SINAP, Shanghai
The Shanghai Synchrotron Radiation Facility (SSRF) is an intermediate energy, third generation light source. In December 2007, electron beam was stored and accumulated in the SSRF storage ring. Since then the accelerator commissioning and beam line installation have continued toward the scheduled user operation from May 2009 onwards. This paper presents an overview of the SSRF status and preparations for user operations.
INFN/LNF, Frascati (Roma)
LAL, Orsay
CERN, Geneva
roma3, Rome
Rome University La Sapienza, Roma
BINP SB RAS, Novosibirsk
KEK, Ibaraki
BINP, Novosibirsk
SLAC, Menlo Park, California
INFN-Roma, Roma
In 2007 DAΦNE was upgraded to operate in a regime of large Piwinski angle, with a novel IR optics, reduced vertical beta at the interaction point, and additional sextupoles providing for crab waist collisions. The specific luminosity was boosted by more than a factor of four, and the peak luminosity was more than doubled with respect to the maximum value obtained with the original collider configuration. The DAΦNE commissioning as well as the first experience with large Piwinski angle and crab waist collisions scheme will be reported.
UCLA, Los Angeles, California
We report on the use of a ultrashort high brigthness relativistic beam from the UCLA Pegasus laboratory RF photoinjector source for probing matter transformation at the atomic scale with sub-100 fs time resolution. The high accelerating gradient and the relativistic electron energy allow to pack more than 107 electrons in less than 100 fs bunch length, enabling the study of irreversible ultrafast phenomena by single-shot diffraction patterns. The experimental setup, and the initial results from the first ever relativistic electron diffraction time-resolved study will be discussed.
KEK, Ibaraki
A 6-m cryomodule, which includes four Tesla-like 9-cell cavities, was assembled and installed in the STF tunnel in April, 2008. After cooldown of the cryomodule, high power tests of four cavities had been carried out at 2 K from September to December, 2008. A cavity package consists of a 9-cell niobium cavity with two HOM couplers, an input coupler with a cold and a warm rf window, and a frequency tuning system with a mechancal and a piezo tuner. The performance as a total sc cavity system was checked in the cryomodule test with high rf power. One of the cavities was achieved a stable pulsed operation at 32 MV/m higher than the specific operating gradient (31.5 MV/m) in ILC. The maximum accelerating gradients (Eacc,max) obtained in the vertical cw tests was maintained or slightly improved in the cryomodule tests with a pulsed operation of 1.5 msec and 5 Hz. Compensation of Lorentz force detuning at 31 MV/m was successfully demonstrated by using piezo tuner and pre-detuning.
Muons, Inc, Batavia
Fermilab, Batavia
Funding: Supported in part by USDOE Contract. DE-AC05-84-ER-40150 and by FRA DOE contract number DE-AC02-07CH11359
Magnetrons are low-cost highly-efficient microwave sources, but they have several limitations, primarily centered about the phase and frequency stability of their output. When the stability requirements are low, such as for medical accelerators or kitchen ovens, magnetrons are the very efficient power source of choice. But for high energy accelerators, because of the need for frequency and phase stability–-proton accelerators need 1-2 degrees source phase stability, and electron accelerators need .1-.2 degrees of phase stability–-they have rarely been used. We describe a novel variable frequency cavity technique which will be utilized to phase and frequency lock magnetrons.
CLASSE, Ithaca, New York
Funding: Work supported by NSF Grant PHY 0131508
The RF power required to phase-stabilize the Cornell University ERL main linac cavities is expected to be driven by microphonic-noise. To reduce the required RF power we are exploring the possibility of active compensation of cavity microphonic noise with the cavities in the Cornell ERL injector cryomodule. The Cornell ERL injector cryomodule houses five elliptical 2-cell SRF cavities developed for the acceleration of a high current (100mA) ultra-low emittance beam and is currently undergoing extensive testing and commissioning. Each of the five cavities is equipped with a blade tuner; each blade tuner integrates 4 piezoelectric actuators and vibration sensors for the active compensation of cavity detuning. This paper presents first results of active frequency-stabilization experiments performed with the Cornell ERL injector cryomodule cavities and their integral blade/piezoelectric fast tuners.
Fermilab, Batavia
Euclid TechLabs, LLC, Solon, Ohio
KEK, Ibaraki
The accelerating gradient required for the ILC project exceeds 30 MeV/m. With current technology the maximum acceleration gradient in SC structures is determined mainly by the value of the surface RF magnetic field. In order to increase the gradient, the RF magnetic field is distributed homogeneously over the cavity surface (low-loss structure), and coupling to the beam is improved by introducing aperture "noses" (re-entrant structure). These features allow gradients in excess of 50 MeV/m to be obtained for a singe-cell cavity. Further improvement of the coupling to the beam may be achieved by using a TW SC structure with small phase advance per cell. We have demonstrated that an additional gradient increase by up to 46% may be possible if a pi/2 TW SC structure is employed. However, a TW SC structure requires a SC feedback waveguide to return the few GW of circulating RF power from the structure output back to the structure input. Advantages and limitations of different techniques of exciting the traveling wave in this structure are considered, including an analysis of mechanical tolerances. We also report on investigations of transient processes in the SC TW structure.
KEK, Ibaraki
Kyoto ICR, Uji, Kyoto
The inspections of the superconducting RF cavities seem essential in achieving high accelerating gradient. The Kyoto camera system is a good tool to survey a defect location and to be analysis a defect shape in the inner surface of the superconducting rf cavities. The cavity inspections of the AES, ACCEL, ZANON and STF Baseline cavities were inspected to study relations between a defect shape and a heating gradient of the superconducting rf cavities. The STF Baseline #5 and #6 cavities with each surface treatment (as received, after Pre-EP, after EP-1, and after vertical test with EP-2) were inspected to trace a changing spots shape. The full inspection of the EBW seam, the HAZ (heat affected zone) and hot spots region were carried out before EP-2 process and a vertical test then the shape analysis of a discovered spots was done. The vertical tests of these cavities with T-map of fixed 9-cell type were measured at STF from September 2008. The inspection and shape analysis of these cavities were made after vertical tests for based on T-map data. The result of vertical tests and changing a shape of a discovered spots with EP-2 process will be presented.
Diversified Technologies, Inc., Bedford, Massachusetts
Funding: U.S. Department of Energy SBIR Program
Constructing and supporting a wide range of RF amplifiers for research accelerators at is costly at present. This is because amplifiers to date have been designed for a single application, and have little commonality in their design and control interfaces. Diversified Technologies, Inc. (DTI) is developing a modular RF amplifier design for a wide range of amplifier requirements. Amplifiers built on this model have common design, controls, and spares, independent of frequency or power. The amplifier design combines a solid-state RF driver, power conditioning, and controls with a high-power vacuum electronic device, giving high performance at a low cost. In this paper, DTI will describe results of the first implementation of the amplifier, which delivers 20 kW CW at 704 MHz.
HRC, Santa Clara, California
Funding: Work performed under the auspices of the U.S. Department of Energy SBIR Grant No. DE-FG02-08ER85197.
The salient features and design parameters of a dual resonant ring system configured for evaluating the high gradient performance of 11.424 GHz TW linear accelerator structures are presented; and the inherent rapid protection mechanism that automatically limits energy deposition during breakdown of the structure, and minimizes RF source reflections, is discussed. The diagnostic characteristics of the RF bridge load monitors and their unique capability of detecting the power imbalance caused by a feedback loop phase change of less than 2 parts in 10000, representing a 2 to 3 degree phase change of the linac structure, is described. The transient and steady-state power apportionment within the ring system is analyzed; and, in considering initial high power tests using an 18-cavity CLIC/KEK/SLAC structure, the results indicate that the demonstration of an unloaded average accelerating gradient of 108 MV/m will require a source power of 26 MW.
KEK, Ibaraki
The super-conducting RF test facility (STF) at KEK has been functional since 2005, and the STF phase-I, which involves the testing of a cryomodule with four superconducting cavities, was performed at the end of 2008. In this test, intense study of the power distribution system for the possible linear collider scheme was performed. Linear power distribution and tree-like distribution were compared and also the effects of eliminating circulator are studied. Current status of RF source of KEK STF are reported.
FZK, Karlsruhe
CAP Konstanz, Konstanz
Max-Planck Institute for Metal Research, Stuttgart
In a synchrotron radiation source coherent synchrotron radiation is emitted when the bunch length is comparable to the wavelength of the emitted radiation. To generate coherent THz (far IR) radiation, the ANKA storage ring is operated regularly with a dedicated low-alpha optics. Typical effective pulse lengths are of the order of 1 ps and below. In order to characterize the THz emission and beam oscillations in this mode a femtosecond laser system has been set up. This allows resolving the Terahertz electrical field by electro-optical sampling in a ZnTe crystal. The laser system consists of a 500 MHz repetition rate oscillator that can be phase locked to the repetition rate of the synchrotron. First results are presented. In contrast to previous approaches the high repetition rate is used in conjunction with a high frequency detection scheme in order to significantly increase the sensitivity of the detection. The discussion will concentrate on the limits in synchronization by locking the laser to either the bunch clock, a stripline signal in the ring or the visible light emission co-propagating with the THz radiation. The observations are compared to calculated pulse shapes.
