CLIC Main Linac studies N Blaskovic CLIC Beam
CLIC Main Linac studies N. Blaskovic CLIC Beam Physics meeting 1
Acknowledgements • Thanks to the CLIC Beam Physics team, in particular to: – Daniel Schulte – Andrea Latina – Edu Marin – Chetan Gohil N. Blaskovic CLIC Beam Physics meeting 2
Contents • Energy spread studies at 380 & 350 Ge. V • MICADO orbit correction studies N. Blaskovic CLIC Beam Physics meeting 3
Energy spread studies at 380 & 350 Ge. V with D. Arominski, C. Gohil, D. Schulte & R. Yang N. Blaskovic CLIC Beam Physics meeting 4
Energy spread • Energy spread develops in the bunch from – RF time-varying profile – Bunch wakefield effect N. Blaskovic CLIC Beam Physics meeting 5
RF contribution off-crest RF phase 25° 0° head of bunch N. Blaskovic on-crest tail of bunch CLIC Beam Physics meeting 6
Wakefield contribution head of bunch N. Blaskovic tail of bunch CLIC Beam Physics meeting 7
Combined contribution off-crest 15° smallest energy spread for RF phase ~ 15° 0° head of bunch N. Blaskovic RF phase 25° on-crest tail of bunch CLIC Beam Physics meeting 8
Emittance growth • Bunch tail experiences a defocusing force due to transverse wakefields, leading to emittance growth and ultimately beam break-up • Mitigate emittance growth by using BNS damping (Balakin, Novokhatsky and Smirnov, HEACC 1983) N. Blaskovic CLIC Beam Physics meeting 9
BNS damping • Introduce energy spread along the bunch so that tail has a lower energy than head • Thus, quadrupole focusing (stronger for bunch tail due to lower energy) balances transverse wakefield’s defocusing force • Lower energy tail can be achieved by accelerating on-crest at start of linac • Subsequently remove this energy spread by accelerating off-crest at end of linac N. Blaskovic CLIC Beam Physics meeting 10
Poor BNS damping original assume no incoming energy spread low energy spread huge emittance! original N. Blaskovic CLIC Beam Physics meeting 11
Good BNS damping original assume no incoming energy spread poor good poor original good RF: 0° N. Blaskovic RF: 0° RF: 20° RF: 30° CLIC Beam Physics meeting 12
380 Ge. V design • Baseline 380 Ge. V design simulated with: – Nominal bunch charge: 5. 2 x 109 particles – Nominal bunch length: 70 um • Energy spread optimised at 0. 35% (std) at the end of the linac by D. Schulte et al. • Latest RF phases selected by C. Gohil for low emittance & 0. 35% std energy spread: main linac RF: 6° N. Blaskovic RF: 6° RF: 14° RF: 30° CLIC Beam Physics meeting 13
RF phase combinations 190 Ge. V linac injection at 9 Ge. V N. Blaskovic RF: 0° RF: 0° RF: 15° RF: 0° RF: 30° RF: 15° RF: 30° RF: 30° RF: 15° RF: 30° RF: 6° RF: 14° RF: 30° CLIC Beam Physics meeting optimum 14
Simulation • • 200 slices, 2000 macroparticles/slice Emittance calculated relative to bunch axis No incoming beam offset Incoming RMS energy spread: 1. 6% RMS vertical quad position error: 10 nm Incoming horizontal emittance: 920 nm Incoming vertical emittance: 10 nm RF accelerating gradient: ≤ 72 MV/m N. Blaskovic CLIC Beam Physics meeting 15
RF phase scan • Scan sequentially fourth, third and second sectors’ RF phases from 0° to 30° main linac N. Blaskovic RF: 0° RF: 0° RF: 30° RF: 30° RF: 29° RF: 30° CLIC Beam Physics meeting 16
RF phase scan 2 nd sector scanned 0° to 29° (last 2 sectors at 30°) all sectors at 0° emittance growth < 1 nm 10 nm incoming emittance N. Blaskovic 3 rd sector scanned 0° to 30° (last sector at 30°) last sector scanned 0° to 30° CLIC Beam Physics meeting 17
Further RF phase scan • Using first sector phases > 0° • RF phases of all 4 linac sectors scanned together from 0° to 30° • For each phase, sequentially increase: – Last sector’s phase up to 30° – Preceding sectors’ phases up to 30° N. Blaskovic CLIC Beam Physics meeting 18
