Here are a few notes on today's meeting. Feel free to correct, add or
comment on anything...

======

Present: Alex, Carlos, Eric, Franck, Malek

1.- Malek on nDVCS proposal:

a) Generally agreed to start writing the proposal under the assumption of
one single Q2=1.9 GeV2 and two beam energies, with the goal of separating
DVCS^2.

b) Latest ELOG entry by Malek (Results/151) shows the statistical
uncertainty expected in the extracted coefficients assuming same
luminosity as previous experiment (and a single beam energy). Some
concerns raised (by me) about the fact that no systematic uncertainties
were considered in that study. I'll try to be more clear in a separate
email.

c) Malek described the first plot of that same ELOG entry which he claims
explains the bad chi2 on fits of the unpolarized cross section on the
neutron. He attributes that asymmetry to the higher inefficiency of some
blocks. He may try to estimate this inefficiency using a physics channel
(pi-?) and correct for it if possible. Notice that in principle this
asymmetry could be due to some sine-like physics channel, although all
contributions to the unpolarized cross section should be cosine-like and
therefore cancel on that asymmetry.

d) CONCLUSION: Malek will write within the next week a first sketch of the
proposal to be discussed in the next meeting. If enough time, he will look
into point c) above.


2.- Eric on nDVCS paper status:

a) Discussed Markus (and Marc's) comments to the paper. None of them seem
major. Eric will implement the changes by the end of the week and send it
to us. After agreement (or further discussion in next week's meeting if
needed), he will send it to the Hall A Collaboration (hopefully next
week).

b) Some discussion on whether to include Simonetta's new predictions into
the plots. No clear agreement reached, to be discussed later (we can
produce the plot anyway and decide later).

c) Some discussion on the infamous factor 5 in the linear combination of
Ju and Jd we measured (Marc's remark). Eric might ask Marc to think about
it, but it's probably a model-dependent result with no much physics
foundation.


3.- Franck on long paper preparation:

a) He will write a sketch soon and ask for contributions.

b) Small discussion on whether 1 or 2 (proton and neutron) long papers.
Discussion postponed after the long paper is more advanced.


NEXT MEETING: next Tuesday July 17th same time (unless nothing to discuss)
------------


On Mon, 9 Jul 2007, Carlos Munoz Camacho wrote:

>
> Dernier rappel ;)   A demain! - Carlos
>
> On Fri, 6 Jul 2007, Carlos Munoz Camacho wrote:
>
>>
>>  Bonjour,
>>
>>    Un rappel du meeting qu'on avait prevu mardi prochain (10 juillet) a
>>  10h/16h. Dans l'agenda:
>>
>> >   - nDVCS 6 GeV proposal status (Malek, 20+10')
>> >   - E03-106 paper status (Eric, 5')
>> >   - pi0 paper (Charles/Pierre, 5')
>> >   - Long paper: plan & todo list (Franck, 5')
>>    - (?) Note Charles "asymmetries & sections efficaces" (ELOG:Results/150)
>>
>>  Si quelqu'un a des transparents, merci de les envoyer ou poster a l'avance.

  - Accelerator has made some progress during the weekend. 5.75GeV setup
was no problem but still lots of RF faults. Hall A lockdown still
scheduled for wednesday 1600.

  To Do List until wednesday 1600
--------------------------------------------------------------------------

  General:

  - New HV cards need to be put into EPICS. 2 HV spares need to be located
(Sirish).
  - Get results from survey (Pierre).
  - Verify remote controlled switches, test, and write procedures (Alex).
  - Write procedures (Cath).


  Calorimeter:

  - Check Z-axis, check vertical limit switch position (Carlos).
  - Check-out DC current signals and read-out on all channels (Cath).
  - Pedestal run, commit it to DB (Carlos).
  - Leave HV and LV on for 2hr or so before taking any data.
  - LED scan, infer new HV, new LED scan to check (Cath, Carlos).
  - Long cosmic run (Carlos).
  - LED scan once in a while to check long term stability (Cath, Carlos).
  - Ask Barbosa for spare LED board.


  Tagger (Malek, JM)

  - Test all channels.
  - Pedestal run. Commit it to DB.
  - Leave HV and LV on for 2hr.
  - Take cosmics. Check with MC (some day).


