User name sawatzky
Log entry time 11:24:19 on March 12, 2005
Entry number 141298
This entry is a followup to: 141293
The negative deadtimes are quite small (-5 to -2). Browsing the code in
/adaqfs/home/a-onl/scaler/dtime/ suggests that the most probable culprit
is a too coarse/incomplete calculation of the "corrected" triggers
(intended to take correlations between triggers into account -- ie. a T5
is also a T1 and a T3).
The upshot is that a livetime value > 1 is computed under our trigger
conditions. Thus deadtime (=100.*(1 - livetime)) goes < 0.
FWIW, the numerator and denominator of the datamon livetime calculation
are each around 260, so it wouldn't take a big error/omission in the
trigger correction calculations to produce what we see...
(I know, more detail than necessary. It's been an uneventful shift :-)