Contributed talk (15 + 5 min, 30 slides) ACHIEVABLE SPACE-CHARGE TUNE SHIFT WITH LONG LIFETIME IN THE CERN PS & SPS Elias Métral Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 1/30
2 π SC ISSUES IN THE PS (Q x,y ª 6.2, ξ x,y ª -1) LHC beam ~ Round Small vs. mechanical aperture ~ ½ million turns at low energy Linear coupling to damp a horizontal head-tail instability CNGS beam ~ Flat (ratio of ~2) Fills the mechanical aperture INTRODUCTION Largest DQ sc,y ª -0.25 Trajectory + Montague resonance Horizontal emittance should be kept small (ª 4 times the vertical emittance in SPS: 5-turn CT extraction + emittance exchange in TT10) Vertical aperture in the SPS SC ISSUES IN THE SPS (Q x,y ª 26., ξ x,y ª 0.1-0.5) Generally not a problem (alone) for proton beams î Interplay with other mechanisms (ecloud etc.)? Could be a problem for ions (foreseen DQ sc,y ª -0.1 for ~ 40 s) Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 2/30
CONTENTS INTRODUCTION ON LHC AND CNGS BEAMS IN PS & SPS PS Crossing the integer or half-integer resonance Montague resonance Space charge driven resonance phenomena Decoherence without and with space charge at PS injection Why Q h is Ø to ~ 6.1 for high-intensity (e.g. CNGS) bunches? RW instability with LHC beam without and with space charge SPS Beam lifetimes studies with protons and large DQ sc,y (ª -0.2) Working point studies with a pencil proton bunch Results from early ion (Pb 82+ ) commissioning in 2007 CONCLUSION Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 3/30
INTRODUCTION (1/7) Duoplasmatron = Source î 90 kev (kinetic energy) LINAC2 = Linear accelerator î 50 MeV PSBooster = Proton Synchrotron Booster î 1.4 GeV PS = Proton Synchrotron î 25 GeV SPS = Super Proton Synchrotron î 450 GeV LHC = Large Hadron Collider î 7 TeV SPS operates above transition (~ 21 GeV) for LHCtype beams LHC beam Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 4/30
INTRODUCTION (2/7) Duoplasmatron = Source î 90 kev (kinetic energy) LINAC2 = Linear accelerator î 50 MeV PSBooster = Proton Synchrotron Booster î 1.4 GeV PS = Proton Synchrotron î 13 GeV SPS = Super Proton Synchrotron î 400 GeV Carbon target î 400 GeV SPS operates below transition (~ 21 GeV) for CNGS beam CNGS beam Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 5/30
9.2 10 12 ppp @ 26 GeV/c (01/09/04) LHC beam in the PS INTRODUCTION (3/7) 10 30 INTENSITY [10 12 ppp] 8 6 4 2 4 b + 2 b 25 20 15 10 5 p [GeV/c] ε required x, y ( 1σ, norm) 3.5 μm @ 450 GeV/c 0 150 650 1150 1650 2150 TIME IN THE CYCLE 4000 [ms] 0 LHC beam in the SPS 500 3500 450 ~ 6 10 5 turns at low energy ~ 5 10 5 turns at low energy INTENSITY [10 10 ppp] 3000 2500 2000 1500 1000 ~ 3.3 10 13 p at 450 GeV/c (i.e. 4 72 = 288 bunches with ~ 1.15 10 11 p/b) 400 350 300 250 200 150 100 MOMENTUM [GeV/c] 500 50 0 0 Elias Métral, HB2008 workshop, Nashville, Tennessee, 0 USA, August 500025-29, 2008 10000 15000 20000 6/30 25000 TIME IN THE CYCLE [ms]
3.42 10 13 ppp @ 14 GeV/c (Monday 27/09/04 12:47) CNGS beam in the PS INTRODUCTION (4/7) INTENSITY [10 13 ppp] 4 3.5 3 2.5 2 1.5 1 0.5 0 150 250 350 450 550 650 750 TIME IN THE CYCLE [ms] 7 6 16 14 12 10 8 6 4 2 0 p [GeV/c] 5.26 10 13 ppp @ 400 GeV/c Saturday 02/10/2004-09:30 ( 1σ, norm) 12 μm @ 400 GeV/c required ε x, y I required = 4.4 10 13 p/p INTENSITY ENERGY 500 400 INTENSITY [10 13 ppp] 5 4 3 2 1 300 200 100 ENERGY [GeV] 0 0 Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 7/30 0 1000 2000 3000 4000 5000 6000 TIME IN THE CYCLE [ms] CNGS beam in the SPS
LHC beam (in the SPS) INTRODUCTION (5/7) CNGS beam (in the SPS) V 200MHz [10 4 V] BCT [5â10 10 p] BCT [10 11 p] V 200MHz [2 10 4 V] V 800MHz [10 3 V] Beam momentum [GeV/c] Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 8/30
INTRODUCTION (6/7) Over the years the intensity / bunch in the PS for the LHC increased for several reasons Initial scenario = debunching/rebunching (84 b of 1.15E11 p/b on h = 84 at top energy) î 8 b on h = 8 at injection Then, triple and double splittings instead + gap for kicker (72 b of 1.15E11 p/b on h = 84 at top energy) î 6 b on h = 7 at injection Then, compensation for the losses in the SPS (72 b of 1.3E11 p/b on h = 84) î 6 b on h = 7 at injection with more intensity / bunch î Losses of few % on the injection plateau are now observed (without emittance growth) Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 9/30
