Present:
M. Ito, D. Sober, E. Clinton, M. Wood, A. Teymurazyan, K. Hardy,
M. Grisson, L. McLaughlin, N. Woods, K. Broadnax, D. McNulty,
A. Gasparian, V. Mochalov,
On Phone: D. Dale, R. Miskimen.
Secretary: V. Mochalov
Agenda
1. Appointment of secretary: all
2. Review of last meeting's minutes: all
3. Tagger Timing: Eric
4. The Mochalov Report: Vasily
5. Any new business: all
Meeting protocol:
1. Secretary for next week: Kelly Hardy
2. Previous week meeting minutes are absent.
3. Tagger Timing (Eric).
The information discussed is presented on Eric timing page:
(see http://www.jlab.org/~eclinton/15July2005)
Eric investigated the influence of the tagger timing windows to the
inefficiency of tagger reconstruction. He presented the number of
non-reconstructed combinations as a function of timing window for
information in adjacent channels (EE, ET and TT correlation).
Main conclusion is that inside 28 nsec window (14 ns cut) inefficiency is
less than 1% for all EE combinations.
Comments concerning the talk:
3.1. TT combinations should be ignored, but ET correspondence should
be studied accurately.
3.2. it is necessary to show reconstruction efficiency (and accuracy
of the efficiency) as a function of timing window width.
3.3. sometimes 4 electrons are reconstructed in tagger when there
is only two hits in raw data.
3.4. background is over estimated
Recommendation:
-- it is necessary to show reconstruction efficiency
(and accuracy of the efficiency) as a function of timing
window width.
-- (as a response of 3.2 and 3.3) Eric and Aram will prepare the
note next week with description the strategy of reconstruction
and will discuss it with E.Pasyuk
-- when time window cuts will be defined and final reconstruction
will be prepared Aram have to redo his flux analysis.
4. Vasiliy presented short summary of software technical meeting.
All information is presented on the page:
http://www.jlab.org/primex/weekly_meetings/slides_2005_07_15/2005-July-15-vm.html
Protocol of the Software technical meeting (2005-July-14) is saved at:
http://www.jlab.org/%7Emochalov/hycal_software/soft_meetings/2005-July-14.html
4.1. Leakage algorithm outside detector is under development (Vasiliy).
Recommendations - some investigations to show the accuracy of the
method are required
4.2. Error flags for tagger (Eric) were presented.
They will be discussed after DAQ errors will be fixed
(See Eric page: http://www.jlab.org/%7Eeclinton/14July2005/index.html)
4.3. Pile-up or rate-dependence investigation was presented by Kelly
(See Figures on: http://www.jlab.org/%7Ehardyk/)
There is no evidence of the effect.
Recommendations:
Dustin will check possible pile-up effect (or gain dependence) for LMS.
4.4. Mark introduced future implementation of the exceptions.
Real implementation will be presented (discussed) later.
4.5. Different discussions at software technical meeting:
4.5.1. How to work with "bad pedestal channels".
4.5.2. It was long discussion about loosing of information
about prescale factors.
The main conclusion - the prescale factors should not influence
on cross-section result.
4.5.3. DAQ decoding - failed to meet with Dave -
try to do it next week (also tagger aspects will be discussed).
4.6. Error flags (representative from each detector)
hycal is introduced, Eric presented for tagger.
Another libraries not presenting result:
* veto, veto2hycal (Mike)
* trigger, tac, ps, pflux (Aram)
* timing_links (Vasiliy, Dustin - next Thursday)
4.7. Other activity (not discussed)
Current activity:
* cross-check of the new algorithm code
-- under investigation (Dustin)
* discussion about timing links
(Dustin and Vasiliy - next Thursday will be discussed).
* Compton - prepare for new skim files and future calibration (Elena)
* beam position activity (see Mike G. today)
* trigger data base activity
New activity:
* shower leakage inside detector (Vasiliy)
* new hycal timing reconstruction (Vasiliy)
* final tagger calibration:
Discussion on tagger:
-- New corrections of tagger energy reconstruction are prepared
by Stepan Stepanyan. Vasiliy will speak with Dave how to introduce them.
-- It is required to know exact electron beam Energy --
we need to extract the information from Hall-C data
-- the tagger energy reconstruction also is sensitive to beam position.
Dan Sober is working on these corrections
5.1. Dustin presented the information about LMS stability
(really first talk of the meeting)
All pictures are saved in the directory:
http://www.jlab.org/primex/weekly_meetings/slides_2005_07_15/lms/
It was found that:
5.1.1. The "lms gain" correction (comparison between first and second
calibration) gives:
for crystal: mean shifted +1%, sigma 1%
glass: mean shifted -1%, sigma 2%.
(see picture at:
http://www.jlab.org/primex/weekly_meetings/slides_2005_07_15/lms/lms_gain_rel_gain_begin_end_Diffs.eps)
Comments:
-- final calibration required to understand systematic
shift between glass and crystals.
-- we need to check calibration using pair-production and Compton
data.
5.1.2. Kelly did not see pile-up (rate dependence effect).
Nevertheless Ilya found the effect -- see figure:
http://www.jlab.org/primex/weekly_meetings/slides_2005_07_15/lms/lmsgain.ps
Unfortunately there is now LMS information in the files when
different current was studied.
Dustin will repeat Ilya analysis.
5.2. Mike G. responded on question from previous meeting to understand
why sigma in bpm's is small in some events.
It was found that small sigma is corresponding to the epics event
with really law statistics (beam trips).
The data presented at
http://www.jlab.org/primex/weekly_meetings/slides_2005_07_15/beamMonTalk2.ps