CONQUEST OF DISTANCE BY WIRE TELEPHONY 381 



better type of filament, providing a more profuse emission of electrons and a 

 much longer life, could be used in place of deForest's tantalum filament. 

 All in all, Arnold painted an exceedingly intriguing picture regarding the 

 practically certain prospects of developing a really good telephone amplifier 

 from the deForest Audion. There was, of course, considerable chagrin 

 that these prospects had not been recognized much earlier in the Telephone 

 Company's research work on repeaters, but no time was wasted in attempts 

 to develop alibis. Arrangements were made for Arnold to spend most of 

 his time on the vacuum tube job and several assistants were provided. 

 From then on, the work on the mercury arc amplifier element slowed down. 

 To get full freedom in the development and use of the improved audion, 

 patent rights were purchased from deForest and later on, as commercial 

 use approached, it became desirable to obtain patent rights from other 

 outside inventors, American and foreign, who had been working on elec- 

 tronic repeaters. 



Arnold was the first worker in the electronic field to determine the physi- 

 cal laws of operation of the 3-element high vacuum tube, this being his 

 initial personal contribution to the development. In the concurrent and 

 subsequent experimental work it was found that the grid condenser of the 

 deForest circuit was a basic factor, along with the previously mentioned blue 

 haze, in the paralysis of the Audion as an amplifier, when large input cur- 

 rents of the magnitudes involved in wire telephony were employed. Under 

 these conditions it was found that the grid condenser acted as an electron 

 trap which, by piling up a negative charge on the grid, would cut off the 

 plate current and block the tube, even if the vacuum should be high enough 

 to prevent blue haze phenomena. Consequently, the condenser was 

 eliminated from the grid circuit. For some time, a grid leak was substituted 

 (a very high resistance, grid to plate). This was subsequently replaced by 

 a battery inserted in the grid circuit to maintain the grid at a positive po- 

 tential relative to the filament; this held the grid impedance to a definite 

 value and improved stability. Early in 1914, Arnold began using a negative 

 "C" battery in the grid circuit and thereby increased the sensitivity and 

 the stability. This led to the potentiometer input method of controlling 

 gain. Still later, it was learned that Lowenstein had anticipated Arnold in 

 the use of the negative C battery, and patent rights were obtained. 



In carrying out the development of the vacuum tube repeater by "methods 

 of pure science which were brought to its study in the spirit of research," 

 many workers were involved and a host of problems had to be sohed. A 

 new manufacturing art had to be created in the research laboratory. Op- 

 timum conditions and means for connecting the vacuum tube elements into 

 the repeater circuit had to be worked out. From the beginning, these 

 involved the 22-type repeater circuit, discussed later on. 



