the tagger OR

What we call the tagger OR is a logical OR of the 19 discriminators monitoring the left-side phototubes on the top 19 tagger focal plane paddles. Each input to the OR generates an edge-triggered internal pulse of fixed duration; these are continuously OR'ed together to produce a single signal that we call the tagger OR. Note that this is different from what CLAS calls the tagger OR. The pulse duration is manually adjustable in the tagger electronics racks. In this model it is currently ignored (set to zero). The parameters in the model that concern the tagger are the linear slope $r_{_T}$ of the tagger OR in counts per nA, the tagging coincidence window width $t_{_T}$ and the scaler dead time $d_{_T}$. The linear slope is the proportionality constant between beam current and the tagger OR rate in the linear region at low intensities.

In the model, every signal in the Radphi detector arises either from the interaction of a tagged photon (or its prompt reaction products) or it does not. These two are called trues and accidentals in the following discussion. By a tagged photon is meant a bremsstrahlung photon that was generated in the same electromagnetic cascade that contained an electron which produced a pulse over discriminator threshold in the left side of one of the tagger focal plane counters. The distinction between trues and accidentals is in the physics of the event that caused them, not in the temporal proximity of the tagger and detector signals, although the above restriction to prompt reaction products excludes trues where the two signals are separated by hundreds of nanoseconds. At finite rates it is practically impossible to distinguish trues from accidentals on an event-by-event basis. Nevertheless, as long as the events are independent, the two categories of events are distinct from each other and can be treated as two separate populations, to be added together in the end.

The content of the model is embodied in the following three formulae.

$$S_{_T} = r_{_T} I\, e^{-d_{_T}\,r_{_T}\,I}$$ (1)

$${\cal F}_{acc.} = 1-e^{-t_{_T}\,r_{_T}\,I}$$ (2)

$${\cal F}_{true} = 1$$ (3)

where $S_{_T}$ is the rate observed in the scaler counting the tagger OR signal and ${\cal F}_{true}$ [ ${\cal F}_{acc.}$] is the probability that a true [accidental] hit in the detector will occur in coincidence with the tagger OR signal, within the tagging coincidence window $t_{_T}$. There are some restrictions built into the model.

1.

Beam current $I$ must be high enough that the singles rates in the tagger phototubes can be ignored. The contribution from dark current should be less than 1% for $I$ above a few nA.

2.

The tagging counter discriminator dead time, together with the internal pulse duration in the module that forms the tagger OR, have been ignored. With the widths set to reasonable values, these effects are small compared with those which have been included in the model because they enter multiplied by the rate in a single tube (a couple MHz at most) rather than by the full tagger OR rate, in contrast to $t_{_T}$ and $d_{_T}$. If they were to be included explicitly then the two-pulse resolving power of the tagger would also have to be included. At a rate of 2MHz per tagging counter and a 20ns discriminator gate width, tagger electronics dead time depresses the tagger OR by about 4%. At present the stability of our beam current measurement is no better than that.

3.

It is assumed that beam factors that affect tagger performance, such as beam tune and steering, are constant so that $r_{_T}$ can be treated as a constant. For the set of runs throughout the June 1998 commissioning period that were used for this study, this assumption was upheld to within 2%.

4.

It is assumed that the tagging coincidence window width $t_{_T}$ is wide enough to contain the entire coincidence peak. This assumption is implied by Eq. 3. For a well-timed setup this should be satisfied for $t_{_T}$ of 6ns or greater.

Figure 1: Tagger OR scaler rate vs. beam current measured at the beginning of the June run period.

Figure 2: Tagger OR scaler rate vs. beam current measured at the end of the June run period.

The parameters $r_{_T}$ and $d_{_T}$ were extracted from a two-parameter fit of measured tagger OR rates vs. beam current taken during the June run. The results from a scan taken early in the run are shown in Fig. 1. The dotted line shows the linear extrapolation with $r_{_T}$ from low rates. The saturation of the measurements gives a value for the scaler dead time $d_{_T}$ of $95$ns. This large dead time is completely out of line with the specification for the scaler (200MHz), and indicates that there is an additional dead time hidden somewhere in the system that is not in the model. It turned out that the tagger discriminator widths had not been properly set. When this was done, a new scan produced the data shown in Fig. 2, now shown all the way out to 200nA. A new fit yields 10ns for $d_{_T}$, which is in good agreement with the width of the tagger OR pulse width seen on the oscilloscope. Although the model is merely descriptive at this point, getting reasonable values for the parameters provides a good check that the electronics is working as expected.