Advanced Accelerator Research Department - Feedback ... · Advanced Accelerator Research Department...
Transcript of Advanced Accelerator Research Department - Feedback ... · Advanced Accelerator Research Department...
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Advanced Accelerator Research Department- Feedback & Dynamics Group -
J.D.Fox, C. Rivetta, T. Mastorides, O. Turgut
March 18, 2011
C. H. Rivetta ARD - Monthly meetings March 18, 2011 1
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
1 Introduction
2 Accelerating RF station - Beam dynamics interaction
3 Intra-bunch Instabilities: E-clouds / TMCI
4 Conclusions and Future plans
C. H. Rivetta ARD - Monthly meetings March 18, 2011 2
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Group’s Goal
Combine Beam dynamics, feedback system, signal processingand electronics hardware skills to conduct R & D to overcomefundamental limits in accelerators related to the beam intensityand luminosity.
1 Accelerating RF station - Beam dynamics interactionCoupled-bunch instabilitiesRF noise effects on beam distributionOptimal configuration of RF station
2 Intra-bunch Instability: Electron Clouds and Strong Head-TailInstabilities at SPS
Appropriated Modeling of the bunch dynamics for feedback controldesignBunch stabilizationLimitations in the feedback design due to noise, signalperturbations and hardware.
C. H. Rivetta ARD - Monthly meetings March 18, 2011 3
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Group’s Activities
Accelerating RF station - Beam dynamics interactionStudy of the Beam-RF station as a unique dynamical system.Beam Stability, RF station stability, RF station noise impact onbeam dynamics.RF station configuration tools and diagnostics based on beamperformance.
Intra-bunch Instability: Electron Clouds and Strong Head-TailInstabilities at SPS
Simulation Tools including feedback system.Reduced models and System Identification.Data Analysis.Hardware development for machine studies - Kicker design report.
Proposals: Timing & Synchronization, Multi-rate control of linearAccelerators
C. H. Rivetta ARD - Monthly meetings March 18, 2011 4
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Accelerating RF station - Beam dynamics interaction
Coupled-bunch instabilities - Setting of RF stations
Coupled-bunch instabilities are defined by the impedance andassociated circuitry of RF stations.
Cavity Controller (LLRF) feedback loops are employed to reduce theaccelerating fundamental impedance to achieve stable operation.Cavity Controller settings are critical for the stability of both the beam andthe RF station. Defines level of RF noise and perturbation of theaccelerating RF stations.
RFcav.
Supply
Klystron
Power
KlystronPolar Loop Modulator
V
DemodLLRF boards Beam
BeamPhase Loop
PhaseDetector
SetpointRF Reference
CavitySum
400.8 MHz
SynchroLoop
VCXO
−40 −30 −20 −10 0 10 20 30 4010−5
10−4
10−3
10−2
10−1
100
101
Mode Number
Gro
wth
Rat
e |σ
n| (s−
1 )
Growth Naked CavityDamping Naked CavityGrowth Direct LoopDamping Direct LoopGrowth Direct/Comb OnDamping Direct/Comb On
C. H. Rivetta ARD - Monthly meetings March 18, 2011 5
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Accelerating RF station - Beam dynamics interaction
With LARP/SLAC support and in collaboration with the CERNBE-RF group, SLAC personnel have successfully developed asuite of tools to remotely commission and optimally configure theLHC RF stations.
Remote operation was crucial under the new stricter CERN policies whichprevented tunnel access when the magnets are energized.
−2 −1.5 −1 −0.5 0 0.5 1 1.5 2−40
−30
−20
−10
0
10
Frequency (MHz)
Gai
n (d
B)
FitData
−2 −1.5 −1 −0.5 0 0.5 1 1.5 2−400
−200
0
200
400
Frequency (MHz)
Phas
e (d
egre
es)
The tools were essential forthe Winter 2010 and 2011commissioning.Tools reducedcommissioning from 1.5days/station to 1.5hours/station, increasedconsistency and reliability.
