Blast analysis meeting on Thursday 10/19/2006

From: Michael Kohl (kohlm@mit.edu)
Date: Tue Oct 17 2006 - 13:59:15 EDT


Hi,

we're meeting this coming Thursday 10/19/2006 at 9am at Bates.
See the previous minutes of last time below.

If you can't attend, call in at
Bates conference call +1-866-867-8301, passcode 4073393

Preliminary agenda:
-p(e,e') inclusive analysis (Tavi)
-d(e,e'n) followup (Eugene, Vitaliy)
-aob.

Meeting minutes of 10/05/2006:
==============================
-Status and quality of v3_4_21 calib and recrunch (MK)
  +See ppt
  +Quality of recrunch and calib is very good, the time-dependency of
   the offsets is generally well under control. Few spots with "problems"
   remain. TOF calib is good <1ns (absolute) for entire BLAST dataset.
  +Recrunch v3_4_21/lrd complete for 2004 deuterium (7002-12143), 2005
   deuterium (13278-16506) and 2004 hydrogen (12144-13278)
  +Looking at three time correlations, for (e,e'p), (e,e'n) and
   (e,e'gamma) events. If the calibration is correct, the quantities
   dT=T-Tcalc are expected to peak at zero, where T is the
   time-of-flight for the proton, neutron, gamma, respectively, and
   Tcalc the expected time-of-flight from tracking information. The
   widths of dT result from convolution of uncertainties in T, Tcalc.
  +Proton peaks for 16x16=256 TOF paddle combinations
  +Neutron, photon peaks for 2x(8+14+14)=72 neutron detector channels
  +Proton peaks used to check TOF-TOF calib, neutron+photon peaks to check
   TOF-NeutronDet calib
  +e'p peaks very good and consistent for entire dataset,
   broken down into channels, deviations of peak positions up to
   1-2 ns, total mean of ~0.4ns, sigma<1ns.
  +Photonpeaks:
   L15_L,NC_L jumps in certain channels: at certain run numbers related to
   validity ranges of cosmics calib for flasher monitor. Effect on total
   yield rather small, not clear wether it is worth to spend the effort and
   fix it. Rather, the affected detector channels/run numbers should be cut
   out. I will send a list of runs and detector channels who have this
   problem.
   L20,NC_R: have double peaks, not correlated with time, cercut, adc
   ->could be misidentified e,e'n events in e,e'gamma channel
  +Neutronpeaks:
   Time dependence of tdc offsets well under control; however peak
   positions generally 3-5ns off zero. Could be due to wrong absolute
   offsets, wrongly assumed tdc slopes (ch/ps), slightly wrong
   pathlength (different pathlength of cosmics tracks and real neutral
   tracks in production runs; detector thickness: n-interaction inside
   detector material, whereas cosmic ray produces signal right at the
   surface after entering the detector.
   -> It's necessary to determine time peak position bar-by-bar for the
   clean e,e'n event sample. Apply shift as time correction for
   neutron time of flight, and/or length correction of the neutral
   track (as a scale-factor correction). This shift is then valid for a
   full dataset, not time-dependent; consider 2004 and 2005 separately.

-d(e,e'p):
  +See ppt
  +Comparison v19 data with MC: Cerenkov cut on/off, plot p_e for
   various Q2 bins. Good agreement at lowest Q2, but then increasing
   disagreement towards higher Q2 or theta_e. This is a potential
   candidate to explain the Q2 dependence of hPz. Remember hPz was
   left-right consistent only at the lowest Q2 bin.
  +Data-MC discrepancy in pe vs. Q2 has substructure (->nonlinear t2d?)
  +Agreement for proton momentum much better than for electron momentum
  +Discrepancy largest in third Q2 bin 0.3-0.4: Data has low-momentum
   excess tail for electrons and high-momentumn excess tail for protons
   which is not seen in MC and which is not explained by the Cerenkov
   acceptance
  +Apply EG's kinematic corrections and energy loss and compare again
  +Use same normalization factor for MC in each plot
  +Vary acceptance cut for phi (+-12deg) and/or z (+-15cm)
  +Look at location of Mmiss peak (supposed to be delta-function like)
   instead of maximum of p_e (is continuous)
  +Break down into tof bars and cerenkov boxes
  +comparison data-MC for pmiss components ->crucial for asymmetry
   evaluation

-d(e,e'n):
  +See ppt
  +Comparison data-MC for 2004 (v18)(!) and 2005 (v21)
   -> need to update 2004 part, then compare 2004 v18/v21 and for
   v21 2004/2005
  +Excess of electron momenta around 0.6GeV/c in MC, not in data
  +pmiss tail in 2005 longer than in 2004 -> =more efficiency for
   high-pmiss neutron tracking in 2005 with converters and better calib?
  +neutron solid angle acceptance data-MC agrees well, some edge effects
  +pmiss components: discrepancy in pmiss_par and pmiss_perp,
   good agreement in pmiss_oop; theta_cms wider in data than in MC
  +third Q2bin: yield in 2005 much lower compared to 2004
   ->inefficiency in left-sector electron-TOF ("dip")??
  +L15 hits with angle >46deg -> Problem with neutral track assignment?
  +timecorr: wrong offset(additive) or wrong Lnn (multiplicative)
   -> time correction to be done bar-by-bar, not as a function of Q2
  +MC: Geant-based+recon?
  +Adt,AedV vs reaction effects and vs. pot., vs. pm,thcms

+-------------------------------------+--------------------------+
| Office: | Home: |
|-------------------------------------|--------------------------|
| Dr. Michael Kohl | Michael Kohl |
| Laboratory for Nuclear Science | 5 Ibbetson Street |
| MIT-Bates Linear Accelerator Center | Somerville, MA 02143 |
| Middleton, MA 01949 | U.S.A. |
| U.S.A. | |
| - - - - - - - - - - - - | - - - - - - - - -|
| Email: kohlm@mit.edu | K.Michael.Kohl@gmx.de |
| Work: +1-617-253-9207 | Home: +1-617-629-3147 |
| Fax: +1-617-253-9599 | Mobile: +1-978-580-4190 |
| http://blast.lns.mit.edu | |
+-------------------------------------+--------------------------+



This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:33 EST