Trigger Studies

Since at the trigger level, the reactions $\pi^- p \rightarrow \gamma n$ (B.R. $=$ 0.39) and $\pi^-p\rightarrow\pi^0n$ (B.R. $=$ 0.61) generate a strong two-photon background, the weaker two-photon signal had to be isolated for the $\pi ^-p\rightarrow \gamma \gamma n$ experiment. Determination of our two-photon trigger was based on good selection efficiency for two-photon events and good rejection efficiency for single-photon events.

Table 4.1: Definitions of the loose, 2-D, 4-D and tight triggers derived from the multiplicities and patterns of the $C$ and $D$ trigger scintillators. $C$ counters: the number of the $C$ hits and the gap in the $C$ hits. $D$ counters: the number of valid $D$ hits and the limit on the valid $D$ range.

Trigger Name	Fig. ID	No. of $D$	Range of $D'$s	No. of $C$	Gap in $C'$s
		Scintillator		Scintillator
		Sectors		Sectors
loose	1	$\geq$2	$\pm$2	2,3	0
2-D	2	$\geq$2	$\pm$2	2	1
4-D	3	$\geq$4	$\pm$2	2,3	0
tight	4	$\geq$4	$\pm$1	2,3	1

During the May-June 1997 test run, $\pi^- p \rightarrow \gamma n$ and $\pi^-p\rightarrow\pi^0n$ trigger rates versus various trigger conditions were measured. Due to practical considerations, a CH$_2$ target rather than a liquid H$_2$ target was used, and the carbon background was subtracted from pion stops measured on a carbon target, and trigger rates were scaled from CH$_2$ to liquid H$_2$. Figure 4.2 shows the trigger rate for the loose, 2-D, 4-D and tight trigger conditions at magnetic fields of 1.2 kG and 1.8 kG. The most noticeable effect was found to be the very large decrease in the trigger rates between the loose and 2-D triggers (rates of 30-110 $s^{-1}$) and the 4-D and tight triggers (rates of 1-3 $s^{-1}$).

Figure 4.2: The measured two-photon trigger rate scaled to liquid hydrogen following $\pi ^-$ stops in a CH$_2$ target as a function of magnetic field and trigger condition. For definition of the trigger ID's see Table 4.1.

The trigger rates also slowly decreased with increasing field. Inspection of the recorded events indicated for the loose and 2-D (i.e. $\geq$2 valid $D$) triggers single photons dominate the rate, whereas, for the 4-D and tight (i.e. $\geq$4 valid $D$) triggers photon-pairs dominate the rate. Thus along with the $\overline{A}\cdot\overline{A'}\cdot\geq$ 2$C$ requirement, a $\geq$4 $D'$s requirement effectively rejected single photon events from two-photon events.

For the final production run of April-May, 1999, in addition to requiring a tight trigger, an excellent way to achieve further reduction in the trigger rate was found by rejecting the back-to-back photon pairs from at-rest $\pi^-p\rightarrow\pi^0n$ by eliminating triggers due to back-to-back $C$ counters at the AHC trigger level, followed by rejecting the back-to-back hits in the drift chamber cells, called the $SSP$ cut. Reduction factors of 1.9 in the trigger rates were obtained by rejecting triggers where the opposite $C$ fired (i.e. if $C$1 fires $C$7 is not allowed), and 8.5 by rejecting triggers where any of the three opposite $C'$s fired (i.e. if $C$1 fires $C$6, $C$7 and $C$8 is not allowed). Application of the $SSP$ cut meant that a $\geq$3 valid $D$ trigger rather than a $\geq$4 valid $D$ trigger could be used, which allowed increased overall acceptance.

Thus for $\pi ^-p\rightarrow \gamma \gamma n$ final data taking, as described in Section 3.4, a trigger file mC3_gC1_rD1_OC3 was employed which rejected back-to-back photon-pairs from $\pi^o\rightarrow\gamma\gamma$ events, required two or more $C$ counter hits (n$C\geq$2), three or more $D$ counter hits (n$D\geq$3) and as part of the $CD$ hit pattern requirement, required that at least three valid $D$ counters fire.

In the following chapter, we describe the analysis of the $\pi ^-p\rightarrow \gamma \gamma n$ data.