Operators: vziskin zwart karpiusp
7AM Hall opened to fill snakes. We had some intermittent problem with the Ion polarimeter so no atomic fraction was measured. AF should be measured after snake fill.
At 8AM Pete Binns notices no gas flow into ABS. He turned off RF and is checking bottles.
Prior to 6:30 AM running was very smooth and we had time for some analysis.
Looking at the difference in the asymmetries for the left sector and the right sector we observed the following:
1) The distributions for the reconstructed proton vertex and the reconstructed electron vertex do not agree. For electrons on the right side the centroid of the reconstructed proton distribution is 0.45 cm greater than than the electron vertex. Further there is a tail in the distribution where the proton vertex is ~10 cm downstream of the electron vertex. This could be interpreted as pointing to a background source ~5 cm beam right.
For electrons on the left side the centroid of the reconstructed proton vertex disagree by 1.7 cm, the proton vertex is downstream of the electron vertex. However for electrons on the left there is no tail in distribution of the difference between electron and proton vertices.
We added these offsets to cuts.C. If this is not done the constraint of vertex agreement will cause the asym for electrons on the left vs electrons on the right to sample different longitudinal sections of the target.
Attached are three plots that show the left and right asymmetry for three different sections of the target. Upstream.ps is cut on -5<z<-15, center10cm.ps is cut on -10<z<10, and downstream is cut on 5<10<15 cm. The center +/- 10 cm cut is tighter than the 15 cm used previously. Also there is pretty good agreement left and right for the low theta point. The plots for the upstream and downstream section seem to indicate lower target polarization or greater dilution as expected.
TZ VZ PK
Vitaly suggested the possibility of flipping the the direction of the longitudinal holding field (if it is not possible to flip the target with rf transitions). Seems like this would be an nice approach to understanding another class of false asymmetries.
This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST