Post on 18-Jan-2018
description
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
LCLS - Accelerator System OverviewLCLS - Accelerator System Overview
Patrick Krejcikon behalf of the LCLS Team
Stanford Linear Accelerator Center
LCLSLCLS
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Ti:Sa laserRG Gun
Laser HeaterUpstream linacL0
Linac1
RF PhotoInjector 135 MeVx,y = 1mm mrad, q=1 nCz = 3ps, 0.05 % 0.05 %
Linac2
Linac3
BC1 chicane 250 MeVz = 0.63 ps
Existing SLAC structurewarm, copper linacS-band 2856 MHz120 Hz
X-band harmonic cavity
BC2 4.5 GeVz = 0.07 ps
UndulatorL =130 m
X-rays1.5 Å
LTU 14.1 GeVz = 0.07 ps, 0.01 % 0.01 %
1 kmLoLa transverse cavity
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Accelerator Issues in the SLAC Design
IssuesIssues Low emittance injectorLow emittance injector
Cold beam, low Cold beam, low 0.05 % 0.05 %
Bunch compressionBunch compressionCoherent Synchrotron Coherent Synchrotron RadiationRadiationLongitudinal space chargeLongitudinal space charge
Beam stabilityBeam stability
Design solutionsDesign solutions RF photoinjectorRF photoinjectorLaser heaterLaser heaterTwo magnetic chicanesTwo magnetic chicanesRF linearization with RF linearization with higher harmonic X-band higher harmonic X-band cavitycavityDiagnosticsDiagnosticsFast feedbackFast feedback
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Vrf Electron bunch S-band rf = -25º
energy
time
Electron bunch with higher energy tail
…tail begins to catch up
…fully compressed
additionalX-band rf = -160º
Bunch Compression Dynamics
Longitudinal tracking with litrack – P. Emma
Remove spike in distribution with addition of X-band
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Diagnostic challenges
MeasurementMeasurement Measurement of ultra-Measurement of ultra-short bunch profilesshort bunch profilesShot-by-shot Shot-by-shot measurement of bunch measurement of bunch lengthlengthBunch timing Bunch timing measurementmeasurement
DevicesDevices RF transverse deflecting RF transverse deflecting cavity “LOLA”cavity “LOLA”Terahertz coherent Terahertz coherent spectral power spectral power measurementmeasurementCoherent radiation Coherent radiation autocorrelationautocorrelationElectro-optic bunch Electro-optic bunch profilingprofiling
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Bunch Length Measurements with the RF Transverse Deflecting Cavity
yy
Asymmetric parabola indicates incoming tilt to beam
Y = A * (X - B)**2 + CA = 1.6696E-02 STD DEV = 1.3536E-03B = 28.23 STD DEV = 3.084C = 1328. STD DEV = 8.235RMS FIT ERROR = 23.63
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
****
***
**
**
*
****
********
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
E
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
-80 -40 0 40 80SBST LI29 1 PDES (S-29-1)
1.7
1.6
1.5
1.4
1.3
X103
E
MANUAL STEPPING. STEPS = 30
1-APR-03 20:21:16
Cavity on
Cavity off
Cavity on- 180°
Bunch length reconstructionMeasure streak at 3 different phases
z = 90 m
(Stre
ak s
ize)
2
0 180
2.4 m 30 MW
LoLa*
An S-band DLW structure with a TM11 transverse deflecting mode at 2856 MHz
*Loew, Larsen 1964
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Accelerator Issues in the SLAC Design
IssuesIssues Low emittance injector
Cold beam, low Cold beam, low 0.05 % 0.05 %
Bunch compressionBunch compressionCoherent Synchrotron Coherent Synchrotron RadiationRadiationLongitudinal space chargeLongitudinal space charge
Beam stability
Design solutionsDesign solutions RF photoinjectorLaser heaterLaser heaterTwo magnetic chicanesRF linearization with higher harmonic X-band cavityDiagnosticsDiagnosticsFast feedback
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Limitation from Coherent Synchrotron Radiationrrzz radiation coherentradiation coherent
Energy spread from CSR increases x
P. Emma
With CSR:With CSR: //00 14 14
no CSR:no CSR:
Limits compressionhypothetical 1 m bunch
CSR instabilityCSR instability amplifies noise in initial charge distribution
S. Heifets, S. Krinsky, G. Stupakov, S. Heifets, S. Krinsky, G. Stupakov, SLAC-PUB-9165, March 2002SLAC-PUB-9165, March 2002
33101066
Microbunching*
and further emittance growth
* First observed by M. Borland (ANL) in * First observed by M. Borland (ANL) in LCLS LCLS ElegantElegant tracking tracking
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Two-stage bunch compression approach – P. Emma
IssuesIssues At At low energieslow energies if bunch if bunch is compressed too much is compressed too much space chargespace charge spoils spoils emittanceemittanceAt At high energieshigh energies if bunch if bunch is compressed too hard is compressed too hard synchrotron radiationsynchrotron radiation adds large energy spreadadds large energy spread
Design solutionsDesign solutions Compress in two stagesCompress in two stagesLimit low energy Limit low energy compression so space compression so space charge not a limitcharge not a limitSecond compression to Second compression to final bunch length at final bunch length at higher energy, but with higher energy, but with weaker bends to limit weaker bends to limit synchrotron radiation.synchrotron radiation.
