Synchronous Detection vs Pulse Counting for Sensitive Photomultiplier Detection Systems

Synchronous detection is compared experimentally with pulse counting, as techniques for use with photomultipliers in very sensitive optical detection systems. The criterion for comparison is the signalto-noise ratio, defined as the ratio of the value of a measurement to its standard deviation. Care is taken to obtain optimum S/N with each method. An EMI 9558B tube is used. Some observations are as follows. (1) For light levels below the dark current equivalent level, pulse counting gives somewhat better S/N than synchronous detection. The improvement in S/N is by a factor of three at 6 x 10(-17) W, the lowest illumination tested. The advantage of pulse counting increases as illumination decreases. On the other hand, for light levels of the order of the dark current equivalent level or larger, synchronous detection, pulse counting, and simple de current measurement give substantially equal results. (2) With this tube, signal-to-noise ratios are only slightly improved by cooling. The improvement in S/N is of the order of a factor of 1.5 for pulse counting and of the order of 3 for synchronous detection.