Fermilab, Batavia
Kentech Instruments Ltd., Wallingford, Oxfordshire
A fast chopper capable of kicking single 2.5 MeV H- bunches, spaced at 325 MHz, at rates greater than 50 MHz is needed for the Fermilab High Intensity Neutrino Source (HINS). Four 1.2 kV fast pulsers, designed and manufactured by Kentech Instruments Ltd., will drive a ~0.5m long meander made from a copper plated ceramic composite. Test results showing pulses from the prototype 1.2 kV pulser propagating down the meander will be presented.
LANL, Los Alamos, New Mexico
The Los Alamos Neutron Science Center (LANSCE) accelerator is an 800 MeV proton linac that drives user facilities for isotope production, proton radiography, ultra-cold neutrons, weapons neutron research and various sciences using neutron scattering. The LANSCE Refurbishment Project (LANSCE-R) is an ambitious project to refurbish key elements of the LANSCE accelerator that are becoming obsolete or nearing end-of-life. The conceptual design phase for the project is funded and underway. The 5 year, $170M (US) project will enable future decades of reliable, high-performance operation. It will replace a substantial fraction of the radio-frequency power systems (gridded tubes and klystrons) with modern systems, completely refurbish the original accelerator control and timing systems, replace obsolete diagnostic devices, and modernize other ancillary systems. An overview of the LANSCE-R project will be presented. The functional and operating requirements will be discussed, the proposed technical solutions presented, and the plan for successful project execution while meeting annual customer expectations for beam delivery will be reviewed.
CEA, Bruyeres-le-Chatel
The design of a 15 MeV, 2 kA peak current, electron accelerator for the DEINOS project is presented. It is dedicated to a new radiographic facility. The accelerator design is based on a dc photo-injector and a rf superconducting linac. Up to twenty electron micro-pulses, 100 ps time duration and 200 nC bench charge are emitted at 352 MHz repetition rate from a CS2Te photocathode and accelerated to 2.5 MeV in the dc diode before injection into a superconducting linac. A general description of the main accelerator components and the beam dynamics simulations are presented.
Diversified Technologies, Inc., Bedford, Massachusetts
Funding: U.S. Department of Energy SBIR Program
Diversified Technologies, Inc. (DTI) is developing a driver for a kicker strip-line deflector which inserts and extracts charge bunches to and from the electron and positron damping rings of the International Linear Collider. The kicker driver must drive a 50 Ω terminated TEM deflector blade at 10 kV with 2 ns flat-topped pulses, which according to the ILC pulsing protocol, bursts pulses at a 3 MHz rate within one-millisecond bursts occurring at a 5 Hz rate. The driver must also effectively absorb high-order mode signals emerging from the deflector. In this paper, DTI will describe current progress utilizing a combination of high voltage DSRDs (Drift Step Recovery Diodes) and high voltage MOSFETs. The MOSFET array switch, without the DSRDs, is itself suitable for many accelerator systems with 10 - 100 ns kicker requirements. DTI has designed and demonstrated the key elements of a solid state kicker driver which both meets the ILC requirements, is suitable for a wide range of kicker driver applications. Full scale development and test are exptected to occur in Phase II of this DOE SBIR effort, with a full scale demonstration scheduled in 2009.
POSTECH, Pohang, Kyungbuk
PAL, Pohang, Kyungbuk
NFRI, Daejon
Funding: This work is supported by KAPRA and POSTECH Physics BK21 Program.
An intense L-band electron linear accelerator is now being commissioned at CESC (Cheorwon Electron-beam Service Center) for industrial applications. It is capable of producing 10 MeV electron beams with 30 kW average beam power. For a high-power capability, we adopted the traveling-wave structure operated with the 2π/3 mode at 1.3 GHz. The structure is powered by a 25 MW pulsed klystron with 60 kW average rf power. The rf pulse length is 7 μs while the beam pulse length is 6 μs due to the filling time in the accelerating structure. The accelerating gradient is 4.2 MV/m at the beam current of 1.45 A which is the fully beam-loaded condition. In this paper, we present details of the accelerator system and commissioning results.
