TU4005
Design of a Sub-Picosecond, High Charge Electron Linac*

H. BLUEM, C. PAULSON, M. REUSCH, A. TODD, Northrop Grumman Corp.

A photocathode-based electron linear accelerator, and associated transport line, which is capable of delivering sub-picosecond electron pulses of up to 3 nC has been designed. The gun is 2.5 cells with a magnesium photocathode, followed by a four cell booster accelerator. These are designed to produce an electron bunch at a nominal energy of 15 MeV with a longitudinal phase space which is optimized for magnetic compression inside a 90 degree achromatic bend. The predicted performance of this beamline (obtained through PIC simulations and a nonlinear beam transport code) is a charge of > 2 nC in a pulse width of < 0.7 ps. This same linac beamline is also designed to deliver in excess of 90 nC in somewhat longer pulses. We will present the results of our simulations and the detailed design of this linac and transport system.

*This work was supported by Argonne National Laboratory under contract # 972582401.

TU4006
Development of an Upgrade of the CEBAF Acceleration System*

J.R. DELAYEN, TJNAF

Long-term plans for CEBAF at Jefferson Lab call for achieving 12 GeV in the middle of the next decade and 24 GeV after 2010. Such energies can be achieved within the existing footprint by fully populating the accelerator with cryomodules capable of providing 3 to 4 times as much voltage as the design value of the existing within the same length. In particular, this requires for the development superconducting cavities capable of operating at gradients above 12 MV/m and Q close to 10¹⁰. An R&D program for the development of the cryomodules is under way and will be presented, as well as various options for the upgrade path.

* Work supported by the U.S. DOE Contract # DE-AC05-84ER40150

TU4007
Latest Developments from the S-DALINAC and its Free- Electron- Laser *

M. BRUNKEN, S. DÖBERT, R. EICHHORN, H. GENZ, H.-D. GRÄF, T. HAMPEL, S. KOSTIAL, U. LAIER, H. LOOS, A. RICHTER, B. SCHWEIZER, A. STASCHECK, O. TITZE, T. WESP, Institute für Kernphysik Tu-Darmstadt

The FEL at the superconducting electron accelerator S-DALINAC has been operated at wavelengths between 6.6 and 7.8 µm corresponding to electron beam energies between 31.5 and 29.6 MeV respectively. In house developments like active stabilization of the optical cavity's length based on a laser interferometer and a fast 10 channel on-line-spectrometer for the near infrared laser radiation were incorporated and used for investigation of the properties of the FEL itself. In addition a simple focusing system produced power densities of 10⁸ W/cm² and enables us to perform first ablation experiments from different tissues. To optimize the complicated bunching process, the low energy part of the S-DALINAC was carefully studied by tracking calculations in order to achieve bunch lengths < 3 ps corresponding to peak currents > 1.5 A, necessary to operate the FEL reliably in the saturated regime. For operation at higher energies using two recirculations of the electron beam, longitudinal tracking simulations show that off crest acceleration together with nonisochronous recirculation introduce longitudinal stability resulting in significant improvement in energy spread. As some of the sc cavities showed Q degradation due to organic contamination, eight cavities were cleaned by a method involving organic solvents and buffered chemical polishing. During the same shutdown the beam transport to the experimental areas was redesigned maximizing its acceptance and rebuilt.

* Supported by BMBF under contract number 06 DA 820 and by DFG under contract number RI 242/12-1.

TU4008

Contribution withdrawn.

TU4009
1GeV Linac at Fermilab*

M.B. POPOVIC, A. MORETTI, C.W. SCHMIDT, R. NOBLE, FNAL

A proposed linac injector for a new proton source complex at Fermilab is assumed to have a kinetic energy of 1 GeV. This linac would be sized to accelerate ~100 mA of H^- beam in a ~200 microsecond pulse at a 15 Hz repetition rate. This would be adequate to produce ~1.0E14 protons per pulse allowing for future increases in antiproton neutrino prodution as well as providing a source for a possible muon collider. An alternate proposal is to add 600 MeV of side coupled cavity linac at 805 MHz to the existing 400 MeV Linac. This addition may either be in a new location or use the present Booster tunnel. A discussion of these possibilities will be given.

*Work supported by the U.S. Dept. of Energy under contract No. DE-AC02-76CH03000.

TU4010

Contribution withdrawn.

TU4011  (oral poster TU3001)
Design of the 200-MeV Proton Linac for the Japan Hadron Facility

T. KATO, Y. YAMAZAKI, KEK

A 200-MeV proton linear accelerator for the JHF has been designed. A peak current of 30 mA with 500 micro-sec pulse duration will be accelerated at a repetition rate of 25 Hz. The linac consists of a 3-MeV radio-frequency quadrupole linac (RFQ), a beam-transport line from the RFQ to the DTL (MEBT), a 50-MeV drift tube linac (DTL) and a 200-MeV separated-type drift tube linac (SDTL). A frequency of 324 MHz has been chosen for all of the rf structures. A future upgrade plan of up to 400 MeV is also considered, in which annular-coupled structures (ACS) of 972 MHz will be used in an energy range of above 150 or 200 MeV. There are three distinct features in the design. The first is stable operation with high performance for a beam-loss problem. The second is a total high shunt impedance, achieved by adopting the SDTL structure. The third is the adoption of klystrons of 2.5-MW peak output power rather than conventional 200-MHz triodes. The details of the design, including the results of beam simulation studies using two kinds of focusing methods (equipartitioning focusing and constant phase-advance focusing in transverse motion) are presented.

TU4012
Development of a Commissioning Plan for the APT Linac*

L. W. FUNK, WSRC; K. R. CRANDALL, Techsource; J. D. GILPATRICK, E. R. GRAY, L. J. RYBARCYK, T. P. WANGLER, LANL

The APT proton linac will have an average beam power of 170 MW, which exceeds by two orders of magnitude that of the next largest proton linac. Safe, successful and prompt commissioning of this accelerator will be the culmination of a years-long series of activities which will include: ensuring that the design incorporates those features necessary for a successful demonstration of technical performance; a careful and tightly integrated program of subsystem acceptance testing and installation; thorough planning to ensure the timely availability of infrastructure, resources and trained staff; and a comprehensive test plan. All of these activities require as a starting point a complete identification of all performance characteristics that must be measured to confirm that the linac meets its design objectives. This paper will present our present understanding of the technical objectives of the commissioning program, review the impact of that understanding on the design of the accelerator, and summarize the issues which remain to be addressed and the plan to address them.

* Work supported by the U.S. Department of Energy

TU4013  (oral poster TU3002)
A Linac for the Spallation Neutron Source*

A. J. JASON and the SNS LINAC Design Team, LANL

The Spallation Neutron Source Project (SNS) accelerates H^- ions to an energy of 1.0 GeV with an average current of 1-mA for injection in an accumulator ring that produces the short intense burst of protons needed for the spallation-neutron source. The linac will be the most intense source of H^- ions and as such requires advanced design techniques to meet project technical goals as well as to minimize cost. In particular, low beam loss is stressed for the chopped beam, placing strong requirements on the beam dynamics and linac construction. Additionally, the linac is to be upgraded to the 2- and 4-MW beam-power levels with no increase in duty factor. We give an overview of the linac design parameters and discuss the special challenges and status of the effort.

* Work supported by the Office of Energy Research, Basic Energy Sciences, of the US Department of Energy; and the Oak Ridge National Laboratory.

TU4014
Status Update on the Low-Energy Demonstration Accelerator (LEDA)*

H. V. SMITH, JR., J. D. SCHNEIDER, LANL

As part of the linac design for the accelerator production of tritium (APT) project, we are assembling the first 10-MeV portion of this cw, 100-mA proton accelerator. The primary objective of this low-energy demonstration accelerator (LEDA) is to verify the design codes, gain fabrication knowledge, understand beam operation, and improve prediction of costs and operational availability for the full 1000- to 1700-MeV APT accelerator. This paper provides an update on this past year's progress that includes extensive beam tests of the LEDA injector using the Chalk River Injector Test Stand (CRITS) radio-frequency quadrupole (RFQ) accelerator, which results in production of a cw, 100-mA, 1.25-MeV RFQ output beam; fabrication, assembly, tuning, and installation of the 6.7-MeV LEDA RFQ; upgrades to the facility; installation and testing of the 350-MHz RF system; installation of the HEBT; and installation (as well as prototype testing) of the LEDA ogive beam stop. First tests with the 6.7-MeV, 100-mA, cw beam from the RFQ are scheduled for late fall. References will be given to many detailed papers on LEDA at this conference.

* Work supported by the US DOE, Defense Programs.

TU4015
Beam Dynamics Design of the 211 MEV APT Normal Conducting LINAC*

L. M. YOUNG, J. H. BILLEN, H. TAKEDA, R. L. WOOD, LANL

This paper presents the design of the 6.7 to 211 MeV Normal Conducting Linac for the Accelerator for Production of Tritium (APT) project. This linac was designed with the new version of the PARMILA [1]. This linac accepts the beam from the 6.7 MeV radio frequency quadrupole without a separate matching section. At about 10 MeV, it has a smooth transition in the length of period from 8 to 9 in quadrupole focusing lattice. This adjustment of the period was needed to provide sufficient space for the quadrupole focusing magnets and beam diagnostic equipment. The linac consists of the coupled cavity drift tube linac [2] up to 100 MeV and coupled cavity linac above 100 MeV.

*Work supported by the U. S. Department of Energy.

Classification Category: A02

TU4016  (oral poster TU3003)
First Results of the Two-Beam Funneling Experiment*

A. FIRJAHN-ANDERSCH, A. SCHEMPP, J. THIBUS, E. WINSCHUH, H. ZIMMERMANN, IAP-Univ. of Frankfurt

High intensity accelerator concepts for Heavy Ion Inertial Fusion (HIIF) injectors require small emittance, high current and high energy beams. The improvement of brightness in such a driver linac is done by several funneling stages at low energies, in which two identically bunched ion beams are combined into a single beam with twice the frequency, current and brightness. For the Heavy-Ion Driven Ignition Facility (HIDIF) we have proposed the use of an two-beam accelerator structure which provides two beams within one cavity and a single r.f. deflector structure which bends the two beams to one common axis. The progress of the experiment and first beam-test results will be presented.

* This work is supported by the BMBF

TU4017  (oral poster TU3004)
Present Performance of the CERN Proton Linac

C.E. HILL, A. LOMBARDI, E. TANKE, M. VRETENAR, CERN

The original 1973 design specification of the CERN 50 MeV Proton Linac was for a 150 mA beam but this intensity was rarely used. Preliminary tests for the high brightness beam required for LHC indicated that 170 mA could be produced for short pulses (30 µs). Since then further optimisation has enabled the 170 mA to be delivered reliably, within the nominal emittances and dispersion, in long pulses (120 µs) to the user (PS Booster, about 80 m downstream of the linac). The improvements will be described along with the steps envisaged to attain a goal of more than 180 mA.

TU4018

Contribution withdrawn.

TU4019
Fermilab Linac Operations, Studies and Improvements*

L.J. ALLEN, C.W. SCHMIDT, M.B. POPOVIC, FNAL

The Fermilab Linac operated as an injector for the Tevatron Fixed Target Program, Booster studies and for neutron production for cancer treatment during 1997 and 1998. Operational reliability was 98% of the scheduled operating time. The beam current was generally in the 45 to 48 mA range. After completion of the Fixed Target Program, the Linac embarked on a series of studies. The studies were intended to verify visual survey data indicating a misalignment of the low energy linac, correct the misalignment, and check the linac RF phase and amplitude settings. Operations data and studies results will be presented.

