ID221: Distributed Implementation Plan of a Large, Distributed Accelerator Control System
Bill DeVan, Oak Ridge National Laboratory
Dave Gurd, Los Alamos National Laboratory
Steve Lewis, Lawrence Berkeley National Laboratory
John Smith, Brookhaven National Laboratory
The proposed National Spallation Neutron Source (NSNS) is an accelerator-based 1 - 4 MW pulsed neutron source to be built in Oak Ridge, Tennessee. The facility has five major sections - a "front end" consisting of a 65 Kkev H- ion source followed by a 2.5 MeV RFQ; a 1 GeV Linac; a storage ring; a 1MW spallation neutron target (capable of 4 MW); and the conventional facilities to support these machines. These components will be designed and implemented by four collaborating institutions: Lawrence Berkeley National Laboratory (Front End), Los Alamos National Laboratory (Linac); Brookhaven National Laboratory (Storage Ring); and Oak Ridge National Laboratory (Neutron Source and Conventional Facilities). It is proposed to implement a fully integrated control system for all aspects of this complex. The integration will be based upon the widely-used EPICS control system toolkit. This paper discusses the technical and organizational issues of planning a large control system to be developed collaboratively at four different institutions, the approaches being taken to address those issues, as well as some of the particular technical challenges for the NSNS control system.
Submitted by: David Gurd, Los Alamos National Laboratory
Full address:Lansce-8,MS H820 Los Alamos National Laboratory Los Alamos,NM,87545 USA
Email Address: firstname.lastname@example.org
Phone: (505) 665-6128
Fax: (505) 665-5107
Keywords: Distributed Controls, Integration, EPICS
ID222: Experience Using EPICS on PC Platforms
J.O. Hill, LANL P.O. Box 1663, Los Alamos, NM 87545 USA (LANL)
M. Clausen, Deutches Elektronen-Synchrontron (DESY)
Z. Kakucs, Deutches Elektronen-Synchrontron (DESY)
K.U. Kasemir, Universitat Osnabruck, Fachbereich Physik, D-49069 Osnabruck, Germany
The Experimental Physics and Industrial Control System (EPICS) has been widely adopted in the accelerator community. Although EPICS is available on many platforms, the majority of implementations have used UNIX workstations as clients, and VME- or VXI-base processors for distributed input output controllers. Recently, a significant portion of EPICS has been ported to PC hardware platforms running Microsoft's operating systems, and also Wind River System's real time VxWorks operating system. This development should significantly reduce the cost of deploying EPICS systems, and the prospect of using EPICS together with the many high quality commercial components available for PC platforms is also encouraging. A hybrid system using both PC and traditional platforms is currently being implemented for LEDA, the low energy demonstration accelerator under construction as part of the Accelerator Production of Tritium (APT) project. To illustrate these developments we compare our recent experience deploying a PC-based EPICS system with experience deploying similar systems based on traditional (UNIX-hosted) EPICS hardware and software platforms.
Submitted by: Jeffrey O. Hill
Full address: LANL P.O. Box 1663, Los Alamos, NM 87545 USA
E-mail address: email@example.com
Fax number: 505 6655107
Keywords: epics windows control system
ID223: Recommendations for the use of fieldbuses at CERN in the LHC era
G. Baribaud, R. Barillere, A. Bland, D. Brahy, H. Burckhart, R. Brun, E. Carlier, L. Jirden, F. Perriollat, D. Swoboda
Fieldbuses are increasingly used in a variety of control applications at CERN both for the machines and for the experiments. In order to avoid the proliferation of many types of fieldbuses within CERN it was decided to make recommendations concerning their usage.
A working group examined the requirements for the use of fieldbuses particularly within the context of the LHC project and compared them with existing industrial products. Design, implementation and maintenance aspects are taken into account
The paper will discuss the investigation procedure to define the requirements and the selection criteria. The recommendation concerns three buses for which technical, administrative and commercial support will also be provided.
Submitted by: G. Baribaud
Affiliation: CERN, Switzerland
Keywords: Accelerators, controls, experiments, fieldbus, standard
ID224: A VME DSP Serial Signal Generator
Anne Mason, FNAL
A commercial VME mother-board with mezzanine module, incorporating dual Super Harvard Architecture (SHARC) digital signal processors (DSPs) and a FPGA, is customized to produce machine data (MDAT), beam synchronous Cclocks (BSClks), and Tevatron Clock (TCLK) serial data link signals. In association with a master VME CPU module performing finite state machine operations, also performs TCLK time line generation (scenarios of machine reset events). Rapid hardware, firmware and software development using Altera MAX+Plus FPGA tools, Analog Devices DSP Development Software, and Alex Computer Systems APEX- Lite support software are outlined.
