1、P.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004LCLS Diagnostics and Commissioning Workshop Injector,Linac and Undulator Diagnostics and Beam Position Monitors P.KrejcikP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23
2、,2004Context of Diagnostics in CommissioningReview scope of proposed diagnosticsEmphasize that diagnostics themselves need commissioningConsider if full features(resolution,automation)are needed at beginning of commissioningImplicit sequence of commissioning:e.g.feedbacks after BPMs commissioned;sli
3、ce parameters need prof.monitors and TCAVsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Readiness of diagnostic systemsWhich SLC diagnostics should be preserved?Technology choices still being made on some new systemsBPM modules trying to attain desired r
4、esolutionProf monitor cameras resolution,controls integration,data rateSome diagnostics are turnkey systems,others are R&D projectsR&D still required for ultrafast diagnosticsCSR THz power bunch length monitorsEO bunch profilingP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.edu
5、September 22-23,2004Dynamic Aspects of CommissioningInitially diagnose a wildly mis-tuned and unstable machineYet the same diagnostics should ultimately have finesse to optimize SASE operationDeal with imperfect and uncalibrated settingsDetective work for finding hardware faultsQuantify magnitude an
6、d sources of jitterP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Diagnostics Roadmap for electronsTrajectoryresolutionPositionAngleEnergyBeam sizeresolutionEmittanceEnergy spreadBunch length&DTdevelopmentLongit.profileSingle shot rmsSlice parametersresol
7、ution Emittance Energy spreadBunch chargeFEEDBACKStabilizationresponseJitter characterizationNoninvasiveInvasive SetupTuning120 HzP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Accelerator System Diagnostics*upstream linacL0RFgunL3L1XL2 E s sEBC1BC2DL2DL1
8、undulatorLTUDumpE,D DEgegex,y*See also P.Emma talk how optics is optimized for diagnosticsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Beam Position Monitoring requirementsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember
9、 22-23,2004Beam Position MonitorsStripline BPMs in the injector and linac(existing)and in the LTUDifferencing large numbersMechanical precisionFabrication by printing electrodes on ceramic tubesDrift in electronicsDigital signal processingCavity BPMs in the undulator,LTU launchSignal inherently zero
10、 at geometric centerC-band(inexpensive)signal needs to be mixed down in the tunnelP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Stripline versus Cavity BPM SignalsPf700 MHz500 MHzBP filterADCx4119 MHzClock24th harmonicDigitalprocessingRF inControlsysteml
11、/4lStriplineMixerLO synced to RFIF noise(resolution)minimized by removing analog devices in front of ADC that cause attenuation drift minimized by removing active devices in front of ADCC-bandcavityDipole mode coupler5 GHzP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptem
12、ber 22-23,2004Simplistic View of Digital BPMsIs the purely digital approach the best way to go?Must always maximize signal to noise for best resolutionSo eliminate any cause of attenuation:couplers,hybrids,active devices etc.This also eliminates drift which causes offsetsOther approaches also try to
13、 do this:e.g.AM to PM conversion with a hybrid and then digitizeMight as well digitize first,eliminate the middle men,and do the conversions digitallyUltimately left with calibrating the drift in the BPM cables,because ADCs are now very stable.P.KrejcikLCLS Diagnostics and Commissioning Workshoppkrs
14、lac.stanford.eduSeptember 22-23,2004Linac stripline BPMsi-tech.siP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Analysis of Test Signals in the“Libera”module S.SmithMeasured signal to noise ratio implies resolution of 7 mm in a 10 mm radius BPMIdentified
15、fixable artifacts in data processingP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Cavity beam position monitors for the undulator and LTUCoordinate measuring machine verification of cavity interior X-band cavity shown Dipole-mode couplersR&D at SLAC S.Sm
16、ith X-band cavity shown Dipole-mode couplersNLC studies of cavity BPMs,S.Smith et alP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004C-band beam tests of the cavity BPM S.Smith25 mm200 nm Raw digitizer records from beam measurements at ATFcavity BPM signal
