A High Level RF (HLRF) system consisting of power amplifiers (PA's) and ferrite loaded cavities is being built by Brookhaven National Laboratory (BNL) for the Spallation Neutron Source (SNS) project. Four cavities were built and are being tested. Each cavity has two gaps with a design voltage of 10 kV per gap and will be driven by a PA directly adjacent to it. The PA uses a 600 kW tetrode to provide the necessary drive current. All the PA's were built and are being tested at BNL prior to shipping to ORNL. A dynamic tuning scheme used to help compensate for the effect of beam loading was implemented and tested.

A high level RF system (HLRF) consisting of power amplifiers (PA's) and ferrite loaded cavities is being designed and built by Brookhaven National Laboratory (BNL) for the Spallation Neutron Source (SNS) project. It is a fixed frequency, two harmonic system whose main function is to maintain a gap for the kicker rise time. Three cavities running at the fundamental harmonic (h=l) will provide 40 kV and one cavity at the second harmonic (h=2) will provide 20 kV. Each cavity has two gaps with a design voltage of 10 kV per gap and will be driven by a power amplifier (PA) directly adjacent to it. The PA uses a 600kW tetrode to provide the necessary drive current. The anode of the tetrode is magnetically coupled to the downstream cell of the cavity. Drive to the PA will be provided by a wide band, solid state amplifier located remotely. A dynamic tuning scheme will be implemented to help compensate for the effect of beam loading.

Two AC dipoles with vertical and horizontal magnetic field have been proposed at RHIC for applications in linear and non-linear beam dynamics and spin manipulations. A magnetic field amplitude of 380 Gm is required to produce a coherent oscillation of 5 times the rms beam size at the top energy. We take the AC dipole frequency to be 1.0% of the revolution frequency away from the betatron frequency. To achieve the strong magnetic field with minimum power loss, an air-core magnet with two seven turn winding of low loss Litz wire resonating at 64 kHz is designed. The system is also designed to allow one to connect the two magnet winding in series to resonate at 37 kHz for the spin manipulation. Measurements of a half length prototype magnet are also presented.

Average beam currents of 40 A will be present in the Spallation Neutron Source. Even though the entire cycle time is only one synchrotron oscillation the longitudinal phase space determines peak beam current and momen- tum spread. Both factors play a role in space charge and instability dynamics. Longitudinal simulations with beam loading and longitudinal space charge have been done in the design phase.

Two AC dipoles with horizontal and vertical oscillating magnetic fields will be installed in RHIC. They will provide coherent oscillations for beam dynamic studies and betatron function measurements. The AC dipole with horizontal magnetic field will also be used to induce a full spin flip for RHIC polarized proton experiments. This note discusses the applications of the AC dipoles in RHIC and their expected parameters.

are done to determine the length of the power coupler inner conductor and pickup probe for the Nb cavity at the fixed axial location. Size and location of both the fundamental power coupler and the pickup probe can be decided from the simulation results.

The rf system for the synchrotrons of the spallation neutron source is designed to accelerate 1.4 {times} 10{sup 14} protons/pulse to an energy of 3.6 GeV. Injection energy is 600 MeV. The synchrotron repetition frequency is 30 Hz, with a 50% duty factor. The choice of operating frequency is somewhat arbitrary. The authors propose a low frequency of 1.3 to 1.6 MHz, which is the second harmonic of the revolution frequency. The advantages of such a low frequency system are: (1) There will be two bunches in the machines and the time between bunches will be sufficiently long to allow for the rise time of the extraction kicker. No missing bunches will be necessary, which simplifies injection, and transient beam loading problems are avoided. (2) With only two bunches there are no unstable coupled-bunch modes of longitudinal instability. (3) In multi-gap low frequency cavities the transient time factor is essentially unity because the rf wavelength is much longer than the cavity dimensions. (4) Cavities in this low frequency range are basically lumped-element type structures, where the sources of the inductance and capacitance are clearly identified. This allows effective control of higher order mode impedances in such cavities. (5) Ferrite-loaded low-frequency cavities are necessarily low impedance structures; ferrites are lossy. This low impedance makes it possible to achieve system stability without large amounts of feedback in a heavily beam loaded system. (6) BNL has a good deal of experience in building rf systems in this range of frequency, voltage, and power level. This report outlines the essential parameters of a practical rf system for the synchrotrons of the Spallation Neutron Source. The design uses materials, ferrites and vacuum tubes, that are commercially available and with which the laboratory has recent experience.