Author's name: Anne Mason
Full address: Fermilab MS 307 POB 500 Batavia, IL 60510 US
E-mail address: firstname.lastname@example.org
FAX number: (630)-840-8590
Keywords: DSP, VME, clocks, data, commercial
ID225: ARCHITECTURE OF THE APS REAL-TIME ORBIT FEEDBACK SYSTEM*
J Carwardine, F Lenkszus
Argonne National Laboratory
The APS real-time orbit feedback system reduces dynamic orbit disturbances in the APS positron storage ring in order to stabilize the x-ray beams for the users. The system is entirely digital and uses an array of digital signal processors (DSPs) to perform orbit corrections and implement the feedback regulator at a 1kHz rate. The DSPs are contained in a total of 21 VME crates distributed around the storage ring which communicate with the APS controls network and behave as standard input-output controllers (IOCs). Each crate contains a Motorola 68040 processor which runs EPICS software and provides the interface between the EPICS environment and the DSPs. The DSP software combines 'C' code with assembly language for time-critical and vectorized functions. Each IOC accesses data from up to 16 beam position monitors, and controls several corrector magnets. A key system component is the dedicated reflective memory network which provides essential data transfer between the VME crates at up to 29 MBytes/second. One crate is used as a system master and performs real-time data analysis such as real-time Fourier analysis of 320 channels of beam position data. This paper describes the system architecture and inter-processor communication, and discusses performance issues.
* Work supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. W-31-109-ENG-38.
Prefered session: Feedback, Control Theory, Automation
Keywords: orbit correction, feedback, storage ring, digital signal processing
The submitted manuscript has been created by the University of Chicago as Operator of Argonne National Laboratory ("Argonne") under Contract No. W-31-109-ENG-38 with the U.S. Department of Energy. The U.S. Government retains for itself, and others acting on its behalf, a paid-up, nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government.
Prime Author: John A Carwardine
Full address: Advanced Photon Source, Argonne National Laboratory Building 401 9700 South Cass Avenue Argonne Illinois 60439 USA
Tel (630) 252 6041
Fax (630) 252 5291
ID226: WWW-based Accelerator Data Warehouse*
A. Chan, G. Crane, I. MacGregor, S. Meyer and R. Sass
Stanford Linear Accelerator Center
In addition to real time data needed for on-line control of an accelerator, there is a need for prompt access to other data required to support and maintain an operating accelerator complex. This includes data on trouble-reporting, spare parts, fabrication, calibration, drawings, etc. This paper focuses on the latter infrastructure data, which is necessary to maintain an operational facility at peak efficiency. We discuss using the World Wide Web (WWW) interface and data repositories to link these data, which are generally segregated by departments and resides on different computer platforms and software. This method of using the WWW, and of linking existing databases rather than rebuilding systems, has allowed us to build a data warehouse at a fraction of the typical programming time. A cornerstone of this is the collaborative effort across departments to conform to a site-wide Formal Device Name, derived from the real-time control system. The disparate databases are linked through this Formal Device Name field. The functionality of a data warehouse in an accelerator environment is described. One can pick a device and find the actual components for the Formal Device Name, calibration measurements, fabrication history, associated cables, modules, and operational maintenance records for the components. Information on inventory, drawings, publications, and purchasing history are also part of the database implementation. Lessons will be discussed regarding database design that can maximize the possibility of presenting a coherent view of data to accelerator operations. A highly reusable relational database design for a 'generic component' that can encompass any type of component and any type of measurement is also discussed. This design is suited for the engineering and fabrication processes of the accelerator and detector, where there are thousands of unique component types, obviating the need to build a database for each distinct system. Data access is also through a WWW interface, while templates are available for international collaborators to collect data off-line. New WWW/database technology that can streamline the work, methods to integrate this with WWW cgi scripts, and the use of off-the-shelf software and tools such as Remedy (WWW-based trouble-reporting software) and Oracle CASE will be discussed.
*Work supported by Department of Energy contract DE-AC03-76SF00515.
