moving coil is in series with and carries the same current as the fixed coil, whence the angle of torsion is proportional to the √mean mean square value of the alternating current, and in addition the use of a third Siemens electro-dynamometer, arranged so that the moving coil has its own separate terminals, and is not in series with the fixed coil.

If then two alternating currents of equal period, from either the same or different sources, flow through the two independent coils, the periodic time of oscillation of the moving coil being yery large compared with that of the current, the angle of torsion is proportional to the mean product of the simultaneous instantaneous values of current throughout the period, and is called the split dynamometer " reading.

66

=

If A and A ̧1 maximum values, and A, A1 the mean values of two simple periodic alternating currents, one of which lags behind the other by an angle a, then the ordinary dynamometer will give A = 4,2 and A1= 1⁄2(A ̧1)2. On passing these currents through the split dynamometer its reading & would be1⁄2Ã1 cos. a, and

The following method is quite general, and does not assume that the current is a simple sine function of the time, but does assume that there is no magnetic leakage, i. e. that the number of lines cutting the primary and secondary are the same. This is not true in all types of transformers in full load, but is nearly so in closed magnetic circuit types.

Since the split dynamometer gives no reading on open secondary circuit, this method is useless for determining the open circuit" losses.

Apparatus. — Two ordinary Siemens electro-dynamometers AA, and one split dynamometer (4); transformer T to be tested; non-inductive resistance L (such as a bank of lamps to take up the secondary load) (p. 395); alternator D; switches S1 S.; voltmeters V1 V2; non-inductive Wattmeter W, inserted merely for the purposes of comparison.

1 2

Observations. (1) Connect up as in Fig. 100, and adjust the instruments to zero where necessary. See that all lubricating cups in use feed slowly and properly, then start D.

(2) S being open, close S1, and adjust the speed and excitation.

so that V1 reads the normal voltage required for the primary at the normal frequency of the transformer. Note the readings of

A1, V1 and W.

10

(3) Close S2 and adjust L so that A, reads about of the maximum secondary current, the voltage V, being kept at 100 by

1

varying the excitation. Now note the readings of A, A1, A2, V1, V2 and W.

(4) Repeat 3 for about 10 secondary load currents, rising by about equal increments to the maximum allowable, and tabulate as follows

(5) Plot curves having values of Ws as abscissæ, with efficiencies

(121) Measurement of the Efficiency of Multiphase Alternating Current Transformers.

Introduction. The determination of the efficiency of ordinary single-phase transformers has already been fully considered in the preceding pages.

The present test does not differ materially in principle from those in question, and practically the only difference is in the method of measuring the power absorbed and developed by the multiphase transformer, and which possesses some characteristic differences from that used in the case of the ordinary singlephase transformer.

Most of the preceding methods are equally applicable in the present case whether the transformer is of the two or the three phase type. The reader should refer to p. 247 for the method of measuring electrical power in two and three phase alternating current circuits, where a more detailed description of them will be found. If in the present instance, as in fact with any others, the rheostats or circuits in which the load is to be absorbed are strictly non-inductive, i. e. are of the nature of incandescent lamps or water rheostats, then providing such load-absorbing devices operate equally on each of the sections of the circuit, thus maintaining a balanced system, the output can quite accurately enough be obtained from the ammeter and voltmeter readings in the manner set forth on pp. 247–251.

For the present test we will assume that the efficiency of a 3-phase transformer is required by, say, the single conversion method.

Apparatus.-The 3-phase transformer to be tested, of which P is the primary winding and S the secondary shown in Fig. 101, with the star or open winding; two non-inductive Wattmeters W1 and W; three Parr's direct-reading dynamometer ammeters A, A1 and A2; three voltmeters V, V1 and V2; 3-phase variable rheostat R (non-inductive) capable of operating equally on each line (p. 403); source of 3-phase current E; two 3-throw switches SSS and SS1 S1.

Note. If the 3-phase rheostat R is not non-inductive, then two additional Wattmeters will be necessary in the secondary

circuits connected up in precisely the same way as those shown in the primary circuit, the secondary output being then given by the sum of their readings at any particular load.

Observations. (1) Connect up as in Fig. 101, and adjust all the instruments to zero, levelling such as require it. See that all lubricating cups in use feed slowly and properly if the source of 3-phase current supply E is controllable.

(2) With S1 S1 S1 open, close SSS, and adjust the speed of the generator so as to give the proper periodicity for the transformer and then the excitation, so as to have the desired voltage, shown by Vacross the primary.

1

Note the respective Wattmeter readings W1 and W, and if possible that of A in addition to V. Then (W1+ W2) = the noload primary input = the magnetizing losses.

1

1

(3) With R at its maximum, close S1 S1 S1 and note the readings of all the instruments for some ten or twelve secondary loadcurrents from the smallest to the maximum permissible, rising by about equal increments at a time for constant secondary voltage.

(4) Calculate the secondary loads (Ws) from the relationWs= √3 A1V1 = √√3 AV1⁄2, etc.,

1

2 29

and tabulate as follows-

(5) Measure the resistance of the transformer coils by means of either the Wheatstone Bridge or Potential Difference method. (6) Plot the following curves between

Efficiencies ʼn as ordinates and secondary loads (√3 A1V1) as

abscissæ.

η

1

Power Factor as ordinates and secondary loads (√3 A1V1) as

abscissæ.

Inferences.-State clearly all that can be inferred from your experimental results.

(122) Efficiency and Output of Multiphase Alternating Current Rotatory Convertors.

Introduction. The rapid development of multiphase alternating current machinery, but perhaps more especially of that particular class of the same, known now commonly by the name of the Rotatory Convertor, marks one important epoch in the history of this all-important and ever-increasing branch of industry— Electrical Engineering. There are several different types of transformers, but all come under one or other of two main heads. (1) Static transformers or convertors with no moving parts. (2) Rotatory transformers or convertors having moving parts, and on which latter their very existence depends. The former of course include the ordinary every-day transformer which we are so accustomed to see.

The type 1 transforms electrical energy of one species at a