muTr Calibration Procedure for data taking - MJL - 7/26/01
==========================================================

To take calibration data:
* log into phoncs account on phoncs0
* make sure to do a setuponcs
* make sure DAQ is working (e.g. by checking with standalone run control)
* beam should be off and no run with the "grand" partition should be going
* cd ~phoncs/mutr/calib
* make sure no muTr run control is running. You can use
  "psit" to look for a muTr run control and "killit" to
  kill any you find. Also look for muTr GUI runcontrol
  and stop it if you find one.
* run calibration script to take a series of runs:
  Just type "docalibNS" (or docalibN (North only), docalibS (South only))
* if the DAQ on North or South hangs the respective log files may
  show an infinitely repeating report of the same number of events
  and never reach the required 120 events. If so you will need to
  use ps to find the guilty calibN.csh or calibS.csh script, kill it,
  and then use psit to find the rc (118 for South, 120 for North) and
  kill it with killit (e.g. killit phoncs/RunControl120 for North).
* note what log file is being written to (the "docalibNS" script
  prints out the log file name) (Make sure this log file
  doesn't fill up the phoncs home disk with error messages -
  yes it once happened)
* a quick check of the data can be done as follows:
  - ddump -c 10 -f rc-0023108-MUTR.S-3.prdf | more
  - where you would use a run number near the end of the series
    (see log file) and if the pulser was running ok you should
    see that the last 3 samples for most channels have adc values
    significantly different than the 1st sample indicating
    non-pedestal adc values. The 1st of the 9 runs should have
    negligable differences between the 4 samples since it is
    taken with the pulser amplitude set to zero.
  - also look through the log file searching for "collected" and
    you should see 9 runs taken that collected more than 100 events
    each.
* automatic analysis will be done of a successful calibration run
  initiated by a cron job running in the phnxmutr account on va011
  and results should appear on the calibration results web page
  (http://www.phenix.bnl.gov/WWW/publish/phnxmutr/calib_results/calib_results.html)
  10 or 15 minutes later. Zero-suppression thesholds and the database
  will also be updated as part of this process.
* often the copy to the web page fails because the samba mount has
  gone away. This can (sometimes) be refreshed by logging into the
  phnxmutr@va011 and doing 'mount /mnt/win/phenix/'. Then cd to
  ~phnxmutr/calib, identify the respective analysis directory (.e.g
  North-11Sep02_2229 and do 'domove North-11Sep02_2229' which will
  put the results on the web)


Notes:
* calib.csh uses run control to take a series of 9 runs at
  calibration DAC amplitudes of 0x00, 0x10, ... 0x80
* data is recorded into /common/buffer2/calibdata or
  /common/buffer1/calibdata
  in prdf files with names like, e.g.
  rc-0016429-MUTR.S-0.prdf
  (where 16429 is the run number, the "-0" is the station-1
  data file, etc.)