BNL, Upton, Long Island, New York
The Brookhaven AGS Main Magnet Power Supply (MMPS) is a thyristor control supply rated at 5500 Amps, ±9000 Volts. The peak magnet power is 50 MWatts. The power supply is fed from a motor/generator manufactured by Siemens. The generator is 3 phase 7500 Volts rated at 50 MVA. The peak power requirements come from the stored energy in the rotor of the motor/generator. The motor generator is about 45 years old and Siemens is not manufacturing similar machines in the future. We are therefore investigating different ways of storing energy for future AGS MMPS operation. This paper will present simulations of a power supply where energy is stored in capacitor banks. Two dc to dc converters will be presented. The switching elements would be IGCT's made by ABB. The simulation program used is called PSIM Version 6.1. The control system of the power supply will also be presented. The average power from the Long Island Power Authority (LIPA) into the power supply will be kept constant during the pulsing of the magnets at ±50 MW. The reactive power will also be kept constant below 1.5 MVAR. Waveforms will be presented.
BNL, Upton, Long Island, New York
Dual partial snake scheme has provided polarized proton beams with 1.5*1011 intensity and 65% polarization for RHIC spin program. To overcome the residual polarization loss due to horizontal resonances in the AGS, a new string of quadrupoles have been added. The horizontal tune can now be set in the spin tune gap generated by the two partial snakes, such that horizontal resonances are avoided. This paper presents the accelerator setup and preliminary results.
BNL, Upton, Long Island, New York
ALBA, Bellaterra (Cerdanyola del Valles)
Since 2001 RHIC has experienced electron cloud effects, which have limited the beam intensity. These include dynamic pressure rises – including pressure instabilities, a reduction of the stability threshold for bunches crossing the transition energy, and possibly slow emittance growth. We report on the main observations in operation and dedicated experiments, as well as the effect of various countermeasures including baking, NEG coated warm pipes, pre-pumped cold pipes, bunch patterns, scrubbing, and anti-grazing rings.
BNL, Upton, Long Island, New York
Full spin flip in the presence of full Siberian snake has been achieved by using an rf dipole or solenoid as spin flipper at IUCF and COSY. This technique requires one to change the snake configuration to move the spin tune away from half integer. However, this is not practical for an operational high energy polarized proton collider like RHIC where beam lifetime is sensitive to small betatron tune change. An new technique of achieving full spin flip with the spin tune staying at half integer is proposed. This paper presents the design of RHIC spin flipper along with simulation results.
BNL, Upton, Long Island, New York
Snake depolarizing resonances due to the imperfect cancellation of the accumulated perturbations on the spin precession between snakes were observed at the Relativistic Heavy Ion Collider~(RHIC). During the RHIC 2005 and 2006 polarized proton runs, we mapped out the spectrum of odd order snake resonance at Qy=7/10. Here, Qy is the beam vertical betatron tune. We also studied the beam polarization after crossing the 7/10th resonance as a function of resonance crossing rate. This paper reports the measured resonance spectrum as well as the results of resonance crossing.
The work was performed under the US Department of Energy Contract No. DE-AC02-98CH1-886, and with support of RIKEN(Japan) and RenaissanceTechnologies C orp.(USA)
BNL, Upton, Long Island, New York
To achieve the RHIC polarized proton enhanced luminosity goal of 150*1030 cm-2 sec-1 on average in stores at 250 GeV, the luminosity needs to be increased by a factor of 3 compared to what was achieved in 2006. Since the number of bunches is already at its maximum of 111, limited by the injection kickers and the experiments' time resolution, the luminosity can only be increased by either increasing the bunch intensity and/or reducing the beam emittance. This leads to a larger beam-beam tuneshift parameter. Operation during 2006 has shown that the beam-beam interaction is already dominating the luminosity lifetime. To overcome this limitation, a near-integer working point is under study. We will present recent results of these studies.