Submitted by: Steven Meyer
Full Address: Stanford Linear Accelerator Center P.O. Box 4349, Mail Stop 17 Stanford, CA 94309, USA
E-mail Address: email@example.com
Fax Number: (415) 926-3882
Keywords: Database, WWW, Operations, Software, Off-the-shelf
ID227: The CDR of Computer Service System for SSRF Project
SSRF Project Team Abstract
The SSRF ( Shanghai Synchrotron Radiation Facility )project is largest scientific project to be approved by Chinese government soon before 2000 years in CHINA. It consists of 100MeV LINAC 2 or 3 GeV Booster and Storage Ring. Most of pre - Research and Design (R/D) will start from the early of next year. The location selection of SSRF project will be fixed before this year. This paper will describe its present progress and its computer service system including its essential consideration of computer network, accelerator control system , Beam diagnosis system , Computer Simulation system , User service system for light beam line and Office automation for a new LAB-Shanghai National Laboratory of Synchrotron Light Facility in the forthcoming future.
Submitted by: Dekang LIU
Full address: P. O. Box 918 Beijing, CHINA 100039
E-mail Address: LIUDK@BEPC3.IHEP.AC.CN
FAX number: 86 -10 68213374
Keywords: SSRF, control, beam ,diagnosis, OA, .
ID228: A Configurable RT OS Powered Fieldbus Controller for Distributed Accelerator Controls
A.Matioushine, A.Sytin, P.Vetrov, S.Zelepoukine, (IHEP)
M.Clausen, B.Schoeneburg (DESY).
A General Purpose Fieldbus Controller (GPFC) is primarily designed as a compact and economical solution for introducing advanced programming and operating environment at the fieldbus level of accelerator control systems. Its scalable modular architecture enables a number of applications, ranging from a highly intelligent fieldbus node upto a powerful networked host driving several fiedlbuses. GPFC is implemented as a standalone IndustryPack(R) (IP)carrier, so that a wide range of commercially available plag-on IP modules (CAN, Profibus, Ethernet, digital and analog I/O, etc.) can be used to configure it to application needs. GPFC may have a single board configuration hosting 2 IP modules, or an extended (dual board) one which hosts 4 IPs. GPFC is powered with the Motorola ColdFire(TM) CPU and runs Vxworks(R), the real time operating system from Windriver Systems. GPFC, configured with low cost and highly reliable real-time CAN communications, seems to be an efficient solution for interfacing SEDAC crates, which are widely used at DESY for equipment controls. The GPFC project is jointly run by DESY and IHEP. GPFC implementation for SEDAC control is described and, additionally, using GPFC in the EPICS based distributed control system is discussed.
- IndustryPack (R) -- GreenSpring Computers, Inc.
- ColdFire (TM) -- Motorola Inc.
- Vxworks (R) -- Windriver Systems, Inc.
Telephone: +7 (095) 217 5853
Fax: +7 (095) 230 2337
IHEP, Institute for High Energy Physics 142284 Protvino, Moscow region, Russian Federation.
DESY, Deutsches Elektronen Synchrotron Notkestrasse 85, D 22607 Hamburg, Germany.
Keywords: Fieldbus, IndustryPack, Controller, Vxworks, EPICS.
ID229: PC/MULTIBUS BASED CONTROL SYSTEM
A.Alexandrov, V.Alferov, V.Ershov, A.Gussak, S.Kozhin
V.Kovaltsov, A.Kuznetsov, I.Lobov, S.Petrov, E.Sherbakov
Institute for High Energy Physics, Protvino, Russia
The system is used for controls of electrical power equipment. It includes PC's: Communications (DOS), Server, Console (WINDOWS), Alarm (DOS),connected by ETHERNET, and MULTIBUS 1 crates, connected to Communication PC by MIL 1553. Multibus EUROcrate 6U height,220 mm depth houses modules: -Power Supply with its Controller; -Intel 80186 based Crate Controller; -MIL 1553 Remote Terminal; -Input Register; -Output Register; -ADC; -RS-232 Multiplexer. Remote controllers can be connected to Crate Controller by RS 485 Field Bus. Implementation of low consumption components permits to exclude fan cooling. Alarm PC (Industrial) performs the filling in circular buffer with data for retrospective analysis in case of 70 GeV Proton Synchrotron Rump Generator failure.
Submitted by SSC Institute for High Energy Physics.
