Get documents about "akre llrf rf 2006 apr fac
Shared by: HC12080905298
-
Stats
- views:
- 8
- posted:
- 8/8/2012
- language:
- English
- pages:
- 73
Document Sample
Linac Coherent Light Source (LCLS)
Low Level RF System
Injector Turn-on December 2006
April 20, 2006
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Safety First and Second and Third…..to Infinity
Hazards in the LLRF system
RF 1kW at 120Hz at 5uS = 0.6 Watts average,
2 Watt average amps at 2856MHz,
60W average amps at 476MHz
Hazards – RF Burns
Mitigation – Avoid contact with center conductor of energized
connectors. All employees working with LLRF systems are
required to have the proper training.
110VAC Connector
Hazards - Shock
Mitigation - Don’t touch conductors when plugging into outlet.
All chassis are inspected by UL trained inspector.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Scope of Work – Injector Turn-on
Linac Sector 0 RF Upgrade WBS 1.02.04.03.01
All 3 RF Chassis completed and Installed
Control Module ready for test – John Dusatko
Sector 20 RF distribution system WBS 1.02.04.03.02
Phase and Amplitude Controllers (PAC) – 6 units in Design
Phase and Amplitude Detectors (PAD) – 1 unit in Design
Phased Locked Oscillator – Use SPPS unit for Turn On
LO Generator – Design 90% Complete and tested
Multiplier – 476MHz to 2856MHz – Complete
4 distribution chassis - Complete
Laser Phase Measurement – in Design – not required for turn on
LLRF Control and Monitor System WBS 1.02.04.03.03
1 kW Solid State S-Band Amplifiers – 5 units – in Fab, 2 done
PAD – 12 units as above in design
PAC – 6 units as above in design
Bunch Length Monitor Interface – awaiting Specs
Beam Phase Cavity WBS 1.02.04.03.04
Will use single channel of PAD Chassis
Pill box cavity with 2 probes and 4 tuners - Complete next month
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LCLS Layout
P. Emma
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Control system spans Sector 20 off axis injector to beyond Sector 30
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LCLS RF Jitter Tolerance Budget
Lowest Noise Floor
Requirement
0.5deg X-Band = 125fS
Structure Fill time = 100nS
0.50
Noise floor = -111dBc/Hz
@ 11GHz 10MHz BW
-134dBc/Hz @ 476MHz
X-band X-
RMS tolerance budget for
<12% rms peak-current jitter or
<0.1% rms final e− energy
jitter. All tolerances are rms
levels and the voltage and
phase tolerances per klystron
for L2 and L3 are Nk larger,
assuming uncorrelated errors,
where Nk is the number of
klystrons per linac.
P. Emma
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Slow Drift Tolerance Limits
(Top 4 rows for De/e < 5%, bottom 4 limited by feedback dynamic range)
Gun-Laser Timing 2.4* deg-S
Bunch Charge 3.2 %
Gun RF Phase 2.3 deg-S
Gun Relative Voltage 0.6 %
L0,1,X,2,3 RF Phase (approx.) 5 deg-S
L0,1,X,2,3 RF Voltage (approx.) 5 %
(Tolerances are peak values, not rms) P. Emma, J, Wu
* for synchronization, this tolerance might be set to 1 ps (without arrival-time measurement)
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Linac Sector 0 RF Upgrade
LCLS must be compatible with the existing linac operation including PEP timing shifts
MAIN LINAC (SECTOR 0) RF/TIMING SYSTEM
Master Oscillator is
located 1.3 miles
1 1.3 Miles to
Sum
from LCLS Injector 476MHz MASTER Fiducial Main Drive Line (MDL) LCLS Injector
MASTER PEP PHASE AMPLIFIERS to RF 476MHz RF plus
OSCILLATOR SHIFTER 476MHz 360Hz Fiducial
Measurements on +-720 Degrees
in 0.5mS
To:
Main Linac - 2 miles
January 20, 2006 SLC COUNTDOWN
CHASSIS 476MHz
Damping Rings
PEP
at Sector 21 Divide to 8.5MHz NLCTA
End Station A
FFTB
show 30fS rms jitter 8.5MHz ORION
in a bandwidth from Master Trigger
Generator MTG
Fiducial Generator
Syncronized to:
360Hz
10Hz to 10MHz 360Hz Line
Sync.