  Proton Array (Gagik, David, Charles)

  - Repair one channel (amplifier).
  - Pedestal run. Commit it to DB.
  - Leave HV and LV on for 2hr.
  - LED runs, gain variation versus DC current (continuous LED).
  - Check time stability by several LED runs.


  DAQ (Alex, Matthieu, Michel, Bob)

  - MSQL DB for Proton-Array + tests.
  - Zero-suppression for Proton-Array + tests.
  - MUX issue need to be solved (until 2pm monday).
  - TS setup for multiple cosmic
  - Start and End of run scripts need to be moved to adaql2.
  - Final DB.
  - Timing with pulser (after 2pm monday).
  - DAQ stress test.


  HRS

  - S2 timing (Rob).


  Beamline (Charles, Sirish)

  - Check if all elements are functional, including Moeller, EP, Harp,
ARC, BCM, BPM, ...


  Misc

  - Scalers to be plugged in + tests (Alex).
  - Put everything to ground (Jack).

Present: Hari, Zafer, Pierre, Srish, Franck

Beam requirements:

Beam size: 50 < x < 150  75 < y < 250  with X:Y is 1:1.5
                       (ideal 80 by 120)
Beam position stability: +- 0.1 mm

Energy: 5.75 GeV

Polarization: as large as possible, >85% with super-lattice.

Energy spread: 6x10-5

Current: 3uA except during calibration ~80-100 uA
          ( no raster during 3 uA )
           current measurement accuracy of 1%
FFB on Energy and position mode

SLI will be monitored and Target OTR will be inserted on demand

Beam thru Compton Chicane with background of 200 Hz/uA or better

30 Hz pseudo-direct report.

  Present: B. Michaels, S. Nanda, K. McCormick, E. Voutier, J-M. Fieschi,
F. Sabatie, C. Munoz Camacho, M. Beaumel, P. Bertin, C. Ferdi, A.
Camsonne, A. Chanteleauze

  DAQ/Alex

  Electronics rack in spectrometer hutt. Remote controled plug is
working and will allow reboot w/o operator present. clrpc48 (PA/veto DAQ
when in the EEL room) is now in the CH for a few tests. Calorimeter crate
works fine. PA trigger has issues with a couple channels, needs
investigation (after B. Michaels is gone). Moving the acquisition to adaq
cluster in progress... some issues still. Bob Michaels has postponed his
vacations for another week to help setup the DAQ (big thank you!). Few
last channels of ADC for DC level monitoring need some work, Kathy will
try and do that before leaving.

  PA/Charles:

  Comparison of cosmics between reality and MC in progress, results next
week. Currently buidling a power supply for LED pulser. Last ADC cable for
DC monitoring has been built.

  VETO/Jean-Marc:

  No more HV problems. Two channels had shorts after the patch panel
boards went for repair. Just got them back from electronics group, will
test them again. VETO ready for transport at the end of the day today.

  Calo/Catherine/Kathy:

  - Catherine summarized her work on the LEDs. The corresponding technical
note can be found in the Elog and shortly will appear on the web page.

  - Four "broken" blocks have been repaired. 9 out of 11 spares are
already in the Hall. No clean work can be done in the Hall due to dust and
so on. If additionnal work on the calorimeter blocks needs to be done, we
have to do it some place else. All spares are already repaired.

  Misc:

  - Need to check position of epics/motor control place with 25m cable and
install them.

  - Meeting at 10am on friday with A. Areti for beam issues.

  Present: Charles, Gagik, David, Catherine, Alex, Jean-Marc, Franck,
Carlos, Matthieu, Eric, Malek, Sirish.

  Franck/Installation:

  - Electronics rack will move into the Hall on thursday afternoon.
Calorimeter unstacking will take place in the clean room on
thursday-friday. Blocks will be brought 5 by 5 to the Hall for
repair/stocking/restacking. DVCS detectors+platform will go into the Hall
early next week. RG213 will be moved at about the same time.

  - Few remaining question to Jack: RG213 cable management (carriages?).

  Sirish/Documents:

  - Need to write COO, ESSAD, RSAD. TSOP may not be needed (will know
soon). ESSAD is what we should focus one. Each subsystem should write
about 1-2 pages describing installation and associated hazard and security
features.

