Minutes of CALCOM Analysis Meeting, May 9 1997.
                   =================================================

 Agenda:                 
        Offline analysis teams for commissioning runs     -  V. Burkert
        Update on EC gain matching                        -  L.C. Smith
        Summary of February test run analysis             -  B. Niczyporuk

Note: There will be no Friday CALCOM analysis meetings during the upcoming 
photon run. 
==========================================================================
Reports:

Volker Burkert  presented a plan to form offline analysis teams for shift taking
during the commissioning runs. The reason for having special analysis teams is
to quickly obtain infomation about the detector performance, running conditions,
etc. which will be needed as input into the decisison making about whether or 
not data taking for a specific commissioning task has been completed. During 
commissioning run conditions may have to be changed frequently for the study of 
specific detector performances and calibration procedures. 
A variety of skills will be needed for the members of the teams, from low 
level (ADC, TDC) detector analysis, knowledge about RECSIS, to more specific 
analysis procedures (e.g. straight tracks analysis, MIPs analysis, etc.). 

For the June commissioning run the CLAS_Offline group will organize these 
teams based on the run plan which is currently being developed.


Bogdan Niczyporuk:  (no report received)
=================
Bogdan presented a summary of his analysis of the February test run, covering
all areas of detector analysis and efficiency studies for drift chambers,
TOF, Cherenkov counters and calorimeters. Some of the methods used to define 
efficiencies prompted heated discussions, especially in the area of electron 
identification. 
Everyone interested in the results should contact Bogdan directly. (V.B.)


Cole Smith:   - Update on EC Gain Matching
==========
Cosmic ray runs are continuing with the forward angle calorimeter in
an attempt to use MIP tracks in single pixels to gain match the 
photomultiplier tubes to within 10%.  This technique assumes MIP
muons deposit fixed energy/length in each pixel so measured energy
is function only of pixel distance from PMT and pixel stack length.
Attenuation lengths and gains are then derived from exponential fits.

So far data-taking has been limited to Sectors 3,5,6, which have favorable
acceptance for cosmic muons. Typical runs last 2 
hours and provide 3.6M events, 2-3% of which survive single pixel cut.
After several iterations of fitting and HV adjusting, the RMS/mean 
of the distribution of mean pixel energies (uncorrected for attenuation)
is as follows (best case is expected value due to scintillator 
attenuation assuming perfectly matched gains):

	   Measured	  Best Case
Sector   Inner  Outer   Inner  Outer   
  3      11.1%   8.9%    2.2%   2.6%
  5       7.6%   7.2%    3.1%   2.7%
  6      15.2%  12.2%    4.4%   3.5%
  
The substantially better results for Sector 5 simply reflect a larger
number of iterations for this sector.  Once the intrinic gains of
the PMTs are known from the laser calibration system this process will 
converge more quickly.