310 likes | 528 Views
Space charge in LEIR?. Michael Bodendorfer February 27 h , 2014. With the help of Django Manglunki , Maria-Elena Angoletta , Alan Findlay, Christian Carli , Sergio Pasinelli , Gerard Tranquille , Jerome Axensalva and Maxim Andersen And many, many thanks to:
E N D
Space charge in LEIR? Michael Bodendorfer February 27h, 2014 With the help of DjangoManglunki, Maria-Elena Angoletta, Alan Findlay, Christian Carli, Sergio Pasinelli, Gerard Tranquille, Jerome Axensalva and Maxim Andersen And many, many thanks to: Elena Benedetto and Vincenzo Forte
Overview LINAC3 & LEIR
Pre-LHC era: • Ion chain with new Linac3 for SPS fixed target ion operation (via Booster) • Not suitable for LHC: large emittances and low brightness (brightness limited already at ECR ion source) MotiviationLEIR came from LEAR • Proposals to increase brightness for LHC: • Laser-ion-source to increase brightness (Studied in the 1990s) • Accumulation in synchrotron with electron cooling • Electron-Cooling fast at low energy and for “heavy” (high charge state) ions • Studied experimentally between 1994 and 1997 with LEAR • Chosen as viable solution, which can be ready on time for LHC (expected around 2006) • Construction of LEIR, re-using most of the LEAR hardware • LEIR transforms several long low density pulses from Linac3 into dense short bunches,useful for LHC in 2006
LEIR – Low Energy Ion Ring From LINAC3 & to PS Extr. Septum Injection RF Extr. Kicker Beam Ecooler
Overview LEIR
Quadrupole triplet Extraction Bρ=4.8Tm 72.2MeV D=0 Injection Bρ=1.14Tm 4.2MeV RF Symmetry Optics Ejection Kicker 2 D=10m D=10m Ejection Kicker 1 Quadrupole doublet E-Cooling D=0
LEIR performance, 2013 and LIU IonsWhy upgrade? ALICE wants by 2035: 10nb-1 From: Performance of the injectors with ions after LS1 D. Manglunki for the LIU-Ions team, CERN, Geneva, Switzerland, 2013
Continuous electron cooling RF capture @ 1780ms (Coasting beam) Up to 50% loss A LEIR NOMINAL cycle (Qh=1.82; Qv=2.72) Extraction @ 2880ms (Master timer) Magnetic ramp start @ 1823ms 7 injections. First at 215ms, then spaced 200ms. 200μs long. B-field Intensity in 1010 charges vs. cycle time: 0 to 3.6s
Electron beam shifts the ion beam in energy space Uc = Cathode potential Ue = Space charge potential of electron beam Ui = Space charge potential of ion beam Cold electrons Hot electrons Uc Uground Ue Ie >> Ii Toroidal magnet Less hot ions Hot ions Electron cooling Δveli Ui Δveli vele Ekin_e=qe-(Uc+Ue+Ui) RF adjust Frevcorr necessary Cooling Cooling rate:
Higher accumulated intensity (before RF-capture) • beam is lost at RF-capture • Adjusting Frevcorr -> RF-capture successful Waterfall diagram of Tomoscope analysis: Trev. Tomoscope @ RF capture before Tomoscope @ RF capture after Measurements from Nov. 29th, 2012
Oracle output shows improvement: original Frevcorr improved Frevcorr improved original Jan 21st, 2013
Qx - Qy= -1 2Qx - Qy= 1 4Qy = 11 Qx + 3Qy = 10 3Qy = 8 2Qx + 2Qy = 9 TWISS parameters (MADX) Tune measurement:
LEIR multi-turn injection (proposed by D.Möhl and S.Maury) CO remains fixed Closed orbit of injected beam 1stturn of incoming beam After 3 turns End of injection (71 turns) After collapse of the bump Range: Δp/p=4x10-3 Low momentum High momentum x Inj. beam Septum Bump+DxΔp/p DxΔp/p 200μs (71turns) Bump D=10m Mechanism: • Bumper moves orbit inwards • Momentum ramping moves orbit outwards • Betatron amplitude for incoming beam remains constant time Towards machine centre() Stack “parked” with negative momentum offset Pictures from: Low Energy Ion Ring LEIR, C. Carli, 14th March 2012
Longitudinal Schottky spectrum 21. Nov. 2012 4.5*1010 charges 1st inj. 2st inj. p-ramp: 4x10-3 3st inj. 4st inj. Dense beam 5st inj. 6st inj. Center: 35MHz Span: 330kHz 7st inj. Time 3. Feb. 2013 8*1010 charges Less dense beam Center: 36MHz Span: 313kHz Momentum
Typical LEIR NOMINAL cycle (sampler output, Nov. 21st, 2013) 1st inj. 2st inj. 3st inj. 4st inj. 4.5*1010 charges Dense beam 5st inj. 6st inj. 7st inj. Insufficient cooling and/or space charge?
Where/when is there space charge in LEIR? 1st inj. 2st inj. 3st inj. 4st inj. Coasting beam Dense beam 5st inj. At RF-capture? 6st inj. Can S/C cravate detach from design tune? 7st inj. During cooling? During multi-turn inj.?
Findings so far: • Accurately modelling LEIR is a challenge (MADX-PTC: 90° BHN bending magnets, design specifications inconsistent with actuallity,…) • Identified three potential origins for space charge, maybe more? • Simulation activities (actual and planed): • MADX-PTC optical model vs. machine parameters by LOCO-analysis • PTC-Orbit fit for ions? (may need source code modification) • PTC-Orbit flat-bottom, multi-turn injection, RF-capture and magnetic ramp • Wanted: • Space charge during/after multi-turn injection? • Space charge at RF-capture? • Electron cooling simulation (effect) • Potential reward: • Compare LEIR with existing simulations (PSB,…) • Understanding better PTC-Orbit (Performance, stability, accuracy, convergence behavior) BIG THANKS to S/C WG ! ! ! • Off to the movies!
S. Hancock, 21.2.2013 εlong = 8.1eVs Pb54+/bunch = 5.0E8 εlong = 10.8eVs Pb54+/bunch = 5.1E8
Electron cooler loss rates for different Pb charge states Loss-rate coefficients measured for lead ions of different charge states and different machine settings. Experimental Investigation of Electron Cooling and Stacking of Lead Ions in a Low Energy Accumulation Ring J. Bosser, C. Carli, M. Chanel, CERN, CH{1211 Geneva 23, Switzerland April 27, 1999
LEIR horizontal tune NOMINAL (6 injections) Restricted maxima search Cycle time [ms] Fractional tune
ΔQ for programmed radial offset from-20mm to +20mm 1 time step: 20ms Cycle time [ms] 10mm -5mm 20mm -15mm 5mm 15mm -10mm -20mm Difference of Horizontal tune to reference tune (@ 0mm beam offset)
Measurement from +10mm programmed radial offset Linear extrapolation of Δp/p Linear extrapolation: -20mm to +20mm radial offset Cycle time [ms] Measurement from -10mm programmed radial offset Δp/p