IDEAL CONTINUOUS CURRENT DYNAMO.

59

tached, and the brush B, touching the half cylinder to which the wire g is attached. But, at the same time, the direction of electro-motive force in the two wires has been reversed, the wire c d entering the right-hand side of the field, and a b entering the left-hand side. Consequently the external current flows in the same direction as before, growing from zero to a maximum when the crank stands horizontally on the left, and again diminishing to zero when it is vertical. Graphically represented, the current is of the character shown by the curve, Fig. 15, the abscissæ being consecutive angles of the crank, and the ordinates being proportional to the sines of these angles.

Fig. 15.

2T

It should be noted that the reversal of current always takes place when the electro-motive force is zero, and consequently the change in the contact with the brushes from one commutator plate to the other takes place without sparking. To increase the power of the machine, we can replace the single rectangle of wire by a coil of many turns. Fig. 16. Hitherto we have tacitly assumed that the space contained within the wire coils forming the armature contains air or other non-magnetic substance. The lines of force passing between the polar surfaces SN have to leap across a considerable air space, and if by some means we could shorten that portion of their path which lies entirely in air, we would facilitate the flow of lines and increase the strength of the magnetic

field. Roughly speaking, we may take it that air offers to the lines of force about 700 times the resistance of iron, and if we can contrive to fill part of the space between the polar surfaces with iron, a considerable increase of electro-motive force, and consequently of current, will be the result. The space available for this purpose is that contained within the armature coil; in other words, to increase the power of the machine we must wind the armature coils over an iron core. An early dynamo constructed on this principle is that of Siemens, invented in 1855, and provided with the so-called shuttlewound armature. The core consists of an iron cylinder Fig. 16.

provided with two deep longitudinal grooves placed opposite so that the cross-section resembles a double T with rounded heads. The wire is wound into these grooves, and the two ends of it are joined to the plates of a twopart commutator. Fig. 17 shows a cross-section of this armature. In the first machines the core was in one solid piece, but it was found to heat considerably on account of internal currents. It is well known that if a solid body of metal be rapidly rotated between two powerful magnet poles it becomes hot. The reason for this phenomenon is that the outer portions of the metal in cutting through the lines of force become themselves the seat of electro-motive forces acting at right angles to the direction of motion and to the lines, and powerful

SHUTTLE-WOUND ARMATURE.

61

currents are started parallel to the axis which run in opposite directions, up on one side and down on the other side of the axis. In a solid armature core there is nothing to check the flow of these currents but the resistance of the metal, which, on account of the large cross-sectional area, is extremely low. These wasteful currents are consequently very strong, and not only absorb much power, but they also weaken' the current generated in the copper wire by induction. To avoid their creation, it is necessary to subdivide the mass of the core by planes at right angles to the axis, and to insulate as much as possible the subdivided portions from each Fig. 17.

SIEMENS SHUTTLE-WOUND ARMATURE.

other. This can be done either by cutting deep narrow circular grooves in the core, or by building it up of thin discs insulated from each other either by paper discs or by being coated with some insulating paint. These armatures are not much used for dynamos at the present day, having been replaced by more perfect forms to be described presently; but they are still extensively employed for small electro-motors. By referring to Fig. 15 it will be seen that the counter-electro-motive force of these motors is a variable quantity depending on the angular position of the armature. If the heads of the double T core are opposite the field magnet poles, the coil is at right angles to the lines of force and the counter-electromotive force is zero. This happens precisely at the moment

when the brushes touch simultaneously both plates of the commutator, and are therefore short circuited. A current sent through the motor while at rest in this position cannot start it, and this condition is expressed by saying that the armature has two dead points. When at work the momentum of the armature is sufficient to carry it over the dead points, and, apart from the inconvenience to have to start the motor occasionally by hand, these dead points present no mechanical imperfection. But it might be thought that they present a serious electrical imperfection for the following reason : The strength of the current which is allowed to pass through the motor at any given moment depends partly on the electrical resistance of the motor, and partly on its counter-electromotive force at that particular moment. But since at the dead points there is no counter-electro-motive force, the strength of the current will be a maximum, whilst at those moments the mechanical energy produced is nil. We assume here that the motor is fed by a current flowing under a constant electro-motive force, which is the case most commonly met with in practice. We have now to distinguish between two cases: the motor may be either series wound or shunt wound. If the former, the current in passing through the motor whilst the armature is at a dead point has only to overcome the resistance of the field magnet coils. If the armature is in the position of greatest counter-electro-motive force the current has to overcome not only that, but also the combined resistance of field magnet and armature coils. In that position the mechanical energy of the armature is at its greatest value, but the strength of the current is a minimum. We find, therefore, on the one hand, that the strength of the field magnets (which depends on the current) is least at

EFFECT OF SELF-INDUCTION.

63

the very moments when the armature is in a position to exert most power, and on the other hand, that it is greatest when the armature is at its dead points and cannot exert any power. From the foregoing we should expect that twice during each revolution a great waste of current must take place when momentarily the brushes are short-circuited by the commutator. Although the time during which such short circuits lasts may appear to our senses very brief, it would in comparison with the speed of electric phenomena be still considerable, and have an appreciable effect on the economy of the motor. But there is one circumstance which greatly tends to mitigate the evil effect of the dead points just described, and this is the property of electric currents called self-induction. It can best be described as a kind of inertia opposing any sudden change in the strength of the current. If a circuit contains a coil of wire surrounding iron, as in the present case, the field magnets, the self-induction is so great that it requires an appreciable time to change the strength of the current. The increase of current at the dead points is, therefore, checked by this property of selfinduction, and the current, instead of being subjected to abrupt and violent changes, becomes simply undulatory. The case is different if the motor be shunt-wound and fed from a source of constant electro-motive force. Since the field magnet coils are excited independently from the current which passes through the armature, their selfinduction cannot in any way steady that current, and abrupt changes in its strength and great waste of electrical energy must occur at the dead points. This is a matter of considerable practical importance, and shows that motors with shuttle-wound armatures should never be used coupled up otherwise than armature and field