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

is more limited and care must be taken to choose filters which give, with the photocell, a response close to the peak of the film sensitivities. No specific rules for the selection of the filters can be given except that a set is satisfactory if it predicts filter changes that agree with photographic tests. Once a set of filters has been decided upon, it may be used as long as the print material has the same sensitivity distribution. The most satisfactory filter set found for control of printers using Eastman Color Print Safety Film, Type 5381, was: red — Kodak Wratten Filter No. 70; green — Kodak Wratten Filter No. 16 + No. 61; blue — Kodak Wratten Filter No. 35 + No. 38A. With this set, predictions or transfer of color and intensity balance from one printer to another are correct to 0.02 log E when printing through color-compensating filters. When transferring a balance from this condition to a printer equipped for printing with narrow spectral bands, prediction is correct to about 0.05 log E. The set of narrow-band printing filters used was: red — Kodak Wratten Filter No. 29; green — Kodak Wratten Filter No. 16 + No. 61 ; blue — Kodak Wratten Filter No. 35 + No. 38A + No.2 A. In addition to the tricolor filters, the present instrument has a small disk of flashed opal and one of Corning 9780 infrared-absorbing glass 1^ mm thick over the photocell. These are necessary when using the instrument as a color densitometer since the dyes have little density to the infrared. These precautions are probably not necessary when using the photocell as a photometer, but no tests have been made to verify this conclusion. The Densichron probe containing the photocell is small enough to fit easily in the gate of a Bell & Ho well printer. By rotating the probe slowly through a small angle, it is quite easy to get a reproducible maximum intensity reading. In some printers where the probe does not fit in the gate, a curved rod of trans parent plastic may be used to conduct the light to the cell. In this case a mechanical guide should be used to locate the probe and rod in the gate, since it is difficult to get reproducible readings when the assembly is held by hand. When a printing balance is known on one printer, or exposing device, it is frequently necessary to set up the same balance on other printers. By making the appropriate changes in the tricolor readings from the first printer to compensate for differences in speed or time of exposure, these new readings may be set up on the second printer by adjusting the filter pack and timing shutter until the Densichron shows the desired values. This procedure usually brings the printer in balance or so close to it that one photographic test is adequate before starting the printing of full-length pictures. When a printer test is off balance or shows improper exposure, and density readings on the print, or experience suggests that a change should be made in the printing light, the photometer is more reliable than the catalog densities of the compensating filters. Figure 3 shows spectrophotometric curves of a well-known set of compensating filters. Inspection of these curves shows that the addition of a Kodak Color Compensating Filter CC-50C to the filter pack would seriously disturb the blue and green light balance as well as make the desired correction to the red intensity. By taking red, green and blue photometer readings before and after each filter change, the balance may be corrected exactly to the prescribed set of values. The densities of the complete set of color-compensating filters as read by the Densichron through the filters given are listed in Table I. If measurements are not made by a photometer in the printer gate, these values may be used to estimate the amount of neutral density being introduced in the light beam by the addition of one or more of these filters. For printer control, daily readings are made through red, green and blue filters C. A. Horton: Color Printing Control 241