C. H. Rivetta ARD - Monthly meetings March 18, 2011 6
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
RF noise effects on LHC longitudinal beam emittance
The noise power spectrum of the RF accelerating voltage can strongly affect thelongitudinal beam distribution and contribute to beam motion and diffusion.
A theoretical formalism relating the bunch length growth with beam dynamics,accelerating voltage noise, and RF system configurations was developed atSLAC.
Studies and measurements were then conducted to estimate the bunch lengthgrowth for various operational configurations and to identify the RF componentscontributing to this effect.
100 101 102 103 104 105 106 107−180
−160
−140
−120
−100
−80
−60
−40
Frequency (Hz)
Cav
ity P
hase
Noi
se (d
Bc/
Hz)
BPL OnBPL Off
100 101 102 103 104 105 106 107−180
−160
−140
−120
−100
−80
−60
−40
Frequency (Hz)
Noi
se (d
Bc/
Hz)
G = 1125G = 281G=140G=20G=5G=0
C. H. Rivetta ARD - Monthly meetings March 18, 2011 7
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Longitudinal beam emittance - RF noise threshold
100 80 60 40 20 0 20 40 60 80 100130
120
110
100
90
80
70
Frequency (Hz)
Inje
cted
Noi
se 1
0log
(rad
2 /Hz)
70 dBc/Hz76 dBc/Hz82 dBc/Hz85 dBc/Hz88 dBc/Hz94 dBc/Hz100 dBc/Hz
0 2000 4000 6000 8000 10000 12000150
200
250
300
350
400
Bun
ch L
engt
h (p
s)
Time (s)
Based on noise studies and measurements a validated relationship betweenbunch length growth rate and noise in the accelerating RF stations was defined.
Studies were conducted with proton and ion beams.
A threshold level for the noise power spectrum was calculated to achievedσdt = 2.5ps/hr .
C. H. Rivetta ARD - Monthly meetings March 18, 2011 8
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Intra-bunch Instabilities: Electron Clouds / StrongHead-Tail instabilities at SPS
One of the limits in particle accelerators to increase current is due to intra-bunchinstabilities.
Large research effort has been conducted to understand and quantify the differentinstabilitiesMitigation for e-clouds induced instabilities have been proposed and tested: Coating,grooves in vacuum chambers, etc.Mitigation for transverse mode coupled instabilities are under study at SPS:Chromaticity
Our approach is to investigate the feasibility of controlling e-cloud and strongHead-Tail instabilities via feedback. Requires measuring/driving transversallydifferent areas of the bunch (nsec - psec scale)
!!!!!"#$!%!#&'()*+,!%!
-)./+,!
012.3!
4,5.+66)27!!$3822+(!
9+,:;!")6'(;!
9.! 9<!
y1…y64
Vb
Vb1…Vb64
Receiver + ADC Vc1…Vc8-16
9=!
Vy1…Vy8-16
+
Noise
C. H. Rivetta ARD - Monthly meetings March 18, 2011 9
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Intra-bunch Instabilities: Electron Clouds
Simple Observations from SPS Data
SPS MDs: June 2009, SPS injection 26GeV, Charge: 1E11p/bunch, separation 25 nsec.,
Time domain Vertical pick-up signals: SUM and DIFF - Extracted Vertical displacement(Data sampled 20 ps/point)
0 50 100−400
−300
−200
−100
0
100
200
300
400
500 Vertical displacement of bunch 47
slice
SUM
/ D
IFF
sign
als
(a.u
)
0 50 100−1
−0.8
−0.6
−0.4
−0.2
0
0.2
0.4
0.6
0.8
1
slice
Verti
cal d
ispl
acem
ent (
a.u)
SUMDIFF
0 50 100−400
−300
−200
−100
0
100
200
300
400
500 Vertical displacement of bunch 119
slice
SUM
/ D
IFF
sign
als
(a.u
)
0 50 100−1
−0.8
−0.6
−0.4
−0.2
0
0.2
0.4
0.6
0.8
1
slice
Verti
cal d
ispl
acem
ent (
a.u)
SUMDIFF
Two batches: First 72 bunches stable, (e.g. bunch 47), second set of 72 bunches E-cloudinstabilities, (e.g. bunch 119). Time span: 2.6 nsec.
