Operators: tancredi ajmasch tong
Smooth data taking empty target. Experiment needs little attention, compton
also doing fine. Wch R11 still holds no hv (another broken wire?). Made runlist.outbend
Analysis of reverse field data:
the inbending particles are attributed negative charge, obviously this is a
flipped sign. Acceptance is extended to ca 15 deg. Most of the ++ background
is gone. Fraction of elastic events/events with a track went from a few to ca 30 %.
One first problem is the resolution (see next 4 attach) which has deteriorated as seen from elastic H run. Also the zL-zR distribution is worse. These problems
do not seem related to mutltiple scattering of protons for forward angle
scattering, since they stay for a cut on the e- angle. After consulting with cc2 we realize that there is a redundancy in the definition of the sign of the B field (look for his email). So chances are that tracking used the wrong field direction. Will test crunch the new H data today.
Finally the last problem is a sizeable L/R ambiguity (last but one attachament,
with angular distribution of elastic events in the Red/Right and bLue/Left sectors).
We seem to have poor recon efficiency at forward angles in the red sector
which makes us lose (backward as they always are) protons in the blue sector.
This could be confirmed by comparing spectra that used only the tracking info
from a given sector (i.e. no coplanarity, no qwodd but use CC, even if "efficiency"
is different). Problem seems to be there also with fast fit. We have seen similar
things back in june and were related to a bad wch box: there are no obvious bad
boxes for these data.
Neglect the last attachment, I simply could not delete it. Did not progress much
with LADS code: I did update the library though, so it is a matter of testing
the getters with some low level stuff like ntuple.C.
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:30 EST