  - Still some money to spend quickly.

  Gagik/Proton Array:

  - repaired one of the worst-case block (horizontal tower). Only took
20min to replace the base which didn't work. Now, all PA channels are
fully operational.

  Carlos/Calorimeter:

  - Performed tests using continuous LED. Gain stability was better than
1% at 20uA and better at lower anode current.

  - Performed stability tests, using chronical LED scans over a period of
10hrs. Clearly the system needs about 1-2 hrs to warm up. After this,
stability is better than 0.3% (might be even better, since the test is
limited by the statistical accuracy of the LED scan itself).

  Eric/Veto:

  - Patch panels are being repaired by local electronics group. Everything
will be ready by the end of the week.

  - Writing 2 tech. notes, will be available shortly.

  Alex/DAQ:

  - problem with 2 ARS boards. Had to use spare. Now no spare anymore.
Faulty boards will be repaired/changed. Monday-Wednesday next week, Alex
and Bob will work together on DAQ, and integration into Hall A.

  - Will try new PowerPC CPUs.

  - Some remote switched have been ordered. Might need one more.

  Franck/DB:

  - script to send coda information to MySQL database on a run by run
basis has been written.

Information for run 4127, End of Run Comment:
Run_type=LED,beam_energy=0,beam_raster=off,comment_text=Chain stability,daq_mode=DVCS only,dvcsarm_distance=0,dvcsarm_theta=0,leftarm_p=,leftarm_theta=,rightarm_p=,rightarm_theta=,target_type=Dummy
 

  Present: A. Camsonne, J-M. Fieschi, A. Chanteleauze, F. Sabatie, C.
Munoz Camacho, M. Beaumel, C.E. Hyde-Wright, D. Hayes, G. Gavalian, S.
Nanda, K. McKormick, J. Segal, E. Voutier, M. Mazouz

  -> Jack and everyone/Installation:

  - 08/02 :         DAQ side un-cabling. Preparation for transport.

  - 08/03 :         Move DVCS electronics to Hall A.

  - 08/03-09 :    * Plug-in DVCS DAQ to Hall A DAQ with Bob's help.
                  * Detector side un-cabling in clean room.

  - 08/09 :         Move DVCS detectors in Hall A.

  - 08/10-09/03 : * Unstack calorimeter, repair, stack again.
                  * Standalone DAQ tests.
                  * Pull RG213 from spectrometer hutt down to pivot.
                  * Slow control + HV cabling (decide where to put it).
                  * Patch Panels installation. Plug in RG213 and DAQ side.
                  * Write security documents.
                  * DVCS detector on platform on 09/03.

  - 09/04-09/16 : * Plug-in cables on the detector (RG213, etc).
                  * DAQ timing with Bob Michael.
                  * Detectors check out (about 8-10 days of tests w/ DAQ).

  Also discussed a few minor issues with Jack.

  -> Charles and Gagik/Proton-Array:

  Charles presented a possibility to actually look for elastic protons
into the proton array while doing the spectrometer check-out with HRSleft
(e-) at 20 degrees. To illuminate two towers, one can do different
settings moving back the DVCS stack by a total of about 80cm at the
furthest setting. We all agreed it would be beneficial to do such a quick
test while doing the HRSleft check-out.

  Gagik showed latest results on the P-A. Everything looks okay except one
block (badly located) which appears to be dead (most likely PMT). This
will be investigated when the DVCS detector is in the Hall.

  -> Eric/VETO:

  The HV short was investigated to come from patch panel. Electronics
group will help repair the electronic boards of the patch panel. Should be
done before end of next week.

  Other than that, everything looks fine.

  -> Carlos/Calorimeter

  Showed new results about block positioning (see ELog for details).
Determination of this block center is accurate to within 1mm2, but after
investigation, it seems that the amplitude dependency versus a move of
around 1mm is 5% (which remains to be understood). The LED method seems to
have a systematic error at least that size just because of positioning
error. With new position from scans (vastly improved the X-Y matrix of
block position), the discrepancy between last cosmics gain-matching and a
new LED run goes down to about 9% which is the record so far. A vote of
what method would be closer to the real electron or photon events during
the experiment ended up largely in favor of the LED method, we will see :)
Work on the broken blocks (broken solderings actually) will start as soon
as tuesday next week. Kathy has agreed to setup a table to work on this in
the clean room.

