Main INDEX
Monthly INDEX
PREV
NEXT
Make New Entry,
Make Followup Entry
User name R. Michaels
Log entry time 21:03:53 on November 20, 2008
Entry number 250089
This entry is a followup to: 250057
Followups:
keyword=re: CRL for RICH
Just to clarify ....
In consultation with Dave Abbott, today Alex and I did the following to the RICH CRL:
1. Cut on event type. Only T3 and T5 are read out, the rest are rejected.
(Note that the ADCs and gassiplex are cleared after each event in any case.)
2. Reduce the wait loop to be 1.5* the max theoretical wait time. If the the wait loop
times out, the data are flagged bad.
3. Eliminated a secondary, uncessary wait loop (related to point 2).
Note, we think the combination of 1,2,3 should reduce deadtime, which might have been
up to 4 msec previously under some conditions. The wait loop had been ~12 times the
max theoretical. We'll see ...
4. (From Alex before), don't read out if more than a certain maximum number of hits.
Note, this can cause truncation of data ! However, those events are probably junk.
It needs to be checked in offline analysis to see if cut is ok (it looked conservative to me).
5. Cleanup code -- eliminated confusing blocks of commented-out code
Please report on the deadtime performance with RICH and please analyze those rings
carefully to see if we did nothing wrong.
A copy of this log entry has been emailed to: camsonne@jlab.org,kalyan@jlab.org