Transcript Physics Requirements and Technology Choices for LCLS

Breakout Session: Controls
Physics Requirements
and Technology Choices for
LCLS Instrumentation & Controls
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Outline
Beam position monitors
Issues for the undulator cavity BPMs
Issues for signal processing
Power supplies and controllers
Pulsed operation of DL1 for diagnostics
Low level RF
Source and synchronization issues
Feedback and x-band regulation
Bunch length monitors
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Cavity Beam Position Monitors
Frequency choice
Cavity Iris should be masked from SR
Vacuum chamber dimensions for the undulator are now
chosen
12 mm aperture
is close to X-band cutoff
Evaluating two frequency choices (Z. Li)
Issues
5 mm
10 mm
BPM location with respect to quadrupoles
Resolution in combination with beam-based alignment
with EM quads
Signal processing
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Undulator Cavity BPM locations
with respect to quadrupoles
Quadrupole and BPM mounted adjacent on the undulator
support cradle to ensure 1 um beam based alignment
resolution
Also need to keep the distance between the electron beam
and the undulator segment axis to less than 70 microns rms
Considering beam position measurement options at downstream
end as well
Quad BPM
assemblies
Optional wire monitors,
Train-linked undulator sections – see H.-D. Nuhn presentation
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Cavity BPM Signal Processing
X and Y cavity at each undulator plus ~1 phase reference
cavity per girder
High-frequency x-band signal is attenuated in a short
distance
Incorporate a local mixer to IF at the cavity
Only a simple passive device in the tunnel
Temperature stable
Relatively low radiation loss environment
Distribution of reference x-band oscillator signal in the tunnel
Choose intermediate frequency to match into the RF front end used
for stripline
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Digital BPM Signal Processing
Use same RF front end for stripline BPMs and output from
first mixer for cavity BPMs
Initial desire to use a commercially produced BPM
processing module (Libera)
We obtained a try out Libera module
Integration into the control system not proceeding fast enough, e.g.
could not access raw data in the module.
Present design solution
Commercial VME 8 channel digitizer
RF front end from discrete, commercial components
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Power supplies and controllers
Requirements
Stability of 1E-5 for bunch compressors
fast response for feedback correctors
Integrate with epics controls
reliability
Design solution
digital controller/regulator
developed at PSI and further developed at Diamond
commercially supplied power modules
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Power supplies and controllers
Status
Test power supply delivered from PSI
controlled from an epics IOC
long term current stability tests into resistive
load are underway
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
PSI Power Supply 12 hour Test
<2.5E-5
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Pulsed operation of DL1 for diagnostics
VA
TR
EA
TM
CU
UM
EN
MP
S
HA
MB
ER
PU
TC
LO
A SS NG
EM B EL
BL LO
Y
WS
SP
SE O OL
AL S
FR
OM
VA
OD
E
UH
G ATV A
E VALLME
LV TA
ES L
PO
H OL RC
DE U PIN
R
E CA
TH
LV
E
•Propose to allow option of
pulsing DL1 bends
•allow pulse stealing at ~1 Hz
into the spectrometer line
•monitor beam profile and
energy spread
•Potentially combine with
pulsing of the transverse cavity
•Laminated magnet
•Experience at SLAC with
damping ring DRIP magnets
•Keep two dipoles in series
•Need to maintain 1E-4 stability
•Laminate magnets now
•Develop pulsed supply later
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Pulsed operation of DL1 for diagnostics
•1 Hz pulsed into the
spectrometer line
•monitor beam profile
•Investigate further if transverse cavity
can be optimized for slice measurements
in the spectrometer line
April 7-8, 2005
LCLS FAC
•monitor energy spread
Patrick Krejcik
[email protected]
Low Level RF
Feedback and x-band regulation
Question that arose last time was how to
distinguish drift in the X-band system from errors
in the S-band system
Solution is to keep X-band regulation fixed, and
compensate errors with the S-band system only
