Hello, long email, there are conclusions at the end.
First I want to make one important correction. I recently discovered that
in a previous montecarlo study of the q2 acceptance the formula
4.*E0*Pp*sin(theta/2.)**2 was used (Pp hadron momentum) which brings a
systematically lower value of Q2 (by a factor about 0.3-0.5)
This generated the erroneous perception that our data coul extend down to
Q2 of 0.05 and therefore scattering angles as small as 15 deg were possible.
This is now corrected for ep, ed elastic (see first two plots). I also
attached the original wrong plot for comparison. Again, it is not an
intrinsic problem of the montecarlo, but simply of the user macro.
I then did a little study of the electron angle acceptance using the blast
montecarlo. Although in a sector TOF bar #0 is at a nominal angle of 19
deg, this is not the same acceptance covered by the 6 wch super-layers.
The montecarlo generates elastic events over a broad angular range (from
15 degrees on). Depending onhow this data is analyzed generated events
can result in singles or coincs.
Then I get the following results (field bending referred to e-) for the
minimum electron angle in deg, for ep scattering
Polarity theta (coincs) theta (singles)
Inbending 25.9 24.8
Outbending 20.8 20.8
In going from 25.9 to 20.8 (the effect of reversing the blast polarity)
Q2 went from 0.133 to 0.09, and the ep cross section goes from 1.6 ub/sr
to 4.5 ub/sr. So we won a lot of cross section.
For inbending electrons we loose some acceptance compared to data analyzed
requiring one track or MC singles. This may be because of protons being
bent "out" of the last tof paddle. Given other measured and blastmc
results, this effect seems not to be due to energy loss, the proton energy
ihere is very high (70 MeV)
For outbending electrons the acceptance is determined by the minimum
electron angle for both polarities. Of course the proton energy is now
lower and - in the data - detection thesholds could be a problem.
Note that the detection threshold for montecarlo is infinetely small
However this also does not seem to be our problem, since with reverse
field we reconstruct many events around 20deg. If anything, for ep
elastic, we have too much low-Q2 data than too litte.
This is shown in the last attachment, where the vertical line is the
montecarlo prediction.
Conclusions
-----------
_ The minimum expected angle for ep elastic scattering is 20.9 deg for
singles and coincs, and is determined by the BLAST acceptance at forward
angles
_ For non-bending particles (infinite momentum or neutrals) this would correspond
to tof paddle #1 (from 0). Indeed recent data shows that our elastically scattered
electrons start to show up above paddle #2, and in paddle #3.
_ We clearly have data in that region, where the proton energy at the target is
48 MeV which is well above the punchthrough energy in BLAST (this is 40 MeV for
D2, and about 20 MeV for protons)
_ TOF thresholds are therefore not a particular issue for elastic scattering at these
angles
_ There was a confusion about the Q2 acceptance and minimum angle in the
acceptance. It seems to be coming all from a stupid error.
_ We seem to systematically mis-reconstruct some data at forward angles, but this is
not inconsistent with the poor accuracy in reconstruction either (sigma 1.6 deg)
-- ________________________________________________________________________________ Tancredi Botto, phone: +1-617-253-9204 mobile: +1-978-490-4124 research scientist MIT/Bates, 21 Manning Av Middleton MA, 01949 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST