This file: ~cdaq/gen01/Documents/moller_howto.txt

Initial draft: B. Zihlmann, Aug 2001
Updates: M. Zeier, Sep 2001


GEn01
How to do a moller run 
----------------------

0.0) setup all GUI's needed for moller run
     these are supposed to be allready there
     in the moller work space on the x-term above
     the cdaqs3 terminal.

0.1) check He level in the moller solenoid
     If closed open the fill valve (E91017) to 10% and then after
     a while stepwise to 20%, where it stays during the whole
     moller run.
     Valve EV91015 stays at 21% always.
     Nitrogen should be on PID.

1.0) check that the collimators are at the right place
     the correct positions must me determined experimentally
     once for each energy. 
     Howard Fenker uses the monte carlo (MC)
     to get good starting values for the quadrupole currents
     and collimator settings which in turn need to be optimized
     in the first moller run.

1.1) Polarized Target: Move carbon target or empty target
     into position

1.2) Turn off slow raster.

1.1) set solenoid to STANDBAY (SDBY) in the meantime
     this will turn on the heater so the solenoid will
     be ready to ramp up.

2.0) Ask MCC to tune for moller: this means
     CENTER BEAM AT 3C16, 3C17 AND 3C18 IN X AND
     Y TO ZERO +- 0.5MM. usually the beam is centered
     very well at 3C16 and 3C18 while it is slightly off
     at 3C17 by +0.5mm in x and y.
     (Corrector 3H01H needs to be used for that)

2.2) turn on Power Supply PS for Q2 (but 0 current!) 
     and enable the alarm of moller Q2 (same pannel as HMS, SOS)

2.1) ramp up moller solenoid to 3 Tesla (press "go set") 
     This will steer the beam horizontally positive by about 4mm
     and vertically positive by about 1.5mm at 3C18

2.2) ask MCC to steer beam back to zero in x/y at 3C18
     keeping position fixed at zero at 3C16/17

2.3) ramp up small quadrupole. For the 2.33GeV beam this
     is 136 Amps and will steer the beam vetically negative by
     2mm and horizontally by -1mm.
     To ramp the magnet the fields "ramp down" and "peak current"
     can't be empty or 0.
     peak current: 150
     ramp down (A/sec): 30 
     A nonzero value in the  current
     readback does not necessarily mean that the magnet is powered. 

2.4) ask MCC to steer beam back to zero in x/y at 3C18
     keeping position fixed at zero at 3C16/17

2.5) ramp up big quadrupole (Q2) to field. In the case
     of 2.33 GeV beam energy this is 360.6Amps. This should
     not move the beam at 3C18 vertically and only very small
     horizontally (~0.5-1.0mm). if the beam moves much more the 
     tune is bad.

3.0  turn off beam and move in moller target (pos #2 = 10um foil)

3.1  turn beam on again and check rates (visual scaler in moller electronics):
     Typical values for 10um foil and 100nA:
     L single: 60k
     R single: 130k
     Coin: 4k
     Acc: 60

4.0) setup DAQ for moller run (configure for run type moller)
     
4.2) take data at 100 nA
     typically 100000 events for a ~1% absolute statistical error.

5.0) the runs are analyzed in the directory 
     ~cdaq/moller/anal/e93026/
     the program is called "replaynt"
     
5.1) the results of the run are writen so file and screen.
     the value with accidental background subtracted is the 
     correct figure for the polarization.

5.2) a kumac from Howard  mscan.kumac can be used to scan through
     the most important moller spectras. 
     a) check that the threshold in the discrimnators does not
        cut into moller peak in the ADC spectra
     b) check that the ridge in the 2dim. spectra of the 
        hodosocpes is on the diagonal
     c) check that the projection of this diagonal looks
        parabolic with the minium in the center. best results
        are optaned by looking at the multiple hits spectrum

repeat 4.1 and 5

after run is over:
   -turn off beam
   -retract target
   -ramp down magnets
   -close He fill valve
   -ask MCC to retune beam to previous settings


Expert: Howard Fenker
        office: 7431
	pager: 820-9799
	cell: 890-9512