INTRODUCTION (7/7) Reminder in 2000 (with 1.15E11 p/b at extraction instead of 1.3E11 now) Intensity [10 10 ppp] 3 b + 3 b Time [ms] R. Cappi et al., HEACC2001 DQ sc,y ª -0.21 î ~ No losses on the injection flat-bottom Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 10/30
Crossing the integer or half-integer resonance in the PS (1/2) Qy Q vy 6.2 6.1 5.9 6.1 6.2 5.9 Q x Q y Case 1 6.16 6.24 Q Qx hx M. Giovannozzi et al., PAC2003 Horizontal bunch profile + Gaussian fit 800 600 Qy Q vy 6.2 6.1 Q x 400 Case 3 200 6.11 Qy 6.24-40 -20 20 40 [mm] 5.9 6.1 6.2 5.9 Q Qx hx Regime of loss-free core-emittance blow-up Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 11/30
Crossing the integer or half-integer resonance in the PS (2/2) S. Cousineau et al., EPAC2004 Measurements ORBIT simulations Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 12/30
Montague resonance in the PS (1/2) E. Métral et al., HB2004 STATIC CASE (constant tunes from injection to the measurement point) 40 35 Q y = 6.21 Asymmetrical stop-band predicted by simulations 30 25 20 15 Emit_H (norm, 2 σ) [μm] Emit_V (norm, 2 σ) [μm] Emit_H from 3D simul. Emit_V from 3D simul. 10 5 6.15 6.17 6.19 6.21 6.23 6.25 Horizontal tune Fully 3D PIC code IMPACT ΔQ = 0.054 Δ = 0. 109 inc, x0 Q inc, y0 Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 13/30
Montague resonance in the PS (2/2) DYNAMIC CASE (the horizontal tune was changed linearly from 6.15 to 6.25 in 100 ms) Trev = 2.3 μs î ~ 44 000 turns ( σ ) ε 2 x,norm ( σ ) ε y 2,norm Mixing due to longitudinal motion î See also I. Hofmann et al., EPAC04 Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 14/30
Space charge driven resonance phenomena in the PS (1/6) Mechanism anticipated by G. Franchetti & I. Hofmann, which involves Space charge tune spread Nonlinear (octupole) resonance Synchrotron motion QQy y v Regime of loss-free core-emittance blow-up Q Qy y v Regime Regime where where continuous loss loss occurs occurs î Due Due to to longitudinal motion motion 4 Q x = 25 Q Q h Qxx Particles diffuse into a halo Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 15/30 Q Q h Qxx
Space charge driven resonance phenomena in the PS (2/6) G. Franchetti et al., ICFA workshop, Bensheim, Germany, October 2004 6.265 I oct = 20 A 6.25 Q x Loss dominated regime Emittance growth dominated regime Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 16/30
Space charge driven resonance phenomena in the PS (3/6) Q x = 6.27 180 ms Regime where continuous loss occurs î Due to longitudinal motion N b 10 [ 10 ] Octupole ON (40 A) 50% of losses Undershoot due to the measurement device Time [ms] 170 300 1200 1200 ms Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 17/30
Space charge driven resonance phenomena in the PS (4/6) By lowering the working point towards the resonance 4 Q x = 25, a gradual transition from a regime of loss-free core emittance blowup to a regime dominated by continuous beam loss has been observed, as expected by Ingo&Giuliano Emittance growth in good agreement with predictions The observed maximum losses (~30%) were still much larger than predicted (~8%) at COULOMB05 Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 18/30
Space charge driven resonance phenomena in the PS (5/6) Results from Ingo&Giuliano at HB2006 Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 19/30
Space charge driven resonance phenomena in the PS (6/6) This mechanism is sometimes @E10 D observed in the PS with the LHC beam LHC when the tunes are not correctly set 150 @kturns D 125 100 200 150 100 50 75 50 25 @ns D 180 160 800 900 1000 1100 1200 LHC @ms D 0 140 0 500 1000 1500 @ns D 2000 50000 100000 150000 200000 @turns D Courtesy S. Hancock Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 20/30
Decoherence without and with space charge at PS injection (1/2) Measurements from F. Blas with a nominal LHC bunch MD 29/10/2003 1.4 GeV Flat 0.8 0.6 0.4 Delta h [25mm/V] 0.2 0-0.001 0 0.001 0.002 0.003 0.004-0.2-0.4-0.6-0.8 2 ms TFB OFF TFB ON -1 time [s] Start = C320 = Kick time Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 21/30