  -> Kathy/Slow control

  Kathy figured out the right setup for the VME scan ADCs and tested them.
Everything looks good. P-A GUI is there, Calorimeter GUI is there, just
miss the Veto GUI. The Veto guyz will interact with Kathy so that the GUI
is done. Good news for Kathy and bad news for us, Kathy got a job in
Washington State! Congratulations Kathy !!!

  -> Sirish/TSOP and misc items

  Standard operating procedures need to be written shortly. Sirish will
start and spokespersons will fill in the blanks.

  A survey meeting is absolutely necessary. We will try to schedule it as
soon as Pierre is back (August 5th).

  Some money remains to be spent, if you have items, let him know...

  Present: Charles, Eric, Catherine, Alex, David, Carlos, Gagik,
Jean-Marc, Matthieu, Malek, Astrid, Franck

  Malek and Jean-Marc/Tagger:

  All paddles have been tested with cosmics. Now will come back to paddles
which were tested "bad" during first try. This was DAQ problem though.
Analysis and decode classes are finished. Testing in progress. Next step
is comparison of results from gain match with what was done at Grenoble
for the 11 different shapes. If time permits, plan to redo everything.

  Charles and Gagik/Proton Array:

  Currently taking cosmic ray data for 4 towers at a time. The problem
with one tower (which was supposed to be a short or something) turned out
to be a cross-wire problem, so no disassembly was necessary. Comparison
with absolute gain calibration in ODU turns out rather badly. Geometry is
very different however (shape of blocks does matter for cosmics but not
for a LED). Monte-Carlo comparison will be coming soon. The LED pulser
board needs to feed power to 20 LED boards. It seems that the current
drawn is very high (4Amps total) which makes the VME power supply sag...
Solution is probably to plug in an external power supplu for that.
Fernando Barbosa is giving them some limited support on that. DC
monitoring has also issues, Kathy is helping to sort this out.

  Catherine and Carlos/Calorimeter:

  Carlos reports on his study to get find block centers. Most information
is in the Elog. Important things: position IS different depending which
LED is lit or which combination of bias for several lit LEDs. However,
position does not seem to depend on bias (good news). Also a good news,
the weighted average of the position for LED1 and LED2 and LED3 turns out
to be the same as the position of LED1+2+3 (as it should be). Conclusion
of this study by the end of the week. Catherine is writing final scripts
for getting number of photoelectrons, gains, doing linearity checks and so
on, so that it is fully automated when data taking starts. Most light
leaks have been blocked now thanks to Matthieu Jean-Marc and Alex.

  David/Spectrometer:

  David is working on the script to produce the HANA tree. It will be
available shortly. Gagik made a point that we should freeze (for now),
HANA's version so that everyone uses the same. Carlos will investigate
which version is most suited, most likely the last one. Also, Carlos will
write merging and Spectrometer classes in DVCS package once David has
given him the HANA output tree.

  Alex and Franck/DAQ:

  The change of VME CPU solved the DAQ hanging problem from last week.
Weird "empty" events noticed by Malek turns out to be a real bug for the
proton array/tagger trigger. Michel Brossard is working on that and will
fix this issue. Note that those events are only there when this trigger is
used as a real trigger. The calorimeter trigger doesn't have this issue.
Last few ARS modules will arrive next week. Our DAQ computer dvcsdaq had a
disk failure on 07/20. Had to change the disk and recover the data. Now
done. Zero-suppression mode for P-A is underway. Needs some more work,
especially getting the offsets right, and full blown test. DAQ GUI
programming underway (for now only for "newled" configuration). For now,
prestart and end run submit Elog entries with summary file. Need to make
entries in database as well. Will work on that now.

