Film Lighting Using Metal-Halide Lamps

Abstract

When a discharge lamp is run from an alternating supply, the intensity of light from it varies between maximum and minimum values at twice the supply frequency. This so-called intensity “ripple” can, if such lamps are used as light sources in motion-picture film work, lead to a cyclic variation in frame-to-frame exposure. This in turn can cause the overall luminance of the displayed picture to fluctuate. In this paper, an investigation of the visibility of such luminance fluctuations over the frequency range 0.1–12.5 Hz is described. Limits of film exposure variation are recommended, below which these picture luminance fluctuations should not be visible, taking into account a nonlinear transfer characteristic. The magnitude of film exposure variation which can occur in practice is discussed, based on the assumption that the variation in lamp intensity is sinusoidal in character, as is the frequency of the luminance fluctuations which occur when the film is replayed. Using these factors, theoretical relationships are derived between camera frame frequency and shutter angle, and lamp supply frequency and intensity ripple magnitude, such that picture luminance fluctuations due to the presence of the ripple component are not perceptible. Camera shutter angles in the range 135°-235°, camera frame frequencies of 24 Hz and 25 Hz, and lamp supply frequencies both in the range covered by conventional public electricity supplies and also of values much higher than this are included. In this latter case a “safe” lamp supply frequency is assessed, above which the effects of intensity ripple in causing film exposure variation may be discounted. The effect of ripple asymmetry (successive maxima of the ripple waveform having alternately higher and lower magnitudes) is discussed. The characteristics of two currently available metal-halide discharge lamps are examined, and experimental techniques for measuring intensity ripple under standardized conditions are described.