Control System Overview Bob Dalesio Control System Group

Control System Overview Bob Dalesio Control System Group Leader NSLS-II DOE Status Review June 9 -11, 2009 1 BROOKHAVEN SCIENCE

Outline • Design Requirements • Technical Requirements & Specifications • Cost & Schedule Baseline & Staffing • Status • Construction • Standards • Technical Development • Recent Accomplishments • Near Term Plans • Conclusions 2 BROOKHAVEN SCIENCE

Control System Requirements – 1 of 3 • • Bunch Length 1 -40 psecs 2. 642 usec ring revolution Injection system runs up to 1 Hz Top off every 1 minute Top off bunch train ~20 nsec Top off damping time 10 -50 msecs (no extraction) 80% Fill – 1056 of 1320 buckets filled • Orbit stability to. 3 microns at injection • • 5 Hz updates to operators of up to 1000 chosen parameters Archive up to 20, 000 parameters once every 5 seconds continually Must scale to support 150, 000 physical I/O connections and 400, 000 computed variables 99. 7% availability during operation • different fill patterns • high charge single bunch at mid-gap 3 BROOKHAVEN SCIENCE

Control System Requirements – 2 of 3 • • 5 KHz RF Feedback on beam phase 10 k. Hz orbit feedback, (100 usec loop time) • Slow orbit correction • • • 10’s of Hz Data Collection for RF loop correction. 200 Me. V to 3 Gev. V in 400 msec ramp in booster for gap changes Transverse feed back Bunch to Bunch • BPM Requirements • 360 BPMs (12 per cell) • 90 Corrector PS (3 per cell) • 180 Slow PS (6 per cell) synchronized with fast orbit feedback • 1 Hz model based control • 1 H and 1 V stripline • RF cavity control • • • Turn by turn on demand Fast 10 k. Hz output Slow 10 Hz data ADC Raw data Optical clock below 100 fsec 4 BROOKHAVEN SCIENCE

Control System Requirements – 3 of 3 • • 80 psecs pulse to pulse timing jitter. • • 20 msec equipment protection mitigation Transient Recording • • Provide data for all control aspects (no hidden parameters) Conventional facilities • During top off, some beam lines will need 1. 1 - 1. 8 psecs of timing jitter Timing resolution • • • 2 nsec for bunch to bunch timing 200 psec for single bunch injection Provide electron detector as event for beam line 10 k. Hz power supply read backs triggered from timing sys All data available to system with revolution identifier for turn by turn data correlation. • Take coherent turn by turn orbit data for up to 800 channels for 1024 turns • Latch the last 10 seconds of data from all parameters in the storage ring • Beam line needs 1 msec archiving over 1 minute for temperatures and positions • Tunnel and experimental floor Vibration < 25 nm from 4 -50 Hz • Thermal stability of storage ring tunnel enc +-. 1 dgc • Experimental floor +-. 5 dgc 5 BROOKHAVEN SCIENCE

Budget - $300 M ($30 M for Controls) • Manpower by FY 09 • • 4 High Level Application Engineers: Nikolay Malitsky, Guobao Shen, Kunal Shroff, Marty Kraimer (contract), 2 RDB Architects: Don Dohan, Gabriele Carcassi 10 Project Engineers: Joe Delong, Kiman Ha, Yong Hu, Jayesh Shah, Michael Davidsaver, Yuke Tian, Daron Chabot, David Dudley, Huijuan Xu, Wayne Lewis 1 EPICS Expert: Ralph Lange (on sabbatical) 1 Network Expert: Robert Petkus 1 position left to fill Material • • • IOCs crate, processor, and monitoring (1. 3 Million) $1 M network hardware & 360 K timing distribution $400 K Servers and Control Room Consoles $400 K Timing System Hardware $200 K test hardware 6 BROOKHAVEN SCIENCE

