Tagger accidentals rate scan, Fall 2019

From GlueXWiki
Revision as of 07:49, 31 October 2019 by Jonesrt (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

The purpose of this rate scan is to study systematic effects that enter into the correction for accidental tagging coincidences that is a necessary part of any cross section measurement using the tagged photon beam. The ideal starting point in this discussion is a perfect bunched beam structure with uniformly populated bunches passing through the radiator every 4.08 ns. The number of tagged photons in each beam bunch is then an independent Poisson random variable, all with the same mean value as long as the beam current is stable. This is only a starting point; we know that corrections to this simple model are needed. So far, the only correction we have studied is one that accounts for random bunch-to-bunch intensity variations. Are there non-random intensity variations that we can quantify and take into account? Are there rate-dependent effects in the accidental statistics that need to be understood in order to measure precise cross sections at high rates?

To explore these issues, we should attempt to measure a known process, using a well-understood detector and trigger, and push it up to the highest rates possible with the tagger. In the tagger microscope, this limit is reached at approximately 5 MHz per fiber. The well-understood detector and trigger is the pair production process in the converter and the pair spectrometer coincidence trigger. This is the best option because the rates in the PS using the 75 micron Be converter are low enough to prevent concerns about saturation in that part of the measurement. This remains true even at rates a factor 5 higher than standard GlueX running, as proved in 2018 when we ran with the 750 micron converter at 200 nA without problems. This rate is higher than anything we have every run with in the GlueX detector, so I propose that these scans be done with the GlueX detectors turned off, at least those close to the beam.

The predicted rates per tagger channel vs electron beam current are given in the table below, for a series of beam currents proposed for these scans. Rates are computed assuming the Al 40 micron (4.5x10-4 rad.len.) radiator and the 5 mm collimator. There is no need to shift the collimator to the 3.4mm hole, so the standard feedback stabilization on the beam position at the active collimator should be enabled for these runs.

step e- beam current (uA) TAGM rate (MHz/chan) TAGH rate (MHz/chan) PS trigger rate (kHz)
1 0.125 0.3 1.1 3 (?)
2 0.25 0.6 2.2 6 (?)
3 0.5 1.2 4.4 12 (?)
4 0.7 1.6 (off) 16 (?)
5 0.9 2.2 (off) 22 (?)
6 1.2 2.9 (off) 29 (?)
7 1.6 3.8 (off) 38 (?)
8 2.1 5.0 (off) 50 (?)

For a single scan, separate DAQ runs should be taken at each current setting.

  • minimum number of triggers per step: 10 million
  • minimum run length per step: 10 live minutes

We already have a lot of data at rates comparable to steps 1 and 2, but it is important for this study to take all of these runs in a row, under constant beam conditions apart from increasing current. These runs could be taken concurrently with amorphous physics running if the reduced beam current is not an issue.

If everything goes smoothly, this scan should be carried out twice, with a gap between to give time for the data to be analyzed. This is the first time the tagger channels have been operated at these rates, and adjustments to their operating points (adc readout threshold, discriminator threshold, bias voltage) will probably be needed in the case of the TAGM.