This paper deals with the factors that influence the accuracy of the measurements performed with a calorimeter, developed to detect the optical signal emitted by the spark generated to facilitate experiments with electrostatic discharge. Experiments are described which were performed with a special calorimeter and a compound optical source consisting of a spark gap, a human-body-model electrostatic discharge (ESD) signal generator, and four LEDs. The spark gap was used as a versatile and more powerful source of optical emission than the human-body-model circuit; the LED source was used for calibration and alignment. For control, the spectrum of the spark discharge was measured with a spectrometer and a broadband photodetector. The calorimeter was used as a means for the determination of the energy contained in the optical signal, yielding the value of the spark-gap emission in the fundamental units for energy. The long-term stability of the system was measured, and the system response was studied for threshold optical signals. The calorimeter detectivity, defined as the overall instrument sensitivity, measured in volts per joule, was D=4.3×107 V.J-1 and its detection limit 2.3×10-13 J. The system's response to the IEC standard human-body-model circuit was consistent with the measurements of the system's detectivity. Most parts of the developed apparatus have been tested in an industrial environment.