In order to prevent oscillations, which could not fail to occur after each emission of a current (so that the helix, instead of returning to a position of equilibrium and stopping there, would go beyond it and alternately uncover the slits, a a'), the apparatus is provided with a liquid deadener. To this end, the prolongation, v, carries a piece, o, which dips into a cup containing a mixture of glycerine and water.
We shall now describe the differenzialrecorder. Opposite the two slits, a and a', there are two powerful converging lenses, l and l', whose foci coincide with two sorts of selenium plate rheostat, z and z'. The result of this arrangement is that as soon as one of the slits, as a consequence of the displacement of the helix, r, allows a luminous fascicle to escape, this latter falls upon the corresponding lens, which concentrates it and sends it to the selenium plates just mentioned. Under the influence of the luminous rays, the resistance that the selenium offers to the passage of an electric current instantly changes. At M and M' are placed two horseshoe magnets whose poles are provided with pieces of soft iron that serve as cores to exceedingly fine wire bobbins, d. These polarized pieces are arranged in the shape of a St. Andrew's cross, and in such a way that the poles of the same name occupy the two extremities of the same arm of the cross, an arrangement very clearly shown in Fig. 2.
Between the poles of the magnets, M and M', there is a permanent magnet, A, movable around a vertical axis, i. Four spiral springs, f, whose tension may be regulated, permit of centering this latter piece in such a way that when the current is traversing the spirals of the polar bobbins it is equally distant from the four poles, n, s, s', and n'. Under such circumstances it is evident that a difference in the power of attraction of these four poles, however feeble it be, will result in moving the magnet, A, in one direction or the other around its axis. The energy and extent of such motion may, moreover, be magnified by properly acting upon the four regulating springs.
The bobbins of the magnet, M, are mounted in series with the selenium plates, z, the local battery, B, and a resistance box, W. Those of the magnet, M', are in series with z', B', and W'. The local batteries, B and B', are composed of quite a large number of elements. The current from the battery, B, traverses the selenium plates and the bobbins of the magnet, M, and returns to B through the rheostat, W; and the same occurs with the current from B'. The two currents, then, are absolutely independent of one another.
From this description it is very easy to see how the system works. Let us suppose, in fact, that the current which is traversing the spirals of the helix, r, has a direction such that the helix in its movement approaches the pole, S; then the prolongation, v, will uncover the slit, a, which, along with a', had up to this moment been closed, and a luminous fascicle escaping through a will strike the lens, l', and from thence converge upon the selenium plates, z'. This is all the duty that the line current has to perform.
The luminous rays, in falling upon the selenium plates, z', modify the resistance that these offered to the passage of the current produced by the battery, B'. As this resistance diminishes, the intensity of the current in the circuit supplied by the battery, B', increases, the attractive action of the polar pieces of the magnet, M', diminishes, the equilibrium is destroyed, and the piece, A, revolves around the axis, i. If the polarity of the line current were different, the same succession of phenomena would occur, save that the direction of A's rotation would be contrary. As for the rheostats, W W', their object is to correct variations in the selenium's resistance and to balance the resistances of the two corresponding circuits. The magnet, A, will be combined with a registering apparatus so as to directly or indirectly actuate the printing lever. The entire first part of this apparatus, which is very sensitive, may be easily protected from all external influence by placing it in a box, and, if need be, in a room distant from the one in which the employes work.
Figs. 3 and 4.
The differenzialrecorder alone has to be in the work room.