Зображення сторінки
PDF
ePub

dynamo is regulated to the proper voltage, the main switch Mis then closed. On the blade of the latter there is a catch which engages with the blade of the field switch F and locks them together, also forming an electrical contact. When the main switch

Zox

B+

B

F

T

M N

M is opened in order to throw the machine out of circuit, it brings the blade of the field switch with it, the two remaining in electrical connection; the field being now selfexcited dies away gradually as the dynamo slows down. The switch N may then be opened in order to entirely disconnect the machine from the circuit. When the dynamo is required again, the catch that locks the two switches together is withdrawn, and the field switch is closed independently, the other steps having already been described. This method combines the advantages of self and 'bus excitation without involving complication. The 'bus bars B + and B – in Fig. 32 may be indefinitely extended in either direction; and any number of dynamos may feed current to them, the connections and operation being the same for each. If there happened to be no current on the 'bus bars, owing to accident or the stopping of all the machines, any one of them could be started as a self-exciting dynamo by closing the switches F and M.

Fig. 32.

A

C

Donshea Method of Shunt Field Excitation.

The excitation of a compound generator is conveniently and effectively accomplished by causing current from the other machines to flow through its series coil; for example, the switches E and F in Fig. 30 are first closed, as already explained. But a compound dynamo working alone would have to build up its own magnetism the same as a series or shunt machine. In the case of the last named, it excites more readily the higher the resistance in the external circuit, hence the latter should be opened if possible. But with a series or compound the excitation is facilitated by shortcircuiting the main terminals so that a strong current will flow in the series field coils. This should be tried, however, only when the machine refuses to generate under normal conditions; and the

short-circuit should be carefully applied for brief periods, otherwise a very excessive current may be produced. The causes and remedies for a dynamo failing to generate are given in Vol. I., p. 362, and more fully in Practical Management of Dynamos and Motors by Crocker and Wheeler, p. 155.

Feeder Regulation. It has been stated on page 51 that the ordinary practice in the smaller electric-lighting plants is to operate the generators at a fixed voltage, or to supply a somewhat higher voltage as the load in amperes increases, in order to make up for the drop in pressure on the conductors. The rise in voltage is produced by hand regulation, over-compound winding, or by an automatic regulator, as already described. In large systems of distribution the various feeders (Fig. 12) may differ greatly in length, and may be very unequally loaded, the latter condition being in a continual state of change. It therefore becomes necessary to independently control the different feeders, for which purpose several methods are employed, as follows:

1. Connecting and disconnecting feeders at various points on the mains; 2. Feeder rheostats; 3. Auxiliary 'bus bars; and 4. "Boosters."

The first of these methods is represented in Fig. 33, in which a pair of mains, P Q and R S, are supplied with current from the

[blocks in formation]

generator D by means of feeders A, B, and C, the corresponding return feeders being shown below. It is assumed that there are considerably more lamps burning at J and L than at K, hence the voltage would be higher at K than at J and L, if all three feeders were in circuit. This difference will be reduced by opening the switch F, thereby disconnecting the feeder B, as indicated. The

small amount of current required for the few lamps at K will be supplied from both directions by the feeders A and C with very little drop. If nearly all the lamps at L were turned out, the feeder A could also be disconnected by opening the switch E; but if every lamp were put in use, or if the load on the mains happened to be uniform, then all three switches, E, F, and G, should be closed.

The distribution of potential on uniformly loaded mains supplied by feeders at symmetrical points is illustrated in Fig. 34. Each of the mains E K and S T is supposed to consist of 1,800 feet of No. 2 (A. W. G.) copper wire, and supplies ten equidistant groups of lamps, each taking ten amperes. These mains are fed with current from the generator D by feeders A and B, which lead to points 400 feet from each end. The portion EFG of the mains

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

Fig. 34. Uniformly Loaded Mains and Uniformly Distributed Feeders.

is identical with the arrangement shown in Fig. 16, and so is the portion HJK; hence the distribution of potential will be the same as in that case, and is represented in Fig. 35, the highest pressure being 111 and the lowest 109 volts. With a perfectly symmetrical arrangement of the lamps, there will be no flow of current in either direction between G and H; hence those sections of the mains could be removed without affecting the electrical conditions. But in case there were more lamps at one end than at the other, then there would be a transfer of current through GH, which would tend to equalize the pressure. The feeder A is assumed to consist of 1,600 feet of No. 0000 wire having a resistance of about .08 ohm, and carrying 50 amperes; since it supplies 5 groups of lamps, each taking 10 amperes. The drop upon it is 50 x .08 = 4 volts, and the same amount for the return feeder a; consequently the potential at the generator D must be 111 + 4 + 4 = 119 volts.

If the other feeder B consists of 600 feet of No. 1 wire, its resistance will be about .075 ohms, which is approximately the same, and would cause a similar drop. In this instance the resistances of two feeders are designed to be about equal by making their

[merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small]

cross-sections proportional to their lengths. If the feeder B were made of the same size wire as A, the drop upon the former would volts, in which case it should be

only be 100% as much, or 1

[ocr errors]

supplied with 111 + 11⁄2 + 13 =

114 volts instead of 119, and one of the following methods of feeder regulation would be required. The same statement applies if the currents in the feeders differ considerably when their resistances are approximately equal.

Feeder Regulation by Resistance. In such cases a feeder rheostat, or "feeder equalizer," consisting simply of a variable resistance, may be placed in series with each feeder, as represented in Fig. 36, the current capacity of the rheostat being sufficient for the maximum current conveyed by the feeder. In operating such a system a certain amount of resistance R is introduced into the circuit of feeder A that is lightly loaded, in order that the pressure which it supplies to the mains shall not be too high compared with that of the more heavily loaded feeder B, which has less resistance inserted. Thus by adjusting the arms of the rheostats R and S, the voltage at the ends of the feeders A and B may be made equal for all loads, or the potential at the outer end of B may be raised a little above that of A, in order to make up for the greater drop in the mains at H than at F.

The voltage at the farther ends of the feeders may be determined by running extra conductors, WW, called "pressure wires," from the generating-plant to the point at which the feeder is con

nected to the mains. The actual potential is read directly on the voltmeter V, since the current in the wires W W is so little that there is no appreciable drop upon them even when they are quite small. Another method consists in subtracting the drop IR on the feeders from the voltage Vat the generators; that is, the potential at the ends of the feeders P = VIR, in which I is the current and R the resistance of a given pair of feeders. For this purpose an amperemeter may be put in series with each pair of feeders, and it can be calibrated to give the drop upon them by simply multiplying its scale numbers by the total resistance of the two feeders. A still more perfect device is the so-called compensated voltmeter. This has the ordinary coil which measures the voltage of the generators, and an additional coil that carries a cer

[blocks in formation]

tain fraction of the feeder current, the effect of the latter being to slighty oppose the former, so that it reduces the deflection of the pointer, and indicates the pressure at the outer end of the feeder. The objection to the resistance method of regulation is the loss of energy that it entails, the amount of which is represented by the percentage that the voltage supplied to a given feeder is lowered. The present practice tends toward the use of the two following methods of feeder regulation in preference to the two already described, particularly in the larger systems.

Auxiliary 'Bus Bars are often employed in stations or plants of considerable size, in order to avoid the loss of energy which inevitably occurs when "dead" resistance is used for regulation. This method is represented in its simplest form in Fig. 37, C and D being two dynamos, one of which, D, is connected to the main 'bus bar F, and generates the ordinary potential required to supply the shorter feeders, or those that are lightly loaded, such as B.

« НазадПродовжити »