Syncs Fiducial to
8.5MHz Damping Ring
360Hz Power Line
8.5MHz Damping Ring
and 360Hz Power Line 476MHz RF Distribution
PEP PHASE SHIFT ON MAIN DRIVE LINE MDL RF with TIMING Pulse – Sync to DR
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Linac Sector 0 RF Upgrade Status
New Low Noise Master Oscillator – Done
New Low Noise PEP Phase Shifter
RF Chassis – Done
Control Chassis – In Test
New Low Noise Master Amplifier – Done
Main Drive Line Coupler in Sector 21 – Done
Measurements
Noise floor on 476MHz of -156dBc/Hz
Integrated jitter from 10Hz to 10MHz of 30fS
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Sector 20 RF Distribution
Main Drive Line (MDL)
476MHz RF
360Hz Fiducial
LCLS Sector 20 RF Reference System MDL to Linac Sectors 21 to 30
From Sector 0 (2km) PEP and Research Yard
RF HUT Coupler
476MHz Ref. 100uW
FSJ4-50 0.8dB/30ft
476MHz
LASER LOCK
LCLS 476MHz PLL Reference
TIMING SYSTEM RF CONTROL
FIDO IQ Modulator
RF CONTROL
Offset adjust LASER
120Hz Track/Hold
TRBR TRBR
119MHz Sample and Hold PLL
with DAC offset adjust LASER Diode
and Error Monitor Output
2830.5MHz LO Gen
119MHz Phase
2856MHz
RF MONITOR 2856MHz in IQ Modulator Control
119MHz 4x 476MHz LASER Diode
Mixer Monitor
0dBm OUT 13dBm OUT Phase Noise
IQ Modulator to adjust
Measurement
2830.5MHz to 2856MHz Phase
RF MONITOR
Divide 112 to 25.5MHz
RF CONTROL
SSB Mix to 2830.5MHz
+13dBm in +13dBm in 4X to 102MHz IQ Modulator
476MHz to 2856MHz 476MHz to 2856MHz 2856MHz
MULTIPLIER MULTIPLIER 25.5MHz out
+7dBm +7dBm
102MHz out 2830.5MHz out
+17dBm +17dBm
RF CONTROL RF CONTROL RF CONTROL LO Phase Monitor
IQ Modulator IQ Modulator IQ Modulator RF MONITOR
2856MHz 102MHz 2830.5MHz
2Watt Amplifier 2Watt Amplifier 2Watt Amplifier
Diode Detector Diode Detector Diode Detector
2856MHz 102MHz 2830.5MHz LO
16 Way Distribution Digitizer Clocks 16 Way Distribution
20dBm each 16 Way Distribution 20dBm each
20dBm each
Gun
L0A Gun
Gun
L0A
L0B
Phase Critical Cables
L0B L0A
L0B L0TCAV
L0TCAV
L1S
L1X
L0TCAV
L1S
L1S
L1X Laser <140ft < 700fSpp
LINAC L1X
EXPERIMENTS LO Phase Monitor
RF MONITOR Gun < 100ft < 400fSpp
2856MHz from Sector 21 LO Phase Monitor
RF MONITOR
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Sector 20 RF Distribution System Status
Phase Locked Oscillator – 476MHz
Initial Turn On use SPPS Oscillator
May modify control to achieve better stability during 2007
LO Generator – 2830.5MHz
Design complete – Prototype tested – 25MHz SSB modulator board done
2856MHz IQ Modulator prototype near completion
Multipliers - 476MHz to 2856MHz – Done
Phase and Amplitude Control (PAC) Unit
In Design – IQ Modulators and Amplifiers selected – See Next Section
Phase and Amplitude Detector (PAD) Unit
In Design – Testing Mixers, Amplifiers, Filters – See Next Section
Amplifiers – not ordered yet
Laser Phase Measurement System – Design Started
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Control System
Distributed Control System
Microcontroller based IOC Control and
Detector Modules
Ethernet Switch
Central Feedback Computer
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Control and Monitor System Klystron Station
TRIG
LCLS RF HUT ENET
2830.5MHz LO Trig & Ethernet
Amp / Splitter
From Klystron Drive Coupler
TCAV 20-5 3dBm LO
RF Gun 20-6 PAC OUT
PAD
L0A 20-7 KLY BEAM Voltage
L0B 20-8
L1S 21-1
SPARE
L1X 21-2
102MHz
Clock In Out
LCLS RF HUT 240ft = 2.5dB 1/2 Superflex
102MHz Clock
Amp / Splitter = 1.6dB LDF4
= 3dB LDF1 23dBm
TCAV 20-5
RF Gun 20-6
L0A 20-7
L0B 20-8
L1S 21-1
L1X 21-2
13dBm
102MHz Clock In
LCLS RF HUT SSSB
2856MHz RF PAC
Amp / Splitter 240ft = 17dB 1/2 Superflex
= 10dB LDF4 Coupled Out Coupled Out
TCAV 20-5 3dBm In 2856MHz Out 17dBm In 2856MHz Out
To IPA
RF Gun 20-6 Klystron Drive
SSSB Control Control & TRIG
L0A 20-7
Trig & Ethernet
L0B 20-8 ENET
L1S 21-1
TRIG
L1X 21-2
140ft = 25dB 3/8 Superflex KLYSTRON STATION
RF HUT
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Control and Monitor System Status
1 kW Solid State S-Band Amplifiers – 5 units
1kW amplifier modules currently in test
Existing amplifier support design under review
Phase and Amplitude Detectors – 11 dual chan units
Preliminary Design Complete
Evaluating amplifiers, mixers, and filters
Phase and Amplitude Controllers – 6 single chan units
Preliminary design complete
Evaluating mixers and amplifiers
Bunch Length Monitor Interface
Need Specifications
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Beam Phase Cavity Status
Measurement of beam phase to RF reference phase. The result will be
used to correct timing of laser to RF reference. Cavity is located
between L0A and L0B.
