Download 433_1.PDF

34 GHz, 45 MW Pulsed Magnicon
Oleg A. Nezhevenko*, Michael A. LaPointe*, Vyacheslav P. Yakovlev*,
Jay L. Hirshfielcf, Gennady V. Serdobintsevt1i, Gennady I. Kuznetsov11,
Boris Z. Persov11, and Alexander Fix*
* Omega-P, Inc., New Haven, CT 06511, USA
Yale University, New Haven, CT 06520, USA
1
Budker Institute of Nuclear Physics, Novosibirsk, 630090 Russia
^Institute of Applied Physics, Nizhny-Novgorod, 603600 Russia
f
Abstract. A high efficiency, high power magnicon at 34.272 GHz has been designed and built
as a microwave source to develop RF technology for a future multi-TeV electron-positron linear
collider. The tube is designed to provide a peak output power of ~45 MW in a 1 microsecond
pulse, with a gain of 55 dB, using a 500 kV, 220 A, 1 mm-diameter electron beam. The status of
the tube itself as well as the near-term experimental program is presented.
INTRODUCTION
The motivation for development of a millimeter-wave magnicon for a future
normal conducting collider arises from an expectation of exploiting the customarilyassumed scaling (in rough proportion to frequency) of the dark current limit [1] for the
maximum accelerating gradient that can be sustained by a copper accelerating
structure. The maximum gradient will be constrained below the dark current limit by
rf breakdown [1], and further limited in practice by surface fatigue due to pulsed
heating that can affect structure lifetime [2]. Further experimental tests to determine
the maximum achievable accelerating gradient under a variety of conditions must be
carried out before the absolute limits will be known.
The magnicon is an RF source, based on circular deflection of an electron beam,
whose main features are high power and high efficiency [3,4]. These properties make
the tube especially attractive for accelerator applications. Furthermore, since RF
cavities in the magnicon are significantly larger than in a klystron at the same
operating frequency, magnicons can be designed for higher peak and average power.
The first magnicon developed at Budker INP was a fundamental harmonic
amplifier at 915 MHz. It operated with an efficiency of 73% using a 300 kV, 12A
electron beam [5]. The measured output power was 2.6 MW and the pulse width was
30 jisec. In experimental tests also at Budker INP [6], a second harmonic magnicon
amplifier operating at 7.0 GHz achieved an output power of 55 MW in a 1.1 jisec
pulse, and a repetition rate of 3 Hz, with a gain of 72 dB and an efficiency of 56%.
CP647, Advanced Accelerator Concepts: Tenth Workshop, edited by C. E. Clayton and P. Muggli
© 2002 American Institute of Physics 0-7354-0102-0/02/$19.00
433
This device is driven at 3.5 GHz and uses a 430 kV, 230 A beam with an area
compression ratio of about 2300:1 [7]. These encouraging results prompted Omega-P,
in close collaboration with the Naval Research Laboratory (NRL), to undertake
design, construction and evaluation of an 11.424 GHz, 60 MW pulsed magnicon [8].
At present, the tube is conditioned for power levels of up to 15 and 25 MW for 1.0 and
0.2 jisec pulse widths respectively, and already has been used for high power tests of
active RF pulse compressors and dielectric-lined accelerating structures [9].
34 GHZ MAGNICON AMPLIFIER
In scaling magnicon amplifiers to higher frequencies (consequently, smaller
physical dimensions), a few design problems arise at high power due to the limitations
imposed by cathode loading, breakdown field and pulse heating of the cavity walls.
The concept of a third harmonic magnicon amplifier is introduced to overcome these
limitations [10]. In general, a magnicon (similarly to a klystron) consists of four
major components, namely: electron gun, magnet, RF system and beam collector.
The resulting design parameters of this amplifier are given in Table 1.
Table 1; 34.3 GHz magnicon parameters.
Operating frequency, GHz
Power, MW
Pulse duration, us
Repetition rate, Hz
Efficiency, %
Drive frequency, GHz
Drive power, W
Gain, dB
Beam voltage, kV
Beam current, A
Beam diameter, mm
Magnetic field, deflecting cavities, kG
Magnetic field, output cavity, kG
34.272
44-48
1.5
10
41-45
11.424
150
54
500
215
0.8-1.0
13.0
22.5
The complete engineering design [11] of the 34.3 GHz magnicon is presented in
Fig. 1.
The gun design [12] calls for a cathode current density of 12 A/cm2, and a
maximum surface electric field strength of 238 kV/cm on the focus electrode. Beam
compression in this gun is only partially electrostatic (500:1). Higher electrostatic
compression would lead to a higher electric field at the focus electrode, and would
require a magnetic field of about 13 kG at the edge of the pole piece, leading to
undesirable saturation in the iron. Thus a magnetic compression of about 2:1 occurs
as the beam passes through the hole in the pole-piece and into a ~5 kG field, and a
further factor of 3:1 occurs adiabatically as the magnetic field gradually rises to 13 kG.
The resulting compression ratio of 3000:1 is comparable to the 2300:1 compression
ratio for the 7 GHz magnicon [7,8]. It is found in the 34.3 GHz magnicon gun design
that 95% of the current is within a diameter of 0.8-mm [12].
434
Before assembling the magnicon tube, the gun and beam collector were assembled
and been tested up to the design power of 100 MW in jisec pulses at the Yale Beam
Physics Lab [12]. Initial conditioning up to ~515 kV was carried out without beam
current. To achieve this, a matched load was connected to the primary of the pulse
transformer. After cold conditioning, the gun was conditioned and tested hot up to
-480 kV and -200 A. To reach 500 kV, small modifications in the modulator are
required. The measured beam current is in excellent agreement with the design value,
with differences within the measurement error or better than ±2%. This indicates that
the critical gun dimensions are within required tolerances and consequently, that the
electron optics should be close to the design.
FIGURE 1 34.3 GHz magnicon amplifier tube: 1-electron gun, 2-RF system, 3-output waveguide
(WR28), 4-WR90 waveguide, 5-superconducting coils, 6-iron yoke, 7- cryostat, 8-beam collector.
435
The RF system consists of seven cavities: one drive (#1) and three gain cavities (#24), two "penultimate" cavities (#5-6), and one output cavity (#7). The shapes and
dimensions of the cavities are chosen to avoid monotron self-excitation of
axisymmetric modes, and harmonic frequency modes [11]. All cavities of the
deflection system are about 1.25 cm long and their diameters are about 3.0 cm. There
are four WR90 waveguides built in the body of deflecting system. One of them is for
the drive cavity, and the rest are for measurements in cavities #3, 5 and 6. These
waveguides are also used for pumping. The length of the output cavity (3.15 cm) and
its shape were optimized to achieve maximum efficiency, absence of parasitic
oscillations, and acceptable surface electric fields [10]. The diameter of the output
cavity is about 1.75 cm. Power extraction is by means of a set of four WR28
waveguides with an azimuthal separation A0=7i/2 that couple to both field
polarizations [10]. The RF system is made as a brazed monoblock that allows baking
up to 400° C. Each of the four magnicon output is loaded by a waveguide watercooled vacuum load preceded by a directional coupler.
A superconducting solenoid (see Fig. 1) provides a magnetic field of 13 kG in the
deflection system and 23 kG in the output cavity. The magnet coils consist of three
independently driven sections for adjustment of the magnetic field profile. At present,
the magnet is at the final stages of adjustment prior to installation.
Presently, the magnicon is assembled, cold tested, baked out and is under the
operating vacuum of ~5-10~9 Torr. The RF tests are expected to commence
immediately after the magnet is installed. A photograph of the assembled magnicon is
shown in Fig. 2.
Experimental Program
Operation of the 34.3 GHz magnicon will allow the establishment of a unique
accelerator class test facility, which gives the possibility of providing a wide range of
investigations including:
a)
Research on the breakdown of accelerating structures of different designs.
One example could be a 19-cell structure [13] with minimized electric and
magnetic field enhancement. The magnicon could test it up to an
accelerating gradient of 300 MeV/m.
b)
Experimentation on metal fatigue caused by pulse heating, and
consequently determination of accelerating structure longevity. The use of
-2 MW of magnicon power will allow a temperature rise of approximately
500 C° to be reached at the surface of novel designed test cavity.
c)
Development of high power components to drive the test structures and
cavities that were recently started [14]. The list includes the following:
output window(s), (probably in an overmoded structure), mode converters
436
for
for the output window(s),
window(s), phase
phase shifters,
shifters, power
power splitters
splitters and
and power
power
combiners,
combiners, low-loss
low-loss transmission
transmission lines,
lines, etc.
etc.
d)
d)
For
For the
the most
most advanced
advanced breakdown
breakdown experiments
experiments with
with traveling-wave
traveling-wave
accelerating
accelerating structures,
structures, compression
compression of
of RF
RF pulses
pulses isis inevitably
inevitably required
required and
and
is
is included
included in
in future
future plans.
plans. [15].
[15].
FIGURE
FIGURE 2.
2. The
The assembled
assembled 34.3
34.3 GHz
GHz magnicon.
magnicon.