  Everybody/Run Plan:

  Draft run plan can be found at the following URL:

  http://www.jlab.org/~sabatie/stuff/runplan.ps

  Spectrometer check out will happen first. This will most likely be done
at lowest possible angle (19 degrees) where elastic scattering is still
possible (at the very limit of the momentum capabilities of the HRS left
and the a few ten Hz rates). It might actually be possible to put
protons in the proton-array in that configuration, but most likely needs
to pull back DVCS detectors a bit. Deep inelastic is also possible at low
angle with a few hundred Hz counting rates. So information will be given
to Ed Folts that Left HRS will be at 19 degrees, with negative polarity
and maximum momentum at start-up. Please verify run plan for the rest of
the settings. Colimator box issue: in or no? if we might have to use it,
knowing that we can put it in the "no collimator" setting, we should
require it + survey. Right HRS at 70 degrees. Polarity not chosen yet.

  For target issues: will make a push to get at least one beer can for
hydrogen, machined cell is okay for LD2, if we can't have the beer can for
that as well.

  Present: Bob, Charles, Eric, Malek, Pierre, Jean-Marc, Alex, Catherine,
Astrid, Gagik, David, Xiachao, Sirish

  Alex/DAQ:

  - Been working on PA/tagger crate lately.
  - Issue pending: crate is hanging after around 20 to 30k events, waiting
for an inhibit signal. More work needed to solve problem.
  - Bob Michael sent us a list with some propositions. Important
information: we will have only one trigger supervisor (left arm), even for
right spectrometer. We would like to have best S2 inter-paddle timing as
possible. Rob can help us take care of this.
  - General issues about DAQ and commissioning: trigger gate width
(40-60ns at the moment)? Pulse location in the gate? How many ARS channels
to read? whole 128ns? less? The possibility to record less channels should
be implemented and tested before run starts. Position of the pulse in the
ARS sample?

  Gagik and Charles/PA:

  - Take cosmic events. Issues with max number of events (see above). They
require MC file with "vertex" information for cosmics to simulate PA
paddles and trigger. Gagik presented a pulse parametrization which works
well for PA. LED for PA is not working too well, several issues to solve
(problem with LED pulse board?). Should start thinking (and implementing
DAQ-wise) strategy for data taking (and testing during the commissioning
period): 0-suppression mode, DVCS mode (read only PA blocks where proton
can land assuming it is a DVCS event).

  Malek/tagger:

  - taking cosmics, helped by Alex. 20 paddles have been tested. Can only
test a few because of cable issues. Currently working on DVCS software.

  Catherine and Astrid/Calo:

  - LED with new script (all blocks at once). Same results as last time.
  - Astrid showed that there was a 16% spread in photo-electron yield
using information from cosmic runs and LED. This needs some more work.
More news next week.

  General:

  - we will soon start a Elog thread on items that are missing (mostly
cables at the moment). Everybody should add to it so we know where we stand.
 

attendance:

P.Y. Bertin, C.E. Hyde-Wright, G. Gavalian, S. Nanda,
A. Camsonne, C. Ferdi, A. Chantelauze, M. Mazouz



Reports:

P. Bertin,

HV Cables for Tagger on order, not yet delivered
Need 60 lemo to BNC adapters for Tagger
Added note to elog regarding preference for 15cm target cells.
Need Beam-Line expert for experiment (Charge, BPM, Raster, Moller Beam-
energy)
Target will have two loops (one H2, one D2, no spares).  Each loop has
5cm and 15cm cell.  Target will also have standard solid targets.



C.Hyde-Wright

Working on ODU grant renewal.
Refurbished two scintillators (and support)for Proton Array Cosmic-ray 
coincidence trigger.
Delivered monitor, keyboard, mouse for proton array DAQ computer in EEL
Delivered prototype PA block for Tagger Cosmic trigger (Hypertronix--SHV 
adapter also available).
LED pulser boards are functional, but final design switch to biasing led 
cathode rather than anode complicates the use.  LEDs can only be turned 
off reliably by setting bias to 5V (DAC=FFFhex)  This also requires 
adjusting the potentiometers on the PA LED boards.
One LED board and one low voltage board appear to be shorted 
(independently).



G. Gavalian:

Cosmic ray runs completed for two proton array towers.
Porting Calo ARS decoding software to Proton Array.
Will add spare cables to tray (HV and signal).
Working on PA HV controls



S. Nanda:

Compton Polarimeter will be a separate responsibility from other beamline 
issues during experiment.
Compton Polarimeter will be an excellent Halo monitor, together with
SLI (Synchrotorn light Interferometer).
Will purchase ADC monitor cables for Tagger, if we give specs.



