Docstoc

ILC_rf_nagaitsev

Document Sample
ILC_rf_nagaitsev Powered By Docstoc
					@ Fermilab




ILC bunch-compressor and linac rf requirements



              Sergei Nagaitsev
                  Fermilab
                Feb. 9, 2006
@ Fermilab   The Baseline Machine (500GeV)




                               Nick Walker – GDE meeting Frascati 8.12.05



               ILC@FNAL - Nagaitsev                                         2
          @ Fermilab     Gradient (yet to be demonstrated)


                Cavity   Qualified    Operational     Length*       energy
                 type    gradient      gradient

                           MV/m           MV/m           Km          GeV

initial         TESLA       35            31.5          10.6         250

upgrade           LL        40            36.0          +9.3         500

                                                    * assuming 75% fill factor

            Total length of one 250 GeV linac  10km
            Need for each linac:      7872 Nb cavities
                                      984 cryomodules
                                      328 Klystrons
                                 …many km’s of waveguides

                             ILC@FNAL - Nagaitsev                                3
       @ Fermilab       About the ILC Bunch Compressor (from PT)

   Two-stage bunch compressor
      Stage 1 @ 5 GeV (DR extraction energy) – 2 Klystrons (?)
      Stage 2 @ ~15 GeV -- 20 Klystrons(?)
      Necessary to get large compression factor with large longitudinal emittance
       from DR
   Two final bunch lengths to consider
      1 psec RMS (nominal)
      0.5 psec RMS (LowQ / HighLum)
   Two configurations for each final length
      “A” configurations
         • Total 180° longitudinal phase rotation (90° per stage)
         • Easier transverse tolerances (lower energy spreads) at expense of
            tighter RF and longitudinal tolerances
      “B” configurations
         • Undercompress in stage 1 for 90° total rotation
         • Easier longitudinal tolerances at expense of tighter transverse
            tolerances
      “B” configurations are baseline, but being able to achieve “A” might be useful
       as well




                                ILC@FNAL - Nagaitsev                                    4
@ Fermilab                     Luminosity vs. IP Offset for Nominal ILC
                                     Parameters (from M. Church)
                        luminosity vs. IP offset (ILC nominal)
                0
                                                                        y = m1*M0*M0+m2*M0*M0*M0*M0

                -5                                                                        Value       Error
                                                                              m1    -0.00015487 2.5352e-06
               -10                                                            m2     1.6059e-10 8.2611e-12
                                                                        Chisq              2.3921         NA
dLum/Lum (%)




               -15                                                          R             0.99938         NA

               -20

               -25

               -30

               -35                                                                                  calculated with
                                                                                                    Guinea-Pig
               -40
                 -600   -400      -200          0        200         400           600
                                          IP offset (um)

                               e- focus                IP                      e+ focus
                e-                                                                                   e+
                                                         IP offset




                                                               waist offset



                                            ILC@FNAL - Nagaitsev                                                      5
       @ Fermilab                 2% integrated luminosity reduction; BC300B

                                                6                     12                       10 cm
N  30000                           m  10     m          ps  10        sec       c  3 10 
                                                                                                        sec

i  0  N  1         e  160 m            p  160 m

                    e                                  p 
            
t 1  rnorm N  0                              
                                      t 2  rnorm N  0     
                    m                                  m 

                            2
                     t
           0.00015487
L( t)                         1                   L( 114)  0.98       -- From M.Church
                  100

                       t 1i  t 2i  
Lum                             
           1
                      L                         ( Lum  1)  100  1.971 -- % luminosity reduction
      N                2 
                 i                    


  e                                       e    1
        0.533ps                                          0.252          degrees of 1.3 GHz at BC
   c                                             
                                           c 2.117ps                     (sensitivity number f rom M. Church)



                                                ILC@FNAL - Nagaitsev                                            6
               @ Fermilab                Luminosity distribution for 30000 bunches

                       1.2 10
                             4




                        1 10
                             4




                           8000
# of bunches




                 1
               H           6000


                           4000


                           2000


                                 0
                                     0         5               10             15       20
                                                              0
                                                           H
                                                    Luminosity reduction, %
                                                                                   .


                                              ILC@FNAL - Nagaitsev                          7
    @ Fermilab       PT Slide from Snowmass Conference


    Tolerances on Mean Phase and
     Amplitude of BC RF System
Blue = BC1 drives tolerance, Red = BC2 drives tolerance, Purple = both systems
                                 about equal
Parameter      1 Stage      300 “A”      300 “B”      150 “A”      150 “B”



BC RF            0.2%        0.1%         0.15%        0.08%         0.1%
Amplitude                                 ??                        ??
BC RF            0.07°       0.05°        0.12°        0.03°        0.06°
Phase
                                        0.25                       0.12


                             ILC@FNAL - Nagaitsev                                8
     @ Fermilab   Time scales needed for rf specs


1.   Bunch-to-bunch (100s of ns)
2.   Train-to-train (few Hz)
3.   Long term (Seconds to days)

Bunch patterns:
1. One pilot (1% intensity) bunch alone – should be able
    to get all the way through
2. One pilot (1%) bunch followed by a single (nominal)
    bunch (10 us) apart




                         ILC@FNAL - Nagaitsev              9
   @ Fermilab                     RF specs

 Bunch compressor (BC) may require a separate rf
  spec (nominal 300B: 0.25 deg rms, 0.15% (??) rms).

 Main Linac (all energy jitter):
    Each bunch has an rms energy spread of 0.1%
    Colliding bunches have luminous region rms energy spread
     of 0.39 GeV (0.15%)
    Bunch-to-bunch energy jitter adds in quadrature
    Factor of 50 due to SQRT(N_klystrons)
    >1 deg, 1% (rms) uncorrelated rf phase/amplitude
    0.5 deg, 0.1% correlated


 Still need to look at extracted bunch (from DR)
  energy jitter.

                      ILC@FNAL - Nagaitsev                      10
   @ Fermilab                Groups of specs

 20% energy error can get the beam thru the Linac
 2% energy error to get to the IP
 Two groups of specs: inside-the-feedback-loop,
  outside-of-the-loop
 Outside: MO, Timing distr, Reconstruction -- SNS
  attained: .1 deg btw adjacent cavities, Tesla
  prototype: 0.5 degree rms over 15 km over long
  time scale
 Inside: LLRF, HLRF and rf distribution
    Assumes we can measure and sum up 24 cavity probes
     without errors
    Some cables (from cavities to LLRF crate) will be 100 m
     long



                      ILC@FNAL - Nagaitsev                     11
       @ Fermilab   What About Beam-Based Feedback?

   This slide from Marc Ross

   Energy:
      Certainly will have train-to-train feedback
         • Takes care of variation on time scale of seconds or longer
      Intra train feedback dubious
         • Takes ~10 usec to change cavity voltage ~1%
         • Takes ~50 usec for signal to reach front of linac from IP
         • May not be possible to make an intra-train energy feedback which has speed and
           range required
   Arrival time:
      Will have train-to-train feedback if we can make reliable IP instrument to
       measure arrival time difference
         • Otherwise, we’ll rely in a dither feedback
         • In either case, need seconds to many minutes to detect and correct arrival time
           problems
      Intra-train has similar issues to energy




                                  ILC@FNAL - Nagaitsev                                       12

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:2/16/2012
language:
pages:12