ANL, Argonne, Illinois
SLAC, Menlo Park, California
In recent years, we have explored application to the Advanced Photon Source (APS) of Zholents'* crab-cavity-based scheme for production of short x-ray pulses. Work concentrated on using superconducting (SC) cavities in order to have a continuous stream of crabbed bunches and flexibility of operating modes. The challenges of the SC approach are related to the size, cost, and development time of the cavities and associated systems. A good case can be made for a pulsed system** using room-temperature cavities. APS has elected to pursue such a system in the near term, with the SC-based system planned for a later date. This paper describes the motivation for the pulsed system and gives an overview of the planned implementation and issues. Among these are overall configuration options and constraints, cavity design options, frequency choice, cavity design challenges, tolerances, instability issues, and diagnostics plans.
*A. Zholents et al., NIM A 425, 385 (1999).**P. Anfinrud, private communication.
ANL, Argonne, Illinois
ANL, Argonne, Illinois
The Intense Pulsed Neutron Source (IPNS) Rapid Cycling Synchrotron (RCS) accelerates 3.2x 1012 protons from 50 MeV to 450 MeV at 30 Hz. During the 14.2 ms acceleration period, the RF frequency varies from 2.21 MHz to 5.14 MHz. The beam current is limited by a vertical instability. By analyzing turn-by-turn Beam Position Monitor (BPM) data, large amplitude mode 0 and mode 1 vertical beam centroid oscillations were observed in the later part of the acceleration cycle. The oscillations develop in the tail of the bunch, build up and remain localized in the later part of the bunch. This vertical instability was compared with a head-tail instability that was intentionally induced in the RCS by adjusting the trim-sextupoles to make the horizontal chromaticity positive (below transition). It appears that our vertical instability is not typical head-tail instability. More data analysis and experiments were performed to characterize the instability.
BNL, Upton, Long Island, New York
Fermilab, Batavia, Illinois
ANL, Argonne, Illinois
Rutgers University, The State University of New Jersey, Piscataway, New Jersey
The A0 photoinjector is being reconfigured to test the principal of transverse to longitudinal emittance exchange as proposed by Emma et. al., Kim and Sessler, and others. The ability to perform such an exchange could have major advantages to FELs by reducing the transverse emittance. Several schemes to carry out the exchange are possible and will be reported separately. At the Fermilab A0 Photoinjector we are constructing a beamline to demonstrate this transverse to longitudinal emittance exchange. This beamline will consist of a dogleg, and a TM110 5 cell copper cavity followed by another dogleg. The beamline is designed to reuse the bunch compressor dipoles of the photoinjector, along with some existing diagnostics. Beamline layout and optics discussed along with inital data. Future possibilites of performing a similar experiment at the proposed NML facility at Fermilab are also discussed.
ANL, Argonne, Illinois
In the IPNS RCS, a single proton bunch (h=1) is accelerated from 50 MeV to 450 MeV in 14.2 ms. The bunch experiences an instability shortly after injection (<1 ms). During the first 1 ms, the beam is bunched but little acceleration takes place; thus, this period of operation is similar to that of a storage ring. Natural vertical oscillations (assumed to be tune lines) show the vertical tune to be rising toward the bare tune value, suggesting neutralization of space charge and a reduction of its detuning effects. Neutralization time near injection ranges from 0.25 ms - 0.5 ms, depending on the background gas pressure. Oscillations move from the LSB to the USB before disappearing. Measurements made with a recently installed pinger system show the horizontal chromaticity to be positive early but approaching zero later in the cycle. The vertical chromaticity is negative throughout the cycle. During pinger studies, two lines are observed, suggesting the formation of islands. Neutralization of the beam space charge implies the generation of plasma in the beam volume early in the cycle which may then dissipate as the time-varying electric fields of the beam become stronger.
ANL, Argonne, Illinois
The Advanced Photon Source storage ring is normally operated with 100 mA of beam current. A number of high-current studies were carried out to determine the multibunch instability limits. The longitudinal multibunch instability is dominated by the rf cavity higher-order modes (HOMs), and the coupled-bunch instability (CBI) threshold is bunch-pattern dependent. We can stably store 200 mA with 324 bunches, and the CBI threshold is 245 mA. With 24 bunches, several components are approaching temperature limits above 160 mA, including the HOM dampers. We do not see any CBI at this current. The transverse multibunch instabilities are most likely driven by the resistive wall impedance; there is little evidence that the dipole HOMs contribute. Presently, we rely on the chromaticity to stabilize the transverse multibunch instabilities. When we stored beam up to 245 mA, we used high chromaticity, and the beam was transversely stable. The stabilizing chromaticity was studied as a function of current. We can use these experimental results to predict multibunch instability thresholds for various upgrade options, such as smaller-gap or longer ID chambers and the associated increased impedance.
ANL, Argonne, Illinois
The horizontal bursting instability was first observed in a single bunch in the APS in 1998, soon after operation began. Above the instability threshold, the bursting is characterized by exponentially growing bunch centroid oscillations that saturate, then decay, repeating quasi-periodically. More recently, bursting was also observed with multiple bunches in both the horizontal and vertical planes, showing that this is not purely a single-bunch phenomenon. On the other hand, the multibunch instability threshold is strongly dependent on bunch spacing, and the dependence is markedly different for the two transverse planes. Depending on the bunch spacing, the bunch-to-bunch oscillations are sometimes coupled, sometimes not. In this paper, we discuss the threshold in terms of the chromaticity required to stabilize the beam. We present instability imaging data using a streak camera that shows the bunch-to-bunch oscillation phase, and turn-by-turn beam position histories that give the bursting time dependence for different bunch spacings. Finally, we discuss the machine impedance and measured tune shift with current.