Further RF phase scan C. Gohil’s phase (0. 35% energy spread): sector 1: 6° sector 2: 6° sector 3: 14° sector 4: 30° all sectors at 0° N. Blaskovic CLIC Beam Physics meeting 19
Comparison 190 Ge. V linac injection at 9 Ge. V N. Blaskovic CLIC Beam Physics meeting 20
Comparison 190 Ge. V linac injection at 9 Ge. V 1. 6% energy spread from RTML optimum N. Blaskovic CLIC Beam Physics meeting reduction in final energy spread by increasing final RF phase at end of main linac 21
Comparison 190 Ge. V linac injection at 9 Ge. V 1. 6% energy spread from RTML optimum 10 nm incoming emittance N. Blaskovic optimum CLIC Beam Physics meeting reduction in final energy spread by increasing final RF phase at end of main linac Final emittance < 11 nm 22
380 Ge. V operation • The target energy spread of 0. 35% (std) is achieved at the end of the main linac whilst preserving a small emittance growth N. Blaskovic CLIC Beam Physics meeting 23
350 Ge. V operation • N. Blaskovic CLIC Beam Physics meeting 24
Optimum bunch length & charge reduce bunch charge increase bunch length N. Blaskovic CLIC Beam Physics meeting 25
Optimum bunch length & charge • Scan RF phase combinations over a range of bunch lengths and charges to find Emittance* Energy spread 380 Ge. V 18 nm 0. 35% Target < 14 nm < 0. 20% whilst preserving an acceptable bunch charge for good luminosity * Emittance with respect to machine axis N. Blaskovic CLIC Beam Physics meeting 26
target Emittance with respect to machine axis N. Blaskovic CLIC Beam Physics meeting 27
target reached at 90% charge & 110% length Emittance with respect to machine axis N. Blaskovic CLIC Beam Physics meeting 28
Optimum bunch length & charge • Energy spread (std) reduced from 0. 35% at 380 Ge. V to <0. 20% at 350 Ge. V with: – Bunch charge: 90% of nominal – Bunch length: 110% of nominal N. Blaskovic CLIC Beam Physics meeting 29
Optimum bunch length & charge • Total luminosity falls but peak-over-total luminosity increases from 60% to 64% D. Arominski N. Blaskovic CLIC Beam Physics meeting 30
MICADO orbit correction studies with D. Schulte N. Blaskovic CLIC Beam Physics meeting 31
MICADO • Typically, one-to-one (1 -to-1) correction uses all BPMs and all correctors • However, the beam orbit only contains a few significant Fourier components • The MICADO routine selects only a few correctors to perform a global orbit correction, thus reducing the number of setting errors and sources of noise N. Blaskovic CLIC Beam Physics meeting 32
MICADO • N. Blaskovic CLIC Beam Physics meeting 33
One MICADO iteration • residual orbit at m BPMs N. Blaskovic position at m BPMs response matrix gain CLIC Beam Physics meeting settings of n correctors 34
MICADO iterations • The ~10 correctors are set and the MICADO routine is repeated for the next train, i. e. , a new subset of ~10 correctors is selected to perform the correction N. Blaskovic CLIC Beam Physics meeting 35
MICADO literature • MICADO routine: – B. Autin & Y. Marti, “Closed orbit correction of A. G. machines using a small number of magnets”, CERN-ISR-MA/73 -17, 1973 • MICADO simulations for the ILC: – A. Latina, G. Rumolo, D. Schulte & R. Tomas, “Feedback studies”, proceedings of PAC 2007, Albuquerque, New Mexico, USA, pp. 2841– 2843 N. Blaskovic CLIC Beam Physics meeting 36
Running MICADO • Run ATL ground motion for 10 hours starting from a perfectly aligned machine • Run MICADO for 100 consecutive iterations assuming no additional ground motion during this period • Run for 100 machines and average results • Figure of merit is the reduction in the emittance growth in the linac versus MICADO iteration number N. Blaskovic CLIC Beam Physics meeting 37
