270 likes | 412 Views
LIU evolution. H. Damerau, R. Garoby, B. Goddard, S. Gilardoni, K. Hanke , D. Manglunki, M. Meddahi , B. Mikulec, L. Ponce, E. Shaposhnikova , M . Vretenar Acknowledge work from ABP, ABT, RF, OP, BI, STI, EPC , MSC, VSC , …. and US-LARP collaborators. Contents.
E N D
LIU evolution • H. Damerau, R. Garoby, B. Goddard, S. Gilardoni, K. Hanke, • D. Manglunki, M. Meddahi, B. Mikulec, L. Ponce, E. Shaposhnikova, • M. Vretenar • Acknowledge work from ABP, ABT, RF, OP, BI, STI, EPC, MSC, VSC, …. • and US-LARP collaborators
Contents • Post LS2 HL-LHC requirements • Corresponding LIU beam requirements • 2012 proton injector chain performance • Limitations and mitigations • Unknowns, risks and remaining issues • Post LS2 expected performance • Conclusion
Corresponding required beam parameters through injector chain Assumptions: B. Goddard, Chamonix 12
2012 Performance and limitations at SPS ejection Large improvement is required for either 25 or 50 ns beam! 6
Limitations: 25 ns space charge/brightness • PSB: exy [mm] -0.42 Nb [e12] / DQy, with 160 MeV injection • Confident to run with DQy of at least -0.4 • Need 3.4e12 in 1.9 mm: potential for 3.4e12 in 1.7 mm (OK) • PS: exy [mm] -0.26 Nb [e12] / DQy , with 2 GeV injection • Confident to run with DQy of at least -0.26 (operational beam reached -0.28) • Need 3.2e12 in 2.0 mm: potential for 3.2e12 in 3.0 mm (Not OK) • Cannot digest fully what PSB can provide • Need to increase DQy to -0.37 to reach the required brightness • SPS: exy [mm] -1.22 Nb [e12] / DQy, with Q20 optics at 26 GeV • Reasonably confident to run with DQy -0.15 (single bunch brightness) • Need 2.5e11 in 2.1 mm: potential for 2.5e11 in 2.0 mm (OK) For 25 ns: suggests limit will be PS, as PSB delivers more than PS can digest
Limitations: 50 ns space charge/brightness • PSB: exy [mm] -0.42 Nb [e12] / DQy, with 160 MeV injection • Confident to run with DQy of at least -0.4 • Need 2.7e12 in 2.2 mm: potential for 3.4e12 in 1.7 mm (easy!) • PS: exy [mm] -0.26 Nb [e12] / DQy , with 2 GeV injection • Confident to run with DQy of at least -0.26 (operational beam reached -0.28) • Need 2.6e12 in 2.4 mm: potential for 2.6e12 in 2.4 mm (Just OK) • SPS: exy [mm] -1.22 Nb [e12] / DQy, with Q20 optics at 26 GeV • Reasonably confident to run with DQy -0.15 (single bunch brightness) • Need 4.0e11 in 2.5 mm: potential for 4.0e11 in 3.3 mm (Not OK) • Need to increase DQy to -0.2 to reach the required brightness For 50 ns: PSB and PS OK, limit in SPS
Other limitations and actions • SPS TMCI : single bunch • Q20 now used: above ~3.6e11 per bunch (Q’=0) • PS longitudinal phase plane: • Should increase limits from present ~1.7e11 to ~3e11 p+/b while keeping the beam splitting quality • Combine effects of different actions: • New longitudinal damper using dedicated broad band cavity • Fast RF feedback systems upgraded on all High Power RF systems • One turn-delay feedbacks renovated • More RF voltage at 40 MHz installed for better SPS capture efficiency • PS-SPS transfer parameters • Larger longitudinal emittance required for beam stability both in the PS and SPS, but its leads to higher capture losses in the SPS. • Simulations predict a significant gain by optimising the PS bunch rotation with spare RF cavities. First MD results are very encouraging – Helga Timkó et al.