See next slide
Source and synchronization issues
noise and stability issues in oscillator and
distribution
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Demonstration of L1 S-band adjustment
to compensate Lx errors – courtesey Juhao Wu
X-band phase error of + 5o,
fixed with L1 S-band
adjustment: phase +2.1°,
voltage - 2.1 %
April 7-8, 2005
LCLS FAC
X-band amplitude error of 5%,
fixed with L1 S-band
adjustment: phase +0.61°,
voltage 0.18 %
Patrick Krejcik
[email protected]
Low Level RF Source and synchronization
Present design concept:
Microwave crystal oscillator phase locked to SLAC MDL
– low noise in the low frequency band
Gun laser oscillator mode locked to crystal oscillator –
low noise in the high frequency band
Under evaluation
Derive the LLRF 2856 MHz from crystal oscillator or
from laser optical output
Distribute LLRF over copper
or optional optical fiber
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
RF/Laser distribution
proposed by Ilday et al, MIT at the SLAC Timing workshop
Optical-laser
synchronization module
Master laser
oscillator
Upgrade path:
Fiber distribution
system
RF-optical
synchronization module
Low noise crystal
microwave oscillator
Baseline
Cu Coax distribution
LLRF to klystron
Linac MDL
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
RF stabilization – Ilday et al, MIT
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Derivation of LLRF from laser – F. Omer Ilday, MIT
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Synchronizing Gun and User Lasers – F. Omer Ilday, MIT
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
BC1, BC2 Single-shot Bunch Length Detectors
Non-intercepting detector for off-axis
synchrotron radiation
Reflected through a port to:
Spectral Power detector
Single shot Autocorrelator
THz
power
detector
THz
autocorrelator
B4 Bend
CSR
Bunch Compressor Chicane
April 7-8, 2005
LCLS FAC
Vacuum port with
reflecting foil
Patrick Krejcik
[email protected]
Bunch Length Monitor Issues
The CSR we now understand is dominated
by Coherent Edge Radiation
Same spectral and angular distribution
characteristics as transition radiation
Need to account for interference effects from
adjacent magnets
Experimental investigation at SPPS planned
Can also learn from UCLA expt at BNL-ATF
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Bunch Length Monitor Issues
Need practical experience in evaluating
window materials
Detectors (pyrometers, Golay cells, bolometers)
Autocorrelator designs (mirrors, splitters,
detectors)
New development of single-shot
autocorrelators
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
End of Presentation
Backup slides
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Power supply controller system layout
EPICS
I
O
C
8 ch
VME card
5MHz
Optical fiber
Power Supply
DSP Controller
Monitor
signals
AC
line
April 7-8, 2005
LCLS FAC
PWM
signal
ADC
Card
DCCT
load
PWM
AC Converter
Patrick Krejcik
[email protected]
PSI Digital Power Supplies
Fast Optical Link
(5 MHz)
Optical Trigger
Master
DSP
Controller
ADC/DAC
Card
0..6
Slaves
DIO
PWM
Signal
Power
Converter
I
U1..4
DCCT
Magnet
Courtesy A. Luedeke, PSI
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Stripline versus Cavity BPM Signals
P
l/4
Stripline
f
700 MHz
RF in
l
500 MHz
BP filter
C-band
cavity
~5 GHz
Dipole
mode
coupler
April 7-8, 2005
LCLS FAC
Digital
processing
ADC
x4
Mixer
LO
sync’ed to RF
IF
Control
system
119 MHz
Clock
24th harmonic
• noise (resolution) minimized
by removing analog devices in
front of ADC that cause
attenuation
• drift minimized by removing
active devices in front of ADC
Patrick Krejcik
[email protected]
SPPS Laser Phase Noise Measurements
476 MHz
M.O.
MDL
3 km
fiber
~1 km
Ti:Sa
laser osc
VCO
x6
2856 MHz
2856 MHz
to linac
EO
diode
Phase
detector
scope
April 7-8, 2005
LCLS FAC
– R. Akre
Patrick Krejcik
[email protected]
Courtesy F. Jenni, PSI
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]
Energy and Bunch Length Feedback Loops
E
Vrf(L1)
Φrf(L2)
E
Φrf(L3)
E
DL1
Φrf(L1) sz
E
Vrf(L0)
DL1
Spectr
.
L1
L0
Φrf(L2)
sz
BC2
BC1
L2
L3
BSY
50B1
DL2
Beam based feedback will stabilize RF F,A
Against drift and jitter up to ~10 Hz
But no diagnostic to distinguish drift of X-band
Linearization, higher-harmonic RF has the tightest tolerance
No unique beam measurement
April 7-8, 2005
LCLS FAC
Patrick Krejcik
[email protected]