*Work supported by the U.S. Department of Energy under contract No. DE-AC02-76CH03000.

TU4020
Moscow Meson Factory Linac - Operation and Improvements

S. K. ESIN, L. V. KRAVCHUK, P. N. OSTROUMOV, V. L. SEROV, INR RAS

During 1997-1998 the Moscow Meson Factory Linac (MMFL) provided 3800 beam hours for experiments, isotope production, and systems reliability upgrades. The new beam diagnostics computer controls exert a very important impact on the beam quality and operational efficiency of the linac. A crucial experiment was conducted that revealed the cause of the discrepancy between the calculated and the measured capture efficiency. A safe triple protection circuit was introduced after a hole was accidently burned in the isotope production channel beam pipe. A new injection scheme implementation is underway which must increase the average beam current from the recent 70 mcA to about 200 mcA. During the March 1998 shift the MMFL worked for the Pulsed Neutron Source for the first time.

TU4021
Status of a New Switchyard Design for LANSCE*

R. W. GARNETT, E. BJORKLUND, M. BORDEN, C. ROSE, F. SHELLEY, J. D. ZUMBRO, LANL

The details of a new design for a section of the LANSCE 800 MeV beam switchyard will be discussed. At present, the switchyard is used to deliver a proton beam to experimental Area-A and an H^--ion beam down Line-D. The total H^- repetition rate is 120 Hz. Of that 120 Hz, 100 Hz is transported to the Weapons Neutron Research (WNR) area. The remaining 20 Hz is injected into the Proton Storage Ring (PSR). In order to provide H^- beam to other experimental areas without interfering with the PSR operations, a new design of the switchyard was completed. The new design incorporates the use of pulsed kicker magnets to deflect a fraction of the H^- beam, normally delivered to the WNR area, down Line-C at the demand of the experimenters.

*Work supported by the U. S. Department of Energy Contract W-7405-ENG-36.

TU4022
Understanding Accelerator Reliability

C. PIASZCZYK, Northrop Grumman Corp.

Over the period of several years, the author studied reliability of both planned and existing accelerator systems. This involved collection and analysis of data gathered at various operating accelerator facilities and mathematical modeling of accelerator design concepts. Although the surveyed accelerator facilities are complex and different in details, they all share common operational characteristics. Moreover, they typically consist of just a few typical components: ion sources, focusing magnets, magnet power supplies, RF windows, circulators, rf amplifiers, high voltage components, etc. It should be therefore possible to use the accumulated operational experience to optimize the envisioned future concepts and to extend the operating performance of existing machines. In the process, accelerators can serve as a testbed for developing an understanding of the reliability of repairable systems.

Classification Category: A05

TU4023
Status of 200 MeV Electron LINAC and Its Applications

Y.J. PEI, Y.F. WANG, S. DONG, X.F. LUO, G.R. HUANG, G.C. WANG, W. WANG, Y.X. LI, University of Science & Technology of China

The 200 MeV Electron LINAC is an injector of Electron Storage Ring which was the first dedicated synchrotron Radiation facility in China. The LINAC has been running well for ten years, since 1987. Typical operation parameters are energy of 200 MeV, current of 50 mA and energy spread of 0.8%. Some improvements are described in this paper, such as developing an energy stabilized system, adding a post-transport line in order to guide the beam into a nuclear physics experimental hall, developing a two mode combined pulser for the electron gun to meet single-bunch and multi-bunch mode running in the storage ring. Some applications are briefly described here also. Some research results are given, such as isotopes products: 123I, 11C, 15O, 18F etc.

TU4024
Status of PLS 2 GeV Electron Linac Performance*

M.H. CHO, K.R. KIM, J.S. OH, S.H. PARK, S.S. PARK, I.S. KO, W. NAMKUNG, PAL-POSTECH

The 2 GeV electron linac at the Pohang Accelerator Laboratory (PAL) has been operated continuously as a full energy injector for the Pohang Light Source (PLS) since its completion on Dec. 7, 1994. There has been continuous effort to make the linac system more stable and reliable, such as the modifications of the klystron modulators, temperature stabilization of the main rf drive line, installation of one more klystron-modulator station with the addition of two sets of acceleration sections. The average high voltage operation time of the klystrons (E-3712, Toshiba) installed at the very beginning (8 tubes are survived out of total 11 tubes) has reached near 32,000 hours as of May 1998. Current overall system availability is over 90%. In this paper, we report the major linac system upgrade for the nominal 2.5 GeV operation as well as the system performance and relevant machine statistics such as lifetime of klystrons and thyratrons, and overall system's availability, and others.

* Work supported by POSCO and MOST, Korea.

TU4025  (oral poster TU3005)
Achieving 800kW CW Beam Power and Continuing Energy Improvements in CEBAF*

C.E. REECE, TJNAF

During the past year, CEBAF at Jefferson Lab has demonstrated its full capacity of sustained 800 kW beam power. All systems performed as intended. The energy stability at the design parameters of 4.0 GeV, 200 microamps CW beam was measured to be better than 3E-5 rms. During the Fall of 1997, physics experiments were conducted using 4.4 GeV beam. Having demonstrated the benefits of in situ helium/rf processing of SRF cavities for increasing the energy reach of CEBAF, a program of processing all installed cryomodules has begun. This processing has proven effective against the principal chief gradient limitation of the SRF cavities in CEBAF: discharges at the cold rf waveguide window, induced by electron field emission in the cavities. Approximately half of the cavities are limited by such effects. Sustained operation above 5.0 GeV has been scheduled, and we are preparing for physics runs above 5.5 GeV. Successes and difficulties encountered on this upgrade path will be presented.

*Work supported by the U.S. Department of Energy, contract DE-AC05-84ER40150.

TU4026
Stability of the CLIC Drive Beam Accelerator

D. SCHULTE, CERN
(R. Ruth, presenter)

The drive beam of the Compact Linear Collider (CLIC) needs to have a current of several amperes. The time structure of this beam is discussed. First results on longitudinal single bunch effects are presented and achievable bunch lengths and sensitivity to jitter of the gradient, initial energy and charge are analysed. The transverse stability of the beam is discussed based on the present cavity model. The requirements on the damping and detuning of the cavities are detailed. A beam-based alignment technique is presented and the stability with respect to jitter and ground motion is investigated.

TU4027
High Current Transport and Acceleration at the Upgraded UNILAC

W. BARTH, J. GLATZ, J. KLABUNDE, GSI Darmstadt

The UNILAC will be upgraded as a high current injector for SIS. The Wideroe section will be replaced by an RFQ-IH-type accelerator. This paper focuses on beam dynamics studies in the stripper section, the poststripper Alvarez accelerator and the beam transport line to SIS including a second stripper section. The beam dynamics investigations are faced with considerable space charge effects along the UNILAC. Most severe space charge forces occur after the charge state jump in the gas stripper at 1.4 MeV/u. A new design of the charge separation system will be presented. Furthermore, simulation results for matching to the Alvarez linac, emittance growth effects in the poststripper linac, equipartitioning and longitudinal mismatching will be discussed. The beam transfer to SIS will include a modified foil stripper and charge separation system, the beam brilliance loss due to foil straggling and space charge forces will be calculated. Beam experiments at the present UNILAC with high current beams supported the design studies, experimental results will be reported.

TU4028

Contribution withdrawn.

B.R. POOLE, G.J. CAPORASO, Y. J. CHEN, LLNL

A fast stripline beam kicker and septum are used to dynamically switch a high current electron beam between two beam lines. The transport of the beam through these structures is determined by the quality of the applied electromagnetic fields as well as temporal beam induced effects due to the wakefields produced by the beam. In addition, nonlinear forces in the structure will lead to emittance growth. The effect of these issues is investigated analytically and by using particle transport codes. Due to the distributed nature of the beam-induced effects, multiple macro-particles (slices) are used in the particle transport code, where each slice consists of an ensemble of particles with an initial distribution in phase space. Changes in the multipole moments of an individual slice establish electromagnetic wakes in the structure and are allowed to interact with subsequent beam macro-particles to determine the variation of the steering, focusing and emittance growth during the beam pulse.

*This work was performed under the auspices of the U.S. Department of Energy by the Lawrence Livermore National Laboratory under contract No. W-7405-Eng-48

TU4030
Higher-Order-Modes and Beam Breakup Simulations in the Jefferson Lab FEL Recirculating Linac*

L. MERMINGA, I. CAMPISI, TJNAF

The Jefferson Lab FEL driver accelerator is a recirculating, energy-recovering superconducting linac designed to attain a CW beam current of 5 mA. Measurements of the external Q's of the TE₁₁₁ and TM₁₁₀ dipole passbands of all 10 superconducting cavities yield values much higher than anticipated for some of the TE₁₁₁ resonances. These enhanced Q's are related to strong polarization of the modes along and orthogonal to the fundamental power coupler. The reasons of such polarization will be discussed together with solution for this potentially harmful problem. Threshold currents for the multipass beam breakup (BBU) instability were calculated to evaluate the safety margin available in the operation of the accelerator. TDBBU simulations were done with measured HOM parameters and the threshold current was determined to be 27 mA. The dominant contribution to the instability comes from a few high Q resonances of the y-polarization of the 4/5 TE₁₁₁ mode. If these modes were adequately damped--which might be possible in the future with minor hardware changes--then simulations predict that the threshold current would increase by almost an order of magnitude. The results of these measurements and simulations, as well as an analytical treatment of the BBU during energy recovery, are presented.

* Work supported by the U.S. DOE Contract # DE-AC05-84ER40150.

TU4031
TESLA Injektor Simulations with MAFIA

P. SCHUETT*, T. WEILAND, TU-Darmstadt

The TESLA Test Facility (TTF) is a test bed for the superconducting option of a linear e⁺/e^- collider in the TeV regime. The injector delivers a 20 MeV beam with 8 nC charge per bunch of 1 mm length. It consists of an RF-gun, a superconducting 9-cell Capture Cavity and a chicane bunch compressor. The MAFIA modules TS2 and TS3, the particle in cell programs in 2D and 3D, as well as the module L, a tracking program, have been used to study the particle dynamics in the TTF, e.g. the emittance growth due to space charge effects in the bunch compressors. Dedicated interfaces for the three different simulation tools allow to study components separately and to hand over beam data from one accelerator section to another or one simulation tool to another respectively.

* Work supported partly by DFG (Deutsche Forschungsgemeinschaft).

TU4032
Halo Formation in 3-D Bunches with Different Phase Space Distributions

A.V. FEDOTOV, R.L. GLUCKSTERN, Univ. of Maryland; S.S. KURENNOY, R.D. RYNE, LANL

TU4033
Study of the Transverse Beam Motion in the DARHT Phase II Accelerator*

Y.-J. CHEN, T. L. HOUCK, LLNL; W. M. FAWLEY, LBNL

The accelerator for the second-axis of the Dural Axis Radiographic Hydrodynamic Test (DARHT) facility will accelerate a 4 kA, 3.5 MeV, 2 ms long electron current pulse to 20 MeV. The energy variation of the beam within the flat-top portion of the current pulse is ± 0.5%. The performance of the DARHT Phase II radiographic machine requires the transverse beam motion to be much less than the beam spot size which is about 1.5 mm diameter on the x-ray converter. In general, the leading causes of the transverse beam motion in an accelerator are the beam breakup instability and the corkscrew motion. We have modeled the transverse beam motion in the DARHT Phase II accelerator with various magnetic tunes and accelerator cell configurations by using the BREAKUP code. The predicted sensitivity of corkscrew motion and BBU growth to different tuning algorithms will be presented.

