Technical Note
radphi-2004-301

Tagging Photons with the Radphi Detector



Richard Jones
University of Connecticut, Storrs CT

March 10, 2004

Abstract:

The analysis of data taken with the Radphi experiment in summer 2000 has been hindered by the lack of a reliable method for associating an event in the detector with the energy of the photon in the beam that produced it. In principle the energy of the beam is tagged on an event-by-event basis by a coincidence between the detector and one of the channels of the tagging counters. This association is complicated by the fact that accidental coincidences with the tagger make event-by-event association ambiguous. Furthermore many of the interactions in the target that produce sufficient gamma energy to satisfy the trigger also include charged particles to which the calorimeters are largely insensitive. The result is that studies that look for association between the tagger channel and total energy in the detector have shown little if any correlation. This report shows how these problems can be overcome. An excellent agreement between the tagger channel and measured energy is demonstrated, in quantitative agreement with the resolution expected based upon prior studies of meson widths. The analysis procedures that incorporate the method are explained, and the total event yields obtained using photon tagging are provided.

The Radphi experiment took data during summer 2000 using the tagged photon beam line in Hall B at Jefferson Lab. The photon beam had an incoherent bremsstrahlung spectrum with an end-point of 5.65 GeV. The photon source was operated in two modes, the first with a thin radiator of approximately $10^{-4}$ radiation lengths and 130 nA of electron beam current and the second with a thicker radiator and 77 nA. Both configurations correspond to an inclusive rate of $5\cdot 10^7$ /s in the tagging focal plane. The experimental trigger was based on a triple coincidence between three thin scintillator layers in the barrel detector referred to as the bsdAND . The range of interest in the beam spectrum was defined by a subset of the Hall B tagging counters subtending from 75% to 95% of the end-point energy. The electronic sum of the signals from these 19 counters called the taggerOR was combined in coincidence with the bsdAND to ensure that every trigger had a hit in at least one tagging counter within the tagging time window of 20 ns. Introduction of a veto from beam halo counters upstream of the detector, denoted the upvOR , completes the definition of the level-0 trigger as bsdAND $\cdot$ taggerOR $\cdot$ $\overline{\mbox{{\em upvOR}\ }}$.

The upv veto was used to suppress interactions coming from scraping of the beam on material upstream of Radphi, and just served to gate off the experiment for 40 ns whenever a halo track was seen. The upvOR rate varied from 1.4 MHz near the beginning of the run down to around 400 KHz where it remained for the latter third of the run period, after the accelerator operators had improved the beam alignment. The effect of the upvOR veto on the experiment was simply to reduce the live time of the experiment by a factor of 2-5%. Its effects are ignored in the following analysis.