1906 · Three-Electrode Vacuum Tube · Work 804 of 1,871
A small glass valve of 1906 on a workbench in a workshop in New York at night, a round clear glass bulb five centimetres across with a glass neck at each side, inside it a glowing filament, a small flat rectangular metal plate set at a slant and a finely waved wire grid, from one neck a thick wrap of black insulating tape with two black rubber cables, from the other neck thin wires in red and green insulation running to a headset with a leather band, a hand in a white shirt sleeve holding the headset, warm kerosene lamplight from the left.
What it was used for
As described by John L. Hogan Jr., who worked in de Forest's laboratory, a Grid Audion was usually a 6-volt, low-candle-power incandescent lamp with a tantalum filament, a platinum plate of about 10 x 15 mm roughly 3 mm from the filament, and a platinum grid wire of about number 22 between them; Hogan counted six or more distinct varieties. Hogan treats the tube as a detector of electric waves. Hogan thought it suited to wireless telephony because, carefully adjusted, it reproduced the higher harmonics of the voice.
What was new
In 1908 Hogan explains the action as the shattering of a column of conducting gas by the received impulse; the glass globe shields that gas column from the air currents that had made flame and arc detectors unreliable. He calls the Grid Audion the most sensitive type designed so far. The barb is in the same article: the sensitive condition was hard to find and harder still to maintain, some tubes lost their quality quickly in use, tests by government and industry had been unsatisfactory, and makers of wireless telephone apparatus saw almost no demand and supplied crystalline thermo-electric detectors instead. Hogan judges the tube not yet dependable enough for serious radio communication.
What the sources leave open: The article "The Audion; A Third Form of the Gas Detector" (Modern Electrics, October 1908, pp. 232-233, correction November 1908, p. 275) is by John L. Hogan Jr., not by de Forest; the judgments are therefore attributed to Hogan. No amplifying action appears in this text. The Audion patent numbers were not verified here and are omitted.
Connections
Every connection names its kind. “Forerunner” on its own would be a claim.