LOLA: Past, present and future operation FLASH Seminar 1/2/29 Christopher Gerth, DESY 8/5/29 FLASH Seminar Christopher Gerth 1
Outline Past Present Future 8/5/29 FLASH Seminar Christopher Gerth 2
Past - Present - Future Installed at TTF2 in 23 Designed at SLAC for particle separation in 1966 (4 LOLAs). As diagnostic tool installed at SPPS/LCLS in 2. Will be moved from ACC7 to new position upstream of SASE undulators in shutdown 9/29. 8/5/29 FLASH Seminar Christopher Gerth 3
LOLA operates as streak camera Camera Transverse density distribution LOLA Screen vertical vertckal y [mm] 1 2 3 4 -Fy Fy -.4 -.2.2.4 x [mm] horizontal 5 6 1 2 3 4 s [mm] longitudinal 8/5/29 FLASH Seminar Christopher Gerth 4
LOLA operates as streak camera Camera Transverse density distribution: LOLA Screen vertical 1 2 vertical y [mm] 3 4 -.4 -.2.2.4 x [mm] horizontal 5 longitudinal 6 1 2 3 4 s [mm] 8/5/29 FLASH Seminar Christopher Gerth 5
LOLA operates as streak camera Camera Transverse density distribution LOLA Screen vertical longitudinal 1 vertical y [mm] 2 3 4 -.4 -.2.2.4 x [mm] horizontal 5 longitudinal 6 1 2 3 4 s [mm] 8/5/29 FLASH Seminar Christopher Gerth 6
Methods Longitudinal Resolution y } } ζ LOLA Y [mm] -8-6 -4-2 2 4 6 Estimation of LOLA off 8-3 -2-1 1 2 3 X [mm] Y [mm] -8-6 -4-2 2 4 6 Measurement of y as a function of the RF-phase LOLA on y 8-3 -2-1 1 2 3 X [mm] Typical values: 8/5/29 FLASH Seminar Christopher Gerth 7
The Present Setup Current position of LOLA RF gun Dump ACC1 ACC2 ACC3 ACC4 ACC5 ACC6 BC2 BC3 LOLA Undulator magnets OTR screen for bunch profile / slice emittance meas. OTR screen for long. Phase space meas. 8/5/29 FLASH Seminar Christopher Gerth 8
Setup LOLA at position ACC7 LOLA Installed in 23, Collaboration DESY- SLAC Frequency: 2.86 GHz Length: 3.6 m Maximum deflecting voltage ~ 25 MV @ 2 MW input power Beam direction 8/5/29 FLASH Seminar Christopher Gerth 9
On-crest operation Slice Emittance Measurements: Longitudinal Phase Space: γε x [µ m] I [ka] x [mm] 4 3 2 1 5 4 3 2 1.5 -.5-6 -5-4 -3-2 -1 ζ [mm] 49 MeV,.6 nc E/E [%].2.1 -.1 -.2 -.3 Single shot -.4-6 -4-2 2 ζ [mm] 65 MeV, 1nC, compressor chicanes switched off 8/5/29 FLASH Seminar Christopher Gerth 1
Operation @ SASE conditions Slice Emittance Measurements: γε x [µ m] 14 12 1 8 6 4 2 Slice emittance ~2µm ~1 µm Longitudinal resolution ~ 8 µm (RMS) I [ka] x [mm] 1.6 2 1.2.8.4 1-1 Current profile Head 494 MeV,.7 nc -.15 -.1 -.5.5.1 ζ [mm] Increase in slice emittance in the peak current region: Cause? In contradiction to SASE theory!? 8/5/29 FLASH Seminar Christopher Gerth 11
CSR effects I [ka] Current profile: RF-phase of module ACC1 adjusted in simulation 2 1.5 1.5 Measurement Simulation -.2 -.15 -.1 -.5.5.1.15 ζ [mm] γ ε x [µ m] 14 12 1 8 6 4 2 Slice emittance Measurement Simulation no CSR -.2 -.15 -.1 -.5.5.1.15 ζ [mm] Simulations by M Roehrs with ASTRA (K. Flöttmann) and CSRTrack (M. Dohlus) 8/5/29 FLASH Seminar Christopher Gerth 12
Phase-space Tomography PhD thesis M. Roehrs γε x [µ m] I [ka] 16 14 12 1 8 6 4 2 2 1.6 1.2.8.4 Slice emittance Current profile Tomography Least squares -.15 -.1 -.5.5.1 ζ [mm] longitudinal slice position, thickness ζ = 8 µm x [mrad] Reconstructed phase space:.2.1 -.1 -.2-2 -1 1 2 x [mm] 1.8.6.4.2 8/5/29 FLASH Seminar Christopher Gerth 13
Phase-space Tomography PhD thesis M. Roehrs γε x [µ m] I [ka] 16 14 12 1 8 6 4 2 2 1.6 1.2.8.4 Slice emittance Current profile Tomography Least squares -.15 -.1 -.5.5.1 ζ [mm] x [mrad] Reconstructed phase space:.2.1 -.1 -.2-2 -1 1 2 x [mm] 1.8.6.4.2 longitudinal slice position, thickness ζ = 8 µm 8/5/29 FLASH Seminar Christopher Gerth 14
Phase-space Tomography PhD thesis M. Roehrs Measured distribution in the peak current region 2-dimensional Gaussian fit to the peak 8 µm, 1.2 ka 2 µm,.7 ka typical: 2-4 m normalized emittance,.5 1. ka peak current FEL radiation not saturated (measurements during FEL study shifts) peak current may change downstream of LOLA (e.g. in the dogleg) 8/5/29 FLASH Seminar Christopher Gerth 15
Summary: Present LOLA does not measure the bunch parameters at the undulator; dogleg contributes to the compression process Special optics required for LOLA measurements - online measurements during user operation not possible Temporal resolution not sufficient to fully resolve the SASE spike in short-pulse mode (without 3 rd harmonic cavities). Even slice emittance values are of limited use (at least for comparison with SASE theory). Tomographic methods required to access the SASE spike Transverse space charge effects slightly distort the tomographic method. Inclusion could improve reconstruction. 8/5/29 FLASH Seminar Christopher Gerth 16
Good bye Michael Dr. Michael Röhrs Young Scientist Prize at the Int. FEL Conference 28. 8/5/29 FLASH Seminar Christopher Gerth 17
New location of LOLA Shutdown Sept 29: LOLA will be moved upstream of the SASE undulators RF gun Dump ACC1 ACC2 ACC3 ACC4 ACC5 ACC6 BC2 BC3 ACC7 sflash LOLA Undulator magnets 8/5/29 FLASH Seminar Christopher Gerth 18
Goals for new setup standard optics for on-line bunch profile monitor - optics layout with optimum LOLA performance so that QUAD settings do not have to be changed. - matching into SASE undulators (4 options, energy dependent) Design of spectrometer beamline has started - longitudinal Phase Space measurements - dump for commissioning (1-3 bunches) Quad scan for slice emittance measurement - special optics needs to be designed: change of phase advance from LOLA to screen by 18 deg in x with upstream QUADs while keeping the beam sizes in x and y constant at the screen. 8/5/29 FLASH Seminar Christopher Gerth 19
New Lattice Layout 8/5/29 FLASH Seminar Christopher Gerth 2
Later Upgrade with UND7 Installation of undulator UND7 should be possible with minor modifications 8/5/29 FLASH Seminar Christopher Gerth 21
LOLA Resolution Resolution: vertical beam size at OTR / shear parameter Beam size at OTR: Shear parameter: Deflecting voltage: Maximum input power at LOLA: 18MW => +6m waveguides => 12 MW Maximum deflecting voltage V : 25 MV => 2.5 MV Beam size at screen: Not resolution limited by optical system Has to fit on the screen (with timing jitter, ~.2mm*S) 8/5/29 FLASH Seminar Christopher Gerth 22
Online Bunch Profile Measurement Layout has to be integrated into overall lattice - check optics for SASE compatibility (matching, focusing, steering,..) 8/5/29 FLASH Seminar Christopher Gerth 23
S2E Simulations Bunch.6.4.2 Short bunch mode without 3 rd harmonic Space charge forces Operation with 3 rd harmonic Courtesy of I. Zagorodnov E/E[%] -.2 -.4 -.6 -.8-1 CSR -.6 -.5 -.4 -.3 -.2 -.1 ζ [mm] 677 MeV,.5 nc 8/5/29 FLASH Seminar Christopher Gerth 24
Energy Spectrometer / Dump Energy resolution: Aim: σ δ = 5 kev (5 1-5 @ 1GeV) D x 1 m σ x = (ß ε) ½ < D x σ δ ß 4 m Degrading effects: 1. Energy spread induced by off-axis longitudinal fields of LOLA 2. CSR effect in dipole magnet 3. Other effects: optics, chromaticity 8/5/29 FLASH Seminar Christopher Gerth 25
Uncorrelated Energy Spread Energy spread induced by LOLA Uncorrelated energy spread induced by LOLA due to off-axis longitudinal accelerating fields (for round beams): Behrens-Gerth Relation: => The energy spread induced by LOLA is anti-proportional to the time resolution. 8/5/29 FLASH Seminar Christopher Gerth 26
Uncorrelated Energy Spread Elegant simulation: Energy spread induced by LOLA 8/5/29 FLASH Seminar Christopher Gerth 27
Correlated Energy Spread Slice energy spread and centroid tilt due to longitudinal accelerating fields of LOLA: - Initial Gauss bunch - ß y = 4m, V = 8.5 MeV - ß y = 7m, V = 2.5 MeV Both cases have the same time and energy resolution but different beta-functions and deflecting voltages V. A smaller beta-functions leads to a larger phase advance within LOLA and stronger shearing and therefore a larger correlated energy spread. 8/5/29 FLASH Seminar Christopher Gerth 28
CSR effects Longitudinal Phase Space at position of OTR screen in energy spectrometer Simulation of measurement at position of OTR screen in energy spectrometer - Bend angle 1 deg - Bend angle 5 deg - Bend angle 1 deg - Bend angle 5 deg Correlated energy chirp induced by LOLA partly compensated by CSR No big difference between 5 and 1 deg bend angle in terms of BD 1 deg option gives technical advantages 8/5/29 FLASH Seminar Christopher Gerth 29
CSR effects Slice energy spread induced by CSR negligible compared to energy spread induced by LOLA! 8/5/29 FLASH Seminar Christopher Gerth 3
S2E Bunch Simulation of longitudinal phase space measurement - S2E bunch upstream LOLA (I Zagorodnov) - S2E bunch measured on screen 8/5/29 FLASH Seminar Christopher Gerth 31
Conclusion: Future Current layout of new LOLA section (MATCH) allows for online bunch profile measurements. Layout is being integrated and optics is checked for compatibility with SASE operation. Technical design has started (check for collisions, space for pumps and bellows, design of dipole chamber, ) Measurements are being simulated with S2E bunches. 8/5/29 FLASH Seminar Christopher Gerth 32