Radio Broadcast (May 1928-Apr 1929)

AS I HI EEOHKgSIEE SrKTTT ~ BY CARL IMUHhk Z Photographic Data For Broadcasters H. B. MARVIN, during the discussion on his paper, "A System of Motion Pictures with Sound,"* said, "The latter i (variable area type of sound record on film) requires a sound track which has a high degree of contrast and that is all. We aim at an exposure which will give us a density of 1.3 and develop 1 to a gamma of 1, but these are not critical. The variable density system requires fairly close control of exposure and development in order to eliminate distortion. . . ." It is as necessary for broadcast engineers who are transferring their allegiance to sound motion pictures to learn something about the photographic end of the business as it is for the movie people to acquire some familiarity with audiofrequency technique. While almost everyone knows something about photographic contrast, development, etc., a more scientific understanding of such terms is required by the professional. A useful work in this field is Photography as a Scientific Implement, written collectively by a group of authors, published by Blackie and Son, Ltd., in England, and distributed in the United States by the D. Van Nostrand Company. Chapter IV, on "The Theory of Photographic Processes and Methods," by S. E. Sheppard, is referred to in the present discussion of photographic exposure and development, which is not intended as more than an introduction and basis for further study. We may begin with a few definitions. The opacity of a photographic image is defined as the ratio of the incident normal light to the emergent light. The transparency is the reciprocal of the opacity. It follows that opacity may theoretically be anything from one to infinity while transparency varies between zero and unity, the latter corresponding to perfect transparency. The Briggs logarithm of the opacity is called the density. The electrical engineer will note the analogy to transmission of energy along telephone lines. The ratio of the electrical energy impressed on a line to the energy received at the other terminal is analogous to the opacity of a photographic image in the field of optics, and the tu method of reckoning transmission loss and gain corresponds to photographic density, which is likewise a logarithmic function (See Radio Broadcast for September and October, 1926, pages 405 to 408 and 506 to 509, respectively, for a discussion of telephonic gain and the standard transmission unit). * Transactions, Society of Motion Picture Engineers, XII, No. 33, p. 86, 1928. 2.0 TIME IN MINUTES FIG. 3 It may be shown experimentally that, approximately, D = pM where D is the density, p is a constant, and M is the mass of metallic silver per unit area on a negative. The photographic process is essentially the transformation of silver from the ionic to the metallic state. The above terminology and relationship were worked out by the photometric physicists Hurter and Driffield, and they also introduced the characteristic curve of a plate or film shown below as Fig. 1, in which densities are plotted as ordinates against logioE, E being the exposure, which is the intensity of the light to which the plate was exposed multiplied by the time of exposure. L0G10E FIG. 2 This photographic characteristic curve which, it will be noted, is similar in shape to the static plate current-grid voltage graph of a vacuum tube, shows the relationship between two logarithmic functions: that of the opacity of a negative and the degree of exposure to light which produced the opacity. The region convex to the X-axis at the left is known as the region of underexposure; the middle part, which is sensibly a straight line, is called the region of correct exposure; the portion to the right where the curve bends and becomes concave to the X-axis is the region of over-exposure. In the middle, where the curve is straight, the slope or tangent of the angle which the straight line makes with the X-axis is called the development factor {gamma). It is a convenient measure of the photometric contrast of the negative in question, and is also known as the contrast factor. Mathematically the straight line portion of the curve is represented by the equation D = y (logioE — logioi) where D is the density, T (gamma) the development factor of the negative, E the exposure (intensity of light x time) and i is a constant corresponding to the value of the exposure where the extended straight line cuts the X-axis. If E is held constant, therefore, it follows that D = k",', that is, the density equals a constant times gamma. However, there is more to the idea of gamma than the relation of density to exposure. The chemical development also plays a part, and in general the value of gamma increases with the time of development up to a limit of extreme contrast, called gamma infinity (rM). Fig. 2 shows a family of curves illustrating the variation of gamma with the time of development T, up to the maximum T^. For each time of development there is a definite value of gamma corresponding to the slope of the line, and a 185 definite linear variation of density with degree of exposure. All the curves intersect at the point logioi on the horizontal axis. The value of gamma, while it increases with time of development, does so at a decreasing rate (since less and less silver remains to be acted on) and ultimately reaches a limit which is largely fixed by the constitution of the plate in question. If plates which have been given a series of exposures increasing in geometrical proportion are subjected to different development times and the resulting densities are measured, a graph of density against time of development has the saturation form shown in Fig. 3. The decreasing slope is what would be expected. It is the same picture as that of a temperature-time variation in a heat run on a transformer, or many other chemical and electrical processes. The meaning of Marvin's statement as cited above should now be clear. It indicates that in the process of sound recording by the variable area method the exposure is regulated so that the density, as plotted in Fig. I, would be about 1.3 (corresponding to an opacity of almost 20) and subsequent development is timed so that, in Fig. 2, the appropriate curve would be one making an angle of 45 degrees with the horizontal axis and therefore having a tangent of 1.0. Of course, as far as regulation of exposure goes, the time is fixed by the constant movement of the film at the rate of 90 feet per minute, but the result desired may be secured by properly setting the intensity of the recording light, which is constant in the variable area system of recording. In the variable density system, since the recording is accomplished by audio-frequency variations in the intensity of the light source, it is difficult to avoid wave form distortion caused by movement above or below the straight portion of the characteristic of Fig. 1. To some of the boys who are more given to reading Liberty than poring over technical treatises, the above discussion may seem unduly theoretical. If it seems so, it is merely because they are not at home in the field of photography, which, being older than radio, has a more extensive literature and at least as involved a technique. The discussion, as a matter of fact, is most elementary and less technical, probably, than much of the broadcast material which has appeared in this department in past years. The trouble is that it involves penetration of a new field to those of us who come from the radio side of the business. An understanding of it is indispensable to anyone who wants to approach such (Concluded on Page 200) Characteristic Curve LOG 10 E FIG. I