Decoherence without and with space charge at PS injection (2/2) HEADTAIL simulations from E. Benedetto (PHD thesis) 0.015 0.01 0.005 0-0.005-0.01 Bunch centroid motion [m] 20 40 60 80 100 Time [turns] Comparison between HEADTAIL and theory (with chromaticity only) 7μ10-6 6μ10-6 5μ10-6 4μ10-6 3μ10-6 2μ10-6 1μ10-6 Bunch rms emittance[m] 20 40 60 80 100 Time [turns] Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 22/30
Why Qh is Ø to ~ 6.1 for high-intensity (CNGS) bunches in the PS? (1/2) Does not go in the good direction for the classical space-charge effect (as it pushes the bunch closer to the integer resonance ) Could have been explained by the Montague resonance (this is why it was studied ) The other mechanism is simply the change of trajectory Some similarities with observations at SNS reported by John Galambos (plenary talk) Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 23/30
Why Qh is Ø to ~ 6.1 for high-intensity (CNGS) bunches in the PS? (2/2) Changing the trajectory in the injection transfer line + injection process parameters, it is possible to keep a high tune (~ 6.20) with the same amount of losses! 10 10 The injection is at C170 C Time [ms] Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 24/30
RW instability with LHC beam in the PS without and with space charge î Measurements (and theory) seem to be confirmed by HEADTAIL simulations without including space charge (see PAC07 paper) No beam losses with linear coupling Q x Q y ξ x ξ y = 6.22 = 6.25 = 0.5 = 1.0 Beam losses measured without linear coupling m = 6 Time (20 ns/div) î Next step (challenge): Simulate the real case with space charge! Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 25/30
Beam lifetimes studies with protons and large DQ sc,y (ª - 0.2) in the SPS H. Burkhardt et al. (CERN-AB-2003-013 ABP and CERN-AB- 2004-056) Done on a ~ nominal (1.2E11 p/b) LHC-type bunch at 14 GeV/c (instead of 26 GeV/c) DQ sc,y ª - 0.30 with initial parameters 2003 î DQ sc,y ª - 0.17 (rather good lifetimes over 100 s) 2004 (see picture) î Working point at that time: Q x,y = 26.184 / 26.13 (it was observed that better conditions were reached by Q y to ~ 26.20) DQ sc,y ª - 0.24 with meas. parameters in the SPS after 1 s î Conclusion: Should be OK with ions Courtesy H. Burkhardt Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 26/30
Working point studies with a pencil proton bunch in the SPS F. Roncarolo et al. (AB-Note-2006-008 ABP (MD)) Done on the LHC pilot bunch (5E9 p/b) at 26 GeV/c Stop-band of the integer resonance: 0.015 in H 0.020 in V Identification of possible stable operating region: 26.11 < Q x < 26.14 26.16 < Q y < 26.18 Courtesy F. Roncarolo Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 27/30
M om.@ G e V êcd an d B C T@ 1 0 8 c harge s D 400 300 200 100 0 Results from early ion commissioning in the SPS in 2007 D. Manglunki et al., EPAC08 0 2000 4000 6000 8000 10000 12000 14000 Time @msd TT60.BTV.610317 4 bunches of 910 7 ions (Pb 82+ ) - Intensity: - Design: 295.2 10 8 charges - Achieved: 10% smaller than design - Transverse emittances (rms, norm): - Design: 1.2 μm - Achieved: 25% smaller than design (the smaller the better!) (- Tunes optimization: 26.13 / 26.25) ~ 7 s long injection plateau instead of ~ 40 s in the nominal scheme Beam seen at the beginning of one of the extraction lines towards LHC (TT60/TI2) Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 28/30
CONCLUSION (1/2) PS Few % of beam losses are observed during the long (1.2 s = 0.6 million turns) injection flat bottom in the PS with the LHC beam for LHC (DQ sc,y ª -0.25) Almost no beam losses in the previous years when the PS did not have to compensate for the SPS losses (DQ sc,y ª -0.21) Chromaticities are high (not corrected): Good for the head-tail instability (slower rise-times) Not so good for the beam lifetime Next challenge: Simulate the PS low energy resistive-wall instability with both linear coupling (used to stabilize the beam) and space charge over 0.6 million turns! Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 29/30
CONCLUSION (2/2) SPS Detailed studies at low energy with Pb 82+ ions revealed that although the space-charge detuning was as high as ~ - 0.1, ~ no transverse blow up was observed over periods of the order of one minute, confirming the expectations based on studies with protons Elias Métral, HB2008 workshop, Nashville, Tennessee, USA, August 25-29, 2008 30/30