316

INVENTION'S OF NIKOLA TESLA.

value, as it is quite easy to adjust the dimensions of the circuitso that oscillation is maintained. The employment of a steadye. m. f. in the high frequency conversion affords some advan-tages over the employment of alternating e. m. f., as the adjust-ments are much simpler and the action can he easier controlled.But unfortunately one is limited by the obtainable potential dif-ference. The winding also breaks down easily in consequenceof the sparks which form between the sections of the armatureor commutator when a vigorous oscillation takes place. Besides,these transformers are expensive to build. It has been found byexperience that it is best to follow the plan illustrated at ina.In this arrangement a rotating transformer g, is employed toconvert the low tension direct currents into low frequency alter-nating currents, preferably also of small tension. The tensionof the currents is then raised in a stationary transformer t. Thesecondary s of this transformer is connected to an adjustable con-denser c which discharges through the gap or discharger dd, placedin either of the ways indicated, through the primary p of a dis-ruptive discharge coil, the high frequency current being obtainedfrom the secondary s of this coil, as described on previous occa-sions. This will undoubtedly be found the cheapest and most con-venient way of converting direct currents.

The three branches of the circuit a represent the usual casesmet in practice when alternating currents are converted. InFig. lb a condenser c., generally of large capacity, is connected to thecircuit l containing the devices 11, mm. The devices mm are sup-posed to be of high self-induction so as to bring the frequency ofthe circuit more or less to that of the dynamo. In this instancethe discharger d d should best have a number of makes and breaksper second equal to twice the frequency of the dynamo. If notso, then it should have at least a number equal to a multiple oreven fraction of the dynamo frequency. It should be observed,referring to i b, that the conversion to a high potential is alsoeffected when the discharger d d, which is shown in the sketch, isomitted. But the effects which are produced by currents whichrise instantly to high values, as in a disruptive discharge, areentirely different from those produced by dynamo currents whichrise and fall harmonically. So, for instance, there might be in agiven case a number of makes and breaks at d d equal to justtwice the frequency of the dynamo,'or in other words, there maybe the same number of fundamental oscillations as would be pro-