The Cine Technician (1935-1937)

76 The Journal, of tJie Association of Ciiie-Tecliiu'ciaus Fcl)ruar\ lO.v^ gular aperture which is focussed accurately on the rectangular sHt by the objective B. M is an oscillograph mirror, which vibrates in accordance with the impressed signal, and so varies the length of the illuminated portion of the slit. The slit is focussed on the film and so produces X o.oo; " i \i J-' ~ FHEQUENCr IN CVCLES .■"-■:« 'JECONi I'l : .: l.i.^scs f.-.r various i;i;.';:i wii)t!i or ;'./:i!mil'n <:< Fig. 2. on the negative the dark line of varying width which was seen in the first slide. With 100"„ modulation the width of the exposed line varies from zero to 84 mils. The optical set-up here produces a reduction in size, and the width of the image in the direction of travel of the film is usually ooVtt" This dimension should be as small as possible. In variable density systems a narrow slit is also imaged on the film, but exposure variation is accomplished in one or other of two quite distinct ways. Either the intensity of the light falling on the slit is varied, or the width of the slit is varied, the intensity remaining constant. Intensity variation is usually achieved by means of a glow tube — a special adaptation of the neon-lamp principle. The method of slit width variation is the more important, however. The slit in this case consists of a light metal ribbon doubled round and stretched on a bridge, the assembly being generally known as a light valve. The gap between the ribbons is capable of exact adjustment and the tension is controllable. The ribbons are placed in an intense transverse magnetic field, so that a current round the ribbon circuit causes the two parts to open, or close, according to its direction and strength. The ribbons are focussed on the film by means of an achromatic objective giving an optical reduction of two to one. The intensity of the image is thus constant, but its width varies from zero to I -mil., or whatever the dimension may be. Exposure thus varies on a time scale, a given point on the film taking a longer time to pass through a wider image (when the ribbons are open) and so receiving more exposure. It will have been noted that the optical slit, focussed on the film, is a feature of all recording systems. Its nature and effects will now be considered in more detail. We have taken the frequency of 9000 cycles per second as the upper lin.it of practice, and simple arithmetic shows that at a film speed of 90 feet i)er minute this corresponds to a wavelength of 2 mils. — 2 thousandths of an inch. Now just as fine detail cannot be drawn with a blunt pencil, so short wavelengths cannot be recorded with a wide slit image. The same appHes to reproduction. Fig. 2 shows how high frequency loss is affected by image width. It will be seen that at 9000 cycles the loss is reduced from 10 to I D.B. by reducing image width from 1 -5 mil. to -5 mil. Tlie actual losses are always a little greater than the theoretical figures, owing to imperfect image formation, the effective image width being thus somewhat greater than the calculated value. The most modern practice consists of a |-mil. slit with a specially corrected objective, giving an effective image width only slightly in excess of |-mil. Since most recording is done on positive type emulsion, the lens correction should be most effective in the upper blue and violet region, and not carried out by coincidence of blue and yellow images. The considerations relating to processing of the sound track have now to be examined. This part of the subject is more interesting from the standpoint of variable density and will be so considered. There are three definitions which have first to be made. The transmission (generally expressed as a percentage) of a photographic image is the proportion of the total light incident on the film which is transmitted through it. Density is the logarithm of the reciprocal of the transmission. Hence a transmission of 50"'o (or -5) is equal to a density of the log. of 2, which is -3. If a series of exposures of different values are made on a plate or film, and the densities produced after development are measured, it is found that the density is proportional to the log. of the exposure over a limited range. The factor of proportionality is called the gamma, or contrast factor, and it depends on the nature of the emulsion and the degree of development. The curve produced by plotting density against log. of exposure is called an H. & D. or Hurter and Driffield curve, and three typical examples are shown in Fig. 3. The two steep ones are of positive TIMi SCAce H*5 CUffvtS Fig. 3. type emulsions and development, and the one with smaller slope is a record of the type of negative development used for variable density recording. These curves are produced with the aid of an instrument known as a sensitometer, which is an apparatus for producing a scries of known exposures, in whicli each step is a certain multiple of the preceding one. This can be done either by keeping the time of exposure constant and varying the intensity of illumination, or by maintaining the intensity at a suitable value and varying the time. The latter method is the more convenient, the time ratio between successive steps being generally the square root of two, which is the same thing as to say that it has a logarithmic scale in which the separations are -15. The measurement of contrast ob