The summary of the past few hrs is at the end. Please read the plan first
For each setting, we are only going to measure with the TOF's set of a gain of 1300
above pedestals (commis/hv/tof_1300.sql), all thresholds at 31.3 mV. For each setting
I prepared the relevant trigger files. Be aware with the paddle HV and
thresholds.
A-shift: collect data for LTOF 4-7, only with the above TOF setting. Summarize in
logbook data analysis (also, update your macro..) In parallel with the TOF
data taking, you should try to find a setting for LC1. You may want to try
adding another 10 % to HV to the present values (see hv_defaults.cerenkov), reduce
the LC1 delay if data reaches to TDC channel 4000 (use first 4 channels of LEdgeDel)
or maybe lowering the threshold (first 4 channels of LEdgeDisc). 1 or 2 hrs of beam
should be enough to see a difference in LC1
B-shift 10/27: At backward angles we get less rate. We are happy to have about 1,000
counts for determining the efficiency. Assuming that the L4-7 setting has
enough statistics by now, prepare to move paddles to RTOF 0-3. This requires
moving the start counter. I have to be informed well in advance before that moves happens.
Re-adjust cables in the D-tunnel.
Collect & analyze data
At these forward angles we colleced all of our data in 4 hrs. So you should be able to
finish this setting in your shift. Again, starting from the present (factory-"recommended")
values, try to find a decent RC0 efficiency
C-shift 10/27: Hopefully with help from B, move to R4-R7... same as above
A-shift 10/28: Given the above Cerenkov information, study the C efficiency on a more
open trigger (use central region of TOF). USE THE BEAM.
B-shift 10/29 Hall will be open. Cerenkov flasher studies.
next week: depending also on the Wch plans, we may go directly to eep runs (Wihtout lead foils!)
or continue on to the L8-11 and R8-11 regions. The data of this weekend provide an
independent check. Most of our efficiencies will be hopefully easier to study now that we are gain
matched. That includes all remaining detectors.
**********************************
Summary saturday 10/26
The very low counting rate of the past 2 shifts was also due to the
threshold of the start counter. We typically always used 31.3 mV but never
worried about "rates". Apparently, at 21.3 mV we increase the rate of good
events. We also increased the SC voltage by 50 V
Note updated threshold and HV values for the paddle scintillators (logbook)
We retuned the slits, obtaining the reduction of a factor of 10-20 of the
open trigger rate (any_coincs) with a minor compromise onthe lifetime
(10-20% on empty target). With these slits, lifetime at 2.0 sccm is 6-8
minutes. Injection current with 2.0 SCCM target reached 100 mA
consistently (snake off tune)
In 4 hrs we are able to collect more data than the past 16 hrs. We have
about 1000-2000 events for each tof from which we can estimate
efficiencies. More careful analysis shows results 90 % for L0 and 96 % for
both L2 and L3. Two aspects are very important:
a) There are still showers coming from the ring, and at this low rate,
also during injection (we can trigger on standby at injection, so we
decided to ramp TOF hv to zero, rather than standby.)
THEREFORE make sure that - when studying a given tof - no other TOF
fired. Events with more than one particle should not enter the analysis
for the efficiency
b) that reduces most, not all, of the bad stuff. A target event must pass
the start counter before the paddle and must be timed by the paddle.
Therefore you should cut around the "self-timing" peak of each paddle
to make sure you have events from the target.
c) note also that there is geometrical acceptance for events to go through
-say- LTOF 1 and paddle #0
Cutting deeper into the self-timing peak the efficiency numbers go up and we get 91%, 98%, 98%.
We have a problem with paddle #1 (LTOF #1) which we can not understand. Most the events there
seem bad. On the scope I do not see any evident timing problem (each paddle has a different
tube, a different rise time). LTOF#1 however is counting reasonably so I expect
the problem is somewhere else (tdc cable??). If this paddle keeps giving problem we should
change it at the next cofniguration change (a "spare" is in the hall).
We took 4 hrs of data with the TOF_1500.sql setting and 4 hrs at the TOF_1300.sql.
In both cases, we get the same TOF efficiency. We clain the TOF_1300.sql to be of "high
efficiency".
****
Cerenkov efficiency:
a very similar analysis yielded 60 % for the cerenkovs. Increasing the voltage by
10 % yielded 65 %. We can go up another 10 % or maybe lower our threshold (?). Note that
the cernkov efficiency canb e studied in various ways, this is a first pass. This set up
has too low a count rate for detailled studies.
*******
software: the efficiency macros are tof_eff.C and cc_eff.C. And that is for everybody.
Need I say more ?
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:28 EST