TRANSACTIONS OF SECTION G. 565 
FRIDAY, SEPTEMBER 6. 
Joint Discussion with Section A on Wireless Telegraphy.—See p. 401. 
The following Papers were then read :— 
1. The Production of Electrical Oscillations by Spark-Gaps ummersed 
in Running Liquids." By W. H. Eccrms, D.Sc., and A. J. 
Maxower, M.A. 
The most common method of generating electrical oscillations for the purposes 
of wireless telegraphy is by the discharge of a condenser through an inductance 
coil and air-gap in series. When a large amount of energy has to be handled 
there is a tendency for a permanent arc to form between the electrodes, and this 
reduces the efficiency of the apparatus. In the present paper is described a series 
of experiments designed to determine whether the efficiency can be increased by 
making the spark occur in a flowing liquid. The liquids employed in the experi- 
ments were water and transformer oil. Although the results show that the 
efficiency attained is only of the same order as that which can be attained by 
the air-spark, the details of the experiments are of interest and suggest methods 
that would be useful in certain contingencies. 
The simplest form of discharger worked with consisted of an ebonite tuhe 
through the sides of which the electrodes protruded to form a small spark-gap 
in the middle of the bore. When water was passed down the tube, very steady 
oscillatory current could be obtained with voltages of about one thousand volts 
across the gap. When oil was used the voltages had to be of the same order as 
those employed with air-gaps in order to obtain the best results, and, at the 
same time, the gaps had to be smaller than air-gaps. Other forms of dis- 
charger were made which provided for the spark to take place under great 
hydrostatic pressures, but these showed no consistent electrical advantage over 
the dischargers operated with liquids at atmospheric pressure. In the paper 
damping curves are given for various degrees of coupling between the primary 
and secondary oscillatory circuits. These show that the water spark follows 
nearly the same laws as an air-spark, and that the spark in oil exhibits the 
phenomenon known as impact excitation. Dischargers of the type described 
possess the advantage over the open air-spark that their discharge can be made 
practically noiseless. 
2. The Impedance of Telephone Receivers as affected by the Motion of 
their Diaphragms.* By Professor A. E. KENNELLY and G. W. Pierce. 
The authors made a research on the effect of frequency upon the electrical 
properties of various telephone-receivers, between the limits of 430 and 2,400 
cycles per second. The Rayleigh modification of the Wheatstone bridge was 
used, and the measurements were conducted with constant voltage (0°3 to 10 
volt in different series) at the telephone-receiver terminals. 
It was found that the vibration of the telephone diaphragm materially affected 
the resistance and inductance of the receiver. Measurements were therefore 
made both with the diaphragm arrested, or damped, and with the diaphragm 
free. In the instruments tested, the ‘damped resistance’ was found to be a 
simple quadratic function of the frequency, and also to be in inverse ratio to the 
inductance, over a considerable range of frequency. 
The ‘damped impedance’ of a telephone-receiver, when plotted with react- 
ance ordinates and resistance abscisse, increases when the frequency is increased, 
advancing regularly along a smooth curve. The ‘free impedance,’ however, 
moves around a closed loop, in that portion of the frequency range lying in the 
neighbourhood of the fundamental frequency of the diaphragm. There are thus 
1 See 7'he Electrician, September 13, 1912. 2 Tbid, 