KEK, Ibaraki
Two synchrotron light sources of the Photon Factory storage ring (PF-ring) and the Photon Factory advanced ring (PF-AR) have been stably operated at KEK. PF-ring covers the photon-energy range from VUV to hard X-ray using a 2.5 GeV (sometimes 3.0 GeV) electron beam. PF-AR is mostly operated in a single-bunch mode of 6.5GeV to provide pulsed hard X-rays. Recently, the operation has progressed to realize a so-called top-up injection at PF-ring. In a single-bunch mode, the continuous injection to preserve a constant beam current of 51 mA has been carried out since February 2007. In addition, the injection with continuing the experiments has been successfully operated in a multi-bunch mode since October 2008. At PF-AR, sputter ion pumps have been extensively reinforced to prolong the beam lifetime and to reduce the frequency of sudden lifetime drops by substituting for distributed ion pumps, which are considered as one of the dust sources. In this conference, we present the recent progress of the operation at PF-ring and PF-AR including machine developments.
ASCo, Clayton, Victoria
Argonne National Laboratory, Office of Naval Research Project, Argonne
Monash University, Faculty of Science, Victoria
SLAC, Menlo Park, California
This paper describes the implementation of a neural network hybrid controller for energy stabilization at the Australian Synchrotron Linac. The structure of the controller consists of a neural network (NNET) feed forward control, augmented by a conventional Proportional-Integral (PI) feedback controller to ensure stability of the system. The system is provided with past states of the machine in order to predict its future state, and therefore apply appropriate feed forward control. The NNET is able to cancel multiple frequency jitter in real-time. When it is not performing optimally due to jitter changes, the system can successfully be augmented by the PI controller to attenuate the remaining perturbations.
NSRRC, Hsinchu
A low run-out rotating sample stage is under development to realize a precise resolution within 30 nm on the horizontal plane for the end-station of transmission X-ray microscope (TXM) at NSRRC. The main assembly consists of a commercial rotation stage with run-out less than 1 μm, six capacitive sensors, one master ball, one flat and a horizontal adjusting stage. Error sources (including the profile of the master ball, run-out of the master ball in horizontal and vertical directions, flat plate) are separated from stage and the sensor readings can be down to the nanometer level. A feedback method is proposed to compensate the systematic errors and keeps the samples with little run-out and axial motion in the level of several tens nanometer. The details and tests of the rotation stage are presented in this paper.
ANL, Argonne
Funding: Work supported by U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
The APS storage ring uses DC/DC converters to power the magnets. High resolution for current regulation is desired for future improvement. It is calculated that at least 20- to 21-bit digital pulse width modulation (DPWM) is required in the proposed digital control system. This paper proposes a digital control system that adopts a new DPWM topology to achieve 21-bit DPWM without gigahertz system clock. The proposed topology uses a combination of a field-programmable gate array (FPGA) and a serializer chip TLK2541 from TI. The FPGA calculates the desired PWM signals and sends them to the TLK2541 chip. Then, the TLK2541 generates corresponding high-resolution DPWM pulses. An FPGA development board has been used to develop a prototype system to verify the proposed DPWM generation topology. This paper discusses the circuit topology and the experiment results.
LBNL, Berkeley, California
Funding: This work was supported by the Director, Office of Science, U. S. Department of Energy under Contract No. DE-AC02-05CH11231.
The Advanced Light Source (ALS) is a third generation synchrotron light source that has been operating since 1993 at Berkeley Lab. Recently, the ALS was upgraded to achieve Top-Off Mode, which allows injection of 1.9GeV electron beam into the Storage Ring approximately every 30 seconds. Modifications required for Top-Off operation included replacing the booster dipole and quadrupole magnet power supplies to increase the peak booster beam energy from 1.5GeV to 1.9GeV. Each new power supply was originally controlled by an analog controller that performs the current feedback loop and, in concert with other modules in the control chassis, determines the output of the ramped power supply. The new digital power supply controller performs the current feedback loop digitally to provide greater output stability and resolution. In addition, it provides remote monitoring of feedback loop signals, interlocks, and status signals, as well as remote control of the power supply operation via Ethernet. This paper will present the ALS Digital Power Supply Controller module requirements and design.
NSRRC, Hsinchu
Corrector magnets of TLS storage ring are served with linear power supplies (corrector magnet power supplies), with some modifications the long-term output current stability and ripple of these linear power supplies were improved from 500 ppm to 50 ppm. But these linear power supplies are very low efficiency、low power factor and about 20Hz low frequency response bandwidth that waste power、noisy and unable to serve fast orbit correction. MCOR30 is a modular switching power converter with smaller volume、high efficiency and above 100Hz frequency response bandwidth, replacing these linear power supplies with MCOR30s that could save power and increasing orbit correction response.
NSRRC, Hsinchu
The Agilient 6682A power supply is used as a dipole magnet power supply of Booster to storage ring (BTS) transport line, its output current stability is less than 100 ppm although specification is 1000 ppm. The performance of Agilient 6682A is quite good for TLS operational requirement but not suitable for less than 10 ppm output current stability general requirement of power supplies of TPS. Circuitry modification of Agilient 6682A to reach less than 10 ppm output current stability is hard to implement; but utilize analog programming function of Agilient 6682A with adding an external current control loop the output current stability of Agilient 6682A could be improved to less than 10 ppm.
KEK, Tsukuba
Tokyo City University, Tokyo
NIRS, Chiba-shi
J-PARC (Japan Proton Accelerator Research Complex) is a new accelerator facility to produce MW-class high power proton beams at both 3GeV and 50GeV. The Main Ring (MR) of J-PARC can extract beams to the neutrino beam line and the slow extraction beam line for Hadron Experimental Facility. The slow extraction beam is used in various nuclear and particle physics experiments. A flat structure and low ripple noise are required for the spills of the slow extraction. We are developing the spill control system for the slow extraction beam. The spill control system consists of the extraction quadrupole magnets and feedback device. The extraction magnets consist of two kinds of quadrupole magnets, EQ (Extraction Q-magnet) which make flat beam and RQ (Ripple Q-magnet) which reject the high frequent ripple noise. The feedback system, which is using Digital Signal Processor (DSP), makes a ramping pattern for EQ and RQ from spill beam monitor. Here we report the construction status of the extraction magnets and the development of the feedback system.
IHEP Beijing, Beijing
The prototype of pulsed power supply for injection painting bump magnets of CSNS/RCS is being developed. This pulsed power supply consists of IGBT H bridges in series and parallel. The pulse current of the prototype is 18000A, the voltage is about 3KV and the equivalent frequency is about 1MHz. This paper will introduce this prototype in detail.
LBNL, Berkeley, California
Traditional rf processing systems have involved heterodyned rf processing based on mixing a Local Oscillator to up and down convert rf signals through a baseband I/Q or Mag/Phase processing channel. These systems were traditionally custom engineered for each accelerator application. Recent technical developments in rf processing and the development of sufficiently fast reprogrammable digital processing functions lead to development of general-purpose rf processing functions which can incorporate a mix of heterodyned and direct digital down/up-converted processing ("software radio"). This general-purpose approach allows one design of hardware to be applicable to many rf processing tasks, where the firmware and software in the programmable functions define the application. An example design, with applications to linac LLRF control loops and electro-optic timing reference stabilization is presented.
BARC, Trombay, Mumbai
Indian Institute of Technology Bombay, Mumbai
TIFR, Mumbai
As a first step towards development in digital domain, a computer model of a self excited loop (SEL) has been created using MATLAB/SIMULINK. The behaviour of a resonator and a power amplifier combination has been approximated using two first-order differential equations. The square of the amplitude of the RF field in the resonator acts as a driving force for the motion of mechanical modes of the resonator, which are individually represented as second order systems. A key element is the limiter, which has been modelled as a feedback loop, to achieve constant output amplitude. The model has been created in the I-Q domain for computational efficiency and close correspondence with actual implementation. To study the field stabilisation, proportional amplitude and phase feedback loops have been appended to the model of the SEL. In this paper we discuss the details of the model and results from simulation. Initial experimental results are also presented.
University of Pennsylvania, Philadelphia, Pennsylvania
TRIUMF, Vancouver
The quality factor of superconducting radio-frequency cavities typically degrades with increasing field at moderate gradients before the on-set of field emission. The origin of the so called medium field Q-slope is not fully described and understanding it would be important in order to develop a cavity design or treatment which minimizes this effect, allowing us to produce cavities with reduced cryogenic losses. This paper will present an analysis of the medium field Q-slope data measured on cavities at different frequencies treated with buffered chemical polishing (BCP) at TRIUMF. The data is compared with existing models and agreements-discrepancies will be highlighted.
BNL, Upton, Long Island, New York
Funding: Work performed under contract number 126615 for Brookhaven Science Associates, LLC.