C. H. Rivetta ARD - Monthly meetings March 18, 2011 10
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Intra-bunch Instabilities: Electron Clouds / StrongHead-Tail instabilities at SPS
R & D goal
Goal is to have a minimum prototype to fully understand the limitations offeedback techniques to mitigate E-cloud / TMCI effects in SPS.
High LevelSimulation
Reduced Model
ControlDesign
System Design -Implementation
Measurements Validation Tests Commissioning
R & D areasNon-Linear Simulation Codes (CMAD-WARP) - Real Feedback Models - Multi-bunchbehaviorDevelopment and Identification of Mathematical Reduced Dynamics Models for thebunch.Control AlgorithmsMD Coordination - Analysis of MD data - Data Correlation between MD data /Multi-particle resultsStudy and Development of Hardware Prototypes
C. H. Rivetta ARD - Monthly meetings March 18, 2011 11
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Next MD, we want to drive the bunch using the existent SPS kicker.
Critical to prove feasibility of driving different areas of the bunch, understand thepower level and kicker gain to design the new kicker prototype.
Combine results and studies from different areas of our R & D effort:multi-particle codes, bunch modeling and identification.
Test SLAC hardware - Back-end, Synchronization with SPS machine - Timing
Amplifier +Kicker
BeamExcitationVert. Displ.Vc Vb
Synch.Oscilloscope
Input OutputOff-‐Line Processing
Bunch Dynamics Reduced Model
Using a particular signals(VC), drive a beam throughthe amplifier and Kicker.
Measure the verticaldisplacement
Based on Input- Outputsignals, estimate the bunchdynamics and reduced model.
C. H. Rivetta ARD - Monthly meetings March 18, 2011 12
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Studies using CMAD multi-particle simulation code
CMAD multi-particle code is being up-dated to include realistic models of thefeedback control system.
Includes frequency response of the real hardware, allows to estimate andquantify the vertical displacement of different areas in the bunch.
Studies are conducted to analyze the impact of different signals driving thebunch.
0 10 20 30 40 50 60 70 80 90 1000
0.5
1x 10−3 Signal applied to the bunch
Mom
entu
m
0 10 20 30 40 50 60 70 80 90 100−5
0
5x 10−3 Y displacement
Y−ce
ntro
id [m
]
0 10 20 30 40 50 60 70 80 90 100−0.01
0
0.01
Y [m
]
Turns
C. H. Rivetta ARD - Monthly meetings March 18, 2011 13
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Studies using CMAD multi-particle simulation code
050
100
010
0
100
200
TurnsDAC samples
DAC
sig
nal (
a.u.
)
050
100
0
500
0.5
1
x 10−3
TurnsBunch SlicesM
omen
tum
[eV−
m/s
ec]
0 20 40 600
2
4
6x 10−4
Bunch Slices
Mom
entu
m [e
V−m
/sec
]
Turn # 20
0 20 40 60−4
−3
−2
−1
0
1x 10−3
Bunch Slices
Y [m
]
Turn # 29
0 20 40 60 800
1
2
3
4
5
6x 109
sample
Num
ber p
artic
les/
slic
e
0 20 40 60 80−1
−0.5
0
0.5
1Z displacement
Z [m
]
sample
0 50 1001.1
1.1
1.1001x 1011 Charge
Turn
Num
ber o
f par
ticle
s
C. H. Rivetta ARD - Monthly meetings March 18, 2011 14
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Studies using CMAD multi-particle simulation code
020
4060
80100
0
20
40
60−3
−2
−1
0
1
2
3
x 10−3
Turns
Y displacement
Bunch Slices
Y [m
]
C. H. Rivetta ARD - Monthly meetings March 18, 2011 15
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Hardware Development
In collaboration SLAC AE ControlElectronics Eng. group, developmentof hardware necessary for MDs andprototypes for evaluation of feedbacksystem performance.