IUAC, New Delhi
The frequency control of the superconducting quarter wave resonator at IUAC is currently accomplished by mechanical and electronic tuners which are operated in the time scale of seconds and hundreds of milliseconds to a few tens of microseconds respectively. Due to presence of microphonics, input rf power in the range 200-300 W was required to control the resonator for a typical field of 3-5 MV/m achieved with 6 watts dissipation. Implementation of a novel idea to damp the mechanical vibration with the help of SS-balls has helped to reduce rf power below 100 W. Though resonators are working fine at this power level, we are investigating whether further reduction of rf power is possible using a piezo actuator to control the drift of frequency. The piezo tuner working in hundreds of milli seconds range with the dynamic phase control scheme will share a substantial load from the electronic tuner. As a result, the resonator's phase lock loop will remain locked for less rf power. The initial test results of the piezo tuner will be presented.
GANIL, Caen
INFN/LNS, Catania
ALTEK, San Gregorio (CATANIA)
The SPIRAL2 LEBT line uses a single chopper situated in the line section common to protons, deuterons and A/Q=3 ions. The paper describes the design and the test of the power circuits, based on standard components and working up to 10 kV, at a 1 kHz repetition rate.
RRCAT, Indore (M.P.)
CERN, Geneva
Funding: Department of Atomic Energy, India.
CERN is building a 352.2 MHz, 3 MeV RFQ based Test Stand as first part of LINAC 4. Extending its collaboration with DAE of India, CERN approached us to design and develop a high voltage pulsed modulator for 1 MW LEP klystrons (planning their reuse). Three design schemes were proposed out of which an all solid state bouncer compensated modulator was chosen for follow up development. The main features of the modulator are: no gas tube crow-bar, all solid state construction low rise/fall times and high stability of the flat top. The major specifications are output voltage upto 110 kV, output current upto 24 Amp, pulse duration 800 μs, PRR 2Hz, pulse droop <1% and ripple on pulse top <0.1%. The energy in klystron arc is restricted to 10J. Based on these principles, a modulator has been developed and constructed at CERN and is currently undergoing tests with a klystron while another one with similar development is in the final stages of integration/evaluation at RRCAT. The present paper describes the topology, simulation results, protection strategy and integration aspects of the pulse modulator and would briefly summarize the results.
The work is done under DAE CERN Collaboration under NAT Protocol.
RIKEN Spring-8 Harima, Hyogo
RIKEN/SPring-8, Hyogo
A klystron modulator power supply for XFEL project at SPring-8 has been developed, which concepts are a compact body, a low noise, and a good stability. The cooling system of the power supply is one of the most important key for the stable modulator. For example, temperature change of insulation oil in the tank caused drift of the klystron voltage, and higher oil temperature deteriorates insulation oil and electric components. We adopted simple and compact cooling systems utilizing natural conviction cooling, because of low costs, limited space, and maintenance free. In order to estimate the requisite cooling ability, we designed four types of cooling panels and measured the natural conviction heat transfer coefficient between the oil and each cooling panels. Using the results, we designed cooling systems composed of water cooling panels placed on the side walls and a water pipe hanged from the ceiling panel. The temperature of the inner oil of the power supply in the rated operation was suppressed below 43 degree C, which is agreed with our expectation. In this paper we present the design and ability of the power supply, and the key point of oil cooling.
KEK, Ibaraki
SUT, Noda-shi, Chiba
University of Tokyo, Tokyo
The multi beam electron tube with a lot of beam pipes is required for the low applied voltage and the high frequency because the efficiency has a limit according to the perveance. However, the total heater power becomes too high if many thermal cathodes are used. Thus the cold cathode such as the carbon nano tube (CNT) is suitable for such a multi beam electron tube. Further the cold cathode has the advantage to work as a switching device since the metal grid close to the cathode can be used. The design and the fundamental test of the partial model will be presented.
Diversified Technologies, Inc., Bedford, Massachusetts
Funding: U.S. Department of Energy SBIR Program
Diversified Technologies, Inc. (DTI) has developed high power, solid-state Marx Bank designs for a range of accelerator and collider designs. We estimate the Marx topology can deliver equivalent performance to conventional designs, while reducing acquisition costs by 25-50%. In this paper DTI will describe the application of Marx based technology to two different designs: a long-pulse ILC focused design (140 kV, 160 A, 1.5 ms), and a short-pulse design (500 kV, 265 A, 3 us). These designs span the known requirements for future accelerator modulators. For the ILC design, the primary challenge is minimizing the overall size and cost of the storage capacitors in the modulator. For the short-pulse design, the primary challenge is high speed operation, to limit the energy lost in the pulse rise-time while providing a very tight (± 3%) voltage flattop. Each design demands unique choices in components and controls, including the use of electrolytic capacitors in the ILC Marx design. This paper will review recent progress in the development and testing of both of these prototype Marx designs, being built under two separate DOE Phase II SBIR grants.
