Subject: I endorse the new calibrations
From: Bob Nolty (nolty@hep211.cithep.caltech.edu)
Date: Fri Jul 21 2000 - 20:44:18 EDT
I have examined the 27 weeks for which new calibrations were produced
a couple of months ago (sets 7466 through 8534) and find them quite
usable. The MASK file created as a by-product of the calibration
procedure is useful, but is not sufficient for eliminating hardware
errors. I had to supplement the MASK with a few microcuts of my own
to get acceptable results.
http://www.hep.caltech.edu/macro/protected/notes/newcal.ps
shows beta distributions (n.b. not 1/beta). The top histogram is
using old calibrations and my old microcuts. The middle histogram is
old data, new calibrations, implementing the mask. It looks horrible,
but the problems are isolated to just a few supermodule-weeks or
ierp-pair-weeks. The bottom histogram is old data, new calibrations,
implementing the mask and some newly-computed microcuts.
Comparing the top histogram to the bottom, the top seems to have fewer
mistimed background events, but both are quite clean. The bottom
contains not quite 1% more events, which means that even with the mask
implemented, fewer channels failed my objective microcut criteria so
the new calibrations increase the number of usable channels.
The microcuts I had to implement are due to the problems seen in
http://www.hep.caltech.edu/macro/protected/notes/newcal-problems.ps
All of these plots were produced after implementing the mask; these
are the problems that remained. The microcuts I implemented were to
cut interERP events involving 2/3 or 3/4 for sets 7689 and 7928, and
to eliminate all SM2 hits for sets 8430, 8487 and 8534.
Bob
This archive was generated by hypermail 2a24 : Fri Jul 21 2000 - 20:44:23 EDT