* The work was performed under the auspices of the U.S. Department of Energy by LLNL under contract W-7405-ENG-48, and by LBNL under contract AC03-76SF00098.

TU4034  (oral poster TU3006)
Controlling Backstreaming Ions from X-ray Converter Targets with Time Varying Final Focusing Solenoidal Lens and Beam Energy Variation*

Y.-J. CHEN, A. C. PAUL, LLNL

Performance of high resolution x-ray radiography facilities requires high current electron beams to be focused to a millimeter spot size on an x-ray converter throughout the entire current pulse. Within a single pulse, the heating of the target by the electron beam will lead to rapid desorption of surface contaminants. The space charge field of the electron beam will pull ions out of this plasma layer into the beam. These backstreaming ions can form an ion focusing channel and overfocus the electron beam. As the ions move upstream in time, the net focusing strength increases. The final beam spot size on the target would then change in time and typically be larger than intended. We have studied the possibility of mantaining a constant final spot size for the entire current pulse by using either a time varying final focusing solenoidal field or beam energy variation to compensate the time varying ion focusing effects. However, the required time varying rate for both the solenoidal field and the energy variation are large. The predicted final focusing with various beam parameters will be presented.

* The work was performed under the auspices of the U.S. Department of Energy by LLNL under contract W-7405-ENG-48.

TU4035
EM-PIC Simulations of e-Beam Interaction with Field Emitted Ions from Bremsstrahlung Targets *

P. W. RAMBO, S. BRANDON, LLNL

We investigate electron beam defocusing caused by field emitted ions from the bremsstrahlung target of a radiography machine using fully electromagnetic particle-in-cell simulations. This possibly deleterious effect is relevant to both current radiography machines (FXR) and machines being planned (DARHT and AHF). A simple theory of the acceleration of ions desorbed from the heated target, and subsequent beam defocusing due to partial charge neutralization is in reasonable agreement with the more detailed simulations. For parameters corresponding to the FXR machine (I=2.3 kA, E=16 MeV), simulations assuming space-charge-limited emission of protons predict prompt beam defocusing. Time integrated spot-size measurement, however, is dominated by early-time small spot brightness, and so is not a sensitive diagnostic. Comparisons are made to available FXR data. We also investigate use of recessed target geometries to mitigate field emitted ion acceleration; only modest improvements are predicted.

* This work was performed under the auspices of the U. S. Department of Energy by Lawrence Livermore National Laboratory under Contract No. W-7405-Eng-48.

TU4036

Contribution withdrawn.

TU4037

Contribution withdrawn.

TU4038
Model-Independent Analysis of Beam Dynamics*

J. IRWIN, C.X. WANG, Y. YAN, K. BANE, Y. CAI, F.J. DECKER, M. MINTY, G. STUPAKOV, SLAC

In this paper, we report a novel method to analyze the beam dynamics based on BPM readings. This analysis is independent of any machine models. Instead, it relies on the ensemble of readings from a large number of BPMs for a large number of pulses. In addition, pulse-by-pulse information of beam parameters and machine status are used. By taking advantage of the spatial correlation and temporal structure of all these signals, our analysis allows observation of beam dynamics at a level far beyond the single BPM resolution. Further more, unexpected beam dynamics as well as machine information can be revealed. Both simulation and experimental results from SLC will be presented.

*Work supported by the Department of Energy under Contract No. DE-AC03-76SF00515

TU4039
Multiple Coupling and Beam Loading of a RF Cavity

H. SAFA*, LANL and CEA/Saclay

Future high power accelerators are aiming at increasing the power transferred to the beam through radiofrequency (RF) cavities. Consequently, multiple main coupler drives may be required to withstand the high RF power needed per cavity. An analysis including multiple couplers and beam loading is described here, featuring the interaction between all couplers, the beam and the cavity. In this description, the beam is shown to act as if it were merely an additional drive for the cavity. Special focus is given to the case of superconducting cavities in a continuous (CW) operation mode. Some important conclusions derived from this analysis are briefly discussed as, for example, power coupler fault conditions or a way to handle the commissioning of an accelerator in a CW mode.

* Work performed while on assignment at Los Alamos National Laboratory.

TU4040
Analytical Treatment of Single Bunch Stability in a Linac

G. GUIGNARD, J. HAGEL, CERN

Single bunch stability is analysed by solving the equation of motion of the particles travelling in a linac, for a Gaussian distribution of charge, a linear variation of the transverse wakefield along the bunch, a smooth focusing and negligible acceleration. The treatment is based on a non-standard perturbation expansion that has been specifically developed for this study and preserves at each order the intrinsic detuning likely to stabilise the resonant break-up. It provides a closed expression for the tune shift along the bunch resulting from BNS damping and autophasing, methods proposed in the past to control the emittance, and a first order solution for the transverse off-sets within the bunch. The analytic result obtained makes it possible to study the behaviour of the solution and compute the emittance dilution in specific cases. The present theory is a useful complement to the numerical simulations done with the MUSTAFA code for the Compact LInear Collider scheme (CLIC). It also gives an interesting as well as comprehensive view of the physics involved in the single bunch motion and the damping of the instability.

TU4041
Application of the Mode-Matching Technique to the Analysis of Waveguide Arrays

A. JOESTINGMEIER, M. DOHLUS, N. HOLTKAMP, DESY

In this contribution the mode-matching technique is applied to compute the absorption characteristics of a two-dimensional array of rectangular waveguides. This analysis is motivated by a proposal of a broadband absorber for TESLA. Outside the waveguide array the so-called Rayleigh expansion is used which represents the electromagnetic field in terms of space harmonics whereas the complete modal spectrum of eigenmodes is taken into account inside the waveguides. In the case of normal incidence the validity of the presented method is confirmed by MAFIA computations. The absorption characteristics for various angles of incidence are calculated for a broad frequency range. The accuracy of the results is checked by a study of convergence. It is shown that the Rayleigh expansion has to be modified if the waveguide array is excited by an ultrarelativistic beam instead of an incoming plane wave. For an array of parallel-plate waveguides numerical results (beam impedance, wakefield, loss parameter) are presented and compared with those obtained by other methods.

TU4042
Simulation Study of the Bunching Section of Xband Klystrons

S. MATSUMOTO, S. MICHIZONO, H. TSUTSUI, Y.H. CHIN, S. FUKUDA, KEK

A precise simulation of a klystron is useful to understand its operation and preferable for its design work. For this purpose, MAGIC code, which is a 2.5 dimensional fully electromagnetic and relativistic PIC code for self-consistent simulation of plasma physics problems, is used. Simulation works for the production of the beam, the modulated beam motion and the beam-cavity interaction at the output cavity as well as the gain cavities can be done by this code. In this article, the simulation of the buncher section of a relativistic klystron, such as KEK XB72K or SLAC XL-4, is reported. The process of the modulation on the beam motion and the evolution of rf current is observed.

TU4043  (oral poster TU3007)
Electron Gun Simulation Using MAGIC

S. MICHIZONO, H.TSUTSUI, S. MATSUMOTO, Y. H. CHIN, S. FUKUDA, KEK

In a future linear collider such as Japan Linear Collider (JLC) and SLAC NLC, high power klystrons at X-band (11.424 GHz) will be used. In these klystrons, the beam (>350A) should be compressed to around the radius of 3 mm and the area-convergence ratio of the beam is more than 100. In such a high beam intensity, the precise beam trajectory is analyzed by the suitable gun simulation.

In this report, electron-gun regions of some klystrons (S- and X-band) are simulated by MAGIC, which is the 2.5-D fully electromagnetic and relativistic particle-in-cell (PIC) code for self-consistent simulation of plasma physics problems. Simulation results indicate that the profile of the current density is almost same to the EGUN and the perveance is larger than the EGUN predicted value and almost same to the measured value. The particle data (including momentum) from the electron gun are exported to the next rf interaction region and the whole structure of the klystron can be simulated by MAGIC.

TU4044
High Energy Beam Transport Beamline for LEDA*

W. LYSENKO, LANL; M. SCHULZE, General Atomics

This paper describes the High Energy Beam Transport (HEBT) beamline for the Low Energy Demonstration Accelerator (LEDA), which is part of the Accelerator Production of Tritium (APT) project. Space-charge effects are important in this 6.7 MeV 100 mA CW beam. Simulations are presented showing that maintaining strong focusing (no large magnet separations) and minimizing beam size variations (avoiding rapid increases in the charge density) in the upstream part of the beamline are important for avoiding halo formation. The beamline design was obtained with the help of the TRACE 3-D design code and the beamline properties were verified with different 2-D and 3-D particle-in-cell (PIC) simulation codes. Comparisons show that 3-D effects are important and that TRACE 3-D predicts rms beam properties well. The PIC simulations are important for determining the presence of beam halo, which we define as a long tail in the charge distribution. The PIC simulations make clear predictions concerning the presence or absence of beam halo as a function of the HEBT tune. HEBT experiments at LEDA are proposed to verify the accuracy of the simulation codes' ability to predict halo.

*Work supported by the U.S. DOE, Defense Programs.

TU4045
Beam Dynamics Simulation of the Spallation Neutron Source Linear Accelerator *

H. TAKEDA, T. BHATIA, J. H. BILLEN, LANL

The accelerating structure for Spallation Neutron Source (SNS) consists of a drift-tube linac (DTL) from 2.5 to 20 MeV, a coupled-cavity drift-tube linac (CCDTL) to 96 MeV, followed by a conventional coupled-cavity linac (CCL) to 1 GeV. The CCDTL and CCL have regularly spaced drift lengths between segments of accelerating structure for the quadrupole focusing magnets and diagnostics. Across transitions between structure types we maintain continuous average focusing strength per unit length, both transversely and longitudinally. We discuss details of the transverse and longitudinal matching between the 402.5-MHz DTL and the 805-MHz CCDTL. We present end-to-end multiparticle simulations of the beam emerging from the RFQ through the rest of the linac. High cavity stored energy helps to reduce the amplitude of cavity field variations caused by chopping the beam. We discuss the beam characteristics along the linac at the design field level and at the peaks and valleys of the field over a chopping period.

* This research is sponsored by the Division of Materials Sciences, DOE, under contract DE-AC05-96OR22464 with Lockheed Martin Energy Research Corp

TU4046
Using of LIDOS.RFQ Codes for CW RFQ Designing

B. BONDAREV, A. DURKIN, S. VINOGRADOV, MRTI; J.-M. LAGNIEL, R. N FERDINAND, CEA/DSM/DAPNIA/SEA

RFQ linac with CW current ~ 100 mA and proton output energy 5-7 MeV can be used as beginning part of linac for accelerator-driven transmutation technologies and applications. There are several restrictions using such an RFQ linac. The radiation purity criterion (minimum of particle lost) is the main requirement. The RFQ channel must have high transmission (>95%), low RF field intensity (no more then 1.7 Kilpatrick) and a minimum of particles lost with energies above 3 MeV. It is also very important to study the influence of input beam parameter deviation on beam transmission as well as influence of RFQ channel parameter deviation. The three level code package LIDOS.RFQ helps to solve all above problems. The optimal choice of main RFQ channel parameters can be made using the first level codes. The second level codes are used for correct RF fields calculations taking into account the real vane geometry. Beam simulation in real RF fields is a main task of the third level codes. It is possible for LIDOS.RFQ to calculate non-symmetrical perturbations because there are no assumptions about symmetries both in beam distributions and in RF field distributions. The new algorithms are discussed as well as some advantages over much used codes.