Electronics will use single
channel of PAD Chassis
Pill box cavity with 2 probes
and 4 tuners
Cavity Electronics will use
single channel of RF Monitor
Cavity in fabrication
Complete – May 2006
Bake – June 2006
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Controls Engineering Requirements
When beam is present, control will be done by
beam-based longitudinal feedback (except for T-
cavs); when beam is absent, control will be done
by local phase and amplitude controller (PAC)
Adhere to LCLS Controls Group standards:
RTEMS, EPICS, Channel Access protocol
Ref: Why RTEMS? Study of open source real-time OS
Begin RF processing of high-powered structures
June 2006
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
External Interfaces
LLRF to LCLS global control system
PVs available for edm screens, archiving, etc over
controls network
LLRF VME to beam-based longitudinal feedback
from feedback: phase and amplitude corrections at 120
Hz over private ethernet
from LLRF: phase and amplitude values
(internal) LLRF VME to LLRF microcontrollers
from VME: triggers, corrected phase and amplitude
from microcontrollers: phase and amplitude averaged
values at 120 Hz, raw phase and amplitude values for
diagnostics
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Sector 20 PAC and PAD Control
VME IOC
Ethernet Switch
Arcturus Coldfire
13 PADs
FIFO
ADC
13 PACs
FPGA
DAC
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
EPICS PANELS
Single Pulse
Diagnostic Panels for
PADs are Running
Remaining Software
History Buffer Select
PVs
Multi pulse data
analysis, correlation
plots
Local RF Feedback
loops
Links to global
Feedback loops
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF Status Summary
Linac New Low Noise Source – RF components installed, Controls Feb06
RF Distribution – Prototyping underway (R. Akre, B.Hong, H. Schwarz)
Monitor Controller Board (J. Gold, R. Akre, Till Straumann)
Single channel prototype for ADS5500 tested to specifications
Four channel ADS5500 board – layout complete (SNR 70dBFS)
Switched to LTC2208 16bit 130MSPS ADC (Prototype in test) (SNR 77dBFS)
RF Monitor Board in preliminary design (H. Schwarz, B.Hong)
Testing mixers
Control Boards (J. Olsen)
Fast Control Board – All but slow ADCs for temp and voltages tested and low level drivers
written
Slow control board – use fast board
RF Control Board in preliminary design (H. Schwarz, B. Hong)
Software (D. Kotturi, Till Straumann)
EPICS on RTEMS on Microcontroller done
Drivers – data collection interrupt routine done
Algorithms – PAD 90% complete PAC in progress
Calibration routines – Need specifications
Collision free Ethernet
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Schedule
RF Distribution Design Complete May 2006
RF Hut Distribution System installed August 2006
PAC design Complete June 2006
PAD design Complete July 2006
PAC and PAD minimal operational software complete
Ethernet testing with multiple PACs and PADs???
Single S-Band station – hardware installed Sept 2006
4 other S-Band Stations – November 2006
Feedback software interfacing???
Test and debug with Klystrons On – December 2006
X-Band Station January 2007
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
End of LLRF RF Talk
Backup for RF Talk
Mostly Correct
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
DESIGN PHILOSOPHY
Reliability is inversely proportional to the number of
connectors.
Stability is inversely proportional to the number of
connectors.
Measurement accuracy is inversely proportional to
the number of connectors and the amount of
Teflon,which is typically found in connectors.
Cost of maintenance is proportional to the number
of connectors.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Electro-Optical Sampling Timing Jitter
200 mm thick ZnTe crystal
Single-Shot (20 Shots)
e- <300 fs
Ti:Sapphire
laser
e- temporal information is encoded
on transverse profile of laser beam
170 fs rms
Adrian Cavalieri et al., U. Mich.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
MPS – PPS Issues
Addressed by Controls Group
Not Reviewed Here
Vacuum
New vacuum system summary to be fed to each
klystron existing MKSU.
PPS System
Injector modulators will be interlocked by Injector
PPS system.
PPS requirements for radiation from the injector
transverse accelerator needs to be determined.