NSLS-II is a new ultra-bright 3GeV 3rd generation synchrotron radiation light source. The performance goals require operation with a beam current of 500mA and a bunch current of at least 0.5mA. The position and timing specifications of the photon beam place tolerances on the phase stability of the RF cavity fields of less than 0.15 degrees jitter. This study develops computational methods for the construction of LQG controllers for discrete-time models of single-cavity rf systems coupled to rigid-bunch beams able to meet this tolerance. It uses Matlab’s control-systems toolbox and Simulink to synthesize the LQG controller; establish resolutions of state variables, ADCs, DACs, and matrix coefficients that, in a fixed-point controller provide essentially undiminished performance; simulate closed-loop performance; and assess sensitivity to variations of the model. This machinery is applied to NSLS-II-, CLS-, and NSLS VUV-ring models showing exceptional noise suppression and bandwidth. Thoughts are given on the validation and tuning of the rf model by machine measurements, DSP implementations, and future work.
CEA, Gif-sur-Yvette
The acceleration of non relativistic particles, with a velocity lower than light velocity, in an RF cavity is more complex than for relativistic particles. Non-linear behaviours appear on the accelerator voltage because of the phase slippage inside the cavity. Moreover, a superconducting RF cavity is sensitive to various perturbations like mechanical vibrations (microphonics) and Lorentz force detuning. These perturbations produce a significant detuning of the cavity, leading a strong instability for the amplitude and phase of the field because of the narrow bandwidth of the accelerating mode. We will present a simulation approach of the cavity and its LLRF system control in order to ensure proper cavity operation under perturbations in the framework of the SPIRAL2 project.
CRE, Wuppertal
DESY, Hamburg
For future applications in Light Sources and Large Scale Linear Accelerators we have developed a fully digital LLRF system which overcomes the intrinsic problems of analogue and semi digital LLRF systems by realizing all functions in the high speed cores of FPGAs. Due to its modular design using either the ATCA or the VME form factor the LLRF system can be configured conveniently according to the specific requirements of the accelerator to control the rf field in individual resonators or in a combination of cavities. The LLRF input stage can be custom designed for rf frequencies of up to 3.9 GHz. The hardware and software architectures of the Cryoelectra digital LLRF control system are presented.
CERN, Geneva
JINR, Dubna, Moscow Region
Hardware commissioning of the LHC RF system was successfully completed in time for first beams in LHC in September 2008. All cavities ware conditioned to nominal field, power systems tested and all Low level synchronization systems, cavity controllers and beam control electronics were tested and calibrated. Beam was successfully captured in ring 2, cavities phased, and a number of initial measurements made. These results are presented and tests and preparation for colliding beams in 2009 are outlined.
CERN, Geneva
In the framework of the LHC project the concept has been developed of a global digital signal processing unit (DSPU) that implements in numerical form the architecture of low-level RF systems [1]. The approach, using an FPGA as core for the low-level system, is very flexible and allows the upgrade of the signal processing by modification of the original firmware [2]. The achieved performances of the LHC 1-Turn delay Feedback are compared with project requirements. The PS Transverse Damper DSPU, with automatic loop delay compensation adapting to the beam’s time of flight and Hilbert Filter for single pick-up betatron phase adjustment, is presented. A modified DSPU with digital inputs for the LHC Transverse Damper is also presented.
[1] V. Rossi, CERN SL-2002-047-HRF, CERN, Geneva, July 2002.
[2] V. Rossi, CERN BE-2009-009, CERN, Geneva, January 2009.
Fermilab, Batavia
Funding: Work supported by U.S. Department of Energy under contract DE-AC02-76CH03000
Progress has been made at Fermilab on the development of feed-forward and feed-back algorithms used to compensate SCRF cavity detuning, which is caused by Lorentz Forces and microphonics. Algorithms that have been developed and tested for the 1.3GHz (ILC-style) SCRF cavities (Capture Cavity II) will be reported.
KEK, Ibaraki
NETS, Fuchu-shi
Digital low-level rf (LLRF) control system has been installed in many linear accelerators to stabilize the accelerating field. In the digital LLRF system, the rf signal is down-converted into intermediate frequency for sampling at analog-to-digital converter (ADC) and the number of ADC required for vector sum feedback operation is equal to the number of cavity. In order to decrease the number of the ADCs required, a digital LLRF control system using different four intermediate frequencies has been developed at STF (Superconducting RF Test Facility) in KEK. This digital LLRF control system was operated with four superconducting cavities and the rf field stability under feedback operation was estimated. The result of the performance will be reported.
KEK, Ibaraki
JAEA/J-PARC, Tokai-Mura, Naka-Gun, Ibaraki-Ken
NETS, Fuchu-shi
Upgrade of J-PARC linac has been planed using 972 MHz rf system. The rf field regulation is required to be less than ±1% in amplitude and ±1deg. in phase. The basic digital llrf concept is same as the present 324 MHz llrf system using a compact PCI crate. The main alterations are rf and clock generator (RF&CLK), mixer and IQ modulator (IQ&Mixer) and digital llrf algorithm. Since the typical decay time is faster (due to higher operational frequency than present 324 MHz cavity), chopped beam compensation is one of the main concerns. Performance of the digital feedback system using a cavity simulator is summarized.
KEK, Ibaraki
NETS, Fuchu-shi
Vector-sum control of 4 superconducting cavities is examined at STF in KEK. The digital llrf control is carried out and the stabilities of rf fields are obtained. Various studies such as feedback margin necessary for enough field regulation, effects of perturbations of cavity detuning or klystron HV and so on. Performance degradation by elimination of circulators is also studied from the viewpoint of llrf system.
KEK, Ibaraki
NETS, Fuchu-shi
In STF phass-1, four-cavities are operated with vector-sum feedback (FB) control. The FB control instabilities arising from passband of TM010 mode other than π mode with FB loop-delays were measured. Further, a feedforward (FF) table was used in combination with FB control, which improved the flatness of the flat-top region. A method for reduction of overshoot in FB + FF operation is also proposed. By electrically developing a quasi-beam, the response for quasi-beam injection was also measured, and the correction on beam-loading was performed.
I-Tech, Solkan
In a feedback system the disturbances added in the receiver section are one of the major contributors to the amplitude and phase fluctuations of the fields in the RF cavities that are being controlled. It is therefore crucial to thoroughly evaluate the receiver section of the control system. Measurement results of parameters like amplitude noise, phase noise, coupling between RF channels, linearity and temperature dependent drifts of the receiver are presented. We also discuss what the influences of some of the measured parameters on phase and amplitude stability of the RF fields are. Finally, we summarize the results of the measurements and their impact on the future development of the Libera LLRF system.
NSRRC, Hsinchu
The direct RF feedback has been adopted in storage ring to reduce the beam loading effect for maximizing the stored beam current. Its performance in reducing beam loading is determined by the operational parameters, including the feedback gain, RF phase shift and the loop delay time. This paper presents a mathematical method, based on the Pedersen model, to study the effects of the direct RF feedback on beam loading. Through an example, the influences of different operational parameters on the performance of the direct RF feedback is analyzed by examining the characteristic equation of the feedback loop. The Nyquist criterion is applied for the determination of system stability.
PAL, Pohang, Kyungbuk
SLAC, Menlo Park, California
In PAL, We are preparing the 3GeV Linac by upgrading the present 2.5GeV Linac and new 10GeV PxFEL project. The specification of the beam energy spread and rf phase is tighter than PLS Linac. In present PLS 2.5 GeV Linac, the specifications of the beam energy spread and rf phase are 0.6%(peak) and 3.5 degrees(peak) respectively. And the output power of klystron is 80 MW at the pulse width of 4 microseconds and the repetition rate of 10 Hz. In PxFEL, the specifications of the beam energy spread and rf phase are 0.1%(rms) and 0.1 degrees(rms) respectively. We developed the modulator DeQing system for 3GeV linac and PxFEL. And the phase amplitude detection system(PAD) and phase amplitude control(PAC) system is needed to improve the rf stability. This paper describes the microwave system for the PxFEL and the PAD and PAC system.
SLAC, Menlo Park, California
Funding: Work supported by the U.S. Department of Energy under contract # DE-AC02-76SF00515 and the US-LARP program
A non-linear time-domain simulation has been developed that can determine technical limitations, effects of non-linearities and imperfections, and impact of additive noise on the interaction of the beam with the Impedance Control Radio Frequency (RF) systems [1]. We present a formalism for the extraction of parameters from the time-domain simulation to determine the sensitivity of the beam longitudinal emittance and dilution on the RF system characteristics. Previous studies [2], [3] have estimated the effect of a noise source on the beam characteristics assuming an independent perturbation source of the RF voltage and a simplified beam model with no coupling. We present the methodology for the time-domain simulation study of the dependence of the accelerating voltage noise spectrum on the various RF parameters and the technical properties (such as non-linearities, thermal noise, frequency response etc.) of the Low Level RF (LLRF) system components. Future plans to expand this formalism to coupled bunch studies of longitudinal emittance growth in the LHC at nominal and upgraded beam currents are briefly summarized.
SLAC, Menlo Park, California
ANL, Argonne
DESY, Hamburg
Funding: *Work supported by the DOE under contract DE-AC02-76SF00515
The FLASH L-Band superconducting (SC) accelerator facility at DESY has a LLRF system that is similar to that envisioned for ILC. This system has extensive monitoring capability and was used to gather performance data relevant to ILC. In particular, waveform data were recorded with beam off for three, 8-cavity cryomodules to evaluate the input rf stability, perturbations to the SC cavity frequencies and the rf overhead required to achieve constant gradient during the 800 μs pulses. In this paper, we discuss the measurements made in September 2008 and the data analysis procedures, and present key findings on the pulse-to-pulse input rf and cavity field stability.
SLAC, Menlo Park, California
CERN, Geneva
Funding: Work supported through SLAC/DOE Contract DE-AC02-76-SF00515 and US LARP CERN collaboration.
The LHC Low Level RF System (LLRF) is a complex multi-VME crate system which is used to regulate the superconductive cavity gap voltage as well as to lower the impedance as seen by the beam through low latency feedback. This system contains multiple loops with several parameters which must be set before the loops can be closed. In this paper, we present a suite of matlab based tools developed to perform the preliminary alignment of the RF stations and the beginnings of the closed loop model based alignment routines. We briefly introduce the RF system and in particular the base band (time domain noised based) network analyzer system built into the LHC LLRF. The main focus of this paper is the methodology of the algorithms used in the routines within the context of the overall system. Measured results are presented which validate the technique. Because the RF systems are located underground in a location which is relatively un-accessible even without beam and completely un-accessible when beam is present, these tools will allow CERN LLRF experts to maintain and tune their LLRF systems from a remote location similar to what was done very successfully in PEP-II at SLAC.
TRIUMF, Vancouver
The rf control system for the 1.3 GHz TRIUMF e-linac elliptical superconducting cavities is a hybrid analogue/digital design. It is based in part on an earlier design developed for the 1/4 wave superconducting cavities of the ISACII linac. This design has undergone several iterations in the course of its development. In the current design, down-conversion to an intermediate frequency of 138MHz is employed. The cavity operates in a self-excited feedback loop, while phase locked loops are used to achieve frequency and phase stability. Digital signal processors are used to provide amplitude and phase regulation, as well as mechanical cavity tuning control. This version also allows for the rapid implementation of operating firmware and software changes, which can be done remotely, if the need arises. This paper describes the RF control system and the experience gained in operating this system with a single-cavity test facility.
SOLEIL, Gif-sur-Yvette
After two years of operation, the SOLEIL 3rd generation synchrotron light source is delivering photons to 20 beamlines with a current of 250mA in multibunch or hybrid modes, and 60 mA in 8 bunch mode. The radiation control of the beamline hutches is performed at 300 mA, but recently a 455mA current was stored during machine tests following the installation of the second RF cryomodule. It is foreseen to reach the maximum current of 500mA in the early 2009 and to operate in top-up mode from then on. The new transverse feedback loop has enabled to improve the performance of the single bunch and multibunch beams. The beam position stability is in the range of few micrometers thanks to the efficiency of the fast orbit feedback. Fifteen insertion devices are now installed in the storage ring, ten others are under construction, and a cryogenic undulator is under development. A big effort is being taken in order to compensate the effects of these insertion devices on the machine performance. The good operation performance achieved in 2007 (first year) has been improved in 2008 during which ~4 000 hours will have been delivered to the users with a 95.5% availability and a 30 hours MTBF.
SLAC, Menlo Park, California
Funding: Work supported by Department of Energy contract DE-AC03-76SF00515.
The beam stability for the Linac Coherent Light Source (LCLS) at SLAC is important for good X-Ray operation. Although most of the jitter tolerances are met, there is always room for improvement. Besides the short term pulse-to-pulse jitter, we will also discuss oscillation sources of longer time cycles from seconds (feedbacks), to minutes (cooling systems), and up to the 24 hours caused by the day-night temperature variations.
SLAC, Menlo Park, California
Funding: Work supported by Department of Energy contract DE-AC03-76SF00515.
The electron beam for the Linac Coherent Light Source (LCLS) at SLAC is accelerated by disk-loaded RF structures over a length of 1 km. The mainly longitudinal field can sometimes exhibit transverse components, which kick the beam in x and/or y. This is normally a stable situation, but when a klystron, which powers some of these structures, has to be switched off and another one switched on, different kicks can lead to quite a different orbit. Some klystrons, configured in an energy and bunch length feedback, caused orbit changes of up to 1 mm, which is about 20 times the σ beam size. The origins and measurements of these kicks and some efforts (orbit bumps) to reduce them will be discussed.
SLAC, Menlo Park, California
ASCo, Clayton, Victoria
Funding: Work supported by Department of Energy contract DE-AC02-76SF00515. This work was performed in support of the LCLS project at SLAC
The Linac Coherent Light Source is an x-ray Free-Electron Laser (FEL) project being commissioned at SLAC. The very bright electron beam required for the FEL is subjected to various sources of jitter along the accelerator. The peak current, centroid energy, and trajectory of the electron bunch are controlled precisely at the highest repetition rate possible with feedback systems. We report commissioning experience for these systems. In particular, there is high frequency content in the electron bunch current spectrum, and we report its impact on the systems. Due to the coupling of the betatron motion and the dispersion component of the electron trajectory, a fast in-line model* is incorporated. For the longitudinal feedback, we report the performance of two different configurations: one with RF system as direct actuators, which are nonlinear, and the other with artificially formed linear energy and energy-chirp actuators. Since the electron bunch is compressed to a final peak current of 2 to 3 kA, coherent synchrotron radiation and other wakefields are included for precise control of the electron bunch parameters. Machine performance is compared to start-to-end simulations.
*P. Chu et al., these PAC09 proceedings
BNL, Upton, Long Island, New York
The reduction of beta* beyond the 1m design value at RHIC has been consistently achieved over the last 6 years of RHIC operations, resulting in an increase of luminosity for different running modes and species. During the recent 2007-08 deuteron-gold run the reduction to 0.70 from the design 1 m achieved a 30% increase in delivered luminosity. The key ingredients in allowing the reduction have been the capability of efficiently developing ramps with tune and coupling feedback, orbit corrections on the ramp, and collimation at injection and on the ramp, to minimize beam losses in the final focus triplets, the main aperture limitation for the collision optics. We will describe the operational strategy used to reduce the b*, at first squeezing the beam at store, to test feasibility, followed by the operationally preferred option of squeezing the beam during acceleration, and the resulting luminosity increase obtained in the Cu-Cu run in 2005, Au-Au in 2007 and the deuteron-Au run in 2007-08. We will also include beta squeeze plans and results for the upcoming 2009 run with polarized protons at 250 GeV.
SLAC, Menlo Park, California
INFN/LNF, Frascati (Roma)
Funding: Work supported by the U.S. Department of Energy under contract number DE-AC03-76SF00515.
The use of normal conducting cavities and an ion-clearing gap will cause a significant RF accelerating voltage gap transient and longitudinal phase shift of the individual bunches along the bunch train in both rings of the SuperB accelerator. Small relative centroid position shifts between bunches of the colliding beams will have a large adverse impact on the luminosity due to the small beta y* at the interaction point (IP). We investigate the possibility of minimizing the relative longitudinal position shift between bunches by reducing the gap transient in each ring and matching the longitudinal bunch positions of the two rings at the IP using feedback/feedforward techniques in the LLRF. The analysis is conducted assuming maximum use of the klystron power installed in the system.
SLAC, Menlo Park, California
INFN/LNF, Frascati (Roma)
Funding: Work supported by the Department of Energy under contract number DE-AC03-76SF00515.
We present a possible design for a fast luminosity feedback for the Super-B Interaction Point (IP). The design is an extension of the fast luminosity feedback installed on the PEP-II accelerator. During the last two runs of PEP-II and BaBar (2007-2008), we had an improved luminosity feedback system that was able to maintain peak luminosity with faster correction speed than the previous system. The new system utilized fast dither coils on the High-Energy Beam (HEB) to independently dither the x position, the y position and the y angle at the IP, at roughly 100 Hz. The luminosity signal was then read out with three independent lock-in amplifiers. An overall correction was computed based on the lock-in signal strengths and beam corrections for position in x and y and in the y angle at the IP were simultaneously applied to the HEB. With the 100 times increase in luminosity for the SuperB design, we propose using a similar fast luminosity feedback that can operate at frequencies between DC and 1 kHz, high enough to be able to follow and nullify any vibrational beam motion from the final focusing magnets.
JAI, Oxford
ATOMKI, Debrecen
STFC/DL/ASTeC, Daresbury, Warrington, Cheshire
We present the results of beam tests of the FONT4 ILC prototype intra-train beam feedback system. The system comprises a stripline BPM, a fast analogue BPM signal processor, a custom FPGA-based digital feedback board, a high-power fast-response drive amplifier, and a stripline kicker. The hardware was deployed at the Accelerator Test Facility at KEK. Trains comprising three electron bunches were extracted from the ATF damping ring, with bunch spacing c. 150ns. The feedback loop was closed by measuring the position of bunch 1 and correcting bunches 2 and 3. We report the performance of the feedback, including gain studies, the correction dynamic range, latency measurement, and quality of the beam position correction. The system achieved micron-level bunch stabilisation with a latency of c. 140ns.
