There was an unanticipated consequence to the shift from the Hall C to the Hall A chopper slit on Friday day shift: the reference signal that the Musson box locks to at low beam currents shifted by 1/(1497 MHz) = 668 ps relative to the beam pick off signal used at high currents. The reference signal is the divide-by-16 YO signal we used last year. The change in chopper slits means that the beam arrival time has changed by +/- one cycle of 1497 MHz relative to the divide-by-16 YO. What does this mean? A comparison of time spectra taken at 1 uA and 40 uA between restoration of beam Friday day and correction of the problem Saturday swing should show a relative shift of 668 ps. Also a comparison of time spectra at 1 uA taken before and after the problem was corrected should show a similar time shift. Unfortunately, we had just taken some 1 uA TOF data when the problem was discovered. Those half hour worth of runs were retaken. We will have to readjust this again on Monday when our chopper slit is repaired.
It is also worth noting that there are still problems with the Musson box in the beam pickoff mode, problems that generate LTD errors.
Beam was restored to Hall C after the chopper slit switch at about 17:00. A deadtime studies program was undertaken with confusing results (g0log 70895). Measured NA CFD deadtimes were large and varied significantly as a function of detector and threshold. However, looking at proton pulses and the associated CFD output did not show any cases where a proton pulse arrived and the CFD did not fire. Kaz came in to do a RHWP/IA/PZT scan and set up the calibration parameters for beam feedback. Jianglai and Kaz tried using a version of the feedback that incorporated a 3x3 matrix to determine the correction but found it rapidly walked off the end of the PZTy correction range. Reverted to the 2x2 PZT with 1x1 IA correction algorithm and it ran smoothly (70975).
The shift was dedicated to taking long TE data runs on LH2 with beam feedback on. Again, the beam feedback results look quite promising (70975). At the end of the shift another RHWP/IA/PZT scan was performed and the measured slopes seemed rather stable (70940 (the IA slope measured by Kaz and Jianglai is incorrect in this log entry but I do not have access to the correct number right now)).
The day shift was dedicated to Moller systematics (71016). The plan included runs at 10 uA as opposed to the normal 1-2 uA and a scan of the effect of laser phase on polarization. Preliminary analysis indicates little to no dependence of polarization on any of the parameters varied today. It was found that the Moller solenoid did do a little steering of the beam (70984).
No beam due to energy spread minimization program in Hall A. John Musson is supposed to have a firmware upgrade for his namesake box. Various other activities including GMS checkout and DNL runs are scheduled. May put the flyswatter in to take some one hour runs on LH2+FS when beam comes back.
No beam due to chopper slit repairs. I'm sure we'll find something to do. Tentatively, put flyswatter in and set FSD to tungsten radiator position. We would then start up taking the 10 hrs of asymmetry data on radiator+FS that the background working group wants. This data plus 30 hours on radiator alone and 25 on LH2 alone would be used to decide what configuration we should run in in January to nail down our background asymmetry. Should we wait on this until problems with the Musson box are nailed down better?