Radiation levels will be measured during testing
in the Klystron Test Lab – Feb 06.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Bandwidth of S-Band System
Upper Frequency Limit – 10MHz
Beam-RF interaction BW due to structure fill time
< 1.5MHz S-Band Accelerators and Gun
~10MHz X-Band and S-Band T Cav
Structure RF Bandwidth ~ 16MHz
5045 Klystron ~ 10MHz
Lower Frequency Limit – 10kHz
Fill time of SLED Cavity = 3.5uS about 100kHz
Laser – Needs to be measured ~ 10kHz
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Noise Levels
RF Reference Single Side Band (SSB) Noise Floor
2856MHz RF Distribution -144dBc/Hz
-174dBc/Hz @ 119MHz (24x = +28dB +2 for multiplier)
2830.5MHz Local Oscillator -138dBc/Hz
Integrated Noise
-138dBc/Hz at 10MHz = -65dBc = 32fS rms
SNR = 65dB for phase noise
Added noise from MIXER (LO noise same as RF)
SNR of 62dB
ADC noise levels
SNR of 70dB – 14bit ADS5500 at 102MSPS
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Phase Noise – Linac Sector 0
OLD MASTER OSCILLATOR NEW MASTER OSCILLATOR
-133dBc/Hz at 476MHz -153dBc/Hz at 476 MHz
340fSrms jitter in 10MHz BW 34fSrms jitter in 10MHz BW
Integrated Noise - Timing Jitter fs rms
Integral end 5MHz 10kHz
Integral start 1M 100k 10k 1k 100 10
Aug 17, 2004
Sector 30 27 30 33 38 75 82
Jan 20, 2006
Sector 21 15 19 20 20 8 17
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Sector 20 RF Distribution Cable Errors
Temperature Coefficient of 2.8ppm/ºF and
Cable length is 1200ºS/ft
All Cables except LASER are less than 100ft
Distances feet and errors in degrees S total range
RF Hut Down Linac Wall Injector Total
Unit Ft degS ft degS ft degS ft degS ft degS DegS
Laser 8 0.054 25 0.017 10 0.014 10 0.007 85 0.58 0.68
Gun 8 0.054 25 0.017 10 0.014 10 0.007 40 0.27 0.37
L0-A 8 0.054 25 0.017 10 0.014 10 0.007 30 0.21 0.31
B Phas 8 0.054 25 0.017 10 0.014 10 0.007 20 0.14 0.24
L0-B 8 0.054 25 0.017 10 0.014 10 0.007 20 0.14 0.24
L0-T 8 0.054 25 0.017 10 0.014 10 0.007 10 0.07 0.17
L1-S 8 0.054 25 0.017 50 0.068 0.14
L1-X 8 0.054 25 0.017 60 0.081 0.16
Temperature Variations: RF Hut ±1ºF : Penetration ±0.1ºF : Linac : ±0.2ºF
Shield Wall ±0.1ºF : Injector ±1ºF
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF System Topology / Specifications
Linac
Sector 0 RF
Num ber of cables per device
MDL Reference cables are
8ft and can drift +-50fS
476MHz PLL Cable Drift Based on
L0, L1 - 5 Klystrons
2830.5MHz LO Temperature variations Most Devices
Specifications
Amp / Splitter PAD and temp co of 5ppm/degC are in tunnel
100fS rms jitter
+-2.3pS drift Laser 1 +-680fS Laser
Laser
RF Gun 5 RF Gun +-370fS RF Gun
L2 - 4 Sectors 2 +-310fS
Specifications L0A L0A L0A
70fS rm s jitter 2 +-240fS
Phase Cavity Phase Cavity Phase Cavity
+-5pS drift
L0B 2 L0B +-240fS L0B
L3 - 6 Sectors L1S 4 +-140fS
L1S L1S
Specifications
150fS rms jitter L1X 2 +-160fS L1X
L1X
+-5pS drift
1 +-500fS
L2 Ref L2 Ref L2 Ref
RF HUT
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF Monitor Signal Counts
ADC Chan Cnt Chassis Count/Location
Distribution (5~2850MHz, 4<500MHz) 4 1Hut
RF Gun 9 1Kly 1.5Hut
Beam Phase Cavity 2 0.5Hut
L0-A Accelerator 4 1Kly 0.5Hut
L0-B Accelerator 4 1Kly 0.5Hut
L0-T Transverse Accelerator 4 1Kly 0.5Hut
L1-S Station 21-1 B, C, and D Acc 6 1Kly 1.0Hut
L1-X X-Band accelerator X-Band 5 1Kly 0.5Hut
S25-Tcav 4 1Kly
S24-1, 2, & 3 Feedback 0
S29 and S30 Feedback 0
Total Chassis 7Kly 6Hut
Total into Hut IOC 12
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF Control Signal Counts
Distribution (3~2850MHz, 3<500MHz) 6 IQ Mod
RF Gun 1 Klystron
Beam Phase Cavity 1 IQ mod
L0-A Accelerator 1 Klystron
L0-B Accelerator 1 Klystron
L0-T Transverse Accelerator 1 Klystron
L1-S Station 21-1 B, C, and D accelerators 1 Klystron
L1-X X-Band accelerator X-Band 1 IQ Mod
S25-Tcav 1 Klystron
S24-1, 2, & 3 Feedback 3 Klystrons
S29 and S30 Feedback 2 IQ modulators 476MHz
Total modulators 11 Fast 8 Slow 19 modulators
Totals at ~2856MHz 14 modulators