CERN, Geneva
Phase I collimators, equipped with Carbon-Carbon jaws, effectively met specifications for the early phase of LHC operation. However, the choice of carbon-based materials is expected to limit the nominal beam intensity mainly because of the high RF impedance and limited efficiency of the collimators. Moreover, C/C may be degraded by high radiation doses. To overcome these limitations, new Phase II secondary collimators will complement the existing system. Their extremely challenging requirements impose a thorough material investigation effort aiming at identifying novel materials combining very diverse properties. Relevant figures of merit have been identified to classify materials: Metal-diamonds composites look a promising choice as they combine good thermal, structural and stability properties. Molybdenum is interesting for its good thermal stability. Ceramics with non-conventional RF performances are also being evaluated. The challenges posed by the development and industrialization of these materials are addressed in a collaboration program, involving academic and industrial partners and complementing material research with an innovative design.
MPI-P, München
CANDLE, Yerevan
YSU, Yerevan
DESY, Hamburg
At DESY the PETRA accelerator has been converted into a new 3rd generation high-brilliance synchrotron radiation facility called PETRA III. For the first commissioning in spring 2009 a positron beam is used. In the future it is also foreseen to operate the synchrotron light source with an electron beam. Ion effects pose a potential problem to the electron beam operation of PETRA III. In this paper, a weak-strong simulation code is employed to study the ion effect issues in detail for different operation scenarios.
BNL, Upton, Long Island, New York
We investigate AC response of accelerator vacuum chambers to external magnetic field, when the wall thickness is comparable or greater than the skin depth. Good agreement was established between experimental measurements, analytical modeling, and ANSYS simulations. Based on the results we suggest a transfer function model for electrically thick vacuum chambers with arbitrary transverse cross-section.
ANL, Argonne
Funding: Work supported by U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract No. DE-AC02-06CH11357.
The Advanced Photon Source (APS), a third-generation synchrotron light source, has been in operation for eleven years. The monopulse radio frequency (rf) beam position monitor (BPM) is one of three BPM types now employed in the storage ring at the APS. It is a broadband (10 MHz) system designed to measure single-turn and multi-turn beam positions, but it suffers from an aging data acquisition system. The replacement BPM system retains the existing monopulse receivers and replaces the data acquisition system with high-speed analog-to-digital converters (ADCs) and a field-programmable gate array (FPGA) that performs the signal processing. The new system has been installed and commissioned in a full sector of the APS. This paper presents the results of testing of the beam position monitor which is now fully integrated into the storage ring orbit control and fast feedback systems.
LNLS, Campinas
UNICAMP, Campinas, São Paulo
This paper reports on the LNLS experience with the digital electron beam position monitor Libera Brilliance through the realization of several standard accelerator physics experiments, taking advantage mainly of the equipment’s turn-by-turn capabilities.
BNL, Upton, Long Island, New York
Funding: Work supported by Brookhaven Science Associates, LLC under Contract No. DE-AC02-98CH10886 with the U.S. Department of Energy.
RHIC BPM system average orbit was originally calculated by averaging positions of 10000 consecutive turns for a single selected bunch. Known perturbations in RHIC particle trajectories, with multiple frequencies around 10 Hz, contribute to observed average orbit fluctuations. In 2006, the number of turns for average orbit calculations was made programmable; this was used to explore averaging over single periods near 10 Hz. Although this has provided an average orbit signal quality improvement, an average over many periods would further improve the accuracy of the measured closed orbit. A new continuous average orbit calculation is currently under development and planned for use in the 2009 RHIC run. This paper will discuss the algorithm, performance with a simulated beam signal, and beam measurements.
CLASSE, Ithaca, New York
Funding: Support provided by the US National Science Foundation and the US Department of Energy.
A central component of the operation of the Cornell Electron Storage Ring as a Test Accelerator (CesrTA) for ILC Damping Rings R&D is the characterization of electron cloud growth in each of the principal vacuum chamber types in use in the storage ring. In order to facilitate measurements in chambers with tightly constrained external apertures, retarding field analyzers have been developed that can be deployed in regions with as little as 3mm of available aperture. We report on the design, fabrication, characterization and operation of devices that are presently deployed in CESR drift, dipole, and wiggler chambers.
INFN/LNF, Frascati (Roma)
In the more than decennial history of DAΦNE, the Frascati e+/e- collider, the positron beam has always shown more difficulty to store high current than the e- beam. Given that the two rings are identical, many types of measurement have been tried to figure out the problem and to solve it, but eventually only one technique has presented a crucial utility: the modal grow rate measurement. In principle this method could be implemented using a commercial spectrum analyzer with the right software procedure inside. Nevertheless it is much easier and faster to record data by the bunch by bunch feedback diagnostics and to use for analyzing the offline feedback programs. A large campaign of data taking has been done in DAΦNE main rings during last fall. A comparison with grow rate records from previous years has point out clearly the difference with 2008 DAΦNE performance showing the way to solve the beam current limit. In particular, measurements have been done versus different machine conditions. Very fast horizontal instability present only in the e+ ring has been characterized showing linear behavior versus beam current. These data have been used to figure out the current limit problem.
LANL, Los Alamos, New Mexico
Funding: US DOE
The Compact Reconfigurable Input/Output (cRIO) hardware manufactured by National Instruments (NI) is evaluated as a wire scanner motion controller. This particular configuration utilizes a NI cRIO-9074 system combined with various C-series modules for wire scanner motion control I/O. Programs for this system have been written in LabVIEW and a majority of the motion-control functionality has been programmed into the cRIO's FPGA in order to provide the fastest motion control processing possible with cRIO. Additional topics of interest include, cRIO-based resolver-to-digital conversion and closed-loop, stepper-based motion control
CERN, Geneva
IN2P3-LAPP, Annecy-le-Vieux
OXFORDphysics, Oxford, Oxon
CEA, Gif-sur-Yvette
To reach the design luminosity of CLIC, the movements of the quadrupoles should be limited to the nanometer level in order to limit the beam size and emittance growth. Below 1 Hz, the movements of the main beam quadrupoles will be corrected by a beam-based feedback. But above 1 Hz, the quadrupoles should be mechanically stabilized. A collaboration effort is ongoing between several institutes to study the feasibility of the “nano-stabilization” of the CLIC quadrupoles. The study described in this paper covers the characterization of independent measuring techniques including optical methods to detect nanometer sized displacements and analyze the vibrations. Actuators and feedback algorithms for sub-nanometer movements of magnets with a mass of more than 400 kg are being developed and tested. Input is given to the design of the quadrupole magnets, the supports and alignment system in order to limit the amplification of the vibration sources at resonant frequencies. A full scale mock-up integrating all these features is presently under design. Finally, a series of experiments in accelerator environments should demonstrate the feasibility of the nanometer stabilization.
KEK, Ibaraki
IN2P3-LAPP, Annecy-le-Vieux
ICEPP, Tokyo
University of Tokyo, Tokyo
The ATF2, accelerator test facility has been developed confirming techniques for obtaining super low emittance beam for future particle accelerators. Here, the converged beam size is designed to be 37 nm, and a precision beam size monitor using interference fringes as a reference called Shintake monitor is used for measuring it. In order to measure the beam size with resolution of better than 10%, relative position between the beam and the interference fringes should be stabilized within few nanometers. Highly rigid tables and mounts for the Shintake monitor and final focusing magnets are adopted with highly rigid floor to ensure relative position stability. Then, the Shintake monitor can be stabilized against the beam, since the beam fluctuates coherently with the final focusing magnets. On the other hand the interference fringes are stabilized against the Shintake monitor with precise phase control system. As a result, relative position between the beam and the interference fringes is stabilized based on rigidity of tables, mounts, and floor between them. We will present our conception for stabilization and results of vibration measurements for the Shintake monitor.
NSRRC, Hsinchu
TPS (Taiwan Photon Source) is a new 3GeV ring to be constructed at NSRRC Taiwan. A motorized magnet girder system with 6 cam movers on 3 pedestals had been designed and tested to provide 6-axis precise adjustments. With 3 consecutive girders to form one section, there will be 72 girders in the whole ring. In order to align the girders precisely and quickly with less manpower, considering the deformation of the floor and limited space in the tunnel also frequent earthquakes in Taiwan, a whole-ring girders auto alignment system was thus proposed. This system consists of touched sensors between consecutive girders and laser PSD system between straight section girders in addition with electric leveling sensors on each girder. The system operating algorithm had been defined and program also fulfilled to be tested on a 3 girders prototype system. The detailed system design and testing results would be described in this paper.
NHMFL, Tallahassee, Florida
Muons, Inc, Batavia
Magnetic confinement fusion reactors (tokamaks) require the development of magnets capable of generating large fields under stringent structural constraints. High temperature superconducting magnets which are well suited to this application are however vulnerable to quench occurrence during operation. Temperature and strain sensors based on fiber optics are being developed as a countermeasure to this contingency. Optical fibers with Bragg gratings are amenable to embedding within superconducting magnets to monitor temperature, strain, irradiation, and to detect quench occurrence. In a length of YBCO tape where quench propagation velocities are slow, we show that it is possible to detect the event occurrence using fiber optic sensors even with a sampling rate as low as 1 Hz. This preliminary result demonstrates the feasibility of using fiber optic sensors to monitor the temperature and strain condition along the length within a coil. These sensors could be used to provide feedback to or trigger magnet protection systems. This would be an invaluable method for mitigating damage to superconducting magnets and increasing up-time for reactors.
CERN, Geneva
In the compact linear collider (CLIC) the tolerances on dynamic imperfections are tight in the main linac. In particular the limited beam delivery system bandwidth requires very good RF phase and amplitude stability. Transverse motion of the beam line components is also of concern. The resulting tolerances are detailed in the paper.
CERN, Geneva
For several years, a large variety of beams have been prepared in the LHC injectors, such as single-bunch and multi-bunch beams, with 25 ns, 50 ns and 75 ns bunch spacings, nominal and intermediate intensities per bunch. As compared to the nominal LHC beam (i.e. with nominal bunch intensity and 25 ns spacing) the other beams can be produced with lower transverse emittances. Beams of low transverse emittances are of interest during the commissioning phase for aperture considerations and because of the reduced long-range beam-beam effects. On the other hand machine protection considerations might lead to prefer nominal transverse emittances for safe machine operations. The purpose of this paper is to present the results of controlled transverse emittance blow-ups using the transverse feedback and octupoles. The procedures tested in the SPS in 2008 allow to tune the transverse emittances up to nominal values at SPS extraction.
NSRRC, Hsinchu
Since the diagnostic system built with the new BPM system upgrade in TLS, we can observe and analyze the orbit stability more clearly and systematically. The disturbances to cause orbit fluctuation mainly come from power supply ripple, ground vibration, ID effects and etc. Removing the disturbed source is a straight, effective but inactive solution. Orbit feedback system is therefore adopted to suppress the remaining noise. In this report, we will evaluate the orbit stability in TLS and present the efforts we have done to improve the orbit stability.
NSRRC, Hsinchu
Stability of photon beam intensity, I0, is one of the most important performance merits of a modern light source. The photon intensity measured at dragon beam line (BL11) is routinely used as a reference signal for I0 stability measurements. At NSRRC, a highly stable I0 intensity is maintained in most percentage of the user beam time. Meanwhile, glitches of I0 intensity up to few tens of percentage had been observed once every few operating hours, which was a puzzle before its reason had been identified later. A spontaneous large variation of photon intensity (I/I0) caused difficulties for users operating their experiments. Here, we report our development of a dedicated electronic circuit with functionality of single-gate, which was very helpful to clarify the puzzle of I0 glitches observed at NSRRC.
NSRRC, Hsinchu
Commissioning of new digital BPM system for TLS is done recently. The new BPM system could support functionalities of turn by turn data, post-mortem and 10Hz slow data acquisition. 10 kHz fast data translation through Liberas grouping mechanism also succeeded to acquire all bpm data and integrate into the orbit feedback system. Various tests are performed systematically to confirm its performance and reliability and will be discussed in this report. We also present the functionalities and infrastructure of the related diagnostic tools. It could record 10 sec orbit data simultaneously via hardware and software event trigger at 10 kHz. Turn by turn and post mortem are also supported through embedded EPICS IOC. More integrated software tools and environment will continue to be developed for future operation.
NSRRC, Hsinchu
The orbit feedback system of the TLS has been deployed for a decade to stabilize electron closed orbit. As the upgrades of digital bpm electronics and switching power supply, the infrastructure of orbit feedback system has also been dramatically modified and rebuilt. The most primary works for the upgrade plan have been done including installation of new bpm and power supply since it was first proposed. After the ordered computer blade ready and the current updated rate raised from 1 kHz to 5 kHz or even 10 kHz, the system will evolve to a newly fast orbit feedback system. It is new scheduled to be commissioned in 2009 spring and can be expected to achieve a submicron stability of the electron beam at a bandwidth of at least 60 Hz.
BINP SB RAS, Novosibirsk
The VEPP-4M electron-positron collider is now operating with the KEDR detector for high-energy physics experiments in the 1.5−2.0 GeV beam energy range. Parallel with these experiments, the VEPP-4M scientific team carries out a number of accelerator physics investigations. A new registration system for the Touschek polarimeter has been put into operation. A new NMR-based system for suppression of the guide field ripples has been developed. The counting rate of the Touschek particles has been measured as a function of the beam energy in the range from 1.85 to 4 GeV. The measurement results can be claimed at the future super B and C-Tau factories. For simultaneous measurement of the transverse beam position and inclination angle an X-ray multi-pinhole camera has been designed, manufactured and installed at the VEPP-4M. To suppress the longitudinal instability caused by high-order modes of the RF cavities, a feedback system has been developed.
USTC/NSRL, Hefei, Anhui
The stability of synchrotron radiation source is of great significance for users, and an accurate and reliable photon beam position monitor (PBPM) is essential for success of synchrotron radiation experiments. Recently, we development a new PBPM called staggered pair photon beam position monitor for photon beam position measurement in Hefei Light Source (HLS). Its main advantage is to reduce the influence of bunch size. Usually, difference over sum (Δ/Σ) method is used to process the photon beam signal. Two new methods are put forward, which are a ratio method and a log-ratio method. For photon beam with Gaussian distribution, differences among methods of Δ/Σ, ratio and log-ratio are introduced. Some calculating results are given for three signal processing methods. Comparing those three methods of position signal processing, log-ratio method is found to have the widest range of linearity, and can obtain identical beam position with different bunch size. Based on that, we also compare staggered pair monitor with double-blade monitor. The staggered pair monitor is found to have higher sensitivity, as well it can ignore the influence of bunch size.
ANL, Argonne
Funding: Work supported by U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
The Advanced Photon Source storage ring has several bunch fill patterns for user operation. In some fill patterns the single-bunch beam charge is as high as 16 mA. We installed a bunch-to-bunch feedback system that aims to overcome high-charge beam instability and reduce the required chromatic correction. Due to the drive strength limitation, we decided to first commission the feedback system in the vertical plane. We present our preliminary results, some of the issues that we have experienced and resolved, and our plan to expand the system to the horizontal plane.
Melbourne
ASCo, Clayton, Victoria
Recently a transverse bunch-by-bunch feedback system was commissioned to combat the resistive-wall instability in the storage ring. The system successfully controls the vertical beam motion so 200 mA can be stored with all in-vacuum undulators at minimum gap and a slightly positive chromaticity setting. The FPGA that comes with the feedback system also provides powerful possibilities for diagnostic measurements. Results will be presented for a) growth/damp measurements to quantitatively characterise the resistive-wall instability, b) bunch-by-bunch diagnostics such as tune chromaticity and c) initial bunch-cleaning attempts in conjunction with a APD bunch purity measurement system.
LNLS, Campinas
The main facility of the Brazilian Synchrotron Light Laboratory is a 1.37 GeV Synchrotron Light Source. The accelerator ring can be filled with up to 148 electron bunches and the initial current of 250 mA decreases down to 150 mA at the end of the user’s shifts. The beam energy is ramped down to 500 MeV, the current is refilled and the energy is ramped up again to 1.37 GeV for a new shift. Coupled-bunch instabilities excited by different sources can negatively impact the light source performance either lowering the brilliance of the beam or causing beam losses in the energy ramps. The upcoming new insertion devices and beamlines are pushing up the beam stability requirements even more. We present the current status of a digital feedback system that is being designed for controlling transversal and longitudinal beam instabilities.
CLS, Saskatoon, Saskatchewan
I-Tech, Solkan
A high-purity single-bunch operating mode, required for time-resolved experiments, has been introduced into the CLS Storage Ring. The newly deployed Transverse Feedback System, which uses the Libera Bunch-by-Bunch system as the feedback processor, has added features that inherently enable bunch cleaning. The bunch purification mechanism is based on a frequency modulated signal that drives the unwanted bunches into betatron oscillations to remove them from the Storage Ring. Bunch purities of 10-6 are achieved, limited only by the leakage rate from adjacent bunches.
Dimtel, San Jose
LBNL, Berkeley, California
IHEP Beijing, Beijing
BEPC-II is a two ring electron-positron collider designed to operate at 1 A beam currents. Longitudinal and transverse coupled-bunch instabilities have been observed in both electron and positron rings. In this paper we present measurements of both transverse and longitudinal instabilities with the identification of active eigenmodes, measurements of growth and damping rates, as well as of the residual beam motion levels. The measurements will then be used to estimate the growth rates at the design beam currents (yet to be achieved). We will also demonstrate how such data is used for specifying power amplifier and kicker parameters.
