1 / 24

Drive beam generation

Drive beam generation. B. Jeanneret CLIC-ACE , may 2009 W ith material and advice from E. Adli , C.Biscari , R. Corsini , D.Schulte , F. Stulle. 326 klystrons 33 MW, 139 m s. combiner rings Circumferences delay loop 72.4 m CR1 144.8 m CR2 434.3 m.

adelio
Télécharger la présentation

Drive beam generation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Drive beamgeneration B. Jeanneret CLIC-ACE , may 2009 Withmaterial and advicefrom E. Adli, C.Biscari, R. Corsini, D.Schulte, F. Stulle

  2. 326 klystrons 33 MW, 139 ms combiner rings Circumferences delay loop 72.4 m CR1 144.8 m CR2 434.3 m drive beam accelerator 2.38 GeV, 1.0 GHz CR1 CR1 1 km delay loop CR2 326 klystrons 33 MW, 139 ms drive beam accelerator 2.38 GeV, 1.0 GHz 1 km delay loop CR2 decelerator, 24 sectors of 868 m BDS 2.75 km BDS 2.75 km BC2 BC2 245m 245m IP1 e- main linac , 12 GHz, 100 MV/m, 20.85 km e+ main linac TA R=120m TA R=120m 47.9 km CLIC 3 TeV booster linac, 9 GeV, 2 GHz or 4 GHz ? BC1 e- injector 2.4 GeV e+ injector, 2.4 GeV e+ DR 365m e- DR 365m

  3. Drive beamparameters 3 TeV 500 GeV R. Corsini, ACE sep08 106 A 2.54 GeV

  4. Frequency multiplication,transfer, compression • OPTICS : DL, CR1, CR2, DBTL, TAs • BEAM DYNAMICS • GRADIENT AND BUNCH POPULATION • PHASE ERROR, PHASE CONTROL

  5. CR1 arc cell C. Biscari, BJ PAC09 Quad-triplet : Sextupoles : All thiswouldideallyrequires 3 families of sextupolesatwell Defined phase advances (x and y) In practice : • - Not enough ∆µ/per cell , ratio D/ßinadequate ,… • dp = 0.02  large higher-ordereffectswithmanysextupoles • presently, explore with 1 familyonly

  6. CR1 optics and 6D-tracking Withonly one family of sextupoles Spaced by ∆µx=3π/2 δp OK up to Jx,y= 3 • Furtherwork : • Workoff-energy to avoid large ∂p <0 • Full 6D trackingwitherrors, etc

  7. PresentTurn-Around • 80m-long vs. 146m/CR1 • Twice more dipoles (accept. CSR) • Larger D/r • stronger quads • More aberrations, … • Correcting longitudinal and transverse phase space simultaneously not possible COST : Presently DBTL + 48 TA : 3% of total LTA 2LTA : 3.7% of total • Need new TA optics • similar to CR1 with LTA ≈ 150m

  8. Delay Line • Presently : Layoutis a ring • Constraint : LDL = 73 m (beam train length) • As with TA : transverse chromaticeffects • SAME CURE : longer length • Go to a ‘Ω’ – shape Lloop = 135m Lstraight = 62 m Total extra-length : Lloop+Lstraight-∆L = 123 m (smallextra-cost ) Straight segment : Veryuseful for beam diagnostics AfterLinac FromLinac To CR1

  9. CR2 / Down to tunnel /DBLongTLOptics • CR2 • LCR2 = 438 m • Not studiedat 2ndorder • But lengthcomfortable a priori not problematic • Down to tunnel • No studied • Common tunnel with Main Beam • Main beam more critical (vertical bends vs. Smallεy, mayrequireεy⇔εx) • A priori, not critical for Drive Beam • Long Transfer line • L= 20 km, simple FODO • Long cellsavoids large chromaticdetuning and filamentation • Needslimitedbeambasedalignment • No critical issues, see CLIC Note 760

  10. Beamdynamicsfrom DL  DECEL • Presented by R. Corsiniat ACE-sep08, seealso CLIC Note 761 • N-bunchresistivewallinstabilities • Largeeffect (3000 bunches, 5 1010e-(8 nC), spaced by 2.5 cm • W⊥ ~ a-3 CURE : use vac. Chamberwith large radius (copper) • Long transfer line (L= 20 km, large β’s) : a = 100 mm • Otherlines : : a = 25 mm • Vacuum • e- ionise and confine +ions • ∆µ betweenhead and tail of trains correctability, filamentation • Beam/ion instabilities • CURE : p < 10-10 Torr . Reachablewith getter + bake-out • ISSUE : “not more than“ COST (budgetedalready) • Emmitancegrowthsmall

  11. RF deflectors in DL,CR1,CR2 • OK in CTF3 • But largercurrent in CLIC 100 A vs. 12A, whilehigherenergyhelps • Need more work in order to fix a ‘level of criticality’ vs. Collective effects • optimize ring tune for min. transverse excitation by deflector(R.Corsini, D. Schulte) Transverse position errorat the entrance of the Decelerator • ∆x/σβ =0.3 for <5% increase of amplitude at the DECEL exit (E. Adli) • Atdeflectors ( β=15m, ε=2⋅10-8 m) , σ’= 4⋅10-5 rad • Deflector kick Kd = 5⋅10-3 rad • Beamwillpassthroughndefl = 9 deflectors (DL 2, CR1 3, CR2 4) Remains to translate this Intodetailedspecification Similarly, withnkickkickers (CR1,CR2,in-TA) :

