFCG systems, the optimal crowbar time in performance of the new system to be built and to provide the designFT-HFCGs is not always at the peak of the discharge current, and is requirements for the driving capacitor bank.also highly dependent on the prime power-field coil circuit parameters.Experimental data will be given, where appropriate, that confirms the 2A-4: Miniature 100-kV Explosively Driven Prime Powerresults of simulations. Conclusions from the full analysis of this study Sources Based on Pb(Zr0.95Ti0.05)O3 Ferroelectricwill be given in the form of guidelines which can be used whenconsidering possible designs for FT-HFCG systems. Ceramics S. I. Shkuratov1, J. Baird1, E. F. Alberta2, W. S. Hackenberger2, A. H. Stults3, L. L. Altgilbers4 1 Loki Incorporated, Rolla, MO, United States 2 TRS Technologies, Inc, State College, PA, United States 3 U.S. Army Aviation Research, Development and Engineering Center, Huntsville, AL, United States 69
  • 69. 4U.S. Army Space and Missile Defense Command, Huntsville, AL, and disadvantages. One possible significant disadvantage of manyUnited States generators is high g-force survivability. To address this issue, a different FCG design is being developed. This new generator, called aA new technology for the production of high-quality lead zirconate Shock Wave Generator (SWG), replaces the standard air filledtitanate Pb(Zr0.95Ti0.05)O3 (PZT 95/5) ferroelectric ceramics for armature with a powder fill, which under pressure transitions from aexplosive pulsed power applications has been under development by dielectric to a metallic state. Unlike previous SWG designs, which wereTRS Technologies for the past few years [1,2]. Their PZT 95/5 implosive, these generators are explosive; i.e., the metallic shock waveceramics was previously tested by HEM Technologies [1,3] and generated in the powder is moving radially outward. This movingKTECH Corp. [1]. In the present paper, we report on a new approach metallic shock front is used to convert the chemical energy of highfor utilizing TRS’s PZT 95/5 ceramics in explosive pulsed power. We explosives into electrical energy. In this paper, we will report on recentpropose new operational ideas for high voltage shock-wave experimental studies of these generators.ferroelectric generators (FEGs). Loki Inc. designed, constructed andtested several series of explosively driven generators based on these Cylindrical Ferro Electric Generators Waveshaping 2A-7:ideas. As a result of these efforts, Loki developed miniature shock Techniques and Performancewave FEGs having diameter 38 mm and a total volume less 100 cm3 Z. S. Roberts1, F. Rose1, S. Rendall1, L. Altgilbers2, A. Stults3,that are capable of producing output voltage pulses with amplitudes J. Sweitzer3exceeding 100 kV. The duration of these high-voltage pulses ranges 1 Radiance Technologies, Huntsville, Alabama, United Statesfrom 3 to 4 ms. 2 USASMDC, Huntsville, Alabama, United States 3 AMRDEC, Huntsville, Alabama, United States[1] E. F. Alberta, B. Michaud, W. S. Hackenberger, B. Freeman, D. J.Hemmert, A. H. Stults, L. L. Altgilbers, in PPC2009 - 17th IEEEInternational Pulsed Power Conference 2009, Article number 5386193, This paper will review a current research and development effort aimedpp. 161-166 (2009). at developing a small form factor cylindrical PZT Ferro Electric[2] E. F. Alberta, R. Guo, A. S. Bhalla, Ferroelectric Review, Vol. 3, pp. Generator (FEG). Two options will be discussed – shaping the PZT1-327 (2001). crystal into a conic section and shaping the acoustic pulse from the[3] S. L. Holt, J. T. Krile, D. J. Hemmert, W. S. Hackenberger, E. F. explosive wave train, both techniques will be discussed in terms of easeAlberta, J. W. Walter, J. C. Dickens, L. L. Altgilbers, A. H. Stults, in of implementation, fabrication and performance achieved. We will alsoPPPS-2007 – 16th Pulsed Power Plasma Science 2007, Vol. 2, Article discuss corona induced failure in FEG devices and methods to mitigatenumber 4652397, pp. 1177-1180 (2007). this effect. Representative data will be presented for FEG devices made from EC-64 and 95/5 PZT for both geometries. Scaling relations will be presented and discussed in terms of physical dimensions and2A-5:Ferroelectric Generator Design for Multiple Initiation materials.of Blasting CapsA. H. Stults 2B: Microwaves II: Microwave and RF Sources, AntennaeAviation and Missile Research Development and Engineering and SystemsLaboratory, Redstone Arsenal, AL, United States Monday, June 20 15:30-17:30 CC 10C-DFerroelectric generators (FEG) have been used in the past because oftheir excellent high voltage characteristics. As the anti-ferroelectric - Operational Performance of the Horizontal Fast Rise 2B-1:ferroelectric phase change of PZT 95/5 has substantial current EMP Pulser at the Patuxent River EMP Test Facilitygeneration capabilities, a switchless FEG has been designed to have D. W. Belt, A. D. Mazucsuitable characteristics that initiate arrays of commercial blasting caps. 5.4.4.6, Naval Air Systems Command Aircraft Division, Patuxent River,This unique FEG and arrays of blasting caps have been tested in many MD, United Statesconfigurations and data is presented for 36, 64, and 128 cap arrays inseries. These arrays represent an inductive load for the FEG and For the past two years, the Naval EMP test facility located at theconstitute the major factor in the coupled response of the FEG-array Patuxent River Naval Air station has been working with the L-3 Pulsecircuit. The switchless design shows that reasonable characteristics are Sciences Division to develop and procure the new Horizontal Fast Riseutilized through good explosive design of the plane-wave generator EMP (HFREMP) pulser. The HFREMP pulser was designed toversus the more common peaking switch approach. compliment and upgrade the Horizontally Polarized Dipole (HPD)Description of the electrical characteristics of commercial blasting caps simulator and will greatly improve its test capabilities for future EMPwill be explained in terms of their initiation requirements. The testing. The new HFREMP pulser utilizes two battery charged, dualresistance and inductance of the cap arrays will be presented for 36, 64, opposing Marx generators (+/- 3 MV) with two pulse compressionand 128 cap arrays. The resulting waveforms for both voltage and circuits per Marx, to generate an EMP pulse with a peak amplitude ofcurrent in the FEG-blasting cap system will be shown and explained. up to 77 kV/m at a distance of 24.5 m away from the source. Combined with the two dual compression circuits, a 200 PSIG center switch2A-6: Shock Wave Generators allows the HFREMP pulser to generate a fast rise time of 1.3 ns to 2.8B. L. Freeman1, G. C. Newsom1, J. W. Guthrie1, L. L. Altgilbers2, ns. By utilizing the multiple onboard diagnostics and free fieldM. S. Rader2 measurements, the operators are able to adjust various switch timing1 parameters/settings to achieve a multitude of peak amplitudes, rise Directed Energy Div, Ktech Corporation, Albuquerque, NM, UnitedStates times, and Full Width Half Maximum (FWHM) values required for2 EMP testing. In this paper, we will examine a few of the operational SMDC, US Army, Huntsville, AL, United States parameters and the resulting performance, demonstrating the versatility and capability of the new HFREMP pulser.Flux compression generators are high-energy power supplies that arecapable of delivering kilojoules to megajoules of energy to loads. Thereare different types of generators, where each type has its advantages 70
  • 70. High Power Microwave Threat Simulator Facility at2B-2: A draw back for some applications exist where the available number ofWhite Sands Missile Range array apertures is small, for it defines the ratio of bandwidth toR. Blundell repetition rate.Department of Army, White Sands, United States Performances of an Ultra Compact, High-Power, 2B-4:The U.S. Army has recently commissioned a new high power Monocycle Pulse Former for WB and UWB Applications.microwave threat simulator facility to support electromagnetic High P. Delmote, J.-P. Dupéroux, F. Bieth, S. PinguetPower Microwave effects testing activities at WSMR. The threat EMW, Institut franco-allemand de recherches de Saint-Louis (ISL),simulator employs Reltron microwave tubes built by L3PS. These Saint Louis, Francesources operate in burst mode at low repetition rate. The high voltagedrive for the Reltron tubes is a Marx/PFN. The system is capable of This paper presents the design and the performances of an ultraproviding threat level field strengths over a large illumination area at 15 compact, general-purpose, and high-power ultra-wide band (UWB)meters from the source. The test volume is an open portion of the test source named GIMLI. The system was developed for dual use, fromrange, allowing for a variety of test configurations to be considered. laboratory to battlefield applications. The power supply is a coaxialWhen fully implemented, the threat simulator will cover the frequency Marx generator composed of specifically designed stages. In a 12 stagerange from 550 MHz to 3 GHz. configuration, the rise time can be less than 20 ns (measured on a 50 ohm load) with an operating voltage reaching values up to 360 kV A Novel HPM Array System Based on Mode Locking2B-3: (with an open circuit configuration). An ultra compact (less than 2 litres)Multi Frequency pulse forming stage (PFS) is directly connected to the output of theO. S. Zucker1, P. K. L. Yu2 Marx generator. It is composed of a pulse sharpening assembly made1 Polarix Corporation, San Diego, CA, United States up of a peaking and a multi channel grounding spark gap running under2 University of California, San Diego, La Jolla, CA, United States a high pressure of nitrogen. These switches are followed by a monopulse-to-monocycle converter module which is based on a Blumlein coaxial line. The bipolar signal at the output of the PFS has aHPM systems fall into two general categories. The first uses total duration which can be adjusted from 1 to a few ns by the use ofmoderately high power, quasi CW sources in the MW range with pulse different lengths of Blumlein module. For example, the smallest deviceduration around a millisecond and are generally of narrow bandwidth generates a signal composed of two Gaussian pulses. A positive one isbut with large average power. In the second category we have very high followed by a negative one and the two are separated by less than 500Peak power – in the GW range, but of very short pulse duration, ps with rise times less than 250 ps. Measured on a 50 ohm dedicatedtypically in the 100ns range, and are of a large bandwidth. These ultra wideband resistive load, the peak-to-peak output voltage issources, due to their pulse power technology are limited in average tunable up to 400 kV. With a right adjustment (in pressure and inpower and pulse repetition frequency. Some HPM applications require distances inside the electrode chamber) the maximum dV/dt can reachbeam steering either by mechanical means or by phasing an array. 2.10^15 V/s. If a more important slope is required, it is possible toSome applications, such as counter electronics, require the continuous insert a pre-peaking stage between the Marx generator and the PFS.sweeping of the beam over a large forward area. Sweeping arrays Using this stage allows to get performances in the order of 5.10^15 V/srequire phase shifters that are costly and difficult if they are required to (rise times lower than 150 ps) but to the detriment of the maximumaccommodate a large bandwidth and Power. peak to peak level which decreases to 200 kV. The use of a coaxial 50This paper describes a new and unique HPM system with the following Ω output enables to connect the GIMLI source to many different typescharacteristics: i)it utilizes an ensemble n, of narrow band high power of antennas, allowing the radiation of WB or UWB electromagneticCW sources each of a different frequency, ω1, ω2, ...ωn; ii) the signals. For instance, high-power radiation tests were performed withindividual source frequencies are incremented in frequency by a fixed the pulser connected to a specific half TEM ridged horn. The resultsincrement Δω and all are harmonically locked (as in a mode locked showed that the electrical field acquired at 10 m was higher than 140laser); iii) each of these sources feed a specific single radiating aperture kV/m peak to peak.of the array. Thus the apertures are incremented in space and infrequency simultaneously.Such a system produces multiple interference patterns; all travel 2B-5:Delay Time Distribution of High Power Microwavetransversely to the Poynting vector at a constant velocity Δω/Δk, which Surface Flashoverinterfere with each other to produce a narrow angled beam consistent J. Foster, H. Krompholz, A. Neuberwith the array gain. This search light fashion sweep repeats at Center for Pulsed Power and Power Electronics, Departments offrequency Δω produces temporal pulses at the target with temporal Electrical and Computer Engineering, Texas Tech University, Lubbock,width of 1/nΔω at a repetition rate of Δω, and with frequency TX, United Statesϖ=ω1+nΔω/2.In one example, a 3GHz system using 100 array apertures powered Breakdown phenomena in a high power microwave (HPM) systemeach by one narrow band 1MW CW source, with individual frequencies present unique obstacles for the further development of HPM systemsincremented at Δω=1MHz. The resulting beam frequency, pulse width, and technologies. The non-uniformity of a high frequency electric fieldpower, beam angle, repetition frequency and sweep rates are; 3.05GHz, and the statistics associated with breakdown in general along with the10ns, 100MW, 0.01radians, 1MHz and 106 Radians/s, respectively. stochastic nature of naturally occurring electron generating mechanismsA variation of the concept for UAV platforms and short range have introduce significant challenges for predicting as well as preventingcorresponding lower power but need to operate above 10GHz to allow breakdown occurrences within an HPM system. Characterizing andfor sufficient number of apertures. explaining the initial stages of breakdown without an obvious source ofThe advantages of the system are as follows: i) single frequency, CW initiatory electrons has proven to be difficult due to the unexplored(narrow band amplifiers) sources are simple and cheap; ii) digital processes involved. An experiment utilizing an S-band multi-megawattfrequency synthesizers to produced individual phase locked frequencies HPM pulse is used for observing and characterizing an alternating fieldare common in the wireless field; iii) no phase shifters are used; iv) induced plasma sheath across a dielectric surface in a controlledarbitrarily short pulses and very high sweep rates are produced simply atmospheric environment. In order to minimize experimental deviations,by choosing the appropriate n and Δω. 71
  • 71. an external continuous wave UV lamp is used to provide a relatively A Mobile Pulsed Ring-Down Source Array Using Low 2B-7:constant source of initiatory electrons through the process of Power Solid State Radiatorsphotoemission. This drastically reduces the statistical waiting time for D. Reale, J. Mankowski, S. Holt, J. Walter, J. Dickensflashover initiating electrons to appear in the high field region, however, Center for Pulsed Power and Power Electronics, Dept. of Electricalprimarily due to avalanche statistics, variations are still observed and Computer Engineering, Texas Tech University, Lubbock, TX,experimentally. A statistical model has been developed to describe the United Statesvariations in measured flashover delay times (i.e. the time from theapplication of the HPM pulse to the sharp drop in transmitted power) invarious gases and pressures. This model utilizes an exponential We are currently investigating the technologies required for a mobiledistribution sampling procedure to predict the surface flashover delay Pulsed Ring-Down Array. In particular, the capabilities of commericaltimes for a variety of conditions. In the absence of a constant source of of the shelf (COTS) GPS positioning and timing systems for a pulsedbreakdown initiating electrons, it is assumed that field detachment from ring down array were evaluated. Additionally, Monte Carlo simulationsion clusters is the primary source for initial electrons. The theoretical are performed to evaluate the potential performance of the array usingmodel can be used to characterize this generation rate of electrons due current GPS capabilities and to determine the necessary requirements toto field detachment based on the observed variations in flashover delay achieve the desired system performance. Error sources within thistimes. A supporting experiment that utilizes a continuous wave UV simulation include the position error of each element, the jitter of thelamp and a DC electric field is used to measure low level current due to closing switch, and the timing synchronization error between elements.photoemission from the dielectric window. Photoemission from Error distributions for the simulation are based on the results of stand-dielectrics has not been explored in detail in the past, and a decisive alone device testing. The array is tested by adding an artificial delayobservation is a strong pressure and gas dependence of the UV-induced that represents an error source. The simulation is verified by runningelectron flux, probably due to backscattering and subsequent surface each error source independently and comparing the results to theattachment of low-energy photoelectrons. An explanation of the model experimental data. A system of low power radiators based on a soliddescribing these phenomena is presented along with a comparison of state stacked MOSFET switch were built to represent the Pulsed Ringcalculated and experimental results. Estimations of seed electron Down Source (PRDS) elements for experimental purposes. The testinggeneration rates due to UV illumination and field detachment are also methodology and results are presented.given for various gases and pressures. Reducing Both the Physical Size and Operational 2B-8: Investigation of the Transmission Properties of High2B-6: Frequency of Helical Antennas by Means of DielectricPower Microwave Induced Surface Flashover Plasma LoadingS. Beeson, J. Foster, J. Dickens, A. Neuber M. B. Young, K. A. OConnor, D. A. Crosby, R. D. CurryElectrical and Computer Engineering, Texas Tech Univeristy, Lubbock, Dept. of Electrical & Computer Engineering, University of Missouri,United States Columbia, MO, United StatesThe generation of a high power microwave (HPM) pulse in a vacuum Helical antennas have long been known as an excellent choice for pointenvironment for the purpose of radiating into atmosphere, necessitates to point communications and are commonly used for transmitting andthe use of a dielectric window to separate the vacuum side from the receiving signals greater than 500MHz. As the operational frequency ofatmospheric side of the system. Under certain conditions (high RF-field helical antennas decreases, their physical size increases. Air-coreand/or low pressure) surface flashover develops across the atmospheric helical antennas that operate below 500MHz are inherently large, soside of the window resulting in a severe loss in HPM transmission. An large that they are impractical for everyday use. In order to overcomeexperimental setup has been constructed to observe pulsed RF-plasma this issue, the University of Missouri-Columbias Center for Physicalparameters such as transmission and absorption coefficients as well as and Power Electronics has been actively researching the properties ofplasma relaxation times. The setup utilizes an S-band magnetron dielectric loaded antennas. Based on laboratory experiments, it hasoperating at 2.85 GHz to produce a 4 MW, 3 µs pulse for inducing been found that the addition of a dielectric material within the core ofdielectric window flashover in atmospheric gases (Air, N2, and Argon the helix will translate into both a reduction in the operationalfor basic studies) at pressures ranging from 10 to 200 Torr. Using a frequency of the antenna and a reduction in its physical size. Using thishigh power microwave switch, the rise time of the pulse is shortened to knowledge, a 22 turn, 2.413cm diameter by 15.24cm tall helicalfrom ~600 ns to ~50 ns to allow for better comparison between antenna was designed around a core with a dielectric constant of εr=45experimental results and theoretical models that frequently assume an to achieve an operational frequency of 495MHz. This results in an 87%ideal square pulse. With air as background gas, the transmitted power reduction in operational frequency and a 99.8% reduction in thewas measured to be attenuated by 50 dB at 10 Torr and 20 dB at 155 antennas physical volume compared to that of a 495MHz air-coreTorr. While it was previously demonstrated that adding sub-mm sized helical. CST Microwave Studio was used to simulate the antennametallic points to the surface of the window reduced flashover delays designs derived from modifications to Krausss formulas. This papertimes by as much as 60%, the transmitted power levels were largely describes the simulation, design, construction, and results of a dielectricunaffected after flashover occurred (attenuation values within 0.1 dB of loaded helical antenna that yields a significant reduction in both itsthe values mentioned above and within apparatus resolution). physical size and operational frequency.Additionally, a continuous wave (CW) 10 GHz, low power (~1 W) This work was supported by ONR under contract no. N00014-08-1-source is used as an external signal for probing the 2.85 GHz HPM 0267.induced plasma in order to determine the plasma relaxation time andbetter characterize the transmission coefficients of the plasma.Presented in this paper is a discussion of the pressure dependenttransmission at 2.85 GHz and the CW source implementation alongwith relevant experimental results for the aforementioned gases andpressures. 72
  • 72. 2C: Pulsed Power Systems I: Generators and Networks pulse width (FWHM) are 37 ns and 65 ns, respectively. The basic principle of this technique and the implementation will be discussed inMonday, June 20 15:30-17:45 CC 11A-B this paper.2C-1:Status of the 2 MA Driver for Creating 2 MG Magnetic High Power HV Generators of Sequential Two 2C-3:Fields for Cluster Fusion Experiments 1 2 3K. W. Struve , J. W. Argo , R. D. Bengtson , D. I. Headley , 1 Nanosecond PulsesJ. W. Kellogg1, S. M. Lewis3, H. J. Quevedo3, M. E. Savage1, G. Remnev, Y. UsovB. S. Stoltzfus1, C. J. Waugh1, M. Wisher3 National Research Tomsk Polytechnic University, Tomsk, Russian1 Pulsed Power Sciences Center, Sandia National Laboratories, FederationAlbuquerque, NM, United States2 Ktech Corporation, Albuquerque, NM, United States There are different applications of high power HV generators of bipolar3 Dept. of Physics, University of Texas, Austin, TX, United States pulses as well as sequential two pulses when pulses of the same or opposite polarities are divided by tens of nanoseconds to hundreds ofNow that the 400 kA version of a pulsed-power driver for magnetic microseconds. High power HV generators of bipolar pulses werecoils for cluster fusion experiments has been completed, we are developed in seventies as feeding sources of streamer chambers andcurrently assembling a 2 MA version for generating magnetic fields up means for measuring ion charges and their mass in the explosiveto 2 MG. Both versions are intended to drive single-turn, single-use 1- emission plasma of the vacuum discharge. Afterwards, differentcm diameter magnetic field coils that enclose laser-produced, high- versions of special generators of two sequential pulses with time spandensity deuterium cluster plasmas in vacuum. The driver is being built in nanosecond range were used for explosive emission plasmaat the Sandia National Laboratories* and laser-plasma experiments are formation and plasma ions acceleration when the second pulse wasbeing done at the University of Texas in Austin. The coils provide axial positive and also for increase in electron beam generation efficiencymagnetic fields to slow radial loss of electrons from the plasma. Peak when both the first and second pulses are negative. Bipolar pulses werefield with the 400 kA system is about 40 T, and will be up to 200 T for used for feeding accelerating gaps of induction accelerators with seriesthe 2 MA upgrade, with a current rise time of 1.7 µs for both systems. electron energy multiplication. Generators of two sequential pulses andFor these experiments the driver output must pass through a vacuum generators of negative pulse train of more than two pulses were used tofeed-through to couple to a short transmission line in vacuum that is study the electron beam transportation. In recent years, the generatorsterminated with a single-turn coil. This is done with a cylindrical of bipolar pulses of about 1 ns pulsewidth and subnanosecond pulseinsulator and conical transmission lines. A description of the device, width range are widely used in ultraband microwave pulse generators.and results of initial tests of the high-current version of the driver will The report presents the review of designs of the bipolar pulsebe given. generators and generators of two sequential pulses separated in time for* This work was supported by the Laboratory Directed Research and purposes of ion and electron beams generation. Different designs of theDevelopment program at Sandia National Laboratories. Sandia is a HV generators of nanosecond pulses are considered.multiprogram laboratory operated by Sandia Corporation, a LockheedMartin Company, for the United States Department of Energy’s 2C-4: A High-Power, High-Energy Pulsed Power GeneratorNational Nuclear Security Administration under Contract DE-AC04- for High-Impedance Loads94AL85000. M. J. Parker1, B. M. Novac1, I. R. Smith1, P. Senior1, G. Louverdis2 1 Electronic and Electrical Engineering, Loughborough University,2C-2: A Linear-Transformer-Driver (ltd) with Multiple Self- Loughborough, United Kingdom 2Triggered Switches Security Sciences Department, Dstl, Sevenoaks, United KingdomA. J. M. Pemen1, Z. Liu2, E. J. M. van Heesch11 Electrical Engineering, Eindhoven University of Technology, The supply to a high-impedance load of a high current pulse having aEindhoven, Netherlands relatively short rise time presents a significant challenge to the pulsed2 Zhejiang University, Hangzhou, China power designer, and a generator for such an application has been developed recently at Loughborough University as part of a majorThe LTD (Linear-Transformer-Driver), in fact a kind of Inductive- research programme. As will be described in the paper, the generator isVoltage-Adder, is gaining importance with regard to transient plasma based on an inductive storage technique, with an exploding wire arraygeneration. A LTD is based on multiple discharge branches, where in the primary circuit of a high-voltage (0.5 MV) transformer poweredeach branch consists of an energy storage capacitor and a switch. All by a 400 kJ capacitor bank. At a time close to that at which thebranches are magnetically coupled, and discharged simultaneously for exploding wire array generates its maximum voltage, closure of ahigh power pulse generation. switch in the transformer secondary circuit results in a current pulse inA major problem related to the conventional LTD is the excess of 10 kA and with a rise time of about 1.5 µs being fed to a load,synchronization of the switches in all branches within a short time having a resistance between 4 ohms and 30 ohms and an inductanceinterval. In this paper, we propose a simple method to realize automatic between 15 µH and 30 µH. A novel feature of the system is that the usesynchronization of the switches in all branches. of an unusual ground connection to the high-voltage transformerWe will first introduce our various topologies for very high peak power enables a greatly increased energy transfer to the load at a GW level ofgeneration, such as a multiple switch transmission-line-transformer, power. During preliminary testing, a major design challenge arose fromand a multiple switch Blumlein configuration. Next we will describe the very large internal forces generated in the transformer, whichthe novel concept of a multiple switch LTD (or inductive voltage necessitated an innovative solution. A dedicated 2D model wasadder). This concept has been verified on an experimental setup with therefore developed, which enabled the forces, dynamic resistance, andten discharge branches. Each branch is composed of a 1.3 nF high- a detailed distribution of the currents, temperatures and magnetic andvoltage capacitor and a spark-gap switch. It is shown that all ten electric fields to be calculated. Experimental results will be presented tobranches can be synchronized automatically, without any external confirm the accuracy of theoretical predictions provided by the model.trigger circuits. When the setup was charged up to 19 kV, an outputcurrent of about 3 kA was obtained. The risetime (10-90%) and the 73
  • 73. 2C-5:Bipolar Pulse Generation Based on Unipolar Solid- this type of circuits. The presented circuit uses four on-off switches perState Marx Modulator with Blumlein Line Stack stage, two for charging the energy storing capacitors, one for theJ. P. Mendes1,2, H. Canacsinh1,2, L. M. Redondo1,2, J. O. Rossi3 positive pulse and another one for the negative pulse. This reduces the1 Nuclear Physics Center from University of Lisbon, Lisbon, Portugal switch count in relation to topologies that use two for each pulse mode;2 Lisbon Engineering Superior Institute, Lisbon, Portugal however the circuit performance presents some limitations when3 Instituto Nacional de Pesquisas Espaciais, São Paulo, Portugal driving capacitive type loads, which is not the case for with the existing higher number of semiconductors topologies. A comparative analysis regarding the number of switches per cell, semiconductor losses andIn this paper solid-state and transmission line technologies are merged hold off voltage of the semiconductor is made. A circuit topology isin order to achieve a simple, flexible and compact bipolar Hybrid obtained as a compromise in terms of operating performance,Modulator. In fact, the connection of a solid-state positive Marx complexity and switching losses. A five stage laboratory prototype ofgenerator with a Blumlein stack enables the generation of high-voltage this circuit is assembled using 1200 V IGBTs and diodes, operatingbipolar output sub-microsecond pulses into a load. Bipolar high-voltage with 1000 V dc input voltage and 1 kHz frequency, giving 5 kV bipolarpulses are a very important tool for food processing, medical treatment, pulses, with 2.5 microseconds pulse width and 5 microsecondswaste water processing. These applications require pulses to produce relaxation time into resistive, capacitive and inductive loads.high electric fields, dozens of kV/cm, to ease the electroporation ofbiological cells, in which pulses with over 1 µs width charge the outercell membrane and can lead to the opening of pores, temporary or 2C-7: Bipolar Solid State Arbitrary-Waveform Marxpermanent, the latter usually resulting in cell death. However, pulses Generator for Capacitive Loadsmuch shorter than 1 µs, affect intracellular structures without adversely L. M. S. Redondo1,2, H. Canacsinh1,2, M. R. Gomes1 1affecting the outer cell membrane, being an important probe for Nuclear Physics Center from Lisbon University, Lisbon, Portugal 2intracellular interface. In this new developed modulator, bipolar pulses Lisbon Engineering Superior Institute, Lisbon, Portugalare obtained by connecting a solid-state positive Marx modulator to aload through a Blumlein stack, using only one power supply. Sub- A bipolar arbitrary-waveform generator has been developed, using anmicrosecond bipolar pulse generation is accomplished from the innovative bipolar solid-state Marx type modulator, for driving resistiverepetitive charge follow by the subsequent discharge of the Blumlein and capacitive type loads. In a variety of applied and fundamentalstack by the solid-state Marx generator, where pulse widths are fields, there is an increased need for the possibility of applying pulsedchanged by varying the length of the transmission lines and the time power voltage pulses in arbitrary or multilevel mode. This is importantbetween the two pulses is defined by the width of the Marx unipolar whenever there are dv/dt constraints, or the applications yield betterpulse. As both the solid-state Marx generator and the cascaded results if the voltage waveform differs from the typical rectangularBlumlein line are voltage multipliers circuits with, respectively, n and k shapes. In view of the potential industrial implementation of thisfactor this system gives higher flexibility for obtaining a high voltage modulator, a number of key aspects have been taken into account: a)output, v0=nkVdc, from a lower power supply, Vdc. In order to obtain modular construction, in order to enable scaling to different power andhigh efficiency, the rise time of semiconductors used in the Marx voltage levels, using the same hardware; b) high availability, whichgenerator should be less than 100 nanoseconds, so that the rise and fall calls for redundant operation capability after component failures; c)times of the unipolar pulses obtained at the Marx generator output failure management that includes avoidance of mechanical destructionwould be of the same order. For this, low-voltage fast MOSFET (i.e. due to high surge currents or arcing. The bipolar Marx type solid-stateMetal Oxide Field Effect Transistor) and HV SiC JFET (i.e. Junction topology presented in this paper is capable of multilevel or arbitraryField Effect Transistor) are used as switches in the Cascode circuits. waveform generation, both for positive and negative pulses, since eachThe results obtained from a 10 kV laboratory prototype that produces cell can be set on or off independently. The modulator consists ofbipolar pulses of 100 ns width at 1 kHz repetition rate (from a 1 kV dc multiple modular Marx type stages, with five on-off IGBT switchespower supply with four Marx stages and two Blumlein stages) are each, allowing high operating flexibility, such as the discharge of thediscussed using several load conditions. load capacitances both after the negative and positive pulses. The crucial component of the modulator is the control unit that is capable of2C-6: Optimizing Repetitive Bipolar Solid-State Marx optically turn on an off, in an arbitrary mode, each of theGenerators semiconductor switches in the stages; implying a higher operatingH. Canacsinh1,2, L. M. Redondo1,2, J. F. Silva3,4 complexity and cost that is compensated by the large flexibility, which1 Instituto Superior de Engenharia de Lisboa, Lisbon, Portugal enables the generation of multiple different voltage levels and2 Nuclear Physics Center from Lisbon University, Lisbon, Portugal waveform types. In this paper, we will discuss the results obtained from3 Instituto Superior Técnico, TU Lisbon, Lisbon, Portugal a 5-stages modulator, each with1 kV, for 100 Hz to 1 kHz frequency,4 Center for Innovation in Electrical and Energy Engineering, Lisbon, for rectangular, triangular and other type of waveforms, achieving aPortugal maximum of 11 voltage levels (5 positive, 5 negative and zero volts), into resistive and capacitive type loads.A novel solid-state Marx type topology capable of generating bipolarhigh-voltage pulses is presented and discussed against two other similar 2C-8:A Pulsed Power Generator with 20 Synchronous High-circuits in terms of number of semiconductors used, losses and Voltage Output Pulsescapability to deal with various load conditions performance. In a G. Deng, Z. Liu, B. Wang, K. Yanvariety of modern pulsed power applications, such as liquid food Department of Chemical and Biological Engineering, Zhejiangsterilization, water decontamination and air pollution control, the best University, The Industrial Ecological and Environmental Researchresults are achieved with bipolar pulses (i.e. positive and negative). Institute, Hangzhou, ChinaThis imposes solid-state modulator systems with higher complexity,using additional switches in comparison to the needed in unipolar pulse In pulsed power technology, synchronous pulses are required in manygeneration. Nevertheless, topologies using additional semiconductors cases. In this paper, a generator with 20 synchronous output pulses willhave the capability for flexible operation with different frequency and be described. This setup was developed by using a TLT (Transmissionduty ratio conditions, as well as, load conditions, which can be Line Transformer) based multiple-switching circuit topology. Itindustrially attractive. For this reason, it is very important to optimize 74
  • 74. consists of 20 capacitor banks (5.6 nF), 20 spark-gap switches, and a through rapid cooling restructure or alloy the surface to improve20-stage TLT. In each stage of the TLT, five RG218 coaxial cables, corrosion or wear resistance properties. This technique was applied tocovered by magnetic cores, are used. And they are put in parallel at the solve the corrosion problems of future heavy liquid metal (Pb, PbBi)input side and in series at the output side for high-voltage generation. cooled nuclear power plants. It has been proven that alloying theAll 20 switches are made of three electrodes, and assembled in one surface with an appropriate concentration of Al solves the corrosioncompact chamber under a pressure of 2 bar. At the input side of the problem in liquid lead at temperatures above 700 °C. For treatment ofTLT, the capacitor banks, spark-gap switches and the TLT are fuel cladding elements a novel electron beam facility with radialconnected in a way that all the switches can be synchronized converging electron flow has been developed.automatically like in a Marx generator. Once the synchronization Substantial progress has also been made in the development andprocess is completed, the energy in the capacitors will be delivered into optimisation of electrodynamic fragmentation facilities. Using adequatethe 20 independent loads via the TLT, and all the output high-voltage high current switch electrode materials, low loss capacitors andpulses are simultaneous and identical. Within the present work, the optimised component arrangements for minimum energy loss, servicecapacitors are charged up to 50 kV by a resonant charging source. The intervals of the high repetition rate pulse generators used in suchtotal energy per shot is about 140 J and the theoretical peak power is 5 facilities could be extended to beyond one year. ElectrodynamicGW. For each output pulse, it has a peak voltage of 150-250 kV, a fragmentation facilities are now marketed by our Swiss industry partner,current of 0.5-1 kA, an energy per shot of ~7 J, and a rise time of about in particular in the fields of mineral exploration and processing. Pulsed10 ns. electric fields of microsecond duration with electric field strengths of few to tenth of kV/cm can induce pores in the membrane of biological cells. This effect is used to extract foodstuff from plant cells, to2C-9:Inductive Adder Based Method for Generating accelerate the drying of biomass for energetic use or to reduce bacterialElectromagnetic Pulse with Controllable Timing contamination in waste water. Nanosecond pulsed electric fields canX. P. Yan, Z. Wang, P. Li, X. H. Hao, C. D. Yu, J. T. Wang also effect the growth of plats and fungi. It has been found that forState Key Laboratory for Mechatronical Engineering and Control, small energies of 100 J/kg (5 kV/cm, 100 ns, 100 pulses) plantBeijing Institute of Technology, Beijing, China seedlings show a leaf area increase by a factor of 2, whereas the mycelium covered area is 3 times larger.In this paper, a novel method for generating the electromagnetic pulses,in which both of timing and peak value are controllable, is proposed. 3A: Accelerators and Beams I: LTDs and High CurrentFor the proposed method, the capacitor is utilized as the battery storage Acceleratorsand the hydrogen thyratron is considered as the switch for powersupply. By simultaneously triggering multiple hydrogen thyratroni, the Tuesday, June 21 09:30-12:00 CC 10A-Bachievable voltage of electromagnetic pulse can be increasedremarkably with a peak voltage of 50kv at least. By changing the Linear Transformer Driver (LTD) Research for 3A-1:voltage of the capacitor or the number of superimposed hydrogen Radiographic Applicationsthyratroni, the peak value of the generated electromagnetic pulses can J. J. Leckbee1, S. R. Cordova1, B. V. Oliver1, T. J. Webb1, M. Toury2,be adjusted to the desired value. In addition, owing to the rational M. Caron2, R. Rosol2, B. Bui3, T. Romero3, D. Ziska3overall arrangement in the proposed method, the pulse front edge is 1 Sandia National Laboratories, Albuquerque, NM, United Statescontrolled to be within 20ns. Meanwhile, with the aid of DSP based 2 CEA-DAM, Polygone dexpérimentation de Moronvilliers,auxiliary circuit, the interval of electromagnetic pulses can be Pontfaverger-Moronvilliers, Francepredetermined, which results into the electromagnetic pulses with 3 K-Tech Corporation, Albuquerque, NM, United Statescontrollable timing as well as peak value and a minimum pulse intervalof 10us. Simulation results show that our proposed method achieves allthe anticipative targets and is effective for generating the The linear transformer driver (LTD) is a promising pulsed powerelectromagnetic pulses. technology being developed at Sandia National Laboratories, Commissariat à l’énergie atomique (CEA), and other institutions with applications including high voltage flash x-ray radiography and highPL2: Plenary 2 current z-pinch drivers. A LTD cavity is similar to a traditionalTuesday, June 21 08:00-08:50 CC 12A-D inductive voltage adder (IVA) cavity, however the primary energy storage is packaged inside the cavity. The resulting architecture is more compact than a traditional Marx generator driven IVA.PL2-1: Status and Recent Progress in Pulsed Power A 1-MV LTD built by the High Current Electronics Institute in Tomsk,Applications at Karlsruhe Institute of Technology Russia was delivered to Sandia in 2004 for radiographic testing andG. Mueller1, W. An1, T. Berghofer1, M. DelGiacco1, C. Eing1, evaluation. The system has been tested with various large area electronR. Fetzer1, B. Flickinger1, W. Frey1, H. Giese1, M. Gottel1, C. Gusbeth1, beam diodes (LAD) and the self-magnetic pinch (SMP) diode. TheA. Heinzel1, P. Hoppe1, A. Jianu1, F. Lang1, K. Leber1, M. Sack1, system was used to evaluate the LTD technology when coupled to aG. Schumacher1, J. Singer1, R. Straessner1, L. Wegner1, vacuum transmission line and dynamic impedance diode loads. A 1-A. Weisenburger1, F. Zimmermann1, V. Engelko2 MV LTD built by International Technologies for High Pulsed Power1 Institute for Pulsed Power and Microwave Technology, Karlsruhe (ITHPP) in France was delivered to the CEA at the PolygoneInstitute of Technology, Eggenstein-Leopoldshafen, Germany d’expérimentation de Moronvilliers (PEM) during 2010 and represents2 Efremov Institute of Electrophysical Apparatus, St. Petersburg, Russia the first commercially developed LTD system. The CEA LTD is a prototype module of a 8-MV LTD system that is being considered forIn this paper we review the progress that has been achieved at radiographic applications. A review of these 1-MV prototype programsKarlsruhe Institute of Technology (KIT) for several industrial scale will be presented.projects and in basic investigations based on intense particle beams and In addition, at Sandia, the 1-MV LTD system has been upgraded frompulsed power technologies. seven series cavities to 21 series cavities with an output voltage of 2.5Using intense large area pulsed electron beams thin layers at the surface MV. The 2.5-MV accelerator has sufficient vacuum electron flowof materials can be heated adiabatically above the melting point and current to evaluate the effects of electron impact on the cavity vacuum 75