FEL/Duke University, Durham, North Carolina
PAL, Pohang, Kyungbuk
Dimtel, San Jose
Funding: work supported by US Air Force Office of Scientific Research medical FEL grant FA9550-04-01-0086
The coupled bunch mode instabilities (CBMIs) caused by vacuum chamber impedance limit and degrade the performance of the storage ring based light sources. A bunch-by-bunch longitudinal feedback (LFB) system has been developed to stabilize beams for the operation of a storage ring based Free Electron Laser (FEL) and the High Intensity Gamma-ray Source (HIGS) at the Duke storage ring. Employing a Giga-sample FPGA based processor (iGP), the LFB is capable of damping out the dipole mode oscillation for all 64 bunches. As a critical subsystem of the LFB system, kicker cavity is developed with a center frequency of 938 MHz, a wide bandwidth (> 90 MHz), and a high shunt impedance (> {10}00 Ω). First commissioned in summer 2008, the LFB has been operated to stabilize high current multi-bunch operation. More recently, the LFB system is demonstrated as a critical instrument to ensure stable operation of the HIGS with a high intensity gamma beam above 20 MeV with a frequent top-off injection to compensate for the substantial and continuous electron beam loss in the Compton scattering process. In the future, we will perform detailed studies of the impedance effects using the LFB system.
INFN/LNF, Frascati (Roma)
The SuperB project consists of an asymmetric (4x7 GeV), very high luminosity, B-Factory to be built at Roma-II University campus in Italy, with the ambitious luminosity goal of 1036 cm-2 s-1. To achieve the very challenging performances, robust and powerful bunch-by-bunch feedback systems are necessary to cope with fast coupled bunch instabilities in rings with high beam currents and very low emittances. The SuperB bunch by bunch feedback should consider the rich legacy of previous systems, the longitudinal (DSP-based) feedback built in 1993-97 and the recent “iGp” feedback system designed in 2002-06. Both were designed by large collaborations between Research Institute (SLAC, DAΦNE@LNF/INFN, ALS@LBNL, KEK). The core of the new system will be the digital processing module, based on powerful FPGA components, to be used in longitudinal and transverse planes. Off-line analysis programs, as well real-time diagnostic tools, will be included. The feedback impact on very low emittance beams have to be carefully considered. A MATLAB simulator based on a beam/feedback model is also foreseen for performance checks and fast downloads of firmware/gateware code and parameters.
INFN/LNF, Frascati (Roma)
In this paper the horizontal feedback upgrade for the positron DAΦNE ring is presented. After having completed the analysis of the e+ current limit behavior, a feedback upgrade has been turned out necessary. For the success of the crab waist experiment in the 2008 year, a fast solution to implement the upgrade has been necessary. It has been considered if a simple power increase would be the best solution. The lack of power combiners and of space for other two power amplifiers has brought to a different approach, doubling the entire feedback system. The advantages of this implementation respect to a more traditional power amplifier doubling are evident: two feedback kicks every revolution turns, better use of the power amplifiers, greater reliability, and less coherent noise in the system. Measurements of the two feedbacks have shown a perfect equivalence of the new and the old system: in fact the resulting damping rate is exactly the double of each system taken individually. A description of the implementation is presented together with the performance of the system.
JAI, Oxford
ATOMKI, Debrecen
STFC/DL/ASTeC, Daresbury, Warrington, Cheshire
We present the design of prototype fast beam position monitor (BPM) signal processors for use in inter-bunch beam-based feedbacks for linear colliders and electron linacs. We describe the FONT4 intra-train beam-based digital position feedback system prototype deployed at the Accelerator test facility (ATF) extraction line at KEK, Japan. The system incorporates a fast analogue beam position monitor front-end signal processor, a digital feedback board, and a fast kicker-driver amplifier. The total feedback system latency is less than 150ns, of which less than 10ns is used for the BPM processor. We report preliminary results of beam tests using electron bunches separated by c. 150ns. Position resolution of order 1 micron is obtained.
JAI, Oxford
The major goals of the final focus test beam line facility ATF2 are to provide electron beams with a few tens nanometer beam sizes and beam stability control at the nanometer level. In order to achieve such a level of stability beam based feedback systems are necessary at different timescales to correct static and dynamic effects. In particular, we present the design of intra-train feedback systems to correct the impact of fast jitter sources. We study a bunch-to-bunch feedback system to be installed at the extraction line to combat the ring extraction transverse jitters. In addition, we design a bunch-to-bunch feedback system at the interaction point for correction of position jitter due to the fast vibration of the magnets in the final focus. Optimum feedback software algorithms are discussed and simulation results are presented.
LBNL, Berkeley, California
BNL, Upton, Long Island, New York
The performance of feedback control systems is limited by latency. The hardware-based fast communication system described here offers means for deterministic, fault-tolerant data transmission for feedback systems requiring low-latency communications, such as orbit feedback and Radio Frequency (RF) controls.
SLAC, Menlo Park, California
LBNL, Berkeley, California
BNL, Upton, Long Island, New York
CERN, Geneva
Funding: Work supported by Department of Energy contract DE–AC03–76SF00515 and the US LARP program.
The SPS at high intensities exhibits transverse single-bunch instabilities with signatures consistent with an Ecloud driven instability. While the SPS has a coupled-bunch transverse feedback system, control of Ecloud-driven motion requires a much wider control bandwidth capable of sensing and controlling motion within each bunched beam. This paper draws beam dynamics data from the measurements and simulations of this SPS instability, and develops initial performance requirements for a feedback system with 2-4 GS/sec sampling rates to damp Ecloud-driven transverse motion in the SPS at intensities desired for high-current LHC operation. Requirements for pickups, kickers and signal processing architectures are presented. Initial lab measurements of proof-of-principle lab model prototypes are presented for the wideband kicker driver signal functions.
SOLEIL, Gif-sur-Yvette
Ever since the first user operation, the digital transverse bunch by bunch feedback system developed at SOLEIL has successfully been operated, achieving a stable multibunch beam at the highest intensity in the two planes at zero chromaticity with a single chain working in a diagonal mode. Since then a vertical stripline, optimised to generate large deflections to combat the strong single bunch headtail instability, was installed to construct another chain. The combined use of the two chains allowed enhancing the feedback performance. In particular, by differentiating the feedback gain between high and low intensity bunches, the system is capable of working in hybrid filling modes. In parallel, online applications were integrated into the control system to allow measuring the tunes by selectively exciting a single bunch, damping and growth rates, and analysing the bunch by bunch data in frequency or time domain for post-mortem purposes. Future plans including installation of a horizontal stripline and a noise reduction by avoiding the baseband conversion of the beam signal are also discussed.
SOLEIL, Gif-sur-Yvette
SOLEIL is the French 2.75 GeV high brilliance third generation synchrotron light source delivering photons to beam-lines since January 2007. Reaching micrometer to sub-micrometer level stability for the photon beams is then necessary but very challenging. Since September 2008, a fast orbit feedback has been running in daily operation. The performances of the system will be presented together with comparison with the ones previously achieved with the slow orbit feedback system. Status of the interaction of both feedback systems will be discussed. Moreover new X-BPMs have been installed on dipole and undulator based beam-lines; a total number of 9 vibration sensors (velocimeters) are now installed in the storage ring tunnel, on the experimental slab and outside the building in order to help to locate the different noise sources. Detailed results will be presented and debated.
USTC/NSRL, Hefei, Anhui
A bunch-by-bunch analogue transverse feedback system at the Hefei Light Source (HLS) is to cure the resistive wall instability and the transverse coupled bunch instabilities. The kicker of the feedback system has four 21-cm-long electrodes of stripline type mounted in a skew 45°. Calculation and Simulation of the transverse kicker are shown.
USTC/NSRL, Hefei, Anhui
IHEP Beiing, Beijing
SINAP, Shanghai
IHEP Beijing, Beijing
SSRF, Shanghai
In this paper, we introduce the BxB transverse feedback systems at Hefei Light Source (HLS), which employ an analog system and a digital system. The construction and commissioning for two feedback systems, as well as the instability analysis of beam and the experiment result of the feedback system in HLS are also presented in this paper.
USTC/NSRL, Hefei, Anhui
As low injection energy and multi-turn injection at HLS, the task of diagnosing and curing coupled-bunch instabilities becomes ever harder. The transverse analog feedback system has been redeveloped to improve effect, recently. In this paper, the new improved designs are described and new system's commissioning results are discussed. The transverse coupled bunch instability at 200MeV injection status is also experimentally studied.
USTC/NSRL, Hefei, Anhui
JASRI/SPring-8, Hyogo-ken
Hefei Light Source (HLS) is an 800MeV storage ring with bunch rate of 204 MHz, the harmonics of 45, and circumference of 66 meters. HLS injection works at 200MeV, where the multi-bunch instabilities limit the maximum stored current. A digital transverse bunch-by-bunch feedback system has recently been commissioned at HLS to suppress the multi-bunch instabilities during injection. We employ the SPring-8 FPGA based feedback processor and modified it at NSRL to process horizontal and vertical oscillation signals, independently and simultaneously by one single processor. The design of the digital transverse feedback system and the experiment results are presented in this paper.