M. Mazouz:

All channels of Tagger tested on scope, ok.
Broke 4 PMT in installation, 9 spares remaining.
Need a 1A +/-5 V power supply.



A. Camsonne

Fixed a bug in ARS addressing.
Calibrated Proton Array, Calorimeter, timing in cosmic ray setup.
Will create separate CODA DAQ for Calorimeter cosmic ray running and
PRoton ARray+Tagger running.
Assisted with Proton Array DAQ



C. Ferdi

Presented results on Cosmic ray and LED calibration of Calorimeter (see 
Elog).
Mean signal from Cosmic rays is 40 photoelectrons, with a dispersion block 
to block of 2 pe (except four blocks known to be poorly assembled).



A. Chantelauze.

Wrote report on internship, will defend by videoconference tomorrow.





Next Meeting, TUESDAY, July 13, after the Hall A meeting.

Minutes of E00-110 DVCS meeting 6/30/04
Proton Veto (Grenoble)
A-Layer assembled
B-layer started
PMT amplifier bandwidth was reduced to 100MHz to eliminate
oscillation.
50 Ohm output impedance added also,
Will try to restore output impedance to 10 Ohm
  (10 channels tested 7/2)
C.H.-W. agreed to provide paddle for Cosmic Ray
trigger for tagger (done 7/2).

Proton Array:
two Cosmic ray trigger paddles and mounting brackets now
available.
Tagger shielding/mounting cone conflicts with 1/2"Al
shielding plates for forward towers.  Shielding plates
will be trimmed.
Login on/off for DC LED does not function.
Potentiometers on LED pulser boards must be adjusted to
allow 1) maximum light output for pulse LED 2) Turn off
all LED by setting voltage DAC to FFF(Hex).
Sirish has promised to purchase a high quality monitor for
Proton Array DAQ in EEL (to be reclaimed later for
Compton Polarimeter).

Alex Camsonne:
Working on EPICS control of XY table
Working on Proton Array DAQ

Bob Michaels will be on vacation Aug 10-30.

General Discussion:

- Tagger soon received in few days and guys from Grenoble coming next week
- Need organisation and device (oscilloscope, computer, coda...) in the EEL
for working (Tagger people, Proton array people and Calorimeter people)
- Dismount Calorimeter before moving it in the Hall A and mounting it in the
Hall A (It takes one week, no modification in the schedule)
- Bob Michaels in vacation after Happex and until 23 or 25/08 (Bodo could
replace him)
- 50 % beam time in addition ( all calculations done with 70% polarisation
and now for Happex 86% polarisation). Don't tell this to Kees and Larry...
- Need meeting with accelerator's guys after Happex (Sirish has asked the
maximum current we need which is linked to the efficiency machine: 5micro
for proton and 6 micro for neutron)
- Need one guy in charge of the luminosity monitor during the experiment


Gagik has presented some results about the calibration of the left
spectrometer (other elog report), Alexander about the DAQ (in attachment)
and Astrid and me about the LED studies (other elog report). 

Catherine

  Presents: G. Gavalian, D. Hayes, M. Mazouz, C. Ferdi; A. Chantelauze, A.
Camsonne, P. Bertin, F. Sabatie, R. Gilman

  Minutes:

  -- M. Mazouz: showed pictures of the tagger fully assembled in Grenoble.
It will be sent to Jlab before this weekend and will be here next week.
Malek also talked about MC results for the tagger. New simulation with new
mechanical description of tagger system shows that rates are everywhere
less than 1MHz at 4MeV threshold except for one paddle. Malek suspects
that positionning of side plates creates an unrealistic top-bottom
asymmetry and will investigate this. Malek also showed results
of MC wrt what paddles are hit when a certain proton array block is hit
(by a proton). He will put this into correspondance table so that it can
be used in the trigger if necessary, or for quick offline analysis.

  -- Astrid: showed updated results on the LED monitoring. Removing one of
the collimator's plate increases the number of photoelectrons by a factor
10. A new light guide is currently under fabrication, which should improve
light collection and maybe allow one to use better collimator (with 2
plates). Cross-talk, which was a problem in the first system studied last
year, seems to be completely gone now (good news!). PMT linearity was
studied as well. Comparing the mean of the integral for two LEDs and then
for the sum of the two, linearity was determined to be of order 1.4%. This
result was however not recovered when comparing photo-electron yields.
This remains to be understood. Lastly, Z-dependence was checked to be of
the order 1% per mm, which is okay to work with since the estimated z
spread of blocks on the face of calorimeter should be within these specs.