Beneficial Occupancy starts Feb, 2011 7 BROOKHAVEN SCIENCE

Control System Software Standards Software standards are being evaluated • • • EPICS as the control system IRMIS used for all configuration data: lattice, components, wiring etc. Embedded Real-Time Operating System choice: RTEMS / Linux Workstations running Debian Commissioning physics applications: Matlab Middle Layer Toolkit – MMLT Model Control: Elegant (APS) and Tracy Online simulation running under EPICS Process Database (Diamond) PV Data and PVAccess being used for client/server environment on HLA Mercurial used source/release control Experiment Control Tools: SPEC (ubiquitous) and GDA (Diamond) Operator Tools: CSS w/ BOY, BEAST. Extending for multi-channel BROOKHAVEN SCIENCE 8

Control System Hardware Standards • • • Hardware Components are being selected Dell Linux development workstations Micro Research Finland – Timing System PLC Solutions – Allen Bradley Non-VME IOCs (240) • Moxa VME crates (90) • Weiner CPU Boards • Motorola • Intel Board? ? MVME 3100, Motor Controllers • Delta Tau Network Switches • Brocade Beam Position Monitors, Power Supply Controllers, FOFB hardware is developed in-house. 9 BROOKHAVEN SCIENCE

Embedded Device Control RF Clock Distribution 45 MHz +T 0 + - Settling time on BPMs Fiducial Distribution 1 Hz +0. 0 usecs - BPM to Compute Controller 384 bits = 64 bits * 6 BPMs 4. 9 usecs = 384 bits over 100 MBit enet +4. 9 usecs - Compute Controllers to each other 10, 290 bits = 30 nodes * 384 bits BPM Controller /OC EPICS PCIx Interface BPM Controller Prev Cell Core Controller BPM Controller I/OC EPICS PCIx Interface 10. 5 usecs = 10, 290 bits over 2 GB enet +15. 4 usecs – Compute local matrix 0 usecs + 15. 4 usecs - Communicate t Power Supply Controllers 24 = 4 PS * 4 bytes each 3. 5 usecs +18. 9 usecs – loop complete settling time for magnets PS Controller Core Controller I/OC EPICS PCIx Interface PS Controller communicate diagnostic waveforms etc… 100 MB ENET +200 usecs – start again 10 Next Cell Core Controller 2 GBit ENET BROOKHAVEN SCIENCE

BPM Version of DFE 11 BROOKHAVEN SCIENCE

High Level Applications Design Goal Lattice Physics Applications (Thin Client) Optics Resp. Matrix (S) Model Server Physics Applications (Thick Client) Measured Orbit Differences Conversions Name Mapping Configuration Parameters Optics Deviations Gradient Errs & Corrections Beam R. M. Diff’s Data Server Mid Level Data Client/Server Application / Family Channel Access Computed Data EPICS Client/Server Need to port Need to develop Under development IOC Process Database Existing Dipole Quad Sext. Corr. BPM RF Designed lattice & installed hardware seq 12 BROOKHAVEN SCIENCE

IRMIS Web Based Reports Editor Component Type Component Lattice EPICS Database Name Mapping Wiring Scripts Component Type EPICS Database Components Lattice EPICS Database Wiring EPICS Database Name Mapping Wiring Component Lattice Name Mapping Files for control Lattice EPICS Database Complete FY 08 Complete FY 09 13 Name Mapping BROOKHAVEN SCIENCE

IRMIS - Component applications 14 BROOKHAVEN SCIENCE

IRMIS - Cabling applications 15 BROOKHAVEN SCIENCE

Random Points • We had some very early help in getting our • • plan right from Mark Heron and Dave Gurd. They identified key budget areas that were missing. Identifying areas of development early, makes it possible to make a difference for this project. There is only 3 years from start of the project to start of commissioning Getting requirements from physicists is one of the more challenging parts of a project. We are most successful by talking to the control groups at other similar projects. 16 BROOKHAVEN SCIENCE

Conclusions • Our capabilities are defined by our group, • • and we have been very fortunate to attract such great talent. The developments that we have undertaken are going well in hardware, relational database tools, and high level application architecture. These could be useful in other environments. The resources available to meet the demands of modern facilities are such that we must expand the areas where we 17 BROOKHAVEN SCIENCE