Plotting emittance growth in the main linac (i. e. subtracting the 10 nm emittance at the start of the linac) Each point shows the mean and error over 100 machines 2 seconds N. Blaskovic CLIC Beam Physics meeting 38
Correction gain • A gain of 1 leads to the reduction in beam emittance in the fewest iterations, at the risk of introducing too much noise from BPM and corrector errors • Assume 100 nm BPM resolution and 1 nm error in the corrector setting N. Blaskovic CLIC Beam Physics meeting 39
Gain scan Best gain: 0. 4 N. Blaskovic CLIC Beam Physics meeting 40
Corrector step size • The corrector step size can be set, such that the corrector setting is rounded the nearest corrector step • It is used to simulate the resolution with which the corrector position can be set N. Blaskovic CLIC Beam Physics meeting 41
Gain scan Best corrector step size: 1– 10 nm N. Blaskovic CLIC Beam Physics meeting 42
Number of chosen correctors • The number of correctors chosen by the MICADO routine, e. g. 5, 10, 20 or 40, can be varied to find the optimum number N. Blaskovic CLIC Beam Physics meeting 43
Gain scan ≥ 10 correctors gives a fast convergence in 30 iterations traditional 1 -to-1 MICADO beats traditional 1 -to-1 N. Blaskovic CLIC Beam Physics meeting 44
Number of chosen correctors • Starting with the same initial conditions, emittance growth in main linac: – ~0. 04 nm after 30 MICADO iterations – 0. 061 nm using traditional 1 -2 -1 correction • As a smaller number of correctors reduces the e�ect of corrector setting errors, 10 correctors are chosen for each MICADO iteration for the subsequent studies N. Blaskovic CLIC Beam Physics meeting 45
DFS • Dispersion Free Steering (DFS) consists in correcting both orbit and dispersion • The beam is steered to minimise both: – the offsets of the nominal-energy beam from the BPM centres – the di�erences of the trajectories of beams at di�erent energies • DFS overcomes systematic errors due to BPM o�sets N. Blaskovic CLIC Beam Physics meeting 46
MICADO-DFS • 2 m residuals weight N. Blaskovic m positions & m position differences response gain settings of matrix n correctors CLIC Beam Physics meeting 47
MICADO-DFS • Instead of applying the ATL motion, the BPMs have been randomly vertically o�set to demonstrate the DFS performance • 1 -to-1 steering is performed before doing either MICADO or MICADO-DFS N. Blaskovic CLIC Beam Physics meeting 48
1 -to-1 correction Gain scan MICADO cannot improve on 1 -to-1 correction in presence of BPM offsets MICADO-DFS overcomes limitation due to BPM offsets N. Blaskovic CLIC Beam Physics meeting 49
MICADO-DFS • Emittance growth reduced to under 60 nm • Traditional DFS would yield 16 nm • Investigate by using perfect conditions (perfect BPM resolution, no corrector setting error, gain of 1) • Vary number of correctors selected by MICADO N. Blaskovic CLIC Beam Physics meeting 50
Gain scan ≥ 200 correctors required N. Blaskovic CLIC Beam Physics meeting traditional DFS 51
MICADO-DFS • MICADO-DFS matches traditional DFS performance but requires at least around a third of correctors to converge N. Blaskovic CLIC Beam Physics meeting 52
Conclusions • The MICADO routine selects the best 10 correctors (from 576) for orbit correction • Starting with the same initial conditions, emittance growth in main linac: – ~0. 04 nm after 30 MICADO iterations – 0. 061 nm using traditional 1 -2 -1 correction • MICADO-DFS matches traditional DFS performance but requires at least around a third of correctors to converge N. Blaskovic CLIC Beam Physics meeting 53
Thank you for your attention! N. Blaskovic CLIC Beam Physics meeting 54
350 Ge. V vs 380 Ge. V (relative to bunch axis) N. Blaskovic CLIC Beam Physics meeting 55
- Slides: 55