  • 75. insulators and general LTD operation. Data from power flow FWHM of less than 75ns). Over 600 shots have already been made andexperiments at 2.5 MV with a LAD load will also be presented. are being analysed to determine the reliability and repeatability of this system. In this report we provide a description of the generator and the measured performances.3A-2: Development of a 1 MV Ultra-Fast LTD GeneratorF. Bayol1, F. Cubaynes1, R. Delplanque1, P. Genez1, C. Legras2,M. Parzych2, M. Toury3, M. Caron3, M. Mouillet3, A. A. Kim4 3A-4: Experimental Validation of the First 1-MA Water1 ITHPP, Hameau de Drèle, 46500 Thégra, France Insulated Mykonos LTD Voltage Adder.2 SEIV, Av du Val dOr, 33703 Mérignac, France M. G. Mazarakis1, M. E. Savage1, W. E. Fowler1, R. G. McKee1,3 CEA, DAM, DIF, 91297, Arpajon, France D. H. McDaniel1, K. W. Struve1, W. A. Stygar1, A. A. Kim2,4 HCEI, Acad. Ave. 2/3, 634055 Tomsk, Russia V. A. Sinebryukhov2, P. Wakeland3 1 01671, Sandia National Laboratory, Albuquerque, United States 2A 1 MV linear transformer driver (LTD) generator has been Pulsed Power Department, High Current Electronic Institute, Tomsk,manufactured by ITHPP and SEIV and delivered to Russia 3CEA/DIF/Polygone d’Expérimentation de Moronvilliers, in order to Engineering Department, KTECH, Albuquerque, Unites Statesdemonstrate the capability of this technology for flash X-rayradiography applications. This generator is composed of 10 cavities; The LTD technological approach can result in very compact deviceseach cavity includes 16 bricks representing two capacitors (GA 35436, that can deliver very fast, high current and high voltage pulses straight100kV, 8nF) and one multi-gaps air pressurized switch. To reduce the out of the cavity without any complicated pulse forming and pulseresistive loss in the switches, the inter-electrode gaps have been compression network. Through multistage inductively insulated voltagereduced from 6mm in initial design to 4mm, and the dry air pressure adders, the output pulse, increased in voltage amplitude, can be appliedhas been slightly increased. Each cavity has been tested individually on directly to the load. Because the output pulse rise time and width can bean automated testbed using a low inductive resistive load (~1.2nH) and easily tailored (pulse shaped) to the specific application needs, the loadextensive diagnostics, including a set of sixteen capacitive probes in may be a vacuum electron diode, a z-pinch wire array, a gas puff, afront of each switch to control their timing sequence. The cavities liner, an isentropic compression load (ICE) to study material behaviordeliver more than 100kV at ±100kV charging voltage into a matched under very high magnetic fields, or a fusion energy (IFE) target. Ten 1-load of 0.55-0.6Ω in a ~70ns FWHM pulse. Around 300 shots have MA LTD cavities were originally designed and built to run in a vacuumbeen made on each cavity (to condition the switches) before to or Magnetic Insulated Transmission Line (MITL) voltage adderassembly the generator. The 10 cavities are stacked in series on a stiff configuration, and after successful operation in this mode, weremetal frame and can move on rails with precise and reproducible modified and made capable to operate assembled in a deionized watermechanical alignment. Compression of the cavities is provided by the insulated voltage adder. Special care has been taken to deaerate thedesign of this frame. The center cathode of the output MITL is a 3.5m water and eliminate air bubbles. Our motivation is to test thelong straight conductor that is cantilevered and hold on a support, advantages of water insulation compared to the MITL transmissionmoving on rails, to allow easy and fast insertion and removal. All the approach. The desired effect is that the vacuum sheath electron currentsubsidiaries are also integrated on a dedicated lateral support, so the losses and pulse front erosion would be avoided without any newwhole generator is movable on these main rails. A high voltage trigger difficulties caused by the de-ionized water insulator. Presently, we havegenerator with 20 outputs (2 trigger cables per cavity) has been used to assembled and are testing a two-cavity, water insulated voltage addertrigger the cavities synchronously with a very low jitter. The with a liquid resistor load. Experimental results of up to 100kVpremagnetizing pulse is applied globally to all cavity cores through the capacitor charging will be presented and compared with circuit codesMITL cathode by using a movable contact under vacuum that is simulations.pneumatically controlled. The whole system is computer controlled andallows easy and reliable remote operation. The generator has been 3A-5: Experiments with the 2-Meter-Diameter, 810-KA LTD-successfully tested with a large area ~6Ω e-beam diode, in more thanone thousand shots at full charge voltage. It proves to be very III Accelerator Cavity J. R. Woodworth1, B. S. Stoltzfus1, W. E. Fowler1, W. A. Stygar1,reproducible (a few percent) and has a very low jitter (<2ns, 1σ over 20 M. E. Sceiford1, M. G. Mazarakis1, D. Anderson2, M. J. Harden2,consecutive shots). R. White3, J. R. Blickem4, A. A. Kim5 1 1671, Sandia National Laboratories, Albuquerque, NM, United States3A-3: Tests of a 1 MV LTD generator at CEA 2 National Security Technologies, Albuquerque, NM, United StatesM. Toury1, M. Caron1, L. Magnin1, M. Mouillet1, F. Bayol2, 3 L3 Communications, Pulse Sciences Division, San Diego, CA, UnitedF. Cubaynes2, R. Delplanque2, J. Leckbee3, B. Oliver3 States1 4 CEA, DAM, DIF, Arpajon, France Ktech Corporation, Albuquerque, NM, United States2 5 ITHPP, Thegra, France High Current Electronics Institute, Tomsk, Russia3 Sandia National Laboratories, Albuquerque, New Mexico, USA We are performing experiments with a 92-kV, 810-kA, 74.6-GWSince the summer of 2010, a 1 MV LTD generator is being tested at Linear Transformer Driver cavity. This cavity generates a ~100 nsCEA/DIF/Polygone d’Experimentation de Moronvilliers to power pulse from DC-charged capacitors in a single step. Ourdemonstrate the capability of this technology for flash X-ray experiments start with the existing 100-kV, 490-kA LTD-II * cavityradiography applications. This generator is composed of 10 cavities. and are making a number of improvements to it that are aimed atThe output voltage is added in series along a MITL which transfers the increasing the cavity’s peak output power and better understanding itspower to a large area electron beam diode. Different AK gaps of the operation. We are making improvements to the gas switches, thediode have been tested to qualify the generator on different load capacitors, and the magnetic toroids as well as heavily instrumentingimpedances. In a matched mode (near 6 Ohms), the peak voltage is in the cavity. These experiments have increased the cavity’s outputthe order of 1 MV and the pulse width is less than 95 ns (FWHM). The current into a matched load by 65% without increasing its volume.output pulse width is increased by the single diameter cathode that do Sandia is a multiprogram laboratory, operated by Sandia Corporation, anot match the increasing generator impedance (single stage matched Lockheed-Martin company, for the United States Department of 76
  • 76. Energy’s National Nuclear Security Administration under Contract No. proposed LTD-driven accelerator would require 588,000 switches. SoDE-AC04-94-AL85000. many switches would make the conventional triggering method*M. G. Mazarakis et al. Phys Rev. STAB 12,050401 (2009) significantly more difficult. It is necessary to develop a new triggering technique to meet the requirement of LTD technology. This paper proposes a new triggering technology based on reversed-LTD principle.3A-6:Conceptual Designs for an Upgrade of the Sphinx Z- We name it LTD-trigger. It has a similar structure to LTD device, but itPinch Driver operates reversely in principle relative to LTD. LTD-trigger is anF. Lassalle1, A. Loyen1, A. Georges2 inductive voltage divider contrast to LTD – inductive voltage adder. Its1 CEA GRAMAT, 46500 Gramat, France high voltage input between central metal conductor and outer conductor2 CEA DIF, Moronvilliers, France corresponds to the output of high voltage pulse in LTD. On the other hand, the output of LTD-trigger corresponds to the input from capacitorThe SPHINX machine [1] is based on the microsecond LTD and switch in LTD. With one fast pulsed power supply of LTD-trigger,technology and is used as a radiation effects simulator. It operates in multichannel fast triggering pulsed voltage to trigger correspondingdirect drive mode with 16 branches of 10 LTD stages each and delivers LTD switch can be produced by the linear inductive transformera 5.5MA, 800ns current on a Z-pinch load. With 2.2 MJ stored energy (magnetic core). It eliminates the delay time and jitter fromand a load 7cm radius made of a single array of aluminum wires, total intermediate switches in normal triggering method. It can match theradiation is 6TW, 300kJ with 25kJ above 1keV. delay time of switches in different module of LTD automatically. TheWe investigate several conceptual driver designs to improve radiation paper analyzes the working principle of LTD-trigger by establishing aoutputs from aluminum and argon loads and also to be able to drive circuit model and simulation. Taking some influence factors intohigher Z loads like titanium, stainless steel or copper. The analysis consideration, the results shows that such LTD-trigger profiling isfocuses on two architectures based respectively on the 100ns LTD possible. At present a test stand and a proof-of-principle experiments istechnology [2],[3],[4] and the microsecond LTD technology [1]. Two underway.designs giving approximately same driver size (building availabilityspecification) and comparable driver cost (funding specification) are Foil-Flyer Electro-Magnetic Accelerator - 3A-8:compared. Details on LTD stages designs and drivers architectures are Experimental Campaigngiven. K. Omar1, N. Graneau1, M. Sinclair1, B. M. Novac2, I. R. Smith2These designs correspond respectively to a 0.8MJ, 6MA, 200ns driver 1and a 9MJ, 19MA, 1400ns driver. Yield analysis is based on Hydrodynamics Division, AWE, Reading, United Kingdom 2performances measured on fast Z-pinches (200ns) and performances Pulsed Power Group, Loughborough University, Loughborough,extrapolated at currents above 6MA for slow Z-pinches (800ns). United KingdomThis analysis shows that the microsecond driver, compared to the 200nsone is more efficient in terms of cost/Joule of radiated energies. An experimental programme, developing a foil-flyer electromagneticImprovements on long implosion time plasma radiation sources and/or accelerator (FFEMA) is being conducted at AWE. A high currentimprovements on compact amplification systems connected upstream pulsed power generator driving a mechanical foil-flyer accelerator isthe load [5] would further confirm in the future the interest of the being used to provide tailored impulse profiles into targets for testingcompact and low cost microsecond LTD technology. material properties at high strain rates. This work has required the[1]. F. Lassalle et al., “Status on the Sphinx Machine Based on the design and construction of both an experimental platform to conductMicrosecond LTD Technology”, IEEE Trans. On Plasma Sciences, the experiments, named AMPERE, as well as the development of aVolume 36, Issue 2, Part 1 pages 370-377, April 2008 number of computer models to predict the electrical and mechanical[2]. A.A.Kim et al.; “100 GW Fast LTD Stage”; Proceedings of 2004 performance of the flyer and its interaction with a target.SHCE Conference; pp. 141-144 (LTD with oil and pressurized air A joint effort has been initiated with Loughborough University toinsulation) model the system. The independent models developed at the two sites[3]. W. A. Stygar et al. ; « Architecture of petawatt-class z-pinch will be used to predict the total current and its distribution in the circuit,accelerators », Phys.Rev. 10, 030401 (2007) taking into account conductor geometry and the variation of the[4]. A.V. Kharlov et al.; “Investigation of a Linear Transformer of resistance of a series dynamic resistor with increased deposited energy.Megaampere Level at Operation on Resistive-Inductive Load”; The various codes will all be used to ultimately model the mechanicalProceedings of 2010 SHCE Conference; pp. 261-264 (LTD with performance of the foil-flyer.atmospheric air insulation) A number of experiments have been conducted and the predictions of[5] A. S. Chuvatin et al.; « Design Criteria and Validation of a Vacuum corresponding models have been compared to the measured data.Load Current Multiplier on a Mega-Ampere Microsecond Inductive Current, Voltage and foil-flyer velocity are the primary diagnosticStorage Generator » ; Laser and Particle Beams (2010), 28 , 361-369 measurements. Both the experimental and predicted results will be described in this paper.3A-7:A New Triggering Technology for LTD Switches Basedon Reversed-LTD Principle Technology for Large Scale Trigger System Based on 3A-9:K. Liu PCSS Triggered V/n SwitchDepartment of Information of Sceince and Technology, Fudan L. Zhou, J. Deng, L. Chen, M. WangUniversity, Shanghai, China Institue of Fluid Physics, CAEP, Mianyang, ChinaLTD technology has been paid attention recent years as an advanced Large current facilities have been used for a wide variety of HEDPpulsed power system own to its unique advantages such as compactness, experiments, such as radiation-physics, equation-of-state, plasma-versatility and modular configuration with low voltage unit over Marx physics, astrophysics etc. Linear transformer driver (LTD) is a newgenerator with capacitor stacks. But there exiting a primary technical approach that can deliver fast high current and high voltagedisadvantage of a LTD accelerator is that many more switches are pulses with very compact devices. One of the most difficult technicalrequired to be triggered synchronously with low jitter than a problems faced in building large LTD facilities is that there are toocomparable Marx-driven accelerator. According to reference [1], One many switches to trigger, so the technical scheme of trigger system is 77
  • 77. critical. This paper presents a new technical scheme for construction of United States 3large scale trigger system. The technical scheme is based on traditional Communications and Power Industries, Beverly, MA, United Statestechnology, namely Marx generator and PFL. The most significantfeature lies on the output switch—a V/n switch triggered by PCSSs. The results of our effort to virtually prototype a conventionalThis technical scheme combines the V/n switch and PCSS, using PCSS magnetron capable of an RF output power exceeding 1.0 MW areas the trigger switch of the main spark gap switch, then lowers the presented. Magnetron design evaluation is carried out via numericalcharge voltage of PCSSs to about 40 kV and provides pulse charging simulation using the 3D Improved Concurrent Electromagneticconditions about 500 ns, meanwhile, takes the advantage of low jitter, Particle-in-Cell code (ICEPIC). Three major magnetron designs werelow optical trigger energy requirements of PCSSs. A trigger generator examined. They differed primarily by the number of resonant cavitieswhich can deliver 10 channels high voltage voltage pulses (~100 kV, they possessed as well as cathode length, radii and RF extraction~100 ns FWHM, ~30 ns rising time ) was built, the design and initial method. Eighteen, sixteen, and fourteen vane designs were simulated.test results were given. Each magnetron is designed to oscillate in the π mode between 600-900 MHz. Additionally, each design was refined to achieve a minimum3A-10: High Pulse Currents mode separation of 200 MHz between the π mode and the nearestI. V. Lavrinovich, N. A. Ratakhin, V. F. Feduschak, A. A. Erfort competing mode, the π-1 mode. In all cases, RF power is extractedHDE, HCEI SB RAS, Tomsk, Russian Federation axially via coaxial lines. For voltages ranging from 40 to 55 kV, output powers typically range from 300 kW to 1.1 MW. Output power efficiencies typically rest at 75% or above. For higher diode voltages,The paper reports on the design, tests, and applications of current in the range of 61-64 kV, simulations yield 1.3 MW output power at agenerators up to 1 MA with a rise time of 50–250 ns. The generators minimum efficiency of 85%.use only low-inductance pulse capacitors developed at IHCE SB RAS.A capacitor and a switch form a single unit of which (or of a number ofwhich) the current generators are composed. The cylindrical capacitors Hysteresis Dependence of Mode Separation on Time- 3B-2:are wound round a dielectric core of diameter 80–90 mm inside which Varying Applied Voltage in a Magnetron with Diffractiona switch can be located. The HCEIcap 50-0.1 capacitor rated at 0.1 uF Outputand 50 kV with a switch built in its interior part is a unit of total M. I. Fuks, E. Schamilogluinductance 16 nH. Four these units are operated in parallel in a HCEI Department of Electrical and Computer Engineering, University offlash 800-10 X-ray station. The operating current of a HCEIcap 50-0.1 New Mexico, Albuquerque, NM, United Statescapacitor is no greater than 50 kA, and this corresponds to 500 kA/uF.The density of stored energy is 208 J/l with an active volume of 0.6 l.The lifetime of the capacitor operated as part of the HCEI flash 800-10 In magnetrons with diffraction output (MDO - a relativistic magnetronstation is 1000 pulses. The capacitor lifetime attained in preliminary with axial extraction of radiation through a horn antenna in which thetests is 5 000 pulses. The HCEIcap 50-0.25 capacitor rated at 0.25 uF cavities of the anode block are continued up to a cross sectionand 50 kV with a switch built in its interior part is a unit of total exceeding the cut-off one for the radiated mode) any eigenmode can beinductance 16 nH. Four these units are operated in parallel in an X- used as the operating one. This is unlike conventional magnetrons withpinch radiographic device. The operating current of a HCEIcap 50-0.25 asymmetric extraction of radiation where only non-degenerate modescapacitor is 110 kA, and this corresponds to 440 kA/uF. The density of (pi-mode or the 2pi-mode) can be used, a consequence of which couldstored energy is 414 J/l with an active volume of 0.8 l. The lifetime of be mode hopping, resulting in overheating/erosion of the magnetronthe capacitor approximates 1000 pulses, and this value is high because electrodes. For the highly efficient gigawatt MDO [1] with fixedthe X-pinch is not a dummy load and the capacitor operates essentially applied voltage V=400 kV, the critical magnetic field H=4.527 kOein the short-circuit mode. The HCEIcap 100-0.2 capacitor rated at 0.2 was found to separate regions of synchronous interaction of electronsuF and 100 kV was tested in an X-pinch generator. The capacitor with with neighboring eigenmodes, the pi-mode with frequency f=2.34 GHza multigap switch built in its exterior part and a load is a unit of and the 2pi/3-mode with f=2.63 GHz [2].inductance 18 nH. The density of stored energy is 416 J/l with an activevolume of 2.4 l. The operating current is 200 kA at a voltage of 90 kV, In this present work using the MAGIC [3] particle-in-cell code weand this corresponds to 1000 kA/uF. By now, the lifetime of the show the hysteresis behavior of the boundary between these twocapacitor is several hundreds of pulses; however, it is clear that an operating regions as a function of the time-varying applied voltage foradditional dummy load is required. The HCEIcap 100-0.1 capacitor the magnetic field H. We demonstrate that the hysteresis loop israted at 0.1 uF and 100 kV has a switch built in its interior part. The successively wider for shorter voltage pulses, and for a short pulse ofinductance of the unit is 35 nH. The density of stored energy is 250 J/l order 100 ns the loop becomes so wide that it overlaps both voltagewith an active volume of 2 l. The operating current is 70 kA at a regions (2pi/3-mode: 0.31 – 0.40 MV; pi-mode: 0.40 – 0.495 MV). Tovoltage of 90 kV, and this corresponds to 700 kA/uF. Ten these initiate magnetron operation in a given mode, a short voltage rise timecapacitors as part of a current generator ensure a current of 700 kA in on the order of 1 ns, which is less than the characteristic time of70 ns. increasing oscillations (the quality factor Q that is close to the minimum diffraction Q [4] divided by the radian frequency) for the corresponding voltage region is required.3B: Microwaves III: High Power Microwave DevicesTuesday, June 21 09:30-12:00 CC 10C-D The extension of the region of single mode operation in the highly efficient, high power MDO can be used in applications where an optimal driving voltage pulse cannot be assured, as, for example, when Virtual Prototyping of a 1.0 MW Conventional3B-1: an explosive magnetic flux compression generator is used. At presentMagnetron Design low efficiency vircators are typically used in such cases [5].T. P. Fleming1, P. J. Mardahl1, M. Lambrecht1, J. Keisling2, M. Tracy31 Directed Energy Directorate, Air Force Research Lab, Kirtland Air [1] M. Fuks and E. Schamiloglu, “70% Efficient RelativisticForce Base, United States Magnetron with Axial Extraction of Radiation through a Horn Antenna,”2 Scientific Applications International Corporation, McLean, VA, IEEE Trans Plasma Sci., vol. 38, 1302-1312 (2010). [2] M. Liu, C. Michel, S. Prasad, M. Fuks, E. Schamiloglu and C.-L. 78
  • 78. Liu, “RF Mode Switching in a Relativistic Magnetron with Diffraction analysis is implemented from 0 to 2π to account for all the phase angles.Output,” Appl. Phys. Lett., vol. 97, 251501 (2010). By calculating the kinetic energy of the electron as it exits the cavity,[3] B. Goplen, L. Ludeking, D. Smith, G. Warren, “User-Configurable we show that the cavity can be self-excited. The particle-incell codeMAGIC for Electromagnetic PIC Calculations,” Comp. Phys. Comm., MAGIC [4] is used to simulate the dual cavity reltron with dualvol. 87, 54-86 (1995). extraction. From our simulations, we obtain the resonance frequency of[4] N.F. Kovalev, B.D. Kol`chugin, V.E. Nechaev, M.I. Petelin, M.M. the two hot cavities and the microwave frequency generated. WeOfitserov M.M., E.I. Soluyanov and M.I. Fuks, “Relativistic Magnetron compare our results with the single cavity reltron to demonstrate thewith Diffraction Output of Electromagnetic Radiation,” Sov. Tech. advantage of using a dual cavity.Phys. Lett., vol. 3, 430-431 (1977). REFERENCE[5] A. Neuber, A. Young, M. Elsayed, J. Dickens, M. Giesselmann, M. [1] R.B. Miller, W.F. McCullough, K.T. Lancaster, C.A. Muehlenweg.Kristiansen and L.L. Altgilbers, “Compact High Power Microwave Super-Reltron Theory and Experiments. IEEE Trans. on Plasma Sci.Generation,” Rev. Sci. Instrum., vol. 76, 104703 (2005). Vol. 20, pg. 332 – 343. June 1992. [2] S. Soh, E. Schamiloglu, R.B. Miller, Circuit Model of UNM Reltron Cavity, IEEE International Power Modulator and High Voltage3B-3:Amlitude and Phasing Control of Superradiative Conference, 2010.Pulses by the Magnetic Bias of Saturated Ferrite [3] R. B. Miller, Electronic Irradiation of Foods: An Introduction to theV. V. Rostov1, A. A. Elchaninov1, A. I. Klimov1, I. V. Romanchenko1, Technology (Food Engineering Series), Springer, 2005.G. A. Mesyats2, M. I. Yalandin3 [4] B. Goplen, L. Ludeking, D. Smith, G. Warren, “User-Configurable1 SB RAS, High Current Electronics Institute, Tomsk, Russia MAGIC for Electromagnetic PIC Calculations,” Comp. Phys. Comm.,2 RAS, Lebedev Physical Institute, Moscow, Russia vol. 87, 54-86 (1995).3 UB RAS, Institute of Electrophysics, Ekaterinburg, Russia 3B-5: An "Energy Efficient" Vircator-Based HPM SystemThe superradiation of high current electron beam in the relativistic J. Walter, J. Dickens, M. KristiansenBWO [1] allows one to achieve not only extremely high microwave Texas Tech University, Center for Pulsed Power and Powerpower but also, under some conditions, very stable parameters of HPM Electronics, Lubbock, TX, United Statesgeneration (peak power, central frequency and RF phase relativelyvoltage rise time) from one pulse to another. Because of the rise time ofthe cathode voltage should be optimal, a ferrite-loaded section of A common figure of merit utilized for HPM sources is thecoaxial transmission line with bias magnetic field was used. Saturation instantaneous power efficiency at the time of peak output power. Thisof ferrites in longitudinal direction was provided by DC magnetic field conversion efficiency from the electron beam power to radiatedattained 70 kA/m. The length of nonlinear line was about 20 cm and it microwave power disregards the overall energy efficiency of thecould not only reduce the rise time of the voltage front, but also change system, which is vitally important in systems intended for practicalthe time delay of the pulse delivery to the tubular explosively-emissive application. Optimizing the overall system energy efficiency allows acathode. Microwave pulses with sub-gigawatt peak power level and reduction in volume, weight, and prime power requirements that makesFWHM of 1 ns were produced with the use of low magnetic field for a much more practical and fieldable system. Texas Tech University(about 0.5 T) in the slow wave structure. The cathode voltage and e- has developed a vircator-based laboratory HPM system that has a highbeam current were 330 kV and 3 kA, respectively. Conditions of stable end-to-end energy efficiency relative to other HPM systems. Theoutput power value with controllable phase distribution from one pulse system utilizes a sealed-tube vircator source that requires no externalto another were found. Besides, such phasing control in the vacuum pumping subsystem. For rep-rate use, the tube contains anconfiguration of two-channel BWOs based HPM source [2] provided integrated low-power sputter ion pump. The lack of need for anpreassigned shift of resulting interference pattern of the radiation in externally applied magnetic field saves complexity, weight, andfree space. potentially power. The vircator tube is simple and robust. The source is[1]. A.A. Elchaninov, et al., IEEE Trans. Plasma Sci., 2004, vol. 32, No driven by a Marx generator utilizing pulse forming networks (PFNs)3, pp. 1093–1099. instead of lumped capacitors. Utilizing PFNs allows the driver to apply[2]. A.A. Elchaninov, et al., Technical Physics, 2011, vol. 56, No 1, pp. a more ideal pulse shape to the source than a traditional Marx,121–126. enhancing the performance of the source and reducing the amount of*This work is supported in part by RFBR, Grants: 10-08-00420, and energy wasted in the rise and fall of the pulse. The system has11-02-00097. 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 1003B-4: Design of a Dual Cavity Reltron MW. The system has been demonstrated at burst repetition rates asS. Soh1, E. Schamiloglu1, R. B. Miller2 high as 100 pps for short (10-pulse) bursts. Results will be presented1 Department of Electrical & Computer Engineering, University of New for the system operating from both battery and wall power.Mexico, Albuquerque,NM, United States2 Ktech Corporation, Albuquerque,NM, United States 3B-6:A High Voltage Pulsed Power System for Repetitive Vircator TestingThe reltron uses a modulating cavity to modulate the velocity of P. Norgard, K. R. Clements, R. D. Curryelectrons as they transit through the cavity [1]. The modulating cavityconsists of a main cavity and a side coupling cavity radially joined to Dept. of Electrical & Computer Engineering, University of Missouri,the main cavity [2]. Columbia, MO, United StatesIn the design of a dual cavity reltron, we add a second cavityimmediately after the first cavity to increase the velocity modulation of The University of Missouri – Columbia (UMC) is beginning researchthe electrons. The two cavities are isolated from each other by micron into high power microwaves and their effects upon various systems. Athin wire grids and are coupled only through the electron beam. high power microwave (HPM) laboratory is, therefore, beingUsing the relativistic equation of motion [3], we perform a 1D analysis established at UMC to provide the capacity to perform the tests. Theof the electron’s velocity and position as it traverses the cavity. The initial HPM source will be based on a virtual cathode oscillator, or 79
  • 79. vircator. The vircator requires a stable pulse-top voltage and will be Solid state amplifiers begin to replace traditional vacuum tuberequired to be repetitively driven. The UMC has modified an existing technology (e. g. Klystrons) in several applications. They offer therepetitive pulse generator to achieve the required pulse characteristics. perspective of lower cost, better reliability and reduced maintenance.The UMC vircators will be driven by a water pulse forming line output So far, predominately silicon based MOSFETs were used as activewhose output pulse is transformed to higher voltage by an induction device in the amplifier stage, because of the proven high powervoltage adder cell. The coaxial water pulse forming line (PFL) has an capability and adequate gain at the desired frequency band. However,impedance of 4.8 ohm and produces a 125 kV pulse that is 70 ns in Si devices suffer from a relatively low breakdown voltage which limitslength. A modulator has been developed to provide the requisite 250 J the maximum operating voltage and RF output power. Silicon carbideenergy pulse at repetition frequencies up to 100 pps in continuous mode (SiC) has a larger band gap of 3,26 eV which enables devices tooperation. The output voltage from the PFL is delivered to a matched- operate at a higher supply voltage and higher junction temperatureimpedance transmission line and transformed to 250 kV by an compared to typical silicon (eg. LDMOS) devices. We present ourinduction voltage adder to provide sufficient voltage for the vircator to second generation of high power semiconductor RF amplifier modulesoperate efficiently. Simulations of the driver connected to a vircator based on SiC normally-on vertical JFETs that can operate at VHFmodel were used to verify performance. Simulation results are frequencies. Due to the fast and robust body diode of the SiC-JFET, nodiscussed showing driver performance under rep-rate conditions. external freewheeling diode or circulator needs to be integrated which makes the RF-module very compact and reliable. The SiC transistors were optimized for high frequency operation. The3B-7: 3-D PIC Simulation of Virtual Cathode Oscillator devices have a pinch off voltage of -18V and deliver a maximumW. Jiang, K. Ito pulsed drain saturation current of 35 A. In a single ended amplifier testDepartment of Electrical Engineering, Nagaoka University of circuit, the transistor provides at 160 V supply voltage an output powerTechnology, Nagaoka, Japan of 1000 W at an input power of 110W. This correspond to a gain of 9,6 dB. The efficiency was 55%. The maximum supply voltage was limitedA new configuration of virtual cathode oscillator has been proposed by the test setup. However, much higher RF output powers can beand studied by using three-dimensional particle-in-cell simulations. achieved, as the maximum breakdown voltage of the SiC-vJFET isThis microwave oscillator is featured by wide range of operable power 3400V.level and high beam-to-microwave efficiency. It requires weak external We have built RF amplifier modules that have a PCB size of 100 x 160magnetic field to guide and confine the electron beam. The particle- mm and operate in a parallel push-pull topology with four transistors infield interaction occurs at the resonance between the virtual cathode parallel in each leg. We choose the circlotron configuration as it offersoscillation and the electromagnetic mode of the structure, which is fully symmetrical operation and requires only one grounded voltagesubstantially different from previous vircators. Intitial simulation source while the classical parallel push-pull circuit needs two. The RF-results have given microwave conversion efficiency over 20 %. module includes also a capacitor bank as a fast reservoir of energy for powering the 8 transistors.3B-8:Design and Implementation of Dual Independent The RF-modules were tested at an operation frequency 150 MHz with a pulse width of 100 µs at a duty cycle 1:10000. The test setup requires aVircators Driven by a Single Pulsed Power Source transformation network at the output of the amplifier. It provides anK. R. Clements, R. L. Druce, R. D. Curry, N. G. Kinsey, P. Norgard, impedance transformation between the optimum impedance where theS. R. Ashby transistors deliver the maximum output power and the 50 Ohm systemDept. of Electrical & Computer Engineering, University of Missouri, with the high power attenuator on the other side. The RF output powerColumbia, MO, United States was measured with a spectrum analyzer 10 µs after start of the RF output pulse. The RF-modules delivered 10,5 kW output power at 240The Center for Physical and Power Electronics at the University of V supply voltage and showed a gain of 11,2 dB. Stable operation atMissouri-Columbia has been investigating the interaction of solid-state various loads has been demonstrated.components with RF signals. To facilitate this research, dual,independent vircators are currently being designed and fabricated that 3B-10: Energy Efficiency of High Power Microwave Systemsare to be driven by a single pulsed power source. J. T. Krile, M. KristiansenEach vircator source is manually tunable over a range of 800 MHz to Center for Pulsed Power and Power Electronics, Texas Tech1.5 GHz by varying the back plane spacing in the anode-cathode University, Lubbock, TX, United Stateschamber. The vircators are driven by a single pulsed power source that,in conjunction with a linear transformer driver (LTD), will have anoutput of 250 kV driving a 16 ohm load impedance with a 70 ns pulse. High Power Microwave (HPM) system efficiency has typically beenThe designed output of each vircator is 10 to 20 MW. Each vircator’s calculated by dividing the peak output power by the peak input powerRF signal is transmitted to an anechoic chamber via waveguide. or the input power at the time of the peak output power. This method isWe will present details of the design, including FDTD-PIC simulations typically chosen over energy based calculations because of theusing MAGIC and electromagnetic simulations using CST Microwave difficulties in obtaining an accurate power envelope for certain HPMStudio, as well as construction details. The initial operational results sources. Power based efficiency calculations have also traditionallywill be presented. We will show results from investigations into current excluded any sub-systems which can be highly inefficient. However, assharing of the vircators and the effect of this dual arrangement on HPM systems begin moving from lab to field operation the poweroutput frequency. based efficiencies 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 major3B-9: High Power SiC Solid State RF-Modules limiting components of any mobile system. A prime example of anR. Irsigler1, R. Baumgartner2, M. Hergt1, T. Hughes1, O. Heid1 inefficient sub-system typically ignored in power based efficiency1 CT T P-HTC, Siemens AG, Erlangen, Germany calculations is the magnetic field generation (MFG) system of some2 CT T DE HW1, Siemens AG, Munich, Germany relativistic HPM sources. For a non-permanent, ambient temperature MFG system the minimum pulse width is typically limited by a capacitive discharge into a large inductive electromagnetic coil. Long 80
  • 80. pulses are needed, on the order of 100 µs or greater, to establish the capacitor through a resistive load. The railgap was triggered by a highrequired magnetic field. The situation is even worse for voltage pulse applied to a knife-edge trigger electrode in the switchsuperconducting magnet systems that, in addition to auxiliary cooling using a Maxwell 40168 trigger amplifier. A divider network ensuredsystems, must be ramped up slowly, on the order of seconds to minutes, the trigger did not swing relative to one of the switch sides, and ato avoid quenching. Given that most HPM pulses are on the order of 10 capacitor and resistor in series with the trigger supply acted as ato 100 ns, the energy efficiency of these MFG systems can be less that peaking circuit to increase the dV/dt of the trigger. The high voltage0.1%. Energy losses can also be found in the power conditioning discharge circuit achieved repetition rates up to 5 Hz. Capacitorsystems which include Marx generators, pulse forming lines/networks, lifetime was evaluated as a function of energy, voltage, discharge rate,capacitive stores, and impedance matching lines. These losses can be and repetition rate. Capacitor failure modes were examined usingmagnified by the impedance mismatches that occur as a result of the scanning electron microscopy. These results and an explanation of thetime varying impedance of most HPM sources. Additional energy observed differences are reported.losses can be incurred in the transmission of the HPM pulse through anantenna structure and in the conversion of the main fuel source, i.e. Test for End Connection Integrity of Metalized Film 3C-2:diesel, gasoline, grid power, or batteries, to a usable form. Theefficiency of HPM systems proposed for real world, mobile Capacitorsapplications should be calculated end-to-end and include any necessary S. Qin1, X. Qi2, T. R. Jow3, S. Boggs1 1sub-systems. The total energy efficiency is the product of the energy Institute of Materials Science, University of Connecticut, Storrs, CT,efficiencies of all the system components necessary to produce an HPM United States 2pulse and provides a more accurate figure of merit with which to judge GE Global Research, Niskayuna, NY, United States 3competing HPM systems. An overview of the proposed methodology Army Research Laboratory, Adelphi, MD, United Statesas well its application to some currently available systems will bepresented. The wire-arc-metal-sprayed end connections of metalized film capacitors limit their performance for high current discharge3C: Components II: High Energy Density Storage, applications. Because the spray does not connect to the filmTransmission Lines and Diagnostics metallization continuously along the film edge and because the spray penetrates to varying depths along the edge, locations of very highTuesday, June 21 09:30-12:00 CC 11A-B current density can occur along the edge of the film. As a result, large voltage drops can occur in the metallization or between the3C-1:High Voltage, Repetitive Pulsed Charge Discharge metallization and the end connection, which causes local discharge nearTesting of Prototype Capacitors peak current. Such discharges are caused by high current, not the highS. L. Heidger1, F. Dogan2, A. Devoe3, D. Brown4, M. Domonkos1 voltage applied to the capacitor. If a capacitor winding is placed in an1 AFRL/RDHP, Air Force Research Laboratory, Kirtland AFB, NM, under-damped LRC discharge circuit, the current through the capacitorUnited States and the voltage across the capacitor are nearly 90o out of phase, which2 Material Science & Engineering, Missouri Institute of Science & means that the peak current occurs when the voltage across theTechnology, Rolla, MO, United States capacitor is approximately zero. If partial discharge is observed near3 Presidio Components, Inc., San Diego, CA, United States the peak current under these conditions, it must be caused by the high4 SAIC, Albuquerque, NM, United States current and not the applied voltage and is therefore caused by end connection discharge, which indicates poor end connection quality. We have developed a solid state discharge circuit with integrated partialPrototype, multilayer ceramic capacitors fabricated using nano-grain discharge detector to evaluate the quality of end connections with asized, ultra-high purity, titanium dioxide dielectric were investigated single high current discharge, and we have demonstrated a strongfor high voltage, repetitive pulsed operation. Capacitor performance correlation between this test and winding discharge life. Such a test canwas highly sensitive to the processing conditions. Several iterations of be very useful to the industry. If the quality of individual windings cancapacitors are being fabricated to optimize the processing conditions. be assured, a large capacitor made from many such windings will haveProcessing parameters including powder processing, slurry constituents, greatly improved performance and be much more reliable.electrode configuration, binder, solvents, sintering temperature andtime, and post processing techniques were varied to improvemechanical and electrical properties. The most promising approaches 3C-3:Effects of Nd Doped BaTiO3 Nanoparticles on thewere used to fabricate multilayer capacitors and investigated for Dielectric Properties of Nd-BaTiO3/PVDF Compositessuitability for pulsed power applications. Analysis was based on M. F. Lin1, J. Z. Lim1, P. S. Lee1,2 1comparisons of the stored energy, loss, breakdown strength, lifetime Department of Material Science and Engineering, Nanyangand failure modes in pulsed and DC operation. Experimental grade Technological University, Singapore, Singapore 2multilayer capacitors with capacitance from 0.5 nanofarads to 2.75 T emasek Laboratories, Research Techno Plaza,, Nanyangmicrofarads were tested at voltages ranging from 1,000-15,000V. The Technological University, Singapore, Singaporecapacitors had very low loss tangents typically less than 0.001.Repetitive discharge tests from 1000-2000V were performed using an During the past decade, the advantages of reduced size and weight haveUltravolt 2C24-P-250 regulated DC/DC converter that was operated at not been fully exploited in high energy capacitors for portable power200 milliamp constant current, to charge the test capacitors to full devices. Metallized polymer film capacitor has many desirablevoltage, and discharged through a load resistance that was varied to properties such as light weight, low cost and lower thickness. Theachieve a discharge rate of less than 1 microsecond. Test capacitors fluoropolymers families are of special interest because of the highwere subjected to charge/discharge cycles by alternately triggering breakdown electric field, high releasing energy density, low energy loss,IGBTs at repetition rates that were varied from 1 to 200 Hertz limited and absence of remnant polarization after the discharge cycle. Polymersby the 250W power supply. Repetitive discharge tests from 5000V to in general possess relatively low dielectric constant values; therefore,15,000V were performed using a discharge circuit designed to yield a the improvement on high capacitance density is of intense interest. On200 nanosecond pulse with a near-critically-damped condition. The the other hand, BaTiO3 is one of the electronic materials with a highcapacitors were charged using a DC high voltage supply with a resistor dielectric constant and excellent ferroelectric properties. It can be usedin series. A miniature railgap switch was used to discharge the 81
  • 81. to produce high energy density capacitors in the middle voltage level absorption features were observed in the convolute. Lithium was usedfrom several hundred volts to 1 kV. At 1 kHz, the pure BaTiO3 powder as a localized dopant to establish the location of the plasma observedhas dielectric constant of 1013, with dielectric loss of 0.221 and within the convolute. Stark broadening of the H-alpha absorptionbreakdown strength of 1.17 MV/m at 2 mm film thickness. In this work, feature was used to infer the electron density within the convolute. Linewe report the formation of Nd-BaTiO3/PVDF composites for high shapes and intensities were analyzed to give the electron density andenergy capacitors applications. Synthesis of the BaTiO3 nano-particle plasma composition for various times. Peak electron densities rangedand doping of BaTiO3 with various at% Nd with particle size less than from 10^17 to 10^18 cm^-3 depending on load type. The measured100 nm of narrow size distribution has been carried out using sol-gel emission intensity and plasma density are strongly correlated with themethod. The composite was prepared by simple blending with PVDF convolute voltage. Comparisons between experimentally inferredfollowed by drop coating procedure. The dielectric constant and values and LSP simulations [1,2] will be shown.breakdown field strength were tested using metal-insulator-metal [1] D. V. Rose, D. R. Welch, T. P. Hughes, R. E. Clark, and W. A.capacitors. Experimental results show that the composite exhibits Stygar, “Plasma Evolution and Dynamics in High-Power Vacuum-dielectric constant of 115.9 at 100 Hz with improved energy density. Transmission-Line Post-Hole Convolutes,” Physical Review Special Topics - Accelerators and Beams, vol. 11, 2008, pp. 1-10. [2] D. V. Rose, D. R. Welch, R. E. Clark, E. A. Madrid, C. L. Miller, C.3C-4: Cygnus Dose Quality Mostrom, W. A. Stygar, M. E. Cuneo, C. A. Jennings, B. Jones, D. J.E. C. Ormond1, S. R. Cordova1, D. S. Nelson1, I. Molina2, J. R. Smith3, Ampleford, and K. W. Struve, “ZR-Convolute Analysis and Modeling:D. E. Good4, M. D. Hansen4, D. J. Henderson4, K. Hogge4, S. S. Lutz4, Plasma Evolution and Dynamics Leading to Current Losses,” PulsedC. V. Mitton41 Power Conference Proceedings, vol. 17th, 2009, pp. 1153-1158. Sandia National Laboratories, Albuquerque, NM, United States *MRG was supported by the Stockpile Stewardship Graduate2 Great Basin Technology Inc., Albuquerque, NM, United States Fellowship awarded by the KRELL institute in conjunction with the3 Los Alamos National Laboratory, Los Alamos, NM, United States DoE/NNSA. Sandia is a multi-program laboratory operated by Sandia4 National Security Technolgies, LLC, North Las Vegas, NV, United Corporation, a Lockheed Martin Company, for the United StatesStates Department of Energy’s National Nuclear Security Administration under Contract DEAC04-94AL85000.The Cygnus Dual Beam Radiographic Facility consists of two identicalradiographic sources each with a dose rating of 4-rad at 1 m, and a 1- Design of a Diagnostic System for Use in Optical and 3C-6:mm diameter spot size. The development of the rod pinch diode wasresponsible for the ability to meet these criteria. The rod pinch diode in Vuv Spectroscopy of Explosive Emissiona Cygnus machine uses a 0.75-mm tungsten diameter tapered anode rod J. M. Parsonextended through a 9-mm diameter cathode aperture. When properly Electrical Engineering, Texas Tech University, Lubbock, TX, Unitedconfigured, electrons born off the aperture edge can self-insulate and Statespinch onto the tip of the rod creating an intense, small x-ray source.The Cygnus sources are utilized as the primary diagnostic on single- This study focuses on the design of a diagnostic system for use inshot, high-value experiments. In such applications there is an emphasis optical and vacuum ultraviolet (VUV) spectroscopy on explosiveon reliability, shot to shot reproducibility and x-ray dose quality. In this emission processes at the cathode in a high power vacuum sealed tube.work reproducibility and quality will be diagnosed using lithium The system consists of a high power pulsed Nd:YAG laser with afluoride (LiF) thermoluminescent dosimeters (TLD’s). Both azimuthal highly tunable optical parametric oscillator (OPO, 200 to 2600 nm), aand axial measurements will be included. 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* Sandia National Laboratories is a multi-program laboratory managed explosive emission occurring at the cathode forms a plasma frontand operated by Sandia Corporation, a wholly owned subsidiary of propagating across the A-K gap negatively affecting the impedance ofLockheed Martin Corporation, for the U.S. Department of Energys the gap thus shortening pulse length and spoiling the desired consistentNational Nuclear Security Administration under contract DE-AC04- low vacuum within the sealed tube. The scope of this project is to probe94AL85000. the plasma formed between the A-K gap to determine the species emitted off the cathode. The timing integration of each sub-system is3C-5:(INVITED) Spectroscopic Measurements in the Post- very critical as the window for measurement is approximately 200 ns. A compact, fiber optically coupled, battery-operated, low jitter (~500Hole Convolute on Sandias Z-Machine ps), fast risetime (~20 ns) pulse trigger generator has been designed andM. R. Gomez1, R. M. Gilgenbach1, M. E. Cuneo2, R. D. McBride2, incorporated as the trigger source in a trigatron triggering scheme forG. A. Rochau2, B. Jones2, D. J. Ampleford2, D. B. Sinars2, J. E. Bailey2, the Marx Generator. Preliminary jitter measurements of 20-30 ns haveW. A. Stygar2, M. E. Savage2, M. Jones2, A. D. Edens2, M. R. Lopez2, been seen on the fully erected Marx Generator. The intent of this paperE. Stambulchik3, Y. Maron3, D. V. Rose4, D. R. Welch41 is to discuss the details of the various sub-systems and the timing Nuclear Engineering and Radiological Sciences Department, between them, along with species identification results gained fromUniversity of Michigan, Ann Arbor, MI, United States optical / VUV spectroscopy analysis.2 Sandia National Laboratories, Albuquerque, NM, United States3 Weizmann Institute, Rehovot, Israel4 Voss Scientific, LLC, Albuquerque, NM, United States 3C-7:Thermal Imaging Diagnostics of the High-Current Pulsed Electron BeamIn large pulsed power machines, the current from several magnetically G. Kholodnaya1, Y. Isakova1, V. Koghevnikov2 1insulated transmission lines is combined into one anode-cathode gap Tomsk Polytechnic University, Tomsk, Russian Federation 2using a post-hole convolute. Sandia’s Z-Machine utilizes a 12-fold- Institute of High-Current Electronics, Tomsk, Russian Federationsymmetry, double post-hole convolute. The peak current losses in thisdevice were measured to be as high as 20% for certain load types. A A method for the rapid control of the pulsed electron beams parametersstreaked visible spectroscopy diagnostic was developed to observe is presented here. This technique provides a registration of the totalplasma emission in the Z-Machine. Continuum emission and electron beam energy, as well as the spatial distribution of energy 82
