Journal of the Society of Motion Picture and Television Engineers (1950-1954)

ECHO DELAY IN PICTURE ELEMENTS = (i) T\TT .06 Fig. 12. Relation between echo amplitude and echo delay for double-excursion K = 1 in characteristic. Thus a compromise value of a is possible for each echo delay plotted in Figs. 9 and 10, to permit approximate constancy in the resulting excursions R and $. A further point is to be noted in the application of equations (2). This is that, at very short echo delays, the major part of the phase distortion consists of a linear component proportional to frequency. In practice this linear component would be interpreted as merely that much more (or less) undistorted delay in the system. Thus it would be subtracted out, and only the residual curve (which is approximately a cubic) would be kept as a measure of the phase distortion. The question of exactly what to subtract, over the range of echo delays, is not easy to answer categorically. It is clear that for the larger echo delays, where there are several cycles of scallop, the linear component of the scallop close to zero is not significant, and nothing should be subtracted. In order to avoid entering into difficult mathematics, only a simple rule will be used here. The curve, with linear term subtracted, is shown as the fine dotted curve G'(w0r) in Fig. 11. It is derived from the middle auxiliary diagram at the top of the figure. Here the phase excursion aG' is taken with respect to the initial linear component of the scallop as shown by the dot-and-dash line. It is seen to be negative, and its absolute value is plotted in the lower diagram. As the fraction of the cycle used (or echo delay) increases G' rapidly becomes large and crosses curve G, for which no subtraction of Pierre Mertz: Television Transmission Echoes 585