Automated Beam Steering and Shaping 2001

December 3rd and 4th 2001, San Jose, California USA

Home

Session Timetable

Objectives

Discussion

Registration

Hotel Information

Committees

Speaker and Author Information

Programme

These are the currently confirmed talks and contributed presentations (no talk, only a web page describing the work). Abstracts are listed below.  See the discussion page for much more information. Please email me, Greg White, if you would like to contribute.

Confirmed Talks

Applied Accelerator Improvement Techniques

Future Accelerator ABS Systems

Beamline and Model Diagnostics

Software, Frameworks and Sharing

Managerial and Tactical Issues

Contributed Presentations

The following presentations will be prepared as web pages only - there will not be a talk. When they are ready the URL link will be made.

Abstracts and Further Information

The following are extracts of emails, or my own summary, and in some cases real abstracts, regarding the confirmed talks above. Real abstracts will replace these as they come in. 

RHIC Injection Correction, Wolfram Fischer, BNL

The RHIC injection correction application was developed with an emphasis on robustness and ease of use. The application can display orbits (closed orbit and injection orbit) at the end of the transfer line and the first sextant in the ring. Also embedded is a loss monitor display in the same range, a beam oscillation and a tune display. Thus, a fast determination of the injection conditions with respect to orbit, coherence, tune and coupling is possible. With a single button the injection orbit can be corrected.

Evolution of Beam-Based Feedback Systems for the SLAC Accelerators, Linda Hendrickson, SLAC, for the SLC/PEPII/E158 feedback team

A generalized beam-based feedback system was implemented for the Stanford Linear Collider (SLC) starting in 1990.  The system became an essential tool for operations and grew to include 50 control loops for the SLC alone, with additional systems supporting other experiments.  In addition to stabilizing the position, angle, intensity and energy of the beams at various locations, a higher-order optimization system maximized luminosity at the interaction point using subtolerance excitation, or "dithering".  The same software has been used for control of other SLAC beamlines including the Final Focus Test Beam (FFTB) and fixed target experiments.  For the PEPII B factory, a colliding storage ring, the system was adapted to stabilize the injected beam and the stored orbits for the electron and positron rings, and to keep the beams in collision.  An upcoming fixed target experiment, E158, will have stringent requirements for beam stability.  Special-purpose feedback loops are needed to control a variety of parameters associated with the polarization asymmetry.  In addition a higher-order feedback will stabilize the linac phase to control the energy spread of the beam.

Experience with the SLC feedback system uncovered some performance limitations of the original design.  While developing designs for the NLC feedback systems, simulations and beam experiments have been performed.  Recent experiments have demonstrated that prototype improved systems achieve better performance.

Longitudinal Multi-Bunch Feedback with Accelerator Diagnostics at PEP-II, ALS, DAFNE, BESSY and the PLS, John Fox, D . Teytelman, SLAC

The latest generation of high-current particle accelerators and synchrotron light sources achieve their intensity performance goals with operating currents that require active feedback to control instabilities. Such wideband feedback systems present technical challenges, but also offer novel accelerator measurement and diagnostic capabilities. Our group has developed a general-purpose instability control architecture, which is in operation at labs in the US, Europe and Asia. The digital signal processing formalism of the architecture allows numerous accelerator, beam and feedback diagnostics - this talk will highlight the principles of transient-domain measurements, in which the beam is allowed to develop unstable motion for a short interval (typically 10 - 30 ms) before being re-damped by the feedback system. Such measurements quantify the performance of the feedback system, and reveal modal information about beam instabilities. The essential purposes of these measurements are to calculate modal growth and damping rates. These measurements of the beam can also be used to measure the frequency-resolved complex impedances as seen by the beam through the tune shifts of individual modes, and the synchronous phase transients produced by uneven fills. Results illustrating these techniques will be presented with examples from several installations.

Beam-Based Alignment of the LCLS FEL Undulator, Paul Emma, SLAC

The Linac Coherent Light Source (LCLS) [] is an x-ray FEL based on a long, permanent magnet undulator comprised of multiple segments allowing space for electron beam position monitors (BPMs) and quadrupole focusing magnets. The 15-GeV electron trajectory within the undulator must be straight to <5 mm over a 10-m FEL gain length in order to maximize electron-photon overlap, and minimize phase slippage. This alignment precision is not achievable with available mechanical survey methods. For this reason, an empirical beam-based alignment technique is developed that uses BPM readings in conjunction with large, deliberate variations in the electron energy. We describe here a detailed alignment simulation including BPM and magnet-mover calibration errors, quadrupole gradient errors, and dipole field errors which demonstrate the required alignment can be achieved.

Foreseen ABS corrections for CERN CTF3 (including acceleration). Frank Tecker, CERN.

The Clic Test Facility CTF3 at CERN aims at testing main ingredients of the power production scheme for a future CLIC linear collider. One key issue is the pulse compression and bunch combination in a combiner ring using RF deflectors to increase the bunch repetition frequency. This requires a stable injection and a good control of the trajectory.