Classification Category: D04

TU4047
Multi-fiducial Techniques for Tracking Large Phase Space Distributions in Non-Linear Fields

S.M. LIDIA, LBNL

One of the challenges in tracking intense beams through linacs is to account for the differences in nonlinear forces experienced by portions of the beam separated by large regions of phase space. In many situations, the high-order map generated by a single fiducial trajectory can fail to adequately capture or describe the beam dynamics of distant particles. I describe here a technique which overcomes this difficulty by piecing together lower-order maps induced by multiple fiducial orbits. This atlas of maps can more accurately track the evolution of a beam with large phase space extent. I discuss applications of this technique to simulating beam dynamics in two-beam accelerators.

TU4048
Production of Negative Hydrogen and Deuterium Ions in Microwave-Driven Ion Sources*

D. SPENCE, K. R. LYKKE**, ANL

We report progress we have made in the production of negative hydrogen and deuterium atomic ions in magnetically-confined microwave-driven (2.45 GHz) ion sources. The influence of source surface material, microwave power, source gas pressure, and magnetic field configuration on the resulting ion current will be discussed. Results strongly suggest that, at least in our source, vibrationally excited molecular hydrogen the precursor to atomic negative ion production, is produced via a surface mechanism demonstrated by Hall et al. [1] rather than via a gas phase reaction as is generally believed to be the case in most ion sources.

*Work supported by USDOE via Argonne LDRD funds.
**Present Address: National Institute of Standards and Technology, Washington, D.C.

[1] R. J. Hall, I. Cadez, M. Landau, F. Pichou, and C. Schermann, Phys. Rev. Lett. 60, 337 (1988).

Classification Category: T01

TU4049

Contribution withdrawn.

TU4050

Contribution withdrawn.

TU4051  (oral poster TU3008)
Doppler-Shift Proton Fraction Measurement on a DC Proton Injector*

J. H. KAMPERSCHROER, General Atomics; J. D. SHERMAN, T. J. ZAUGG, LANL; A. H. ARVIN, A. S. BOLT, SRS

A spectrometer/Optical Multi-channel Analyzer has been used to measure the proton fraction of a dc proton injector. This technique was pioneered by LBNL and was subsequently adopted by the international fusion community as the standard for determining the extracted ion fractions on neutral beam injectors. Balmer-alpha emission from excited neutral atoms, created in charge-changing collisions of beams ions with background gas, is observed at an angle relative to the beam axis. The resulting Doppler-shifts allow the light emanating from the three extracted ion species to be distinguished from one another. While the proton emission rate is orders of magnitude lower on the Low Energy Demonstration Accelerator injector (beam current of tens of mA) than on fusion neutral beams (tens of A), S/N ratios of 200:1 on the proton peak and 20:1 on the molecular ion peaks have been obtained by operating with 30-50 micron slits and integrating the light for 16 s. Preliminary analysis indicates a proton fraction of ~90%, a value in good agreement with results obtained by magnetically sweeping the ion beam, collimated by a slit, across a Faraday cup.

*Work supported by the U.S. Department of Energy.

TU4052

Contribution withdrawn.

TU4053
Beam Simulations for the H^- Injector Upgrade at LANSCE*

R. R. STEVENS, JR., W. INGALLS, O. SANDER, B. PRICHARD, JR., J. SHERMAN, LANL

An upgrade program is being carried out to increase the beam current from our present H^- injector to provide 200 microamperes current in the proton storage ring at LANSCE. In order to meet this objective, the injector must provide at least 30% more current than presently available [1]. More optimal operation of the ring, however, requires a factor of two higher peak current from the injector to reduce circulating losses in the ring. At this higher current, reduced beam emittance is needed to limit beam losses in the linac. Beam simulations have been carried out to model the operation of the existing injector and to determine what changes are needed to obtain higher beam currents. A collaboration with the Lawrence Berkeley Laboratory is now in progress to modify the present ion source to produce 40 mA of H^- beam with reduced beam emittance. Appropriate changes will be made in our 80 kV accelerating column and preinjector beam line in order to accelerate and transport this higher current beam with acceptable beam size and emittance growth. Experimental results for the initial phase of this program have been obtained on our ion source test stand, and these results will be compared with predictions of the simulations.

* Work supported by the U.S. DOE, Defense Programs.

Classification Category: T01

TU4054
Ion Source Development for LANSCE Upgrade *

M. WILLIAMS, R. GOUGH, R. KELLER, K. LEUNG, D. MEYER**, A. WENGROW, LBNL; O. SANDER, W. INGALLS, B. PRICHARD, R. STEVENS, LANL

The upgrade of the Los Alamos Neutron Science Center (LANSCE) Facility will require high intensity H^- beams. Lawrence Berkeley National Laboratory(LBNL) has been contracted by Los Alamos to develop an H^- ion generator that can meet the LANSCE upgrade requirements. Specifically, the output current of the new H^- ion source will be increased from 16 to 40 mA and maintain approximately the same emittance. In order to meet these requirements, LBNL has selected the surface-conversion multicusp ion source. The H^- output current of the present LANSCE surface-conversion source saturates at ~ 20 mA as higher discharge power is applied. Previous experimental studies at LBNL demonstrated that if the source is operated with a magnetic filter, the H^- output current increases linearly with the discharge power. During the past year, we have investigated several source and filter configurations. By placing the converter parallel to the source axis, we are able to installed six tungsten filaments for the discharge. These filaments are located inside the cusp magnetic fields of the side wall magnets which provide the filtering effect. With this new arrangement, the H^- output current was improved from 20 to over 40 mA at the prescribed 12% duty factor (1 msec pulses at 120 Hz).

*This work was supported by Los Alamos National Laboratory and the U.S. Department of Energy under Contract No. AC03-76SF00098.
** Permanent address: Kernfysisch Versneller Instituut, The Netherlands.

TU4055
High Power Test Results of the First SRRC/ANL High Current L-band RF Gun*

C.H. HO, S.Y. HO, G.Y. HSIUNG, J.Y. HWANG, T.T. YANG, SRRC; M. CONDE, W. GAI, R. KONECNY, J. POWER, P. SCHOESSOW, ANL

There is a joint effort between the SRRC (Synchrotron Radiation Research Center, Taiwan) and the ANL (Argonne National Laboratory, USA) for developing a high current L-band rf photocathode gun. We have constructed an L-band (1.3 GHz) , single cell rf photocathode and conducted low power rf test at SRRC. High power rf conditioning of the cavity will be conducted at ANL. In the high power test stage, a pair of solenoids will also be used. In this paper, we report on the construction and high power test results, especially on the dark current beam loading and rf breakdown at the high axial fields (> 90 MV/m) .

*Work supported in part by the National Science Council of Taiwan, R.O.C. under grant No. NSC87-2613-M-213-012.

TU4056
Extraction and Low Energy Transport of Negative Ions *

A. LAKATOS, J. POZIMSKI, A. JAKOB, H. KLEIM, IAP-Univ. of Frankfurt

High perveance negative ion beams with low emittance are necessary for several next generation particle accelerators (e.g. neutrons spallation sources). The production, extraction and transport of these beams have intrinsic difficulties different from positive ion beams. Therefore intensive research is necessary to fulfil the project requirements. The low energy transport from the plasma generator to the entrance of the first accelerator is limitting the current and the quality of the beam. A cesium free H^- volume source based on the high current source for ESS was constructed and integrated into an LEBT section to investigate the influence of beam extraction, electron dumping and space charge compensation on beam current and emittance. The influence of various parameters like residual gas pressure, external magnetic and electric fields on beam propagation, space charge compensation and hence on transmission and emittance growth was studied. Results of the theoretical and experimental work are presented.

* Work supported by BMFT under contract no. 06 OF 841.

TU4057
Stable Ampere Level Emission of Energetic Electrons by Electrically Excited Ferrelectric Ceramics

I. BOSCOLO, A. SCURATI, University and INFN-Milano

Stable emission of energetic electrons from ferroelectric ceramic of PLZT type, i.e. lead zirconate titanate lanthanum doped, in form of thin disks have been obtained with a proper design of the front electrode. The current is about 1 A square centimeter and the energy is a couple of keV. High density, more than 50 A square centimeter, is obtained in the plasma assisted configuration. The electron expulsion from the surface is due to the spontaneous polarization switching induced by a high voltage pulse applied to proper metallic electrodes deposited as thin film on the two surfaces of the disk. In passing from the usual front electrode having the structure of an interconnected grating, to an electrode being a pattern of disconnected patches contained within a ring, the electron emission passes from erratic with samples and decaying with the number of shots to stable in both cases. This change in the behavior is explained by the fact that the sandwich of the continuous electrode and the grating constrains the domain switching and the relevant charge carriers motion within the part of material covered by metal, whilst a quasi-free surface allows the domain switching all over the whole area with the consequent flux of electrons in and out of the sample surface. In the plasma assisted emission, the front electrode must be the normal grating because the plasma in this electrode configuration is homogeneous.

TU4058

Contribution withdrawn.

TU4059
Time-Resolved Imaging for the APS Linac Beams*

A. H. LUMPKIN, W. J. BERG, B. YANG, M. WHITE, ANL

The particle-beam imaging diagnostics for the Advanced Photon Source (APS) injector linac have been enhanced by the installation of optical transition radiation (OTR) screens and the use of Cs-doped YAG crystals as beam profile monitors. Both converters have improved spatial resolution and time responses compared to the standard Chromox (A1₂0₃:Cr) screens used elsewhere in the linac. For the linac macropulse of 30-ns duration composed of 86 micropulses at S-band frequency intervals, only the OTR mechanism is prompt enough to separate individual micropulses. The OTR also allows synchroscan streak camera measurements of the micropulse averaged bunch length. Tests have been performed at 400 to 600 MeV using the gated DC thermionic gun source. Beam sizes less than = 60 µ have been observed with micropulse bunch length of = 2-3 ps using OTR. Images of accelerated beam from the lower-emittance rf thermionic gun will be reported as available.

* Work supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. W-31-109-ENG-38.

TU4060
LEDA & APT Beam Position Measurement System: Design and Initial Tests*

J. D. GILPATRICK, J. POWER, R. B. SHURTER, M. STETTLER, LANL; J. O'HARA, AlliedSignal FM&T/NM

Beam profile measurements are being designed and fabricated for the Low Energy Demonstration Accelerator (LEDA), a 20-MeV, 100-mA-cw proton-accelerator, presently under construction at Los Alamos National Laboratory (LANL). These measurements will provide beam position information for a steering method which centers the beam within the beam pipe for Accelerator Production of Tritium (APT) and LEDA. The beam steering method uses two BPMs and two translatable quadrupole magnets in three successive focusing-quadrupole magnets every five FODO lattice periods. What makes these beam position measurements unique is how they will attain their required accuracy performance and how this accuracy is verified. The position measurement systems consists of micro-stripline beam position monitors (BPMs) and rf coaxial cables, log-ratio processors, on-line error correction sub-systems, and associated algorithms and computer software. This paper will discuss the mapping of the BPM probes responses, the algorithm used to calculate their low beam-energy performance, the log ratio processors and its on-line error-correction, and initial results from bench tests of an integrated position-measurement system.

* Work suported by the U.S. Department of Energy.

TU4061

Contribution withdrawn.

