AGS s CNI Polarimeter C 15 Kin Yip Collider-Accelerator

AGS’s CNI Polarimeter @C 15 Kin Yip Collider-Accelerator Department July 31, 2009

There were ~3750 polarization measurements in 2009. Each measurement nominally has ~40 M events ~2. 6% errors (typically about 270 Mbytes). There were also thousands of so -called “emittance measurements”. But really they were used as a quicker tool (compared to the so-called “Target Scan”) to find the target/beam position. The total size of files for this run stored in HPSS~900 GB.

These measurements were made by the AGS machine physicists to try to study and maximize the AGS polarizations with respect to all sorts of machine parameter changes. This happened all the time, nonstop and actually for many years, whenever it’s not impossible. If there is any suspicion, the issue was dealt with ~immediately whenever possible. Only features/problems that I can recall : Rev. tick (i. e. bunch-zero) signal from the RF group has larger width than last year; Malfunction of one of our two frequency mulitipliers the solution was to swap it with a spare one and afterwards, we have made new ones/spares.

Offline calibration (t 0 and dead-layer) was done at least once a week (so that the carbon peaks would be right). Some special studies have been done just like previous years. One new thing was to look at the polarization in the 1 st and 2 nd half of data-taking period in each AGS cycle: 1 st half 2 nd half 44813 0. 5850 ± 0. 0379 0. 6428 ± 0. 0370 44814 0. 6466 ± 0. 0379 0. 6135 ± 0. 0370 44815 0. 6441 ± 0. 0380 0. 6366 ± 0. 0371 44817 0. 6188 ± 0. 0382 0. 6183 ± 0. 0364 Other special measurements like ramp measurements (both up and down) etc. were also done.

Just one example (I made a quick check before I sent them the data:

Apart from the problems mentioned before, the polarimetry system has been quite stable. The only other thing came to mind was the bad crimp cables (made ~5 years ago ) in the LV which led to loss of 3 out of 8 detector readout. ~Once a month, only in the weekend, I might be called because a process which wouldn’t stop and prevented further measurements. We don’t have an obvious problem of rate issue, probably because our rates in AGS are lower than RHIC, we only take data ~1 s out of each ~4 s AGS cycle (there is time for the DAQ system to deal with the data in-between) and we’ve tried to use thinnest AGS carbon targets ……

A couple things to do : Make a few minor software changes (eg. allowing “emittance measurements” even when we use rev. counter instead of GCC) which I had refrained from changing in the middle of the run. Buy a new PC and install Linux/drivers/softwares as a backup for “cnipol 2” which was bought before the 2006 run when I did an upgrade for the DAQ.