SPS longitudinal instabilitiesP. Baudrenghien, W. Höfle, E. Shaposhnikova, US LARP collaborators • Longitudinal stability: 25 ns beam unstable at 2-3e10 p+/b , Q26 • Presently mitigated with controlled long. emittance blow up (0.6 eVs) and 800 MHz RF voltage used in bunch shortening mode • Q20 has increased the threshold on flat top by factor 1.6 • Both 25 and 50 ns beams, affected • After upgrade (double 200 MHz RF power, 6 shorter assemblies – vs. 4 today -> reduced impedance) : • At reach: 2.3e11 p+/b for 25 ns, and >3.4e11 p+/b for 50 ns • Planned upgrades of the 800 MHz low and high power equipment are mandatory. US-LARP collaboration has been, and still is, essential. • Higher power will only remove limitation from the beam loading, but not from longitudinal instabilities with high intensity since cannot have much larger longitudinal emittance at injection to the LHC -> should find the source of this instability
ecloud • PS: observed but not yet a limitation • Transverse damper proved to be effective to damp instability observed • Studies are ongoing to investigate HL-LHC regime • Mitigations being investigated (coating, double bunch rotation, …) • SPS: potentially a major performance limit (beamloss, vacuum, ecloud instability and incoherent emittance growth) • Scrubbing for 25 ns could be tough…StSt chambers, OP limitations… • But in 2012 machine in good shape – no limit from ecloud. Crucial to have 2 weeks to investigate scrubbing after LS1 (to decide between scrubbing and coating) • aC coating now fully validated as mitigation (still LIU baseline) • HBW feedback could cure vertical single bunch ECI –much work on-going – indispensable US-LARP collaboration (with W. Höfle and team)
‘Operational’ limitations • Mainly in SPS (to date) • Heating of extraction kickers: should be ‘solved’ with final shielded MKE in LS1. Expect limit to be at least twice present beam power -ABP, ABT groups • Outgassing of dump and impact on injection kickers MKP vacuum: extra differential pumping and sectorisation planned. Effect is mainly a limitation for scrubbing and setting up, rather than LHC filling –ABP, ABT, VAC groups • ZS sparking was previously a concern, but 2012 performance has been very good. Still working on ‘ppm’ voltage modulation as mitigation – EPC, ABT, CO groups • New beam intercepting and protection devices needed (major simulation & design efforts from STI) • BI development for measuring with adequate accuracy reduced beam sizes, higher brightness and intensity
Essential MDs LIU Beam studies Review – organised by Giovanni Rumolo - 28 August 2012 Giovanni Rumolo - LMC 17 Oct. 12
Expected performance after LS2 LIU preferred option is 25 ns
Unresolved questions/issues • Electron cloud • aC coating or scrubbing (+HBW damper) in SPS? • Is ecloud going to be an issue in PS? • Prospects for increasing (already ambitious) target for PS longitudinal stability for 50 ns beam? • Prospect for PS significantly exceeding -0.26 in space charge tune shift for 25 ns beam? • Study to assess maximum tolerable Laslett tune shift with no resonance compensation. • One quarter resonance excited by Space charge under study • Alternative schemes : longer bunches, resonance compensation, acceleration/deceleration, dedicated injection optics ….
LIU Project Timeline HL-LHC-LARP 2012 meeting @ Frascati PSB-PS transfer 1.4 GeV 2 GeV Baseline start of installation in the PSB is beginning of LS2 (~January 2018) LS2 for injectors LS1 for injectors 2019 2012 2013 2014 2015 2016 2017 2018 01/2016 Stripping foil and instrumentation test stand during L4 beam reliability run SPS aC coating, 200 MHz power upgrade completed 12/2016 all injection hardware components should be available, “ready for PSB installation”, if needed 07/2016 Start complete H- injection region “montage à blanc” assembly . Injectors commissioned W.Weterings – 11/10/2012
Conclusions (I) • For HL-LHC era, requirements are challenging • 25 ns: need x2 present intensity, >x2 present brightness • 50 ns: need x2.2 present intensity, 30% above present brightness • Baseline LIU does not reach HL-LHC ‘point-like’ requirements. • To even get close: • Need all planned upgrades to be fully effective, and to approach single bunch limits with multi-bunch operation • Need “stretch” loss/blowup levels in injectors, with HL-LHC ≤10% blowup, and losses of around 3%... • Limits are different for 25 and 50 ns production • PSB performance sufficient with 160 MeV injection for both • 25 ns: PS space charge tune shift (2 GeV injection fixed by PSB). • 50 ns: brightness in SPS, PS longitudinal stability, PS beam loading
Conclusions (II) • Not totally impossible to approach HL-LHC request, providing • Losses and blowup can eventually be reduced to “stretch” levels • PS can run with higher space charge tune shift, nearer DQ≈ -0.32 • PS longitudinal stability can improve significantly above present expectation (today at 1.7e11, LIU baseline 3e11, need 3.7e11) • SPS can increase brightness limit, to DQ≈ -0.17 • Realistic time for re-commissioning after LS2 to be foreseen • LS2 will be ‘LIU shutdown’… • Planning and organisation of this should start now! • Support from US-LARP is acknowledged, much needed , and should continue (space charge studies in PS, damper systems in SPS, Beam dynamics through injector chains, SPS LLRF upgrades…)
Q20 optics: experience from deployment • Q20 deployed in SPS and TLs, now working operationally – H. Bartosik , Y. Pappaphilipou , BTP, RF, OP et al. • More stability in SPS and slightly lower exy to LHC as result (0.1-0.2 um) • In the SPS: more orbit fluctuation than with Q26 at the ejection point creating larger oscillations in the TL – being analysed – Eliana Gianfelice et al. • Satellites as well as the 6 b have a bigger profile and more tails – Karel Cornelis – Wolfgang Bartmann Lene Drosdal satellite bunches
Trajectory stability from IQCBPM Variations TI8 Q26 Slightly worse? Q20 Lene Drosdal
Re-commissioning after LS2 • Will not be “plug n play” to recover pre-LS2 performance • Three ‘new’ injectors to startup, together • PSB: new ramp to 2 GeV, (maybe) new H- injection at 160 MeV, new RF, new instrumentation, new beam transfer • PS: new 2 GeV injection, new longitudinal FB, new instrumentation • SPS: (probably) rebuilt after ecloud coating (744 MBs, 216 MQs), ‘new’ RF 200 MHz, new high bandwidth FB, new orbit and BLM system, new scraper and TL collimators • Normal startup: 1-2 weeks per machine • Must count months per machine to recover pre-LS2 performance • A concrete LS2 plus commissioning plan to work on (in 2012) • Will certainly take period LS2-LS3 to reach baseline LIU goals
SPS beam loading • SPS 200 MHz: x2 power, 46 (shorter) cavities, -20% impedance • Will allow 10 MV at extraction for 3 A RF current (now 1.5 A) • Need to operate existing power plants in pulsed mode (0.751.05 MW) • After upgrade: same voltage available as now (if pulsed) for 2.3e11 p+/b (25 ns) and 4.6e11 p+/b (50 ns). E.Shaposhnikova