Full Address: 142248, Protvino,Moscow Region, Russia
FAX 007-(0967)-744907 (for Alferov)
Keywords: Control System, PC, Multibus
ID230: Current DSP Aplications in Accelerator Instrumentation and RF
Brian Chase, Ken Fullett, Anne Mason
Fermilab#160#, P.O. Box 500, Batavia, IL 60510
The Beams Division at Fermilab is applying Digital Signal Processors (DSPs) to a diverse group of control and measurement systems. This paper presents Fermilab's use of the Analog Devices SHARC DSP on both custom VXI and commercial VME cards in a variety of systems. These include, the low level RF systems for the Booster, Main Ring, Tevatron, PET RFQ, Recycler , as well as the Anti-proton Source BPM and Tevatron clock time- line generator.. It explores how these systems take advantage of an architecture that allows for software and hardware re-use, flexibility, and evolution. This architecture has been applied to a range of systems from real-time to computationally intensive.
Author's name: Brian Chase
Full address: Fermilab, MS308, PO Box 500 Batavia, Il 60510
Keywords: DSP, SHARC, VXI
ID232: The Use of PCs in Controlling DESY Accelerators
DESY MKI Hamburg
PCs now play a dominant role in the control of the HERA, PETRA and DORIS machines at DESY. Initially (1992), the control system employed made use of MS-DOS Front ends and MS-WINDOWS Consoles. Today, the HERA control system supports a number of platforms, the preferred configuration using LINUX Front ends and Windows NT Consoles. A handful of front ends are not PCs at all, however, necessitating an interface between the PC and the non-INTEL world. This has been by and large unproblematic, and over the years the PC itself has proven to be a surprisingly robust and versatile element in the control system. Nevertheless, one must contend with frequently changing tools and components, both hardware and software, in an area where the "State-of-the-Art" looks different every six months. We report here on the experiences at DESY concerning the PC as a system choice.
Submitted by: Philip Duval
Full address: DESY MKI Hamburg
ID233: A new Remote data acquisition system and its Application to a Betatron Oscillation monitor
J.Kishiro1, T.Toyama1, S.Igarashi1, Y.Yamaguchi2 and Y.Takeuchi2
1 High Energy Accelerator Research Organization, KEK, Japan
2 Yokogawa Electric Corporation, YOKOGAWA, Japan
A new development of a data acquisition system on the network is now undergoing and an application of that to the betatron oscillation monitor in the KEK 12GeV PS is presented. The betatron tune meter consists of two separate stations, one of that is placed in the accelerator satellite room in which a white noise generator and a high power rf amplifier controller are installed. The other station is placed in the accelerator control room by which a fast beam position signal is taken and automatically FFT analyzed. Both are connected by an optical fiber network. The system are the plug-in modules in a station in which a 64bit RISC chip is implemented as a station and network controller. The modules has a software driver in it and automatically implemented in the system under the plug-and-play architecture. A high speed optical network enables the transfer rate of 250Mbps. The whole system is dedicated on the Windows-NT system (named WINE).
Submitted by: Jun-ichi KISHIRO
Full address: High Energy Accelerator Organization, KEK 1-1 Oho, Tsukuba, Ibaraki, Japan
E-mail address: firstname.lastname@example.org
Fax number: 81-298-64-3182
Keywords: plug-and-play, Network, Betatron
ID234: PLC Application in the Control of CIAE-30 Cyclotron
China Institute of Atomic Energy
(P.O. Box 275, Beijing, 102413, China)
Application of Programmable Logic Controller ( PLC ) brings CIAE-30 compact cycotron into entirely automatic control, from cyclotron start-up until required beam current obtained at target could be performed with a single function key. The main parameters are sequentialy set-up and regulated by PLC requiring no operators during routine production. High anti-interference ability, reliable safety interlocks, real time status monitoring and interchangeable.operation modes of either full or half-full control, makes cyclotron reliable, stable, flexible and could display the dynamic status all the time. Processes of radioisotope production e.g. targetry, chemical analyses, radioactivity monitoring and protection could be also performed with the same PLC. The cycotron has commissioned for two years and proved that the hardware arrangement and software structure of the whole control system are practical and succesful.
