ECLOUD in PS 2 PS SPS update Feb

ECLOUD in PS 2, PS+, SPS+: update Feb. 2007 Miguel Furman (presented by M. Venturini) Center for Beam Physics, LBNL ECL 2 -- CERN, 1 -2 March 2007 (Initial talk presented at CARE-HHH-APD LHC-LUMI-06, IFIC, Valencia, 16 -20 October 2006) LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 1

Summary ã Results for ecloud build-up in LHC injector upgrades (PS 2, PS+, SPS+, …) — Bunch spacings tb=12. 5, 25, 50 and 75 ns — Obtained with code POSINST — Only in a dipole bending magnet, during 1 st injected batch ã Initial results (PPT file) were presented by MF at LUMI 06 — Coarse integration time step (Dt=0. 1 -0. 3 ns) for expediency ã Present (new) results have Dt=0. 02 -0. 07 ns — These are now published in the LUMI 06 proceedings (see also http: //mafurman. lbl. gov/LBNL-61925. pdf) ã Basic conclusion: — New results much more favorable vis-à-vis heat load (roughly speaking, factor 5 -20, depending on the case) — Caveat: time step Dt=0. 02 -0. 07 ns believed to be adequate, but not methodically checked LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 2

Input parameter table (from LBNL-61925, based on FZ’s “psplusetcparameters” and “lhcupgradeparameters”) LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 3

Sample old vs. new results: heat load for PS 2 and PS+ old 5 =2 new ns ng g= in ac p 25 ing= ns pac 50 , s Ge. V 75 acing= Ge. V, sp W/m 75 V, Ge ci spa 75 ns acing=75 50 Ge. V, sp ns dmax 50 , s V Ge 75 V, e G 25 ns 5 =2 ns g in c a sp =75 ns g n i c a p , s 75 Ge. V ing=75 ns c a p s , V e 50 G dmax ã Old: MF talk at LUMI 06 — 21 kicks/bunch, or Dt=(6 -7)x 10– 10 s ã New: MF paper in LUMI 06 proceedings (LBNL-61925) — 201 -251 kicks/bunch, or Dt=(5. 5 -6. 7)x 10– 11 s LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 4

Sample new results: heat load vs. dmax (PS, Eb=50 or 75 Ge. V; SPS, Eb=75 Ge. V; tb=25, 50, 75 ns) old result PS 50 tb 25: 100 W/m at dmax=1. 7 (see slide 4) SPS and SPS+b 25 ns W/m 25 ns PS 2 and PS+ 50 ns 75 ns dmax ã Generally, tb=50 and 75 ns is much more favorable than 25 ns — But 75 ns not significantly better than 50 ns ã SPS ~x 2 higher heat load for same Eb, Nb and tb than PS probably owing to smaller sz and smaller chamber size LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 5

Sample new results: Cu vs. St. 120 -150 W/m for St. PS 2 and PS+ Cu, 1 Te. V s t, s. n 25 Cu, 450 Ge. V St s n , 25 S ns Cu, 25 SP s 50 n , t s. t S Cu, 50 ns dmax LHC ) inal m o (n dmax ã Copper much more favorable than St. — Owing to smaller rediffused component in SE energy spectrum — Subtle mechanism; explained in detail in Sec. IV-B of http: //prstab. aps. org/pdf/PRSTAB/v 9/i 3/e 034403 ã Caveat: Cu and St. St emission parameters need to be re-measured to confirm this! LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 6

Conclusions ã Heat load depends inversely with tb — tb=75 ns is best, closely followed by 50 ns — tb=50 ns much better than 25 ns — tb=12. 5 ns is terrible ã Cu (or Cu-coated) chamber much better than St. — But this conclusion is based on a particular set of measurements of the SE energy spectrum — Re-measure energy spectrum in order to verify this conclusion ã Not much difference in heat load between gaussian vs. flat long. profile for the LHC, at least for tb=50 ns ã Not much difference between PS 2 and PS+, nor between SPS 50 and SPS+a 50, except at high dmax for tb=25 ns ã This investigation is limited to heat load in the dipole bends ã I feel much more comfortable about the relative rankings than about the absolute values of the heat load LHC-ECL 2, CERN, Mar. 2007 M. Venturini, M. Furman, “ecloud in PS 2, PS+, SPS+” update Feb. 07 p. 7