BNL, Upton, Long Island, New York
We explore the possibility of using two non-scaling FFAG with a smaller number of distributed RF cavities for a high power heavy ion driver. The pulsed heavy ion source would consist of an Electron Beam Ion Source (EBIS), fed continuously from a high charge state Electron Cyclotron Resonance (ECR) source. The Radio Frequency Quadrupole (RFQ) and a short 10 MeV/u linac would follow the ion source. Microseconds long heavy ion beam bunches from the EBIS would be injected in a single turn into a multi-pass small aperture non-scaling Fixed Field Alternating Gradient (FFAG) accelerator. The heavy ion maximum kinetic energy is assumed to be 400 MeV/u with a total of 400 kW power for uranium ion beams. Partially stripped heavy ions would be accelerated from 10 MeV/u to 67 MeV/u with a first non-scaling FFAG, while, after further stripping, a second non-scaling FFAG would accelerate from 67 to 400 MeV/u.
BNL, Upton, Long Island, New York
After the last successful RHIC Au-Au run in 2004 (Run-4), RHIC experiments now require significantly enhanced luminosity to study very rare events in heavy ion collisions. RHIC has demonstrated its capability to operate routinely above its design average luminosity per store of 2x1026 cm-2 s-1. In Run-4 we already achieved 2.5 times the design luminosity in RHIC. This luminosity was achieved with only 40% of bunches filled, and with β* = 1 m. However, the goal is to reach 4 times the design luminosity, 8x1026 cm-2 s-1, by reducing the beta* value and increasing the number of bunches to the accelerator maximum of 111. In addition, the average time in store should be increased by a factor of 1.1 to about 60% of calendar time. We present an overview of the changes that increased the instantaneous luminosity and luminosity lifetime, raised the reliability, and improved the operational efficiency of RHIC Au-Au operations during Run-7.
BNL, Upton, Long Island, New York
ITEP, Moscow
The Relativistic Heavy Ion Collider~(RHIC) as the first high energy polarized proton collider was designed to provide polarized proton collisions at a maximum beam energy of 250GeV. It has been providing collisions at a beam energy of 100GeV since 2001. Equipped with two full Siberian snakes in each ring, polarization is preserved during the acceleration from injection to 100GeV with careful control of the betatron tunes and the vertical orbit distortions. However, the intrinsic spin resonances beyond 100GeV are about a factor of two stronger than those below 100GeV making it important to examine the impact of these strong intrinsic spin resonances on polarization survival and the tolerance for vertical orbit distortions. Polarized protons were accelerated to the record energy of 250GeV in RHIC with a polarization of 45\% measured at top energy in 2006. The polarization measurement as a function of beam energy also shows some polarization loss around 136GeV, the first strong intrinsic resonance above 100GeV. This paper presents the results and discusses the sensitivity of the polarization survival to orbit distortions.
BNL, Upton, Long Island, New York
AES, Princeton, New Jersey
CLASSE, Ithaca
Stony Brook University, StonyBrook
Jefferson Lab, Newport News, Virginia
In this paper we present status and plans for the 20-MeV R&D energy recovery linac, which is under construction at Collider Accelerator Department at BNL. The facility is based on high current (up to 0.5 A of average current) super-conducting 2.5 MeV RF gun, single-mode super-conducting 5-cell RF linac and about 20-m long return loop with very flexible lattice. The R&D ERL, which is planned for commissioning in 2008, aims to address many outstanding questions relevant for high current, high brightness energy-recovery linacs.
BNL, Upton, Long Island, New York
Gold ions for the 2007 run of the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory (BNL) are accelerated in the Tandem, Booster and AGS prior to injection into RHIC. The setup and performance of this chain of accelerators will be reviewed with a focus on improvements in the quality of beam delivered to RHIC. In particular, more uniform stripping foils between Booster and AGS, and a new bunch merging scheme in AGS promise to provide beam bunches with reduced longitudinal emittance for RHIC.
BNL, Upton, Long Island, New York
There is significant interest in RHIC heavy ion collisions at c.m. energies of 5-50 GeV/u, motivated by a search for the QCD phase transition critical point. The low end of this energy range is well below the nominal RHIC injection c.m. energy of 19.6 GeV/u. There are several challenges that face RHIC operations in this regime, including longitudinal acceptance, magnet field quality, lattice control, and luminosity monitoring. We report on the status of work to address these challenges and include results from beam tests of low-energy RHIC operations with protons and gold.