Development of excitation board:4 GSamples/sec FPGA + DAC togenerate arbitrary signals andsynchronize with the SPS ring todrive the bunch.
Test and pack 100W - 1GHzbandwidth RF power amplifiers
Development of Matlab interface(GUI) to generate waveforms andcommands for the excitation board.
SPS Ecloud/TMCI CM-15 November 2010
Driven Beam Experiments
Progress - Synchronized excitation code - synch toSPS RF and turn
System can drive selected bunches in trains, 32samples/bunch ( 250 ps/sample)
Tunnel “cart” in progress for 2011 SPS MD
• 400W (4 100W) 20 - 1000 MHZ amplifiersordered- delivery December 2010
• Tunnel high-power loads
• Simple on/off remote control
Design in progress
• How much instrumentation is worth it?
• directional couplers on loads?
• Diode Detectors on reverse power and interlock?
• add equalizer to compensate for cable and kickerresponse? SPS Ecloud/TMCI CM-15 November 2010
Driven Beam Experiments
Progress - Synchronized excitation code - synch toSPS RF and turn
System can drive selected bunches in trains, 32samples/bunch ( 250 ps/sample)
Tunnel “cart” in progress for 2011 SPS MD
• 400W (4 100W) 20 - 1000 MHZ amplifiersordered- delivery December 2010
• Tunnel high-power loads
• Simple on/off remote control
Design in progress
• How much instrumentation is worth it?
• directional couplers on loads?
• Diode Detectors on reverse power and interlock?
• add equalizer to compensate for cable and kickerresponse?
C. H. Rivetta ARD - Monthly meetings March 18, 2011 16
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
MD Plans: Driving the bunch
Identification of Internal Bunch Dynamics: Reduced Model
Inject a random sequence or chirp signal –> measure the vertical displacementof different areas of the bunch.
Comparing the vertical displacement to the input signal we can calculate thereduced model
0 50 100 150 200 250 300 350 400−20
−10
0
10
20Measured & Estimated Output − Sample # 9
Turn
y &
y est
measured outputestimated output
0 50 100 150 200 250 300 350 400−20
−10
0
10
20Measured & Estimated Output − Sample # 10
Turn
y &
y est
measured outputestimated output
0 50 100 150 200 250 300 350 400−20
−10
0
10
20Measured & Estimated Output − Sample # 11
Turn
y &
y est
measured outputestimated output
C. H. Rivetta ARD - Monthly meetings March 18, 2011 17
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Conclusions and future plans
Combining Beam dynamics, feedback system, signal processing and electronicshardware skills, we conduct R & D to improve machine performance andovercome limits in accelerators related to the beam intensity and luminosity.
Understand, model and design accelerator units as a complete dynamicalsystem.
Beam Dynamics <–> Hardware <–> Perturbations: Stability -Performance
Collaborate with CERN LHC RF groups to commission and operate the LHC RFsystems at high beam currents. (T. Mastorides - Toohig Fellow)
R & D on feedback control of intra-bunch instabilities: SPS bunch length feasiblefor state-of-the-art of current electronic systems.
Collaborate with SSRL to improve RF station- longitudinal feedback at highcurrent
Evaluate proposal with LCLS on 1) Timing and Synchronization and 2) Multi-ratefeedback systems.
C. H. Rivetta ARD - Monthly meetings March 18, 2011 18
Introduction RF station / Beam dynamics eClouds/TMCI Conclusions
Acknowledgments
CERN BE-RF group (RF systems - e-clouds/TMCI)
LBL ( e-clouds/TMCI)
SLAC Accelerator Research Division - Initial support for R & D, Graduatestudent support
SLAC Collaborators
AE Control Electronics Eng. group: D. Van Winkle, J. Olsen, J.Dusatco, P. Keahey, A. Young (Hardware)ARD: Mauro Pivi (CMAD Multi-particle simulation code)Past Graduate Students: A. Bullitt, G. Ndabashimiye, M.SwiatlowskiSLAC User: I.Rivetta
Thanks to the audience....Questions??
C. H. Rivetta ARD - Monthly meetings March 18, 2011 19