We will present the different parts of the CTF3 complex where an ABS system is foreseen and explain our ideas to approach the problems. This includes mainly the trajectory correction in the linac with acceleration and the injection into a ring using the RF deflection cavities.

Anticipated ABS Applications at the Next Linear Collider, Peter Tenenbaum, SLAC.

This is to get ideas and suggestions from the workshop.  It includes Roland Muellers’s emphasis on the need to simultaneously solve many problems, some coupled to each other. It also includes a probable need for an automatic beamline diagnostics and reporting agent, and combating ground motion.

Simulation of NLC Intra-Pulse Beam-Beam Feedback, Stephen R. Smith, SLAC

The Next Linear Collider (NLC) is a proposed e+ e- collider designed to study physics in the TeV energy region. The nanometer-scale beam spot size at the interaction point of the NLC makes its luminosity sensitive to beam jitter due to ground motion and other disturbances. A mechanism for aligning the beams to each other which acts during the bunch-train crossing time of 265 ns is expected to maintain luminosity in the presence of pulse-to-pulse beam jitter. I describe a beam-beam deflection feedback system which responds quickly enough to correct beam misalignments within the bunch-crossing time and the simulation tools used to understand design tradeoffs needed to achieve a robust system response.

Verifying Machine Design from the perspective of their suitability for ABS, Yu-Chiu Chao, JLAB/CERN. 

An important aspect of accelerator design that can be more systematically approached than has been, is the performance of its diagnostics and control configurations, or its ABS system in general.  The emphasis here is on the configuration of elements rather than their individual performance.  In this report a brief description will be given to a program for orbit correction system evaluation and optimization, recently developed to study these systems for the LHC transfer lines.  Based on analytic methods, this method evaluates the global performance of an orbit correction system in terms of observability, correctability, correction range and response singularity. In addition, orbit and error envelopes are obtained over the entire beam line in an efficient and rigorous manner, providing insights not easily accessible with conventional tools. The performance of a given configuration can be optimized, both analytically through the elimination of structural defects and numerically through fine-tuning on the performance measures.  Finally, features for failure mode analysis allow the user to diagnose observed performance anomalies, and features for critical-element analysis enable the user to identify weak spots in the configuration.  Application to the LHC transfer lines, and extension to other systems will be discussed.
       The broader issue of systematic study, and possible eventual conclusion on design rules, for ABS systems, including beam profile measurement and matching, transfer matrix measurement and matching etc., will be discussed if time allows.

The Supplemental slides used in the ABS meeting talk.
A SLAC seminar I gave on this subject.

CERN PS injection mismatch corrections based on turn-by-turn profile acquisitions, Christian Carli CERN

Abstract: Turn-by-turn profiles have been made available by connecting a special electronics to one of the existing secondary emission profile monitors installed in the CERN Proton Synchrotron. From positions measured with beams with a momentum offset, the dispersion mismatch is quantified. Turn-by-turn beam sizes allow to compute the betatron mismatch. The measured mismatch is reduced by readjusting quadrupole gradients in the injection line. The method is used to set up a new optics of the transfer line between the Booster and the PS with the aim to reduce the dispersion mismatch with respect to the previous setting.

Review of Coupling Measurement and Correction, James Safranek SSRL

After a brief review of coupling formalism, I will present various methods used to measure and correct coupling, including correction of the difference resonance, direct measurement of transverse beam size, betatron oscillation measurements, and closed orbit measurements.

[James' talk included a review of the work of David Sagan et al at CESR and elsewhere. David has kindly let us reproduce a PDF and slides of a talk he gave this year entitled Linear Lattices: Diagnostics and Correction. Click here for the PDF and here for the PowerPoint version.]

A Full-Order, Almost-Deterministic Optical Matching Algorithm, Yu-Chiu Chao, JLAB/CERN

Abstract: An algorithm was developed to obtain global solutions for beam envelope matching with fixed geometry, taking into account the full order compounded effects of the matching quadrupoles. The method is realized through the following steps:

• Systematic algebraic reduction transforming the system into a 2-dimensional problem, making possible robust and efficient numerical solutions 
• Global solution algorithm based on Gröbner basis methods (implemented through Mathematica)
• Algebraic conversion between thin and thick lenses

Apart from providing truly global solutions to matching problems, often unobtainable through local root-finding algorithms found in common optics codes, this program can also point to potential configuration defects when no real global solution exists.

In actual application, this method can be used to either match a given beam to the design twiss parameters, or, in conjunction with symplectically constrained transfer matrix measurements, correct the transport property of a given beam line segment to high precision. These are done very efficiently because no iterative algorithm is involved.

A brief outline of the method will be given. More emphasis will be devoted to outcome of on-line applications to both twiss matching and transport matching at the CEBAF accelerator.

