VI. Reactive Currents and Compounding 96. Since the polarization due to the power component of the alternating current as synchronous motor is in quadrature ahead of the field magnetization, the polarization or magnetizing effect of the lagging component of alternating current is in phase, that of the leading component of alternating current in oppositon to the field magnetization; that is, in the converter no magnetic distortion exists, and no armature reaction at all if the current is in phase with the impressed e.m.f., while the- armature reaction is demagnetizing with a leading and mag- netizing with a lagging current. Thus if the alternating Current is lagging, the field excitation at the same impressed e.m.f. has to be lower, and if the alter- nating current is leading, the field excitation has to be higher, than required with the alternating current in phase with the SYNCHRONOUS CONVERTERS 251 e.m.f. Inversely, by raising the field excitation a leading current, or by lowering it a lagging current, can be produced in a converter (and in a synchronous motor). Since the alternating current can be made magnetizing or demagnetizing according to the field excitation, at constant impressed alternating voltage, the field excitation of the con- verter can be varied through a wide range without noticeably affecting the voltage at the commutator brushes; and in con- verters of high armature reaction and relatively weak field, full load and overload can be carried by the machine without any field excitation whatever, that is, by exciting the field by armature reaction by the lagging alternating current. Such converters without field excitation, or reaction converters, must always run with more or less lagging current, that is, give the same reaction on the line as induction motors, which, as known, are far more objectionable than synchronous motors in their reaction on the alternating system, and therefore they are no longer used. Conversely, however, at constant impressed alternating vol- tage the direct-current voltage of a converter cannot be varied by varying the field excitation (except by the very small amount due to the change of the ratio of conversion), but a change of field excitation merely produces wattless currents, lagging or magnetizing with a decrease, leading or demagnetizing with an increase of field excitation. Thus to vary the continuous- current voltage of a converter usually the impressed alternating voltage has to be varied. This can be done either by potential regulator or compensator, that is, transformers of variable ratio of transformation, or by a synchronous machine of the same number of poles as the converter, on the same shaft and con- nected in series ("synchronous booster") or by the effect of watt- less currents on self-inductance. The latter method is especially suited for converters, due to their ability of producing wattless currents by change of .field excitation. The e.m.f. of self -inductance lags 90 deg. behind the current; thus, if the current is lagging 90 deg. behind the impressed e.m.f., the e.m.f. of self-inductance is 180 deg. behind, or in opposition to, the impressed e.m.f., and thus reduces it. If the current is 90 deg. ahead of the e.m.f., the e.m.f. of self-inductance is in phase with the impressed e.m.f., thus adds itself thereto and raises it. Therefore, if self-inductance is inserted into the lines between converter and constant-potential generator, and a watt- I 252 ELEMENTS OF ELECTRICAL ENGINEERING less lagging current is produced by the converter by a decrease of its field excitation, the e.m.f. of self-inductance of this lagging current in the line lowers the alternating impressed voltage at the converter and thus its direct-current voltage; and if a watt- less leading current is produced by the converter by an increase of its field excitation, the e.rn.f. of self-inductance of this leading current raises the impressed alternating voltage at the converter and thus its direct-current voltage. 97. In this manner, by self-inductance in the lines leading to the converter, its voltage can be varied by a change of field excitation, or conversely its voltage maintained constant at constant generator voltage or even constant generator excita- tion, with increasing load and thus increasing resistance drop in the line; or the voltage can even be increased with increasing load, that is, the system over-compounded. The change of field excitation of the converter with changes of load can be made automatic by the combination of shunt and series field, and in this manner a converter can be compounded or even over-compounded similarly to a direct-current generator. While the effect is the same, the action, however, is different; and the compounding takes place not in the machine as with a direct-current generator, but in the alternating lines leading to the machine, in which self-inductance becomes essential. As the reactance of the transmission line is rarely sufficient to give phase control over a wide range without excessive reac- tive currents, it is customary, especially at 25 cycles, to insert reactive coils into the leads between the converter and its step- down transformers, in those cases in which automatic phase control by converter series fields is desired, as in power trans- mission for suburban and interurban railways, etc., or to specially design the step-down transformers for high internal reactance. Usually these reactive coils are designed to give at full-load current a reactance voltage equal to about 15 per cent, of the converter supply voltage, and therefore capable of taking care of about 10 per cent, line drop at good power-factors.