  • 82. density over the cross section, and electron energy spectrum. 3C-9:Circuits for Digitally Synthesizing Very Long HPMDiagnostic measurements of the thermal impress of the electron beam Pulses in Compact Geometryon a material with a low bulk density and a low thermal conductivity O. S. Zuckerhave been performed. To register this thermal imprint an infrared Polarix Corporation, San Diego, CA, United Statessensor was used. The method presented here allows for thedetermination of the total energy and cross-section energy densitydistribution of the electron beam, provided that a beam is fully A unique subset of photoconductive switch based HPM uses thin filmsabsorbed into the target. In addition, it provides a record of the spatial as TL. Such films exhibit very high fields and very low impedances.distribution of electrons with energy in a selected range when Combining high-speed turn on with low impedance requires uniqueregistering a thermal print of the beam provided, which it is not fully integration techniques between switches and TLs. The successfulabsorbed by the target. Unlike techniques utilizing radiosensitive implementation of such integration in turn allows the construction of(dosimetric) materials, thermal imaging diagnostics does not require very high power density low voltage sources. Such disparate sourcescostly consumables or extensive amounts of time for processing. To are inherently coherent with one another and allow for spatial beamdetermine the electron energy spectrum according to the temperature combining to very high power.distribution over the depth of a foamed plastic, we calculated a specific There are a variety of circuits that can be utilized for HPM generationrelease of energy from an electron beam (distribution of an absorbed with switched TL sources. Traditionally switched TL based sourcesdose). According to the distribution of the absorbed dose over the depth could generate only a few cycles at a time such that in spite of the highof an absorber, and by solving an inverse problem, we restore the power, the resulting extremely short pulses had a correspondingly lowspectrum of the electron beam. The electron energy distributions were energy. A new class of circuits utilizing switch bypass technologycalculated on prior minimum assumptions by regularization of an ill- circumvents the limitation associated with cumulative switch losses ofposed problem - a Fredholm integral equation [1]. The developed the traditional frozen wave circuits. A key feature of these circuits ismethod allows for the definition of the energy spectrum of electrons the requirement to have the switches buried electrically betweengenerated from different points on the cathode, with a spatial resolution conducting plates while still accessible to external light A particularof 1-2 mm. Thermal imaging diagnostics were performed using the variant of this circuit combines both low switch loss with low TLFluke Ti10 thermal imager, and, as an absorber, a foamed plastic. The conduction loss with great compactness. The tradeoff of these circuitsexperiments were carried on a pulsed electron accelerator TEU-500 will be discussed.with the following parameters: accelerating voltage of 350-450 kV, The photoconductive switch, which is key to the realization of thehalf-high pulse width at maximum of 100 ns, the total kinetic energy above mentioned performance, includes the integration of theper pulse is up to 250 J [2]. Spatial resolution of thermal imaging semiconductor with the energy storing thin films at extremely lowdiagnostics is 0.9 mm, while the sensitivity of the thermal imager Fluke effective inductance.Ti10 provides registration of a thermal imprint for one pulse at a The impedance regime yielding this performance is in the 1/10 of ancurrent density of 0.1-0.4 kA/cm2. Studies on the parameters of a high- Ohm range resulting in very high current and poor match withcurrent electron beam formed by a diode with a flat cathode diameter of traditional HPM sources. Since abandoning the low impedance regime45 mm, made of graphite, copper, multipoint copper, and tungsten have would mean abandoning the above advantages, it behooves us to usebeen carried out. impedance matching techniques to couple to conventional loadReferences: impedances such as antennas.1. A.V. Kozyrev, E.H. Baksht, A.G. Burachenko, V.Yu. Kozhevnikov, Lastly, the laser system has unique requirement. In order to avoid theI.D. Kostyrya, and V.F. Tarasenko Proceedings 16th International more complex mode locked laser with its inherent pre-pulse problems,Symposium on High Current Electronics. Tomsk: 2010. pp. 43-46. which are the preferred source for picoseconds range pulses, other laser2. G. Remnev, E. Furman, A. Pushkarev, N. Kondratiev and D. technologies have to be used.Goncharov. IEEJ Trans. Fundam. and Mater. 124 (6), 491 (2004). These systems related issues will be presented here. Study and Diagnosis of the Power Transformer3C-8: 2P: Components posters II: High Energy Density Storage, Opening and Closing SwitchesBushing Insulation SystemA. K. Mehta Tuesday, June 21 13:30-15:30 Regency BallroomElectrical Engineering, National Institute of TechnologyHamirpur,Himachal Pradesh INDIA, Himachal Pradesh, India 2P-1: High Energy Density Film Capacitors S. Zhang, C. Zou, R. Orchard, D. Kushner, X. ZhouBushings are a critical component in electricity transportation. They are Strategic Polymer Sciences, Inc., State College, PA, United Statesused in substation buildings, transformers, locomotives, and switchgear.Bushings cause more than 15% of transformer failures. The main A series fluoropolymers have been developed in the past several years.purpose of a bushing is to transfer load currents in and out of metal The dielectric, thermal, and mechanical properties can be tailored by(grounded) enclosures at system voltages. The insulation system breaks changing the polymer composition and film production process. Thedown, causes bushing failure results in catastrophic event such as tank dielectric constant varies from 4 to over 50, and the meltingrupture, violent explosion of the bushing and fire. Clearly, the risk and temperature can be as high as 240 deg C. Furthermore, theselikelihood of collateral and personnel damage is a major concern in fluoropolymers are thermoplastic semicrystalline polymers and themsuch an eventuality. can be produced into thin capacitor film with thickness from 2 um toThis research is undertaken to study and diagnosis the power over 10 um by using melt extrusion and biaxial orientation process. Thetransformer bushing insulation system and in-field measurement of capacitor film has been metallized and prototype capacitors have beenpower-factor and capacitance using the Doble M4100 insulation produced. The high energy density capacitors can be used in pulsedanalyzer. The case studies on the different transformers showed how power and power conditioning systems such as medical defibrillators,the trend of moisture and dielectric properties changes with the military weapon systems, and power inverters in hybrid or plug-invariation of power factor and capacitance of insulation system. electric vehicles. 83
  • 83. 2P-2: Study on Self-Healing and Lifetime Characteristics of storage by lifting its operation voltage, a hybrid super-capacitor withMetallized Film Capacitor electrolytic cathode and electro-chemistry super-capacitor anode wasY. Chen, H. Li, F. Lin constructed. At the first, a physical module of hybrid super-capacitorHuazhong University of Science and Technology, College of Electrical element is setup, and its energy storage mechanism is interpretered.and Electronic Engineering, Wuhan, Hubei, China Then, the electric field of the hybrid super-capacitor and porous electrodes is analyzed. According to the relationship of operation voltage and capacitance by the results, some methods are put forwardHigh energy density capacitor is a key device in power supply source in to improve the distribution of electric field and working voltage. SomeElectromagnetic Gun (EMG) system. In order to increase the reliability experiments were made with sample elements of hybrid super-capacitor.of the power source equipment, the lifetime of capacitors must be The results show that the operation voltage of hybrid super-capacitorextended. The increasing of the capacitor’s lifetime is beneficial from element can be over 200V, and density of energy storage can be overits self healing capability. But also the self-healing is the main reason to 2.5kJ/L, it can give higher density of energy storage in application ofthe capacitance loss of metallized film capacitors, and it might finally pulsed power systems.lead to the failure of a capacitor. This paper studies the self-healingcharacteristics of metallized polypropylene (PP) film capacitor, andthen discusses the ways to reduce the self-healing energy and clearing 2P-4: Arc Suppression Snubbers on Energy Extractionarea. Firstly, the research results show that the self-healing energy is Switchgear in the LHC Superconducting Main Circuits ofthe crucial factor which influences the self-healing process. The self- the LHC Collider: Impact on the Vital Quench Protectionhealing energy can be altered or adjusted by several ways. And there is Systemsan optimized electrode configuration in which the capacitor gets a K. Dahlerup-Petersen, F. Formenti, B. I. Panevlongest lifetime under a certain electrical field stress. Then the TE Department, CERN, Geneva, GE, Switzerlandprobability of thermal breakdown of capacitors can be reduced throughchoosing a suitable electrode pattern. Secondly, the paper analyzed theeffect of pressure and interlayer air to capacitance loss of metallized A major challenge for operating a quench protection system for a chainfilm capacitors. For the outer layers, the tension of the film is not strong of superconducting magnets and interconnecting busbars is posed byenough, and there are a large number of air molecules, thus the the mandatory co-existence between the ultrasensitive and rapid quenchcapacitance loss of outer layers is serious. For the inner layers, as the detectors and the high power systems, i.e. the a.c. / d.c. converter andtension increases, a part of air is eliminated from the capacitor, so the the switchgear of the energy extraction (EE) facilities with theircapacitance loss of inner ones is relatively small. The thicker the air occasional, violent actions on the circuit elements. Each of the eight, 13layer is, the more significantly the self-healing area increases. The heat kA, main dipole powering circuits of the recently inaugurated LHCtreatment process is the use of film shrinking by heat to eliminate Hadron collider at CERN represents such a circuit, for which a numberinterlayer air, so it can increase the inner-layer pressure and decrease of mitigation features has been designed and implemented in order tothe self-healing energy. And it is important to consider the adhesion facilitate this coexistence. The introduction of snubber capacitor banksproperty of metal electrode and polypropylene film. For different sheet across the EE opening switches represent the most efficient method toresistance, the temperature of heat treatment has an optimal value, at enhance the commutation process by arc suppression and slow-downwhich the self-healing energy of the capacitor is minimum and the the fronts of the voltage waves penetrating the magnet chains, acting aslifetime of the capacitor is longest. And another way to increase the transmission lines. The paper presents the measured properties andpressure is wrap strengthening. To improve on the winding process, limitations of the combined switch / capacitor-bank assemblies andheat sealing is carried out on the joint when the polypropylene film is describes the design and operation of the associated, short-circuitinserted, thus the winding tension is greater and the interlayer air protection circuits. Furthermore, the article details the improvementsbetween the outer layers can be eliminated from the capacitor. Then, obtained to the original, analogue quench detection facilities and itthrough oil impregnation experimental study, it is revealed that full describes how the application of such a scheme has helped lifting aimpregnation can be achieved in vacuum or under alternating electric number of operational and performance limitations for tools related tofield in atmospheric pressure condition with low compression factor. At the enhanced QPS systems, with specific emphasis on the distributed,last, based on parameter optimization of electrode pattern construction superconducting busbar splice protection system and the so-calledand production technology, including the heat treatment process and oil ‘symmetric’ quench detection tool.impregnation technology, a compact capacitor with an energy densityof 2.0 MJ/m³ was designed. Life test was performed with the goal of 2P-5: Power Triggered Vacuum Switches with Triggeringachieving at least 1000 charge/discharge cycles. And the second of 1.5 Devices and Their Fields of ApplicationMJ/m³ design with discharge current magnitudes up to 100 kA have V. A. Sidorov, D. F. Alferov, G. D. Domashenko, V. P. Ivanovalready been testified in an Electromagnetic Gun (EMG) system All-Russian Electrotechnical Institute, Moscow, Russian Federationapplication. Triggered Vacuum Switches (TVS) owing to high switching capacities Investigations on Increasing the Operation Voltage of2P-3: are widely used in pulse power and pulse technologies. Series andHybrid Supercapacitors Used in Pulsed Power System parallel connection of TVS allows to expand their application area toJ. Song1, L. Zhang2, J. Zou3 voltage level 220 kV and higher, to create devices for switching1 Information Engir., Dalian Ocean University, Dalian, Liaoning, China currents up to 1000 kA by amplitude, to use such devices in industrial2 Electrical Engir., Dalian University of Technology, Dalian, Liaoning, power. Different application areas of TVS define specific demands toChina triggering devices for TVS. In this paper characteristics of recently3 Information Scie. & Engir., Dalian Politechnic University, Dalian, developed triggered vacuum switches RVU-57, RVU-59, modernizedLiaoning, China RVU-43 are resulted. Requirements to triggering devices are defined, various updatings triggering devices are resulted.Super-capacitor is one of the best sources in pulsed power applicationsfor their higher energy storage density. There are two ways to increasedensity of energy storage in capacitors, increasing ratio capacitance orelement operation voltage. In order to increase density of energy 84
  • 84. 2P-6: Protection Against Pulse Overvoltages Based on a University, Morioka, Japan 3Triggered Vacuum Switch Sun-A Corporation, Miyoshi, JapanD. Alferov1, D. Evsin1, V. Filippov2, V. Ivanov1, V. Miroshnichenko2,A. Perunov2, U. Priseko2, V. Sidorov1 A novel concept of synchrotron called induction synchrotron (IS) was1 All-Russian Electrotechnical Institute (VEI), Moscow, Russian developed at KEK in 2006 [1]. In the IS, charged particles areFederation accelerated by pulse voltages generated through 1:1 or 1:2 transformers2 VNITC VEI, Istra, Russian Federation driven by switching modulators employing high power high repetition rate semiconductor switches. The switches are turned on and off byThe scheme of an automatic protection device on the basis of a gate signals which is digitally manipulated from the circulating signalcontrolled vacuum spark gap with an operation time of <1 is proposed. of an ion beam. Consequently, the acceleration is always synchronizedThe results of studying the startup and switching characteristics of this with the revolution of ion beam irrespective of the mass and chargedevice are presented. The high switching capability of the device is state of ion. Thus, an IS is free from the limitation of bandwidth, unlikeensured by the ability of the spark gap to multiply pass pulse currents in RF cavities or amplifiers. Hence this feature allows us to acceleratewith amplitudes of up to hundreds of kiloamperes and durations of up ions in an extremely low velocity region. Acceleration of very slowto hundreds of microseconds and commercial frequency currents of up heavy ions, such as cluster ions is possible [2]. The IS is called theto several tens of kiloamperes for several periods. digital accelerator (DA) because it does not use sinusoidal waves but pulse voltages triggered by digitized beam bunch signals. A switching pulse modulator which generates bipolar pulses is one of the key2P-7: A Two-Stage Breaker with “turnover” of a Capacitor technologies for the DA. In the present DA, the single switching armBank Potential consists of 7 power MOSFETs connected in series, which enables high-O. G. Egorov frequency switching at 1 MHz with an output voltage of 2.5 kV andAtomic corporation, TRINITI, Moscow reg.,Troitsk, Russian peak current of 20 A. However, the ratings of the MOSFET such asFederation voltage and current may cause a restriction factor in the future. A SiC- JFET is one of the promising candidates replacing MOSFET, since inherently they have excellent properties in high electric fields[3].In the papers [1-3] a two-stage switch with a forced transfer of current Therefore we have developed a gate driver circuit optimized for highfrom the first to the second stage has been suggested. Such engineering repetition rate operation of a SiC-JFET and tested a sample device withsolution significantly reduces the switch cost and substantially it in burst mode. As a result, we had confirmed that the SiC-JFET isincreases its resource. The switch circuit in combination with an suited for the digital accelerator[4]. Moreover, we developed a newinductive storage with a high density of energy and low cost could be a device package to attain a continuous mode operation, and obtainedbasis for a compact industrial generator of pulses. This makes it successful results at the operation condition of 600V-7A-200kHs[5].possible to use this equipment with the aim of investigation and Then we have conducted an ultimate operation test of the developeddevelopment of production processes involving pulse powers of several device at 900 V-48A increasing a repetition rate, resulted in failure atGigawatts and duration of some milliseconds. However, the solution 800 kHz. Those test results and inspection results after the experimentssuggested has one drawback in that it shows an insufficient repetition will be described.of pulses. The problem is associated, first of all, with the preparation of [1] K. Takayama et al., “Experimental Demonstration of the Inductioncapacitor banks for the switching process, namely, with their charging Synchrotron”, Phys. Rev. Lett., Vol. 98, No. 5, pp. 054801(1) –to required polarity. 054801(4) (2007)The present paper considers an engineering solution allowing for the [2] T. Iwashita et al., “KEK Digital Accelerator”, Phys. Rev. ST Accel.use of a switch circuit in the mode of a train of gigawatt pulses. The Beams (to be published)mode is attained by self overcharge, i.e. there occurs a potential [3] K. Shenai et al., “Optimum semiconductors for high-powerreverses sing relative to the “earth”. This “turnaround” makes the electronics,” IEEE Trans. Electron Devices, vol. 36, no. 9, pp. 1811-switching processes ready to repeat. The paper estimates the limiting 1822, Sep. 1989interval between pulses in terms of the parameters of the switch [4] K. Ise et al., “Development of a Megahertz High-Voltage Switchingcomponents. Pulse Modulator Using a SiC-JFET for an Induction Synchrotron”,This engineering solution can be used to switch energy of an inductive IEEE Trans. Plasma Sci., VOL. 39, NO. 2, pp. 730-736,FEB. 2011storage of gigawatt powers to different loads such as a source of [5] K. Okamura, et al., “Novel Switching Power Supply for a Digitalmicrowave radiation and high intensity of pulse neutron source; this Accelerator”, in Proc. of The2010 IEEE IPMHVC, 1P11 (in press)finds its application in geophysical investigations, electromagneticaccelerators of armatures etc.Reference 2P-9: Research and Development of Drivers for Pseudospark1. Egorov O.G. Patent Russia, No.2299487, 22.09.2005. Switches2. Egorov O.G. Countercurrent breaking by vacuum interrupter V. D. Bochkov1, D. V. Bochkov1, Y. B. Makeev1, P. A. Bak2,combined in parallel with vacuum switch// Proc. XXIIth ISDEIV, A. N. Panov2, C. J. Pihl3, S. P. Andreason3Japan, 25-29.09.2006. 1 Pulsed Technologies Ltd., Ryazan, Russian Federation3. Egorov O.G. “Development of a combined opening switch for the 2 Budker Institute of Nuclear Physics, Novosibirsk, Russian Federationproject of generators on inductive storages”// Proc. XXIVth ISDEIV, 3 Pulse Power Solutions LLP, Mill Creek, WA, USAGermany, 30.08-03.09.2010, p.530. Data related to design and development of trigger and heater drivers for2P-8:Novel Switching Power Supply for the KEK Digital pseudospark switches (thyratrons TPI- and TDI-type) are presented.Accelerator The drivers (trigger and heater units) are made using up-to-date solid-K. Okamura1, K. Ise2, K. Takaki2, Y. Osawa3, M. Wake1, T. Iwashita1, state elements only and have both manual and remote control withK. Takayama1 CAN (RS-485 or Ethernet) interface. The drivers for TDI-thyratrons1 are made with a single or several outputs, so that the device is capable Accelerator Division 1, High Energy Accelerator ResearchOrganization, Tsukuba, Japan of triggering up to 4 thyratrons. Trigger/heater drivers with high-2 voltage insulation for applications with transient high-voltages Department of Electrical and Electronic EngineeringE, Iwate 85
  • 85. appearing on cathode and hydrogen reservoir of the switch are The structure and mechanism of array microhollow cathode areavailable. Test results of hollow anode TDI-thyratrons (operated introduced in this paper. The structure of microhollow cathode is twovoltage: 45, 75 and 150 kV, peak current up to 200 kA) with the electrodes divided by 200um thick dielectric and threaded by aboutvoltage insulated trigger/heater drivers in pulsed magnetic systems [1] 100um hollow. Array microhollow cathode can produce large-area,are described. While TPI-type thyratrons are a more recent uniform initiative high-energy electrons to ignite main discharge of adevelopment than the TDI series, they are already well established in a switch. With the cathode triggered by this structure, dielectric body-variety of pulsed power systems. Several triggering techniques, with discharge switches can realize uniform mode discharge in dielectricapplication to various switching requirements, have been developed barrier discharge (DBD) under normal air atmosphere. The inductanceand tested. These are the simplest circuit on non-linear elements, of switch can be reduced, so that the switch can operate with fast riseadvanced techniques with the primary discharge circuit triggered via time and nanosecond duration. We focused on the influence of triggersingle, double or triple pulse methods. For applications requiring low discharge current and the characteristics of dielectric on the rising timejitter (less than 0.5 ns), high peak currents, and long life (5-10 years), and pulse width. Through the plasma modeling of dielectric body-triple-pulse triggering has proven to be the preferred method. The discharge, we are supposed to understand the mechanism of switchcircuitry of the BZ-TP/10 trigger unit utilized on the LIU-2 accelerator breakdown in a micro level and a series of influential factors to the[2] is also described. characteristics of dielectric body-discharge switches.1. J. Slough, C. Pihl, V. D. Bochkov, et al, «Prospective Pulsed PowerApplications Of Pseudospark Switches», 17th IEEE International 2P-12: Experimental Studies of a Simultaneously OperatingPulsed Power Conference (2009), pp. 255-259.2. A.V.Akimov, P.V.Logachev, V.D. Bochkov, et al, «Application of Multi-Pin/Plane Corona Stabilised SwitchTPI-thyratrons in a Double-pulse Mode Power Modulator with B. M. Novac1, J. L. Walsh2, I. R. Smith1 1Inductive-Resistive Load», IEEE Trans. on Dielec. and Electr. Insul., Department of Electronic and Electrical Engineering, LoughboroughVol. 17, Issue 3, pp. 718-722, 2010. University, Loughborough, United Kingdom 2 Department of Electrical Engineering and Electronics, Liverpool University, Liverpool, United kingdom2P-10: Optically Triggered Pseudospark Switches withMagnesium PhotocathodesE. B. Sozer, C. Jiang, M. A. Gundersen A very high repetition rate closing switch is essential in all compact pulsed power sources for RF applications, and it is necessary for theElectrical Engineering/Electrophysics, University of Southern self inductance of the switch to be low for it to be properly integratedCalifornia, Los Angeles, CA, United States into the overall system. The paper will describe preliminary experimental work aimed at demonstrating the feasibility of a suitablePseudospark switches have been successfully employed in a variety of multi-pin/plane closing switch. Synchronised operation of the multi-pinpulsed power applications including high power electron beam sources, arrangement is expected to minimise the inductance of the switch tohigh power microwaves, and transient plasma ignition for pulsed well below that of an equivalent single-pin unit, as well as (even moredetonation engines [1-3]. Investigation of magnesium, copper and importantly), avoiding an impedance mismatch when mounted into amolybdenum photocathode candidates for optically triggered coaxial transmission line. The primary objective of the work to bepseudospark switches, also called back-lighted thyratrons (BLTs), has described was therefore to maximise the number of pin-plane gaps thatshown that magnesium has a higher quantum efficiency (up to 1.5×10- could be made to close simultaneously. a feature not possible with5) under BLT-relevant pressure conditions, compared to copper or conventional spark-gap switches. Preliminary considerations led to themolybdenum [4]. Mg-based BLT photocathodes were experimentally use of the corona stabilisation technique as the basis for thestudied by inserting a Mg foil at the cathode back surface. A compact development of the switch. Initially, the paper will present resultsUV laser with an intensity of 3 mJ per pulse was used as the optical obtained when determining the maximum number of pins that could betriggering source for these studies. It was demonstrated that both the used. This will be followed by studies of the simultaneity of closure ofdelay and the jitter of the switch were shortened by a factor of at least two multi-pin configurations, including a spectroscopic investigationthree when the Mg foil was present at the cathode. The electron into the plasma formed during operation of such switches. Finally, theemission mechanisms contributing to this improvement in delay and paper will present conclusions from the experimental programme andjitter for the BLT system are discussed. outline the way ahead for future developments.[1] K. Frank, and J. Christiansen, “The fundamentals of thepseudospark and its applications,” IEEE Trans. Plasma Sci., vol. 17, no. 2P-13: Numerical Simulation of Electric Field in5, pp. 748-753, 1989.[2] C. Jiang et al., “Toward ultracompact pseudospark switches,” Appl. Multichannel Multigap Gas Switches.Phys. Lett., vol. 86, pp. 024105-3 , 2005. A. V. Kharlov[3] F. Wang, et al., “Transient plasma ignition of quiescent and flowing Institute of High Current Electronics, Tomsk, Russian Federationair/fuel mixtures,” IEEE Trans. Plasma Sci., vol. 33, no. 2, pp. 844-849, 2005. Linear Transformer Driver (LTD) technology is actively developed in[4] E. B. Sozer et al., “Quantum efficiency measurements of the Institute of High Current Electronics (IHCE) in Tomsk, Russia. Airphotocathode candidates for back-lighted thyratrons,” IEEE Trans. insulated LTD stages have been developed recently, which are able toDielectr. Electr. Insul., vol. 16, no. 4, pp. 993-998, 2009. operate at 100 kV charging voltage. Capacitor block is the main structural element of the LTD stages. It incorporates two capacitors GA2P-11: A Dielectric Body-Discharge Nanosecond Switch 35426 (40 nF, 100 kV) and multichannel multigap gas switch. Body ofTriggered by Array Microhollow Cathode Discharge the block is founded from epoxy compound. Capacitors are sealedY. Teng, K. Liu, J. Qiu within epoxy in the block body permanently. Multichannel multigap gas switches were developed for use in the capacitor block NumericalDepartment of Information Science and Technology, Fudan University, simulation of the switch has been performed in this work. It has beenShanghai, China shown by calculations that electric field distribution is not equal on the switch gaps even at equal voltage distribution on the switch electrodes, provided by active voltage divider. Main reason for such distortion of 86
  • 86. the electric field is impact of a grounded electrode. Possible solutions consisting of the jitters of the triggered spark gap and the generatorsfor improvement of the switch performance are presented and discussed other spark gaps, and as well the trigger system is of importance. Ain this report. main source of temporal uncertainty is the generation of seed electrons for the breakdown of the over-voltage triggered spark gap. Hence, for only short delays the generation of seed electrons can be enhanced by2P-14: Experiment Study of a Low Inductance Three irradiation with uv light into the spark gap. The light can be emittedElectrode Field Distortion Gas Spark Switch for Linear from a glow discharge in the high-field region of a corona wire aroundTransformer Driver the spark gap electrodes. In order to improve the switching behaviourH. Wei1,2, P. Liu2, F. Sun1, X. Jiang1, J. Yin1, T. Liang1, Z. Liu1, of an over-voltage triggered spark gap, experiments with one spark gapZ. Wang1, A. Qiu1 surrounded by a corona wire arrangement under the variation of the1 Northwest Institute of Nuclear Technology, XI’AN,CHINA, Xian, geometry, oxygen-content of the nitrogen-oxygen atmosphere insideChina the spark gap, the gas pressure, and the applied over-voltage have been2 School of electrical engineering, Xian Jiaotong University, Xian, performed. In the experiments a jitter of the triggered spark gap of lessChina than 30 ns under certain conditions could be achieved.Linear transformer Driver is a novel pulse-power technology to directly 2P-17: Time Jitter Studies of a Corona-Stabilised Closingproduce high power pulses with risetime of 100ns or less. However, it Switchstill places special requirements on gas switches. Besides complex A. Larsson1, D. Yap2, Y. W. Lim2configuration to assemble, the multi-gap gas switch widely used in 1 Temasek Laboratories, National University of Singapore, Singapore,present LTD modules encounters the challenge to further reduce the Singaporerisetime due to its larger size and inductance. In this paper, a low 2 Applied Physics Laboratory, DSO National Laboratories, Singapore,inductance three electrode gas spark switch has been developed and Singaporetested. The switch has two each 7.5mm gaps, and is insulated with pureSF6. The trigger characteristics of the switch are detailedly studied andpresented in the paper. The total inductance of the switch is 65nH. The The switching time jitter is one of the important properties one has toexperiment result indicates that the switch delay is 40ns and the 1-σ consider when selecting the closing switch for a pulsed-power system.jitter is 2.8ns, when operating at 70% of its self-breakdown voltage. In an earlier review, four different closing switch types were identified as of interest for low-jitter systems (A Larsson, Time jitter of gas discharge switches – a review, EAPPC-BEAMS 2010, Jeju, Korea, 10- Characterization of Paschen Curve Anomalies at High2P-15: 14 October 2010), the cold-cathode thyratron, the laser-triggered gasPD Values discharge switch, the field-distortion threeelectrode switch and theW. J. Carey1, A. J. Wiebe1, R. D. Nord1, L. L. Altgilbers2 corona-stabilised switch. In this work, one of them is the object under1 ARC Technology, Whitewater, KS, United States investigation, namely the corona-stabilised switch (CSS). The corona-2 US Army Space and Missile Defense Command, Huntsville, AL, stabilised switch under study is a slightly modified commercial switchUnited States (Samtech CSS-01). The operating gas is SF6 at around atmospheric pressure. The CSS is housed in a metallic cylinder that acts as triggerPressurized gas switches are commonly used in a variety of high electrode at floating potential, which potential is governed by thevoltage pulsed power systems. According to Paschen’s Law, the self- corona discharges. Two thin disks were the main electrodes, amplybreakdown voltage of a gas switch varies with the product of the separated from each other but with short gap distances to the triggerabsolute pressure (p) and gap spacing (d). Plotting the measured electrode. During hold-off, there are corona discharges between bothbreakdown voltage for a given geometry as a function of p*d results in the disks and the trigger electrode, giving a corona current about 0.1 tothe Paschen curve for that geometry. The slope on the right-hand side 1 mA in the external circuit.of the Paschen curve can normally be approximated as linear for typical The main voltage was supplied by a DC power supply via anvalues of p*d. However, recent work in the high p*d region revealed intermediate storage capacitor, and the trigger voltage impulse had anboth a change in slope of the Paschen curve and a significant increase open-circuit rise time of 60 microseconds and was provided by a powerin the variance of the breakdown voltage when compared to lower p*d supply accompanying the CSS. An insulation spark gap was introducedoperation. The Paschen curve was characterized for high p*d values in between the trigger generator and the trigger electrode. Diagnosticsnitrogen. Tests were conducted on a 1 inch diameter spherical spark included traditional current and voltage measurements usinggap switch with various gap spacings and pressures in a high pressure commercial probes, and the main data used for the time jitter studieschamber. Voltage was applied at a fixed ramp rate with a high voltage was registered by an in-house made D-dot probe facing the main gap ofDC power supply. Results demonstrate an upper limit to linear the CSS. The signal from the D-dot probe was also used to interpret theapproximations of Paschen’s Law at high p*d values due to both discharge sequence-of-events in the CSS, including the closing of thenonlinear effects and large statistical variations in the breakdown insulation spark gap, the successive closing of the two gaps in the CSSvoltage for the same p*d value. and the final closure of the CSS switch.Distribution A: Approved for Public Release The time jitter of the CSS switch is found to be less than 5 ns for an operation voltage range of 15 – 50 kV. The pressure of the CSS switch has to be adjusted in the range 50 – 150 kPa for proper operation of the2P-16: Experiments for Reducing the Jitter of an Over- switch. To achieve this time jitter on a system level, the trigger pulseVoltage Triggered Spark Gap rise time and time jitter must be carefully considered.F. Attmann, M. Sack, G. MuellerInstitute for Pulsed-Power and Microwave Technology, Karlsruhe 2P-18: Time Jitter Studies of a Small V/n SwitchInstitute of Technology, Karlsruhe, Baden-Wuerttemberg, Germany A. Larsson1, D. Yap2, Y. W. Lim2 1 Temasek Laboratories, National University of Singapore, Singapore,For a long-term operation of a triggered Marx generator at higher Singaporerepetition rates over-voltage triggering is of advantage. If such a Marx 2 Applied Physics Laboratory, DSO National Laboratories, Singapore,generator has to be synchronized to an external event, a low total jitter Singapore 87
  • 87. The switching time jitter is one of the important properties one has to In a high-powered single pulse device, graphite electrode is better thanconsider when selecting the closing switch for a pulsed-power system. other ordinary metal electrodes for high energy transfer pulse discharge.In an earlier review, four different closing switch types were identified This paper mainly investigates the erosion mechanism of graphiteas of interest for low-jitter systems (A Larsson, Time jitter of gas electrode with the theory of thermodynamics and experiment results.discharge switches – a review, EAPPC-BEAMS 2010, Jeju, Korea, 10- The electrode erosion process of high-power spark gap switch is14 October 2010), the cold-cathode thyratron, the laser-triggered gas analysed in theory. The simplified mathematical model is built. It candischarge switch, the field-distortion threeelectrode switch and the be known the relationship of electrode erosion and transfer charge iscorona-stabilised switch. In this work, one of them is the object under linear. That can be proved by results of experiments.investigation, namely the field-distortion three-electrode switch. Theswitch is realised as a V/n switch, where the third, trigger, electrode 2P-21: The Research on the Trigger Characteristics of atakes a voltage equal to the main voltage divided by a number, n; in ourcase n=5. The test bed for the switch is a self-triggered system where Three-Electrode Spark Gapthe main voltage travels via transmission lines to the main V/n switch, L. Cai, F.-C. Lin, H. Zeng, G. Liu, L. Li, X. Qi, J. Nanwhich holds off the voltage and local and added capacitances ensures College of Electrical and Electronic Engineering, HuaZhong universitythe correct voltage division between the three electrodes, and then to a of Science and Technology, Wuhan, Hubei Province, Chinatrigger spark gap, which is connected between the earthed electrode andthe trigger electrode of the V/n switch. The trigger spark gap is self- Development of a three-electrode spark gap, intended for use intriggered and when it closes, it reverses the polarity of the voltage of medium power modulators, is discussed in this paper. Somethe trigger electrode, and thus the V/n gap is triggered and closed. Both characteristics have been determined experimentally with the pos-posthe V/n switch and the trigger spark gap were pressurised with nitrogen. configuration: the influence of the air pressure on spark gap operationThe V/n switch is designed and constructed, and tested for a hold-off and the influence of the working ratio on spark gap operation. Sparkvoltage of 40 kV. The main data used for the time jitter studies were gap delay time has been measured. And the statistical analysis of theregistered by three in-house made D-dot probes located close to the obtained results is presented in the paper. The study on the relationshiptrigger gap, facing the main gap of the V/n switch and slightly upstream between time delay and working ratio found that trigger mechanismthe high-voltage electrode. One of the D-dot probes was calibrated to would turned into direct breakdown between trigger pin and anode withgive the correct voltage magnitude. The signal from the D-dot probe working ratio higher than the value of 70%. Meanwhile, delay time as awas also used to interpret the discharge sequence-of-events in the function of working ratio is also evaluated, and it is found that workingsystem, ensuring that the trigger gap indeed triggered the V/n switch. ratio over 50% could effectively shorten the delay time. The result inSubnanosecond time jitter was achieved, but the system parameters this paper demonstrate that the operating voltage range between 50%must be carefully considered in order not to have miss-fires. and 70% might be considered for advancing trigger characteristic, prolonging the trigger electrode life and reducing the probability of2P-19: Test Bed for Time Jitter Studies of Laser-Triggered self-fire. In addition, through the simulation, a proper gap distance value could be calculated to make the switch work in good operatingGas Discharge Switches region with certain working voltage.A. Larsson1, D. Yap2, Y. W. Lim21 Temasek Laboratories, National University of Singapore, Singapore,Singapore 2P-22: Modular Trigger Generator for Over-Voltage2 Applied Physics Laboratory, DSO National Laboratories, Singapore, Triggering of Marx GeneratorsSingapore M. Sack, G. Mueller Institute for Pulsed-Power and Microwave Technology, KarlsruheThe switching time jitter is one of the important properties one has to Institute of Technology, Karlsruhe, Baden-Wuerttemberg, Germanyconsider when selecting the closing switch for a pulsed-power system.In an earlier review, four different closing switch types were identified Over-voltage triggering of a spark gap enables a synchronizedas of interest for low-jitter systems (A Larsson, Time jitter of gas operation of several Marx generators in repetitive operation without andischarge switches – a review, EAPPC-BEAMS 2010, Jeju, Korea, 10- increased wear of the electrodes. In order to generate the trigger pulses14 October 2010), the cold-cathode thyratron, the laser-triggered gas and apply them to one spark gap of a Marx generator a modular triggerdischarge switch, the field-distortion threeelectrode switch and the device consisting of a combination of a semiconductor-based pulsecorona-stabilised switch. In this work, one of them is the object under generator and a transformer has been developed. As the power andinvestigation, namely the laser-triggered gas discharge switch. voltage required for triggering depends on stage voltage and strayA test bed for laser-triggering of gas discharge switches has been capacitances of the Marx generator, the modular set-up enables an easydeveloped. The system consist of one Nd:YAG laser providing the adaptation of the trigger generator design to different designs of Marxfundamental, second and fourth harmonics (laser pulse wavelengths of generators. The paper describes the new design of the trigger pulse1064 nm, 532 nm and 266 nm, respectively). Beam splitters take generator.samples of the laser pulse for energy measurements and as a commontrigger signal. After the beam splitter, the laser pulse is coupled into an 2P-23: Evaluation of Experimental Silicon SGTO Devices foroptical fibre that transmits the laser pulse to a lens system that focusesthe pulse in the spark gap. Pulsed Power ApplicationsThe main data used for the time jitter studies is registered by an in- S. Lacouture1, K. J. Lawson1, S. B. Bayne1, M. Giesselmann1,house made D-dot probe located close to the spark gap. The test bed is H. OBrien2, C. J. Scozzie2 1described and initial time jitter studies are presented. Center for Pulsed Power and Power Electronics, Texas Tech University, Lubbock, TX, United States 2 U.S. Army Research Laboratory, Adelphi, MD, United States2P-20: Study on Erosion Mechanism of Graphite Electrode inTwo-Electrode Spark GapH. Zeng, F.-C. Lin, G. Liu, L. Cai, L. Li, F. Yu, G. Hu, N. Liu The development of new semiconductor designs requires that extensive testing be completed in order to fully understand the deviceCollege of Electric and Electronic Engineering, HuaZhong University characteristics and performance capabilities. This paper describes theof Science and Technology, Wuhan, Hubei Province, China 88