  -- David: Cabling of trigger and ARS just started for Proton-Array and
Tagger. Should be finished next week.

  -- Gagik: aggreed to check on the spectrometer setting for check-out so
that we can send a complete setting to Ed Folts asap. He will also check
the mounting of external paddles for cosmics with the proton array.

  Present: P. Bertin, C.E. Hyde-Wright, G. Gavalian, K. McCormick, M.
Mazouz, D. Hayes, A. Camsonne, C. Ferdi, A. Chantelauze, F. Sabatie

  Franck: Discussed the installation meeting with everybody in more
details. The installation schedule is mostly constrained by the first two
weeks while we wait for septa to be desinstalled. There will only be 7-10
days to check-out the detectors in the final configuration before Hall
locked period (starting Sept. 10th). This Hall locked period is still
unclear. We will probably be able to access the Hall in controlled access
if needed.

  David: showed the PA/tagger cabling scheme (trigger and ARS). Charles
and David payed attention to not get PA blocks and corresponding tagger
paddles in the same ARS module because of cross-talk. Attention was not
paid particularly to block-transfer issues since 0-suppression mode will
be chosen and will probably sparcify the data. David used a reversed
numbering scheme as we used in the Monte-Carlo. We decided to use MC
scheme so he will update his slides and make them available.

  Astrid: Details about LED system check-out using charles' light guide.
It turns out the system is very dependent on the alignment and very little
photons make it to the block. More details are available in Astrid's
recent ELOG entry.

  Alex: worked with M. Brossard on the new logic unit. Now operationnal.
Many scalers can be programmed in this unit (it contains a FPGA). Still a
few things to tune in DAQ. New trigger operationnal.

  Gagik: Worked on software a bit. Found a way to simplify library
structure. Will check in changes in CVS.

  Present: Ed Folts, Jack Segal, Sirish Nanda, Al Gavalya, Pierre Bertin,
Franck Sabatie.

  Minutes:

  Installation starts Aug 2nd (monday).

  Aug 2-8

    In the Hall:

        - Left and right HRS detector work (Jack)
        - FPP Rack dismount (Jack)
        - DVCS Patch Panel mount (Jack)

    In the EEL:

        - Detector uncabling (RG213) (us)
        - Clean up, ready up for moving cables, then detectors (us)

  Aug 9-15

        - Install DVCS Racks in left HRS hut
        - String one 25m (XY-table) cable to check where to put
        all slow control racks and HV. Then pick the location and
        install all that.
        - Ed can move the left HRS a bit away from beam line so that
        we can start stringing RG213.
        - At the end of the week, Septa are out.

  Aug 16-22

        - Install BigBite stand (some welding, machining, survey necessary)
        - Start install of scattering chamber.
        - DVCS team works on acquisition.

  Aug 23-29

        - Bring DVCS stand in.
        - Put DVCS stand on Big Bite stand (requires some welding,
        machining, survey, etc)

  Aug 30-Sept 6

        - Cabling (RG213, slow control, HV, ...) starts when DVCS stand is
        on place.

  Tests can begin with full detector as soon as cabling finished (which
should be fast. Should leave about 1week for tests before hall locked
period (Sept 10-20).

  Additional notes:

        - Large Cerenkov detector will be in left HRS.
        - Pion rejector might have bad blocks, needs to be checked with
        Eugene.

---  Charles:

  18 towers are installed, 2 last will be ready by the end of this week,
installed next week. All cables are in. Extension cables are ordered.

  ADC for DC monitoring still not working. Kathy and Michel will look into
this friday.

  LV power supply built, will be tested soon.

  David is looking into a scheme for exclusivity analysis w/o photon
energy (only angles), will be discussed next week.

---  Gagik:

  HRS Calibration tech. note is available on the web page. Working on PA
database.

---  Alex:

  New trigger installed, cables plugged. Everything works. MUX box
installed. Cosmic trigger with external paddles currently under scrutiny.

