Page:Popular Science Monthly Volume 92.djvu/973

 Wireless Work in Wartime

XI. — Radio transmitters using synchronous and quenched gaps

By John V. L. Hogan

��IN last month's article the non-syn- chronous operation of a rotary gap in the wireless transmitter of Fig. 41 was decribed, for conditions which gave two or three sparks for each half-cycle of alternating current power. The curves of condenser discharge are shown in

���Power Imnsf.

��fig.+l

��A diagram of a radio transmitter with a rotary spark gap interposed in the apparatus

Fig. 42, where the divisions along the horizontal line represent six-hundredth parts of a second. Since the dashed wavy line shows the voltages at which the gap will permit a spark to jump, as time goes on, and the solid wavy line indicates the potential available (in the condenser) to produce a spark, it is evident that the discharge must pass whenever the two curves cross.

If we now adjust the studs so that they are somewhat nearer together, permitting the spark to pass at a lower voltage (or if we raise the maximum charging potential to a higher value), it is clear that the overlaps will occur more often and that it will thus be possible to secure four sparks in each half cycle of charging current. By proper selection of the break-down and charging voltages, by changing the wave-form of the charging voltage, and by using a power transformer which will put energy into the condenser quickly after each spark passes, it is possible to get a large number of fairly regular sparks per second with only a low fre- quency of alternating current power.

��The curves of Fig. 42 are not complete, since the secondary condenser voltage will be bound to be reduced by the with- drawal of energy for each spark; never- theless, a sufficiently "quick" or closely- coupled power transformer will build it up again before the next sparking time, so that the general conditions will be as indicated.

For every spark there will of course be produced a group of radio frequency alternating currents in the oscillation circuit. With the non-synchronous meth- od of operation these will not be of the greatest power obtainable from the same amount of input energy. This is because the condenser is not dis- charged at the instant it has been filled to its fullest (maximum potential) point for each spark. It has been pointed out that for a given capacity, the amount of charge depends upon the potential; obviously, then, if the condenser is dis- charged through the spark gap at a voltage of 7500 there will be less energy for conversion into oscillations than if the charge is held until the full potential of 10,000 volts is reached. The greatest utility of the non-synchronous method lies in the fact that with it one is able

���Fig. «

��Time

��Curves showing the operation of the non- synchronous spark gap for a wireless set

to secure a fairly good and moderately high-pitched spark tone from low fre- quency alternating current, even though at some sacrifice of conversion efficiency. By adjusting the gap for best regularity of operation, with the fewest possible

��9.37

�� �