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Transition radiation is coherent at wavelengths longer than the bunch length,>(2)1/2 z
SLAC SPPS measurement: P. Muggli, M. Hogan
Limited by long wavelength cutoff and absorption resonances
0
0.4
0.8
1.2
1.6
-100 -50 0 50 100CombinedCTRInterferogramsSm
zz 9 9 mm
Diagnosing Coherent Radiation1. autocorrelation
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
m 190z
Diagnosing Coherent Radiation2. spectral power
Smooth Gaussian bunch spectrum from BC1
With 5% microbunching
• Measure bunch length
• Detect microbunching
Fixed BW detectorSignal prop. 1/z
Bunch length signal for RF feedback
- J. Wu
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
40 keV rms40 keV rms
‘‘Laser Heater’ for Landau DampingLaser Heater’ for Landau Damping
Z. Huang Z. Huang et al.et al.PR ST ABPR ST AB, , June 2004June 2004
Ti:Sa 800 nm 1.2 MWTi:Sa 800 nm 1.2 MW
Injector e- 135 MeVInjector e- 135 MeV
‘‘Laser heater’ Laser heater’ suggested by suggested by Saldin et al.Saldin et al.
heater heater ONON
0.01 %0.01 %
heater heater OFFOFF
0.1%0.1%
Final energy distribution:
Microbunch instability
Damped
IFELenergy modulation
chicaneenergy mixing
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Electro-Optical Sampling at SPPS Electro-Optical Sampling at SPPS – A. Cavalieri et al.– A. Cavalieri et al.
ee
Ti:Sapphire Ti:Sapphire laserlaser
Single-ShotSingle-Shot
<300 fs<300 fs
170 fs rms170 fs rmsTiming JitterTiming Jitter
Er
Line image camera
polarizer
analyzer
200 200 m thick ZnTe crystalm thick ZnTe crystal
Pol. Laser pulse
Electron bunch
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Energy and Bunch Length Feedback Loops
4 energy feedback loops2 bunch length feedback loops120 Hz nominal operation, <1 pulse delay
Feedback model (J. Wu)PID controller (proportional, integral, derivative)Cascade control for sequential loops (off-diagonal matrix elements)
L0 L1
DL1
DL1Spectr. BC1 BC2
L2 L3BSY 50B1 DL2
Vrf(L0)
Φrf(L2)Vrf(L1) Φrf(L3)E E E
Φrf(L2) zΦrf(L1) zE
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Bode Plot (E/E)
off
onEEEE
//
log20 10
P:0.2; I:0.5
P:0.2
on
off
/arg 0.0
/E EE E
Ø 3 different settings of the PID controller
ØIntegral term dominant
I:0.5
Energy feedback loop response - J. WuP. Emma
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Bode Plot (I/I)
on10
off
/20log
/I II I
P:0.2; I:0.5
P:0.2
on
off
/arg 0.0
/I II I
Ø 3 different settings of
the PID controller
ØIntegral term dominant
I:0.5
Bunch length feedback loop response - J. WuP. Emma
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Summary
Design optimized for emittance preservationMinimize disruption from strong self-fields of the bunchTwo-stage compressionLaser heater reduces instabilitiesDiagnostics and feedback integral part of designFuture expansion to multiple sase beamlinesNew possibilities include enhanced sase and ultra-short bunches!
P. Krejcik
LINAC 2004 – Lübeck, Germany pkr@slac.stanford.edu
August 16-20, 2004
Acknowledgements
R. AkreA. CavalieriE. BongD. DowellP. Emma Z. Huang C. Limborg-DupreyJ. WuAnd many others!
and thank you to the LINAC 2004 organization