Many of these scratches are produced eluting re-spooling by grains of dust imprisoned between the coils. Transverse scratches appear also when attempts are made to flatten, by knocking edgeways on a table, an irregularly wound spool in which the edges of the windings arc not level.
After passing a number of times through an over-lubricated projector, a film is often be spattered with oil and must be cleaned. There are special machines for this cleaning, usually employing carbon tetrachloride, which is a non-inflammable solvent. This liquid sometimes contains traces of sulphur chloride which may attack the film if the latter is re-spooled before complete evaporation of the cleaner. This risk is avoided by using tetrachlorethylene.
Each time re-spooling is done, the joins should be inspected and re-cemented, if necessary, as a partly-lifted join can cause the film to tear.
908. Flicker. In early models of projectors, the luminous beam was cut off only during the time required to replace one frame by another, so that there were 16 dark intervals per second. The flicker due to these alternations of light and darkness are very unpleasant, and even fatiguing. Trial of various means has shown that flicker can be reduced only by increasing the frequency of the dark intervals, and this has led to the use of shutters with two blades, one serving to mask the film while it is being shifted, and the other solely to reduce flicker. Flicker becomes more apparent according as the screen is more brightly illuminated, and the great increase in the power of the illuminants employed of late years has again brought for ward this question, which has not been solved by the many attempts to use shutters with translucent blades.
It has been clearly shown by K. de Proszynski (1911) that the necessary and sufficient condition to eliminate flicker, when the brilliancy of the screen is very large, is to increase the number of light interruptions to about 5o per second, these interruptions being of equal duration and uniformly spaced.
909. Cinematograph Projection. It is not possible to deal here, even summarily, with cinematograph projectors, nor with their mani pulation and attendant necessary precautions, the chief of which latter are, moreover, compul sory by official regulations.
We may however mention that in addition to projectors intended for picture theatres, there are many patterns, of a more or less simplified type, intended for educational purposes (school models usually include arrangements for stop ping the film at any given frame, at the will of the lecturer, so as to permit details to be exam ined), for commercial purposes (portable pro jectors for commercial travellers), and for home projection.
Whatever the projector used, it must be con stantly maintained in an absolutely clean condition, and the various parts which come in contact with the film must be periodically examined with care, so as to permit the replace ment, in time, of worn parts likely to mutilate the film.
This examination should be directed particu larly to the toothed drums (worn teeth tend to cut the film), take-up rollers, guide rollers, pressure rollers (jamming of the rollers prevent ing their rotation, non-parallelism of their axes with those of the drums, faults of alignment of the various rollers and drums), pressure springs of the film guide (unequal or excessive pressure),' the take-up mechanism of the film (excessive friction of the clutch discs causing a great ten sion on the lower supply spool), the intermittent mechanism (play in the intermittent drum, wear of the plateau of the Maltese cross, and on the supply and take-up spools (warped flanges) ). After showing the film, it must be re-spooled so as to be ready for projection again.
910. Stroboscopic Illusions. It is of somewhat frequent occurrence that during the projection of scenes, including rotating objects in motion (carriage wheels, fly-wheels, gears, etc.), these parts seem to be turning at a speed very different from their normal speed, or seem to be at rest, or even to be turning in the opposite direction.
We must first of all note that this illusion manifests itself only in the case of perfectly symmetrical wheels ; it disappears as soon as any given point can be followed (counterweight of a fly-wheel, spot on the rim or on one of the spokes).
If the picture has been taken at the frequency of ib per second and one of the spokes has come (in this interval of 06th second) to take exactly the place of one of the other spokes (that is, if its angular displacement is some multiple of the angle a between two consecutive spokes), the successive images of the wheel will be identical among each other, and the wheel will seem motionless on the screen.
If the speed of rotation is very slightly higher, and if, for instance, the rotation is na e during the time interval between two frames, the image will seem to have turned only through the angle e in the normal direction. If, on the contrary, the speed is very slightly decreased, the angle of rotation being na — e during the time interval between two frames, the image will seem to have gone backwards through the angle e.