  • 88. evaluation of experimental Silicon high power Super Gate Turn Off output characteristics of the SiC module to a Si SGTO with similarThyristors (Si SGTOs) in a unique testing environment. The test bed mesa area will be presented.allows control over many parameters so that specific testing scenarioscan be created to test different parameters of the device. In particular 2P-25: Analysis of Silicon Carbide MOSFET Devices Duringthis test bed is designed to provide high peak current pulses with a longpulse width through the device. The SGTOs are capable of blocking in Pulsed Operationthe forward direction up to 5kV and are also capable of handling K. J. Lawson, S. B. Bayneseveral kilo-amperes when pulsed. The device structure is asymmetric Center for Pulsed Power and Power Electronics, Texas Techso the reverse blocking of these devices is only a couple hundred volts. University, Lubbock, TX, United StatesSince these devices are SGTOs special consideration had to be given tothe gate trigger circuit so that noise would be minimized on the gate Silicon Carbide offers many expanded capabilities for semiconductortherefore preventing false triggering of the devices. The devices were switches in the pulsed power arena. This material has the capability tofirst tested in a single shot configuration at currents up to 1.2kA with a achieve higher blocking voltages and lower on-state resistances than itspulse width of 1ms. After single shot testing was completed the devices silicon counterpart while operating efficiently at elevated temperatures.were then evaluated in repetitive operation with pulsed currents up to To better understand what impacts this can have on pulsed power1kA and a pulse width of 1ms at a rep-rate of 10 shots per minute. applications, switches made with Silicon Carbide are analyzed andDuring this test the voltage, current, and temperature of the devices tested in applications where high current pulses are necessary. Thewere monitored. The rep-rate test ran for 1000 cycles allowing the purpose of this paper is to investigate the performance of high powerdevices to be stressed significantly. After every test the devices were Silicon Carbide MOSFET switches in high current pulsed powerplaced on an Agilent B1505A High Power Device Analyzer so that the environments. The device under test for this study is a PowerExdevice characteristics could be captured and compared to previous MOSFET capable of (DC characteristics) blocking 1200V and forwardresults. This provides a way to determine if the device has degraded conduction of 100A. In order to test this device in a pulsed situation aover the course of the previous test. test bed was designed and built to operate the device with a current pulse of at least 1kA, 3µs duration, and at least a 2kA/µs rise time. To2P-24: Narrow and Wide Pulse Evaluation of Silicon Carbide accomplish these requirements, an RLC ring down test bed was designed and built with a focus on minimizing the parasitic inductanceSGTO Modules of the design. The components and design of this test bed were chosenA. A. Ogunniyi1, H. K. OBrien1, C. Scozzie1, W. Shaheen2, J. Zhang3, such that peak current can vary from 1kA to 5kA and the rise time canA. Agarwal3, V. Temple41 vary from 1kA/µs to 10kA/µs. This test bed will be used to investigate US Army Research Laboratory, Adelphi, MD, United States single shot capabilities of these devices, as well as quantifying their use2 Berkeley Research Associate, Beltsville, MD, United States in repetitive pulse applications. To determine the impact of these tests3 Cree Inc, Durham, NC, United States on the device under test, characteristic curves will be captured using an4 Silicon Power Corporation, Clifton Park, NY, United States Agilent B1505A Power Device Analyzer before and after each test to determine what characteristic changes are caused by subjecting theseSilicon carbide Super-gate turn-off thyristors (SGTOs) are being devices to such extreme current pulses.pursued by the Army as a replacement for the current silicon-based,high-power solid state switches due to silicon carbides attractive 2P-26: Fast, High-Voltage, High-Current SiC Thyristors formaterial properties. With the continuing improvement in materialdefect density, carrier lifetime, and fabrication process, SiC GTOs are Pulsed Poweron pace to becoming commercialized in the near future. H. D. Sanders1, S. C. Glidden2 1 Solid State Switch Division, Applied Pulsed Power, Inc., Batavia, IL, United StatesThe silicon carbide SGTOs used in this study were designed by Silicon 2 Applied Pulsed Power, Inc., Freeville, NY, United StatesPower, fabricated by Cree, and packaged into modules by ArkansasPower Electronics. The packaging design for the module is based onSilicon Power’s silicon SGTO modules. Each module contains four The full potential of SiC pulsed power devices is gradually being7.76 mm x 7.76 mm silicon carbide devices. ThinPak technology is achieved in terms of current, operating temperature, operatingused in place of wire bonds. External circuit boards and bussing unite frequency, and turn-on time. The thyristor device design is one ofthe four SGTOs. particular interest to the pulsed power community, achieving lower on- state resistance, higher voltages, and higher peak current ratings than other device designs such as JFET, MOSFT, or IGBT. Low current andSeveral modules were delivered to the Army Research Lab for pulse low voltage SiC thyristors are now commercially available. However,switching evaluations. The parameters assessed were peak current these devices do not posses the fast turn-on or high current capabilitiescapability, 1000-shot reliability, and current sharing between parallel desirable for pulsed power.switches. The test bed is an RLC circuit with up to 6.5 kJ of energy This paper will describe performance of SiC thyristors, using wire-storage and a large amount of inductance in order to create a 1-ms wide bondless packages developed by Applied Pulsed Power Inc., designedhalf-sine current pulse upon discharge. The repetitive wide-pulse specifically for MW level switching. These devices have been testedswitching of the SiC module resulted in a peak current pulse of 5 kA, using conventional electrical triggering as well as the laser pumpingwhich corresponds to an action rate (I2t) of 1.3 x 104 A2s. The narrow- method previously used with silicon thyristors. These results showpulse evaluation of the SGTO module utilized a pulse forming network significant improvements in turn-on time.with low stray inductance to obtain very high rate of current rise (dI/dt). This work is supported by the Air Force Research Laboratory, ContractThe narrow-pulse evaluation obtained for the SiC module was a peak Number FA9451-09-C-0010current pulse of 8 kA with a base pulse-width of 170 µs, whichcorresponds to an action rate of 5.0 x 103 A2s and a dI/dt of 720 A/µs.This paper will include a full report on the module’s narrow and wide-pulse capabilities evaluated since the pulse work presented at the 2010Power Modulator Conference. Furthermore, a comparison of the pulse 89
  • 89. 2P-27: Laser Pumping of 5kV Silicon Thyristors for Fast importance of the correct functioning of the LHC beam abort system,High Current Rise-Times the risk of SEB has to be evaluated and counter-measures be applied inH. D. Sanders1, S. C. Glidden2, D. M. Warnow2 required.1 Solid State Switch Division, Applied Pulsed Power, Inc., Batavia, IL, Manufacturers of the FHCTs provide only basic data on the expectedUnited States SEB rate (Udc = 2.8 kV for 100 FIT due to cosmic rays). Hence it is2 Applied Pulsed Power, Inc., Freeville, NY, United States necessary to evaluate the SEB for the present application parameters. In order to limit the cost the tests are performed in a less-destructive way possible by limiting the maximum fault current and the I2t integralLasers have been used to control semiconductor switching devices during an SEB. First tests, limiting the current by resistors and fasteither by laser gating or laser triggering. Laser gating requires the thermal fuses, revealed that the FHCT leakage current increased afteroptical source to generate all charge carriers, which would be an SEB (without fatal damage). The stress to the device under test isprohibitively expensive in terms of optical power to use with high being further reduced by the use of an active current limiting circuit.current devices. Laser triggering only generates charge carriers in the Two test stages are planned. In the first the SEB rate at very highgate region of the device, still resulting in a slow turn-on time, as those voltage (4 kV) and under ambient cosmic rays will be determined, tocharges create conducting channels between the anode and cathode. compare it with better documented devices at similar ratings. In theLaser pumping is an alternative that combines these two concepts, second stage the SEB rate will be measured at the maximum operatingseeding all of the initial charge carriers using optical power while using voltage (3 kV) with the device irradiated by a neutron flux with athyristor action to maintain conduction. By selecting the appropriate representative energy spectrum, albeit more intense than the onewavelength, the charge carries can be generated throughout the bulk of expected in the LHC gallery.the thyristor. This significantly reduces the costs of the optical powerwhile still providing a very fast turn-on, as the limit of the turn-on timeis not the rate at which the initial charge carriers can be generated but 2P: Microwaves posters II: High Power Microwaveshow quickly the device can be seeded with photo-generated charge Tuesday, June 21 13:30-15:30 Regency Ballroomcarriers. Previously reported results demonstrated turn-on times of lessthan 40ns to 2500A. These tests were done using commercial devices 2P-29: Pulse Width of a Reflex Triode Virtual Cathodewith windows etched into the anode metalization to permit laserpumping, describing the development of a compact laser diode source. Oscillator A. Roy, A. Sharma, R. K. Menon, S. Mitra, V. Sharma, K. V. Nagesh,Building on previous work, a new generation of silicon thyristorsspecifically designed for laser pumping, were manufactured. These D. P. Chakravorthydevices are created to couple more efficiently with the laser source. Accelerator and Pulse Power Division, Bhabha Atomic ResearchTwo types of devices were fabricated, some with optical windows on Centre, Mumbai, Indiathe anode side and electrical triggering available on the cathode side,and others with optical windows on the cathode side that cannot be Experiments were carried out to generate intense relativistic electronelectrically triggered. Also a higher power compact laser source was beams and High Power Microwaves (HPM) with reflex triode geometrydeveloped. This paper will describe their comparative performances. from 1 kJ pulse power system. The 1 kJ pulse power system is a MarxThis work is supported by DoE Grant DE-FG02-08ER85188 generator based repetitive system capable of operation at 10 Hz. The typical electron beam parameters were 200 kV, 4 kA, and 300 ns, with2P-28: Attempt to a Non-Destructive Single Event Burnout a current density of a few hundreds of amperes per square centimeter. The cathode used was planar graphite and the anode is a wovenTest of Fast High Current Thyristors stainless steel mesh of 70% transparency. Beam generation studiesV. Senaj, L. Ducimetiere were conducted for diode gaps from 8-20 mm. Cathode plasmaTE/ABT, CERN, Geneva, Switzerland expansion velocity has been calculated from the time dependent diode perveance and was found to be 1.7 cm/µs. It was also found that theSingle event burnout (SEB) can become a dominant failure mode of microwave (base not FWHM) pulse width increased from ~ 150 ns to ~high voltage semiconductors in the presence of either natural (cosmic 700 ns as the diode gap is increased from 8 mm to 20 mm. A largerays) or artificial (accelerator) radiation. Ionisation induced by a pulse width is obtained because of longer time of plasma motion acrosscharged particle or by a product high energy neutron nuclear reaction, the gap and reduced rate of the effective gap change. Measuredin a semiconductor region with high electric field, can lead to a local frequency was 2.2 GHz at 15 mm anode-cathode gap and decreasesavalanche ionisation with possible latching of the parasitic component. with the increased gap.The resulting high current density can lead to a local melting and henceirreversible component damage. The SEB failure rate depends strongly 2P-30: Investigations of a Double-Gap Vircator at Sub-on the applied voltage, with a component-specific threshold abovewhich the failure rate rises extremely rapidly. Microsecond Pulse Durations A. S. Shlapakovski, T. Queller, Y. E. KrasikThe beam abort system of the Large Hadron Collider (LHC) comprises15 extraction and 10 dilution kicker magnets per beam. Each magnet is Physics Department, Technion, Haifa, Israelpowered by a separate pulse generator containing two parallel HVswitch stacks. Each stack is composed of 10 fast high current thyristors Results of double-gap vircator operation studies using the 20-Ohm,(FHCT) in series, to empty a storage capacitor charged up to 29 kV. 500-ns high-current single-shot generator at an accelerating voltageDepending on the accuracy of the voltage sharing resistors and the within the range of 400-600 kV are presented. The double-gap vircatorindividual FHCT leakage current a few FHCT might have to sustain is based on a single-mode two-sectional rectangular cavity, which setsmore than 3 kV. Altogether the two beam abort systems comprise over the vircator frequency [1]. In the experiments, different cavity700 FHCTs. geometries were tested corresponding to different operating modes withThe pulse generators are installed in galleries about 100 m underground one to three variations of the RF field in the first, modulating cavityand hence well shielded against cosmic rays. Nevertheless stray section; the frequency changed from 2.0 to 2.3 GHz. An improvedradiation is expected to leak through cable ducts linking the gallery to design of the velvet cathode allowed for significant reduction of thethe adjacent LHC machine tunnel. Due to the high voltage applied on electron beam current losses and as a result, the vircator peak outputthe FHCT over extended periods (up to 20 hours per day), and the 90
  • 90. power increased up to ~200 MW. The studies were aimed at design. A conventional coaxial Vircator will generate the radiation ininvestigations of factors limiting the microwave output pulse duration. TM01 mode, due to its geometrical properties. For a compact HPM-Time-resolved light emission imaging and measurements of the current system, radiation in TE11 mode is preferred when the radiated energytransmitted through the vircator cavity were implemented as non- needs to be focused on a specific target. For operation in TE11 mode aintrusive diagnostic methods. This allowed us to identify such factors sectioned emitter can be used rather than a circumcircular.leading to microwave pulse shortening as the plasma formation that The efficiency of the Vircator can be greatly improved by optimisingdestroys the virtual cathode in the second cavity gap and radial the geometry of the Vircator housing. The impedance of the pulseddivergence of the electron beam that results in microwave generation power supply driving the Vircator and the impedance of the Vircator,delay. Depending on the accelerating voltage, anode-cathode distance depending on the A-K gap and amount of emitting material, is also(diode current varied from 8 to 15 kA) and cavity geometry, the important for maximizing the Vircator efficiency.obtained microwave pulse duration varied from 100 to 350 ns. For the experiments reported on, the Vircator was driven by a 400[1] S. A. Kitsanov, A. I. Klimov, S. D. Korovin, I. K. Kurkan, I. V. kV/320 J compact Marx generator that can be operated repetitively atPegel, and S. D. Polevin, "A vircator with electron beam premodulation 10 Hz. For these experiments, the Marx generator was operating inbased on high-current repetitively pulsed accelerator", IEEE Trans. single shot mode.Plasma Sci., vol. 30, no. 1, pp. 274-285, Feb. 2002. During the experiments presented here, a couple of geometrical features were varied as well as applied voltage. Their influence on the radiated field strength, dominating frequency and bandwidth is reported2P-31: Anode Optimization for a Compact Sealed Tube and discussed.VircatorJ. Walter, J. Vara, C. Lynn, J. Dickens, A. Neuber, M. KristiansenTexas Tech University, Center for Pulsed Power and Power 2P-33: Experimental Study of a Vircator with PremodulatedElectronics, Lubbock, TX, United States Electron Beam C. Möller1, F. Bieth2, P. Delmote2, M. Elfsberg1, T. Hurtig1, S. E. Nyholm1During the development and optimization of a compact sealed tube 1virtual cathode oscillator (vircator) at Texas Tech University, it has Defence and Security, Systems and Technology, Swedish Defencebecome apparent that processes at the anode have a significant impact Research Agency (FOI), Tumba, Sweden 2on tube performance. The impact of the high energy, high current High-Power Microwave Group, French-Greman Research Institute ofdensity (100-200 A/cm2 or higher) beam on the anode will either Saint-Louis (ISL), Saint-Louis, Francedirectly or through thermal processes cause outgassing of neutralmolecules. Neutral gas emission can also be caused by secondarily A vircator (virtual cathode oscillator) is a compact HPM source whereemitted electrons. Many of the neutrals are impact ionized by the no external magnetic field is needed. A vircator may sometimes showelectrons, forming a plasma. The plasma expands, eventually impacting unstable behaviour, such as frequency chirping or frequency andthe anode transparency and (combined with the plasma formed at the amplitude jumps during the pulse. If the frequency is stabilized bycathode) shorting out the anode-cathode gap. This expansion limits the means of a resonant cavity and/or a feedback mechanism the efficiencymaximum pulse width of the radiated microwave power. The residual of the vircator can be increased. One way to stabilize the frequency,evolved gas also negatively impacts the maximum repetition rate of the and increase the efficiency, is to use a feed back mechanism totube. An effort is underway to study the thermal behavior, gases premodulate the electron beam.evolved, and transparency versus time for different vircator anode An experimental study of a vircator where a feedback mechanism ismaterials and material treatments. Several different anode materials are used to premodulate the electron beam has been performed and theunder investigation, including stainless steel, copper tungsten, tantalum, results will be presented. The influence of several different parameters,nickel, and molybdenum. The effect of different treatments on the such as anode-cathode gap distance, emitter diameter and appliedanodes before tube assembly is also being studied. The thermal voltage, has been investigated with respect to frequency content andbehavior of the different materials has been compared. The surface field strength of the generated microwave pulse. The anode-cathodetemperature that is developed during operation is affected by the heat gap distance of the vircator can be changed without breaking thecapacity and conductivity of the material. Under typical conditions, vacuum.many of the materials can undergo surface melting during operation. For the experiments reported on, the vircator was driven by a 400The gases that are evolved during operation have been characterized kV/320 J compact Marx generator.utilizing pressure and residual gas analyzer measurements. The pre-shotbackground pressure in the tube is in the ultra-high vacuum range (~10- 2P-34: Suppression of Leakage Current in a Relativistic8 Torr), and the vircator is not pumped on during firing. After firing,the total evolved gas quantity and the types of gases present are Magnetron Using Various Cathode Endcap Designs C. J. Leach, S. D. Prasad, M. Fuks, E. Schamiloglumeasured, and then the tube is sputter-ion pumped back down to thebackground pressure before the next shot. Measurements have also Electrical and Computer Engineering Dept., University of New Mexico,been made of the anode transparency versus time, utilizing Faraday Albuquerque, NM, United Statescups located behind the anode. The data collected for the differentmaterials and the implications for device construction are presented. Upcoming experimental verification of MAGIC PIC code simulations of a high-power, 70%-efficient relativistic magnetron with diffraction2P-32: Experimental Studies on a Coaxial Vircator, Designed output (MDO) requires leakage electrons to be insulated from the dielectric window affixed to the end of the radiating horn antenna [1].for Operation in TE11 Mode Endcaps in relativistic magnetron systems have been explored beforeM. Elfsberg, T. Hurtig, C. Möller, S. E. Nyholm with great success in increasing output power and efficiency, andSwedish defence research agency, Stockholm, Sweden greatly reducing leakage current; however, none have yet to completely suppress leakage current [2, 3]. The radius of a disk endcap wasA coaxial Vircator is an advantageous design of a microwave radiation extended slightly beyond the inner radius of an A6 anode cavity suchsource for a compact HPM-system. With a limited size and outer that electrons from the interaction space drifting downstream along thediameter it is possible to use a larger emitting area compared to an axial axial B-field lines are stopped. A thin dielectric coating was added to 91
  • 91. lower the endcap surface electric fields on the endcap surface not being submitted to Appl. Phys. Lettersbombarded by these electrons, and to capture any emitted electrons. 4.J. Luginsland and R. Peterkin, “A virtual prototyping environment forPreliminary experiments on a radial-extraction relativistic A6 directed energy concepts,” Computing in Science & Engineering, vol. 4magnetron system (260 kV, 16 ns sinusoidal pulse) with a solid cathode pp. 42-49, March-April 2002.almost entirely suppressed the leakage current. Further endhatoptimization and testing will be done. Extension of these experimental 2P-36: RF Frequency Switching in a Relativistic Magnetronresults through MAGIC simulations will provide a predictive capabilityfor successful endcap design for the MDO system. with Diffraction Output (MDO) M. Liu1, M. I. Fuks1, E. Schamiloglu1, C.-L. Liu2 1[1] M. I. Fuks, and E. Schamiloglu, “70% Efficient Relativistic Department of Electrical and Computer Engineering, University ofMagnetron With Axial Extraction of Radiation Through a Horn New Mexico, Albuquerque, NM, United States 2Antenna,” IEEE Trans. Plasma Sci., vol. 38, pp. 1302-1312 (2010). Key Laboratory of Physical Electronics and Devices of the Ministry of[2] Y. M. Saveliev, S. N. Spark, B. A. Kerr et al., “Effect of cathode Education, Xi’an Jiaotong University, Xian, Chinaend caps and a cathode emissive surface on relativistic magnetronoperation,” IEEE Trans. Plasma Sci., vol. 28, pp. 478-484 (2000). Using MAGIC particle-in-cell computer simulations of a relativistic[3] M. R. Lopez, R. M. Gilgenbach, D. W. Jordan et al., “Cathode magnetron with diffraction output (MDO) we found a splitting of theeffects on a relativistic magnetron driven by a microsecond E-beam dispersion characteristics caused by electron interaction with differentaccelerator,” IEEE Trans. Plasma Sci., vol. 30, pp. 947-955 (2002). longitudinal modes that appear owing to non-perfect matching with a load. For each region of synchronous interaction of electrons with 3D ICEPIC Simulations of A6 Magnetron with2P-35: different transverse eigenmodes the dependence of radiation frequency on the axial magnetic field H compared to the critical value thatTransparent Cathode for Comparison of 3D MAGIC corresponds to the boundary between synchronous regions for differentSimulations axial modes, where bifurcation of the frequencies occurs. This situationC. L. Mendonca1, T. Fleming2, S. Prasad1, E. Schamiloglu1 is similar to the Van der Pol consideration of dynamical systems with1 Electrical and Computer Engineering Department, University of New two stable states separated by a saddle point [1]. We analyze a possibleMexico, Albuquerque, New Mexico, United States scenario for frequency switching for the same transverse field structure2 Directed Energy Directorate, Air Force Research Lab, Albuquerque, including the possibility to switch the frequency for a gigawatt MDONew Mexico, United States by using short, weak, single frequency RF signal when the magnetic field is near the critical value.Ongoing research at the University of New Mexico (UNM) shows [1] M. Liu, C. Michel, S. Prasad, M.I. Fuks, E. Schamiloglu, and C.-L.significant improvement in the start of and the rate of build-up of Liu, "RF mode switching in a relativistic magnetron with diffractionmicrowave oscillation in a relativistic magnetron with the transparent output,” Appl. Phys. Lett., vol. 97, 251501-3 (2010).cathode [1, 2]. In the recent studies conducted at UNM the results 2P-37: Metamaterial Cathodes in Multicavity Magnetronsobtained experimentally and the results obtained numerically, using the A. D. Andreev, K. J. Hendricks3-dimensional particle-in-cell (PIC) code MAGIC, have shown strong High-Power Microwave Division, Directed Energy Directorate, Aircorrelation [3]. Force Research Laboratory, Kirtland AFB, United StatesCurrently, we are conducting ICEPIC [4] simulations, also a 3-dimensional particle-in-cell code, at the Air Force Research Lab A metamaterial cathode made of a metal-thin-wire (MTW) medium(AFRL) of the A6 magnetron with transparent cathode for comparison enables a multicavity magnetron to operate over broader range ofwith UNM’s results. To facilitate this comparison, of the two codes, electric/magnetic fields and to radiate greater microwave power/energyoutput parameters such as microwave power, microwave frequency, compared with a magnetron having the standard thermionic cathode.anode current and leakage current with respect to the axial magnetic When individual elements (lattice of wires) of the MTW medium arefield will be presented. oriented parallel to a non-zero component (azimuthal or radial) of the rf electric field induced within the magnetron, the MTW medium mayPreliminary results indicate agreement between results from ICEPIC exhibit metamaterial properties by proper selection of the wireand MAGIC to within 5% for standard performance parameters. geometry and the lattice topology. When individual elements of theICEPIC simulations, run on parallel architecture with 64 CPUs at a MTW medium are oriented longitudinally, the MTW medium becomesresolution of 1 mm constituted the standard ICEPIC simulation of the a quasi-metamaterial (rodded) medium being transparent for both theA6 Magnetron. These simulations consisted of roughly 6 million active azimuthal and the radial components of the induced rf electric field.grid cells and 16 million particles. ICEPIC confirms oscillation, of the The negative (zero) permittivity ε and/or permeability μ of theA6 magnetron with transparent cathode, at 4 GHz in the 2π-mode. metamaterial (quasi-metamaterial) cathode significantly modify theAdditional simulation results have shown peak power are in good induced rf electric field distribution within the magnetron’s resonantagreement, total current is 16kA, and leakage current is also in good system with corresponding change of its dispersion diagram. The activeagreement. (emitting) MTW medium, placed in the center of the magnetron’s resonant system, facilitates faster start of microwave oscillations and1.M. Fuks and E. Schamiloglu, “Rapid start of oscillations in a either artificial selection or intentional dumping of either desired ormagnetron with transparent cathode,” Phys. Rev. Lett., vol. 95, pp. undesired magnetron operating modes.205101-1-4 (2005). * Andrey D Andreev is an NRC Associate at the AFRL’s Directed2.H.L. Bosman, M.I. Fuks, S. Prasad, E. Schamiloglu, “Improvement Energy Directorateof the output characteristics of the magnetrons using the transparentcathode,” IEEE Trans. Plasma Sci., vol. 34, pp. 606-619 (2006).3.S. Prasad, M. I. Fuks, K. Prestwich, C. J. Buchenauer, and E.Schamiloglu, “Experimental Observation of Fast Start of Oscillationsin a Short-Pulse Magnetron Driven by a Transparent Cathode,” 92
  • 92. 2P-38: Mechanism Analysis of a Kind of Diode over-Current programs. Finally, the experimental results for Slicer mode and SingleOscillation T-line and Blumein Based BP-former are presented.C. YuBeijing Institute of Special Electromechanical Technology, Beijing, 2P: Accelerators and Beams postersChina Tuesday, June 21 13:30-15:30 Regency BallroomWhen some kind of diode works in overflow condition, and voltage 2P-41: Primary Analysis of Switches Trigger Based onvalue given in both sides of the tube is appropriate, electric currentflowing the diode will show the phenomenon of the oscillation. The Secondary Induced Overvoltage of LTD P. Liu1, F. Sun2, H. Wei2, A. Qiu1, Q. Zhang1paper describes this experiment phenomenon, and explains it by the 1voltage-control switchs model.This kind of oscillation is proved Electrical Engineering, Xian Jiaotong University, Xian, China 2possible existence through the time behavior of thin beams drift in the Pulsed Power, Northwest Institute of Nuclear Technology, Xian,infinitely great space. Nonlinear equations mathematical model Chinasimulation on this kind of phenomenon is carried out, which is suitablefor chaos. The result shows that under some kind of starting value and Attention to the operating performance of LTDs (Linear Transformerregulative parameter, the simulation result tallies with the experiment drivers) has made it crucial to address the synchronization of their arrayphenomenon. of gas switches. In this paper, a trigger manner based on secondary induced overvoltage has been proposed for the switches in downstream2P-39: High Power Autonomous Pulse-Train Oscillator LTD cavities. After the closure of the switches in upstream cavities, aE. Nesterov, V. Fortov, Y. Isaenkov, V. Mikhailov, V. Ostashev, voltage pulse would be generated and propagate down the internalY. Semenov, V. Stroganov transmission line. Such a voltage pulse would raise the voltage applied on the open switches in each downstream cavity. As a result of this, theJoint Institute for High Temperatures of Russian Academy of Sciences, switches in downstream cavities are likely to close cavity by cavity. ToMoscow, Russian Federation demonstrate this assumption, the whole circuit model of a LTD module with sixty cavities in series was established in the PSPICE software. ByOutcome of design of powerful nanosecond UWB pulse oscillator simulations, the validity of the trigger manner based on secondaryintended for study of particular electromagnetic compatibility problems induced overvoltage has been demonstrated.is given. For excitation of oscillator we use high voltage drive-pulsegenerator based on resonance transformer. Voltage output of the 2P-42: Circuit Simulation of Saturn with a Reflex Triodegenerator equals approximately 300 kV. Power supply is a set of nickelmetal hydride rechargeable batteries. Dipole-type antenna with length Load*of dipole equal to 0.35 m is triggered by high-pressure spark-gap R. J. Allen, B. W. Weber, R. J. Commisso, S. B. Swanekamp,switch. Oscillator as a whole is combined into monoblock of 40 liters D. P. Murphyin volume and 30 kg in weight. Single pulse of radiation is Naval Research Laboratory, Washington, DC, United Statescharacterized by the following parameters: field range product (FRP)equals 175 kV (with reflector 330 kV), effective radiated power (ERP) A new circuit model for the Sandia Saturn generator [1] wasis approximately 1 GW (3.3 GW). Short-time amplitude instability of constructed to aid analysis of measurements from recent experimentsgenerated pulse is less than 10%. For average pulse-repetition with reflex triode loads[2]. In this configuration, Saturn applies voltagefrequency equal to 400Hz the power of primary source is 1 kW. Energy to only 4 of the 6 vacuum insulator stack levels. Two levels are used tospectrum of radiation is spread mainly over two ranges: the first range drive each side of the reflex triode and two levels are used as ballastis 0.1…0.4 GHz and the second range is 0.7…1.0 GHz. inductance[3]. This configuration is easier to model than others becauseTested oscillator demonstrated high performance reliability, stability of there are no convolutes in the vacuum region. Also, the very lowradiation parameters and good usability. impedance of the reflex triode makes the amount of vacuum flow current very low, which also simplifies analysis. Saturn has 36 identical2P-40: Comparison and Time Domain Analysis of Two pulse generator modules that each drive a water transmission lineCommon Bipolar Forming Methods in UWB Radiators output. These transmission line outputs are bussed together in waterK. Hojatzadeh before driving the vacuum insulator stack. Measured voltage waveforms from all 36 of the water transmission lines are used to driveElectronic and Communication Research Center, Tehran, Iran the simulation. The 94-ns two-way electrical length between the location of the TL voltage probe and the end of the TL makes the TLBipolar pulses are preferable for radiation purposes. For comparison voltages a high fidelity driver for the simulation. The load currentssake, it is assumed usually that the peak-to-peak voltage of the bipolar from the circuit model are compared to measured currents for shortpulse is equal to the amplitude of the unipolar pulse, and that the time circuit and real load shots. A circuit model for reflex triode was used toderivatives in both cases are the same. Bipolar formers are the essential aid in analysis of the reflex triode behavior.parts of most high power ultra wideband radiators. A precise time * Work supported by DTRAdomain analysis of the two common methods of bipolar forming in [1] D.D. Bloomquist, et al., "Saturn, A Large Area X-Ray Simulationultra wideband radiators is presented. Both methods utilize the fast high Accelerator," Proc. 6th IEEE Pulsed Power Conference, Arlington, VA,pressure sharpening and cutting spark gaps and delay lines. Also, the 1987analysis of high voltage insulation capabilities, effects of jitter and [2] D.P. Murphy, et al., This Conferencedelay between the operations of gaps on the output pulse is presented. [3] V.J., Harper-Slaboszewicz, et al., "Polarity inversion on Saturn,"We develop a new model accurately predicts the time dependent Proc. 17th IEEE Pulsed Power Conference, Washington, DC, 2009switching inductance and provides a reasonable representation of thetime dependent switch resistance based on Rompe- Weizel model andvoltage drop across the switch according to Martin formula. Thecapability and weakness of each method from the point of stableoperation are extracted and verified through common circuit analysis 93
  • 93. 2P-43: A Microsecond LTD Stage Designed as a Prototype 40nF capacitors per brick; the capacitors in each pair are charged infor an Upgrade of SPHINX Z-Pinch Driver opposite polarity to +-100kV and switched into the load by closing a 6F. Lassalle, B. Roques, A. Loyen, T. Chanconie gaps spark switch. Switches operate with pressurized dry air (pressureCEA Gramat, 46500 Gramat, France 5ata) and oil insulation is used inside the cavity. Size of this stage is 3m diameter, 0.23m thickness. Second prototype [4] was developed with the goal of reducingThe SPHINX machine [1] developed at CEA Gramat is the first Z- operation and maintenance costs for future high current sub-pinch driver based on LTD technology and is used since 2006 for microsecond LTD drivers. It stores 14.4 kJ and delivers a 660kA, 95kV,radiation effects experiments. 80ns (10%-90% rise time) pulse on a matched load (Rload = 0.14SPHINX uses 160 microsecond LTD stages (16 branches, 10stages in Ohms), using two LTD stages connected in series. Each stage consistsseries per branch). A single stage stores 14.4 kJ and can deliver a on 20 parallel bricks with for each brick two 40nF capacitors charged560kA, 30kV, 270ns (10%-90% rise time) pulse in a matched load to 95kV and a 7 gaps 6 channels spark switch. Like a SPHINX(Rload = 0.054 Ohms). This stage consists of 2 parallel “bricks” with microsecond LTD stage, this prototype can operate with dry air atfor each brick a 4µF storage capacitor charged to 60kV and a 7 gaps 18 atmospheric pressure and needs no oil and no pressurized switches.channels spark switch. Switches operates in dry air at atmospheric Size of this prototype is 2.9m*2.9m with 0.28m thickness.pressure and solid dielectrics are used for insulation inside the cavity Designs, electrical simulations and experimental results from these two(no oil or purified gases are required). Size of this stage is 2.3m*0.46m sub-microsecond LTD prototypes are presented.with 0.43m thickness. [1] F. Lassalle et al., IEEE Trans. On Plasma Sciences, Volume 36,Several driver designs, based on microsecond or sub-microsecond LTD Issue 2, Part 1, pages 370-377, April 2008technologies are investigated to improve the SPHINX machine [2] F. Lassalle et al.; “Conceptual Design for an Upgrade of the Sphinxperformances [2]. We present here a prototype designed to analyse the Z-pinch Driver”; this Conference.microsecond option. [3] A.A.Kim et al., Phys. Rev. Special Topics–Accelerators and BeamsThis prototype will store 200 kJ and simulations give a 5MA, 40kV, 12, 050402 (2009)360ns pulse on a matched load (Rload = 0.008 Ohms). Stage consists [4]. A.V. Kharlov et al.; “Investigation of a Linear Transformer ofon 16 parallel bricks with for each brick a 4µF capacitor charged to Megaampere Level at Operation on Resistive-Inductive Load”;80kV and a 7 gaps 6 channels spark switch. Like a SPHINX LTD stage, Proceedings of 2010 SHCE Conference; pp. 261-264this prototype can operate with dry air at atmospheric pressure andneeds no oil and no pressurized switches. Magnetic core is designed towithstand a 70 mV.s integral and the output vacuum line is 1.4m outer 2P-45: A Compact 2MA LTD for High Energy Densitydiameter. Size of this LTD stage is 3.9m*3.9m with 0.46m thickness. Physics ResearchDesign and electrical simulation from this microsecond LTD prototype S. N. Bland1, R. B. Spielman2, S. V. Lebedev1, J. Skidmore1,are presented. Details are given on design and experimental tests of a G. Burdiak1, J. P. Chittenden1, P. Cong3 1single switch. Imperial College London, London, United Kingdom 2[1] F. Lassalle et al., IEEE Trans. On Plasma Sciences, Volume 36, Ktech Corporation, Albuquerque, USA 3Issue 2, Part 1, pages 370-377, April 2008 Northwest Institute of Nuclear Technology in China, Xian, China[2] F. Lassalle et al.; “Conceptual Design for an Upgrade of the SphinxZ-pinch Driver”; this Conference. We report on a 2MA, 250ns single stage Linear Transformer Driver generator recently installed at Imperial College London. The generator,2P-44: Investigation of High Current Submicrosecond LTD designed and build by Ktech Corporation, requires no oil, water or SF6Stages at CEA Gramat for operation. The capacitors, charged up to 100kV in operation, areA. Loyen1, F. Lassalle1, B. Roques1, F. Bayol2, A. A. Kim3, potted and the low inductance ball gap switches are dry air insulated.B. M. Kovalchuk3 Conduction to the cavity is through a strip line arrangement with a1 CEA Gramat, 46500 Gramat, France nano-iron Magnetic providing ground isolation; allowing multiple2 ITHPP, 46500 Thegra, France generators to be stacked. Data on critically damped shorts at various3 Institute of High Current Electronics, 634055 Tomsk, Russia charge voltages is presented, and the dependence on pulse shape on both voltage and triggering explored. Initial results on using the LTD to drive isentropic compression experiments will also be shown.The SPHINX machine [1] developed at CEA Gramat is the first Z-pinch driver based on LTD technology and is used since 2006 forradiation effects experiments. 2P-46: Temporally Shaped Current Pulses on a Two-CavitySPHINX uses 160 microsecond LTD stages (16 branches, 10stages in LTD Systemseries per branch). A single stage stores 14.4 kJ and can deliver a M. E. Savage1, M. G. Mazarakis1, K. R. LeChien2, W. A. Stygar1,560kA, 30kV, 270ns (10%-90% rise time) pulse in a matched load D. V. Rose3, C. L. Miller3, E. A. Madrid3, W. E. Fowler1 1(Rload = 0.054 Ohms). This stage consists of 2 parallel “bricks” with Sandia National Laboratories, Albuquerque, NM, United States 2for each brick a 4µF storage capacitor charged to 60kV and a 7 gaps 18 NNSA, Washington, DC, United States 3channels spark switch. Switches operates in dry air at atmospheric Voss Scientific, Albuquerque, NM, United Statespressure and solid dielectrics are used for insulation inside the cavity(no oil or purified gases are required). Size of this stage is 2.3m*0.46m An important application for low impedance pulsed power drivers iswith 0.43m thickness. creating high pressures for shock compression of solids. TheseSeveral driver designs, based on microsecond or sub-microsecond LTD experiments are useful for studying material properties under kilobar totechnologies are investigated to improve the SPHINX machine megabar pressures. The Z driver at Sandia National Laboratories hasperformances [2]. Two prototypes were developed and tested at CEA been used for such studies on a variety of materials, including heavyGramat to analyse the sub-microsecond option. water, diamond, and tantalum, to name a few. In such experiments, it isFirst prototype stores 16 kJ and delivers a 1MA, 100kV, 80ns (10%-90% important to prevent shock formation in the material samples. Shocksrise time) pulse in a matched load (Rload = 0.1 Ohms). This prototype can form as the sound speed increases with loading; at some depth inis a single LTD stage based on previous developments made by HCEI, the sample a pressure significantly higher than the drive pressure canSNL and ITHPP [3]. LTD stage consists of 40 parallel bricks with two 94
  • 94. result. The optimum pressure pulse shape to prevent such shocks 2P-48: A Novel High Performance Thyratron Tube Driverdepends on the test material and the sample thickness, and is generally C.-Y. Liunot a simple sinusoidal-shaped current as a function of time. A system Power Supply Group, National Synchrotron Radiation Research Center,that can create a variety of pulse shapes would be desirable for testing Hsinchu, Taiwanvarious materials and sample thicknesses. A large number of relativelyfast pulses, combined, could create the widest variety of pulse shapes.Linear transformer driver systems, whose cavities consist of many The precise trigger timing and the output waveform of the thyratronparallel capacitor-switch circuits, could have considerable agility in tube are required to achieve optimal working condition of the kickerpulse shape. We will show results from initial experiments in pulse system in the booster ring. To meet this requirement, a reliable andshaping on a system with two inductively isolated cavities in series. precise Thyratron driver is needed to trigger the thyratron tube suchEach cavity contains forty pairs of high voltage capacitors and forty that the kicker’s output performance can meet required specification. Ingas-insulated spark gap switches. The capacitors are arranged in a this paper, design and performance of a new Thyratron tube driverbipolar configuration; the spark gap switches must withstand twice the deployed in SRRC is described. This driver is capable of deliveringcapacitor voltage. A pulse applied to the trigger mid-plane initiates Thyratron trigger-pulse with excellent jitter performance, which iseach switch. We have arranged triggering in groups of ten switches in smaller than +/-1nS, and the power transfer ratio is efficient. Theeach cavity, for a total of eight trigger points in the system. The voltages used to bias Thyratron tube are also included in the design offundamental rise time (10%-90%) of each capacitor circuit is roughly this driver. It has been tested and proven to be working well in60 nanoseconds; this defines the fastest possible output pulse rise time. delivering the kicker pulse with excellent stability and reliability.The pulse rise time can be made longer, and given features on the rise,by delaying the trigger to some of the switches. We will show initial 2P-49: On the Dynamics of the Flow along a Cylindrical Selfexperimental results from tests of the two-cavity system. Magnetically Insulated Transmission LineSandia National Laboratories is a multiprogram laboratory operated by J. G. Leopold, R. Gad, I. NavonSandia Corporation, a Lockheed Martin Company, for the United States Dept. of Applied Physics, RAFAEL Labs, Haifa, IsraelDepartment of Energy’s National Nuclear Security Administrationunder contract DE-AC04-94AL85000. We study by PIC simulation the steady state dynamics of the electron flow along self magnetically insulated cylindrical transmission lines of2P-47: Circuit Model Development to Improve the decreasing radii but fixed vacuum impedance. We find that the physicsPredictability of Shaped Current Pulses on Z follows the theoretical models of magnetic insulation but we alsoP. A. Corcoran1, J. P. Davis1, M. Savage2, B. Whitney1, I. Smith1, observe that the flow is governed by regular cycloidal electron motionD. Hinshelwood3, B. Stoltzfus2, H. Hanshaw2, R. Lemke2, K. Struve2, of fixed spatial period which decreases with decreasing MITL radius.V. Bailey1, E. Neau1, T. Wagoner2, C. Jennings2, W. Stygar2 While this is not surprising, such regularity has not been observed1 L-3 Communications/Pulse Sciences, San Leandro, CA, United States before. Moreover, most of the flow electron current is recaptured on the2 Sandia National Laboratory, Albuquerque, NM, United States emitting surface and only few flow electrons reach the edge. As the3 Naval Research Laboratory, Washington, DC, United States radius of the MITL decreases, the latter effect becomes even more pronounced. When retrapping due to load mismatch is present, the flowThis paper describes a new electrical circuit model of the refurbished Z shows additional interesting features.machine (ZR) at Sandia National Laboratories that accurately predictsshaped current pulses used for Dynamic Materials Program (DMP) 2P-50: The Flow Dynamics along Non-Uniform Selfexperiments. Particular pulse shapes are obtained by individually Magnetically Insulated Transmission Linesconfiguring gas-switch trigger time and water-switch gaps in each of J. G. Leopold, R. Gad, I. Navonthe 36 pulse lines. This mode of operation differs from standard Z- Dept. of Applied Physics, RAFAEL Labs, Haifa, Israelpinch operation where all 36 pulse lines are configured identically todeliver the same short pulse simultaneously to the load. Accurate modelpredictions are essential for determining how to configure the pulse We simulate the electron flow along non-uniform self magneticallylines prior to a shot to achieve the desired current-pulse shape. The new insulated transmission lines (MITL) where an upstream large radiuscircuit model includes both 1-D and 2-D networks of transmission line MITL is connected to a downstream small radius MITL of the sameelements and was based on prior models that had been benchmarked to vacuum impedance through a conical section. The onset of unstablemeasurements under Z-pinch mode operation. Recent model flow is characterized.developments are described which include improved switch models,improved coupling between adjacent pulse lines, and runtime 2P-51: PIC Simulations of Power Flow in a Linearoptimizations. These improvements have allowed better predictions in Transformer Driver for Radiographic Applications*less time to determine how to configure the machine. The most T. D. Pointon, D. B. Seidel, J. J. Leckbee, B. V. Oliverimportant of the improvements is a self-breaking water switch model Sandia National Laboratories, Albuquerque, NM, United Statesthat predicts water switch closure given only the switch gap, even whenthe field in the gap initially reverses due to coupling between early-triggered and late-triggered pulse lines. The runtime optimizations The linear transformer driver (LTD) is a promising technology forinclude an extension to parallel processing and a custom user interface. building a compact, high-voltage driver for radiographic applications.Also described are benchmarks to DMP shots, to well defined test-load Prototype 1 MV LTDs have been built at several sites for proof-of-shots, and between circuit codes. principle experiments, but radiographic applications require higher* Sandia National Laboratories is a multi-program laboratory managed voltage, V > 2 MV, and preferably at least 7 MV. At the higher voltage,and operated by Sandia Corporation, a wholly owned subsidiary of there will be substantially greater electron flow current in the centralLockheed Martin Corporation, for the U.S. Department of Energys magnetically insulated transmission line (MITL). The existing 1 MVNational Nuclear Security Administration under contract DE-AC04- LTD at Sandia National Laboratories has recently been upgraded to 2194AL85000. series cavities with an output voltage of 2.5 MV. This system provides the first opportunity to evaluate the effects of substantial electron flow 95