BNL, Upton, Long Island, New York
Tech-X, Boulder, Colorado
ORNL, Oak Ridge, Tennessee
AES, Princeton, New Jersey
Fermilab, Batavia, Illinois
Jefferson Lab, Newport News, Virginia
BINP SB RAS, Novosibirsk
JINR, Dubna, Moscow Region
The physics interest in a luminosity upgrade of RHIC requires the development of a cooling-frontier facility. Detailed cooling calculations have been made to determine the efficacy of electron cooling of the stored RHIC beams. This has been followed by beam dynamics simulations to establish the feasibility of creating the necessary electron beam. Electron cooling of RHIC at collisions requires electron beam energy up to about 54 MeV at an average current of between 50 to 100 mA and a particularly bright electron beam. The accelerator chosen to generate this electron beam is a superconducting Energy Recovery Linac (ERL) with a superconducting RF gun with a laser-photocathode. An intensive experimental R&D program engages the various elements of the accelerator: Photocathodes of novel design, superconducting RF electron gun of a particularly high current and low emittance, a very high-current ERL cavity and a demonstration ERL using these components.
BNL, Upton, Long Island, New York
Acceleration of electrons up to 10 GeV for a future electron-ion collider eRHIC (Relativistic Heavy Ion Collider) could be performed with the energy recovery linac with multiple passes. An energy recovery scheme is required if a superconducting linac is used for acceleration. We report on an attempt to make a combination of a multi-pass linac with non-scaling Fixed Field Alternating Gradient (NS-FFAG) arcs. Two NS-FFAG arcs would allow electrons to pass through the same structure with different energies. The beam will be accelerated by the superconducting linac at the top of the sine function, and returned to the front of the linac by the non-scaling FFAG. This process is repeated until the total energy of 10 GeV is reached. After collisions the beam is brought back by the NS-FFAG and decelerated before being dumped.
BNL, Upton, Long Island, New York
To further improve the the polarized proton (pp) luminosity in the Relativistic Heavy Ion Collider, the beta functions at the two interaction points (IPs) will be reduced from 1.0 m to 0.9m in 2007. In addition, it is planned to increase the bunch intensity from 1.5*1011 to 2.0*1011. To accommodate these changes, the nonlinear chromaticities and the third resonance driving term should be corrected. In 2007, the number of the arc sextupole power supplies will be doubled from 12 to 24, which allows nonlinear chromaticity correction. With the updated field errors in the interaction regions (IRs), detailed dynamic aperture studies are carried out to optimize the nonlinear correction schemes, and increase the available tune space in collision.
IUCF, Bloomington, Indiana
BNL, Upton, Long Island, New York
Two partial helical dipole snakes were found to be able to overcome all imperfection and intrinsic spin resonances provided that the vertical betatron tunes were maintained in the spin tune gap near the integer 9. Recent vertical betatron tune scan showed that the two weak resonances at the beginning of the acceleration cycle may be the cause of polarization loss. This result has been confirmed by the vertical polarization profile measurement, and spin tracking simulations. Possible cure of the remaining beam polarization is discussed.
BNL, Upton, Long Island, New York
The AGS synchrotron employs two partial helical snakes* to preserve the polarization of the proton beam during acceleration in the AGS. The effect of the helical snakes on the beam optics is significant at injection energy, with the effect greatly diminishing early in the acceleration cycle. In order to compensate for the effect of the snakes on the beam optics, we have introduced eight compensation quadrupoles in straight sections of the AGS at the proximity of the partial snakes. At injection the strength of these eight quads is set at a high value but ramped down to zero when the effect of the snakes diminishes. Four of the compensation quadrupoles had to be placed in very short straight sections therefore had to be 'thin' with a length of ~30 cm. The 'thin' quadrupoles were laminated and designed to minimize the strength of the dodecoupole harmonic. The thickness of the lamination was also calculated** to keep the ohmic losses generated by the eddy currents in the laminations below an acceptable limit. Comparison of the measured and calculated harmonics will be presented and the ohmic losses due to the eddy currents, as a function of time during rumping will be discussed.
* H. Huang, et al., Proc. EPAC06, (2006), p. 273.** OPERA computer code. Vector Fields Inc.