TU4062
Laser Diagnostic for High Current H Beams

R. E. SHAFER, LANL

In the last 5 years, significant technology advances have been made in the performance, size, and cost of solid-state diode-pumped lasers. These developments enable the use of compact Q-switched Nd:YAG lasers as a beam diagnostic for high current H beams. Because the binding energy of the last electron is only 0.75 eV, and the maximum detachment cross section is 4 E-17 cm² at 1.5 eV, A 50 mJ/pulse Q-switched Nd:YAG laser can neutralize a significant fraction of the beam in a single 10-ns wide pulse. The neutral beam maintains nearly identical parameters as the parent H beam, including size, divergence, energy, energy spread, and phase spread. A dipole magnet can separate the neutral beam from the H beam to allow diagnostics on the neutral beam without intercepting the high-current H beam. Such a laser system can also be used to extract low current proton beams for cancer therapy treatment, or induce fluorescence in partially stripped heavy ion beams. Possible beamline diagnostic systems will be reviewed, and the neutral beam yields will be calculated for several examples.

TU4063
Horizontal Emittance Measurement in ATF Extraction Line

T. OKUGI, T. HIROSE, Tokyo Metropolitan University; H. HAYANO, K. KUBO, N. TERUNUMA, J. URAKAWA, KEK; S. KASHIWAGI, Graduate Univ. for Advanced Studies

The beam operation of the damping ring in the KEK accelerator test facility (ATF) has been started since January 1997 for the development of the technologies to achieve lower emittances, which are required in the future linear colliders. (The designed emittance in this damping ring is 1.4nm horizontally and 14pm vertically.) There is a beam extraction line in ATF for the beam diagnosis of the beam which reduced the emittance in the damping ring, and the operation of the extraction line has been started since Novenber 1997. However, the position stability at the extraction line doesn't have good performance for the some reasons of longitudinal and transverse beam oscillations in the damping ring and the stability of extraction kickers and so on, the amount of the position fluctuation is 50 to 200micro-m and it affects the beam size measurement at the ATF extraction line. Therefore, we established the jitter subtracted beam size measurement for evaluation of the horizontal beam emittance. In this presentation, the performance of jitter subtraction and the resulted horizontal emittance in the damping ring will be presented.

TU4064
Requirements for RF Control of the Vector Sum for Superconducting Proton Linacs

M. HÜNING, T. SCHILCHER, S.N. SIMROCK, DESY

Superconducting accelerator technology has demonstrated its superior performance in large scale machines such as CEBA at TJNAF and is increasingly used for new accelerator designs. Until now this technology has found its main application in electron accelerators. However nowadays proton accelerator designs for the European Spallation Source (ESS) and the Accelerator Driven Transmutation Technology (ADTT) also study the feasibility of superconducting linacs. In contrast to the highly relativistic electron beams the proton beam exhibits an increased susceptibility to voltage fluctuations in the acceleration system induced by microphonics and dynamic Lorentz force detuning. Although low beam loss is an important criterion for linac design, studies of the longitudinal dynamics appear to be a good indicator for beam stability in presence of fluctuations of the accelerating field. Control of the vector sum of multiple cavities driven by one klystron is desirable for cost reasons but does not allow for control of individual cavity fields. In this paper we study the performance of such a system.

TU4065
Time Delay Compensation for the Digital RF Control at the TESLA Test Facility

H. IMSIEKE, A. KHOLODNYI, T. SCHILCHER, S.N. SIMROCK, DESY

Time delays or dead times between inputs and outputs are an inherent characteristic of digital feedback systems. The time delay limits the maximum allowable gain required for system stability. Modern control theory provides a scheme called Smith predictor which has the potential to improve control performance significantly. The method is based on model internal control which works well if the dynamics of the plant are slow compared to the time delay. In this paper we analyze the performance improvement that can be achieved in the TTF rf control system where the time delay is dominated by computational delay. In this system the time delay of 5 microseconds and sampling period of 1 microsecond are short compared to the cavity time constant of 700 microseconds. Attention is paid to both theoretical and practical aspects. Also investigated is the use of a Smith predictor implementation in hardware to be used for low Q cavities such as the TTF RF gun cavity.

TU4066
RF Control Studies for Moderate Beamline Coupling Between SRF Cavities*

L. DOOLITTLE, D.X. WANG, TJNAF

When an SRF accelerator is designed, there is motivation to move the cavities close together on the beamline. Assuming the beamline apertures are not shrunk as well, this compaction (which will increase the overall accelerating gradient and/or lower the dynamic cryogenic heatload) increases the inter-cavity coupling. Within certain limits, the control system can compensate for this coupling by retuning each of the cavities. This paper describes constraints on the RF system, tuners, couplers, and control systems that are required to provide stable operation of cavities in the presence of intercavity coupling that exceeds the loaded bandwidth of an individual cavity.

* Work supported by the U.S. DOE Contract # DE-AC05-84ER40150.

TU4067
The RF System for the CEBAF Polarized Photoinjector*

M. CROFFORD, C. HOVATER, G. LAHTI, TJNAF

The CEBAF electron accelerator has recently begun delivering spin-polarized electrons for nuclear experiments. Spin-polarized electrons are emitted from a GaAs photocathode that is illuminated with pulsed laser light from a diode laser synchronized to the 3rd subharmonic (499 MHz) of the accelerating cavity frequency (1497 MHz). Up to three experimental halls (A, B and C) are served by the photoinjector each with their own beam requirements. To accomplish this, three independent diode lasers are synchronized and combined to illuminate the GaAs photocathode emitting a 1497 MHz pulse train of electrons. In addition a RF bunching cavity approximately 2 meters down stream from the photocathode is used to compensate for space charge effects at the higher beam currents. The RF system that controls these elements is a modified VME based system. Custom RF VME modules control phase and amplitude for each laser diode and the bunching cavity. Power requirements were satisfied with commercial RF amplifiers, 5 watts for the diode lasers and 10 watts for the bunching cavity. Phase and amplitude drifts are corrected by simple software algorithms in the EPICS Input/output computer on the VME crate. The RF system is compact, simple and allows for easy hardware or software modifications.

* Work supported by the U.S. DOE Contract # DE-AC05-84ER40150.

TU4068

Contribution withdrawn.

TU4069
Waveguide Harmonic Dampers for Klystron Amplifiers *

Y. KANG, A. NASSIRI, ANL

Waveguide harmonic dampers are tested for removing the harmonic frequency power from the klystron amplifiers of the APS linac. A straight coaxial probe antenna with a matched load is used in a damper. A linear array of such dampers is used in the narrow wall of the WR284 waveguide for damping klystron harmonics while decoupling the dominant TE01 mode to the dampers. The klystron harmonics may exist in the waveguide as waveguide higher-order modes above cutoff. Computer simulations are made to optimize the waveguide harmonic damping characteristics of the harmonic dampers. Other waveguide-type harmonic dampers are also discussed.

* Work supported by U.S. Department of Energy, Office of Basic Energy Sciences under Contract No. W-31-109-ENG-38.

TU4070
Development of a 114.24 MHz Sub-Harmonic Buncher Cavity for the KEKB Injector Linac

S. YAMAGUCHI, S. OHSAWA, M. IKEDA, A. ENOMOTO, KEK; Y. IGARASHI, Mitsubishi Heavy Industries, Ltd., Nagoya Aerospace Systems

A 114.24 MHz sub-harmonic buncher cavity was newly developed to improve the bunch purity of the KEKB injector linac. The cavity was designed so that it has high shunt impedance (1.7 times as large as that of the existing cavity) and low maximum surface field at acceleration gap (less than 10 MV/m for input power of 10 kW) under the condition that it can be installed in the existing coils. The cold tests show the unloaded Q-value of 7000 and shunt impedance of 1.2 , which satisfy the designed values. To reduce the probability of rf breakdown, inner wall of the cavity was electric-polished and vacuum pump-out was put at the end wall of the cavity. Copper was selected as the structure's material (it was copper-plated stainless steel for the existing cavity) to improve the water cooling capability and reduce the frequency change due to the gap-distance change which comes from input power's up and down. Results of high-power tests and beam acceleration tests will be described.

Classification Category: T04

TU4071  (oral poster TU3009)
Design, Construction and Operational Results of the IGBT Controlled Solid State Modulator High Voltage Power Supply used in the High Power RF systems of the Low Energy Demonstration Accelerator of the Accelerator Production of Tritium (APT) Project*

J. T. BRADLEY III, D. REES, R. S. PRZEKLASA, LANL; M. C. SCOTT, Continental Electronics Corp.

The 1700 MeV, 100 mA Accelerator Production of Tritium (APT) Proton Linac will require 244 1 MW, continuous wave RF systems. 1 MW continuous wave klystrons are used as the RF source and each klystron requires 95 kV, 17 A of beam voltage and current. The cost of the DC power supplies is the single largest percentage of the total RF system cost. Power supply reliability is crucial to overall RF system availability and AC to DC conversion efficiency affects the operating cost. The Low Energy Demonstration Accelerator (LEDA) being constructed at Los Alamos National Laboratory will serve as the prototype and test bed for APT. The design of the RF systems used in LEDA is driven by the need to field test systems with high efficiency and extremely high reliability before APT is built. We present a detailed description and test results of one type of advanced high voltage power supply system using Insulated Gate Bipolar Transistors (IGBTs) that has been used with the LEDA High Power RF systems. We also present some of the distinctive features offered by this power supply topology, including crowbarless tube protection and modular construction which allows graceful degradation of power supply operation.

* Work supported by the U.S. Department of Energy.

TU4072
Design, Operation, and Test Results of 350 MHz LEDA RF System*

D. REES, J. BRADLEY, K. CUMMINGS, T. HARDEK, W. ROYBAL, LANL

The Low Energy Demonstration Accelerator (LEDA) being constructed at Los Alamos National Laboratory will serve as the prototype for the low energy section of Acceleration Production of Tritium (APT) accelerator. The APT accelerator requires over 244 RF systems each with a continuous wave output power of 1 MW. The reliability and availability of these RF systems is critical to the successful operation of APT plant and prototypes of these systems are being developed and demonstrated on LEDA. The first completed LEDA RF systems are three, 1.2 MW, 350 MHz, continuous wave, klystrons driving a radio frequency quadrapole (RFQ). This paper presents the design and test results for these RF systems including the klystrons, cathode power supply, circulators, RF vacuum windows, and RF components. The three RF systems driving the RFQ use the accelerating structure as a power combiner and this places some unique requirements on the RF systems. These requirements and corresponding operational implications will be discussed.

*Work supported by the U.S. Department of Energy.

TU4073
The RF Power System for the SNS Linac*

P. J. TALLERICO, W. A. REASS, LANL

The initial goal of the SNS project is to produce a 1 MW average, short pulsed beam of protons onto a neutron-producing target. The objective of the SNS RF system is to generate 117 MW peak of pulsed 805 MHz microwave power with an accelerated beam pulse length of 1.04 ms at a 60 Hz repetition rate. The power system is upgradeable in either peak power or pulse length to deliver 2 MW average power to the neutron target. The RF system also requires about 3 MW peak of RF power at 402.5 MHz, which is not discussed here. The challenge is to produce an RF system near minimum cost, that is very reliable, and economical to operate. The combination of long pulses, the ability to double the pulse length in the future, and the repetition rate makes conventional solutions, such as the pulse transformer and transmission line very expensive. A 2.5 MW, modulating-anode klystron is the baseline RF amplifier. We discuss four power systems as candidates for the design. The baseline is a floating-deck modulating anode system. Next is the fast power supply, switching with a rise time below 0.1 ms. This can eliminate the modulator, and drastically reduce energy storage requirements. Next is to use a pulse transformer with an IGBT switch and a bouncer circuit, as in the TESLA modulator. Next is to use a series high voltage IGBT switch, without a pulse transformer. We discuss the advantages and problems of these four types of power systems, and emphasize the first two.