A CERN seminar I gave on this subject with more info

Use of Modular Programs for Accelerator Analysis and Control, M. Borland, L. Emery, APS. Web Site

Abstract: Argonne's Advanced Photon Source (APS) uses a set of modular programs linked by self-describing files for accelerator analysis and control. This approach has been in use since 1994 for commissioning and operation of the APS. Among the tasks performed in this fashion are orbit correction, trajectory correction, beam energy regulation, x-ray beamline steering, beta-function and dispersion correction, generalized optimization, feedforward, and model-independent analysis. Data for performing corrections is variously derived from model or experiment, as needs dictate. After discussing the philosophy and some details of our approach, we demonstrate its application to several aspects of APS operation.

Use of MATLAB Accelerator Toolbox for accelerator modeling and control, Andrei Terebilo, SSRL, Web site

Abstract: At SSRL we have adopted a compact applications programming framework for design, commissioning and future operations of SPEAR3 light source. The components of this framework are: 

1. MATLAB - a commercial package for numerical analysis and visualization.
2. Accelerator Toolbox - a collection of tools to model particle accelerators and beam transport lines in the MATLAB environment 
3. Channel Access Toolbox - tools to control accelerator hardware from MATLAB using the EPICS Channel Access protocol. The talk will discuss the motivation behind this approach and demonstrate interactive modeling and programming examples.

A Prototype of the UAL 2.0 Application Toolkit, Nikolay Malitsky, BNL 

Abstract: The paper presents a prototype of the accelerator commissioning and simulation application toolkit based on the Unified Accelerator Libraries (UAL) framework. The existing UAL 1.x environment has been implemented as an open collection of C++ and Perl packages that address various tasks of accelerator physics. The UAL 2.0 application toolkit has been developing on the top of the Java three-tier infrastructure for integrating it with distributed accelerator control systems. The core part of the toolkit is composed of the Common Accelerator Objects (such as Accelerator, Algorithms, etc.) that form the framework for developing project-specific applications. The toolkit environment is not only limited to Java applications, but also supports the development and integration of high-level scripting codes (e.g. Java Python scripts) and existing C/C++ libraries. The configuration and navigation of the project-specific application system is provided by the XML-based Application Manager. The paper will conclude with applications of the UAL 2.0 toolkit for the SNS project.

Abeans and DataBush: Tools to Write Machine Physicists Applications with Minimal Programming Effort, Igor Kriznar, Mark Plesko,  JSI. Web Site

Abstract: The heart of all our machine physics programs is a Java library called DataBush. It is a redesigned and highly improved version of the databush that we have originally developed for Elettra ten years ago, where it is still used by nearly all machine physics application in order to decouple the physics oriented code from control system implementation details. A device-oriented design of the library gives physicists intuitive access to devices and elements relevant to machine physics. For example, a non-visual Java Bean represents a magnet with it's relevant machine physics parameters. Access to lattice elements is done via Java collections. The graphical user interface is developed visually in a commercial integrated development environment using standard Java/Swing and our own visual Beans. Access to machine resources is provided with the control system Java library called Abeans in a communication platform and control system independent way. Error diagnostics and event handling is part of the Abeans and DataBush frameworks and handled automatically. A demonstration of some linear optics applications will be given, together with a step-by-step development of a new application. Different possibilities of sharing this software will be discussed, in particular how to attach to the underlying control system and existing machine physics codes.

CORBA/CDEV framework for ABS Applications, Michael Boege, PSI. Web Site

Abstract: A distributed client-server model, based on the Common Object Request Broker Architecture (CORBA), has been established to interface beam dynamics applications at the Swiss Light Source (SLS) to essential software packages such as the accelerator physics package TRACY and the Common DEVice (CDEV) control library. CORBA based ABS applications have been extensively used during the commissioning of the SLS booster and storage ring. The inherent flexibility and modular design of the system allows for rapid development of new applications. Recently the Slow ( < 3 Hz orbit correction rate) Orbit Feedback ( SOFB) has been commissioned which reuses thoroughly tested components of the normal orbit correction application. Short and long term orbit stabilities of ~1um have been achieved in both planes (path length changes are automatically adjusted using the RF frequency as an additional corrector within the SVD based correction algorithm). At the same time the SOFB serves as an important step towards the implementation of the Fast Orbit Feedback (FOFB) (4 KHz orbit sampling rate) which is scheduled to be commissioned in December 2001. The integration of various hard- and software components of the FOFB will challenge the chosen CORBA framework.

Requirements of Accelerator Operations, Simon Baird, CERN.

CERN will  present their finding that moving ABS activities into the remit of operations has been “instrumental in the success of ABS”. We hope to follow this talk with a discussion of requirements and development, considering the question: how are requirements and the development process changed when they are driven by operations.

Criteria and properties for interfacing instrumentation to an ABS system, Fabien Perriollat, CERN

This will be a short presentation, followed by discussion lead by Fabien.

Balancing Different Figures of Merit, Roland Mueller, BESSY

Roland's talk will discuss the requirement to simultaneously optimize many non-orthogonal variables, such as how the desire to optimize beam orbit RMS, is constrained by the need to simultaneously control, optimize or decouple from other related parameters, such as energy.