Journal of the Society of Motion Picture Engineers (1930-1949)

For the conditions demonstrated in Fig. 1, good optical performance is attained during only about 40% of the first exposure and 60% of the second exposure, or a total of only 50% during a complete frame cycle. This is the best the projectionist can do. The figures so far presented demonstrate the magnitude of the defect with which we are concerned, since they have been obtained with representative equipment, operating under conditions which might be found in any large theater. For test purposes, the projector was fitted with facilities to determine the various focus positions of the 5-in. focal length f/\ .9 projection lens. The light source was a Hi-Candescent Arc Lamp, with F-2 condensers, burning at 160 amp and delivering about 9000 1m to the screen with the shutter running. All focus settings were made with the aid of Simplex Screen Scopes. The 8power magnification thus provided enabled lens adjustment with greater precision than could have been attained by direct observation of the screen from the projector. The film plane position along the optical axis was measured directly in terms of lens displacement, a dial indicator calibrated in thousandths of an inch being affixed to the lens mount for this purpose. Initial calibration for zero position on the dial indicator was made by focusing the lens to produce a critically sharp image of a conical hole in a flat steel plate, the small end of the hole being in the same plane as the emulsion contacting surfaces of the film trap. Up to this point, the method and equipment are essentially the same as those employed and described by Kolb.3 The addition of a viewing shutter to the equipment enabled observation of successive phases of the cyclicly varying film frame motion (see Fig. 2). The viewing shutter's drive-motor stator was rotatable so that the shutter opening of about 9° could be phased with relation to the synchronously running projector. This stroboscopic arrangement made it possible to view the screen image in small time increments of about 1 msec through all exposure phases of successive frame cycles. The film emulsion position during any specific phase of the exposure periods could thus be established without regard to possible out-of-focus conditions during the remaining unobserved portions of the cycle. Dial indicator readings were then recorded in relation to the phase settings. A contactor on the projector shutter timed a short-duration light flash for establishing correct phase reference. The equipment as described permitted studies of film behavior under actual operating conditions. The technique of air-blast cooling of film, by which opposing air forces of the front and rear jets are adjusted so as to produce a force for positioning the film, was found to be at best a partial solution to the problem. It is possible to move the film by this method and to shift the average focus position; the resultant force, however, acts upon the film continuously, and therefore, cannot correct for the intermittent cyclical frame deformations caused by the internal buckling forces in the film which occur during the two exposure periods. The center of each frame travels over a range of about 0.020 or 0.030 in. This range is not greatly reduced by application of a continuous displacing air force (Fig. 3). The continuous jets produce a shift in average focus position; this, by itself, only slightly alters the ratio of "in focus" and "out-of-focus" intervals. The air serves primarily as a cooling agent, preventing possible damage to the film in the form of embossing or blistering or the formation of permanent buckle. It was felt that, because of the cyclical nature of the film frame deformations involved, any corrective action to neutralize the defects should be similarly cyclical. Hence, the following approach (Figure 4) was tried: Willy Borberg: Reducing Film Buckle 97