* Work supported by the U.S. Department of Energy.

TU4074
High Power RF Component Testing For NLC

A.E. VLIEKS, W.R. FOWKES, R.J. LOEWEN, S.G. TANTAWI, SLAC

In the Next Linear Collider (NLC), the high power rf components must be capable of handling peak rf power levels in excess of 600 MW. In the current view of NLC, even the rectangular waveguide components must transmit at least 300 MW rf power. At this power level, peak rf fields can greatly exceed 100 MV/m. We present recent results of high power tests performed at the Accelerator Structure Test Area (ASTA) at SLAC. These tests were designed to investigate the RF breakdown limits of several new components potentially useful for NLC. In particular we have tested a new type TE01-TE10 circular to rectangular mode converter, a modified (internal fin) Magic Tee hybrid and an upgraded flower petal mode converter. We will also present an alternative 3 dB Hybrid design [1] having improved peak power handling capability.

Classification Category: T04

TU4075
Linear Inductive Voltage Adders (IVA) for Advanced Hydrodynamic Radiography*

M. G. MAZARAKIS, J. D. BOYES, D. L. JOHNSON, J. S. LASH, J. E. MAENCHEN, P. R. MENGE, C. L. OLSON, S. E. ROSENTHAL, D. C. ROVANG, SNL; B. V. OLIVER, D. R. WELCH, Mission Research Corp.; V. L. BAILEY, I. D. SMITH, Pulse Sciences Inc.

The electron beam which drifts through the multiple cavities of conventional induction linacs (LIA) is replaced in an IVA by a cylindrical metal conductor which extends along the entire length of the device and effectuates the addition of the accelerator cavity voltages. In our approach to radiography, the linear inductive voltage adder drives a magnetically immersed electron diode with a millimeter diameter cathode electrode and a planar anode/bremmstrahlung converter. Both anode and cathode electrodes are immersed in a strong (15-50 T) solenoidal magnetic field. The electron beam cross section is the same size as the cathode needle and generates a similar size, very intense x-ray beam when it strikes the anode converter. An IVA driven diode can produce electron beams of equal size and energy as a LIA but with much higher currents (40-50 kA versus 4-5 kA), simpler hardware and thus lower cost. We present here first experimental validations of our technology utilizing HERMES III and SABRE IVA accelerators. The electron beam voltage and current were respectively of the order of 10 MV and 40 kA. X-ray doses of up to 1 kR@1 m and spot sizes as small as 1.7 mm (at 200 R doses) were measured.

*Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy under Contract DE-ACO4-94AL85000.

TU4076
High Performance Pulse Modulater for 80 MW S-Band Klystron in SPring-8 Linac

T. HORI, H. YOSHIKAWA, T. KOBAYASHI, T. ASAKA, H. SAKAKI, S. NAGASAWA, H. YOKOMIZO, SPring-8

The test-stand of a compact pulse modulator is under construction to carry out an acceptance test of spare 80 MW S-band klystrons (E3712 TOSHIBA) and to examine to different type thyratrons (F351 TRITON, CX1937A EEV). This modulator includes several new technologies such as 40 MHz invertor power supply for a charging unit, command charging, tunable slugs of PFN coils with a remote control in order to improve stability of a klystron beam voltage and to reduce fault rates due to a thyratron prefire as well as to simplify the maintenance works. Typical specifications of the modulator were 190 MW peak power, 390kV peak beam voltage, 60pps pulse repitition rates, 2.2 micro-sec flat top pulse width and less than 0.5% beam voltage stability. This paper presents the circuit design, specifications and the result of the running test.

TU4077  (oral poster TU3010)
The Cold Model of the CDS Structure

L.V. KRAVCHUK, V.V. PARAMONOV, V.A. PUNTUS, INR RAS

The Cut Disk Structure (CDS) was proposed as the compensated accelerating structure for high energy linacs. Cold rf model was manufactured to examine the CDS parameters, partially for S-band electron linacs. In agreement with design parameters coupling coefficient near 22% and high shunt impedance are obtained. Results of experiments are described.

TU4078
The Bridge Coupling Cavities in the Separated Drift Tube Linac Structure

L.V. KRAVCHUK, A.S. LEVCHENKO, V.V. PARAMONOV, INR RAS; T. KATO, F. NAITO, Y.YAMAZAKI, KEK

The Separated Drift Tube Linac (SDTL) structure was proposed for medium energy range of proton linacs. The accelerating cavity consists of several SDTL units with focusing lenses between sections. To drive several SDTL sections from single rf source, application of both rf power dividers and coupling bridge cavities seems feasible. Through bridge couplers, the field distribution, in both amplitude and phase, is automatically stabilised among connected sections in the first order. Additional rf power dissipation of an order of several percents is inevitable in the bridge coupler because of its standing-wave operation. In this paper particularities and limitations for coupling bridges in the SDTL tanks are considered. Well known cylindrical bridge cavities may be used. To mantain features of the SDTL structure as the easy-to-do low-cost one, Rectangular Directly Coupled Bridges (RDCB) look more preferable. Developed and realised in the main part of INR proton linac, RDCB operate in TE10n mode and originally have no mode mixing problem. To simplify manufacturing procedure, RDCB design is based on parts of standard rectangular waveguide. The philosophy of the SDTL-RDCB tank and results of preliminary design are presented.

TU4079  (oral poster TU3011)
324-MHz RF Deflector Design and Test

S. FU*, T. KATO, F. NAITO, K.YOSHINO, KEK

In the design of JHF linac, the beam needs to be chopped for the injection into the following rapid-cycling ring. The chopper is decided to locate in the 3-MeV medium-energy transport line between the RFQ and the DTL. The 324-MHz RF deflector (RFD) is adopted as a fast beam-chopper for its various merits, such as compactness and high deflecting field, which benefit for an efficient deflection of the beam within a short beam line. In the application of an RF Deflector for an intense-beam chopper, the beam loss in transient time is a major concerned issue. The shunt impedance should be as large as possible to make the cavity to be powered by the commercially available solid RF source. The RFD cavity is designed by HFSS code to have a very low loaded-Q value to about 10 by means of two large coupling loops, and meanwhile to keep the higher order modes sufficiently far away from the deflecting mode. An Aluminum cold model cavity was made for the test. In this paper, the design, simulation and the test results are described in detail.

* on leave from China Institute of Atomic Energy

TU4080
Optimization on Wakefield Damping in C-band Accelerating Structure

N. AKASAKA, H. MATSUMOTO, T. SHINTAKE, KEK

In the C-band rf-system [1], the choke-mode type cavity is used to extract the HOM power from the beam. It is essential to accelerate low emittance beams along the main linac of the e+e- linear collider. To damp the HOM power, we use ring shaped SiC-ceramics, whose dimensions were carefully determined to optimize the damping performance. This paper describes details of the optimizing process, which utilizes the time-domain computer simulation of the transverse wake field. A 1.8 m long structure is under fabrication and it will be tested with ASSET beam at SLAC in this year.

Classification Category: T05

TU4081
Design Considerations for Multiple-Beam RFQ Structures

V. KAPIN, M. INOUE, Y. IWASHITA, A. NODA, ICR-Kyoto Univ.

There are applications, which require MeV-range multiple-beams consisting of a large number of identical highly packed beamlets. The multiple-beam RFQ (MB-RFQ) arranged as a matrix array of longitudinal rod-electrodes is appropriate candidate. A configuration of MB-RFQ resonator should ensure identical quadrupole fields in every accelerating channel. The MB-RFQ resonators based on TEM transmission lines are studied. The resonators are designed by a periodical multiplication of a 4-rod unit cell in transverse direction. To understand fields of resonator the normal mode technique is applied. The resonator field is expanded into normal modes having simple field patterns. In general case of resonator with an arbitrary number of rods and normal modes, analysis becomes difficult. To simplify problem, only normal modes composed from four normal modes of a 4-rod unit cell are selected. Similar to normal modes of a 4-rod unit cell (coaxial, quadrupole and two dipole modes), selected normal modes of multi-rod resonator have clear field patterns. Novel configurations of MB-RFQ resonators based on these normal modes are generated. The RF properties of resonators are verified with computer simulations done with MAFIA code. Features of beam dynamics in MB-RFQ structures are discussed. The results of beam dynamics simulations are presented.

Classification Category: T05

TU4082
Proton Beam Studies with a 1.25 MeV, cw Radio Frequency Quadrupole LINAC*

G.O. BOLME, T.W. HARDEK, L.D. HANSBOROUGH, D.J. HODGKINS, D.R. KEFFELER, J.D. SHERMAN, H.V. SMITH, R.R. STEVENS, L.M. YOUNG, T.J. ZAUGG, LANL; P.P. BALLEYGUIER, CEA/Bruyeres le Chatel; A.H. ARVIN, A.S. BOLT, M.C. RICHARDS, SRS; J.H. KAMPERSCHROER, General Atomics

A high-current, cw linear accelerator has been proposed as a spallation neutron source driver for tritium production. Key features of this accelerator are high current (100 mA), low emittance-growth beam propagation, cw operation, high efficiency, and minimal maintenance downtime. A 268 MHz, cw radio frequency quadrupole (RFQ) LINAC section and klystrode based rf system was obtained from the Chalk River Laboratories [1] and was previously installed at LANL [2] to support systems development and advanced studies in support of cw, proton accelerators. A variation of the Low Energy Demonstration Accelerator (LEDA) proton injector, modified to operate at 50 keV, was mated to the RFQ and was operated to support advance developments for the Accelerator Production of Tritium (APT) program. High current, proton beam studies were completed which focused on the details of injector-RFQ integration, development of beam diagnostics, development of operations procedures, and personnel and equipment safety systems integration. This development led to acceleration of up to 100 mA proton beam.

*Work supported by the US Department of Energy.

Classification Category: T05

TU4083
The Mechanical Design and Fabrication of a Ridge-Loaded Waveguide for an RFQ

R. VALDIVIEZ, P. ROYBAL, B. CLARK, F. MARTINEZ, D. CASILLAS, G. GONZALES, J. TAFOYA, LANL

A Radio Frequency Quadrupole (RFQ) accelerator with an RF power input of 3 MW and an H+ beam output power of 6.7 MeV at 100 mAmps, continuous duty factor utilizes twelve identical ridge-loaded waveguides to couple the RF power to the RFQ accelerating cavity. The mechanical design and fabrication of the ridge-loaded waveguide are the topics of this paper. The waveguide is rectangular in cross-section, tapers in two planes, and incorporates a center ridge on the wider dimension with rectangular cavities on either side. The ridge reduces the internal height of the waveguide from 14.6 centimeters to 0.158 centimeters. The deposited RF power into the waveguide walls increases with length due to the decreasing cross-sectional dimensions. The deposited power is accommodated as thermal energy and ranges from 0.053 watts/cm² at the start of the ridge to 0.85 watts/cm² at the RFQ interface. The mechanical analyses that were carried out to predict the final design characteristics of the ridge-loaded waveguide, as well as the manufacturing methods and materials used, are also presented.

