-Attending NM,AM,VZ,TL,BT,DH,TS,YX,JC,MK
-NM:
Showing d(e,e') omega spectra for e-right in comparison with MC, for
various Q2 bins and 2dim versus Q2.
Quasieleastic event selection:
0.5 < Q2/(2 m omega) < 1.5 (=1.5 sigma cut).
omega distribution deviates from MC, deviation larger at larger Q2.
Correcting electron momentum such that measured distribution
agrees with simulated distribution. This changes the selected
event sample, in particular at larger Q2.
Comparison of asymmetries before/after the correction: Better
agreement with MC after. Previously smaller asymmetries at higher
Q2 may be explained with larger amount of diluting background events
before correction.
-Shows sensitivity of left sector (spin-perpendicular) and right
sector (spin-parallel) asymmetries w.r.t. GMn and GEn. Both sectors
have sizable dependence on GMn. While in the parallel sector the
GMn^2 dependence seems to come from the numerator, in the
perpendicular case the denominator seemingly shows the variation,
supported by the direction of the asymmetry change with increasing GMn.
-Shows event selection for NDelta inclusive events. Shows beam-target
asymmetries which are sizable. Background contributions may be higher
than in quasielastic case.
AM:
-New analysis of hPz
old value 0.47, new value 0.42, very small statistical error
-Fit done over Q2 range up to 0.5(GeV/c)^2, in both old and new analysis
-New analysis includes:
+background correction
+ad-hoc polynomial correction for e and p momenta
+"commandline-limited-to-256-characters"-bug fixed (i.e. old analysis
used truncated runlist)
-Extracted hPz varies with Q2: 0.46...0.39 for Q2 of 0.15....0.45
-Also hPz from hydrogen has Q2 dependence as CC showed last week
-Is not the case in AS's analysis according to JC
-Suggested to check how this dependence changes before/after applying
kinematical corrections
-Vector-d(e,e'p) as a polarization monitor: Pm has to be small enough.
Q2 has to be sufficiently large in order to have nuclear effects not
dominate compared to PWIA. It also has to be small enough to provide the
statistics. Identify optimum Q2 interval to be used to determine hPz.
-Discussion about energy loss:
+kinematical correction partly due to unconsidered eloss
+should be implemented in reconstruction
+will be a function of pid, momentum (and angle)
+would affect (improve) tracking
+needs pid which however is not fully available at crunchtime
+remaining kinematical discrepancies can then be parameterized
+implementing eloss in reconstruction is only meaningful if also
retracking is done, otherwise it would be equivalent to only apply
ad-hoc corrections to existing track solutions.
+discrepancies in kinematical variables and asymmetries may be due to
systematic reconstruction errors, but may also be due to physics if
nuclear corrections in the deuteron or if nucleon form factors are
slightly wrong. Use hydrogen data to understand systematics of the
detector first.
+In simulating, AM finds that only a small fraction of eloss occurs
within the chamber gas as opposed to expectation
TS:
-Shows d(gamma,pn) events (positive charged track in one sector, neutral
track in the other).
-p-n opening angle versus reconstructed photon momentum q (from Tp+Tn),
-Comparison with Blast acceptance MC events: 2-body photodisintegration
events are clearly identifiable. Opening angle is close to 140 degrees
with little q dependence
-Most d(gamma,pn) events beween 100 and 400 MeV reconstructed photon
energy (beyond the delta the cros section dies out)
-Also pion photoproduction events (3-body phase space) are seen
distinctly, corresponding openening angle of ~100 degrees
-3body final state survives an anticoplanar cut whereas the 2body state
doesn't
-Huge number of low-energy (q<150MeV) (shower?) events, predominantly
small opening angle (can be separated)
-Make future use of constraining the reconstructed photon direction
-Wait for better neutron timing (recrunch!)
MK:
-Preparation of recrunch:
-Want to have "proper" DST from one BlastLib for all runs
-Present DSTs are inconsistent, in particular for older runs. BATs have
only been added after run 12663; some dst-related bugfixes in BlastLib
became active only by August/September
-Need ntuples with improved neutron timing, also for hydrogen runs
-Desirable to have Epics information in dst,lr,flr in order to correct
ABS-related issues (MFT problem during hydrogen running) but not crucial
right now. Issue of charge re-calculation after cutting out sequences of
data.
-Database being filled with time calibration parameters for runs > 11250
-Will start recrunch asap. Any improvements can be done in later
iterations, most of future improvements can then be done starting from
DST
-Diskspace: all analyzed data is on /net/data/9. Will not be enough
space to keep old original crunch, but will start recrunching in new
directory anyway until space becomes a problem
-D. Phillips (effective field theories) is giving a talk at Bates on
01/26/2005.
-No meeting on 01/05/2005, but an extended analysis meeting on 01/06/2005
prior to the Collaboration meeting on 01/07/2005. People should start to
get prepared for this in time.
Merry Christmas to everybody and a Happy New Year!
Regards,
Michael
--+-------------------------------------+--------------------------+ | 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 - 14:07:32 EST