The AGS operates a varied program of proton, heavy ion, and polarized proton acceleration for fixed-target experiments and will soon serve as the injector of these beams into the Relativistic Heavy Ion Collider, RHIC. The new Booster synchrotron extends the range of intensities and masses that can be accelerated. The 1.5 GeV injection energy increases the space charge limit by a factor of four to more than 6 x 1013 protons per pulse. To accommodate the increased beam current the RF system will be upgraded to provide more power and lower impedance to the beam. The flexibility of the RF system will also be enhanced by virtue of a new RF beam control system and installation of individual tuning servos for the ten RF cavities

To reach the design intensity of 1.5/spl times/10/sup 13/ protons per pulse in the AGS Booster, transverse coupled bunch instabilities with an estimated growth rate of 1500s/sup -1/ have to be dampened. A prototype transverse damper has been tested successfully using a one turn digital delay and closed orbit suppression implemented in a programmable gate array. An updated damper, which includes an algorithm to optimize damping for a changing betatron tune, will also be presented.< >

The Band II RF system was originally built to support the Booster operations during the acceleration of heavy ions. Designed to sweep from 0.6 to 2.5 MHz, it was build and successfully tested over a much broader range reaching 4 MHz. Voltages up to more than 20 kV were reached over the design frequency range. The system consists of two stations, each of which is made of one single gap cavity directly driven by a grounded cathode push pull power amplifier. The low Q high permeability ferrites needed in the coaxial cavity in order to reach the lower end of the band make tuning extremely easy. Both systems were thoroughly tested both at single frequencies and on a sweep and are now installed in the ring,ready for operations. Static measurements showed no high-loss effects. The results of the “bench” tests that lead to important performance improvements are given

It has been proposed that the AGS Heavy Ion/Proton Booster be excited directly from the electric power distribution system without intervening an energy storage buffer such as an MG set or a magnetic energy buffer. The average power requirement of the AGS Booster is less than many single-loads presently housed on the lab site. However, the power swing will be the largest single pulsating load on the lab site. The large power swings will impact on the power grid producing utility-line disturbances such as voltage fluctuations and harmonic generation. Thus, it is necessary to carefully evaluate the quality of the electric power system resulting from the interconnection, such that the utility system is not degraded either on the lab site or at LILCO`s substation.

load and the NSLS is at the 69 KV level. However, on the days that the interference was first observed at the NSLS only one-half of the substation transformers at Temple Place were in service. The 13.8 KV tie breaker was closed and the full substation load was supplied from this common bus. Thus the coupling between the pulsating magnet load and the NSLS was at the 13.8 KV level. Establishing the normal two bus configurations at Temple Place appeared to reduce the disturbance. These events suggested a controlled experiment to measure the magnet power swing and the induced powerline flicker; and from these measurements project the flicker on the lab site generated by the Booster operating at full energy. This experiment could corroborate the validity of the electrical models used in analyzing the power flow from the LILCO power grid and its distribution on the Lab site described in Accelerator Division Technical Note 220.

LILCO is studying the effects of the AGS Booster power swing on its power grid. The study is being conducted by GE systems Development and Engineering, Schenectady, New York. In notes, dates November 10, 1987, prepared for a GE-LILCO Progress Review Meeting, the author notes LILCO system resonances that are excited by the heavy ion cycle. The data used by GE for their study, is the power flow required for continuous operation of the Booster, namely a continuous 13MW Power swing and a period of one second. The data used by GE came from BNL reports, used to analyze the power line flicker generated by this pulsating load. It is a worse case study and does not represent the Booster cycle. The Booster must be synchronized with the AGS, which is operated with a period of 3 seconds, when accelerating heavy ions. Thus the Booster duty cycle is 1/3 with a peak power swing of 13MW. The time of one second used to cycle the Booster magnets is arbitrary and can be increased to a maximum of three seconds. The peak power swing and the power spectrum are modified by the Booster duty cycle and period. The spectrum is critical for the GE study of the LILCO grid.

The Alternating Gradient Synchrotron (AGS) booster has two RF systems covering a frequency range from 600 kHz to 4.2 MHz. The low frequency system, also called Band II, operates on a frequency range from 600 kHz to 2.4 MHz. The low frequency system consists of two single gap cavities that sweep from 0.6 to 2.4 MHz, with gap voltages up to 17 kV. The cavities are loaded with TDK SY7 high permeability ferrite rings. Tuning will be accomplished by changing the permeability of the ferrite with DC bias currents as high as 200 A flowing in two figure of eight windings. The cavity will be driven by a 200 kW push-pull tetrode power amplifier, which will be driven by remotely located solid state drivers. Commissioning of the booster is underway with proton beams.<>

A high-level RF system, including a power amplifier and cavity, has been designed and built for the Alternating Gradient Synchrotron (AGS) Booster. It covers a frequency range of 2.4 to 4.2 MHz and will be used to accelerate high-intensity protons, low-intensity polarized protons and heavy ions to the 1.5-GeV level. A total accelerating voltage of up to 90 kV will be provided by two cavities, each having two gaps. The internally cross-coupled, pushpull cavities are driven by an adjacent power amplifier. In order to accommodate beam intensities of up to 0.75*10/sup 13/ protons per bunch, a low plate resistance power tetrode is used. The tube anode is magnetically coupled to one of the cavity's two parallel cells. The amplifier is a grounded cathode configuration driven by a remotely located solid-state amplifier. It has been tested in the laboratory at full gap voltage with satisfactory results.<>