17、versus predicted position at bunch charge 1.6 nC plot of residual deviation from linear response(2p)1/2 szSLAC SPPS measurement:P.Muggli,M.HoganLimited by long wavelength cutoff and absorption resonances00.40.81.21.6-100-50050100CombinedCTRInterferogramsSmDiagnosing Coherent Radiation2.autocorrelati
18、onP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,200410-1710-1510-1310-1110-910-710-50.0010.11010-410-310-210-1100101001000CTRFSpecSigmaz20Mylar12.5_3 Power Spectrum(a.u.)Filter Amplitue(a.u.)Wavelength(m)MylarresonancesSimple model:Gaussian,sz=20 m,d=12.7 m,
19、n=3 Mylar window+splitter Smaller measured width:sAutocorrelation sbunch!Modulation/dips in the interferogram Fabry-Perot resonance:l=2d/m,m=1,2,Signal attenuated by Mylar:(RT)2 per sheetP.Muggli,M.HoganP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Devel
20、opments in autocorrelation techniquesInvestigate other detector types for wavelength dependanceGolay cellBeam splitters without wavelength dependanceSingle shot autocorrelatorCamera records fringes on single shotUse CSR from chicane bedP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stan
21、ford.eduSeptember 22-23,2004Coherent transition radiation wavelength comparable to bunch lengthLINACFFTBComparison of bunch length minimized according to wakefield loss and THz powerPyroelectric detectorfoilGADC-26-24-22-20-18-16-14-120100200300400500600Pyrometer signal arb.unitslinac phase offset f
22、rom crest deg.S-BandFFTB Pyrometer Signal-26-24-22-20-18-16-14-12200250300350400450500energy loss MeVLinac Wake LossLinac phaseWake energy lossTHz powerP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Dither feedback control of bunch length minimization at
23、SPPS-L.HendricksonDither time steps of 10 secondsBunch length monitor responseFeedback correction signalLinac phase“ping”optimumJitter in bunch length signal over 10 seconds 10%rmsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Diagnosing Longitudinal phas
24、e space:Energy spectrum versus Bunch length signal-Muggli,Hogan et alJitter in the compressor phase:Resuting energy profileCorresponding bunch length signalSingle shot measurementsjittersignalP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004LB=1.80 mB=1.60
25、TSPPS Four Dipole ChicanesLT=14.3 m9 GeVBPM-energyProf.Monitor-DEMomentum compactionR56=75 mms sz50 m mms s 1.6%s sz01.2 mmCorrelated energy spreadLinac chirpMeasured energy spreadSR backgroundP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Measured and pr
26、edicted energy spread from wakefield chirp in SPPSSpecial setup to give 100 mm bunch length with more charge at the head of the bunchMeasured at end of linacP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Wakefields change not only the energy spread in the
27、 bunchBut also the centroid energy of the bunch Fast means of determining relative bunch length P.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Relative bunch length measurementbased on wakefield energy loss scanEnergy change measured at the end of the lin
28、acas a function of the linac phase(chirp)upstream of the compressor chicaneShortest bunch has greatest energy lossPredicted wakeloss_For bunch length s s z _Predicted shape due to wakeloss plus RF curvatureP.Emma,K.BaneP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember
29、 22-23,2004Frequency domainSpectral powerSpectrometryAutocorrelation P.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004SPPS Electro Optic Bunch Length Measurement with in-vacuum crystalProbe laserDefining apertureBeam axisM1M2EO xtalGeometry chosen to measur
30、e direct electric field from bunch,not wakefieldModelled by H.SchlarbelectronsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Features of the SPPS Electro Optic SetupTi:SaoscillatorStretcher Shaper Fiberlaunch e-EO xtlpolarizing beamsplitterspimaging optic
31、s150 m fiberOTRP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Features of the SPPS Electro Optic SetupSLM 640-pixelffffTo fiberGrating pairFrom stretcherP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Features o
32、f the SPPS Electro Optic SetupCrystal mounted close to electron beamAvoid wakefields from smaller aperturesZnTe crystal:200 um thickEO coefficient,phase match,phonon resonancesP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004ErLine image camerapolarizeranal