Total into Hut IOC 14 modulators
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Control and Monitor System
LLRF Control and Monitor System
1 kW Solid State S-Band Amplifiers – 5 units
Phase and Amplitude Monitors – 12 units
Phase and Amplitude Controllers – 6 units
Bunch Length Monitor Interface – Need Specifications
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF Control
Required 13 Units
I Control
3
Includes Distribution
1
I BXMP1007
RF In LO RF 2
17dBm RF Out
0dBm 17dBm
Q
Q Control
4
2856MHz Input Monitor 2856MHz Output Monitor
2850MHz IQ Modulator
RF Control Module consist of the following:
Input Coupler, IQ Modulator, Amplifier, Output Coupler
Filters for I and Q inputs
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
RF Monitor
Required 13 Chassis for Injector – Includes Distribution
LO 2830.5MHz : RF 2856MHz
IF 25.5MHz (8.5MHz x 3 in sync with timing fiducial)
Double-Balanced Mixer
Mixer IF to Amp and then Low Pass Filter
Filter output to ADC sampling at 102MSPS
2830.5MHz Local Osc.
Amplifier
To ADC
RF LO
IF
LTC2208 SNR = 77dBFS
MIXER 25.5MHz BP FILTER
102MSPS
2856MHz RF Signal
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
1 kW Solid State S-Band Amplifiers
Design Complete
Two Units on the Shelf
Modules in house – and
tested
Support parts – Some
parts in house
Power Supplies, relays,
chassis on order
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
SLAC Linac RF – New Control
MDL 476MHz
Next Sector The new control system will tie in
1mW
1W to the IPA Chassis with 1kW of
6X
2856MHz drive power available. Reference
Existing
Phase will be from the existing phase
Reference SubBooster
Line reference line or the injector new
Sub Drive Line
RF reference
3
To Next I
Klystron 3kW 1
2 RF LO
IPA 1kW Amp Q
4
2856MHz
20mW High Power IQ Modulator
Phase Shifter
Attenuator
I and Q will be controlled with a
Phase &
Amplitude Klystron
16bit DAC running at 119MHz.
Detector
SLED
Waveforms to the DAC will be set
200MW
in an FPGA through a
-45dB microcontroller running EPICS on
Existing
System Accelerator RTEMS.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Controls Talk
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
LLRF Controls
Outline
Requirements
External Interfaces
Schedule
Date Needed
Prototype Completion Date
Hardware Order Date
Installation
Test Period
Design
Design Maturity (what reviews have been had)
State of Wiring Information
State of Prototype
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Requirements
At 120 Hz, meet phase/amp noise levels
defined as:
0.1% rms amplitude
100 fs rms in S-band (fill time = 850 ns)
125 fs rms in X-band (fill time = 100 ns)
All tolerances are rms levels and the voltage and
phase tolerances per klystron for L2 and L3 are
Nk larger, assuming uncorrelated errors, where
Nk is the number of klystrons per linac (L2 has
28; L3 has 48)
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Engineering Requirements
When beam is present, control will be done by
beam-based longitudinal feedback (except for T-
cavs); when beam is absent, control will be done
by local phase and amplitude controller (PAC)
Adhere to LCLS Controls Group standards:
RTEMS, EPICS, Channel Access protocol
Ref: Why RTEMS? Study of open source real-time OS
Begin RF processing of high-powered structures
May 20, 2006
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
External Interfaces
LLRF to LCLS global control system
PVs available for edm screens, archiving, etc over
controls network
LLRF VME to beam-based longitudinal feedback
from feedback: phase and amplitude corrections at 120
Hz over private ethernet
from LLRF: phase and amplitude values
(internal) LLRF VME to LLRF microcontrollers
from VME: triggers, corrected phase and amplitude
from microcontrollers: phase and amplitude averaged
values at 120 Hz, raw phase and amplitude values for
debug
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
External Interfaces: Laser - Tcav
RF Phase and Amplitude correction at 120 Hz for:
laser, gun, L0-A, L0-B, L1-S, L1-X, T cav
In-house modules sharing VME crate for timing triggers
Temperature monitors
RF Reference/4 = 119 MHz T Cav
476 MHz RF Reference clock distributed to all 30 sectors in the Linac and beyond
stabilized to 50 fs jitter L1-X
L1-S
L0-B
L0-A
gun
Laser and RF ref
PAD
D
Coldfire A
F
I and Q CPU C
A I
Demo- running
D F
dulator
C O
RTEMS s VME Crate at S20
and l
s
EPICS o C E
running
w P V longitudinal,
RF Reference*6 = 2856 MHz U R
stabilized to 50 fs jitter beam-based
PAC
feedback
D
Coldfire A
CPU C 1 trigger
D
running for 4
FPGA A
RTEMS s channels
C
and l of 1k
EPICS o samples
w
Private ethernet Private ethernet Private ethernet
4 kBytes at 120 Hz 8 kBytes at 120 Hz
Controls gigabit ethernet (interface to MCC)
IQ Modulator
gives phase
and amplitude
control
1 trigger to travel
up to ½ sector
away 100 mW
Lin
ac
/A
All except laser RF
cc
ele
La r RF
La
se 1
119 MHz
rat
se
Solid State Sub Booster Laser
r R 19
or
Oscillator
F 1 MH
RF ler
Li z (I& Q)
na
1 kW
20 z
Ou ato
c/
photodiode
A
H (I&
Amps
t (I r R
cc
Klystron
e
&Q F
119 MHz
120 Hz
) I
Q)
60 MW
photodiode
UV
Gun
SLED
n
NB: For the gun, SLED
cavity
(I
cavity is shorted out
&
Q
)
HPRF
240 MW
1 kW 1 kW
60 MW
10' accelerator
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
External Interfaces: L2-L3
RF Phase and Amplitude correction at 120 Hz for:
laser, gun, L0-A, L0-B, L1-S, L1-X, T cav
In-house modules sharing VME crate for timing triggers
Temperature monitors
RF Reference/4 = 119 MHz T Cav
476 MHz RF Reference clock distributed to all 30 sectors in the Linac and beyond
stabilized to 50 fs jitter L1-X
L1-S
L0-B
L0-A
gun
Laser and RF ref
PAD
D
Coldfire A
F
I and Q CPU C
A I
Demo- running
D F
dulator
C O
RTEMS s VME Crate at S20
and l
s
EPICS o C E
running
w P V longitudinal,
RF Reference*6 = 2856 MHz U R
stabilized to 50 fs jitter beam-based
PAC
feedback
D
Coldfire A
CPU C 1 trigger
D
running for 4
FPGA A
RTEMS s channels
C
and l of 1k
EPICS o samples
w
Private ethernet Private ethernet Private ethernet
4 kBytes at 120 Hz 8 kBytes at 120 Hz
Controls gigabit ethernet (interface to MCC)
IQ Modulator
gives phase
and amplitude
control
1 trigger to travel
up to ½ sector
away 100 mW
Lin
ac
/A
All except laser RF
cc
ele
La r RF
La
se 1
119 MHz
rat
se
Solid State Sub Booster Laser
r R 19
or
Oscillator
F 1 MH
RF ler
Li z (I& Q)
na
1 kW
20 z
Ou ato
c/
photodiode
A
H (I&
Amps
t (I r R
cc
Klystron
e
&Q F
119 MHz
120 Hz
) I
Q)
60 MW
photodiode
UV
Gun
SLED
n
NB: For the gun, SLED
cavity
(I
cavity is shorted out
&
Q
)
HPRF
240 MW
1 kW 1 kW
60 MW
10' accelerator
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Design
Design maturity (what reviews have been had):
RF/Timing Design, DOE Review, August 11, 2004
Akre_FAC_Oct04_RF_Timing, FAC Review, October, 2004
Low Level RF Controls Design, LCLS Week, January 25-27, 2005
Low Level RF, Lehman Review, May 10-12, 2005
LLRF Plans for Development and Testing of Controls, LCLS Week, July 21, 2005
Low Level RF Design, Presentation for Controls Group, Sept. 13, 2005
LLRF Preliminary Design review, SLAC, September 26, 2005
LCLS LLRF Control System - Kotturi, LLRF Workshop, CERN, October 10-13, 2005
LCLS LLRF System - Hong, LLRF Workshop, CERN, October 10-13, 2005
LLRF and Beam-based Longitudinal Feedback Readiness - Kotturi/Akre, LCLS Week, SLAC, October 24-26,
2005
LCLS Week LLRF and feedback - Kotturi/Allison, LCLS Week, SLAC, October 24-26, 2005
LLRF, LCLS System Concept Review/Preliminary Design Review, SLAC, November 16-17, 2005 Comments
LLRF Beam Phase Cavity Preliminary Design review, SLAC, November 30, 2005
Docs at: http://www.slac.stanford.edu/grp/lcls/controls/global/subsystems/llrf
State of wiring: percent complete Captar input will be given at time of presentation
State of prototype: PAD (1 chan ADC) and PAC boards built (shown on next pages).Testing.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
