D. J. Liska

Design of the Accelerating Structures for FMIT

Design considerations and concepts are presented for the accelerating structures for the Fusion Materials Irradiation Test (FMIT) Facility. These structures consist of three major units: a 0.1- to 2-MeV radio-frequency quadrupole based on the Russian concept, a 2- to 35-MeV drift-tube linac made up of two separate tanks designed to generate either 20- or 35-MeV beams, and an energy dispersion cavity capable of spreading the energy of the beam slightly to ease thermal loading in the target. Because of probable beam activation, the drift-tube linac is designed so that alignment and maintenance do not require manned entry into the tanks. This conservatism also led to the choice of a conventional vacuum system and has influenced the choice of many of the rf interface components. The high-powered FMIT machine is very heavily beam loaded and delivers a 100-mA continuous duty deuteron beam to a flowing liquid lithium target. The power on target is 3.5 MW deposited in a 1 × 3 cm spot. Because of the critical importance of the low energy section of this accelerator on beam spill in the machine, a 5-MeV prototype will be constructed and tested at the Los Alamos Scientific Laboratory (LASL).

High-Energy Beam Transport in the Hanford FMIT Linear Accelerator

The High-Energy Beam Transport (HEBT) for the Hanford Fusion Materials Irradiation Test (FMIT) Facilitys Linear Accelerator must transport a large emittance, high-current, high-power continuous duty deuteron beam with a large energy spread. Both periodic and nonperiodic systems have been designed to transport and shape the beam as required by the liquid lithium target. An energy spreader system distributes the Bragg Peak within the lithium. A beam spreader and a beam stop have been provided for tune-up purposes. Characterizing the beam will require extensions of beam diagnostics techniques and noninterceptive sensors. Provisions are being made in the facility for suspending the transport system from overhead supports.

DC Power System for Deuteron Accelerator

The Fusion Materials Irradiation Test (FMIT) Facility dc power system provides excitation current for all linac and High-Energy Beam Transport (HEBT) quadrupole and bending magnets, excitation for horizontal and vertical beam steering, and current-bypass shunts.

High Radiation Zone Design of the FMIT High-Energy Beam Transport

The Fusion Materials Irradiation Test (FMIT) deuteron linac, operating at 35 MeV and 100 mA continuous duty, is expected to spill 3 ¿A/m and to lose 10 ¿A at specific bending-magnet positions. The major impact of this spill will be felt in the High-Energy Beam Transport (HEBT), where many beam-line components must be maintained. A modular design concept, that uses segmented termination panels remotely located from the modules, is being employed. Radiation-hardened quadrupoles can be opened, clamshell fashion, to release the water-cooled beam tube r replacement if there is beam damage or lithium contamination from the target. Termination panels contain electrical, water, and instrumentation fit-tings to service the module, and are positioned to allow room for neutron-absorbing shielding between the beamline and the panel. The modular construction allows laboratory prealignment and check-out of all components on a structural carriage and is adaptable to supporting gamma shields. Proper choice of beam tube materials is essential for controlling activation caused by beam spill.

Tuning and Pre-Beam Checkout of 805 MHz Side-Coupled Proton Linac Structures

tuning and checkout of each portion of the accelerator. Methods for tuning accelerating and side cavities before and after they are brazed together into tank sections are described. High power tests of transient behavior, long-term stability, and stopband vs duty factor are outlined and related to structure cooling and frequency errors. Particular attention is given to the problem of unflatness. Progress and pitfalls in measuring unflatness at high power are presented. Experiments to determine the cause and to correct unflatness are described.

Computer Design of UHF Power Amplifier Tubes

Improved computational methods have been developed which can reduce the trial and error process in the development of certain microwave power tubes that employ cavities with cylindrical symmetry. It has been found, for example, that the application of these computer codes to klystron cavities and internal cavity triodes reveal ways in which their performance can be improved and results in better understanding of experimental results. The computed results agree quite closely with measurements made of such quantities as frequency and operating efficiency.

Mechanical Design of the High-Energy Beam-Transport Line for the FMIT 2-MeV Accelerator

The beam-transport line for the high-power 2-MeV Fusion Materials Irradiation Test (FMIT) accelerator is one of the most heavily instrumented ever designed. A wide variety of diagnostics is required to accurately determine the characteristics of the beam that will ultimately be used. Because the machine is only 2 MeV, the packing factor in the high-energy beam transport (HEBT) is high, especially since full-scale FMIT-grade components are used where possible. The HEBTs mechanical design aspects and its instrumentation are described.

Radiation-Hardened Field Coils for FMIT Quadrupoles

Modern accelerators of the Fusion Materials Irradiation Test (FMIT) class deliver enormous power onto their targets. The high beam currents of such machines produce highly activating radiation fields from beam/target interaction and normal beam losses. The 100-mA deuteron beam from the FMIT accelerator produces a backstreaming fast-neutron flux of 1011 n/s-cm2 near the target. In addition, the neutron contribution from distributed beam spill of 3 ¿A/m along the rest of the machine prevents the use of epoxy resin potting materials in all magnet field coils above 10-MeV beam energies. Two special techniques for radiation-hardened field coils have been developed at Los Alamos for use on the FMIT accelerator. One technique uses vitreous enamel coatings on the conductors and appears attractive for the drift-tube quadrupoles. Another method uses a thermally efficient two-layer coil design that has solid mineral-insulated (MI) conductors with indirect cooling coils, all bonded together in a lead matrix. Test results are discussed, along with applications of the quadrupoles in the FMIT facility that reduce gamma exposures during maintenance periods.

The Particle Separator at Los Alamos

There is only one beam separator presently being developed at Los Alamos and this is for the EPICS channel at LAMPF. EPICS (Energetic Pion Channel and Spectrometer) operates from 20 to 300 MeV pion energy. Excessive electron and positron contamination exists in this channel from 20 to 50 MeV and heavy proton contamination persists over the rest of the range. Energy degraders will not work in EPICS due to the high energy resolution required, 1 part in 104. All the contaminating particles generate background, but the protons are particularly undesirable due to their strong ionization signal in detectors which otherwise compounds their ten to one predominance and reduces the signal-to-noise ratio to about 0.01 in experiments involving H+. Experiments comparing cross-sections with 1+ and Hreactions require that proton contamination be reduced by at least 100:1 over most of the energy range and electron and positton contaminations be reduced by at least 10:1 at the lowest energies. To accomplish these goals without loss of energy resolution, an active beam separator is being developed which operates on the electrostatic crossed-field principal. Despite this standard operational approach, the EPICS separator has several unusual features.

Accelerator Field Measurements at High Power

When a coupled chain of resonant cavities is driven at high peak or average power, it may not necessarily respond in the same way as it does at low power. Differences can be caused by such factors as thermal effects, glow discharge and multipactoring which are not normally present at the tuning power level. These factors may be strongly influenced by the geometry of the cavity chain. A continuing effort is being carried on at Los Alamos to make comparative measurements at both low and high power on the 805 MHz LAMPF accelerating structure. This structure and the procedures used in tuning it have been described elsewhere in these proceedings where it has been pointed out that tank-to-tank average fields must be tuned to ± 1-2% of each other for proper beam dynamics, and that the stopband of ?/2 mode resonantly coupled structures must be properly set for stable high-power operation. The question naturally arises as to whether the field distribution and stop-band in tanks tuned at low power remain unchanged at high power. Some differences have been seen earlier in the distribution of electric probe readings which were set very close to each other at low power and tended to scatter rather badly at high power.