  • 95. in a multi-cavity driver at the low end of the voltage needed for electron accelerator using crossed cylindrical laser focusing. Appl.radiography. Phys. Lett. 69, 2175 (1996).We have been developing a 2-D, r-z PIC simulation model of the entire 3 Esarey, E., Sprangle, P., Krall, J., Ting, A. Overview of plasma basedsystem. Each of the 21 cavities is driven with its own external RLC accelerator Concepts. IEEE Transactions on Plasma Science, Vol. 24,circuit that can be independently triggered. Thus, the model can address No. 2, April 1996.various timing issues, such as optimizing time delays between cavities 4 Clayton, C.E. Plasma based acceleration concepts. AIP Conferenceand the effect of jitter. It also provides an opportunity to evaluate Proceedings, Volume 398, pp. 13-22 (1997).circuit models of the system. Electrons are emitted from the centralcathode with a conventional space-charge-limited emission model. The 2P-53: AMBICA-600: A Waterline Driven Gigawatt Pulsedcode computes the local energy deposition of electrons hitting theanode structures, including the dielectric insulator for each cavity. Electron Beam AcceleratorElectron impact on the insulators is of particular interest. Results from R. Verma1, A. Shyam2, T. Patel1, Y. C. Saxena1 1simulations will be presented. Institute for Plasma Research,, Bhat, Gandhinagar, Gujarat, India -________________________________ 382428 2* Sandia is a multiprogram laboratory operated by Sandia Corporation, Bhabha Atomic Research Center,, Autonagar, Vishakapatnam,,a Lockheed Martin company, for the United States Department of Andhra Pradesh, India - 530012Energy’s National Nuclear Security Administration, under contract DE-AC04-94AL85000. A dual resonant Tesla transformer based pulsed electron beam accelerator AMBICA-600 has recently been developed. It comprises of2P-52: Compact High Average Gradient Particle a coaxial waterline charged from single turn Tesla transformer. Water has been used as dielectric medium in the pulse forming line because ofAccelerators Utilizing Photoconductive Switches its high dielectric constant, high dielectric strength and efficient energyO. S. Zucker storing capability that enhances compactness of the system. WaterlinePolarix Corporation, San Diego, CA, United States has impedance of ~5 ohms and generates pulse width of ~60ns. The system is capable of producing intense electron beam of about 300keV,The drive for high average gradient particle accelerators is on two 60kA when connected to impedance matched electron beam diode.fronts: the first is in the high energy physics and the second in medical This paper describes the gigawatt accelerator design, construction andand industrial applications. characteristics of intense relativistic electron beam produced by a fieldThe conventional approach utilizes Klystron pumped cavities, which emission diode.are pumped slowly (microseconds) and dumped fast (nannoseconds).These accelerators operate at under ~ MV/cm gradient. Both 2P-54: Initial Tests of the AWE Hydrus IVA Marxbreakdown electric field and cavity losses due to field emission are T. Warren1, T. DaSilva1, J. Wilson1, K. Hanzel1, V. Carboni1,time dependent. This results in an upper limit on the working field in D. Spelts1, J. Pearce1, W. Saunders1, W. Glazebrook1, K. Thomas2,the cavity and thus on the average gradient. [1] P. Beech2, S. Clough2, I. Crotch2, S. Brown2, B. Stringer2, C. Goes2,There have been a number of concepts proposed which do not have the A. King2, I. Huckle2, J. Burscough2, S. Trenman2, J. Duffy2,above mentioned limitations; key among them are the direct R. Wheeldon2acceleration in a focused laser beam [2], laser plasma based devices 1[3,4] and the photo conductively switched arrays of radial transmission L-3 Communications/Pulse Sciences, San Leandro, CA, United States 2lines (RTL). The RTL approach depends on photoconductive switches, Atomic Weapons Establishment, Aldermaston, Berkshire, Unitedwhich makes this approach viable because they represent the highest Kingdompower per unit volume and the highest switching speed available. In allRTL approaches, the accelerator is composed of a series of centrally The components of the Hydrus Induction Voltage Adder (IVA) areapertured radial disks where adjacent pairs constitute radial approaching completion at L3 Pulse Sciences at San Leandro, CA. Alltransmission lines. A radially convergent TEM wave is launched at the parts of the IVA are being procured by L-3 PS and delivered to Sanradial periphery, which in turn increases in field as a result of the radial Leandro for subassembly. The major IVA subassemblies beingconvergence. Such a concept has the high field only at the aperture and fabricated comprise the Marx, oil line, PFL, cell, and stalk.lasts only for the temporal length of the wave which is also the duration Subassemblies and subsystems are subject to a variety of QA testsof the acceleration. which include high voltage testing of the Marx and its trigger, and aThe RTL approach described here is but one of a variety of potential first-article PFL driving both a dummy load and a first-article cell.approaches based on switched RTLs. The approach described here has The initial electrical tests of the Marx system are described in thisa number of innovations over the more common RTL approaches paper. The status of the in-progress fabrication and QA testing for eachwhich are: of the major subsystems is described in a companion paper at this1. The photoconductive switch orientation allows both switch and conference. The Marx tests will comprise a total of approximately 250dielectric transmission lines to operate at their respective breakdown shots at different charge voltages fired into test loads. These shotsfields allowing increased power and average gradient by choosing a establish the operating parameters and demonstrate the performance ofdifferent geometry. the Marx, its trigger system, gas diverter, pre-fire detection circuit, and2. Matching the pulse length and electrode material (gold) to obtain the controls. Electrical measurements are used to validate the design circuithighest accelerating field in the aperture by allowing the field emission and its ability to meet the IVA system requirements.current to operate just below the EM wave current. The complete system will not be assembled and tested in the US. AllThe resulting concept has the potential of pushing the average field of components of the IVA are to be delivered as subassemblies to AWE inaccelerators to the 5MV/cm range. the UK in mid 2012, for assembly and commissioning in a newReferences: Technology Development Centre.1 Chattopadhyay, S., Whittum, D., Wurtele, J. Advanced Accelerator * Work sponsored by AWE, Aldermaston, UKTechnologies. A Snowmass ’96 Subgroup Summary. Proc NewDirection in High Energy Physics.2 Huang, L.C and Byer, R.L. A proposed high-gradient laser-driven 96
  • 96. 2P-55: Solid State Pulsed Power System for 50 MW X-Band The design of a scanning magnet power-supply used in a 10MeVKlystron industrial radiation accelerator is presented in this paper. TheT. L. Houck, G. G. Anderson, S. G. Anderson, C. P. J. Barty, consideration of linear magnetic field current and principle of triangleG. K. Beer, R. R. Cross, G. A. Deis, C. A. Ebbers, D. J. Gibson, wave modulator is included, control system based on single chip andF. V. Hartemann, R. A. Marsh it’s anti-jamming scheme in industrial circumstance is also introduced.National Ignition Facility/Photon Science and Applications, Lawrence The scanning magnet power-supply can produce continuous triangleLivermore National Laboratory, Livermore, CA, United States waveform current, the frequency can be adjusted by step length of 1Hz ranged from 20Hz to 50Hz, and the peak-to-peak current is 20A.In support of X-band photoinjector development efforts at LLNL, a 50MW test station is being constructed to investigate structure and 2P-58: MAGIC Implicit Particle Pusher Description andphotocathode optimization for future upgrades. The test station will Validationconsist of a 5.5 cell RF gun, 263 nm photoinjector drive laser, low A. J. Woods, L. D. Ludekingenergy beam transport, traveling wave accelerator section, and Missile Products, ATK, Newington, VA, United Statestransport/diagnostic section. A SLAC XL-4 klystron driven by aScandiNova Systems K2-3X HV modulator will supply the power for MAGIC2D [1] has been modified to include an implicit particle-in-cellthe RF Gun and accelerator section. Timing of the laser pulse on the (PIC) update scheme (particle “pusher”) for increased time steps dt andphotocathode with the applied RF field places very stringent stability in beam-plasma simulations. The adjustable damping implicitrequirements on phase jitter and drift. The choice of the K2-3X was update method of Friedman [2] allows selectable attenuation of modesdriven primarily by this 250 femtosecond, or 1- degree phase at 11.424 in a plasma. Our approach has been to first develop a simple particleGHz, specification on the jitter. This specification leads to a voltage pusher one-dimensional (1D) finite-difference code model, namedflatness of <±0.25% for the modulator pulse. Other parameters for the PPUSH1, which incorporates implicit PIC as outlined in [2]. Particlesmodulator include 420-kV peak voltage, 330-A maximum current, flat are started with initial energy and applied electric field, and the motiontop pulse length of 1.6 µs, pulse-to-pulse amplitude stability of <±0.1%, computed with an adjustable damping parameter. We have replicatedand a repetition rate range of 0-120 Hz. the electron motion reported in [2] for a potential well with electricStandard pulse forming lines and networks were considered, but not field linear in excursion distance. This model represents the simplestchosen due to voltage flatness and amplitude stability specifications. form of plasma oscillation. The essential result is that for a given w0*dtThe ScandiNova modulator uses a parallel, solid state switching (=0.42 in the test case), the damping parameter can be chosen betweentechnology that has demonstrated our requirements. Other attractive 0 (no damping) to 1.0 (10x damping in 16 periods for our test problem).features of this technology are the relatively low voltage outside of the Hence, for a mix of frequencies w0 in a complex many-particle system,insulating oil/transformer container (~1 kV) and robustness (no unwanted higher frequency oscillations can be adjustably diminishedcascading switch failures). The switches are rated for 1.7 kV and we while preserving the lower values which may be of interest.expect high reliability at the 1 kV operating value. The relatively low Subsequent to verification against [2], the implicit PIC routine wasvoltage allows for a compact grounding architecture that minimizes extended to three dimensions and relativistic dynamics, tested forelectrical noise. The solid state switches will provide greater time magnetic field effects, and incorporated into MAGIC2D. The Lorentzbetween replacements/failures than gas switch or thyratron based equation particle update code closely resembles the 1D subroutinesystems. checked out previously. Additional successful tests included particleWe will describe the commissioning of the K2-3X modulator and deceleration in self fields to analytically solvable positions, andinitial performance of the modulator/klystron system. acceleration through known potentials to relativistic levels. The newThis work was performed under the auspices of the U.S. Department of method agrees with the previous explicit treatment where both haveEnergy by Lawrence Livermore National Laboratory under Contract sufficiently small time step.DE-AC52-07NA27344, and funded by DHS Domestic Nuclear The new implicit particle pusher in MAGIC2D enables the user toDetection Office. select damping of high-frequency modes thus concentrating on plasma fundamental frequencies of interest. This additional user tool, to be2P-56: The Pulsed Kicker Power Supply Design for TPS available in MAGIC version 3.1.0, can prevent instabilities resultingSynchrotron Light Source from ionization growth beyond the time step limit for explicit particles.Y.-H. Liu, C.-S. Chen [1] B. Goplen, et. al., “User-configurable MAGIC for ElectromagneticUtility Group, National Synchrotron Radiation Research Center, PIC Calculations,” Computer Physics Communications 87 (1995),Hsinchu city, Taiwan (http://www.mrcwdc.com). [2] A. Friedman , “Implicit multiscale PIC and related topics”, Workshop on Multiscale Processes in Fusion Plasmas, UCLA, Jan.The purpose of this paper is to design the pulsed bumper power supply 2005 (http://hifweb.lbl.gov/public/slides/).for TPS synchrotron light source. The 3 usec half sine waveform pulser * Work supported by ATKwas designed for top-up injection scheme. Maximum 3 kV peakvoltage and 1 kA peak current will excite the bumper magnet toproduce 0.3 Tesla magnetic field. The circuit design was estimated and 2P-59: Phenomena Succession at Generation of Picosecondtested in the beginning. The bumper prototype will be setup and the Runaway Electrons Beam in Airdetail measurements also will be finished in the future. M. I. Yalandin1, A. G. Reutova1, K. A. Sharypov1, V. G. Shpak1, S. A. Shunailov1, G. A. Mesyats2 1 Laboratory of electron accelerators, Institute of Electrophysics, Ural2P-57:Manufacture of a Scanning Magnet Power-Supply Branch of Russian Academy of Sciences, Ekaterinburg, RussianUsed in Industrial Radiation Accelerator FederationC. Yu 2 Dept. of physical electronics, Lebedev Physical Institute of RussianBeijing Institute of Special Electromechanical Technology, Beijing, Academy of Sciences, Moscow, Russian FederationChina 97
  • 97. New experimental data, obtained with a study of picosecond runaway the feed electrons is needed.electron beams (REB) in air filled electrode gap with nonuniform E-field are presented. Real time experiments were conducted with the In this paper, we will investigate various ideas for dealing with thishighest accessible picosecond resolution. Interrelation of initiating issue. For example, one could consider the effect of a solenoidalemissive processes on the cathode and of ionizing phenomena in gas magnetic field at or near the end of the MITL on the electron flow.was considered. Alongside with time-of-flight measurements of Such fields have been successfully used in other devices, such as ionelectrons energy two alternative determination procedures of voltage diodes [3] and plasma opening switches [4], to control the flow ofranges and time point of REB injection on accelerating pulse front are electrons. These fields could be produced using the MITL currentdemonstrated. It is shown that REB injection critical field in air directly by adding one or more field coils in series with the MITLcorresponds to classical ideas, and the duration of accelerated beam circuit, or through coils that are pulsed using a separate power supply,could be of 10-20 ps. It is explained that limitation in REB duration due possibly on a longer timescale.to passage of field cathode emission into micro explosion ofheterogeneities is highly improbable. It is proven that REB acquires We will present results showing the properties of the various schemesmaximum energy not more than that which is ensured by accelerating investigated, and compare the relative merits of any that prove viable.E-field, determined by the cathode potential. In the directmeasurements it is shown that picosecond beam is not the consequence 1. K. D. Hahn, et al., IEEE Trans. Plasma Sci. 38, 2652, 2010.of discharge processes, but it vice versa initiates the gap breakdown 2. N. Bruner, et al., PRST Accel. Beams, 11, 040401, 2008.with the development of secondary run away electrons avalanches. The 3. S.A. Slutz, D.B. Seidel, and R.S. Coats, J. Appl. Phys., 61, 4970,current of such avalanches is registered for the first time. In our view, 1987.above mentioned results complete self-consistent picture and explain 4. M.E. Savage, D.B. Seidel, and C.W. Mendel, Jr., IEEE Trans.the cause-effect relationship of phenomena in an air gap with the Plasma Sci. 28, 1533, 2000.generation of picosecond REB flows. This work was supported in part ________________________________by the Russian Foundation for Basic Research under Grant 10-08- * Sandia is a multiprogram laboratory operated by Sandia Corporation,00081 and Grant 11-08-00145, and by the RAS Presidium project “The a Lockheed Martin Company, for the United States Department ofproblems of physical electronics, beams of charged particles, and Energy’s National Nuclear Security Administration under Contract DE-generation of EM radiation in the powerful systems”. AC04-94AL85000.2P-60: Laser Excitation of Electrostatic Eigen Mode of a 2P-62: Plasma Lens for Transformation of Ion BeamsPlasma in Azimuthal Magnetic Field and Electron V. D. Bochkov1, A. A. Drozdovskii2, A. A. Golubev2, D. D. Iosseliani2,Acceleration Y. B. Novozhilov2, S. M. Savin2, V. V. Yanenko2 1M. Kumar, V. K. Tripathi Pulsed Technologies Ltd., Ryazan, Russian Federation 2Physics, IIT Delhi, NewDelhi, India SSC RF Institute for Theoretical and Experimental Physics, Moscow, Russian FederationThe localized electrostatic eigen modes of a plasma with parabolicdensity profile in the presence of azimuthal magnetic field are The problem of focusing of intense heavy-ion beams is an importantinvestigated in the slab model by taking stepwise variations of plasma issue for investigating of high energy densities in a matter. Applicationdensity and magnetic field. The mode amplitude is maximum where of plasma lens to this area of research has a number of essentialmagnetic field is maximum. A mode can be resonantly by beating two advantages. It is important that the focusing capabilities of the lenscollinear lasers with frequency difference equal to the frequency of the depend on the stage of plasma development. Under certain conditions aeigen mode. Such a mode efficiently accelerates energetic electrons, magnetic field is linear, which allows to focus the beam to a very smallexecuting bounce motion in the azimuthal field to high energies. At spot. In other conditions, the magnetic field is nonlinear, which allowsdimensionless mode plasma wave amplitude ap = 0.55 , 10 keV formation of hollow and other interesting configurations of beams. Atelectrons can be accelerated to 30 MeV energy. ITEP the plasma lens has been designed and installed into the exit channel of the TWAC accelerator complex. In accordance with the principal goals of this project a pulse-power generator has been2P-61: Controlling Feed Electron Flow in MITL-Driven developed with two copper-vapor thyratrons (pseudospark switches)Radiographic Diodes TDI1-150k/25 and a stable discharge with a current of up to 250 kAD. B. Seidel, T. D. Pointon, B. V. Oliver and pulse duration of 5 µs was achieved. The plasma lens testingAdvanced Radiographic Technologies Dept., Sandia National started by focusing of a carbon C+6 ion beam. As a start the sharpLaboratories*, Albuquerque, NM, United States focusing of an ion beam with focal spot with diameter of 350 µm has been demonstrated. Then it has been shown that the plasma lensThe electrons flowing in a coaxial magnetically insulated transmission provides formation of hollow beams of ions in a wide range ofline (MITL), if allowed to flow uncontrolled into a radiographic parameters [1]. That allows to consider plasma lens as a possibleelectron diode load, can have an adverse impact on the performance of variant of a terminal lens for realization of inertial thermonuclearthe system. Total radiation dose, impedance lifetime, and spot quality fusion. The plasma lens can be used for transformation of beams with(size, shape, position, and stability) can be all affected [1]. One Gaussian distribution of particles density into beams withcommon approach to keeping such feed electrons out of the diode homogeneous spatial distribution. It appears that it is possible for a caseregion involves greatly increasing the radius of the MITL’s outer of equilibrium Bennetts distribution of a discharge current. This(anode) electrode near the diode and adding a large, rounded requires a sufficiently long duration of a current pulse of >10 µs. Usingprotuberance to the load end of the inner cathode conductor, forcing up-to-date pseudospark switches - deuterium thyratrons with hollowthese electrons to be lost to the to the anode outside the diode itself [2]. anode TDI1-200k/25H [2] significant currents with pulse duration up toWhile this approach can be quite successful, there is an inductance 20 μs have been attained. The first beam tests have essentiallypenalty that reduces the current that can be delivered to the load and confirmed expected result. Calculations and measurements have beenrequires a large volume in the vicinity of the load. For applications performed for a C+6 and Fe+26 beams of 200 MeV/a.u.m. energy. Thewhere this volume is not available, an alternate method of controlling obtained results and analysis are presented. 98
  • 98. 1.A.Drozdowskiy et al., Investigation of the formation of a hollow power microwave sources, for they require no external energy source tobeam in the plasma lens. IPAC10, Kioto, Japan, drive thermionic emission. The objective of the research is to improvehttp://cern.ch/AccelConf/IPAC10 /MOPE040. upon the already impressive performance of cesium iodide doped2.J. Slough, C. Pihl, V. D. Bochkov, et al, «Prospective Pulsed Power carbon fibers by decreasing the work function of the emitters utilizingApplications Of Pseudospark Switches», 17th IEEE International specialized subcoating technologies.Pulsed Power Conference (2009), pp. 255-259. Cathode tests were conducted on a modulator which generated an electron beam. The modulator, designed and built at UMC, uses a modified type E pulse forming network (PFN) switched by a hydrogen2P-63: Investigation of the Mechanism of Electron Current thyratron to provide the electrical energy. The electron gun wasSuppression in an Ion Diode with Magnetic Self - Isolation designed to deliver up to 10 A. The PFN can be configured to deliverA. Pushkarev, Y. Isakova, V. Guselnikov pulse lengths from 2.5 to 5 microseconds at voltages from 50 to 120 kV.Tomsk Polytechnic University, Tomsk, Russian Federation An electron gun was designed around a carbon fiber emitter element that was 3 cm in diameter. The emission fibers were grown ontoPlanar diodes with an explosive emission cathode are widely used for graphite bases by ESLI, and the fibers were several millimeters longgeneration of a wide-aperture pulsed electron and for ion beams with and 8 µm in diameter. Several emitters were prepared, one having nocurrent densities of more then 20 A/cm2. After application of voltage to treatment, one having a conventional cesium-iodide coating, and onethe diode and following plasma formation on the anode, the generation having novel coatings deposited on the carbon prior to cesium-iodideof electron and ion current occurs simultaneously. The current density treatment. The results of the emission research are reported.of protons in a mode of space charge limitation is 2.3% of the electroncurrent density. Current density of heavier ions is much lower. For an 2P-65: The Study of Pulsed Explosive Ion Emissionefficient generation of an ion beam it is necessary to ensure suppression A. Korenev, S. Korenevof the electronic component of the total diode current. In 1973, Sudan Beam & Plasma Technologies, Inc, Mundelein, IL, United Statesand Lovelace [1] first suggested a construction of a pulsed ion diodewith an external magnetic insulation. In a diode with magnetic selfisolation a transverse magnetic field in the a-c gap is formed by a diode The pulsed explosive ion emission was at first time described about 20self-current as it flows. The absence of an additional magnetic field years ago. The following study of this type of ion emission wassource simplifies the design, but still the ion current does not exceed addressed to the understanding of physical aspects of the phenomena.10-15% of the total diode current. In this paper we analyze the possible The experimental and theoretical results of emission, the timereasons for a low efficiency of the electronic component suppression in characteristics and intensity of electrical field in ion diode are presenteda diode with magnetic self-isolation. The experiments were conducted in the paper. The emission characteristics for different ions areon a TEMP-4M ion accelerator [2]. Results of a study of a stripe ion discussed. The main parameters of pulsed ion emission are presented.diode with an explosive-emission potential electrode are presented. It is The some recommendations for design of pulsed ion sources based onshown that the contribution of a weakly magnetized part of the diode to the phenomena are given.the total current does not exceed 5-10%. Studies have shown that aweak suppression of electron current is due to a high electron drift 2P-66: Low-Energy Electron Beam Sourcevelocity, exceeding 1 cm/ns. For a stripe diode of 25 cm long the drift J. Gleizer, V. Vekselman, S. Yatom, J. Felsteiner, Y. Krasiktime of electrons is less than the pulse duration of accelerating voltage Physics, Technion, Haifa, Israeland the coefficient of suppression of electron current is 1.5-2. Thispaper also introduces the design and results of the research of bothannular and conical shaped diodes with magnetic self-isolation. Experimental results of using a ferroelectric-plasma-source-assistedIncreasing the length of the diode to 450 cm provides a reduction of the hollow anode discharge as a source of low-energy electron beams aretotal diode current (when the area of the anode is increased by 2 times). presented. To generate electron beams, the hollow anode auto-biasAn electron current suppression coefficient in a diode of a new design negative voltage was produced by the discharge current flowinghas increased to 3-4. The results of research on homogeneity of the through the resistor connecting the hollow anode and the groundedgeneration of an ion beam over the diode area are also presented. electrode. It was shown that this method allows reliable andThermal imaging diagnostics [3], adapted to a diode set in a double- reproducible generation of low-energy electron beams with an electronpulse mode, have been utilized for this study. energy of several hundreds of eV, electron current density up to severalReferences: A/cm^2 and pulse duration of several tens of microseconds.1. R. N. Sudan and R. V. Lovelace. Phys. Rev. Lett., 31(16), 1174(1973). 2P-67: Shielded-Grid Triode Eliminates Beam-Interception2. G. E. Remnev, I. F. Isakov, A. I. Pushkarev. at all, Surf. Coat. Switching LossesTechnol. 114, 206 (1999). K. E. Williams3. H. A. Davis, R. R. Bartsch, J. C. Olson, D. J. Rej, and W. J.Waganaar. J. Appl. Phys. 82 (7), 3223 (1997). Applied Science and Engineering, LLC, Hartford, WI, United States High current, high frequency pulsed electron guns with non-2P-64:Investigation of a Novel Cathodes for Microsecond intercepting aperture grids can eliminate many of the disadvantagesPulse Conditions associated beam-intercepting grid designs. These disadvantages canP. Norgard, R. D. Curry include: grid heating, thermal expansion, grid distortion, and spuriousDept. of Electrical & Computer Engineering, University of Missouri, grid emission due to deposited cathode material; moreover, the currentColumbia, MO, United States and power losses associated with grid current interception can be damaging to the electron gun and unacceptably high for the grid powerThe University of Missouri – Columbia (UMC) is investigating cathode supply. Grid distortion can cause unwanted spatial and temporalsurface coatings applied to micro-fibrous carbon emitters in an effort to variations in the extracted current and uncontrollable beam optics. Forincrease low field electron emission. Low field, high electron density multi-megawatt electron guns, the grid current interception losses cancathodes are highly desirable components of non-relativistic high be excessive if not destructive. These problems can be largely 99
  • 99. overcome by designing pulsed or dc electron guns with non- electromagnetic (EM) simulation models of high-current LTD cavitiesintercepting aperture grids. When operating such electron guns at or have been developed [D.V. Rose et al. Phys. Rev. ST Accel. Beams 13,near their rated design current and voltage, the grid current interception 90401 (2010)]. An equivalent two-cavity transmission line model ofis zero. However, during the switching transition from cutoff to rated Mykonos-2 using the BERTHA transmission line code is presented.current, there are significant changes in the strength of the grid-cathode The model explicitly includes 40 bricks per cavity and detailedlens. This changing lens effect can alter the beam optics enough to representations of the water-filled transmission line and resistive load.cause the beam to strike internal gun parts, beam line components, or (A brick consists of two capacitors and a switch connected in series.)both. For dc operation this may not be a problem since these losses This model is compared to 3D EM simulations of the entire acceleratoronly occur during turn-on and turn-off. However, for repetitively including detailed representations of the individual capacitors andpulsed electron guns beam current interception losses during switching switches in each cavity. Good agreement is obtained between the twocan be unacceptably high and sometimes destructive; this is especially simulation models and comparisons with available experimental datatrue for high frequency operation. Using the results of a finite element will be presented. Output pulse variations produced by varying relativesoftware code, this paper illustrates these beam interception switching firing times of the two cavities are explored.losses and describes how they can be eliminated altogether using a *Sandia is a multiprogram laboratory operated by Sandia Corporation,shielded-grid triode design. a Lockheed-Martin company, for the United States Department of Energy’s National Nuclear Security Administration, under contract DE- AC04-94AL85000. Sub-Nanosecond Electron Emission from Electrically2P-68:Gated Field Emitting ArraysM. Paraliev1, S. Tsujino2, C. Gough1, E. Kirk2, S. Ivkovic1 2P-70: A Versatile Marx Generator for Use in Directed1 RF section, Pulsed magnets group, Paul Scherrer Institute, Villigen Energy and Effects Testing ApplicationsPSI, Switzerland T. A. Holt, J. R. Mayes, M. B. Lara, C. Nunnally, J. M. Byman,2 C. W. Hatfield Laboratory for Micro-& Nanotechnology, Field-emitter group, PaulScherrer Institute, Villigen PSI, Switzerland APELC, Austin, TX, United StatesField Emitting Arrays (FEAs) are a promising alternative to the Applied Physical Electronics, L. C., (APELC) offers many Marxconventional cathodes in different vacuum electronic devices such as generators with stored energies ranging from 5 mJ to 1.8 kJ. The line oftraveling wave tubes, electron accelerators and etc. Electrical gating Marx generators offered by APELC can be used in a variety ofand modulation capabilities, together with the ability to produce stable applications including flash x-ray, high power RF, high powerand homogeneous electron beam in high electric field environment are microwave, test and evaluation, triggering, and material studies. Thethe key requirements for their practical application. Due to relatively MG15-3C-940PF (MG15), in particular, has seen wide use andhigh gate capacitance, fast controlling of FEA emission is difficult. In integration into many systems over the past several years. The MG15 isorder to achieve sub-nanosecond, electrically controlled, FEA based a 33-J, 50-Ohm source and is capable of limited duty at 150 Hz PRF.electron emission a special pulsed gate driver was developed. Bipolar The generator delivers a maximum output voltage of 300 kV into ahigh voltage pulses are used to rapidly inject and remove charge form matched load. The range of capabilities and custom configurationsFEA gate electrode controlling quickly electron extraction gate voltage. achieved by the MG15 as well as a sampling of the applicationsShort electron emission pulses (<600ps FWHM) were observed in low featuring the use of the MG15 will be presented. Recent additions toand high gradient (up to 12 MV/m) environment. First attempts were the capabilities of the MG15 include sub-ns rise time, jitter of less thanmade to combine FEA based electron emission with radio frequency 2 ns, and an energy density approaching 3.2 mJ/cm3 (90 J/ft3).acceleration structures (1.5 GHz) using pulsed preacceleration. Thegate driver design together with low inductance FEA chip contact 2P-71: Improvements to a Small Scale Linear Transformersystem is described. The results obtained in low and high gradientexperimental setups are presented and discussed. Driver D. W. Bolyard, A. Neuber, J. Krile, M. Kristiansen Department of Electrical and Computer Engineering, Texas Tech2P: Pulsed Power Systems posters I: Electromagnetic University, Center for Pulsed Power and Power Electronics, Lubbock,Launch, Generators and Networks, and Lasers TX, United StatesTuesday, June 21 13:30-15:30 Regency Ballroom A linear transformer driver (LTD) is being constructed at Texas Tech Transmission Line and Electromagnetic Models of the2P-69: University’s Center for Pulsed Power and Power Electronics with aMykonos-2 Accelerator* goal to achieve roughly 100 ns pulse width into an 18 Ohm load andE. A. Madrid1, C. L. Miller1, D. V. Rose1, D. R. Welch1, R. E. Clark1, energy densities similar to or greater than a previously designed andC. B. Mostrom1, W. A. Stygar2, M. E. Savage2, D. D. Hinshelwood3, built 500 J compact Marx generator. A single experimental LTD stage,K. R. LeChien4 previously constructed, tested, and reported on by TTU, has undergone1 Voss Scientific, Albuquerque, New Mexico, United States several improvements as well as the inclusion of additional diagnostics.2 Sandia National Laboratories, Albuquerque, New Mexico, United Low charging voltage testing into a 1.7 Ohm resistive load hasStates experimentally achieved a 10% to 90% risetime of ~80 ns, peak voltage3 Naval Research Laboratories, Washington DC, United States of 4.4 kV, peak current of 2.8 kA, and instantaneous power of 11.54 National Nuclear Security Administration, Washington DC, United MW. Simulations of the LTD stage correlate closely with theStates 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 150Mykonos is a linear transformer driver (LTD) pulsed power accelerator MW from a single LTD stage. The design for a multi-stage LTD usingcurrently undergoing testing at Sandia National Laboratories. magnetic cores with a flux swing of 6 kV-μs will be presented.Mykonos-2, the initial configuration, will include two 1-MA, 200-kV Improvements in diagnostics and modeling of the LTD will also beLTD cavities driving a simple resistive load. Transmission line and 3D discussed as well. 100
  • 100. 2P-72: Design of a Non-Circular Linear Transformer Driver purity of the pulsed output power for both types of amplifiers are alsoK. Morales1, B. Hilko2 presented. Results of this paper can be used in the development of RF1 Directed Energy Integration Branch, Naval Surface Warfare Center, pulse amplifiers for plasma applications in semiconductor waferDahlgren, VA, United States processing, MRI, and radar applications.2 Envisioneering, Inc., Alexandria, VA, United States 2P-74: Square Pulse LTDThe Directed Energy Division of the Naval Surface Warfare Center – A. A. Kim1, M. G. Mazarakis2, V. A. Sinebryukhov1,Dahlgren Division (NSWC-DD), in support of the Joint Non-Lethal V. M. Alexeenko1, S. S. Kondratiev1, W. A. Stygar2 1Weapons Directorate, has developed a preliminary design for a non- Institute of High Current Electronics, Tomsk, Russian Federation 2circular linear transformer driver (LTD). Linear transformer drivers are Sandia National Laboratories, Albuquerque, New Mexico, USAan attractive option for producing high-voltage, high-current pulsesversus the traditional Marx generator. The majority of LTDs to-date are The usual LTD architecture provides sine shaped output pulses thatcomprised of multiple bricks connected in parallel and arranged in a may not be suitable in flash radiography, high power microwavecircular array to generate high currents for driving magnetically production, z-pinch drivers, and certain other applications. A moreinsulated transmission lines (MITLs) or accelerators. However, suitable driver output pulse could be that of a flat or inclined topdevelopment of a compact, rectangular configuration is necessary for (slightly rising or falling). In this paper, we present the design and firstutilization of this technology in non-lethal weapons applications. The test results of an LTD cavity that generates this type of output pulse byNSWC-DD is currently working on building an initial 2 stage LTD including within its circular array some number of third harmonicusing a rectangular bi-plate design scheme. The preliminary LTD bricks in addition to the main bricks.design, analysis, and circuit model will be presented.Distribution A. Approved for public release; distribution is unlimited. 2P-75: 10 Ohm High Voltage Subnanosecond Pulse Generation2P-73: Pulsing Characteristics of RF Amplifier Topologies S. El Amari, D. Arnaud-Cormos, P. Leveque, V. CoudercA. Eroglu Limoges University - XLIM, Limoges, FranceEngineering Department, Indiana University - Purdue University FortWayne, Fort Wayne, United States In the last decade, nonthermal effects of nanosecond pulsed electric field (nsPEF) on biological samples have been studied [1][2]. The shortRF pulse amplifiers have been widely used in medical resonance duration of the pulses opens new possibilities to reach the interior ofimaging (MRI) [1], radar [2], and plasma applications [3] due to their biological cells. These biological experiments require specific pulseseveral benefits. RF pulse sources increase process quality in plasma generators. The challenge consists in developing generators providingapplications for semiconductor manufacturing, and enables selective nanosecond and sub-nanosecond duration pulses. The generatorexcitation in MRI. Pulse generators are used in ultra wideband (UWB) principle is based on a technology developed a few decades ago i.e.radars for military applications to detect buried land mines and for frozen wave generator. The originality of the proposed generator reliesother geo-physical applications. on the use of an ultrafast photoconductive semiconductor switchesThe pulsing profile of RF power amplifiers differs significantly (PCSS) working on its linear photoconductive regime [3]. The pulseddepending on the topology used in its design. Linear amplifier power generator should be impedance matched with the deliverytopologies such Class A, and Class AB have inherently cleaner signal system in order to ensure an efficient energy transfer. The nsPEFoutputs in comparison to switch-mode power amplifier topologies. experiments frequently use delivery systems whose equivalentHowever, there are some disadvantages and limitations associated with impedance is usually rather low (close to 10 Ohm) such as standardlinear amplifiers. RF pulsing with linear amplifier topologies can be electroporation cuvettes. In this case, it seems appropriate to developachieved using dc power supply or driver signal. Pulsing with dc power low impedance generators (10 Ohm). The 10 Ohm generator issupply has certain restrictions on the dynamic response of the pulsed composed of a transmission line and two PCSS, which supports 4 kVoutput signal due to response time of the dc rail voltage. As a result, of static bias voltage for example, placed at each line extremity. Oncehigh pulsing frequencies with small duty cycles are not easily the line is charged by a DC high voltage source (i.e. 4 kV) through aachievable. RF pulsing for linear amplifiers can be also obtained using bias resistor, the PCSS are triggered by optical light. Thereby, adriver signal. RF driver signal only enables two level, On-Off, pulsing stationary wave will be enclosed between the two PCSS due to ausing linear amplifier topologies. In addition, linear amplifiers have progressive and a regressive wave. Those two waves can generate avery poor thermal profile and low efficiency. This limits amount of squared or a bipolar signal depending on the state of the two PCSS. Indeliverable-pulsed power that can be obtained. Hence, linear amplifiers the “on state”, the PCSS exhibits a residual resistance which dependsmight not be cost effective depending on the application. Switch-mode on the semiconductor, on the incident optical pulse energy and on theamplifiers have several advantages when they are used as pulse bias voltage. The PCSS residual resistance value has been properlyamplifiers. They have very good thermal profile and high efficiency. characterized as a function of the optical energy and of the bias voltage.Small duty cycles and high pulsing frequencies can also be achieved The measured “on state” resistance during the switching process wasusing several techniques. However, switch-mode power amplifiers have low (~ 3 Ohm for 4 kV bias voltage) ensuring minimum reflections andsome disadvantages that need to be considered when they used in keeping significant switching efficiency. In this study, a 10 Ohmpulsing applications. Switch-mode amplifiers have very non-linear nanosecond and subnanosecond generator is presented. Two types ofcharacteristics due to their resonant load network design. When they laser are used to activate the PCSS in the linear photoconductiveare terminated with dynamic loads, amplifier response can be regime: a mode locked picosecond (30 mJ) and an ultracompactdramatically impacted. The pulsed output signal can be distorted and microchip nanosecond (50 µJ) laser. Kilovolt, nanosecond andthis may result in a sustained oscillation. subnanosecond pulses with adjustable shape are generated. The rise andIn this paper, we investigate and compare the pulsing characteristics of fall times are in the subnanosecond range depending on the opticallinear and switch-mode RF power amplifiers. Several linear and pulse duration.switch-mode amplifiers are designed and simulated. Their pulsingresponses including rise time, fall time, and overshoot are compared for [1] K. H. Schoenbach et al., “Ultrashort electrical pulses open a newseveral different load conditions. Efficiency, thermal profile and signal 101
