The roller having been provided, take a piece of brass tube exactly so large inside that the roller will fit tightly into it, and cut off a piece the same length as the roller, or, if anything a trifle shorter. You have now to cut, with a saw or otherwise, two diagonal lines in this tube lengthwise, so that the tube is thereby divided into two pieces. Having done this the brass is replaced on the roller and fastened by minute screws, or “Prout’s elastic glue,” to each side of it, so that the roller becomes practically one of brass, with two slits in it. The screws must not project above the brass, but must be well sunk into it, so as to leave the surface smooth: and care must be taken that the screws do not touch both pieces of brass by going right through the roller—they must be very short. The object of cutting the slits in a diagonal direction is that the springs when pressing above and below the roller (see Fig. 10) shall not leave one half of the commutator before resting on the other part. If they do so the commutator will “spark” badly, which injures the fittings, and less current is obtained. Both slits are to be equidistant, and both inclined in the same direction. The roller is fixed on the axle in such a position that the middles of the lines of division are exactly in a line with the middle of the groove of the armature. When all this has been accomplished you will obviously have two conducting surfaces, each reaching over half the cylinder, separated by a small distance at top and bottom, the paraffined wood, of course, being a non-conductor of electricity. The brass tube must be made to fit smoothly round the wood, the surface being free from any irregularities, so that the contact with the springs at the sides may be as perfect as possible. Care must be taken that the brass is really separate all down on both sides. It is a good plan to fasten small splinters of paraffined wood in the slits to make sure.
This having been done, the wire from one end of the coil of the armature must be soldered to one of the semi-circumferences (if I may coin a word) of brass on the wooden roller, and the wire from the other end of the coil to the other semi-circumference. This is done at the end or underneath, not at the top, or it will make the surface rough, and we want it to be as smooth as it can possibly be. The wire must be quite tight up to the end soldered on; there must be no loops, or it will catch in something and be torn off when it comes to revolve.
Fig. 10.—Section of Commutator Put Together.
P P, Pillars supporting springs, S S, which bear respectively on the upper and under sides of the roller, which is covered with brass except for the slits shown in the diagram.
The brass pillars supporting the springs have now to be inserted in the base, at such a distance, one on each side of the roller covered with brass, that the copper springs at the end of the brass ones are exactly one over and one under the brass roller. Of course, if they are put in a line with it, the springs can easily be shifted to the right position by slipping the slits over the screws of the pillars, and screwing down the nuts lightly when they come to the right place. This is very difficult to make intelligible, and I give another illustration of the relative positions of the parts of the commutator which I hope will make all clear. The pillars P P—which were put together as shown in Figs. 8 and 9—are fixed at such distances on opposite sides of the roller R that the springs S S are continually in contact with the brass semi-circumferences, first one and then the other as the armature revolves.
We are now within sight of the end of our task, and to guide off the current that we are going to produce we must screw in two binding-screws at opposite corners of the same end of the base (the end at which the commutator is). The ends of the wire from the magnet are to be brought down through the base and joined to the under part of these binding-screws. Placing the base so that the commutator end of the armature, and not the pulley end, is next to you, the wire from the inner coil of the magnet goes to the binding-screw on your left hand, and that from the outer coil to that on your right hand. The magnet should be wound and placed in such a position that these ends are respectively on the left and right, and then they have only to be joined to the binding-screws in front of them.
But before connecting these wires up, it is necessary to give an initial magnetism to the magnet, which at present has not been magnetized at all! To do this we must make use of another dynamo or a battery and connect the wires coming from the magnet-coil to the terminals of the battery. This having been done, the magnet will attract iron filings or needles, etc., and this shows that it has really become a magnet. Two cells of the chloride battery will be enough to magnetize it as much as it can be magnetized, and enough will remain when the battery is disconnected to start the action when the armature is revolved. Two or three minutes is long enough to connect with the battery.
PART III.
While the current is passing you can try the following experiment, to prove that the wire is wound on all right. If it is not wound as described there will be two north poles or two south poles, instead of one north and one south. Suppose we decide to make the leg on which the wire comes from the outside of the magnet the north pole, the wire from this must be joined to the wire coming from the zinc end of the battery, and the other coming from the inside, between the poles, joined to the wire from the carbon end. Now if, while the current is passing, a magnetized needle is approached to each pole consecutively, and one end of it is attracted and the other repelled in each case, the wire is all right; if both are attracted something is wrong. The needle must have been really magnetized beforehand, or it will deceive you; you can easily test if it is so with an ordinary permanent magnet.