Submitted by: Li Zhenguo
Full address: China Institute of Atomic Energy P.O.Box 275, Beijing,
102413, P. R. China
Keywords: PLC Function key Dynamic status disply Encoder
ID235: REMOT - A Project to Remotely Monitor and Control Scientific Experiments
M. Pucillo (OAT, Trieste, Italy)
M. Barreto (IAC, Tenerife, Spain)
E. Garcia Alcazar (TCP-SI, Madrid, Spain)
F. Genova (CDS, Strasbourg, France)
M. Korten (Forschungszentrum Juelich, Germany)
S. Labajo Izquierdo (GTI, Univ. Pol. Madrid, Spain)
P. Linde (Obs. Lund, Sweden)
W. Lourens (Univ. Utrecht, The Netherlands)
F. Pasian (OAT, Trieste, Italy)
J. D. Ponz (ESA/VILSPA, Spain)
F. Puente (CMA/ULPGC, Las Palmas, Spain)
E. F. Vieira (LAEFF, Madrid, Spain)
REMOT (Remote Experiment MOnitoring and conTrol) is a project developed under the auspices of the TELEMATICS programme of the European Commission, DG XIII. Its objective is to build and validate a generic approach to allow remote control of scientific experiments and facilities that require real-time operation and multimedia information feedback, using communications infrastructure available to European researchers.
The validation of such generic approach is being performed by representatives from the two user communities participating in the project: the Astronomical community, which has telescopes in rather difficult-to-access locations - such as islands and/or mountains - and the Plasma Physics community, that is concentrating expensive experimental facilities in a few places in order to pool resources.
A generic teleoperation system has been built, using as much as possible available elements from other projects or off-the-shelf, plus ad-hoc software modules. The communications infrastructure is based on the Internet, using dedicated network infrastructures where necessary, to ensure that real-time and safety constraints can be met. Over this generic remote operation system, two independent demonstrators have been built in order to show the feasibility of teleoperating scientific experiments in two real environments, and the suitability of the architecture developed.
The successful results of the REMOT project has encouraged the participants to expand its facilities into a flexible, full-fledged system: building such a system is the purpose of the DYNACORE (DYNAmically COnfigurable Remote Experiment MOnitoring and conTrol system) project, which is currently in the final negotiation phase with the EC TELEMATICS programme. The project results will potentially enable large savings, by allowing for shared and more effective access to expensive research facilities. Furthermore, the potential of performing remote experiments will also enable a wider participation in joint research undertakings from all regions of the European community.
Submitted by: Mauro Pucillo
Full address: Osservatorio Astronomico di Trieste, Via G.B.Tiepolo 11, I 34131 Trieste -- ITALY
E-mail address: email@example.com
Fax number: +39 40 309418
Keywords remote control, monitoring, astrophysics, plasma physics, large-scale facilities
ID236: The LANSCE Accelerator Complex Control Room Console Upgrade
S.C. Schaller, E.A. Bjorklund, N.T. Callaway, G.P. Carr, J.A. Faucett, R.W. Garnett, M.A. Oothoudt,
Los Alamos Neutron Science Center, Los Alamos National Laboratory, Los Alamos, NM, USA 87545
The Los Alamos Neutron Science Center (LANSCE) linac and proton storage ring central control room (CCR) consoles are being upgraded. The intent of the upgrade is to move toward a common operator interface, to increase the control room flexibility for simultaneous beam operations and development, to provide a cleaner and more comfortable control room for 24-hour, eight-month-per-year operations, and to improve the reliability and maintainability of the console interface hardware. The three primarily dedicated-function consoles that service two separate control systems are being replaced by a single multifunction console divided into three sections. Each section will be able to control any part of the facility. To eliminate keyboard clutter, the new console will contain six three-headed Sun workstations and several mainly display-oriented X-terminals. Each of the three sections of the new console will thus only have two keyboards. To preserve our investment in VAX-based application programs while the VAX and EPICS control systems are merged, we have provided X-windows emulators for old, character-cell color CRTs and for the old graphics terminals and touch panels. In this paper, we discuss the reasons for the new console design, implementation issues, and the underlying control system capabilities that make such a unified design possible.
Submitted by: Dr. Stuart C. Schaller
Full address: LANSCE-6, Mail Stop H820 Los Alamos National Laboratory P.O. Box 1663 Los Alamos, NM, USA 87545
FAX: (505) 665-0046
Keywords: Operator Interface, Operations, EPICS, LANSCE, LAMPF
ID237: Fieldbuses: WorldFIP provide the most complete solution to control systems Credibility, agility, performances
Patrice Noury (Cegelec, Strategic Marketing Manager)
WorldFIP has been established more than 10 years ago to solve the core aspects of the control and supervision systems for several industrial domains. Once done, additional functions were added to engineer, monitor the system during the operations without perturbations and to manage devices with local intelligence. The key elements of WorldFIP are a solid EMC resistant transmission technique, a high integrity protocol, a just in time transfer mechanism, the support of local intelligence life cycle supported by a highly reliable cost effective technology. These exceptional unrivalled features have permitted to engineer the most complete solution to serve mission critical systems such as the control of power plants, High Voltage switchgears, traction and braking of trains, industrial non interrupted power supply.... Those applications have been provided with benefits such as better control, lower down time, shorter engineering phase increased reliability. Then, whatever the process applications, WorldFIP provides adequate answer for multiple bus level, multiple speeds from 31.25 kbits/s to 5 Megabits/s, multimedia requirements with a cost competitive solution.