33、yzerPol.Laser pulseElectron bunchEO crystalBunch length scanP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Spatial imaging resolution limits time resolutionCrossing angle determines width of time window and temporal resolutionResolution limit then set by
34、crystal thickness and the phase velocity mismatch Crystal material chosen to minimize phase mismatchP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Electro optic resolution limitsFuture experimentsSmaller crossing angleSmaller angle magnifies time coordina
35、te on spatial axisBut reduces the time window to accommodate beam jitterEO polymer filmsStrong EO coefficientMay not last longHigher laser power cross correlation techniques(Jamison et al)Laser amplifier located near beamlineNew chamber T.MontagneP.KrejcikLCLS Diagnostics and Commissioning Workshopp
36、krslac.stanford.eduSeptember 22-23,2004Synchronization of the Laser timingJitter in the laser timing effectsElectro optic bunch timing measurementPump-probe timing for the usersEnhancement schemes using short pulse lasersP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptemb
37、er 22-23,2004SPPS Laser Phase Noise Measurements R.Akre 476 MHz M.O.x62856 MHz to linacMDL3 kmfiber1 kmVCOTi:Salaser oscdiodeEOscopePhase detector2856 MHzP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Energy and Bunch Length Feedback LoopsL0L1DL1DL1Spectr
38、.BC1BC2L2L3BSY 50B1DL2Vrf(L0)rf(L2)Vrf(L1)rf(L3)EEErf(L2)szrf(L1)szE4 energy feedback loops2 bunch length feedback loops120 Hz nominal operation,1 pulse delayProgressive commissioning scheduleP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Undulator trajec
39、tory launch loop to operate at 120 Hz,1 pulse delayDamps jitter below 10 Hzi.e.need stability above 10 Hz!At lower rep.rates,less dampingLinac orbit loops to operate at 10 Hz because of corrector response timeAntidampDampGain bandwidth shown for different loop delays-L.HendricksonClosed Loop Respons
40、e of Orbit FeedbackP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Remaining intra-undulator diagnostics from Bingxin Yang,Lehman Review August 04Location:every long break(905 mm)Diagnostics chamber length:425 mmFunctional componentsRF BPM,Cherenkov detect
41、or,OTR profiler,wire scanner,x-ray(intensity)diagnosticsP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004FY04 accomplishments from Bingxin Yang,Lehman Review August 04Layout of diagnostics chamberOTR profilerCamera module designed Wire scannerScanner design
42、 in progressWire card adapt SLAC designX-ray diagnostics designBeam intensity:double crystalBeam profile:imaging detectorP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Major issues at UCLA workshop from Bingxin Yang,Lehman Review August 04Beam damage of o
43、ptical componentsExample from Marc Ross coupon test,LINAC 2000Saturated FEL beam deposit higher energy densityDesirable informationTrajectory accuracy(Dx1mm)Effective K(DK/K 1.510-4)Relative phase(Df10)Intensity gain(DE/E0.1%,z-)Undulator field qualityP.KrejcikLCLS Diagnostics and Commissioning Work
44、shoppkrslac.stanford.eduSeptember 22-23,2004Rethink x-ray diagnostics(Galayda)from Bingxin Yang,Lehman Review August 04Intra-undulator diagnosticsElectron beam position monitor(BPM)Electron beam profiler(OTR&wire scanner)Low power x-ray Intensity measurements(R&D)Beam loss MonitorFar-field low-power
45、 x-ray diagnostics(R&D)Clean signature from spontaneous radiationSpace for larger optics/detectorsSingle set advantage(consistency,lower cost)Goal=obtain“desirable information”P.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Final Beam DumpSensitive measure
46、ment of beam energyOptimized for energy spread resolution of 4*10-5(P.Emma)Bends smear out microbunchingDispersion hides emittance measurementMight be possible in the vertical planeP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004SummaryDiagnostics integrat
47、ed into the LCLS designAll systems require commissioning time to achieve LCLS resolution requirementsNew diagnostics still require R&D for bunch length and timing Developmental work at SPPS is criticalDiagnostics being developed hand-in-hand with controls and feedbacksP.KrejcikLCLS Diagnostics and Commissioning Workshoppkrslac.stanford.eduSeptember 22-23,2004Appendix谢谢
