Page:1902 Encyclopædia Britannica - Volume 27 - CHI-ELD.pdf/630

 578

DYNAMO

ployed in many of the earliest machines, and especially in alternators, but has not met with much favour for continuous-current dynamos. In this method the outer end of inductor 1 is joined, by a connecting wire passing round the periphery, to the outer end of another inductor 2' situated nearly diametrically opposite under the other pole (Fig. 9). No iron core is now required; the two opposite poles can be brought close together, leaving but a short path for the lines in the air-gap, and any mechanical support on which the wires are wound may be nonmagnetic. By bringing the inner end of inductor 2' along the inside periphery (Fig. 11), a loop is formed, which may be again repeated. If the ends of the loops

thus formed by either method (3) or (4) are connected to a pair of collecting rings, a simple discoidal-ring or disc alternator is obtained, while if they are connected to a split-ring the current may be commuted, and the dynamo gives a unidirected E.M.F. and current in. the external circuit. It is therefore true to say that in almost all dynamos which are used in practice, whether they supply an alternating or a continuous current to the external circuit, the E.M.F. and current in the inductors themselves are alternating. The addition of new loops leads to a coil of many turns, and thence to a number of coils. If the several new loops are wound over each other in the ring, drum, and discoidal-ring armatures, or by the side of each othei in the disc, the depth of the wires in the air-gap would necessitate such a long distance between the pole-piece and the armature core, or between the opposing poles of the disc, that it would be difficult to obtain a sufficiently strong magnetic field. The additional loops must be wound side by side in the first three cases, or on the top of each other in the disc armature. The coil thus has a certain width along the face of the poles, and the effect of such width must be considered in its relation to the width of the poles. The pitch-line being defined as a circle drawn through the middle of the length of the inductors, or, in other words, as the mean path traversed by them in their rotation, it is convenient to refer the widths of pole, interpolar gap, and coil to the pitch, or the distance measured along the pitch-line from the centre of one pole to the centre of a neighbouring pole of opposite sign. In the ring armature, if the E.M.F.’s induced by the turns in any one coil are to be added in series, they must be in the same direction along the inductors. A ring core entirely overwound with one coil in a continuous helix would, give no E.M.F. at the ends; but when the coil is divided into two halves, these may, by means of appropriate connexions, be placed either in series or in parallel, and each half will at any instant give the same E.M.F. The ring armature must therefore be divided into at least two coils, and it is then evident that if the width of each coil exceeds the width of the interpolar gap, the loops will for some portion of one revolution be moving at the same time under two poles of opposite sign, and consequently differential action will be set up, by which one part of the coil will be

inducing an E.M.F. opposed in direction to that of the rest. If this disadvantageous action is to be avoided, either the width of the coil or the width of the pole, or both, must be reduced until the coil is no wider than the interpolar gap. If the coil be wide, the poles must be narrow, so that with a given density the number of lines in a field will be very small; while, on the other hand, if the coil be narrow and the pole wide, the flux of one field may be large, but the number of inductors in the coil must be small. If the poles of the alternator are arranged in two separate circles (as in Fig. 43), all the N poles forming one crown and all the S poles another crown, the width of the poles may be equal to the pitch ; but if they occur alternately, first a N pole and then a S pole round the same circle, their width must be less than equal to the pitch in order to avoid an excessive amount of magnetic leakage between the adjacent pole-edges. Considerations of the inductance of the armature and of its heating in turn limit the feasible width of winding in the armature coils. A compromise has therefore to be made, and experience shows that in practice the best combined use of an armature core and field-magnet of given size is made if the width of the poles of alternate sign is approximately half the pitch, and the width of the ring coil is equal to the ■Fig. 12. width of a pole. The two-pole machine will then have two coils, and the armature core will be only half covered with winding (Fig. 12 i.). The same considerations govern the best width of a drum coil. If the core be entirely overwound (as in Fig. 12 ii., if the blank portions of the core were covered with loops of continually diminishing size), the arrangement is the equivalent of the ring armature with two coils in series, each of width equal to the pitch. The single coil of the bipolar drum may equally well be divided into two coils, the largest loop of each half having a width equal to the pitch ; and the two coils so formed may be connected either in series or in parallel. Again, however, if there is to be no differential action, opposite sides of one coil must never be moving under the same pole, and the innermost or smallest loop must not be less in width than the pole-face. The best proportions are obtained if the width of the inner loop and the width of the field are each approximately equal to half the pitch (Fig._ 12 ii.). The same reasoning can easily be applied to the discoidal ring and disc machines, and will lead to like results. If the bipolar armatures of Fig. 12 are imagined to be cut through to the centre from any point X and opened out, it is evident that any number of pairs of coils similarly spaced to the original pair may be inserted without in any way a,ffecting the action, save to increase the periodicity of the alternating E.M.F. ; an equal number of pairs of poles may then also be added, with a corresponding increase in the total E.M.F., and the multipolar alternators^ of Fig. 13 are obtained. As the wires of a coil pass gradually into or out of action, the total E.M.F. of the coil, being the sum of the E.M.F.’s in the separate inductors, rises or falls gradually, the maximum value being reached when the coil is immediately under a pole in the ring armature, or when the side of a coil is under a pole in the drum and disc machines. vVhen the coil enters under the next pole of opposite sign, the gradual rise and fall are again repeated, but the E.M.F. is in the opposite direction. Thus the passage of a coil through two magnetic fields of opposite direction yields a complete wave of E.M.F. such as is shown in Fig. 6, and the time in seconds taken by the E.M.r. or current to pass through such a complete cycle is the “period of the alternator. The number of complete periods through whicn the E.M.F. or current passes per second is called the periodicity or frequency, and in the bipolar alternator is equal to the number of revolutions per second. In the multipolar alternators of rig. 13 the coils may be arranged either in series or in parallel, since their phases are all alike ; the periodicity of the multipolar alternator is with either arrangement equal to the number of pairs o fields through which a coil passes in one second, so that in genera the periodicity^ x.p, where N = the number of revolutions per 1 The first use of a number of pairs of coils and an equal number of airs of poles was due to Stohrer of Leipzig, Pogg. Ann. Ixi. 41/, io**.