TU4084
Prototype models for the SNS RFQ*

A. RATTI, J. AYERS, R. GOUGH, J. GREER, M. HOFF, R. KELLER, R. MCGILL, J. STAPLES, R. YOURD, LBNL

Prototype models that simulate components of the RFQ which is part of the Spallation Neutron Source front end injector are described. The RFQ operates at 402.5 MHz, a maximum current of 70 mA H^- and 6% duty factor. The copper plated aluminum model is a full size version of one of the four sections of the RFQ and is used to perform low level r.f. measurements of frequency, r.f. tuning, field structure and power coupling port studies. This model is used to benchmark the 3D computer simulation studies and to test assembly and field measurement procedures. A quarter-wave coaxial resonator is used to perform full-power tests of high current seals and joints. This model is also used to test the prototype of the RFQ tuner. The design and construction of all models are presented, as well as test results.

* This work is supported by the U.S. Department of Energy under contract n. DEAC0376SF00098.

TU4085
Initial Design of a Charge-State Multiplier System for the RIKEN RI-Beam Factory Project

O. KAMIGAITO, A. BANDYOPADHYAY*, A. GOTO, Y. YANO, RIKEN

In the RIKEN RI-beam factory project, a Charge-State Multiplier system (CSM) is planned to be placed between the existing heavy ion linac (RILAC) and the ring cyclotron (RRC) in order to increase the charge-to-mass ratio (q/A) of ions without changing the injection radius of the RRC. The CSM consists of an accelerator, a charge stripper and a decelerator. For the accelerator and decelerator sections, drift tube linacs of variable-frequency type will be used, whose rf frequency should be varied from 36 to 76 MHz. The total accelerating voltage required for the accelerator section is about 26 MV and that for the decelerator section is about 11 MV. Initial design of the CSM tanks, based on the IH structure with a movable shorting plate, is described in this paper.

*On leave from VECC, India.

TU4086
The Present Status of Development on Superconducting Cavities at SHI

Y. MATSUBARA, M. HIROSE, T. HORI, Laboratory for Quantum Equipment Technology; H. SAITO, Quantum Equipment Business Center; H. INOUE, M. ONO, E. KAKO, S. NOGUCHI, K. SAITO, T. SHISHIDO, KEK

We have started development on superconducting cavities with KEK from 1997. First, we checked out the effect of electropolishing (EP) for a L-band bulk-niobium single-cell cavity. In the results, we got the relationship between maximum accelerating field and removed thickness from the cavity surface, where small amount of EP's were applied repeatedly, and obtained the high gradient of over 30 MV/m. Next, when fabricating the second, we directly removed as much the thickness as the optimized depth by the first cavity using heavy chemicalpolishing and final light EP in sequence. We are now in the stage of fabricating the third one, 3-cell cavity. In this paper, we will report on these results and activities at Sumitomo Heavy Industries, Ltd.

TU4087
RF Tests on the Initial 2.8m of the 8m Long ISAC RFQ at TRIUMF

R. L. POIRIER, A. K. MITRA, TRIUMF

The ISAC RFQ is an 8 meter long, 4-rod split-ring structure operating at 35 MHz in CW mode. The rods are vane-shaped and are supported by 19 rings spaced 40 cm apart. The rings are unique in that the rf surfaces have been de-coupled from the mechanical support structure to improve dynamic stability. An initial 2.8m section of the accelerator (7 of 19 rings) was installed and aligned in the 8m, square cross-section, vacuum tank to allow rf and beam tests to be carried out. The stringent, ± 0.08 mm, quadrature positioning tolerance of the four rod electrodes was achieved and a relative field variation along the 2.8 meter of the RFQ was measured to be ± 1%, using the standard bead pull method. The results of the signal level measurements were a frequency of 35.7 MHz, a Q of 8700 and a resonant shunt impedance of 292 kilohm -meters. Compared to the 3 ring prototype, this represents a 21% increase in Q and a 30% increase in shunt impedance. The latest results on the full rf power tests on the seven ring structure will be reported.

TU4088

Contribution withdrawn.

TU4089
Development of a Raster Electronics System for Expanding the APT Proton Beam*

S. CHAPELLE, E. L. HUBBARD, M. E. SCHULZE, R. E. SHAFER, T. L. SMITH, General Atomics

A method of expanding the 1700 MeV, 100 mA proton beam of the APT linear accelerator onto the tritium production target has been devised using eight ferrite dipole magnets to deflect the beam in the x and y-plane and paint the 19 cm by 190 cm target uniformly with a rastered pattern. This paper describes the salient issues of the design of the electronics that are unique to the expander, including considerations of redundancy, target safety, and reliability. IGBT-based power modulators drive the raster magnets with triangular current waveforms which are synchronized using voltage-controlled crystal oscillators and phase-locked loops. The fault detection circuitry is also described. Test data are presented for the entire prototype system.

* Work supported by the U.S. DOE under contract DE-AC04-96AL89607.

Classification Category: T07

TU4090
Low to High Energy Beamstops for APT*

D.W. DOLL, T.H. VAN HAGAN, K.M. REDLER, M. FIKANI, G. SPALEK, General Atomics; J.D. SCHNEIDER, F.R. SPINOS, W. FUNK, LANL

Beamstops are required for the commissioning and operation of accelerators. The family of beamstops currently being developed for the DOE-sponsored Accelerator Production of Tritium (APT) program is addressed. The operational range encompasses proton energies of 6.7 MeV at the beginning of the RFQ to 1.7 GeV at the target/blanket, and both pulsed and cw operating modes. An additional beamstop was needed on a companion test facility to validate ion injector performance. This first of the series of beamstops has been built and operated; the second is under construction and will be operational by mid-1998. Particle stopping distance and duty factor drive the size and heat transfer capacity of these beamstops; the need for low neutron production and activation potential drives the material selection. At energies above 6.7 Mev, the preferred beamstop material is graphite with aluminum the choice for structures and helium the choice for coolant. Boronated water is the preferred shielding below 70 MeV, but is less suitable at higher energies because of the creation of Be-7. The evolution of the low-energy demonstration accelerator (LEDA) to 20 MeV and remaining APT beamstops will be presented with performance ramifications and updated status of the hardware and testing.

*This work was sponsored by the Department of Energy under Contract No. DE-AC04-96AL8907.

TU4091

Contribution withdrawn.

TU4092  (oral poster TU3012)
Design of an Ogive-Shaped Beamstop*

T.H. VAN HAGAN; D.W. DOLL, General Atomics; J.D. SCHNEIDER, F.R. SPINOS, LANL

This paper addresses the evolution, design, and development of a novel approach for stopping continuous-wave, non-rastered proton beams. Capturing the beam in vacuo within a long, axisymmetric surface of revolution has the advantages of spreading the deposited energy over a large area while minimizing prompt neutron backstreaming and reducing shield size and mass. Evolving from a cylinder/cone concept, the ogive shape avoids abrupt changes in geometry that produce sharp thermal transitions, allowing the beam energy to be deposited gracefully along its surface. Thermal management at modest temperature levels is provided with a simple, one-pass countercurrent forced-convection water passage outside the ogive. Hydrophone boiling sensors provide overtemperature protection. Photographs and test experience gained to date on the ogive beamstops for the Chalk River Injector Test Stand (1.2 MeV, 100 kW average power) and the Low Energy Demonstration Accelerator (6.7 MeV, 670 kW average power) proton linacs are included.

*This work is a joint GA/LANL effort sponsored by the U.S. Department of Energy under contract #DE-AC04-96AL89607.

TU4093

Contribution withdrawn.

TU4094
Low Stored Energy 100 kV Regulator for Ion Sources at LANSCE*

E. G. JACOBSON, R. L. HAFFNER, W. B. INGALLS, B. J. MEYER, J. S. STELZER, LANL

To minimize accelerating-column damage caused by uncontrolled energy release during arcdown, it is desirable to minimize the available stored electrical energy. In 1988 Los Alamos Neutron Science Center (LANSCE) personnel developed an initial design of a low-stored-energy power supply and planar triode. As priorities for manpower and funding changed over the years, work on the control circuitry was performed intermittently until construction of the new Ion Source Test Stand (ISTS). A 120 kV power supply, which has low output capacitance, and a planar triode have been used to supply the 60 mA, 120 Hz, 12% duty-factor current for the ISTS. The triode's cathode current is controlled by circuitry operating both at power-supply voltage level and at ground level via a fiber optic link. Voltage droop is approximately 600 V during the 1-msec beam pulse. We present the status of the regulator and its special challenges.

* Work supported by the U. S. Department of Energy.

TU4095
Testing of Vacuum Pumps for the APT/LEDA RFQ

K. KISHIYAMA, S. SHEN, LLNL; N. G. WILSON, AMPRO, Inc; D. SCHRAGE, R. VALDIVIEZ, LANL

Two vacuum systems were designed and built for the RFQ (Radio Frequency Quadrupole) cavity which is part of the APT/LEDA (Low Energy Demonstration Accelerator) linac. The gas load from the proton beam required very high hydrogen pump speed and capacity for the RFQ cavity vacuum system. The gas load from the high power RF windows also required very high hydrogen pump speed for the RF window vacuum system. Cryopumps were chosen for the RFQ vacuum system and ST185 sintered NEG cartridges were used in the RF window vacuum system. Hydrogen pump speed and capacity measurements were carried out for the cryopump and the NEG. This paper will discuss the test procedures and the preliminary results of the measurements.

TU4096
Medical and Industrial Applications of C-Band Accelerator Technology

E. TANABE, AET Associates, Inc.; T. SHINTAKE, H. MATSUMOTO, KEK

Electron linear accelerators are commonly used in medical and industrial applications such as radiation therapy, nondestructive testing and radiation processing. Most of these accelerators used in various applications are based on S-band (about 3GHz) microwave technologies. In this paper, C-band microwave frequencies (5 to 7GHz) are reviewed for medical and industrial applications in terms of technological limitation, production cost and reliabilities. Moreover, the C-band technologies of RF components and systems, including RF sources, solid-state modulators, waveguide components and accelerator structures will be discussed for new product designs.

TU4097
Assessing the Suitability of a Medical Cyclotron as an Injector for an Energy Upgrade

J.A. CLARKE, D.M. DYKES, C.W. HORRABIN, H.L. OWEN, M.W. POOLE, S.L. SMITH, V.P. SULLER, Daresbury; A. KACPEREK, B. MARSLAND, Clatterbridge

The 60 MeV cyclotron at Clatterbridge operates as the UK centre for proton therapy, concentrating on treatment of eye tumours; the accelerator is a Scanditronix model MC60PF fixed energy isochronous cyclotron with a high current ion source. Although possible energy upgrades have been previously considered interest has now been reawakened by the activities of the Italian TERA Foundation, which has proposed a compact high frequency booster linac as a potential solution to achieve the 200 MeV needed for a broader therapy programme. The paper reports progress on studies to assess if the Douglas cyclotron is suitable for a test of such a prototype booster linac. The results demonstrate that a cyclotron beam pulse of about 25 microseconds can be achieved by application of amplitude and phase modulation to its RF system. The most recent measurements of output emittance and energy spread of the accelerator, in this unusual pulsed mode, are presented and compared with the normal CW values. Compatibility with the acceptance of the proposed linac is discussed.