PAD – the monitor board
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
PAD – the monitor board
RF Board 2 X 16 bit ADC Control Board
119 or 102MHz Clock
Line Drivers LTC2208
Filters Transformer Coupled Inputs FIFO 2 X 1k words
16bit DATA 16 bit
25.5MHz IF
Chan. 1 DATA
WCLK
IF
RF CHAN 1 RF LO
CONTROL /
INPUT 16bit DATA Arcturus uC5282
ETHERNET
MIXER CS/
Chan. 2 CLK Microcontroller Module
WCLK with 10/100 Ethernet
IF
RF CHAN 2 RF LO
INPUT
MIXER
Control
LO INPUT CPLD
RF - 25.5MHz EXTERNAL EXTERNAL
CLOCK TRIGGER
102MHz 120Hz
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
PAC – the control board
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
PAC – the control board
EXTERNAL
CLOCK TRIGGER TRIGGER
SSSB
Monitor TTL 120Hz
119MHz
60nS NIM
Chassis
MONITOR
PORTS
Temperature Monitor
Forward Power 0-?V
RF BOARD SSSB
Trig
Reflected Power 0-?V
Over Temp 0 or 12V
MATCHING TTL Power Supplie +12V
Power Supply -12V
FILTER 17 to 30uS
NETWORK
I&Q MODULATOR 16bit DATA 16 bit
3
I DATA
I MAX5875 CLK XILINX CONTROL /
RF OUTPUT 2 RF LO
1
2 X 16 bit DAC SPARTAN 3 Arcturus uC5282
To SSSB 119MHz Clock FPGA Microcontroller Module
16bit DATA
ETHERNET
(1MHz to 200MHz) CS/
Q
CLK with 10/100 Ethernet
Q CLK
4
AD8099 Diff Amp
2856MHz Ref
Control Control
DC Power
Temperature Supply Temperature
Monitor t Monitors Monitor t
Thermocouples
DC Power
Supplies ADCs
Control Board
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Additional Slides
The following two pages show an overview
of the LLRF control modules. From these
diagrams, counts of module types, as well as
function and location are seen.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Overview of LLRF at Sector 20
RF phase and amplitude correction and global feedback at 120 Hz for LCLS LINAC S20
RF Dist’n Key:
Indicates located in RF Hut
SPAC Otherwise at Klystron
SPAC Indicates may be needed
SPAC
Laser The maybe is included in
SPAC counts below
SPAC SPAC
Gun PAD PAD
PAC
PAC
PAD
PAD
L0-A
PAC PAD
PAD PAD
PAD
L0-B
PAC
PAD
PAD VME Crate at S20
L0-Tcav Eth
C E running
P V
PAC recvr
U R
longitudinal,
PAD
PAD
beam-based
PAC
feedback.
PAD
PAD
PAD PAC L1-X
PAD PAC
L1-S PAD
Beam Phase PAD
Monitor PAD
S20
Fast PACs: 8
Slow PACs (SPACs): 6
PADs: 19
VME crates: 1
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Overview of LLRF at Sector 24
RF phase and amplitude correction and global feedback at 120 Hz for LCLS LINAC S20
S24
Fast PACs: 4
Slow PACs (SPACs): 2
L24-1 PADs: 2
PAC VME crates: 1
L24-2
PAC
L24-3 VME Crate at S24
PAC
C E
running
Eth
Tcav L24-8
recvr
P V longitudinal,
PAC U R
beam-based
PAD
PAD feedback.
S29
SPAC
S30
SPAC
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Beam Phase Monitor
R. Akre
A. Haase
B. Hong
D. Kotturi
V. Pacak
H. Schwarz
Preliminary Design Review
November 30, 2005
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Outline
•Purpose
•Specifications
•System outline
•Cavity
•Noise Levels
•Analysis
•Long Term Drifts
•Summary
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
L INAC M DL R ef.
Laser Timing Stabilization Feedback
GU N R F F EEDB AC K
2856M Hz In pu ts
R F R EF . GU N-C ELL 1-PH AS/AMPL
L CL S R F
Osci lla tor GU N-C ELL 2-PH AS/AMPL
L ASER
2856M H z R ef Actu ator s
GU N R F AC T UAT OR S
PH ASE GU N PH AS AM PL
ER RO R
L 0A GU N R F R EF.
Ac tu ator
PH ASE E RR OR L0, L 1 t o L2, L3 L 0B GU N R F
B etween Phas e L ASER OSC IL LAT OR PH ASE AC TU AT ORS
L 0, L 1 a nd L 2, L 3 an d L ASER P OW ER
L 0-TC AV1 F EED B ACK
In pu ts
L 1-X L ASER OSC . P H ASE K LYST R ON
B UN CH C H ARGE AMPL IFI ER /
L ASER OSC IL LAT OR PH ASE SLC C ON T ROL
GU N-C ELL 1-AMPL /PH AS
F EED B ACK L 1-S GU N-F OR
GU N-C ELL 2-AMPL /PH AS
L ASER PH ASE & AM PLIT U DE
In pu ts GU N R F AC T UAT OR S
B EAM PH ASE C AVIT Y B EAM PH ASE C AVIT Y
Actu ator s Actu ator s
L ASER PH ASE AC T UAT OR L ASER POW ER
T OROI D
L ASER PH ASE AC T UAT OR
RF GUN
PH AS AM PL
L ASER OSC
OU T LASER
L ASER R F R EF.