  • 101. gateway into biological cells,” Proceedings of the IEEE, vol. 92, n°. 7, 2P-78: Pulsed Power Generator Driven by FPGA and PCp. 1122–1137, 2004. M. Akiyama1, T. Goh1, M. Suemitsu2, T. Sakamoto2, H. Akiyama2,[2] J. M. Sanders, A. Kuthi, Y. H. Wu, P. T. Vernier, et M. A. T. Ueno3Gundersen, “A linear, single-stage, nanosecond pulse generator for 1 Ichinoseki National College of Technology, Iwate, Japandelivering intense electric fields to biological loads,” Dielectrics and 2 Kumamoto University, Kumamoto, JapanElectrical Insulation, IEEE Transactions on, vol. 16, n°. 4, p. 1048– 3 Oita National College of Technology, Oita, Japan1054, 2009.[3] S. El Amari et al., “Kilovolt, Nanosecond, and Picosecond ElectricPulse Shaping by Using Optoelectronic Switching,” IEEE Photonics The control of pulsed power generator using a field programmable gateTechnology Letters, vol. 22, n°. 21, p. 1577-1579, 2010. array (FPGA) has produced an advantage such as a compactness of pulsed power generators and the pulsed power control for many applications. However, the development of pulsed power generator2P-76: Generators of High-Power High-Frequency Pulses using programming is difficult in noise problems and others.Based on Sealed-off Discharge Chambers with Hollow The present work aims to develop a high-performance pulsed powerCathode generator using FPGA and a personal computer (PC). The pulsedV. D. Bochkov1, V. G. Ushich1, A. E. Dubiniv2, I. Y. Kornilova2, power generator consists of a pulsed power generator with magneticI. L. L’vov2, S. A. Sadovoy2, V. D. Selemir2, D. V. Vyalykh2, pulsed compression (MPC), a charger, a controller using FPGA, aV. S. Zhdanov2 universal serial bus (USB) connection circuit and a software running on1 Pulsed Technologies Ltd., Ryazan, Russian Federation Windows PC. The software running on PC is written with Visual C#,2 Russian Federal Nuclear Center – All-Russian Research Institute for and has a graphical user interface (GUI). This easy operational systemExperimental Physics, Sarov, Russian Federation can control several functions, such as output voltage, pulse interval, pulse repetitive rate and pulse number. The developed pulsed power generator driven by FPGA and PC is applicable to many fields such asIn pulse high-voltage low-pressure gas-discharges with hollow cathode sterilization, water treatment and bioelectrics.a full modulation of discharge voltage can occur at the frequency in arange of some tens of Megahertz. This phenomenon can be used togenerate high-power high-frequency pulses. One of the factors, limiting 2P-79: A FPGA Based All-Solid State Nanosecond Pulsed-application of such generators is a vacuum system itself, gas-filling Power Generatorsystem and chamber pressure control. Application of sealed-off gas- C. Yao, X. Zhang, C. Li, F. Guo, Y. Mi, C. Sundischarge chambers makes the high-frequency generators smaller and Chongqing University, Chongqing, Chinamore mobile [1]. Similar generators can be used e.g. to testelectromagnetic field influence on radio-electronic equipment. The In this paper, a newly developed all-solid-state pulsed-power generatorresults of development and tests of repetitively pulsed generator of using metal oxide semiconductor field effect transistors (MOSFETs)high-power high-frequency pulses based on sealed-off discharge for research applications is proposed. The proposed pulsed-powerchambers with hollow cathode enclosed into a metal-ceramic envelope generator has modular design, it consists of three functionally distinctand maximal dimensions of 37 mm in diameter, length of 125 mm and units: a synchronous trigger circuit, a Marx bank and a controller usingweight of 250 gram are described. Pulses with burst duration up to a field programmable gate array(FPGA). The trigger circuit produces1300 ns and peak power up to 1 MW at frequency of 120 MHz with trigger pulses to synchronously driver solid-state switching units and isrepetition rate of 100 Hz were obtained using the chamber of similar electrically isolated by optical fibers. The Marx bank comprises of 12design and equipped with a nitrogen (or hydrogen) source. stages and each stage includes a capacitor, a solid-state switching unit1. A.E.Dubinov, I.Y.Kornilova, I.L.L’vov, S.A.Sadovoy, et al, IEEE and two diodes. The FPGA controller controls the charge voltage andTrans. on Plasma Sci., vol.38, no.11, pp. 3105-3108, Nov. 2010. trigger timing of the proposed generator. And the characteristics of the* This work was supported by Grant RFBR 10-02-00137-a. generator can be changed by altering the programming of VHSIC hardware description language on the FPGA. This generator has the2P-77: High-Current Pulse Generator for Plasma Focus capability of producing repetitive pulses with a voltage up to 10kV,A. V. Nashilevskiy1, V. P. Vinogradov2, G. G. Kanaev1, V. I. Krauz2, pulse width of 200-1000ns, rise time of 35-50ns and repetition rate ofG. G. Remnev1 1-1000Hz with a 200 Ω resistive load and 1kV DC input voltage. The1 all-solid-state design makes the generator compact, reliable and National Research Tomsk Politechnic University, Tomsk, RussianFederation inexpensive to construct. Initial experiments were carried out to verify2 the performance of the proposed generator. Russian Research Center "Kurchatov Institute”, Moscow, RussianFederation 2P-80: Inductive Adder Based Method for GeneratingAn electrical circuit is described and the construction of high-current Electromagnetic Pulse with Controllable Timingpulse generator is represented with output voltage ~ 20kV, current ~ Z. Wang, X. P. Yan, P. Li, X. H. Hao, C. D. Yu, J. T. Wang100kA, operating in a nanosecond range of time (~ 300ns), with pulse State Key Laboratory for Mechatronical Engineering and Control,repetition rate up to 10Hz in a continuous mode and intended for Beijing Institute of Technology, Beijing, Chinaplasma focus supply. In the generator the capacitive storage dischargeis used via toroidal pulse transformer to electrodes of the plasma focus.Low-inductance multichannel gap discharge is used in the generator as In this paper, a novel method for generating the electromagnetic pulses,a switch. A stored energy of the generator is ~ 600J. The results of in which both of timing and peak value are controllable, is proposed.experimental work of the generator in short-circuit conditions and in For the proposed method, the capacitor is utilized as the battery storagethe case of equivalent load are given. This work was supported partially and the hydrogen thyratron is considered as the switch for powerby the Russian Foundation for Basic Research (project 10-08-00287-a) supply. By simultaneously triggering multiple hydrogen thyratroni, the achievable voltage of electromagnetic pulse can be increased remarkably with a peak voltage of 50kv at least. By changing the voltage of the capacitor or the number of superimposed hydrogen thyratroni, the peak value of the generated electromagnetic pulses can 102
  • 102. be adjusted to the desired value. In addition, owing to the rational Modules have been tested for pumping of power gas discharge laseroverall arrangement in the proposed method, the pulse front edge is with peak current of 50 kA. Long run time testing has shown highcontrolled to be within 20ns. Meanwhile, with the aid of DSP based reliability and stability of all parameters.auxiliary circuit, the interval of electromagnetic pulses can bepredetermined, which results into the electromagnetic pulses with 2P-83: Study of the Discharge Channel Evolutioncontrollable timing as well as peak value and a minimum pulse intervalof 10us. Simulation results show that our proposed method achieves all Characteristics in the Flashlampsthe anticipative targets and is effective for generating the X. Li1, S. Jia1, R. Li1, X. Song1, Z. Shi1, H. Li2 1electromagnetic pulses. School of Electrical Engineering, Xian Jiaotong University, Xian, Shaanxi, China 2 Shanghai Institute of Optics and Fine Mechanics, Chinese academy of2P-81: Development of 20kj Fast Driver for Nx-3 Plasma Sciences, Shanghai, ChinaFocusR. Verma1,2, T. L. Tan1, P. Lee1, A. Talebitaher1, H. B. M. Shariff1, This work is devoted to study the evolution characteristics of theS. V. Springham1, A. Shyam3, R. Rawat11 discharge channel (DC) in the flashlamps. Various influencing factors Natural Sciences and Science Education, National Institute of on the DC evolution process are studied, including pre-ionizationEducation, Nanyang Technological University, Singapore, Singapore schemes, installation schemes, charging energies, main pulse spans,2 Institute for Plasma Research, Bhat, Gandhinagar, India with or without pre-ionization pulse and time delays between the pre-3 Energetics and Electromagnetic Division, Bhaba Atomic Research ionization pulse and the main pulse. Besides, the influence ofCentre, Autonagar, Vishakhapatnam, India neighboring flashlamps is studied when working in parallel with other flashlamps. It is found that the electric field distribution in theEfficient, high neutron yield dense plasma focus device requires fast flashlamp working environment (affected by the pre-ionizationrising, high current and high-energy electrical pulse. To fulfill this schemes, installation schemes and neighboring flashlamps) has theobjective, a 20kJ fast capacitor bank capable of delivering ~900kA of most significant influence on the DC evolution process, determiningpeak current within 3 microsecond rise time has recently been the DC shape, while other factors have influence on the DC brightnesscommissioned and tested at Plasma Radiation Source Lab, NIE, and filling degree.Nanyang Technological University, Singapore, for driving the new‘NX-3 Plasma Focus’. The bank consists of eight 12.5 microFarad high 2P-84: Comparative Analysis of High Velocity Projectileenergy density capacitors connected in parallel (each having storedenergy of 2.5kJ at 20kV). For rapidly transferring the energy stored in Images Using MATLABthe capacitors to the load (i.e. dense plasma focus) pseudospark B. M. Huhman1, A. Choi2 1switches have been used. All pseudospark switches, installed on Plasma Physics Division, US Naval Research Laboratory, Washington,individual capacitors are triggered in sync during the cumulative DC, United States 2discharge. To ensure the synchronized operation of pseudospark Global Strategies Group, Crofton, MD, United Statesswitches, an indigenously designed, high performance 8-channelBlumlien pulser based triggering system of low rise time (<10ns) and The Materials Testing Facility (MTF) at the U.S. Naval Researchlow jitter (±1ns) has been used. Individual modules are connected to Laboratory (NRL) routinely captures orthogonal x-ray images ofthe centralized collector plate using 8 nos. of URM67 cables of 3m launch packages (LP) as they travel through the flight tube after alengths. The capacitor bank also facilitates operation at various energy launch. Located about a meter from the target plates, the x-ray imageslevels (ranging from 3kJ to 20kJ). The measured residual inductance provide a useful diagnostic for determining the performance of theand resistance of the capacitor bank (contributed by capacitors, materials used during the test. The images are captured on digital imageswitches, transmission lines and connections) is about ~38nH and plates, a reusable technology that uses a laser to scan the photo-~3.6milliohm respectively. In the presented paper – pulsed power stimulated phosphor particles, thus rendering a digital image that can behardware, electrical performance, protection and design aspect of the processed by a computer software package. While the image alone isnewly developed pseudospark switch based 20kJ fast driver has been useful, it would be beneficial to have a detailed analysis of the LPreported. captured in the x-ray image, such as a comparison between a virgin static image and the post-launch image to determine where damage has2P-82: High Current All Solid State Sources for Laser occurred and to what extent. The MATBLAB script that was developed will calculate the area of the LP in both the test and original versions,Applications and compare the two images. There are a number of technicalV. Efanov, M. Efanov, P. Yarin challenges that were addressed to develop the analysis. Because theFID GmbH, Burbach, Germany projectile is traveling at hypervelocity, it is difficult to obtain a clear image of the projectile, as the pixel intensity value, which measures theCompact high voltage pulse modules with operating amplitude of up to brightness of a pixel, only varies by a few values from that of the300 kV and peak current of up to 10 kA based on fast ionization background. In addition, the location of the projectile on the image isdevices (FID) switches have been developed. Duration of current can always different, owing to variations in the velocity of the LP. Tovary from a fewtens of nanoseconds to hundreds of nanoseconds. address these issues, a series of image enhancement routines wereOperating pulse repetition rate (PRF) in continuous mode can be up to developed. The development of these techniques and user interaction25 Hz. Energy in pulse for a single module is up to 1 kJ. scripts will be discussed in this paper.Turn on (timing) stability of these pulser modules is better than 100 ps,which permits synchronous operation of many modules into one or 2P-85: Application Based General Scaling in Railgunsmoresingle loads. Dimensions of a module with pulse energy in pulse V. W. Sung, W. G. Odendaalof 1 kJ are approximately 120x30x30 cm.Operation of mModules are designed to operate on DC requites input Dept. of Electrical and Computer Engineering, Virginia Polytechnicvoltage of up to 5 kV, plus an external triggering needs a pulse with Institute and State University, Blacksburg, VA, United Statesamplitude of 100 V. 103
  • 103. When it comes to scaling railguns, there is no best or one-size-fits-all by dedicated MOSFET H-bridges and dedicated battery cells.methodology. It is important for the researcher to be able to choose the While the individual structures see but a few hundred volts and underappropriate scaling methods to best match the application in an 2kA, the combined effect induces an accelerating magnetic field aboveexperiment. Previous literature on scaling has specialized on 10 Tesla, consistent with accelerating around a kg to a few km/s in amaintaining constant field parameters such as temperature, magnetic 5m long gun.field, and stress, but this type of scaling may not be appropriate for all The unique features of the system include; i) virtually no magneticapplications. This paper is an overview of railgun scaling methods; the energy left in the gun behind the projectile; ii) no fringe fields externalgeneric railgun equations are tailored for circuit, kinematic, field, to the gun; iii) accelerating area a factor of ~10 larger than rail guns; iv)thermal, and mechanical applications to obtain suitable scaling methods. reduced overall power requirement by as much as a factor of 3 due theAll scaling methods come from the same basic equations, and each reduced inductive component and IR losses.method has at least one or more scaling inconsistency. The importance Other unique features include no individual high current or voltageof inconsistencies and how certain parameters scale is dependent on the components are used MOSFET at above a kA/cm2 and at a fewapplication. For example, one application might require that the circuit hundred volts have recently appeared on the market. Also, since at nobehavior be maintained for the testing of control algorithms. In this point is the power added electrically, an extremely robust design withcase, the temperature and other field quantities are not important; easily replaceable components is possible since the gun componentshowever, in another application these quantities become important for and the power supply are on and the same.studies involving the interaction of magnetic forces on a thermally The mm circuit element scale is chosen for two reasons: i) the smallloaded sliding metal armature. Circuits can be scaled based on pole size reduces the fringe field volume to that scale; ii) the power,geometry, the power supply side, the railgun side, or the system as a voltage and current associated with each of the single turn motorwhole. Kinematic and mechanical scaling would be concerned about elements matches favorable with that available from MOSFET bridgesthe projectiles acceleration, velocity, and displacement and their and allows for no turn low voltage low current geometries.interaction with the projectile and its payload. Field scaling involves MEMS manufacturing of insulated copper conductors in variousthe relation between magnetic field, current density, temperature, and substrates have been demonstrated. The concept has the promise of lowother fields with varying conditions, while thermal scaling would be cost EM gun systems at many different configurations from virtuallyrelevant for studies on cooling and heat management. There is no identical parts.practical way to perfectly and linearly scale a railgun. Each method hasits own tradeoffs and compromises, and some methods are better than 2P-88: Control of Thermal Limitations in Railgunsothers depending on what the researcher values in a study. G. A. Shvetsov1, S. V. Stankevich1, A. G. Anisimov1, S. V. Sinyaev2 1 Lavrentyev Institute of Hydrodynamics, Novosibirsk, Russian2P-86: Sliding Electrical Contact Test Stand Development FederationD. A. Rice, S. D. Kovaleski, J. M. Gahl 2 Institute of Applied Mathematics and Mechanics, Tomsk StateElectrical Engineering, University of Missouri, Columbia, Mo, United University, Tomsk, Russian FederationStates During the last twenty years, considerable attention of researchersThe purpose of this research is to create a test stand capable of working in the areas of pulsed power, plasma physics, and high-assessing the physics of a high-pressure high-velocity sliding electrical velocity acceleration of solids has been given to electromagneticcontact. Currently, no method exists for testing the electrical contact methods of accelerating solids. Projects have appeared whose authorsnear muzzle velocities for rail guns. Lower velocity tests have been believe that the use of electromagnetic forces to accelerate solids willused to simulate conditions near the breech of a rail gun [1], and make it possible to design accelerators with characteristics superior toinspection of rails after several shots has been used to characterize those of light gas guns, explosive accelerators of macroparticles, andseveral effects in the sliding electrical contact [2, 3, 4]. Studies at lower advanced cannon artillery. For high-velocity accelerators of solids, thevelocities do not fully characterize the high-velocity contact, and rail most important are the following two questions: 1) what absoluteinspection does not provide a very thorough analysis of what occurs in velocities can be achieved in a particular type of accelerator for a bodythe electrical contact near the muzzle of a rail gun. This research aims of a given mass, and 2) what is the service life of the accelerator? Anto design a machine that can supplement the data that is currently analysis has shown that the most severe limitations in attaining highavailable. velocities and providing acceptable service life of electromagnetic[1] Bauer, D. P. “Rapid Testing For Multishot Railgun Bore Life”, accelerators are the thermal limitations due to the circuit current. AnIEEE Transactions On Magnetics, Vol, 33. No. 1, January 1997 increase in the current density near the rear surface of the armature in[2] Thurmond, L. E. “Measurement of the Velocity Skin Effect”, IEEE railguns, due mainly to the velocity skin effect, leads to rapid heating,Transactions on Magnetics, Vol. 27, No. 1, January 1991 melting, and vaporization of the armature near the contact boundary.[3] Senouci, A. “Damage of surfaces in sliding electrical contact copper The development of these processes result in a rapid transition to an arc/ steel”, Applied Surface Science, 1999 contact mode, enhancement of erosion, reduction or termination of the[4] Zhao, H. “Friction and wear in high speed sliding with and without acceleration, and destruction of the barrel and accelerated body. Theelectrical current”, Wear, 2001 analysis shows that the ultimate velocity under the heating conditions of solids in railguns can be substantially increased by using composite conductors with controllable electrothermal properties. In the paper, the2P-87: Low Voltage Low Current Massively Parallel High possibility of controlling the current concentration region on thePerformance EM Gun Topology MEMS Based perimeter of the rail-armature interface in electromagnetic acceleratorsManufacturing of conducting solids is analyzed by numerical simulation of unsteadyO. S. Zucker distributions of magnetic fields, currents, and temperatures in a three-Polarix Corporation, San Diego, CA, United States dimensional formulation. The objective of the work is to reduce the rate of Joule heating of armatures during acceleration and to increase theA new concept for an induction type electromagnetic gun is proposed. ultimate (under heating conditions) kinematic characteristics ofMEMS technology is used to construct thousands of millimeter scale 3 accelerators by choosing the optimal shape and structure of armaturesphase half turn independent structures which are individually powered and rails and the electrothermal properties of the armature and rail 104
  • 104. materials. Results from experimental studies of the effect of the force of new type of intercepting missile is simulated byelectrothermal properties of the rail and projectile materials on the MAXWELL3D module of the finite element analysis softwareultimate velocities of projectiles in arcless acceleration of metallic Ansoft12. The analysis of 3-D eddy filed on intercepting missile whichsolids are also presented. It is shown that, by controlling the structure contains magnetoconductive material is very important not only to findand electrothermal properties of the electrodes and accelerated body, it out distribution laws of endured force of intercepting missile, but alsois possible to significantly reduce the thermal limitations during high- to deign a new type of missile.velocity electromagnetic acceleration of solids and increase theultimate kinematic characteristics of railguns in subcritical modes of 2P-91: Structural Optimization of the Single-Stage Inductiontheir operation. Coilgun Based on Finite Element Method B. Zou2P-89:Modeling of High Voltage Fast Recovery Diode in Naval Aeronautical and Astronautical University, Yantai, Shandong,Pulsed Power Applications ChinaY. Zhou1,2,3, P. Yan1, W. Yuan1, D. Zhang11 Institute of Electrical Engineering, Chinese Academy of Sciences, Single-stage induction coilgun (SICG) mainly consists of drive coil,Beijing, China armature and discharge circuit including equivalent capacitance,2 Graduate School, Chinese Academy of Sciences, Beijing, China inductance and resistance, whose structural parameters are crucial to3 School of Automation and Electrical Engineering, Tianjin University projectile velocity and launching efficiency. In this paper the launchingof Technology and Education, Tianjin, China efficiency is defined as objective function in SICG. The mathematical model of structural optimization consists of design variables, constraintHigh voltage freewheeling diodes in capacitor-based pulsed power conditions and convergence criterion is built and computed with finitesupply are of essential importance to the reliability of the system, element method (FEM). How launching efficiency in SICG varieswhich bear great i2t and di/dt stress. The semiconductor device models resulting from changes of the design variables of drive coil, armatureavailable in most circuit simulation software packages do not and discharge circuit is analyzed. Experimental equipments which areaccurately characterize high voltage fast recovery diodes for reverse based on design variables optimized are set up and scaling tests arerecovery modes. A PSpice model for high voltage fast recovery diode carried out. Calculation and test results reveal that structuralis proposed, which is based on the device behavior rather than on the optimization of the SICG can increase the projectile velocity andphysical structure. It can describe the reverse recovery process with launching efficiency effectively.relative accuracy. The model is used to analyze circuits forexperimental railgun system. Extensive simulations have been 2P-92: Design of a Single Stage Supersonic Reluctanceperformed to examine the model characteristics. Coilgun T. S. El-Hasan2P-90:Finite Element Analysis of 3-D Eddy Field on New Electrical Engineering, Zarqa University, Zarqa, JordanType of Intercepting Missile in the ElectromagneticLaunching In this research the design of the reluctance coil gun is discussed. TheP. Sun objective from the research is to develop a theoretical modelMachinery and electronics engineering, Shijiazhuang Mechanical supplemented with FE model for a scaled down demonstrator of aEngineering College, Shijiazhuang, China single stage reluctance coilgun. The design technique is implemented on a small calibre (5.56 mm) projectile and compared to the M16 rifleIn fact, electromagnetic launching system uses the power stored in where such weapons can be easily accessed and tested. The design datapulsed capacitances as magnetic energy to act on intercepting missile for the coilgun is experimentally validated and compared to thoseand accelerate it, and the missile would be launched at high speed. obtained from the M16 test results. The scope for further research toWhen the power current feed in the driving coil, the power current develop a full scale of coilgun capable of firing such a projectile at thewould bring strong changeable pulsed magnetic field, that could bring speed of 960 m/s is still of a prime concern for further research toinductive eddy on the aluminum plate. The inductive eddy interacts replace the existing explosive/chemical propulsion weapons.with strong changeable pulsed magnetic field, so the aluminum plate The design requires optimisation, selection processes and matchingendures electromagnetic force whose direction is inverse compares techniques between all individual components and will consider thewith the driving coil. Conventionally the driving coil is fixed, then the following areas: • Electromagnetic coil design. • Projectile design,aluminum plate which is regarded as intercepting missile could be material selection and magnetic saturation • Structural design of thelaunched at high speed, however, in order to enhance the interception accelerating tube. • Energy source selection. • Power switchingeffectiveness, people hope to use high-strength magnetoconductive selection. • Triggering circuit design. • Prototype Construction. •material, such as steel. So we design a new integrative intercepting Experimental set up and tests.missile which comprises driving coil, steel layer and aluminum layer. The main contribution of this work is emphasised on the design andThe aluminum plate is fixed, the whole missile consists of three parts, optimisation technique of single stage electromagnetic coil and somewhen the power current feed in the driving coil, the intercepting missile important aspects related to the energy storage, switching devices andwould be launched to intercept coming target. Steel could strengthen control circuit. The combined Theoretical-FA model shows promisingthe interception effectiveness, and aluminum could make the results that can be applied to a full scale model of the electromagneticintercepting missile be launched. We synthesize the both of materials’ coil to achieve the required performance results compared to the M16characteristics, that could make interception effectiveness and driving rifle.coil be better. But there exists magnetoconductive material in the The aim is to design a super sonic single stage coil throughout the bestsolution area of 3-D eddy field that makes differential equation of combination of the design variables. This can be achieved bysteady magnetic field be nonlinear, the analysis process would be so maximising the electromagnetic field intensity, generated by the currenthard that the calculated results in traditional ways, such as numerical passing through the coil windings, to achieve higher muzzle velocitiesanalysis, can’t be exact. The finite element analysis is a very good way per single stage. As a prerequisite, the muzzle velocity of the projectileto solve such problems, so the eddy field’s distribution and endured is set to 350 m/s and the coil axial length is set to the calculated 105
  • 105. projectile length which is 14.00 mm. modulator as a whole and also indicate its performance for variousFor a specific applied voltage, the current drown by the coil is operating points, load conditions and changes in input and outputcalculated and used as an input to the FE model to calculate the parameters etc. The modulator in this paper is to be used for Electronelectromagnetic force exerted on the bullet. The results obtained from gun, Plasma ion implantation and medical applications.the analytical solution for the range of 10 – 58 mm of coil outer The model for the pulse modulator consists of a third order transferdiameter at different winding diameter (0.50, 1.00, 1.50 & 2.00 mm) function wherein only the load, pulse transformer, switch and inputare used in the optimization process (ANSYS APDL macros). High capacitor are included. The model also indicates the effectiveness ofmagnetic permeability and high magnetic saturation material such as charging capacitor in such a low input impedance and high outputsilicon iron boron is used and a non-linear solution is performed where impedance transformation of the pulse transformer. The entirethe BH curve for the material is defined through the FE model. modeling and simulation has been carried out using MATLAB and theExperimentally, using high voltage high current IGBTs and bank of output waveforms have been compared with experimental results (only22,000 uF capacitors with 500 volt, it was possible to achieve a muzzle for resistive loads)velocity of 210 m/s using one single stage. Higher velocities could be Keywords: Pulse Modulator, IGBT, Transfer function.achieved if some mechanical issues can be solved. Hence the need forfurther research is valid to achieve the required velocities. 4A: Radiation Sources I: Z and X-pinches and Lasers Tuesday, June 21 15:30-17:30 CC 10A-B2P-93:Analysis of a Linear Reluctance Motor for AircraftElectromagnetic LauncherM. Mirzaei1, S. E. Abdollahi2, M. Ghodsi3 (INVITED) Advanced Load Current Multiplier on 4A-1:1 Electrical Engineering, Amirkabir University of Technology, Tehran, Zebra GeneratorIran A. Chuvatin1, V. Kantsyrev2, A. Astanovitsky2, R. Presura2,2 A. Safronova2, B. LeGalloudec2, V. Nalajala2, K. Williamson2, Electrical Engineering, University of Tehran, Tehran, Iran3 I. Shrestha2, G. Osborne2, M. Weller2, V. Shlyaptseva2, M. Cuneo3, Mechanical Engineering, Tarbiat Modares University, Tehran, Iran L. Rudakov4 1 Laboratoire de Physique des Plasmas, Ecole Polytechnique, Palaiseau,In this paper, analysis and design of a linear reluctance synchronous Francemotor for aircraft launching is presented. The field and armature 2 Physics Department, University of Nevada, Reno, NV, USAwindings are in the stationary or stator part and moving part or 3 Sandia National Laboratories, Albuquerque, NM, USAsecondary is only laminated iron without windings or permanent 4 Icarus Research, Inc., Bethesda, MD, USAmagnets. This structure helps to have light moving part, which isimportant for launching and braking after the airplane take off due tothe low mass. The moving part is robust because of only using iron and An improved load current multiplier (LCM) designed and tested at aalso the temperature rise in the moving part is not a problem with iron high-voltage 100 ns UNR Zebra generator allowed to achieve 1.7 MAsecondary. 2D analytical and numerical finite element methods are in planar wire-array plasma loads and 1.9 MA in constant-inductanceused for the magnetic analysis and design optimization. The linear loads [V. Kantsyrev et al., Bull. Amer. Phys. Soc., v. 55, 244 (2010)]motor has long stator and short secondary, which stator has windings in compared to 1.4-1.6 MA accordingly with the previous LCM versionwhole path of launching. Optimization of stator and secondary is done [A. Chuvatin et al., Phys. Rev. S.T. 13, 010401 (2010)] and to 0.9-1to obtain high thrust force in minumum volume of the machine. Power MA without LCM. The total X-ray yield and power were significantlysupply design for the field windings and inverter design for the increased [V. Kantsyrev et al., High Energy Density Physics 5, 115armature windings will be presented in the full paper. Steady state and (2009); A. Safronova . et al., Bull. Amer. Phys. Soc., v. 55, 244 (2010)].dynamic analysis of motor during launching are done to calculate the These results were obtained without changing Zebra generatoraverage force and force oscillations. Force oscillations minimization is architecture or stored initial energy. The LCM design represents nowdone by changing stator and secondary teeth widths using finite an integral part of Zebra, installed inside the generator between theelement method. Finally thermal analysis of the linear motor is water-vacuum interface and the load. The generator vacuum chamber ispresented. thus the same as in no-LCM experiments, allowing the same load installation procedure and diagnostics arrangement. We present theoretical estimates and numerical calculations for scaling-up of this2P-94: Mathematical Modeling of a Solid State Pulse Power technique at TW-range facilities. This work was supported byModulator DOE/SNL grant 681371 and DOE/NNSA cooperative agreements DE-H. P. Taskar, V. C. Chinde, H. M. Mangalvedekar, N. M. Singh FC52-06NA27586, DE-FC52-06NA27588, and in part by DE-FC52-Electrical, V.J.T.I,Mumbai, Mumbai, India 06NV27616.Abstract: 4A-2: Status of the Z Pulsed Power DriverThis paper discusses the Frequency domain model of a Pulse M. E. Savage1, D. S. Artery2, P. A. Corcoran3, K. R. LeChien4,Modulator. The Pulse Modulator consists of a high voltage dc source M. R. Lopez1, J. A. Lott2, B. S. Stoltzfus1, W. A. Stygar1charging a dc capacitor which supplies power to the pulse transformer 1 Sandia National Laboratories, Albuquerque, NM, United Stateswith a solid state switch (IGBT) connected in series with the pulse 2 KTech, Albuquerque, NM, United Statestransformer. In this modeling the pulse transformer is described by the 3 L3 Pulse Sciences, San Leandro, CA, United Statesequivalent circuit consisting of the series resistor, shunt capacitor and 4 NNSA, Washington, DC, United Statesleakage inductor. An earlier paper describes the numerical validation ofpulse transformer obtained from the practical results and modeled usingthe IEEE standard model of pulse transformer. The results were The Z pulsed power driver at Sandia National Laboratories is used for aencouraging and the model of transformer was developed using second variety of high energy physics experiments. The Z system stores 20order transfer function derived from its equivalent circuit. The megajoules at the present nominal ±85 kV Marx charge voltage. The Zmotivation for this paper has been to extend this modeling in frequency system consists of 36 basically identical modules, each with a DC-domain to the entire modulator. This will help in designing the charged Marx generator, water insulated capacitor, laser triggered gas switch, and water insulated pulse-forming section. High current drivers 106
  • 106. such as Z are able to create energy densities of megajoules per cubic radiation sources in both soft and hard x-ray spectral ranges. Recentcentimeter in ~1 cm3 volumes, used for creating extremes of hardware enhancement at Zebra facility, called the Load Currenttemperature and pressure on a nanosecond time scale. The Z driver has Multiplier (LCM), allowed to increase the maximum current throughdelivered currents up to 26 MA with 85 nanosecond rise time (10%- wire array load up to 1.7 MA. This upgrade has opened new frontiers90%) into fixed inductance and imploding plasma loads. Z has allowing the implosion of heavier and bigger planar wire array loads ongenerated over 300 TW peak X-ray power, and more than 2 MJ total Zebra generator with numerous new applications ranging from theenergy, radiated from imploding plasma shells. Z is also used for Inertial Confinement Fusion to Laboratory Astrophysics. The Wirestudying compression of materials at megabar levels, using magnetic Ablation Dynamic Model (WADM) has been used for the design andpressure from current densities of tens of megamperes per cm. Z is a optimization of these new loads. Such approach allows consideration ofconfluence of pulsed energy storage and switching technologies. DC the effects of wire ablation on implosion dynamics, which is especiallycharged Marx generators are triggered with ten-nanosecond precision, important for heavy LCM loads. The WADM simulations also take intocharging water-insulated coaxial capacitors. The laser triggered gas account non-uniform return current distribution for the planar arrayswitches perform the final nanosecond time synchronization, and self- loads with the specific design of the return current cage.closing water switches further compress the pulse. Issues with the This work was supported by the DOE/NNSA Cooperative agreementssystem include improving the reliability of the components, improving DE-FC52-06NA27586 and DE-FC52-06NA27588.performance of components, and improving our conceptualunderstanding of the entire machine. Many of the Z components are 4A-5:The Impact of Load Configuration and Wire Materialunique: Z requires laser triggered gas switches operating at six millionvolts, water switches operating at three million volts, and large-area on Radiation Yield from Wire Array Plasmas at Universitysolid-vacuum interfaces operating reliably at 140 kV/cm. Scale Z-Pinch Pulsed Power GeneratorsSynchronization with fast load diagnostics requires nanosecond I. K. Shrestha1, V. L. Kantsyrev1, A. S. Safronova1, A. A. Esaulov1,predictability of the load current timing, while the pre-fire probability A. Astanovitskiy1, K. M. Williamson1, G. C. Osborne1, M. E. Weller1,of any switch must be extremely small. We will describe the state of V. V. Shlyaptseva1, N. D. Ouart2 1the machine, and specific measurements of subsystem performance. Department of Physics, University of Nevada, Reno, Reno,NV, UnitedSandia is a multiprogram laboratory operated by Sandia Corporation, a States 2Lockheed Martin Company, for the United States Department of Plasma physics division, Naval Research Laboratory, Washington DC,Energy’s National Nuclear Security Administration under contract DE- United StatesAC04-94AL85000. The configuration as well as elemental composition of Z-pinch loads on4A-3:Particle in Cell Simulations of Plasma Dynamics in the pulsed power generator is an important feature for both total x-rayZ Double-Post-Hole Convolute and Inner Magnetically radiation yield and total hard x-ray (HXR) yield. The various types of loads such as multi-planar wire arrays, compact cylindrical wire arrays,Insulated Transmission Line nested cylindrical wire arrays, and X-pinches made from Al, Fe, Cu,S. W. Vickers1, J. Chittenden21 brass, Mo, Ag, W and Au were designed and tested on the 1.7 MA Hydrodynamics, AWE, Reading, United Kingdom UNR Zebra generator (impedance 1.9 Ω, current rise time 100 ns).2 Plasma Physics, Imperial College, London, United Kingdom Loads were optimized and applied on UNR Zebra generator with or without Load Current Multiplier (LCM). The results on measurementsDynamic Z-pinch experiments on the Z-accelerator at Sandia National of total x-ray radiation yields and total HXR yields from wire array andLaboratory (SNL) exhibit current losses in the double-post-hole X-pinch plasmas that both were found to depend on configuration ofconvolute (DPHC) used to couple the four level magnetically insulated loads and wire materials, will be illustrated. The future work on thetransmission line (MITL) feed in parallel to the single inner MITL. development of new applications of aforementioned loads as well asParticle in cell (PIC) simulation results suggest that the losses are due new loads will be discussed.to a combination of electrode plasma expansion and magnetic This work was supported by the DOE/NNSA Cooperative agreementsinsulation loss in the electron sheath [1,2]. To more efficiently model DE-FC52-06NA27586, DE-FC52-06NA27588, and in part by DE-the plasma dynamics in the DPHC, a hybrid magneto-hydrodynamic FC52-06NA27616.(MHD) and PIC code is under development; PIC algorithms have beenimplemented in the Gorgon [3] 3-dimensional resistive MHD code. 4A-6: Simulations of Dynamic Laser / Plasma X-RaySimulation results of the plasma dynamics in the Z DPHC using thePIC component of the code are presented and compared with results Production*from LSP [1] and Quicksilver [2]. D. R. Welch1, C. L. Miller1, D. V. Rose1, R. B. Campbell2,[1] D. V. Rose, et al., Proc. 17th IEEE Pulsed Power Conf., B. V. Oliver2, T. J. Webb2, D. G. Flicker2 1Washington, DC, p.1153 (2009). Voss Scientific, LLC, Albuquerque, NM, United States 2[2] T. D. Pointon and D. B. Seidel, Proc. 17th IEEE Pulsed Power Sandia National Laboratories, Albuquerque, NM, United StatesConf., Washington, DC, p.1159 (2009). [3] A. Ciardi, et al., Phys.Plasmas 14, p.056501 (2007). A high fluence X-ray diagnostic is being developed by Sandia National Laboratories for use as a point-projection X-ray source for imaging4A-4: Design and Optimization of Planar Wire Array Loads high-density load dynamics on the ZR pulsed power facility. The LSP 3D PIC code was used to model the dynamic interaction, between thefor 1.7 MA LCM Current Regimes at Zebra Facility Z-petawatt IR laser and a high-atomic-number thin foil, whichA. A. Esaulov, V. L. Kantsyrev, A. S. Safronova, I. K. Shrestha, generates the intense short-duration bremsstrahlung X-ray pulse.K. M. Williamson, G. C. Osborne, M. E. Weller, V. V. Shlyaptseva Particle kinematics are explicit for densities near and below critical andDepartment of Physics, University of Nevada Reno, Reno, NV, United implicit for higher densities through solid. Sufficiently small cell-sizesStates and timesteps were used to fully resolve the laser-plasma interaction and subsequent electron dynamics in the foil -- 15 millions cells andDuring the past six years single- and multi-planar wire array loads have 400 million particles were required. An optional treatment of high-been successfully tested at 1 MA currents at Zebra facility as powerful energy electrons using the continuous slowing down approximation of 107
  • 107. energy loss with the Li and Petrasso model [Phys. Rev. Lett. 70, 3059 4A-8:Extreme State of Water Produced by Converging(1993)] was also examined. The particle-field solution used matrix Strong Shock Waves Generated Using Underwaterinversion techniques as opposed to a conventional ADI scheme. The Electrical Wire Array Explosioneffect of parameters such as foil radius, tilt angle, foil support structure, Y. E. Krasik, A. Fedotov-Gefan, S. Efimov, L. Gilburd, V. Gurvich,and net laser energy on the X-ray fluence and energy spectrum were G. Bazalitskyevaluated in both cylindrical and 3D Cartesian coordinates for all angle Physics Department, Technion, Haifa, Israelleaving the target. The simulated X-ray fluence produced a peak X-raydose in excess of 200 mRad, consistent with experimentalmeasurements. Results of generation of extreme state of water with pressure up to 4* This work is supported by the Department of Energy through Sandia Mbar, density up to 4.3 g/cm(^-3) and temperature up to 2.2 eV in theNational Laboratories. Sandia National Laboratories is a multi-program vicinity of the implosion axis of converging strong shock wave arelaboratory operated by Sandia Corporation, a wholly owned subsidiary reported. The shock wave was produced by underwater electricalof Lockheed Martin Company, for the U.S. Department of Energys explosion of cylindrical Cu wire array. A 10 kJ pulse generator with aNational Nuclear Security Administration under contract DE-AC04- current amplitude of 500 kA and rise time of 350 ns was used to94AL85000. explode arrays with varying lengths, radii and number of wires. Hydrodynamic numerical simulations coupled with SESAME equation of state for water and the experimental data of the shock wave4A-7: Research of Non-Cylindrical Wire Arrays on propagation, energy deposition rate to the array and light emission fromANGARA-5-1 Facility the compressed water in the vicinity of the implosion axis were used toE. V. Grabovski1, V. P. Smirnov1, V. Aleksandrov1, M. Fedulov1, obtain the pressure, density and temperature profiles during theI. Frolov1, A. Gribov1, A. Gritsouk1, Y. Laukhin1, S. Medovshikov1, implosion. Dependences of the pressure in the vicinity of the implosionK. Mitrofanov1, G. Oleinik1, A. Samokhin1, G. Volkov1, V. Zaitsev1, axes on the array radius and deposited linear energy density per unitP. Sasorov2, V. Gasilov3, S. Dyachenko3, O. Olkhovskaya3, length were obtained.A. Shevelko4, O. Yakushev41 SRC RF TRINITI, Troitsk, Russian Federation2 4B: Applications I: Fusion, EM, Beam, Laser and Space ITEP, Moscow, Russian Federation3 Applications IMM RAS, Moscow, Russian Federation4 FIAN, Moscow, Russian Federation Tuesday, June 21 15:30-17:30 CC 10C-DInvestigations into the process of wire array compression which, 4B-1:New Self-Magnetically Insulated Connection of Multi-nowadays, generate the greatest energy of a X-radiation pulse, are Level Accelerators to a Common Load for Fusionaimed at target igniting for thermonuclear fusion reaction. For next J. P. VanDevender1, W. L. Langston2, M. F. Pasik2, R. S. Coats2,facilities designing we make experiments for specify codes describing T. D. Pointon2, D. B. Seidel2, G. R. McKee2, L. X. Schneider2arrays implosion and x-ray production. For check of a hydrodynamic 1 VanDevender Enterprises, Albuquerque, United Statespart of codes we used conic wire liners with weight 200-400 mkg, the 2 Sandia National Labs, Albuquerque, United Statesbig radius of a cone 12-20 mm and small radius 5-10 mm. These linersmake less powerful x-ray pulse than cylindrical, but have a complicatedtime profile of the pulse and allow us to specify codes without exact Self-magnetically insulated convolutes channel the currents from multi-assumptions about initial array heterogeneity For a cone angle of > 20° module, pulsed-power accelerators with multi-level vacuum stacks to athe X-pulse has three peaks. Comparison of their relative magnitude single load. Multi-level magnetically insulated transmission linesand duration obtained by experiment and in calculations allows us to (MITLs) followed by a post-hole convolute provide an efficientmore reliably verify some assumptions introduced into the code than solution for low impedance loads. For higher impedance loads requiredwould be the case of cylindrical liners. The 2D RHDM code MARPLE for fusion, systems with post-hole convolutes suffer losses that are still(IMM RAS) is used. The work demonstrates the results of comparison being investigated. We present a set of conservative designbetween the theoretical predictions and experimental evidence. For requirements for such a convolute and use them to design a new type oftesting of codes describing a spectrum of Z-pinch radiation we register convolute called the Clam Shell MITL (CSMITL) that has magneticthe spectrum produced by arrays from various substances – W, Al, nulls only at large radius where the cathode electric field is kept belowplastic. Along with conical arrays the technology of making quasi- the threshold for emission, has only a simply connected magneticspherical wire arrays has been developed by us. The work [1] shows topology to avoid plasma motion along magnetic field lines into highlythat the implosion of quasi-spherical liners in the scheme of double stressed gaps, and has electron injectors that ensure efficient electronhohlraum makes it possible to increase the power flux to target by a flow even in the limiting case of self-limited MITLs. Results from thefactor of 3-4. However no experiments on implosion of spherical or topologically similar MITE experiment and from fully electromagneticquasi-spherical arrays at the facilities of terawatt power level have been three dimensional simulations are used to test and validate the designconducted earlier. One of the problems is the complexity in creating a codes. A reference design for the Z accelerator is derived and presented.quasi-spherical liner with necessary distribution of density. At the The design conservatively meets the design requirements and showsAngara-5-1 facility due to the electrostatic stretching of the cylindrical excellent transport efficiency in three simulations of increasingwire array occurred by an additional electrode the quasi-spherical complexity: electromagnetic fields only with Emphasis and Quicksilver,arrays of 30-60 tungsten wires 6μm thick are formed with a radius of 8- fields plus electron emission with Quicksilver, fields plus electron and12 mm and a mass of 200-400 μg. The implosion of z-pinches formed ion emission whenever the anode is heated above 400 degC withwith such liners was effected by 3-4 MA current with a rise front of Quicksilver.100ns at the Angara-5-1 facility. It was shown that the central part ofsuch a liner forms a hotter area compared with its edges and a hotterone than in the case of cylindrical liners.[1] V.P. Smirnov, S.V. Zakharov, and E.V. Grabovskii, Pis’ma Zh.Eksp. Teor. Fiz. 81, 556 (2005) [JETP Lett. 81, 442 (2005) 108
