Hi,
below are the minutes of the analysis meeting from Wednesday:
-d(e,e'p): kinematic corrections (AD)
+Shows for each Q2bin comparisons of measured and simulated electron
momentum. Differences are up to 2%, at the margin of the
resolution. Pattern of deviations vs. Q2 is sector-symmetric
+Note that definition of Q2 in q.e. scattering depends on two
variables.
+Shows comparison data/MC for quasifree (pm<0.15) events, expect
events to be close to the elastic ridge in a p_e-vs-th_e plot,
attribute deviation to either wrong th_e or p_e or both. Magnitude
of deviation rather points to momentum (would require several degs.)
+Parametrizes momentum corrections based on quasifree events
(pm<0.15) as scale factor with linear dependence in Q2, plots
measured minus expected electron momentum for entire q.e. event
sample after applying this correction. Finds remaining deviations of
~7-10 MeV.
+Comparison is new 2004 recrunch v3_4_17 with an older MC which uses
old geometry however we're looking at physical variables here
+New MC still has the problem of disappearing yield at small
angles. There seems to be a dependence on computer architecture
(=initialization problem?)
Suggestions:
+Should apply corrections from ep elastic (Eugene's) to see how well
they work
+New MC should make use of spin angle profile
+Adam: could you also evaluate the target yield-weighted average for
given spin angle maps, for the d(e,e'p) event sample (see
discussion on spin angle in separate email)
-Spin angle:
+No estimate yet of the error of the average spin angle extracted
from ep elastic asymmetries.
+Strong dependence on reconstruction likely explains large shift of
the spin angle recently reported
+Remember, in the hPz-vs-th_s plot, the two straights for the two
sectors are almost collinear, while the slopes of the two crossing
straights in the corresponding Pzz-vs-th_s plot from ed elastic
have opposite signs, making ed elastic much more robust in
determining the spin angle against changes in reconstruction.
+For extraction of form factor ratio, needs to account for profile in
extraction of observables, use "nominal" spin angle determined by
Chi (on "average" vs. "nominal" see separate message)
+Shows target z-distribution of ep elastic yield broken down in
Q2bins. Not all bins are distributed as naively expected, pointing
to residual reconstruction errors
+Shows the same with spin angle profile overlayed and the resulting
yield-weighted average for each Q2bin. The latter does not(!) vary
much (<0.2deg) around the yield-weighted average of the entire data
sample, but note that the z coordinate reconstructed from data may
not fully reflect reality.
+Suggestion: Does this negligible variation of yield-weighted
averages for various Q2bins also hold for MC-generated ep elastic
target-z yield distributions?
Best 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 - 14:07:33 EST