Submitted by: Patrice Noury
Full address: Cegelec, Strategic Marketing Manager
E-mail: Noury <101657.3034@CompuServe.COM>
ID238: A Fast Feedback System Based on DSP Technology
Xu Weimin ( University of Science and Technology of China, Hefei, 230026)
A realtime feedback system for accelerator beam orbit collection is developed. The system adopts high speed floating point DSP( Digital Signal Processor ) and high performance VME realtime work station. The system is satisfactory for response speed. It has high flexibility and long stability. The basic principle and construction of the system are described. The system test principle and result are presented too.
Submitted by: Xu Weimin
Full address: University of Science and Technology of China, Hefei, 230026
Key words: orbit feedback, DSP ,VME??
ID239: Significance of a Comprehensive Relational Database System for Modern Accelerator Controls
R. Bakker, T. Birke, B. Kuske, B. Martin, R. Mueller, BESSY
The advantages of a central `data warehouse' that holds relevant project data are obvious. Due to their flexibility and given functionalities professional Relational Data Base Management System (RDBMS) (like ORACLE) seem to be better suited for this purpose than Object Oriented Data Base Systems (OODBS) that are optimized for speed and structural sophistication.
In modern accelerator control diverse but multiply connected data areas have to be maintained: Classical DB applications provide the static data necessary to activate real time DBs for hardware control and data aquisition. Recently generic high level software tools have come into use that need complex configuration data. Theoretical models and their connections to operational procedures can be put into a reference framework due to the centralized storage of design data, calibration factors, geometries etc. On the technical side active nodes are increasingly distributed on large (multi-layered) networks. Different computer systems, each optimized for its designed task, provide services different in scope and multiplicity and require appropriate description and configuration data.
RDBM systems support an excellent environment to control data flow and maintainance. Transaction tools allow to import data from genuine sources while conserving ownership and responsibilities. Programmed clients automatically propagate RDBM changes and help to maintain global system consistency. WEB gateways to the RDBM give platform independent and structured access to the data thus providing high DB transparency. Todays RDBM properties even allow to consider archiving system performance monitor data (`logging') with reasonable update frequency.
Submitted by: Dr. Roland Mueller
Institutions Name: BESSY Address of Institution: Lentzeallee 100 D - 14195 Berlin Germany
E-mail Address: firstname.lastname@example.org
Fax Number: +49 30 6392 4632
Keywords: Data Base, Configuration Handling, Quality Control
ID240: CAN: a Smart I/O System for Accelerator Controls - Chances and Perspectives
J. Bergl, B. Kuner, R. Lange, I. Mueller, G. Pfeiffer, J. Rahn, H. Ruediger
BESSY GmbH, Lentzeallee 100, D-14195 Berlin, Germany
Accelerator control systems should be flexible and open for future improvements. The introduction of a field bus based I/O-system is a step in this direction.
At BESSY most of the accelerator devices are interfaced using the BESSY modular I/O-system and the Controller Area Network (CAN), which provides excellent noise immunity and a sophisticated, reliable protocol together with low hardware costs. The rapid installation procedure of the BESSY II booster synchrotron and the experiences with the CAN installations show: CAN is a good choice to achieve an easy supportable I/O-system without restricting future options.
Additional effort for installing a field bus based I/O-level underneath a Standard Model Control System will be widely overcompensated by the robustness and flexibility of this approach. The growing market of CAN based industrial I/O with its increasing number of out-of-the-box interface solutions is yet another reason for choosing CAN for accelerator controls.
Submitted by: Joachim Rahn
Institutions Name: BESSY
Address of Institution: Lentzeallee 100 D - 14195 Berlin Germany
E-mail Address: Joachim.Rahn@bessy.de
Fax Number: +49 30 6392 4632
Keywords: CAN, Fieldbus, Embedded Controller