  • 108. 4B-2:Inverse Diode for Combination of Multiple Modules compression, can be characterized by the pressure of tens of megabars.and Fusion Driver-Target Standoff In such conditions thermonuclear reaction ignition in plasma isJ. P. VanDevender1, D. B. Seidel2, K. A. Mikkelson2, R. D. Thomas2, achievable. The principle feasibility of this approach is based on theV. J. Harper-Slaboszewicz2, B. P. Peyton3 fact that at present All Russian Scientific Research Institute of1 VanDevender Enterprises, Albuquerque, NM, United States Experimental Physics (VNIIEF) possesses the method of producing2 Sandia Natioanl Labs, Albuquerque, NM, United States magnetized heated plasma (the special MAGO plasma chamber) and3 KTech Corp, Albuquerque, NM, United States powerful explosive magnetic generators (EMGs), which provide liner acceleration up to the energies of the order of 20 MJ. The progress of the MAGO system must rely on the research results in the field of highA newly invented, very-high-current Inverse Diode converts the energy density physics. These are the data on the equations of state ofcurrents in many electron beams to current in a single Magnetically various materials (including non-ideal plasma and warm dense matter),Insulated Transmission Line (MITL) for driving a common load. the coefficients determining the radiation-matter interaction (includingElectrons are injected through a thin anode, cross a vacuum gap, and the opacities of dense plasma with multi-charged ions), transportare absorbed in the cathode of the Inverse Diode. The cathode current coefficients (including electrical conductivity of non-ideal plasma),returns to the anode through a load and generates an electric and data on shock waves and their structure (including collisionless shockmagnetic field in the anode-cathode gap. Counter streaming electron waves), data on the interaction of ultrahigh (megagauss) magneticflow is prevented by self-magnetic insulation in most of the Inverse fields with matter, and surface discharges in such fields, the researchDiode and by self-electrostatic insulation where the magnetic field is results on different hydrodynamic and magneto-hydrodynamicinsufficient. Two-dimensional simulations with a 40 MA, 4 MeV, 40 ns instabilities, etc. On the other hand the investigations in the frames ofelectron beam at 3.5 kA/cm2 current density, 5 degree beam divergence, the MAGO system contribute to all these fields of physics and provideand 45 degree injection angle show >85% of the injected electron beam new research methods (both theoretical and experimental, such as thecurrent is captured and fed into the MITL. The electric field slows the use of pulsed power facilities, including EMG), which have alreadyinjected electrons and the magnetic field bends their trajectories as they provided numerous fundamental results and have vast prospects.cross the anode-cathode gap. Therefore, the optimum anode-cathodegap is ~1.4 times the local Larmor radius of the injected electrons. Thedevice can in principle have ~100% current efficiency and ~50 % 4B-5:Simulation of Diocotron Instability in a Magneticallyvoltage and energy efficiency. We are exploring this device for Confined Hollow Electron Beamcombining pulsed power modules and providing standoff between an H. J. Lee1, M. Chung2, S. I. Chung1 1accelerator and a target in a pulsed power driven fusion power plant Department of Electrical Engineering, Pusan National University,called the Compact Plasma Power Station. Busan, South Korea 2 Handong University, Pohang, South Korea4B-3:The Development of a 2.4 MJ Pulsed Power Supply forthe Production of a Compact Toroid Plasma Recently, there is a renewed interest in the diocotron instability for theW. Zawalski, M. Laberge, S. Howard, M. Wight, K. Epp, A. Wong development of a new type of beam collimator system in high-energyGeneral Fusion Inc., Vancouver, Canada colliders such as the Tevatron or the large hadron collider (LHC). The proposed collimator system makes use of the magnetically confined electron beams with a hollow current density profile. When such aThis paper describes the development of a split 2.4 MJ pulsed power hollow electron beam is overlapped with a proton beam, for example,supply to drive current in the formation and acceleration sections of a the proton beam core passes through the central region of the annularCompact Toroid Injector (CTI) for the production of a Magnetized electron layer without perturbations, whereas the halo particles aroundTarget Fusion (MTF) plasma. The injectors formation current is driven the beam core experience nonlinear transverse kicks. By activelyby a 600 kJ 22 kV capacitor bank consisting of 24 individual switched controlling these kicks, one can improve the collimation performanceparallel capacitor modules, while the acceleration and compression and reduce the uncontrolled beam losses. One of the disadvantages ofsection is powered by a 1.8 MJ 44 kV parallel array of 36 capacitor this collimation concept is indeed related with the diocotron instability.modules, each consisting of two 22 kV sub modules arranged in a Marx Linear analyses show that the electrostatic perturbations for an annularconfiguration. Switching for both the single stage formation and two electron layer can grow due to the free energy associated with the shearstage Marx acceleration modules is accomplished using thyratrons in the flow velocity. As a result, the beam cross section of the hollow(psuedospark switches) manufactured by Pulsed Technologies Ltd. electron beam is expected to distort as it propagates, which has alreadyFormation and acceleration current amplitudes are controlled by been observed in many standard nonneutral plasma experiments withadjusting voltage and the number of active modules with additional remarkable agreement with theories. The cross-sectional view of two-control of the acceleration current profile achieved by varying the dimensional particle-in-cell simulation in a cylindrical geometryswitching time of individual acceleration modules. The system indicates that the electron column is eventually converted into a finiteconfiguration including charging, control and safety features are number of vortices. In this study, we report vortex formation of adescribed. The optimization, reliability and performance of the hollow electron columns confined in a Penning trap and the way how tothyratron switches are also discussed. control vortex formation and merging.4B-4: High Energy Densities and MAGO/MTF 4B-6: Substrate Conditions in Low-Temperature GalliumS. F. Garanin Arsenide and Semi-Insulating Gallium Arsenide DuringITMF, All-Rusiian Research Institute of Experimental Physics Terahertz Pulse Generation(VNIIEF), Sarov, Nizhny Novgorod region, Russian Federation S. Ray1, A. K. Alla1, J. Yakura2, P. Kirawanich3, N. E. Islam1 1 ECE, University of Missouri, Columbia, MO, United StatesThe MAGO/MTF approach (MAGO is Russian abbreviation for 2 ECE, Air Force Research Laboratory, Albuquerque, NM, UnitedMAGnetic Compression and MTF is Magnetized Target Fusion) Statesimplies preliminary heating of DT plasma up to the temperatures of 3 EE, Mahidol University, Salaya, Nakhom Pathom, Thailand0.2-0.4 KeV and subsequent isentropic plasma compression with a liner.The energy density level, which is required to achieve at such 109
  • 109. THz radiation have applications in many fields, specifically in biology possible to operate several pulsers synchronously into one or morebecause of their ability to penetrate deep into organic materials while loads.avoiding harmful ionizing radiation effects like X-Rays. Of the many Pulse generators with amplitude of 5-10 kV, rectangular pulse shapemethods used in the generation of THz radiation, the most prominent with duration of 5-10 ns and repetition rate in burst mode of 5 MHzand well known method is the use of Photoconductive Semiconductor have been developed as well.switches (PCSS). Here the movement of carriers generated in the Another new development is a pulser with output amplitude of 200-300substrate through laser action, results in THz radiation. Both semi- V, pulse duration of about 1 ns and PRF in burst of up to 500 MHz.insulating (SI) and Low-Temperature (LT) grown Gallium Arsenide All of these pulsers are fully solid state and have high efficiency and(GaAs) have been used as the substrate material for the PCSS. Even compact size.though the THz radiation is being used in many applications, not muchis known about the substrate conditions that lead to the generation of 4B-8:Meridian Radar System - Klystron Based Transmitterthis particular type of radiation. It is also of interest to understand whysemiconductor materials such as SI-GaAs and LT-GaAs contribute to for Atmospheric ResearchTHz generation while other semiconductor materials do not. P. Kolda1, S. Lenci1, Z. H. Ding2, M. Kempkes3, K. Schrock3 1In our analysis we have compared the substrate conditions and the Commnications and Power Industries, Palo Alto, CA, USA 2characteristics of the two PCSS materials mentioned earlier that is used China Research Institute of Radiowave Propagation, Qingdao, China 3in the generation of THz. We have specifically looked into the role of Diversified Technologies, Inc., Bedford, MA, USAtraps, its density and concentration on collection rate in the substratethat influences THz generation. We have reviewed effects of carrier Communications and Power Industries, Inc. (CPI) has collaboratedrise time on beam width and decay. Analysis was done using the with Diversified Technologies, Inc. (DTI), and the China Researchindustry standard simulation suite that has built-in interface for user Institute of Radiowave Propagation (CRIRP) to produce a transmitterdefined semiconductor parameters such as trap characteristics, mobility operating at a frequency of 500 megahertz and producing 2 megawattsetc on carrier transport. Our analysis shows that with the application of power output for the Meridian incoherent scattering radar. This paperhigher bias, there is an increase in the total current density for both the describes the radar system components and operation. The klystron, theLT-GaAs and SI-GaAs PCSS. The increase in collection can be transmitter and the capabilities enabled by the system for studying theexplained due the increase in the drift velocity at higher Electric fields. density and velocity of the ionized gases present in the ionosphere areHowever, the initial collection profile differs in devices studied. Results discussed.show that in SI-GaAs, because of a relatively higher trap concentration The Meridian Space Weather Monitoring Project (Meridian Project) ofat midgap, the carrier recombination rate is also higher as compared to China will deploy a continent-scale ground-based array of geomagneticthe LT-GaAs. The I-V curves for both the devices show expected field, radio, and optical instrumentation along the 120 East longitudeslopes for trap filled semi-insulating materials, having a linear followed and 30 North latitude to monitor the space environment. Thisby sub and super linear regions [1]. However, the linearity of LT-GaAs environment includes the middle and low atmosphere, ionosphere, andis more prominent than that of SI-GaAs. For the same bias, collection is magnetosphere. The Center of Space Science and Applied Researchfaster in LT-GaAs based PCSS as compared to SI-GaAs. The response (CSSAR), China, is leading the effort. The space environmentof the structures also depends on the contact shapes [2]-[3]. monitoring system of the Meridian project is comprised of UHF incoherent scatter radar (ISR), HF/VHF back- scattering radar, digital[1] Murray A. Lampert and Peter Mark, “Current Injection in Solids.” portable ionosonde (DPS),meteor radar, MF radar, Lidar, airglow meter,Academic Press, New York, 1970. magnetometer etc. The ISR is the most powerful ground-based probing[2] Phumin Kirawanich, S. Joe Yakura, Carl E. Baum, and Naz E. tool for the ionosphere. It is being built in the city of Qujing, YunnanIslam,” Analyzing Enhanced Terahertz-Pulse Power and Frequency province, China, which is uniquely located in the region of a lowUsing a Field-Carrier Transport Approach”, IEEE Antennas and latitude ionosphere anomaly. The ISR can directly measure manyWireless Propagation Letters, Vol. 7, 2008 ionospheric parameters including the electron density, electron[3] Phumin Kirawanich, Susumu J. Yakura, Member, and Naz E. Islam, temperature, ion temperature, plasma drift velocity etc, in the range of“Study of High-Power Wideband Terahertz-Pulse Generation Using about 1000 kilometers. Due to weakness of the scattered signal fromIntegrated High-Speed Photoconductive Semiconductor Switches”, the ionosphere, the ISR must utilize a high power transmitter, high gainIEEE Transactions On Plasma Science, Vol. 37, No. 1, January 2009 antenna and high sensitivity receiver. The Meridian project ISR is scheduled to be completed in the summer of 2011. CPI produced the VKP-8208A klystron to supply the over 2 megawatts4B-7:Compact Nanosecond FID Pulse Generators for of peak power at a frequency of 500 megahertz and a maximum dutyAccelerators, Laser and Discharge Applications factor of 0.05 percent. The beam voltage required to operate theV. Efanov, M. Efanov, A. Kriklenko, A. Komashko, P. Yarin, klystron is 97 kilovolts and the beam current is 40 amperes. TheS. Zazoulin klystron design incorporates an electro-magnet for the focusing of theFID GmbH, Burbach, Germany electron beam. DTI manufactured the transmitter power supply and modulator utilizingNanosecond pulse generators based on solid state fast ionization integrated Insulated Gate Bipolar Transistors (IGBT) architecture ofdevices (FID) switches for use in accelerators, lasers and gas discharge their PowerMod™ design. DTIs PowerMod solid-state modulators andapplications. Generators output rectangular pulses with amplitude of 50 power supplies are built from a series stack of IGBTs, configured forkV and pulse duration of 50 ns at 90% of peak amplitude. Rise time can very high voltage standoff, and operated as a single ideal switch. Gatebe less than 10 ns. Maximum pulse repetition rate is 100 Hz. These drives are highly synchronized to ensure the load on the switch ispulse generators are designed to operate into 50 Ohm load. Life time shared equally between devices. The entire switch can be quickly andtesting has been performed for 20 days of continuous operation. Tests arbitrarily closed or opened, providing sub-microsecond removal ofhave revealed high stability and reliability of operation. stored energy from the load during a short. DTIs pulse modulatorsPulse generators with amplitude of 10-20 kV into 50-100 Ohm and have onboard gate drives, diagnostics, and fault detection. Each switchpulse duration of 1-10 ns have been developed. Maximum pulse operates fully isolated from ground, and can open without damagerepletion rate in continuous operation is 1 MHz. Such pulsers show under fault currents to 10 kiloamperes. The patented technology ishigh stability of amplitude and delay relative to triggering pulses. It is scalable, enabling a modular approach to building switches from 2.5 110
  • 110. kilovolts to over 200 kilovolts, at currents from 10 amperes to 5000 4, August 2009 Page(s):948 - 955amperes. [2] A Novel Design for a Multistage Corona Stabilised Closing Switch Given M.J., Timoshkin I.V., Wilson M.P., MacGregor S.J., Lehr J.M.;4C: Components III: Arc Discharge Switching Proceedings of the 2010 IEEE International Power Modulators and High Voltage Conference, 23 - 27May 2010Tuesday, June 21 15:30-17:30 CC 11A-B 4C-3:Triggered Vacuum Switch and Air Spark Gap for4C-1: Simulation of a 1.0-MA Current Joint with ALE3D Pulsed Power ApplicationsJ. (. B. Javedani, J. M. Solberg, R. N. Rieben, D. A. White, D. A. Goerz X. Duan, M. Liao, J. Zou, C. Zhao, Z. ZhouEngineering, Lawrence Livermore National Laboratory, Livermore, School of Electrical Engineering, Dalian University of Technology,CA, United States Dalian, ChinaThe LLNL Arbitrary Lagrangian-Eulerian code ALE3D is a parallel, The closing switch is the main device to control the power supply of an3D, unstructured-grid finite element code that enables simulations of ElectroMagnetic Launcher (EML) system. Three types of closingstructural mechanics, hyrdodynamics, energetic materials, heat transfer switch which included a triggered vacuum switch (TVS) and two airand magnetohydrodynamics (MHD). The MHD module was added to spark gaps (ASGs) named ASG -1 and ASG -2 with differentALE3D by White-Rieben-Wallin in 2006 to make this code viable in configuration in trigger unit were described for investigating the delaydesign of coupled electro-thermal-mechanical (ETM) systems such as characteristics, high current capacity and erosion characteristics. A LCmagntic flux compression generators, induction heating and resonant circuit was set up and connected with the TVS’s and ASG’selectromagnetic launchers (railguns) [1]. Most recently we have used main electrodes. A design circuit for controlling the TVS and ASGsALE3D to simulate the dynamic effects of high current conductors as was proposed. The trigger energy of the controller can be adjustedthey fail under excessive conditions. A 0.25 MJ, 1 MA capacitor bank conveniently. By means of changing the trigger energy, thewas used to investigate performance of current joints under high experimental data show that the delay time and jitter time decrease withmagnetic pressure in tens of microsecond time-scales. In this paper, we increasing the trigger energy. The delay time and jitter time decreasesfirst give an overview of the ALE3D code, followed by a brief magnificently with increasing the voltage in main gap. The TVS hasdescription of configurational set up of the current-joint test-stand. We better performance in delay characteristics, higher current capacity andalso report on code’s prediction of current and joint deformation rate less erosion than ASGs. Because of the improvement in the cathodewith the experiment for a selected test. Possible suggestions on how to material, ASG -2 has better performance in delay characteristics, higherbetter match the experimental results with ALE3D will be given. current capacity and less erosion than ASG -1.Key words: ALE3D, Computational EM, Magnetohydrodynamics,Current joint1. D.A. White, et. Al. “Coupling Magnetic Fields and ALE Arc Motion Simulation in a Rotating Arc Gap Switch 4C-4:Hydrodynamics for 3D Simulations of MFCGS”, International IEEE Based on a Chain ModelConference on Megagauss magnetic field generation and related topics, H. Junjia, G. Rui, P. Yuan2006. College of Electrical & Electronic Engineering, Huazhong Univ. of Sci.* This work performed under the auspices of the U.S. Department of & Tech., Wuhan, ChinaEnergy by Lawrence Livermore National Laboratory under ContractDE-AC52-07NA27344. A chain model is employed to describe the arc motion in a rotating arc gap switch. The arc is assumed to be a chain consisted in a series of4C-2:The Triggered Behaviour of a Controlled Corona small cylindrical current elements like a string of beads. Each elementStabilised Cascade Switch is assumed to be linked by a certain rules. The element movesM. J. Given1, M. P. Wilson1, I. V. Timoshkin1, T. Wang1, individually which determines the behavior of the whole arc. The chainS. J. MacGregor1, J. M. Lehr2 model is improvement by considering the arc root motion and surface1 University of Strathclyde, Glasgow, United Kingdom drag force. The arc velocity and shape variation during the movement2 Sandia National Laboratories, Albuquerque, USA are simulated based on this model. The simulated results are compared with the experimental results and a well fitted comparison is achieved. From the simulated results, it implies that the surface drag force appliedCorona stabilised switches have been shown to have advantages in upon the arc root is an important factor that influences the arc velocitypulse power switching applications due to their high repetition rates and shape. The arc column diameter is calculated and the relationshipand low jitter. Work performed in recent years by the High Voltage between it and axial magnetic flux density is obtained.Technologies Group has shown that the operating voltage range of suchswitches can be extended by using a multi-gap cascade configuration.One particular multi-gap topology was shown to operate at 100 kV dc 4C-5:(INVITED) High Pressure Sealed Hydrogen Sparkwith a switching jitter of the order of 2ns [1]. It has since been shown Switchesthat by modifying the topology of the corona sources on the electrodes, S. D. Rendall, Z. Shotts, F. Rose, Z. Robertsit is possible to control the grading of the voltage distribution across the Radiance Technologies, Huntsville, AL, United Statesgaps in the cascade [2]. The voltages across each gap and the self-breakbehaviour of the cascade were found to be in close agreement with thevalues predicted from the corona emission characteristics for the tested High Pressure hydrogen spark switches offer excellent high voltageelectrode topologies. This paper reports on a further examination of the capabilities at high repetition rates. To date their availability has beenbehaviour of the corona controlled switching topology, where triggered limited due to the problems of hydrogen diffusion or leakage throughoperation of the switch has been investigated for different voltage the switch structure. In our laboratory, we are researching a series ofdistributions across the cascade gaps. miniature sealed hydrogen switches for a variety of applications. In this[1] A corona-stabilised plasma closing switch. Beveridge, J.R.; work, we will present data for prototype switches in terms of DC hold-Macgregor, S.J.; Given, M.J.; Timoshkin, I.V.; Lehr, J.M.; IEEE off voltage, hold off voltage as a function of dV/dt, and consistency ofTransactions on Dielectrics and Electrical Insulation Volume 16, Issue 111
  • 111. these data as a function of life of the switch. To date, prototype gap exceeds 3 for N2 and 2.5 for SF6-N2. However, the saturationswitches in the range of 20-30 kV DC are functioning perfectly one phenomena of line density of channels in SF6-Ar are not obvious inyear after fabrication. Limited data will also be presented on repetition these experiments. The line density of channels in the higher pressurerate in burst mode operation. Materials selection parameters will be gas is larger than that in the lower pressure gas for each kind of gases.discussed in terms of compatibility and ease of manufacture. PL3: Marx Award Winner4C-6:Stabilities of Water Switches with Three Different Wednesday, June 22 08:00-08:50 CC 12A-DField DistributionsP. Cong, G. Zhang, L. Sheng Practical Circuit Models and Simulations usingNorthwest Intitue of Nuclear Technology, Xian, China Transmission Lines P. A. CorcoranThe stabilities of self breakdown water switches of three different L-3 Communications - Pulse Sciences, San Leandro, CA, United Stateselectrode structures with typical field distributions tested onQiangguang-I accelerator are described. Switch A with semi-spherical Computer based circuit simulations of pulsed power systems are nowcathode has a quasi-uniform field of switch gap; Based on switch A, the standard practice. Models and simulations are used extensively forfield of Switch B having a small semi-ball like part pointed out on the design, predicting the performance, and understanding the behavior ofcathode top is non-uniformed; Switch C with a cylindrical tube as pulse power circuits. These tools were not available in the early days ofcathode has diverse field. All three switches with field being enhanced pulse power when even large machines were designed without anyat cathodes have the same ground plate as anode electrode. The closing simulations, but were gradually introduced as the global informationtime jitter for switches A, B and C is 140ns, 60ns and 30ns respectively technology revolution unfolded.when the non-uniform factor defined as the ratio of cathode peak field Sparked by the difficulty and inefficiency of performing circuitdivided by average field of gap is accordingly 1.7, 3.6 and 6.2, which simulations on remote mainframe computers and by the advent ofshow that the switch stability improves with the field uniformity going personal computers in the mid-1980s, various groups independentlyworse. The minimum relative standard deviation of load current is developed circuit modeling codes that could run quickly on theacquired as 3.9% with switch C as transfer device on accelerator. The relatively limited local platforms of the time. One such effort was atdischarge was kept for at least 250ns from streamer initiating to Pulse Sciences, Inc. (PSI) where a code was developed that used onlybreakdown point by optic monitoring. By comparison of formulas transmission line and resistor circuit elements. This approach was quiteestimations with gap lengths practically employed by different switches different from alternatives that included lumped elements of puretested on facility, two existing empirical expressions were respectively capacitance and inductance and therefore that required solutions toproper for predicting gaps of switch A and C through known effective systems of differential equations. Transmission line based models andarea of cathode, voltage amplitude and effective time of pulse with the simulations have proven to be a natural, accurate, and practical way toexception of switch B. represent pulse power circuits and have become the accepted standardKey words: water switches; breakdown; jitter; accelerator; field within many organizations. Transmission line (TL) based codes have many features that make them Study on the Saturation Phenomena of Discharge4C-7: well suited for pulse power applications. TL codes solve quickly with aChannel Number of a Gas Spark Switch Gap under constant time step because their purely algebraic junction equations are numerically stable. Their simple algorithm enabled relatively smallNanosecond Trigger Pulses scale code development within the pulse power community whereJ. Chang, H. Wang, Q. Zhang, A. Qiu access to the source code continues to enable the ad hoc addition ofSchool of electrical engineering, Xian Jiaotong University, Xian, models that are tailored to pulse power devices like spark gaps, MITLs,China and diodes. Circuit topology is not restricted. TL based codes have the ability to model electro-magnetic propagation in 1-D, 2-D and 3-DMultichannel discharge in gas spark switches is an effective approach geometries with the same fundamental transmission line elements,for the reduction of switch inductance, switch resistance and electrode thereby facilitating the creation of seamless circuits containing separateerosion, as well as the increase of switch lifetime and stability. Higher regions with different dimensions.trigger pulse or larger pulse rise rate (i.e. dV/dt) is helpful for the Transmission line models have been successfully used to design a largeincrease of discharge channel number. In this paper, experimental work number of pulse power machines. These models have been furtherwas carried out to study the variation of trigger pulse amplitude on developed and validated by comparing simulated waveforms to actualchannel numbers of discharge under nanosecond trigger pulses in a machine performance. Examples are presented and described in thisself-designed coaxial field distortion gas spark switch gap in N2, SF6- talk that represent the variety of transmission line models that haveN2 or SF6-Ar gas mixtures, whose pressures varied from 0.1 to 0.2 been developed for pulse power circuits.MPa. The SF6 content varied from 70% to 90% by volume in SF6-Armixtures, and varied from 10% to 80% by volume in SF6-N2 mixtures. 5A: Accelerators and Beams II: High Energy Accelerators,The results show that channel numbers in a unit of length of the edge of Particle Beams and Free Electron Laserstrigger electrode make a great difference among three kinds of gases(i.e. N2, SF6-N2, and SF6-Ar) under a certain pressure. In each kind of Wednesday, June 22 09:30-12:00 CC 10A-Bgases or gas mixtures (N2, SF6-N2 or SF6-Ar) under a certain pressure,the difference of line density of channels among different mixture 5A-1:Extension of the Operating Point of the Mercury IVAratios is smaller compared with that among different kinds of gases from 6 to 8 MV*under the same overvoltage multiple. In view of this, one fitting curve R. J. Allen, R. J. Commisso, G. Cooperstein, P. F. Ottinger,was got based on data of line density of channels varying with J. W. Schumerovervoltage multiple for each kind of gases or gas mixtures despite of Naval Research Laboratory, Washington, DC, United Statesthe difference among mixture ratios. It can be seen from these fittingcurves that the line density of channels does not increase evidently andtends to saturation when applied overvoltage multiple across the switch 112
  • 112. The output voltage of the NRL Mercury generator has been its trigger, and a first-article PFL driving both a dummy load and asuccessfully increased to 8 MV. Mercury was originally designed for 6 first-article cell. The status of the in-progress fabrication and QAMV, 300 kA operation in negative polarity [1]. However, new testing for each of the major subsystems is described in this paper.experiments required a bremsstrahlung x-ray source with a higher Initial electrical tests of the Marx system are described in a companionendpoint voltage [2],[3]. The threshold energy for photofission in paper at this conference. The complete system will not be assembledtargets of interest is about 5.5 MeV and, given the roughly Gaussian and tested in the US. All components of the IVA are to be delivered asshape of the Mercury output pulse, very few x rays above threshold subassemblies to AWE in the UK in mid 2012, for assembly andcould be generated by Mercury with a 6-MV peak output voltage. Also, commissioning in a new Technology Development Centre.the cross section for fission rises sharply above the threshold energy. Itwas therefore decided to increase the output voltage of Mercury to the 5A-3:Design and Performance of the Darht Second Axismaximum possible without imposing undo cost or risk. A new centerconductor was designed for Mercury that increased the effective output Acceleratorimpedance of the MITL from about 23 to 40 Ohms, thereby increasing K. Nielsenthe maximum output voltage to 8 MV. The new center conductor has Los Alamos National Lab, Los Alamos, NM, United Statescarefully designed tapers (steps down in diameter) at each stage of theadder so that the adder cell voltages are equally balanced at the design The Dual-Axis Radiographic Hydrodynamic Test Facility (DARHT) atlimit of 1.3 MV. One limiting factor is the breakdown probability of the Los Alamos National Laboratory consists of two linear inductioninsulator stack in the adder cells. However, simulations have shown accelerators at right angles to each other. The First Axis, operatingthat the new center conductor only increases the breakdown probability since 1999, produces a nominal 20-MeV, 2 kA single pulse with 60 nsa small amount. The 40 Ohm output impedance was selected because it width. In contrast, the DARHT Second Axis produces up to four pulsesallowed us to reuse several pieces of the existing center conductor. in a variable pulse format by slicing micro-pulses out of a longer pulseAluminum was used for most of the new parts. These steps allowed us (~1.6 microseconds flattop) of nominal beam energy and current of 17to save time keep material costs to a minimum. PIC and circuit MeV and 1.8 kA respectively. Bremsstrahlung x-rays are produced bysimulations were performed to validate the design. Mercury has now focusing the electron beam-pulses onto a high-Z target. The DARHTbeen successfully operated for over 100 shots at 8 MV. Facility achieved full operational status in FY2010 with the successful[1] R.J. Allen, et al., “Initialization and operation of Mercury, a 6-MV execution of five dual-axis experiments, including an explosivelyMIVA,” 15th International Pulsed Power Conference, June 13-17, driven flyer plate and four integrated weapon hydrotests. This paper(2005). focuses on the second axis accelerator and its components. The injector[2] D.D. Hinshelwood, et al., this conference. consists of an 88 stage Marx delivering 2.5 MV to a 6.25 inch diameter[3] D.M. Mosher, et al., this conference thermionic cathode. A crowbar switch in the Marx tank can shorten the* Work supported by the Office of Naval Research 2 µs pulse to as little as 200 ns. The beam is accelerated in a Linear Induction Accelerator (LIA) consisting of 74 cells operating at a total5A-2: Status of the AWE Hydrus Fabrication of 14.5 MV, delivering a beam of ~17 MeV and ~1.6 µs flattop at theK. J. Thomas1, P. F. Beech1, S. G. Clough1, I. Crotch1, S. Brown1, accelerator exit. Each cell is driven by an individual Pulse FormingB. Stringer1, C. Goes1, A. King1, I. Huckle1, J. Burscough1, Network (PFN) tuned to provide voltage regulation of +1%. The designS. Trenaman1, J. Duffy1, R. Whealdon1, W. Glazebrook2, V. Carboni2, and performance of the injector, accelerator cells and PFNs areT. DaSilva2, J. Pearce2, W. Saunders2 described herein.1 Hydrodynamics/DMP, AWE, Reading, Berks UK, United Kingdom2 Pulse Sciences, L3 Communications, San Leandro, CA, United States 5A-4: FXR Marx Triggering and Switch Characterization L. K. Tully, M. M. Ong, J. M. Zentler, B. R. Kreitzer, J. E. DunlapThe ten-module Hydrus Induction Voltage Adder, designed by L-3 Lawrence Livermore National Laboratory, Livermore, United StatesPulse Sciences for AWE, builds on previous IVA experience in the US.Each of the ten modules comprises a 1.4 MV induction cell driven by a Characterization testing of the FXR Marx generator switches waslaser triggered gas switched Pulse Forming Line (PFL) in order to initiated in early January 2010. The FXR Marx is a five-stage, dual-provide nanosecond order synchronisation, and hence excellent pulse polarity charged (+/- 35 kV) generator. With an erected outputreproducibility. The PFLs are charged by a single Marx through an oil- capacitance of 100 nF, the open-circuit output voltage is 350 kV. The 1insulated transmission line. The outputs of the cells are added along a uF capacitors are switched via synthetic-air-filled UV-illuminated spark22 metre long 80 ohm MITL to deliver an 11 MV forward going wave gaps. The scope of this test series was to provide data for theto the e-beam diode. The accelerator will be used for flash radiography optimization of the Marx switch pressure ranges and flow rates for bothby AWE utilising a Self Magnetic Pinch diode as the radiographic a traditional FXR Marx module and a modified test stand module.source. This diode operates at approximately 40 Ohms with the result Previous recommendations for a modification to the triggering circuitthat retrapping of the MITL sheath current occurs, reducing the diode were tested against the nominal triggering circuit. Prefire and no-firevoltage to ~ 7.5 MV, but increasing the load current to 200 kA. The curves for operating pressures were constructed, jitter was analyzed,detailed PFL design was previously prototyped and has been chosen to and a stress test was developed to determine reliability from a minimaltailor the output pulse to compensate for the SMP diode’s intra pulse number of shots. This work was performed under the auspices of theimpedance history and hence generate a relatively constant voltage U.S. Department of Energy by Lawrence Livermore Nationalduring generation of the X-ray flash. The components of the Hydrus Laboratory under Contract DE-AC52-07NA27344.IVA are approaching completion at L3 Pulse Sciences at San Leandro,CA. All parts of the IVA are being procured by L-3 PS and delivered toSan Leandro for subassembly. The major IVA subassemblies beingfabricated comprise the Marx, oil line, PFL, cell, and stalk. Ancillarysystems being fabricated comprise the control software, vacuum, waterprocessing, oil processing, core reset, gas processing, data acquisition,and power supply. Subassemblies and subsystems are subject to avariety of QA tests which include high voltage testing of the Marx and 113
  • 113. 5A-5: Xray Diodes Experiments on the Asterix Generator pulse duration about 100 ps are attained in helium, hydrogen, andB. Etchessahar1, R. Rosol1, R. Nicolas1, L. Hourdin1, F. Poulet1, nitrogen at a pressure of about 60, 30, and 10 torr, respectively. UsingL. Magnin1, F. Cartier1, D. Collignon1, T. Somerlinck1, M. Caron1, the extracted data, estimates show that with the pulser SLEP used,B. Cassany2, L. Voisin2, T. Desanlis2, B. Bicrel2, P. Modin2, optimization of all parameters of the gas diode makes possible a SAEBA. Garrigues3, C. Delbos3, I. Soleilhavoup3 current amplitude of about 500 A at a helium pressure of 60 torr. As the1 DIF / PEM, CEA, 51490Ponfaverger-Moronvilliers, France mode of the beam generation is changed due to the decrease in pressure2 CESTA, CEA, 33114 Le Barp, France from the above values, one can gradually control the FWHM of the3 CEG, CEA, 46500 Gramat, France beam current pulse between 100 and 500 ps in all gases studied. In this case, the maximum amplitude of the beam current is about 1 kA and is greater than the maximum SAEB amplitudes. At low pressures (<1 torr)ASTERIX is a 6MV Marx+Blumlein X-ray generator1, dedicated to of helium, hydrogen and nitrogen in the gas diode, the amplitude of thethe testing of electrical components. For our experiments, the vacuum beam current decreases and its duration continues to increase to overdiode has been modified in order to set up flash radiographic diodes. the duration of the pulser pulse with the matched load. It isTwo kinds of diodes have been tested in this way for several years: the demonstrated that SAEB is generated into an angle exceeding 2π sr inself magnetic pinch diode (SMP) and the negative polarity rod pinch the air at the pressure of 1 atm. Generation of the electron beam into thediode (NPRP)2. The SMP diode usually provides high dose but with a solid angle exceeding 2π sr is explained by the fact that at SAEBlarge spot size. The NPRP diode allows <2mm spot sizes but with generation the dense plasma near the cathode has the form close to thelower doses. spherical one. A time-of-flight spectrometer study and reconstruction ofThe first goal of these new experiments is the demonstration of a stable the spectrum from the data on e-beam attenuation confirmed the factand reproducible functioning point for each diode. The other main goal that groups of electrons with two-three characteristic energies can beis a better understanding of the physical phenomena, by the use of new generated in gas-filled diodes. In experiments, electrons of energydiagnostics such as 4 images high speed camera, streak camera and greater than that corresponding to the nominal voltage amplitude acrosspinhole cameras. the gap were detected. X-ray radiation characteristics from a diodeThe NPRP diode consists of a small diameter (few mm) cylindrical filled with atmospheric pressure air were investigated The experimentsanode extending from the front end of the vacuum cell through a thin have shown that massive-anode gas diodes filled with atmosphericannular cathode, held by the central conductor. The optimal pressure air are the sources of soft X-ray radiation with aconfiguration from the previous experiment3 has been upgraded: the subnanosecond pulse length. A source of soft X-ray radiation with thespot size has been reduced without dose reduction: the diode now FWHL less than 200 ps and exposure doze of about 3 mR per pulse hasprovides 45 rads@1m (in Al) with a 1.7 mm spot size (CEA/LANL been created.definition) at 4.5 MV gap voltage.The SMP diode consists of a cylindrical hollow aluminum cathode infront of a planar multi-layers anode (aluminum, vacuum, tantalum, 5A-7: 3D Simulations of the Self-Magnetic-Pinch Diodecarbon). The best performances reached with a not yet fully optimized N. L. Bruner1, D. R. Welch1, B. V. Oliver2, M. D. Johnston2 1diode are 140 rads@1m (in Al) with a 4.5 mm spot size (CEA/LANL Voss Scientific, LLC, Albuquerque, NM, United States 2definition) at 4.5 MV gap voltage. Sandia National Laboratories, Albuquerque, NM, United SatesFor both diodes, the visible images of the diode show the anode andcathode plasmas and the electronic pinch during the pulse; the streak The self-magnetic-pinch diode is being developed as an intensecamera gives the expansion speed of the plasmas; the pinhole cameras electron beam source for pulsed-power driven x-ray radiography.show the connection areas of the electrons to the anode. Recent 2D simulations of this diode included the formation and1. G. Raboisson, P. Eyl, M. Roche, C. Malaval and A. Johan, evolution of a dense (~ 10^{18}/cc) anode surface plasma“ASTERIX, a high intensity X-ray generator”, 7th Pulsed Power simultaneously with the formation and propagation of the thinnerConference, 1989, pp. 567-570. electron beam (10^{13}/cc). This enabled the first modeling of the2. S.B. Swanekamp, C. Vermare, F. Bayol, B.V. Oliver et al., “Angular time-dependent impedance behavior observed on RITS-6. Follow-ondose variations from 4 to 6 MV rod-pinch diode experiments on the simulations have been conducted to determine the effect of 3DASTERIX pulsed-power generator”, 14th Pulsed Power Conference, instabilities on diode performance. These results are presented here.2003.3. B. Etchessahar “Negative Polarity Rod Pinch diode experiment onthe ASTERIX generator”, 37th International Conference On Plasma Multiple Pulsed Power Systems for Free Electron 5A-8:Science, 2010. Maser Amplifier Experiments C. G. Whyte, C. W. Robertson, K. Ronald, A. R. Young, W. He, A. W. Cross, P. MacInnes, A. D. R. Phelps Generation Supershort Avalanche Electron Beam and5A-6: SUPA, Physics, University of Strathclyde, Glasgow, United KingdomX-Ray During Subnanosecond Breakdown in DifferentGases at Pressures from 1 Torr up to 15 AtmV. F. Tarasenko We present details of the pulsed power systems and results from aHigh Current Electronics Institute, Tomsk, Russian Federation reversed guide magnetic field Free Electron Maser (FEM) amplifier(1) experiment using a two-stage depressed collector at Strathclyde University. FEM amplifiers are broadband devices in contrast to theThis paper reports the properties of supershort avalanche electron beam single frequency FEM oscillators(2-4) that have also been constructed(SAEB) and x-ray formed in different gases at pressures from 1 torr up and operated at Strathclyde. The amplifier was driven by a pair of fourto 15 atm and analyses the SAEB generation mechanism. An electron cable stacked Blumlein cable transformers firing with opposedbeam was obtained behind foil at the pressure of sulfur hexafluoride polarities to give a maximum voltage of 345 kV at the high impedanceand xenon in a gas diode up to 2 atm, at the pressure of nitrogen up to 5 diode. The paired Blumlein pulsers use a common field distortionatm and at the pressure of helium up to 15 atm. It is shown that at a spark-gap switch with a 3:1 charge to trigger ratio to ensure timingnanosecond discharge in the atmospheric pressure air and SAEB synchronization. The trigger pulse was provided by a thyratron-current recording through a small area of the foil the pulse duration switched, capacitively-coupled, 50 Ohm line which provides reversebehind the foil is not larger than 80 ps, but from all surface of foil is power protection to the thyratron. The two stage depressed collectorabout 100 ps. It is shown that the highest amplitudes of SAEB with 114
