The fraction of a second required can be practically obtained by expensive shutters; but the writer has found that very considerable accuracy may be ensured by using a rotary shutter which was made for him in mahogany, by a carpenter, for 3s. About an hour's work afterwards, in adjusting it to the lens, blackening it, etc, and it was ready for use. The shutter described is in size suitable for a Dallmeyer's 5 by 4 rapid rectilinear. The shaded parts represent holes, or parts cut away.
The part that fits on to the hood of the lens is shown in Fig. 234, and consists of a piece a, 1/4 in. thick, faced with a thin piece 6, and strengthened at back by a round piece c to give a firm hold for the screw on which the rotary part turns. This last is shown in Fig. 235, and is a 5-in. disc of thin mahogany with a hole cut out as shown, and a cylindrical piece e, which strengthens the centre, and has a small screw at / projecting far enough to tie a thread to its head. The dotted space g shows the form of the regulator; it turns stiffly on e, and serves to reduce the aperture in d. The figures show the proportion that the aperture bears to the whole circumference; 1/4 when fully open, and reduced by the regulator g to 1/3, 1/4 etc., to 1/10 or less if desired. h is a piece of lead attached to the regulator, and is just heavy enough to determine that when at rest this part shall be lowest, and the shatter so kept closed. The revolving piece tarns as easily as possible on a screw that is fixed tightly through b and a into the centre of c. A strip of cardboard fastened all along the edge of 6 as shown at k makes the apparatus practically light-tight when closed, in spite of a little irregularity in the surfaces of d and b.
It is obvious that with such an apparatus, one revolution of the wheel with its aperture fully open will give an exposure of exactly 1/4 of the period of revolution, provided that the motion is regular. By reducing the aperture, the exposure may be diminished to 1/5 or any other proportion desired. The single revolution is obtained by the cord attached at /, by giving it one turn round the cylinder e, and then palling it off as a clock cord is palled by the weight off the drum it is coiled on.
It was at first thought that by using a weight, a more regular revolution would be obtained, which would be more uniform at different times; but a trial soon showed that method to be impracticable unless modified by adding other parts to the shutter and so increasing its cost, its liability to derangement, and its weight. The writer finds that by the simple hand method with his shutter, when the cord is coiled on the cylinder in the direction in which a clock's hands move, he could not produce a revolution which should occupy more than 1/4 second, and unless violence were used, or a decided jerk given, the revolution could not he accomplished in much less time. Probably, indeed, this quarter of a second cannot be varied by more than 1/20 second without knowing it, although the experiments are done at different times; and with ordinary care and a little practice this amount of variation may almost certainly be considerably reduced.
In this particular shutter, therefore, the figures marked on it must be multiplied by 4 to give the fractions of a second indicating the duration of exposure, and 1/10 second is the longest exposure practicable by this method. 1/10 second is perhaps the longest exposure required for ordinary instantaneous work, but this may be doubled, or otherwise increased, with a fair amount of accuracy by putting the string from f over the cylinder, and so drawing the wheel round. The reason of this decrease in rapidity of revolution is obvious, for when at rest the weight h is downwards, and the opening in d on the right-hand side, so that when the wheel moves as first described, the orifice has farther to go before it gets to the lens than when moving in the other direction; in the former case, also, the weight precedes the opening, while in the hitter case it follows it.
With the shutter and lens as above described, the length of the opening on the revolving piece is about 3 times the diameter of the full aperture of the lens, a point which is held by some to be very important; when the opening is reduced to 1/6 of the circumference, it is still equal to twice the lens aperture; and not until reduced to below 1/12 (exposure then equals 1/48 second) is the aperture of the shutter narrower than the diameter of the lens.
The method of timing the revolution of the wheel of such a shutter is very simple. A tuning-fork has a pin bent round the end of one of its prongs, so that the point projects at right angles to the plane of vibration of the fork. It may be sufficiently fixed by binding it on with cotton. By fastening this to the string, or merely holding them together, and drawing the pin point across a smoked glass plate (while the fork vibrates) at the rata which is to be determined, the wavy line on the plate will accurately show the number of vibrations the fork has made daring its movement. The string should be drawn so that the wheel stops with its opening diametrically opposite the lens opening, for if carried out as above described, the wheel will make a fraction more than one complete revolution.
Then measure off from the end of the wavy line a length equal to the circumference of the cylinder e, and count the humps along one side of the measured piece of line. This number, divided into 430 if the fork is an A fork, as used for tuning violins, or into 256, if it is a C fork, will give the fraction of a second occupied by the revolution. The writer, In using an A fork, got numbers such as 107, 111, and so on, which are sensibly a quarter of 430; the time, therefore, was 1/4 second. (H. Chapman Jones.)