  PA trigger installed, will be cabled by PA team next week.

--- Malek:

  Implemented tagger geometry, with associated shielding and mechanical
support. Rates dropped compared to before, all under 10MHz. Final results
next week.

--- Pierre:

  Meeting with survey guy to talk about the report they produced from
first survey. Turns out that we have the 3 euler angles of the calorimeter
wrt the black box well described but the location of the calorimeter
itself is not clear. It will have to be re-surveyed. Pierre suggests
videotaping next survey to know what is being done!

--- General discussion on PA trigger cabling scheme

  The discussion was long and I will only report the conclusion: it would
be great to have 2 GEn bars (10x10x160 cm3) on top and bottom of the
proton array for cosmic ray calibration. Charles will ask Bogdan about
this. In any case, there is probably a smart way of cabling the ARS, the
trigger and so on. Gagik and Charles will think about this for next week,
before the cabling starts.


  Don't forget next meeting: wednesday june 9th, 3pm, A216

  Sincerely,
  Franck

Two different kinds of analysis will need to be done on spectrometer data:

1. Single arm analysis and calibration.

   The purpose being to calibrate and monitor detectors in the
spectrometer, we need to look at and keep many variables : all ADC/TDC
values of VDCs, S1, S2, Cerenkov, Pion Rejector... (no TDC is available in
pion rejector).

2. Coincidence analysis.

   Only a few variables per detector will be enough:
     - S1, S2 and Cerenkov sums
     - Pion rejector 1 and 2 sums
     - S2 X,Y,theta,phi
     - PID, quality of track

   In addition to that, useful physics variable should be kept :
     - delta,theta,Y,phi at the target (vertex)
     - t_vertex
     - Px,Py,Pz, X,Y,Z at the focal plane
     - Px,Py,Pz in Hall coordinates
     - X, Y, theta, phi at the focal plane (to doublecheck)
     - Event number (CODA)
     - Event type (electron,pion,...)
     - Helicity
     - X,Y,Z of BPM
     - S2 tracking (optics_0) (timewalk)
     - Lots of scaler values... (to be defined!)

Several important issues were raised during the meeting:

- Full electron PID tuning with single arm is sufficient. All calibrations
of spectrometer need only single arm except DVCS
detectors-spectrometer coincidence timing! 

- S2 timing done zith cosmics which cross 2 paddles at the same time.
Small cables can be added to correct for timing spread.

- Status of calibrations : which detectors be calibrated after HAPPEx ?

- Timing issues (a dedicated discussion on the DVCS trigger might be
  needed). Bob and Bodo seem to check time at the scope with cosmics. No
software is necessary.

- The right spectrometer may be used as a luminosity monitor. The minimal
  angle where we can go needs to be investigated.

- Necessity of a sieve slits calibration run? Effect of the pointing
accuracy of the spectrometer in the experiment resolution/acceptance. This
needs to be investigated (by who?). A lower energy might be needed (time
issue), a proton array elastic calibration run might take place if a lower
energy is available.

- Necessity of collimator in spectrometer ?

- What targets that will be available to us ?

- Macros for VDC, S1 and S2 efficiency are not "well centralized" and
  not easily available.

- Some interesting variables of HANA seem to be private and
  inaccesible. Requests should be made to the HANA developpers team to
  write the appropiate methods. An example of these variables are the chi2
  and error of track fits.

- We will need to tag a certain version of the analyzer soon (with a
version of ROOT to go with it).

- Automatic splitting of files seem to be ok with everybody. A SILO
  file size limitation exists (2GB). File size for subsequent analysis
should in principle not increase the CODA file size.

- No delayed helicity scheme (to be checked with HAPPEx2).

- Cerenkov will be there during HAPPEx, but after discussion with HAPPEx
  guys, trying to calibrate during HAPPEx is out of the question (VDCs for
  example won't be there).

The mainstream analysis strategy seems to favor the stand-alone + merging
procedure. A merging script which matches event number in both independent
trees needs to be written.

David Hayes and Gagik Gavalian will take care of the spectrometer
analysis.

In next week's collaboration meeting a list of final variables
will be available.

Everyone agreed to have weekly meetings from now on. Wednesdays at 2pm
seems to be a good time, starting in two weeks (after the collaboration
meeting).