Motion-picture camera providing up to \(10^9\) frames per second

In the study of explosion and detonation processes, a very important role is played by the possibility of instantaneous photography with an exposure of the order of \(10^{-7}\) sec. or less. As in ordinary photography, one may distinguish “instantaneous photography,” i.e., the production of individual photographic images with a short exposure, and “instantaneous motion-picture photography”—the production of a continuous sequence of such images. The former is usually carried out with the aid of an “electric shutter”—a Kerr cell (the inertia of Kerr cells does not exceed \(10^{-8}\) sec.). The latter is carried out with the aid of special motion-picture apparatus. Such, for example, is the O’Brien–Milne camera, which gives \(10^7\) frames per second[^1]. However, cameras of this type necessarily have a very complicated optical system, are expensive, and are difficult to manufacture.

Recently a motion-picture camera has been described which, with a simple opto-mechanical system, makes it possible to obtain up to 300 consecutive frames measuring \(10 \times 10\) cm, with a filming rate from \(10^5\) to \(10^9\) frames per second[^2]. This camera is based on a multi-slit shutter located in the plane of the image formed by the objective and projected onto the motion-picture film. The shutter consists of a glass plate coated with a layer of an opaque substance in which, by means of a dividing machine, narrow slits have been cut, forming a regular grating.

At a certain position of the shutter, an image is obtained on the film in the form of a series of parallel lines, each line being an image of that part of the object which is projected by the objective onto the corresponding slit of the shutter. If, for example, the distance between the slits of the grating is 30 times the width of a slit, then the image obtained will occupy \(1/30\) of the entire area of the frame. Now let us move the shutter by the width of a slit. A second system of parallel lines will be obtained on the film, adjacent to the first. It too will occupy \(1/30\) of the frame area. With gradual displacement of the shutter over the distance between two adjacent slits, the image on the film will represent a superposition of 30 separate frames, shifted relative to one another by the width of a slit. These 30 frames will cover the entire area

of the photograph. Only by setting the shutter to the required position can we distinguish an individual frame in this complex picture (Fig. 1).

In most cases rapidly occurring processes are photographed, i.e., the object being photographed is moving or changing. In filming such processes, the multi-slit shutter moves relative to the film at a constant speed. If we view the film through a uniformly moving multi-slit shutter, we shall see the recorded process as it occurred in time. Different shutter speeds will produce different degrees of slow motion.

Fig. 1. Explosion of a powder charge. Photograph obtained with a multi-slit shutter.

The optimal slit width of a shutter of this type is on the order of 0.0025 mm. To obtain \(10^8\) frames per second, it is necessary to move the shutter at a speed of \(250\ \text{m/sec}\). At such speeds, accelerating, decelerating, or stopping it is in practice very difficult. A rotating cylindrical shutter of the multi-slit type is inconvenient: it requires conical slits of variable width and must have a considerable diameter at sufficiently high angular velocities. Therefore the author preferred to move not the shutter itself, but its projection onto the film. For this purpose he used a rotating mirror. The optical system of the motion-picture camera is shown in Fig. 2. The first lens projects the image of the object onto the shutter; the second lens projects the image of the object and of the shutter onto the film, with an intermediate reflection from the rotating mirror. In decoding the photographs, the transition from a given frame to an adjacent one corresponds to a displacement of the shutter by the width of one slit. Displacement of the shutter is accomplished in this case by means of a micrometric screw.

The rotating mirror in this camera can rotate at an angular speed of up to 500 revolutions per second. With a distance from the mirror to the film of 50 cm, this gives a shutter-image speed on the film of 3000 m/sec. For such a shutter-projection speed, with a slit width of

Fig. 2. Optical system of a motion-picture camera with a multi-slit shutter: 1—object, 2—camera lens, 3—multi-slit shutter, 4—photographic film, 5—projection lens, 6—rotating mirror.

0.013 mm, there corresponds a filming speed of \(2.5 \times 10^8\) frames per second. The author believes that, when filming detonation and explosion phenomena, which give much light, it is possible to obtain up to \(10^9\) frames per second.

M. G.

CITED LITERATURE

1. O’Brien and Milne, J. Soc. Mot. Pict. Eng. 52 (January 1949).
2. Sultanoff, Rev. of Scient. Instr. 21, 653 (1950).