ACKNOWLEDGMENTS
ACKNOWLEDGMENTS
This
This work
work was
was supported
supported by
by the
the Division
Division of
of High
High Energy
Energy Physics,
Physics, U.S.
U.S. Department
Department
of
Energy.
of Energy.
437
REFERENCES
1. Palmer, R. B., in Pulsed RF Sources for Linear Colliders, edited by R. C. Fernow, AIP Conf. Proc.
No. 337 (AIP, New York, 1995), pp. 1-15.
2. O.A. Nezhevenko, "On the Limitation of Accelerating Gradient in Linear Colliders Due to the Pulse
Heating", PAC97, Vancouver, 1997, p.3013.
3. O.A. Nezhevenko, "Gyrocons and Magnicons: Microwave Generators with Circular Deflection of
the Electron Beam," IEEE Trans, Plasma Sc,, vol.22 pp. 765-772, October 1994.
4. O.A. Nezhevenko, "Recent Developments in High-Power Magnicons for Particle Accelerators,"
Physics of Plasmas, vol.7 pp. 2224-2231, May 2000.
5. M.M. Karliner, E.V. Kozyrev, I.G. Makarov, O.A. Nezhevenko, G.N. Ostreiko, B.Z. Persov, and
G.V. Serdobintsev, " The Magnicon-An Advanced Version of the Gyrocon," NucL lustrum.
Methods, Phys, Res., vol. A269, pp. 459-473, 1988.
6. E.V. Kozyrev, O.A. Nezhevenko, A.A. Nikiforov, G.N. Ostreiko, G.V. Serdobintsev, S.V.
Schelkunoff, V.V. Tarnetsky, V.P. Yakovlev, LA. Zapryagaev, "Present Status on Budker INP 7
GHz Pulsed Magnicon," AIP Conference Proc, 474, (American Institute of Physics, Melville, NY,
1999), p. 187.
7. Yu.V. Baryshev, I.V. Kazarezov, E.B. Kozyrev, G.I. Kuznetsov, I.G. Makarov, O.A. Nezhevenko,
B.Z. Persov, M.A. Tiunov, V.P. Yakovlev, and LA. Zapryagaev. "A 100 MW electron source with
extremly high beam area compression", Nuclear Inst. and Methods in Phys. Research A 340 (1994),
pp. 241-258.
8. O.A. Nezhevenko, V.P. Yakovlev, J.L. Hirshfield, S.H. Gold, A.W. Fliflet, A.K. Kinkead, "Status of
X-Band Pulsed Magnicon Project," PAC2001, Chicago, June 17-22, 2001,pp. 1023-1025.
9. S.H. Gold, A.K. Kinkead, O.A. Nezhevenko, V.P. Yakovlev, J.L. Hirshfield, A.L. Vikharev, O.A.
Ivanov, S.V. Kuzikov, A.M. Gorbachev, W. Gai, R. Konecny, and J.G. Power, "High Power
Accelerator R&D at the NRL 11.424-GHz Magnicon Facility," these Proceedings.
10. O.A. Nezhevenko, V.P. Yakovlev, A.K. Ganguly, J.L. Hirshfield, "34GHz pulsed Magnicon for
linear colluder application", PAC97, Vancouver, 1997, p.3156.
11. O.A. NezhevenkoJVI.A. LaPointe, S.V. Schelkunoff, V.P. Yakovlev, J.L. Hirshfield, E.V. Kozyrev,
G.I. Kuznetsov, B.Z. Persov, and A. Fix, " 34 GHz Pulsed Magnicon Project", AAC2000
Proceedings, Santa-Fe, June, 2000, AIP Conference Proc. 569, Woodbury, N.Y., 2001, pp. 786-796.
12. V.P. Yakovlev, O.A. Nezhevenko, J.L. Hirshfield, M.A. LaPointe, M.A. Batazova, G.I. Kuznetsov,
"100 MW Electron Gun For A 34.3 GHz Magnicon," PAC2001, Chicago, June 17-22, 2001,
pp.1041-1043.
13.O.A. Nezhevenko, V.P. Yakovlev, J.L. Hirshfield, G.V. Serdobintsev, S.V. Schelkunoff, B.Z.
Persov, "34.3 GHz Accelerating Structure For High Gradient Tests," PAC2001, Chicago, June 1722, 2001, pp. 3849-3851.
14.G.G. Denisov, et. al, "Study of Ka-band components for a future high-gradient linear accelerator,"
these Proceedings
15. A. L. Vikharev, et. al, "Quasi-optical microwave pulse compressor at 34 GHz," these Proceedings
438