Willing to put up a fight, and appreciating the difference between
measuring something consistent with zero against the odds of having a 100%
unpolarized target (also, the 100% molecules scenario is rather
unsatisfying) I went back to the "final" results before calling them
final. Also, since only the WFT unit switches on/off in the target, we are
likely to retain at least the polarization of state1, were the WFT and
thus state3 completely wrong.
The first two attachments are an asymmetry, in the L and R blast sector
and as a function of scatterd electron angle, consistent with a product of
Pe*Pt = 0.05. Being generous (60% beam polarization) this is consistent
with an 8% vector polarized H target. Of course the actual nuclear
polarization of atoms is higher if we allow for molecular dilution. In
particular, the number above corresponds - by eye- to a decent confidence
level in terms of X2.
The are two main differences from results shown recently:
_ I allowed for a sign error in either the helicity or the target. In
other words I allowed that one of the spin directions is attributed
wrong. Indeed this may be for the beam helicity, it has to be checked
out against compton and "unfortunately" we most often flip the beam
helicity during a run. I hope we have compton for runs 564-567 but I'd
not know how to replay them (...). We are sure about the holding field
direction (beam left) and the sign of the target state.
_ I ventured in a cerenkov cut, assuming of course it is not spin
dependent, and you can see I have a rather clean sample in the second
plot (the correlation of L/R sector angles for elastic events in either
sector). Note that the cerenkovs are still not in their optimal setting.
however now they are shielded fromthe B field. Note also that we may not
be critically dependent on the CC only, since with better reconstruction
we could apply cuts on kinematically overdetermined elastic scattering
Other comments:
_ our small measured value and our "false asymmetries" conspire for a
large (relative) systematic error
_ My definition of false asymmetries (see yesterday) is not very solid,
since such "asymmetries" are of type (0 - 0)/(0 + 0) which can become
uncontrollable. But nevertherless they are small.
_ We simply have no sufficient data (only 4 runs sunday 27) for an
analysis of the asymmetry for runs with fixed spin directions.
- hope we get beam soon! tomorrow I'll be reachable by email.
grinning,
-tancredi
-- ________________________________________________________________________________ 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 - 14:07:29 EST