A generalized fast feedback system has been developed to stabilize beams at various locations in the SLC. The system is designed to perform measurements and change actuator settings to control beam states such as position, angle and energy on a pulse to pulse basis. The software design is based on the state space formalism of digital control theory. The system is database-driven, facilitating the addition of new loops without requiring additional software. A communications system, KISNet, provides fast communications links between microprocessors for feedback loops which involve multiple micros. Feedback loops have been installed in seventeen locations throughout the SLC and have proven to be invaluable in stabilizing the machine.

The small beam sizes at the interaction point of a X-band linear collider require mechanical stabilization of the final focus magnets at the nanometer level. While passive systems provide adequate performance at many potential sites, active mechanical stabilization is useful if the natural or cultural ground vibration is higher than expected. A mechanical model of a room temperature linear collider final focus magnet has been constructed and actively stabilized with an accelerometer based system.

In order to stabilize the beams at the interaction point, the X-band linear collider proposes to use a combination of techniques: inter-train and intra-train beam-beam feedback, passive vibration isolation, and active vibration stabilization based on either accelerometers or laser interferometers. These systems operate in a technologically redundant fashion: simulations indicate that if one technique proves unusable in the final machine, the others will still support adequate luminosity. Experiments underway for all of these technologies have already demonstrated adequate performance.

One of the options for controlling vibration of the final focus magnets in a linear collider is to use active feedback based on accelerometers. While commercial geophysics sensors have noise performance that substantially exceeds the requirements for a linear collider, they are physically large, and cannot operate in the strong magnetic field of the detector. Conventional nonmagnetic sensors have excessive noise for this application. We report on the development of a non-magnetic inertial sensor, and on a novel commercial sensor both of which have demonstrated the required noise levels for this application.

Experiments at the FLASH linac at DESY have demonstrated that the Higher Order Modes induced in Superconducting Cavities can be used to provide a variety of beam and cavity diagnostics. The centers of the cavities can be determined from the beam orbit which produces minimum power in the dipole HOM modes. The phase and amplitude of the dipole modes can be used as a high resolution beam position monitor, and the phase of the monopole modes to measure the beam phase relative to the accelerator RF. Beam orbit feedback which minimizes the dipole HOM power in a set of structures has been demonstrated. For most SC accelerators, the existing HOM couplers provide the necessary signals, and the downmix and digitizing electronics are straightforward, similar to those for a conventional BPM.

The Linac Coherent Light Source (LCLS) will be the world’s first x-ray free-electron laser (FEL). To ensure the vitality of FEL lasing, it is critical to preserve the high quality of the electron beam during acceleration and compression. The peak current and final energy are very sensitive to system jitter. To minimize this sensitivity, a longitudinal feedback system on the bunch length and energy is required, together with other diagnostics and feedback systems (e.g., on transverse phase space). Here, we describe a simulation framework, which includes a realistic jitter model for the LCLS accelerator system, the RF acceleration, structure wakefield, and second order optics. Simulation results show that to meet the tight requirements set by the FEL, such a longitudinal feedback system is mandatory.

The PEP2 (e+)(e-) storage rings contain many complex interrelated systems. When the beam aborts, examining a record of the orbit from the time just before the abort can help identify the root cause. At the Stanford Linear Accelerator Center (SLAC) a system has been developed to continuously record beam orbits from Beam Position Monitors (BPMS) into a circular buffer. When the beam is aborted the buffers are frozen and their contents are stored for later analysis. BPM orbits are saved on a turn by turn basis for 2800 turns in both the high energy ring (HER) and the low energy ring (LER). Each BPM Processor (BPMP) can either monitor the HER or the LER, but not both as the readout of the two rings is multiplexed into a single readout channel. Tools exist as part of the SLAC Control Program (SCP) to collect, display, and save the data. A physicist or operator can choose a few BPMS in which to view all 2800 turns to identify the turn in which the beam went awry; then ask for that specific orbit from all of the BPMS in the storage ring to determine the root cause of the abort.

We report on experimental results from the December 2003/January 2004 data run of the Feedback On Nanosecond Timescales (FONT) experiment at the Next Linear Collider Test Accelerator at SLAC. We built a second-generation prototype intra-train beam-based feedback system incorporating beam position monitors, fast analogue signal processors, a feedback circuit, fast-risetime amplifiers and stripline kickers. We applied a novel real-time charge-normalization scheme to account for beam current variations along the train. We used the system to correct the position of the 170-nanosecond-long bunchtrain at NLCTA. We achieved a latency of 53 nanoseconds, representing a significant improvement on FONT1 (2002), and providing a demonstration of intra-train feedback for the Linear Collider.

Keeping nanometer-sized beams in collision is an essential component in achieving design luminosity in a linear collider. The NLC stabilization strategy is conservative by including enough redundancy so that if some part does not work to specification or the incoming beam motion is worse than expected, the beams will still be kept in collision. We show simulation results with both realistic and pessimistic assumptions about the response of the ground motion, inertial stabilization, intratrain and intertrain feedback systems. By providing backup systems, and by allowing some systems to perform more poorly than design, it is possible to have a high level of confidence that the beams can be stabilized successfully.

For high luminosity in electron-positron linear colliders, it is essential to generate low vertical emittance beams. We report on the smallest vertical emittance achieved in single-bunch-mode operation of the Accelerator Test Facility, which satisfies the requirement of the x-band linear collider. The emittances were measured with a laser-wire beam-profile monitor installed in the damping ring. The bunch length and the momentum spread of the beam were also recorded under the same conditions. The smallest vertical rms emittance measured at low intensity is 4 pm at a beam energy of 1.3 GeV, which corresponds to the normalized emittance of 1.0x1.0(-8) m. It increases by a factor of 1.5 for a bunch intensity of 10(10) electrons. The measured data agreed to the calculation of intrabeam scattering within much better than a factor of 2.

Extensive beam-based feedback systems are planned as an integral part of the Next Linear Collider (NLC) control system. Wakefield effects are a significant influence on the feedback design, imposing both architectural and algorithmic constraints. Studies are in progress to assure the optimal selection of devices and to refine and confirm the algorithms for the system design. We show the results of initial simulations, along with evaluations of system response for various conditions of ground motion and other operational disturbances.

First results and beam measurements are presented for the recently installed linac bunch compressor chicane. The new bunch compressor produces ultra-short electron bunches for the Sub-Picosecond Photon Source (SPPS) and for test beams such as the E164 Plasma Wakefield experiment. This paper will give an overview of the first experiences with tuning and optimizing the compressor together with a description of the beam diagnostics and beam measurements. These measurements form the basis for further detailed study of emittance growth effects such as CSR and wakefields in a previously unmeasured regime of ultra-short bunch lengths.

The linac coherent light source (LCLS) is an X-ray FEL project with a 1-nC electron bunch compressed to an rms length of 20 microns at 4.5 GeV, accelerated in 500 meters of SLAC linac to 15 GeV, and then injected into an undulator to generate SASE radiation. The longitudinal wakefield generated by the short bunch in the (S-band) linac is very strong, and is relied upon to cancel the energy chirp left in the beam after bunch compression. Up to now, both the average and the shape of the longitudinal wake of the SLAC linac have been measured and confirmed using bunches ranging down to an rms 500-microns in length. The recent installation of a chicane in the SLAC linac for the sub-picosecond photon source (SPPS), however, allows compression of a 3.4-nC bunch down to 50 mum rms length. We present measurements of the average wakefield, for bunch lengths down to this, LCLS-type scale, and compare with theory

Future Linear Colliders require nanometer stability of the beams at the interaction point. One approach to stabilising the beams is to use feedback based on inertial sensors (accelerometers/seismometers) to control the positions of the final focus magnets. Commercial seismometers developed for geo-science applications have sufficient noise performance (nanometer noise down to a fraction of a hertz), but due to their large size and magnetic sensitivity are unsuitable for use in a linear collider detector. We report on the development of a high sensitivity, compact, non-magnetic inertial sensor for this application. In addition to its use in linear colliders, the sensor is also expected to have application in vibration measurement and control in synchrotron light sources.

To fully characterize the luminosity performance of a linear collider, it is important to simulate the effects of ground motion as well as the beam-beam feedback and other stabilization systems planned to compensate for that motion. The linear collider simulation codes have recently been extended to include both ground motion models and stabilization systems to support the work of the International Linear Collider Technical Review Committee (TRC). This paper discusses the implementation details and the optimization strategies for interpulse beam-beam feedback.