TU4098
A 3 GHz, 200 MeV Proton Linac Booster (LIBO) for Cancer Treatment

U. AMALDI, A. MILLICH, B. SZELESS, M. VRETENAR, E. J. N. WILSON, CERN; J. E. STOVALL, LANL; K. CRANDALL, Crandall Consult; M. WEISS, Tera Foundation

Proton cyclotrons with output energies of 60 to 70 MeV exist in several hospitals and laboratories. A high frequency (3 GHz) booster linac (LIBO) is proposed to upgrade the cyclotron beam to 200 MeV, an energy required to treat deeply seated cancerous tumors. The LIBO, essentially a side coupled structure, can produce beams with a variable output energy. This paper presents the feasibility study of such an accelerator, together with the 1998-1999 milestones, consisting in the machining of a part of the LIBO and testing it with full RF power on a test stand at CERN.

TU4099
Medical Isotope Production with the Accelerator Production of Tritium (APT) Facility

M. R. BUCKNER, WSRC; M. CAPPIELLO, E. PITCHER, LANL; H. A. O'BRIEN, O'Brien & Associates; K. SPICER, MUSC

A feasibility study of producing radionuclides in the Accelerator Production of Tritium (APT) Facility has been completed. The APT is a 1700MeV, 100ma accelerator proposed for the Savannah River Site for national defense purposes. The proton beam produces highly energetic neutrons through spallation in a tungsten target. Secondary spallation occurs in a lead blanket which results in an intense neutron source which is moderated for capture in He-3 to produce tritium. The study concludes that:

• the APT Facility would be ideal for producing diagnostic, therapeutic, and medical research radionuclides,
• many medical radionuclides could be produced with little or not effect on tritium production, and
• new and revolutionary techniques could be developed using radionuclides produced in the APT facility.

When the APT begins operation (~2007), production facilities in the US and abroad may be obsolete and inadequate to meet the growing demand. The APT facility could meet that demand. Results of scoping studies for several promising isotopes will be presented. In addition, the results of a workshop of nuclear medical leaders in which a consensus endorsement was developed for the incorporation of these capabilities in APT will be discussed.

TU4100
Operation of High-Power 8.6 and 17.1 GHz Coaxial Gyroklystrons*

W. LAWSON, B. HOGAN, M. CASTLE, V. L. GRANATSTEIN, M. REISER, X. XU, Institute for Plasma Research-U. of Maryland

At the University of Maryland, we have designed, constructed, and tested a number of gyroklystron tubes operating from X-Band to Ka-Band over the past several years [1]. The purpose of this effort is to examine the suitability of gyro-amplifiers as drivers for advanced accelerator applications such as the Next Linear Collider. With a 440 kV, 160-260 A beam, we were able to produce about 30 MW of peak power in 1 ms pulses near 20 GHz with a second-harmonic tube. The peak efficiencies were near 30% and the large-signal gains were about 30 dB. We are currently conducting a series of experiments with coaxial tubes which are designed to produce peak powers in excess of 100 MW in X- and Ku-Band, increasing the state-of-the-art by a factor of 3-5. Preliminary results have indicated peak powers in excess of 75 MW at 8.6 GHz in a three cavity first-harmonic gyroklystron tube with a gain near 30 dB and an efficiency near 32%. In this paper we will detail the experimental results of this tube and discuss designs and preliminary cold and hot test results of a 3-cavity second-harmonic device, which is expected to give comparable results at 17.14 GHz.

*Work supported by the U. S. Department of Energy.

[1] V. L. Granatstein and W. Lawson, "Gyro-Amplifiers as Candidate RF Drivers for TeV Linear Colliders," IEEE Trans. on Plasma Science, Vol. 24, pp. 648-665 (1996).

Classification Category: T04

TU4101
Experience at Fermilab with High Quantum Efficiency Photo-Cathodes for RF Electron Guns

A. FRY, E. HAHN, W. HARTUNG, M. KUCHNIR, FNAL; P. MICHELATO, D. SERTORE, INFN-Milano

As part of the A0 Photoinjector collaboration at Fermilab, a system was developed at Milano for coating molybdenum cathodes with a layer of cesium telluride, a photo-emissive material of high quantum efficiency (QE). The system incorporates a set of manipulator arms to transfer a cathode from the preparation chamber into a 1.3 GHz RF electron photo-gun while remaining in an ultra-high vacuum environment, in order to avoid the deleterious effects of residual gases on the QE. The electron gun has been operated with a photo-cathode for several months. The cathode preparation apparatus and techniques are described. Results of DC QE measurements in the preparation chamber and pulsed QE measurements in the gun are given. Measurements of the spatial variation in the QE over the cathode surface and of the time dependence in the QE are presented. The results of photo-cathode "rejuvenation" experiments are given.

Classification Category: T01

TU4102
A High Charge Photoinjector for the Pulsed Radiolysis Facility - ELYSE

M. GAILLARD, Universite de Paris-Sud; J.C. BOURDON, T. GARVEY, J. LE DUFF, LAL

The Physical Chemistry Department at the Universite de Paris-Sud and the Laboratoire de l'Accelerateur Lineaire (LAL) will collaborate on a dedicated radiolysis user facility named ELYSE. The irradiation of user's samples will be performed using electron beams of energies varying from 4 to 9 MeV, produced from a laser triggered RF gun. The accelerator group at LAL, benefitting from the experience gained with their experimental RF gun (CANDELA), are responsible for the design and construction of the accelerator. The nominal beam requirements for ELYSE are 1 nC pulses of 5 ps width (FWHM). However there is a strong scientific interest in obtaining bunch charges of 10 nC for the same pulse width. The need to extract such high charges neccesitates the use of Cs₂Te photocathodes with their high quantum efficiency. An essential user requirement is to keep the charge of the integrated dark current during the RF pulse width (3 microseconds) below 1% of the charge of the main beam. We will present the status of our studies aimed at the construction of an RF gun and its associated transport optics capable of achieving these challenging goals.

Classification Category: A01

TU4103
First Performance of the RFD Linac Structure*

D.A. SWENSON, F.W. GUY, K.R. CRANDALL, J.W. LENZ, W.J. STARLING, Linac Systems

A "Proof-of-Principle" prototype of the Rf-focused Drift tube (RFD) linac structure is under construction at Linac Systems. This prototype comprises a 25-keV proton ion source, an einzel-lens-based LEBT, a 0.65-m-long RFQ linac to 0.8 MeV, and a 0.35-m-long RFD linac to 2.5 MeV. These two linac structures, totaling 1 meter in length, will be resonantly coupled together and powered by a collection of planar triodes. A basic description of the RFD linac structure, the beam dynamics within the structure, the tools developed for its study, and the performance capabilities of the structure will be presented. Some description of the mechanical design of the components, the methods used for their fabrication, and the impact of these features on the cost of RFD-linac-based systems for scientific, medical and industrial applications will be presented. Performance tests on this prototype are scheduled to begin in June, 1998. The results of these early performance tests will be presented. * Supported by the U.S. National Institute of Mental Health (NIMH). Classification Category: A02

TU4104
The Performance of the 1.3 GHz Superconducting RF Cavities in the First Module of the Tesla Test Facility Linac

W.-D. MOELLER, DESY, for the TESLA Collaboration

The design goal on the 1.3 GHz 9-cell superconducting RF cavities for the Tesla Test Facility (TTF) is a gradient Eacc > 15 MV/m at a quality factor of Qo > 3e9. The cavities are operated in pulsed mode: 0.8 ms constant gradient with 10 Hz repetition rate. After the vertical acceptance test and prior to the assembly in the linac the cavities were tested in a horizontal cryostat fully equipped with helium vessel, high power input coupler, higher order modes coupler and tuning system. The first 120 MeV beam has been delivered successfully by the TTF. We report about the processing of the superconducting cavity system in the first module and the performance of the cavity system before, during and after operating the linac.

Classification Category: T05

TU4105

Contribution withdrawn.

TU4106
Simulation of Beam Dynamics Including Space Charge in Proton Linac with Errors*

D.V. GORELOV**, P.N. OSTROUMOV, INR

LANA computer code (Linear Accelerators Numerical Analysis), has been modified in order to study the random error effect on beam dynamics in a proton or heavy ion linac. The standard well known set of different errors of the accelerating and focusing channel including geometry misalignments in all directions, magnetic and electric field imperfections and accelerating field instabilities are included into Monte Carlo simulation of the beam dynamics. Tune-up procedures for the longitudinal and transverse motion are simulated. Partial as well as cooperative effects of the errors are analyzed. LANA code includes the space charge forces of the beam using certain approximation. A comparison of the errors effect on the beam parameters with and without space charge in the 60 mA proton linac is presented. As was found the error effect on beam dynamics is a main contribution on halo formation as well as on beam losses.

* Research was done in frames of the collaboration between INR, Moscow and KEK, Tsukuba and as part of the agreement between these institutions.
** At present time author has temporary position at NSCL, MSU.

Classification Category: D01

TU4107
Basis for Low Beam Loss in the High-Current APT Linac*

T. P. WANGLER, LANL; K. R. CRANDALL, TECHSOURCE; E. R. GRAY, F. L. KRAWCZYK, S. KURENNOY, G. P. LAWRENCE, and R. D. RYNE, LANL

We present evidence for the conclusion that the 100-mA, 1700-MeV APT proton linac design will meet its goals of low beam loss operation. This conclusion has three main bases: 1) extrapolation from our understanding of the performance of the 800-MeV LANSCE proton linac at Los Alamos, 2) our theoretical understanding of the dominant halo-forming mechanism in the APT accelerator from physics models and multiparticle simulations, including those using high performance computers, and 3) the conservative approach and key principles underlying the design of the APT linac, which are aimed at minimizing beam halo and providing large apertures to reduce beam loss to a very low value.

* Work supported by the U.S. Department of Energy.

Classification Category: D01

TU4108
Design Simulation for Spot Size Stabilization of ITS/DARHT*

T. J. T. KWAN, D. C. MOIR, B. G. DEVOLDER, C. S. SNELL, M. KANG, LANL

DARHT is a flash x-ray machine under construction at Los Alamos. The induction linac driven electron beam generates x-rays via the bremsstrahlung process in a converter target. It has been found that the electric field developed between the electron beam and the target is large enough to draw ions from the target surface. The ions provide fractional neutralization of the electron beam and cause it to pinch radially inward. Furthermore, due to the expansion of the ion column in the upstream direction the location of the pinch point moves in the same direction resulting in a temporally increasing radiographic spot size detrimental to the imaging process. The self-biased target concept was proposed by Kwan and his colleagues in which the electron charge deposited on the target generates an electric potential which can effectively limit the axial motion of the ion column and thereby stabilize the growth of the spot size. Detailed simulations of the beam dynamics have been carried out using MERLIN and MCNP, and stabilization was achieved at a bias potential of 360kV. A new target chamber was designed for the Integrated Test Stand (ITS) with beam energy 5.6 MeV and current up to 4 kA. Preliminary results indicate stabilization of the spot size at the predicted bias potential. Detailed simulation and experimental results will be presented.

* Work supported by the U.S. Department of Energy.

Classification Category: D01

TU4109
Modeling Beams with Elements in Phase Space*

E. M. NELSON, LANL

Conventional particle codes represent beams as a collection of macroparticles. An alternative is to represent the beam as a collection of current carrying elements in phase space. While such a representation has limitations, it may be less noisy than a macroparticle model, and it may provide insights about the transport of space charge dominated beams which would otherwise be difficult to gain from macroparticle simulations.

The phase space element model of a beam is described, and progress toward an implementation and difficulties with this implementation are discussed. A simulation of an axisymmetric beam using 1d elements in phase space is demonstrated.

* Work supported by DOE, contract W-7405-ENG-36.

Classification Category: D01