A simple method is proposed for describing the detectivity of photoconductive cells.
First it is pointed out that at signal modulation frequencies well below the frequency that corresponds to the time constant of the cell, the responsivity of the cell is independent of the frequency. For the same range of signal modulation frequencies, the noise power (mean‐square noise voltage) per unit frequency bandwidth is inversely proportional to the frequency. It is a very general property of radiation detectors obeying Lambert's law that the detectivity varies inversely as the square root of the sensitive area.
As a result of the facts just stated, one is led to the conclusion that if one multiplies the measured detectivity by the square root of the cell's sensitive area and by the square root of the band width of the noise, and if one divides by the square root of the frequency of the test signal, one obtains a result that is independent of the test band width and test frequency and is independent also of the sensitive area of the cell. The detectivity D so obtained is defined in Eq. (3) and is called the ``detectivity in the reference condition C.''
The detectivity D is not proposed as a replacement for a detailed description of the test conditions and test results. It is an auxiliary quantity to be computed from the test results. It has the advantage over any other single test result that the knowledge of it alone permits a direct and significant comparison of two cells, even though the cells are of different area and are tested under different conditions.
A discussion is included of the various auxiliary ``ground rules'' that must be observed if one wishes to state the detectivity of a cell with complete unambiguity.