LANL, Los Alamos, New Mexico
Funding: Work supported by the NNSA, U. S. Department of Energy under contract DE-AC52-06NA25396.
The LANSCE Refurbishment Project (LANSCE-R) will revitalize the LANSCE accelerator infrastructure. Much of the equipment has been in use for over 36 years and is approaching the end of its design lifetime. As obsolescence issues make like-for-like replacements increasingly more expensive, modern systems with lower costs become a reasonable alternative. As part of the LANSCE-R project, four of the seven HV power supplies for the 805 MHz rf klystrons will be replaced. The present and future requirements for these power supplies influence the selection of replacement options. Details of the HV power supply replacement requirements and the different replacement options will be discussed. One option is to use four 95 kV, 21 A dc power supplies originally installed nearby as part of the Low Energy Demonstration Accelerator (LEDA) project. Significant material and labor cost savings can be achieved by leaving these supplies installed where they are and building a HV transport system to bring high voltage power from the existing LEDA facility to the LANSCE facility. The different replacement options will be compared based on material and labor costs as offset by long-term energy savings.
LANL, Los Alamos, New Mexico
The Los Alamos Neutron Science Center (LANSCE) is in the conceptual design phase of a refurbishment project that will sustain reliable facility operations well into the next decade. The LANSCE accelerator was constructed in the late 1960s and early 1970s and is a national user facility that provides pulsed protons and spallation neutrons for defense and civilian research and applications. The refurbishment will focus on systems that are approaching "end of life" and systems where modern upgrades hold the promise for significant operating cost savings. The current baseline consist of replacing all the 201 MHz rf amplifiers, replacing greater than 75% of the 805 MHz rf systems with a combination of high efficiency klystrons and new klystrons of the existing style, replacing four high voltage systems, and replacing all the low level rf cavity field control systems along the accelerator. System designs and requirements will be presented and the project plan will be discussed.
SLAC, Menlo Park, California
ORNL, Oak Ridge, Tennessee
Funding: Work supported by the U.S. Department of Energy under contract DE-AC05-00OR22725
The rf source High Voltage Converter Modulator systems installed on the Spallation Neutron Source have operated well in excess of 200,000 hours, during which time numerous failures have occurred. An improved IGBT switch plate is under development to help mitigate these failures. The new design incorporates three significant improvements. The IGBTs are upgraded to 4.5 kV, 1200 A, press-pack devices, which increase the voltage margin, facilitate better cooling, and eliminate explosive disassembly of the package in the event of device failure. The upgrade to an advanced IGBT gate drive circuit decreases switching losses and improves fault-condition response. A common-mode choke is incorporated into the H-bridge to decrease dI/dt during a shoot-through condition, to further improve the circuit response to this fault condition. The upgrade design and development status will be presented.
SLAC, Menlo Park, California
Funding: Work supported by the U.S. Department of Energy under contract DE-AC02-76SF00515
A Marx-topology klystron modulator is under development as an 'Alternative Conceptual Design' for the International Linear Collider project. It is envisioned as a smaller, lower cost, and higher reliability alternative to the present, bouncer-topology, 'Baseline Conceptual Design'. The application requires 120 kV (±0.5%), 140 A, 1.6 ms pulses at a rate of 5 Hz. The Marx constructs the high voltage pulse by combining, in series, a number of lower voltage cells. The Marx employs solid state elements; IGBTs and diodes, to control the charge, discharge and isolation of the cells. The developmental testing of a first generation prototype, P1, is nearing completion. Development of a second generation prototype, P2, is underway. Status updates for both prototypes will be presented.
RIKEN/SPring-8, Hyogo
The 8 GeV C-band electron linear accelerator is under construction at the SPring-8 site aiming at generating an FEL X-ray beam in 2010. C-band accelerator technology has been developed initially at KEK for the e+e- linear collider project, and employed at the XFEL project in Japan. Since C-band generates a high gradient acceleration field as high as 35 MV/m, the total length of the accelerator fits within 400 m, including the injector and three bunch compressors. C-band uses normal conducting rf technology, thus it runs in pulse mode at 60 Hz, which is well suited to XFEL operation and is less expensive. The talk will cover the current status of the XFEL project and hardware production.