  • 114. was driven by a separate 7 stage Marx bank with inductive output coaxial drift tube was obtained under the assumption of strong guidecoupling and a variable division ratio high power resistive voltage magnetic field.divider. Synchronisation of the two power supplies was achieved by In this contribution we present a comparison of the analyticalsequential triggering of the short pulse Blumlein generator from the expression for the SCL current in the coaxial drift tube with results ofMarx bank output terminal. This was necessary to circumvent the numerical simulations for infinite and finite drift tubes and theirtiming jitter associated with the seven Marx bank spark gaps. dependence on the entrance azimuthal velocity of the beam (see alsoAdditionally the FEM uses a pulsed electromagnetic helical undulator [Proc. 20th Int. Crimean Conf. “Microwave & Telecomm. Technol.”,driven by an ignitron switched damped capacitor discharge circuit Sebastopol, Ukraine, p. 237, 2010]).delivering 14kA peak current. The FEM has been designed for zero We give analytical expressions for SCL current in the presence ofbeam-wave slippage to achieve maximum instantaneous bandwidth. dielectric loads and find the next to the leading order analyticalThe bandwidth of operation can be further extended by adjustment of corrections to the SCL current expression if the inner conductor isthe beam voltage and undulator field strength. A 1kW TWT has been biased and a dielectric load is present simultaneously (cf. [Phys.used as the input source for broadband measurements and two 25kW Plasmas, vol. 16, 093116, 2009]). It is shown that such anmagnetrons for saturated output power measurements at discrete approximation is very accurate over the whole range of applied biases.frequencies in the range. Results show instantaneous -3dB bandwidth We also discuss corresponding description of a model coaxial gyro-of 30% of centre frequency for fixed cathode voltage and magnetic TWT beam transport channel in which two finite thickness charged-field values. The device can be tuned over a range of 65% of centre particle beams of not necessarily equal currents, dielectric loads nearfrequency by adjustment of the cathode voltage. Amplifier saturated the outer and inner conductors, and the inner conductor bias are present.gain was measured at 30dB, giving a peak output power ofapproximately 1MW at approximately 6% efficiency. Operation of the 5B: Applications II: General ApplicationsFEM with the depressed collector raised the maximum efficiency to 25% Wednesday, June 22 09:30-12:00 CC 10C-D1) Free electron maser amplifier experiments, Whyte, C.G., Jaroszynski,D.A., Cross, A.W., et al.Nucl. Instr. and Meth. A 445, Issue: 1- 5B-1:(INVITED) European Laboratory for Pulsed Power3, pp. 272-275, 2000. Research (EPPL)2) A 32 GHz Bragg free-electron maser (FEM) oscillator with axial E. Spahn1, J.-M. Löeffler2, S. Balevicius3guide magnetic field, Cross, A.W., Ginzburg, N.S., He, W., et al.Nucl. 1 ISL, Saint-Louis, FranceInstr. and Meth. A 407, Issue: 1-3, pp. 181-186, 1998. 2 University of Applied Sciences, Gelsenkirchen, Germany3) Progress of the Strathclyde Free Electron Maser experiment using a 3 Semiconductor Physics Institute, Vilnius, Lithuania2D Bragg structure,Konoplev. I.V., Cross A.W., He W., et al., Nucl.Instr. and Meth. A 445, Issue: 1-3, pp. 236-240, 2000.4) Experimental and theoretical study of 2D Bragg structures for a In this paper a European laboratory in the field of pulsed powercoaxial free electron maser, Cross, A.W., He, W., Konoplev, I.V. et al., research (EPPL) will be presented. The approach for this laboratory is aNucl. Instr. and Meth. A 475,Issue: 1-3, pp. 164-172, 2001. bottom-up one. Bilateral collaborations between the French-German Research Institute of Saint-Louis (ISL) and the other European labs were extended towards multilateral ones, finally ending up in a5A-9: Power Positron Beams for HED Physics common lab (EPPL). The objective of this initiative is to bundle theV. V. Gorev resources of the involved institutions and to bring together research andKurchatov Institute, Moscow 123182, Russian Federation expertise of Europe’s principal pulsed power laboratories with the goal to minimise overlap research, to achieve a critical mass and to establishIn this talk I will discuss the modern theoretical developments and a European network. One purpose of this network (cluster) is to applyreview today situation in power positron beams reseach,including for international research projects (e. g. FP 7, EDA). In additionphysical and technical aspects of generation and absorbtion of the common research projects in the framework of pulsed power will bepositron beams with Gev particle energy and about (0,01-0,1)A beam conducted, courses, workshops, symposia and conferences will becurrent ,"positive charge pumping"process,with emphasis on organized. In the end EPPL is ought to be developed to a Centre ofsuggestions for some HED(High Energy Density) physics Expertise in the field of pulsed power at least in Europe. This commonapplications:1.macroparticle accelerator.,2.proton and heavy ion laboratory consists of six national labs from France, Germany, Belgium,intermediate energy accelerator.,3.soft(about few mev energy Italy, Lithuania and Sweden. The pulsed power topics of EPPL cover adiapason)gaser.I will also describe the physics and conceptual design of broad range. Starting with pulsed power components like switches andthe machines 1-3. pulsed power sources complete pulsed power systems are studied. In terms of measurement technology currently the focus is put on high magnetic field sensors, so called CMR-sensors, being suited for5A-10: Space-Charge Limited Current of Charged-Particle measuring magnetic fields up to 50 Tesla with a very high spatialBeam in Coaxial Drift Tube resolution. In the area of pulsed power applications both military andK. Ilyenko1, G. V. Sotnikov2, T. Y. Yatsenko1 civilian (commercial) ones are envisaged. High power microwaves are1 Department of Vacuum Electronics, Institute for Radiophysics and investigated for security applications and electromagnetic launchers areElectronics of NAS of Ukraine, Kharkiv, Ukraine studied both for military and for space applications. In the civilian area2 National Science Center “Kharkiv Institute of Physics and Technology” possibilities to use pulsed power for generating nanomaterials areof NAS of Ukraine, Kharkiv, Ukraine investigated in detail. In summary EPPL is an international laboratory consisting of six independent European institutions which bundle theirRecently, there is a substantial interest in the development, design, and pulsed power research in an efficient way.investigation of coaxial high-current devices utilizing relativisticcharged-particle beams (see, e.g., [Phys. Plasmas, vol. 16, 043104,2009] and references therein). In [Tech. Phys., vol. 47, p. 535, 2002]for the first time an analytical expression for space-charge limited (SCL)current of charged-particle beam propagating in an infinite grounded 115
  • 115. The PHELIX Pulsed Power Project: Bringing Portable5B-2: The University of Missouri-Columbia has recently developed anMagnetic Drive to World Class Radiography experimental test stand for the purposes of studying the interaction ofC. L. Rousculp, W. A. Reass, D. M. Oro, P. J. Turchi, B. J. Hollander, shock waves with plasmas in air. The Center for Physical and PowerJ. R. Griego, R. E. Reinovsky Electronics has developed a test stand which includes a robust 8” shockLos Alamos National Laboratory, Los Alamos, NM, United States tube capable of producing a reliable shock wave for the study of interaction of the wave front with an atmospheric air plasma. The shock wave is formed by rupturing a mylar film diaphragm separating theThe PHELIX pulsed power project will introduce magnetically driven pressurized driver section from the unpressurized driven section of thehydrodynamics experiments to the Los Alamos National Laboratory’s tube. For reproducibility, an exploding wire is used to rupture theproton radiography facility (pRad). The Precision High Energy-density diaphragm.Liner Implosion eXperiment (PHELIX) has been commissioned at Los A multi-millisecond duration, exploding wire air plasma with electronAlamos. A small footprint capacitor bank consisting of four parallel density of 1014 -1015/cm3 has been developed for generation of theconnected single-stage marx units (~500 kJ) is cable coupled to a plasma interaction with shock waves. The wire plasma is used to firsttoroidal, current step-up transformer to deliver multi-Mega-Ampere, form the plasma and a secondary discharge region is used to further~10 μs current pulses to cm size cylindrical loads. In a sequence of tests heat and impart momentum to the plasma. We have been able tothe performance of each component (capacitor bank and transformer) produce a self-confined, self-stabilized, toroidal plasma using thiswas evaluated and compared to a circuit model. The transformer method of post discharge heating. Using the toroidal geometry hascoupling was observed to be k ~ 0.93. The tests culminated in a liner resulted in interesting hydrodynamic effects, producing a toroidal airimplosion experiment in which an ~3 cm radius, 0.8 mm thick, ~3 cm plasma, or TAP with milliseconds of lifetime.tall aluminum liner was accelerated to a velocity of ~ 1 km/s. The suite A full diagnostic suite of pressure sensors, high speed photography,of machine diagnostics included linear Rogowski coils and Faraday schlieren photography, as well as current and voltage measurements,rotation for current measurements. The experimental diagnostics have been assembled to observe the interaction of the shock wave andinclude B-dot probes, multi-channel photon Doppler velocimetry the TAP. Data on the results of the interaction experiments, as well as(PDV), and single-frame, flash X-radiography to evaluate the the experimental set up, are presented.performance of the high precision liner implosion. Currently, work is This work was supported by The Office of Naval Research underfocused on integrating PHELIX into normal operations with the 800 contract number S-000296.00001. UMC Curry Applied ResearchMeV proton radiography facility. There, high-resolution, high-frame- Associatesrate imaging of hydrodynamic experiments will be possible. The PHELIX Portable Pulsed Power Machine: 5B-5: The Generation of Triggered Shockwaves in Shock5B-3: Hydrodynamics Experiments and BeyondTubes with Exploding Wires C. L. Rousculp, R. E. Reinovsky, W. A. Reass, D. M. Oro, P. J. Turchi,M. E. J. Rudroff, A. Lodes, R. D. Curry D. B. Holtkamp, B. J. Hollander, J. R. GriegoDept. of Electrical & Computer Engineering, University of Missouri, Los Alamos National Laboratory, Los Alamos, NM, United StatesColumbia, MO, United States The PHELIX pulsed power machine is designed to drive cm-scaleThe University of Missouri-Columbia Center for Physical and Power condensed matter, shock physics and hydrodynamics experiments in anElectronics has implemented the use of exploding wires to trigger the architecture compatible with the high resolution, multi-frame protondiaphragm in a 20.3 cm diameter PVC shock tube. The shock tube is radiography diagnostic at the Los Alamos Neutron Science Centerpartitioned into two segments: the driver side and the driven side. The (LANSE). To meet the tightly constrained footprint available forhigh pressure driver side is partitioned from the low pressure side by a experiment on LANSCE, PHELIX employs a compact, modularized0.127 mm thick Mylar sheet with an exploding wire taped to it. The 120KV capacitor bank coupled to a current step-up transformershockwave is generated in the tube by pressurizing the driver side of packaged into a self-contained, removable and portable unit. Each ofthe shock tube to approximately 30 psi and discharging of the energy two capacitor modules is comprised 4 “ATLAS,” 60 kV, 60 kJ,from an 11uF capacitor bank into a 30 AWG copper wire. The 7.3 cm fiberglass encased capacitors switched with a two 120 kV raillong wire vaporizes as the 4.95 kJ of energy is discharged into it at a switches .The output of each module goes through a low-inductance,peak current of 3 kA. The wire, which is taped to the Mylar, behaves reticulated vitreous carbon resistor (R ~ 10 mΩ) to prevent excessivelike a fuse; puncturing the Mylar diaphragm as it vaporizes. This voltage reversal and is cable coupled to the four-turn primary windingscatastrophic failure results in the low jitter generation of a shockwave of a toroidal, multi-filar, air-insulated, transformer. The single-turnwhereby experiments can be timed to the arrival of the shock wave. secondary winding consists of a machined multi-segment, aluminumThe key benefit to minimizing the jitter of shockwave generation toroid that surrounds the primary windings. The secondary windingallows for better timing for shockwave diagnostics and functionality. feeds a short (~10 cm) radial, disk transmission line with a thin (~1mm)Through our experimentation, we have seen shockwave jitter times of solid insulated gap, and the compact liner implosion experimental load.less than 5us. This low jitter allows us to efficiently capture data, as PHELIX provides a very fast (<5 us) rise-time and 4-5 MA peakwell as time external events designed to interact with the shock wave. current drive with low reversalThe optimization of the exploding wire for these experiments, as well PHELIX is also a prototype magnetic drive option for the LANLas the pulsed power driver for the exploding wire, is reported. Materials and Radiation in Extremes (MARIE) initiative. In addition toThis work was supported by Applied Research Associates under liner implosion applications, other types loads connected to thecontract number S-000296.00001.UMCCurry Applied Research transformer such as a linear (rail gun) accelerator for planar impactAssociates experiments, and a dense-plasma-focus have been assessed. Addition of power conditioning switches in the PHELIX circuit offer promise of5B-4:The Interaction of Shock Waves with a High Density increased efficiency for some load configurations, and faster risetimeToroidal Air Plasma for others. In place of the current multiplying transformer, a pulsed,A. Lodes, M. E. J. Rudroff, R. D. Curry high-field magnet has been proposed as a part of a high-energy particle experiment. Conceptual applications of PHELIX to a variety ofDept. of Electrical & Computer Engineering, University of Missouri,Columbia, MO, United States 116
  • 116. experiments that employ the proton-radiography diagnostic will be of non-thermal plasma in aqueous solution (mostly water solution)discussed. environments for a variety of applications in environmental, materials, biomedical, and chemical engineering. High voltage electrical discharge, especially using pulse power input is one of the most5B-6:Design of Pulsed High-Field Magnets for Pion/Muon common methods to generate non-thermal plasma due to the ease ofCollection controlling the electric field distribution and very low powerP. J. Turchi consumption which yield in significantly high energy efficiency inPhysics Division, Los Alamos National Laboratory, Los Alamos, NM, various physical and chemical applications.United States Unlike gas phase electrical discharge, the physical and chemical properties of electrical discharge in aqueous solution are still notBased on the potential for improving collection of negative-muons for thoroughly clear. Plasma properties, such as initiation and mechanismstandoff active-detection of fissile materials, we explore the design of in aqueous solution had been discussed [1]. Various chemical reactionspulsed high-field (15-20T) magnets that might be powered by the generated by non-thermal plasma in aqueous solution were alsoPHELIX capacitor bank and used in experiments with the 800 MeV discussed via chemical diagnostics and computer simulations [2]. Also,proton beam at Line C of LANSCE. The conventional approach to previous literature [3] has discussed several important experimentalcollection is based on superconducting magnets at 2 – 5T and requires factors’ effects, such as pressure, temperature, electrode configurationclear bores in excess of a meter over lengths of a few meters. The gyro- and solution conductivities on electrical discharge induced non-thermalradius for a 35 MeV pion is about 1.7 cm at 20T, a factor of ten smaller plasma in aqueous solution. There were also literatures pointed outthan in a 2T field, so similar reductions of system size are anticipated. some interesting effect of power input on chemical generationExperiments with pulsed high magnetic fields could include the efficiency [4]. However, there is not much information and no detailplacement of the pion-production target in or near the center of a pulsed analysis about the influences of different pulse power inputsolenoidal magnet and, with two such coils separated by several gyro- characteristics on chemical processes of electrical discharge in aqueousdiameters, at the center of a magnetic mirror trap. In both cases, the solution. In the present study, different pulse characteristics producedprotons would be injected along the axis of the magnet at the time of by semiconductor controlled magnetic switch pulse power generatorpeak field, producing negative-pions with a range of energies and were used to generate electrical discharge in aqueous solution.angles (20 – 50 MeV, and 0 – 2pi, respectively). For the single-coil Chemical diagnostics are used to evaluate the plasma chemicalexperiment, these pions would be constrained radially and expelled processes. Hydrogen (H2) and hydrogen peroxide (H2O2) are veryaxially. In the concept of a magnetic mirror trap, some pions would important gas and liquid molecule species generated from plasma-waterescape through the loss-cone, but most would be retained for times interaction. H2 and H2O2 are not only precious products formuch longer than their half-life (26 nsec), resulting in a population of applications (renewable energy, chemical and biological treatment), butnegative-muons held within a magnetic “bottle”. As an alternative also good indicator for plasma chemical processes. In this study, twoapproach, we consider the use of a pulsed azimuthal magnetic field different pulse power generators were used to generate discharge increated around the pion target by a plurality of conductors forming a aqueous solution. Chemical formations are tested under applyinglargely open, coaxial inductor. The minor cross-section of this various power input, peak voltage, pulse raise time, pulse duration andarrangement is shaped to turn pions of different energies in pulse repetitions. Effects of the individual pulse input factor on energyapproximately the same axial direction, thereby providing a magnetic yield efficiency and plasma characteristics will be shown and discussed.“nozzle”; similar arrangements are known elsewhere as a “horn”. In all 1. Locke, B. R., M. Sato, et al. (2006). "Electrohydraulic discharge andthese approaches, the constraints on our design are the gyro-radius of nonthermal plasma for water treatment." Industrial & Engineeringpions with the highest angular momentum, the acceptance aperture Chemistry Research 45(3): 882-905.(~12 cm diam) of the subsequent normal-conducting RF linac, the 2. Mededovic, S. and B. R. Locke (2007). "Primary chemical reactionsstrength and thermal capacity of the pulsed magnet, and the circuit in pulsed electrical discharge channels in water." Journal of Physics D-parameters of the PHELIX capacitor bank, (a two-section, two-stage Applied Physics 40(24): 7734-7746.Marx, comprising a total of eight 34 microfarad capacitors for an 3. Shih, K. Y. and B. R. Locke (2009). "Effects of electrode protrusionenergy of 490 kJ). PHELIX (Precision High Energy-density Liner length, pre-existing bubbles, solution conductivity and temperature, onImplosion eXperiment) was designed for operation with a current step- liquid phase pulsed electrical discharge." Plasma Processes andup transformer to drive liner implosions that would be diagnosed with Polymers 6(11): 729-740.proton radiography at Line C. The pulsed magnets of the present 4. B. R. Locke and Shih, K. Y., “Review of the method to formconsideration would be substituted for the PHELIX transformer. hydrogen peroxide in electrical discharge plasma with liquid water.”Principal concerns are the mechanical and thermal stresses associated Plasma Sources Science and Technology, in press, 2011with operation at upwards of 20T, corresponding to a magnetic pressureof 1600 atm and a temperature rise of a few hundred C per shot. 5B-8:Gas temperature measurements of nano-secondsApproximate hand-calculations and detailed ANSYS simulations are pulsed discharge based ozonizerused to address these concerns. T. Matsumoto1, N. Takamura1, D. Wang2, T. Namihira3, H. Akiyama1Work supported by DTRA/NTDS and NNSA 1 Graduate School of Science and Technology, Kumamoto University, Kumamoto, Japan5B-7:Characteristics of High Voltage Electrical Discharge 2 Priority Organization for Innovation and Excellence, KumamotoInduced Non-Thermal Plasma in Aqueous Solution under University, Kumamoto, Japan 3Pulsed Control Bioelectrics Research Center, Kumamoto University, Kumamoto,K.-Y. Shih, S. Iyonaga, M. Akiyama, N. Aoki, H. Akiyama JapanGCOE, Kumamoto University, Kumamoto, Japan Ozone has some effects such as sterilization, deodorization, andNon-thermal plasma in contact with water molecule generates intense decolorization, because of its strong oxidization power. Therefore theUV radiation, shock wave and most importantly reactive molecular and application field of ozone has spread to water clarification, waste waterradical species (such as hydrogen peroxide and hydroxyl radial). disposal and semiconductor cleaning applications, etc. However, inTherefore, it is an increasing interest in the analysis and development industrial situation, the improvement of energy efficiency for ozone 117
  • 117. generation is demanded in order to spread the ozone processing Science Meeting and Exhibit, Reno, NV.moreover. Recently, it is demonstrated by our research group that the Sandia National Laboratories is a multi-program laboratory managednon-thermal plasma produced using the nano-second pulsed discharge and operated by Sandia Corporation, a wholly owned subsidiary ofhas extremely higher energy efficiency to produce of ozone. Typically, Lockheed Martin Company, for the U. S. Department of Energy’sthe yield of ozone generation were 500 and 300 g-O3/kWh for 10 and National Nuclear Security Administration under contract DE-AC04-30 g/m3 of ozone concentration in oxygen fed. On the other hand, the 94AL85000.maximum ozone concentration has been saturated at approximately 40g/m3. Therefore, for waste water treatments, the ns pulsed discharge 5C: Components IV: Solid State Switchingbased ozonizer still needs to have the higher concentration around at120 g/m3. In this work, since many researchers reported that gas Wednesday, June 22 09:30-12:00 CC 11A-Btemperature in ozonizers strongly affects the characterization of theozone generation, the gas temperature inside of the ozone generation 5C-1:Laser Enhanced Diffusion of Nitrogen in High Purityreactor was experimentally measured using a fluorescence thermometer. Semi-Insulating 4H Silicon Carbide Substrates for Non-As a result, a temperature dependency of ozone concentration was Rectifying Contact Formation to Photoconductiveobserved. Typically, when the gas temperature was raised up to 323 K,the increasing of ozone concentration stopped at approximately 40 Semiconductor Switches W. Sullivan III, C. Hettler, J. Dickensg/m3. Therefore, it is important to cool down the feeding gastemperature in order to obtain the higher ozone concentration. Center for Pulsed Power and Power Electronics, Texas Tech University, Lubbock, Texas, United States5B-9: Simulation and Analysis of Magnetically-Applied- Non-rectifying (ohmic) contacts are essential for efficientPressure-Shear (MAPS) Experiments photoconductive semiconductor switch performance and maximizingT. A. Haill, C. S. Alexander, J. R. Asay breakdown voltage. Fabricating ohmic contacts requires a very heavilyPulsed Power Sciences Center, Sandia National Laboratories, doped surface layer (> 1017 cm-3), and in silicon carbide (SiC) isAlbuquerque, NM, United States typically done by ion implantation. The high energy ions from this process often cause surface and bulk damage, and a high temperatureA new experimental technique to measure material shear strength at anneal is required to repair the crystal structure and activate thehigh pressures has been developed for use on magnetohydrodynamic impurities. This paper investigates the use of a gas immersion laser(MHD) drive pulsed power platforms [1]. The technique is referred to enhanced diffusion system to selectively dope the SiC as an attractive,as Magnetically-Applied Pressure-Shear (MAPS). By applying an low cost alternative to ion implantation. A tripled Nd:YAG laser (14 nsexternal static magnetic field to the sample region, the MHD drive FWHM @ 355 nm) at 80 mJ and 100 mJ (14 MW/cm2 and 18directly induces a shear stress wave in addition to the usual longitudinal MW/cm2, respectively) was used to nitrogen dope two high puritystress wave. Strength is probed by passing this shear wave through a semi-insulating (HPSI) 4H-SiC samples, each with measured nitrogensample material where the transmissible shear stress is limited to the surface concentrations greater than 1020 cm-3. Secondary ion masssample strength. The magnitude of the transmitted shear wave is spectroscopy (SIMS) analysis is presented for the shallow and deepmeasured via a transverse velocity interferometry system (VISAR) doping cases, depths of 60 nm and 1000 nm respectively, as well asfrom which the sample strength is determined. scanning electron microscope surface images and a transfer lengthThe strength of materials is defined as the ability of a material to method contact analysis. It was found that laser fluences above 250sustain deviatoric (shear) stresses. Strength is an important aspect of the mJ/cm2 produced visible surface damage and ablation. Using a oneresponse of materials subjected to compression to high pressure. dimensional thermal model, the experimental data was fit to diffusionBeyond the elastic response, material strength will govern at what coefficients that are significantly greater than typically seen in SiC. Thepressure and to what extent a material will plastically deform. The gas immersion laser doping technique has been demonstrated as aMAPS technique cleverly exploits the property that, for a von Mises viable alternative to ion implantation for selective area doping of SiCyield criterion at a given longitudinal stress, the maximum amplitude bulk photoconductive switches.shear wave that can be transmitted is limited by the strength at thatstress level. 5C-2:Analysis of Silicon Carbide JFET Devices DuringSuccessful fielding of MAPS experiments to measure shear stressesrelies upon correct numerical simulation of the experiment. Complex Pulsed Operation K. J. Lawson1, G. Alvarez1, S. B. Bayne1, V. Veliadis2, D. Urciuoli3wave interactions among forward and reflected longitudinal and shear 1waves, as well as the advancing magnetic diffusion front of the MHD Center for Pulsed Power and Power Electronics, Texas Techdrive, can make the design of the experiment complicated. Careful University, Lubbock, TX, United States 2consideration must be given to driver, sample, and anvil materials; to Northrop Grumman Electronic Systems, Linthicum, MD, United States 3the thicknesses of the driver, sample and anvil layers; as well as to the U.S. Army Research Laboratory, Adelphi, MD, United Statestiming of the interacting waves.This paper will present and analyze the 2D MHD simulations used to The purpose of this research is to investigate the performance of Silicondesign the MAPS experiments. The MAPS experiments are modeled Carbide JFET devices in repetitive pulsed applications. The deviceusing Sandia’s ALEGRA-MHD simulation code [2]. ALEGRA-MHD under test for these results is a research grade JFET with a ratedis an operator-split, multi-physics, multi-material, arbitrary lagrangian- blocking voltage of 1200V and a rated forward current of 10A with aeulerian code developed to model magnetic implosion, ceramic fracture, power density of 200W/cm2. In order to drive the JFET device, aand electromagnetic launch. We will detail the numerical investigations unique gate driver had to be designed and built to provide switchinginto MHD shear generation, longitudinal and shear stress coupling, between two independent rail voltages. The gate driver had to be abletiming of wave interactions, and transmission of shear at material to provide adjustable rail voltages with one rail ranging between 0Vinterfaces. and -40V and the other rail going between 0V and 2.5V. The gate[1] C.S. Alexander, J.R. Asay, and T.A. Haill, J. Appl. Phys., 108, driver is completely isolated to operate on a high-side switch. In order126101 (2010). to test these devices in pulsed switching applications a pulse ring down[2] A.C. Robinson, et al., AIAA 2008-1235, 46th AIAA Aerospace circuit was designed and built to provide a current pulse of 100A (10 118
  • 118. times the rated current) with a charging voltage range between 100V previously taken up by perimeter high voltage termination can now beand 500V. Special consideration had to be given to the design of this used for conduction. The Super-GTO has a cell-based emitter layoutpulse ring down circuit in order to achieve a high di/dt, and therefore optimized for high-action (I2t) applications. Several wafers of 7 cm2reach the target peak current levels. During these tests, the drain to Super-GTOs were successfully fabricated. Individual devices weresource voltage and drain current were analyzed to see how these packaged with Silicon Power’s solder-based ThinPak lid in place ofdevices handle extreme over current situations while being pushed into more traditional wire bonds. The Super-GTOs were switched at thesaturation. These devices were tested over a temperature range of 25°C Army Research Laboratory in a low-inductance pulse-forming network.to 150°C to determine if the performance of the device changes over Pulse current was successfully stepped up as high as 35 kA,this wide range of operational temperatures. After the initial pulse corresponding to a current density of 5 kA/cm2 over the chip’s footprint,testing was completed, the devices were repetitively pulsed at 1Hz, or 7 kA/cm2 over the active emitter area, excluding the edge10Hz, and 100Hz with a charging voltage of 150V and a peak current termination. Compared to Silicon Power’s 3.5 cm2 Super-GTOs ofof 100A for 1000 shots. These sets of tests were also performed over similar design, the new larger device conducted 45% higher currentthe previously mentioned temperature range. Before and after each of density over the chip footprint. The 35-kA current pulse had a width ofthese tests, the devices were placed on an Agilent B1505A Power 150 micro-seconds and an I2t of 9.4 x104 A2s. The 10-90% rise of theDevice Analyzer to determine if these performed tests had any current pulse was 2.5 kA/us, and the maximum on-state forward droppermanent effects on the JFET device characteristics. was 27 V. Given good processing and packaging yields, this larger Super-GTO can greatly reduce the size of high current pulse switches. Additional pulse parameters that will be assessed with these new5C-3:Recombination Lifetime Modification in Bulk, Semi- devices include faster dI/dt, dV/dt immunity, and shot lifetime. ThisInsulating 4H-SiC Photoconductive Switches paper will also include information on developing switch modules ofC. Hettler, W. Sullivan III, J. Dickens parallel 7 cm2 Super-GTOs.Texas Tech University, Lubbock, TX, United States 5C-5:(INVITED) High Current, Multi-FilamentA series of high temperature annealing experiments were performed to Photoconductive Semiconductor Switchingcharacterize the processing parameters that alter the recombination F. J. Zutavern, S. F. Glover, A. Mar, G. M. Loubriel, M. E. Swalby,lifetime in high purity, semi-insulating (HPSI) silicon carbide (SiC). R. T. CollinsAll annealed samples were diced from a single 4H-SiC wafer with ameasured resistivity of greater than 109 Ω-cm. The samples were Sandia National Laboratories, Albuquerque, NM, United Statesannealed for various lengths of time in a PID-controlled hightemperature induction furnace at incremental temperatures up to High current switching is the most critical challenge remaining for1850 °C. The impacts of soak temperature, soak time, and cooling rate PCSS applications in Pulsed Power. Many authors have described thewere examined. A 36 GHz microwave photoconductivity decay advantageous properties of high gain PCSS such as, low optical trigger(MPCD) system was used to measure the transient photoconductivity energy and inductance, sub-nanosecond rise time and jitter, opticalof the as-grown and processed samples. Through numerical processing isolation and control, pulsed or DC charging, and long device lifetime,of the temporal characteristics of the illuminating laser pulse, the provided the current per filament is limited to 20-30A for short pulsephotoconductivity transients were simulated with various (10-20ns) applications. Low energy optical triggering, long devicerecombination lifetimes to fit the experimental MPCD data. The results lifetime, and current filaments are related features of high gain PCSSshow that the as-grown material has an average recombination lifetime that make high current switching a challenge. Since the location andof 6 ns. However after long high temperature anneals, many processed numbers of current filaments can be controlled by delivering parallelsamples demonstrated recombination lifetimes in excess of 100 ns. The “lines” of optical pulses across the insulating gap of the PCSS, theannealing process reduces the concentration of point defects (Z1/Z2, problem of high current, multi-filament PCSS switching is essentiallyEH6/EH7) in the bulk material which serve as recombination centers in the problem of producing a reliable, efficient, multi-line, opticalHPSI 4H-SiC, extending the carrier lifetime. A portion of the annealed delivery system.samples were metalized and the bulk resistivities were measured to In this paper, several classes of optical delivery systems will beverify that the material still maintained semi-insulating properties. discussed: line-of-sight plastic and glass micro-lens arrays, multi-modeFinally, the impacts of increased recombination lifetime in fiber-optic/micro-lens combinations, single-mode fiber-optic/micro-photoconductive switch operation and performance are presented and lens array combinations, and masked PCSS alone and with other opticaldiscussed. “concentrating” components. The application dependent advantages and disadvantages of each approach will be discussed. Results will be shown from specific examples (plastic and glass micro-lens arrays, Development of "Stitch" Super-GTOs for Pulsed5C-4: single and multi-mode fibers and bundles, and high density opticallyPower masked PCSS) that have been tested and demonstrated.H. K. OBrien1, A. Ogunniyi1, C. J. Scozzie1, W. Shaheen2, V. Temple3 The fundamental requirements of multi-filament, high gain PCSS1 RDRL-SED-P, US Army Research Laboratory, Adelphi, MD, United triggering are also being measured and will be reported. Optical triggerStates energy, pulse width, line-width, and spatial density are the key factors2 Berkeley Research Associates, Beltsville, MD, United States in determining the attainable switching efficiency and volume for high3 Silicon Power Corporation, Clifton Park, NY, United States current pulsed power applications. Measurements of these properties and the trade-offs that they imply for switching applications will beNewly designed, high-power silicon gate turn-off thyristors are being presented.evaluated to satisfy the U. S. Army’s need for compact, lightweight Results from a high voltage, high current, high gain PCSSpulse switches. Following the success demonstrating a 3.5 cm2 silicon demonstration will be described. In these tests, three 1 cm gap PCSSSuper-GTO, Silicon Power re-designed the emitter layout and increased switch 2.5 kA in a short (6 ns) Blumlein charged to 60 kV withthe device footprint to create a switch optimized for high current, wide approximately 100 filaments spaced at 30 filaments/cm. This systempulse applications. The 7 cm2 silicon “Stitch” Super-GTO was tests the practical aspects of optical triggering with the line-of-sightdeveloped to block 7 kV. The 2x increase in die size actually results in micro-lens and masked PCSS approaches. Along with these results, wea 2.5x increase in active area because a portion of chip area that was will also discuss our plans to demonstrate 100 kV, 25 kA, 1000 119
  • 119. filament switching with a bank of PCSS. version of the 13.5 kV high di/dt multichip thyristor. Until now,Sandia is a multiprogram laboratory operated by Sandia Corporation, a detailed results have only been published for prototype versions. TheirLockheed Martin Company, for the United States Department of specifications allow them to be used in pulsed power applicationsEnergy’s National Nuclear Security Administration under contract DE- which necessitate a high di/dt, such as active protection [2], pulsedAC04-94AL85000. power modulators for klystron sources, exploding wire discharge [3], high power laser source [4] and microwave generator [5]. The high reliability of the prototype design has been demonstrated on several5C-6:Novel Press Pack IGBT Device and Switch Assembly modules by switching more than 200 times an electrical energy of 40 kJfor Pulse Modulators during 100 µs for each module [2]. The only problem of these devices,P. Bill, A. Welleman, E. Ramezani, S. Gekenidis, R. Leutwyler as compared to a conventional stack of high di/dt thyristors, is a non-Semiconductors, ABB Switzerland Ltd., Lenzburg, Switzerland optimal access to ring gates, which limits the maximum injected gate current to about 800 A. In a next step ABB integrated the sameAs the trend for Pulse Modulator applications nowadays is moving assembly of one diode and three thyristor wafers (91 mm diameter) intostrongly into the direction of switch-off devices, ABB has developed a a commercial hermetically-sealed ceramic press-pack housing. Whilenew IGBT module which is designed to operate under specific pulsed first tests of these commercial devices indicated no difference in thepower, non industrial standard conditions. A press pack IGBT module performance as compared to the prototypes [1], more extensivewill be presented which is optimized for Pulsed Power applications. investigations revealed limitations of the device lifetime. Following theThe device has a collector-emitter voltage Vces of 4500V and is first device failure at 10 kV after 30 shots, we increased the number ofoffering a chip set combination of 85% IGBT and 15% diode dies. The monitoring points before proceeding to test the next ceramic housingdevice has excellent current sharing between the individual chips, no multichip device. These new channels allow us to know exactly what iswire bonding, and it will be shown that the advantages of the new happening before and after device failure.design are high peak current at high dt/dt and high switching frequency We propose to present our work on the improvement of the ceramicas well as immunity to inhomogeneous clamping. The last is very housing reliability. By recording the new probing points and byimportant in case a large number of series devices are used, especially simulating different circuit scenarios, we can demonstrate that thein applications where mechanical ruggedness is required these devices failure comes from the fact that the multichip thyristor does not reopenare superior to conventional press pack IGBT’s. A ready-to-use 20 and that it is permanently damaged at the end of switching. This failurekVdc / 3 kA switch assembly containing 8 pcs of the mentioned leads also to the destruction of a part of the gate unit, which could havedevices in series connection including cooling and driving circuit will been a possible source of failure. After analyzing failed devices, wealso be described. came to the conclusion that the problem is related to modifications of the wafer assembly process. Furthermore, we intend to present results of devices with an improved ceramic housing, which are currently5C-7:Testing of a Low Inductance Stacked Mosfet Switch for under test, promising to show the same pulse-power capability as wePulsed Ring-Down Sources obtained with multichip thyristor devices in the prototype housing.D. Reale, J. Mankowski, S. Holt, J. Walter, J. DickensCenter for Pulsed Power and Power Electronics, Dept. of Electrical [1] S. Scharnholz, “Performance study of a novel 13.5 kV multichipand Computer Engineering, Center for Pulsed Power and Power thyristor switch,” Pulsed Power Conference, 2009, p679.Electronics, TTU, Lubbock, TX, United States [2] E. Spahn, “50-kJ Ultracompact Pulsed-Power Supply Unit for Active Protection Launcher Systems,” IEEE Trans. on Magnetics, vol.A mobile and low voltage (~ 1 to 2 kV) array of Pulsed Ring Down 45(1), 2009, p462.Sources (PRDS) were required to verify the impulse radiating antenna [3] S.L. Yap, “Exploding Wire Discharge for Synthesis ofarray concept.. A solid state switch was chosen for several reasons Nanoparticles,” International Workshop On Plasma Computations &including low-voltage triggering, low cost, and small packaging. Applications, 2008, p60.However, individually packaged solid state devices at this hold-off [4] G. Tasumi, “Design concept and performance considerations forlevel, such as a high voltage MOSFETs, have high on-state resistance fast high power semiconductor switching for high repetition rate andand appreciable inductance. In order to minimize on-state resistance, high power excimer laser,” Review of Scientific Instruments, vol. 68several 500V, low-loss MOSFETs were stacked. To offset the (7), 1997, p2658.increased inductance resulting from MOSFET stacking, several stacks [5] L.E. Kingsley, “Solid-sate power switches for HPM modulators,”were placed in parallel. When connected in this configuration, the Pulsed Power Conference, 1995, p65.MOSFET inductance did not alter the natural oscillations of the coaxialradiator. The layout of the printed circuit board was designed to 5C-9: On-State Resistance Comparison of Semi-Insulatingminimize inductance and mate easily with the coaxial radiator. Results 6H-SiC Photoconductive Semiconductor Switchesfrom testing of the switch assembly are presented and compared with J. Yuan, H. Liu, J. Liu, H. Li, W. XiePSPICE simulations. Inductance of the switch assembly is estimated by Institute of Fluid Physics, China Academy of Engineering Physics,this comparison. Mianyang, China Ceramic Packaging Reliability Study of a 13.5 kV5C-8: In order to investigate on-state resistance of 6H-SiC photoconductiveMultichip Thyristor semiconductor switches (PCSSs), two kinds of 6H-SiC materials with 1 1 1 1B. Vergne , C. Gauthier-Blum , V. Brommer , S. Scharnholz , different dark resistivity and process have been used to fabricate theE. Spahn1, A. Welleman2 PCSSs. The absorption depths of 6H-SiC materials with different1 French-German Research Institute of Saint-Louis (ISL), Saint-Louis, wavelengths and dark current-voltage characteristics of 6H-SiC PCSSFrance were obtained experimentally. Triggered by a laser pulse with varying2 ABB Switzerland Ltd, Semiconductors, Lenzburg, Switzerland optical energies at wavelengths of 1064 nm, 532 nm and 355 nm, the photoconductivity tests of the switches were performed at different biasSince its first announcement in 2009 [1], ISL and ABB Semiconductors voltages. The experimental results show that the SiC PCSS operates inare working together on the steady improvement of the commercial linear mode. The switch made from 6H-SiC with a dark resistivity of 120
  • 120. 6×10^5 Ω•cm can only sustain an operating voltage of 4.2 kV, and its 3P-2: The High-Intensity Leopard Laser for High Energyminimum on-state resistances can achieve tens of ohms, which are Density Physics Research in Nevada Terawatt Facilityabout two orders of magnitude higher than that of GaAs PCSS. Another P. P. Wiewior, A. Covingtonswitch made from 6H-SiC with a dark resistivity of 2×10^10 Ω•cm can Nevada Terawatt Facility, University of Nevada-Reno, Reno, NV,sustain an operating voltage of 30 kV, but its minimum on-state United Statesresistances can be as high as several kilo ohms, which are totallyunacceptable for pulsed power system. The numerical analyses arepresented and compared with experimental results, and the influencing A 50 TW-class laser (Leopard) was developed for simultaneousfactors of the minimum on-state resistance are discussed, which shows experiments with the 1 MA z-pinch generator (Zebra) at the Universitythat increasing the optical energy of incident laser, carrier mobility and of Nevada, Reno. The Leopard laser was built with CPA technologycarrier lifetime can reduce the minimum on-state resistance. From the and is based on a femtosecond Ti:Sapphire front-end. The stretched fscomparison of different 6H-SiC PCSSs, several methods to reduce on- pulses are intensified by a series of optical amplifiers including astate resistance are proposed, and new switches will be fabricated and Ti:Sapphire regenerative amplifier and a series of Nd:glass rod andtested as soon as possible. disks amplifiers. The laser can produce the short 0.35 ps pulses with energy of 15 J and the long 1 ns pulses with energy of 30 J. The output beam diameter is 80 mm. An adaptive optic system improves the3P: Radiation Sources posters focusing ability of the laser and increases its repetition rate. Two beamWednesday, June 22 13:30-15:30 Regency Ballroom paths are available for experiments in the vacuum chamber of the Zebra generator and in the separate vacuum chamber for laser-plasma experiments. When coupled to the Zebra z-pinch, the Leopard laser is a3P-1: Xenon Theta Pinch for ICF Chamber Environment powerful diagnostic tool which can be used to probe the spatial andExperiments temporal properties of dense plasmas. An outline the status, design,M. A. Rhodes, J. Kane, G. Loosmore, J. Latkowski architecture and parameters of the Leopard laser will be presentedLawrence Livermore National Laboratory, Livermore, CA, United along with recent results of coupled laser Z-pinch experiments.States Work was supported by the DOE/NNSA under UNR grant DE-FC52- 06NA27616ICF power plants, such as the LIFE scheme at LLNL, will likelyemploy a high-Z, target-chamber gas-fill (such as Xenon) to moderate 3P-3: Concept Design of Z-Pinch Accelerator for ICFthe first-wall heat-pulse due to x-rays and energetic ions released M. Wang, W. Zou, L. Chen, L. Zhou, W. Xieduring target detonation. Modeling of this gas-fill with radiation Pulsed Power Lab., Institute of Fluid Physics, China Academy ofhydrodynamic codes predicts that the gas-fill initially is rapidly (10eV) Engineering Physics, Mianyang,Sichuan, Chinaand highly ionized (Z>10) by x-rays and ions released from the fusiontarget. The gas-fill then cools to 1-2 eV on a 100-ns timescale. Coolingrates below 1-eV are uncertain because the available codes do not Based on new understanding of physical requirement for inertial-accurately handle the complex atomic physics of high-Z gases at confinement-fusion (ICF) by Z-pinch, the target value of acceleratortemperatures below 1-eV. However, our modeling suggests that gas optimize design should be array kinetic energy but x-ray energy, andcooling will “stall” in the range 0.5-1 eV because the Xenon has few the array kinetic energy should exceed 5MJ/cm for Z-ICF. Large arrayradiation states in this temperature range. We are interested in the gas- diameter and lower pinch velocity is better for fusion target. Forfill state at very long timescales (70-ms) because the expected optimized load, current rise-time expands to several hundreds ofrepetition rate in the baseline LIFE design is currently 15-Hz. Residual nanosecond contributes less to MHD instability. At the same time, onplasma and hot-gas in the target chamber from the previous shot can concerning A-K gap closure and energy loss to electrode of MITL,have deleterious effects on subsequent laser pulses and injected targets. there exist a physical window of about 200 ns rise time for Z-pinchTo reduce the uncertainties of cooling and beam/target propagation fusion driver design. Aim at the accelerator with peak output current ofthrough such gas-filled chambers, we have developed a pulsed plasma about 60 MA and rise-time (10%~90%) of about 150 ns, thesource designed to produce a 2-5 eV Xenon plasma at a density of advantages and disadvantages of several possible pathes are compared.2E16 particles/cc. In this type of plasma source, the resulting xenon Based on scheme of LTD with long MITL, designed a standard LTDtemperature is a balance between the heating power-in vs. the radiated module with output current of 1.44 MA and rise-time of about 12