  12. Incoherent synchrotron radiation • X-ray @ 6 keV : L_absorption = 50 µm (Aluminium) • Impact angle : ~ 0.2 rad • ∆T = 80 oK @ 50 Hz • Risk of rapidageing ∆x = 10 µm TIME STRUCTURE • OveralldE/E of ISR varies between • Trains travelling : • MIN : No DL, 1×CR1, 1×CR2 • MAX : 1×DL, 3×CR1, 4×CR2 • MaxdE/E = 2 o/oo • Converts to dz after last chicane before DECEL • 1% longitudinal z-pitch modulation R56 = -0.33 m • Cure : • make the DL, CR1,CR2 longer accordingly • complicates the feed-forwardφ–correction • (seeslide 20) 24 bunches x 121

  13. Tolerance on Phase and Gradient • Luminosity and understanding/adjustment of the BeamDelivery and of the Final Focus D.Schulte & R.Tomas, PAC09  Not muchmargin on these

  14. Gradient error for MB A bunch of the main beamseealways the gradient as built-up by the same nb = 0.5f0⋅τfill-MBAS≅ 400 DB bunches ahead of it MB DB With N the bunch population , And ‘inc’ as for ‘per sector’ At the entrance of the DECEL Stability of the DB in the DECEL : An overcurrent of 1% over τfill-PETS≡ 10 b. makesthe last bunch of theseunstable Before the end of the DECEL Stability over a whole cycle :

  15. Tolerance on DB bunch population • Losses : constant part isharmless (adjustNsource ) • Erratic part : • transverse n-poles in RF structures, RF deflectors, magnets • Badlyunderstoodoptics (chromaticerrors) • Time dependenterror in kickers • Not explored as yet. • Untilwe know more about loss control : Maybestilloptimistic Need a more elaborate Statistical model Togetherwith :

  16. Timing and feed-back strategy in the tunnel • To relax on δφ : feed-forward& back beforeentering the DECEL Notional, as for now 9 GeV MB as a fastclock Timing line A B DB ∆t to: injectors DB & MB Damp. Rings MB TA DB TAs @50 Hz TA C E D MB to IP Phase from B, EnergyjitterδE from A&B, Phase correction for δE C&D, Phase error MB/DB , to synchronize theinjectorlinacs Open issue: frequency band of δE correction

  17. 326 klystrons 33 MW, 139 ms combiner rings Circumferences delay loop 72.4 m CR1 144.8 m CR2 434.3 m drive beam accelerator 2.38 GeV, 1.0 GHz CR1 CR1 1 km delay loop CR2 326 klystrons 33 MW, 139 ms drive beam accelerator 2.38 GeV, 1.0 GHz 1 km delay loop CR2 decelerator, 24 sectors of 868 m BDS 2.75 km BDS 2.75 km BC2 BC2 245m 245m IP1 e- main linac , 12 GHz, 100 MV/m, 20.85 km e+ main linac TA R=120m TA R=120m 47.9 km CLIC 3 TeV booster linac, 9 GeV, 2 GHz or 4 GHz ? BC1 e- injector 2.4 GeV e+ injector, 2.4 GeV e+ DR 365m e- DR 365m

  18. Phase control - I At 12 GHz • measurement of the 9 GeV MB ( keep in memory for T = [0..23] × 6 μs] ) • error in the 9 GeV MB fromInjector and in the MB-TA • DB TA length must bevery stable ⇔ magnetstability to bespecified • measurement of the DB • error of the feed-forwardadjustement of the chicane • transmission error in the timing system • bandwidthlimit for correction • periodicdE/E structure complicating the measurement (slide 13) • Path-lengths variations to beworked-outeverywhere Scheme : correct the phase withfastfeed-forward in the final chicane and slow feed-back to the injectors. Ingredients : J.Sladenproposed a Solution,down to 20 fs cost

  19. Phase control – II : data L0 Chicane after TA α • Tolerance on α : • 1/10 δztol • 4 magnets L0 = 8mα= 0.1 rad Use staticmagnets + variable deflectors for range δφr=10o Erk willsay ‘red’ or ‘black’ Range insidewhich correction applies Withfeed-forward : relax for linac (Need room for feed-forwardoverallerrorkeep 0.2o for coh & inc in DB Linac)

  20. Decelerator : conclusions (from E. Adli, last ACE) • Simulations gives reasonable confidence for minimum-loss transport of the decelerator beam • Beam-Based Alignment is needed, and Dispersion-Free Steering seems to be an excellent alternative • Dispersion-Free Steering comes almost "for free" with the use of delayed switching • Tune-up procedures must be applied • Simulations need to be benchmarked and technology needs to be proven: TBTS and TBL PETS Internaltolerances, perf,… : not discussedhere r0 = 3.5 mm : nominal max beam enveloppe Riris = 11.5 mm Dynamic tolerances Static tolerances

  21. Critical issues, injector end TA

  22. Summary and close future (CDR) • Most critical issues • Phase control dφ = 0.2o ↔ chicane / feed-forward / timing strategy • Bunchcurrent↔ source and loss control • RF deflectors ↔ design, tolerance,beaminstabilities • ISR absorption ↔ thermal stress, materials • For CDR (proposal) • Full opticsstudy of CR1 ↔ 6D chrom, Lengthadjustment (If OK : DL, CR2 , TA granted) • Coordinated effort for dφ by RF/CO/BI/ABP • Betterstatistical model for Ibunch, sources, losses

  23. Reserve slides

  24. Chicanes and compression at the end of the Turn-around δE δE F. Stulle σδE=3e-3 z z σuδE= 2.5e-4 σ1,z=2mm σ2,z=1mm Presently : R56 = (σ2,z-σ1,z)/σδE = - 0.33 m • This value is large and convertsenergyjittersinto large phase errors • Make use of the larger possible momentumwindow of ∆E = ±2% • σδE=3e-3σδE=6e-3 ⇒ R56 = -0.16 m More comfortable, but a new studyisneeded (both chicane and chirp in DB Linac)

More Related