R ef erenc e AMPLIFIER W ATER T EMP
L ASER
POW ER GU N T U N E
AC TU AT OR F EED B ACK
L ASER PH ASE
In pu ts
AC TU AT OR
GU N-F OR -PHAS
GU N-C ELL 1 GU N-C ELL 1-PH AS B EAM
PH ASE
L ASER OSC . P H ASE L ASER PH ASE & GU N-C ELL 2-PH AS
GU N-C ELL 2 B UN CH C AVITY
AMPL IT UD E?
Actu ator s C HAR GE
W ATER T EMP
Beam timing information from the beam phase monitor will be
used to apply corrections to the timing of the laser on the RF Gun.
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Specifications
Short term (2 second) timing jitter: 100fS rms
Long term (4 day) timing jitter: ±1pS
Range of the above accuracies is ±10pS
Data available at 120Hz
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
System Outline
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Cavity
Frequency = 2856MHz
Q = 6000
Time Constant = 700nS
Temperature Coefficient = 50kH/°C
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
System Critical Noise Levels and
Bandwidths
Cavity Signal – Bandwidth 500kHz
Local Oscillator – Noise Floor –143dBc/Hz
IF Filter – Bandwidth 4MHz
ADC – SNR at input 76dB
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
System Critical Noise Levels and Bandwidths
Filter - Butterworth
Beam Phase Cavity Attenuator 3rd order BandPass
Monitor Port 2.5.5MHz Center
30dBm pk 4.0MHz BW
2dB IL at 25.5MHz
Coupler Attenuator
30dBm pk 23dBm pk
3dBm pk ADC SNR 77dBFS
-174dBm/Hz
-174dBm/Hz Amp
In Tunnel LO RF ADC LTC2208
Monitor Port MIXER IF 2.25Vpp FS
-3dBm pk Transformer coupled
17dBm pk 2Vpp 10dBm pk
-146dBm/Hz 102MHz Clock
Generated from 119MHz Oscillator -129dBm/Hz Within filters BW
Expected SSB Phase Noise Levels -143dBc/Hz
Offset Hz dBc/Hz @ 2830.5MHz
-143dBc/Hz -135dBm/Hz
13dBm
10 -82 -143dBc/Hz
100 -96 -130dBm/Hz Beyond 5MHz from CF
1k -124 -143dBc/Hz <-155dBm/Hz
10k -144
20k -146 <-163dBc/Hz
1 Integrated Noise
2830.5MHz Oscillator -77dBc
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
ADC Linear
Technologies LTC2208
16Bit 130MHz
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF SNR 77.6dBFS 30MHz in Clock 130MHz SFDR 95dB
akre@slac.stanford.edu, dayle@slac.stanford.edu
Analysis
Phase
Time
Calculated Measured Measured
Beam Phase at Data Data
Beam Time Point 1 Point 2
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
I & Q from Waveform
Digital Down Mixing and Normalization
25.5MHz Digitized Signal
1
0.8
Digitized
0.6
Input Signal
Fraction ADC Full Scale
0.4
0.2
0
0.2
0.4
.
0.6
0 10 20 30 40 50 60
Point Number
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Optimization
Optimal Points to use for analysis is 16
point average at points 18 and 120
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Analysis Results
Standard deviation of result = 1.1e-4 or 6.3fS rms jitter
Signal level 20dB lower will give 63fS rms jitter
Sensitivity to frequency change = 0.6fS/2.8kH freq change
Sensitivity to timing change over +-10deg = 1:1
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Long Term Drifts
80ft (1M deg) of ½ inch superflex has TC of 4ppm/degC
Water temp tolerance is +-0.1degF = +-400fS drift
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Summary
Short term (2 second) timing jitter: 100fS rms
63fS rms
Long term (4 day) timing jitter: ±1pS
±0.8pS
Range of the above accuracies is ±10pS
Results
Data available at 120Hz
Simple algorithm in integer arithmetic will allow this
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Feedback Page 1
LOCAL FEEDBACK
LOCAL FEEDBACK
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
GLOBAL FEEDBACK
Feedback Page 2
LOCAL FEEDBACK
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Feedback Page 3 GLOBAL FEEDBACK
LOCAL FEEDBACK
LOCAL FEEDBACK
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
GLOBAL FEEDBACK
Feedback Page 4
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Feedback Page 5
GLOBAL FEEDBACK
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
Feedback Page 6
April 20, 2006 Ron Akre, Dayle Kotturi
LCLS LLRF akre@slac.stanford.edu, dayle@slac.stanford.edu