UW-Madison, Madison, Wisconsin
ORNL, Oak Ridge, Tennessee
Funding: *Work performed under the auspices of ORNL/SNS, ORNL/SNS is managed by UT-Battelle, LLC, for the U.S. Department of Energy under contract DE-AC05-00OR22725
An electron-proton (e-p) instability is observed with increased beam intensity at the Spallation Neutron Source (SNS) in Oak Ridge National Laboratory (ORNL). This paper presents a wide-band, mixed-signal system for active damping of the e-p instability. It describes techniques used for feedback damping, data acquisition, and analysis. The paper also describes analysis strategies to monitor system performance. The mixed-signal feedback damper system includes anti-aliasing low-pass filters, power amplifiers (PAs), analog-to-digital converters (ADCs), reconfigurable field programmable gate array (FPGA) hardware and digital-to-analog converters (DACs). The system will provide feedback damping, system monitoring, and offline analysis capabilities. The digital portion of the system features programmable gains and delays, and equalizers that are implemented using parallel comb filters and finite impulse response (FIR) filters. These components perform timing adjustments, compensate for gain mismatches, correct for ring harmonics, and equalize magnitude and phase dispersions from cables and amplifiers.
DESY, Hamburg
Uni HH, Hamburg
Next generation FEL light sources like the European XFEL require timing stability between different subsystems of 10-20 fs. In optical synchronization systems, the timing information is distributed across the facilities via sub-ps laser pulses travelling on length stabilized optical fibers. Different methods are available for RF extraction from the pulse train. In this paper, we characterize the long-term phase stability of a 1.3 GHz signal gained from the direct conversion of a higher harmonic of the pulse repetition frequency, and from a voltage controlled oscillator locked with a PLL that uses a Sagnac-Loop as balanced optical-microwave phase detector.
DESY, Hamburg
Stockholm University, Stockholm
For the operation of the European X-Ray Free Electron Laser (XFEL), a system wide synchronous low-jitter clock and precise, adjustable triggers must be generated and distributed throughout the approximately 3.5 km long facility. They are needed by numerous diagnostics, controls, and experiments. Fast ADCs require the jitter of the distributed 1.3GHz clock to be in the order of a few pico seconds (RMS) and that it is synchronized to the accelerating RF. The phase of the 1.3GHz clock must therefore be adjustable at every endpoint. Due to cable lengths, and the temperature dependence of the propagation speed, temperature drifts are a serious issue. Therefore a complex monitoring and compensation mechanism has been developed to minimize these effects. Triggers must also be distributed throughout the system to synchronize different control or measurement tasks. The triggers must be adjustable in time in order to compensate for different cable lengths and should have a resolution of one ns but with ps stability. A prototype of this clock and trigger system has been developed and first measurements have shown, that the strong requirements can be fulfilled.
SLAC, Menlo Park, California
Funding: Work supported by U.S. Department of Energy under Contract Nos. DE-AC02-06CH11357 and DE-AC02-76SF00515.
The Linac Coherent Light Source (LCLS) project at SLAC uses a dense 15 GeV electron beam passing through a 131m undulator to generate extremely bright xrays. The project requires electron bunches with a bunch charge of 20pC to 1nC and bunch lengths of 0.020mm (70fs).To measure the beam resolution to 1 micron (rms) for bunch charge > 20 pC in the undulator, a cavity BPM system was chosen. This system can measure the beam position to within a micron. The LCLS Cavity BPM local oscillator subsystem consists of a second order phase-locked loop (PLL) to synchronize with LCLS timing system and injector system. The output of the PLL is distributed to 36 Cavity BPM receivers and 36 high speed digitizers while maintaining good phase noise and low jitter. This paper describes the design of the PLL and how it met the design specifications of 0.1 degree of phase noise at 119MHz and 1 ns of rms jitter.
JLAB, Newport News, Virginia
Funding: Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177.
Jefferson Lab’s CEBAF accelerator operation requires various frequency signals to be distributed along the site. Three signals: 10 MHz, 70 MHz and 499 MHz are synthesized in the Machine Control Center (MCC) while 1427 MHz and 429 MHz are derived from 499 MHz and 70 MHz signals in four separate locations. We are replacing our obsolete 10 MHz, 70 MHz and 499 MHz sources with new sources that will incorporate a GPS receiver to discipline a 10 MHz reference. In addition the MO (Master Oscillator) system will be redundant (duplicate MO) and a third signal source will be used as a system diagnostic. Moreover the 12 GeV Energy Upgrade for CEBAF accelerator will be adding 80 new RF systems. To support them the distribution of 1427 MHz and 70 MHz signals has to be extended and be able to deliver enough LO (Local Oscillator) and IF (Intermediate Frequency) power to 320 old and 80 new 80 RF systems. This paper discusses the new MO and the drive line extension.
CERN, Geneva
UMAN, Manchester
EPFL, Lausanne
Honorary CERN Staff Member, Grand-Saconnex
The LHC phase 2 collimation project aims at gaining a factor ten in cleaning efficiency, robustness and impedance reduction. From the impedance point of view, several ideas emerged during the last year, such as using dielectric collimators, slots or rods in copper plates, or Litz wires. The purpose of this paper is to discuss the possible choices, showing analytical estimates, electro-magnetic simulations performed using Maxwell, HFSS and GdFidL, and preliminary bench measurements. The corresponding complex tune shifts are computed for the different cases and compared on the stability diagram defined by the settings of the Landau octupoles available in the LHC at 7 TeV.
LBNL, Berkeley, California
CERN, Geneva
Funding: Supported by the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.
Single and multi-bunch instabilities on bunch trains driven by electron clouds have been observed in the CERN SPS for some years. In this paper, we present initial results to implement a damping system in a computer simulation of a single bunch vertical instability using the HEADTAIL code. The code simulates the interaction between a proton bunch and a uniform electron cloud that has built up inside of the beam pipe. In all simulations we use typical SPS parameter sets for three different values of the beam momentum : 26 GeV/c, 55 GeV/c and 120 GeV/c. The feedback is implemented as a corrective kick calculated from the vertical centroid of each slice of the electron bunch with a one turn delay. The bandwidth of the feedback is varied by filtering the slice information along the bunch. Initial results indicate that the instability can be damped with a minimum bandwidth of 300 MHz with a relatively high gain.
LBNL, Berkeley, California
BNL, Upton, Long Island, New York
SLAC, Menlo Park, California
CERN, Geneva
Funding: Supported by the US-DOE under Contract DE-AC02-05CH11231 and the US-LHC LARP. Used resources of NERSC, supported by the US-DOE under Contract DE-AC02-05CH11231.
Electron clouds impose limitations on current accelerators that may be more severe for future machines, unless adequate measures of mitigation are taken. Recently, it has been proposed to use feedback systems operating at high frequency (in the GHz range) to damp single-bunch transverse coherent oscillations that may otherwise be amplified during the interaction of the beam with ambient electron clouds. We have used the simulation package WARP-POSINST to study the growth rate and frequency patterns in space-time of the electron cloud driven beam breakup instability in the CERN SPS accelerator with, or without, an idealized feedback model for damping the instability. We will present our latest results and discuss their implications for the design of the actual feedback system.
BNL, Upton, Long Island, New York
LBNL, Berkeley, California
Funding: U.S. Department of Energy
A new approach to embedded device control is being developed by Lawrence Berkeley Laboratory (LBNL) and Brookhaven National Laboratory (BNL). Synchronous device interface (SDI) will be implemented in NSLS-II project as a key communication protocol. This paper describes the design motivation and principles of SDI. It also discusses SDI in fast orbit feedback system and its extension in power supply control system.
PAL, Pohang, Kyungbuk
The prototypes of PBPM of the four blade types were installed in front-ends 1C1 diagnostic beam line. The four-blade PBPM measure both the horizontal and the vertical positions of the photon beam. KeithleyTM picoammeters are used to record the blade current. The position in both vertical and horizontal directions is calibrated by driving the stepping motors of the PBPM through an Industrial computer. PBPM Data Acquit ion Control System (DACS) is based on Window XP platform. The DACS is equipped with an Ethernet-to-GPIB controller (GPIB-ENET/00). Using the GPIB-ENET/100, networked computers can communicate with and control IEEE 488 devices from anywhere on an Ethernet-based TCP/IP network. This is GPIB interfaces four picoammeters and Ethernet-based TCP/IP communicates by Industrial Computer. Developing with LabVIEW for Windows XP, the interface to EPICS is accomplished by means of Win32 channel Access DLL's. Our LabVIEW application program incorporates EPICS-based motor control and PC-based data acquisition, using a National Instruments I/O board, and saves position data to txt files. This paper presents the PBPM DACS for PLS Control System.
SLAC, Menlo Park, California
Funding: Department of Energy contract DE-ACO3-76SF00515
We describe an integrated relational database for beamline element configuration and online accelerator modelling for LCLS. It is hosted in Oracle, from which online controls software, optimization applications and feedback, use a programming interface to acquire the element data and model. Database population is by an automated process starting with a MAD deck, which is processed in Matlab to derive text files that describe the beamline elements whose data are uploaded using Oracle Loader, and the resulting Oracle APEX applications and reports are used for survey, cabling, metrology and other facilities. An automated facility for online model generation creates an XAL online model beamline description file using a database query; the resulting model is then tracked, and results can be loaded back into the database. As such, both the design or extant machine model, of the present and all previous model runs are available, and linked to the relevant element configuration. We present the process flow from the MAD design to the database, the database schema, the database applications, the process of generating a machine model, and some scientific software which uses the database.