The application of (RCA 1P 28)-photomultiplier tubes to the detection of weak light intensities

Two methods are discussed of measuring weak light intensities (varying but slowly with time) with the aid of a photomultiplier tube. 1°. the light-chopping method in combination with a phase- and frequency-sensitive measuring device (i.c. alternating current galvanometer 1) 2) 3) 4) 16). 2°. the counting method, as used, for instance, in scintillation counters 4) 5). Especially the factors determining the detection limit and the reproducibility of these methods are investigated, notably: the internal amplification factor, the quantum efficiency and the dark current of the photomultiplier. These factors were found to be different between ten RCA 1P28 tubes operating under the same condition, with an average spread of about two times and an extreme spread of ten times. The output-current generated by a lightflux of constant intensity, proved to be constant in time within 1% (the error of measurement) when a well stabilized high voltage source is used. The dark current however, turned out to vary very irregularly with time. The noise of the photo- and dark-current measured within a narrow band about 50-Hz appeared to exceed the calculated value by a factor 2–3, which may be due to a Flicker effect. The optimum operation conditions for measuring weak intensities were found to be obtainable with an uniform voltage drop amounting to 60 V–70 V over each stage. The effect of decreasing the thermionic current (the most important component of the dark-current) by cooling the tube with liquid air, on the useful sensitivity was investigated and so was the alteration of the spectral response curve as a function of the temperature