Skip to main content

Voice Over Radio

Fall 2010 | Volume 25 |  Issue 3

In 1900 the United States Weather Bureau hired 34-year-old electrical engineer Reginald Fessenden to develop a wireless system that could distribute forecasts and relay meteorological data. The Canadian-born inventor, a protégé of Thomas Edison, former consultant for Westinghouse, and professor at Purdue and Western universities, moved his family to Spartan accommodations at the Weather Bureau station at Cobb Island, Maryland, 60 miles southeast of Washington, D.C., in the Potomac River. After a year of hard work, Fessenden and his assistants had succeeded in transmitting Morse code 50 miles to Arlington, Virginia.

He also pursued the more difficult task of transmitting sound. His first efforts used the same “spark-gap” equipment that he and others were developing for wireless telegraphy. By operating at a higher frequency and improving the sensitivity of the components, Fessenden understood that he could transmit almost continuous waves (instead of dot-dash signals) and reproduce them in the receiver. The result would be a rapidly varying electric current that would duplicate the original sound when heard through telephonic headphones. On December 23, 1900, as darkness fell and a light snow dusted Cobb Island, Fessenden succeeded in making the first radio transmission of voice ever, sending a signal between two 50-foot-high wooden masts a mile apart.

In hopes of improving his wireless telephony apparatus, Fessenden looked for something to replace the coherer, a detector of electromagnetic waves that was part of all early radio setups. The coherer amounted to a tube of metal filings inserted within a circuit. If no radio waves were present, the filings were randomly oriented and had a fairly high resistance. But when the coherer was acted upon by a wave, the filings lined up and completed the circuit. The coherer worked better than any other wave detector, but it had numerous deficiencies, not the least of which was that it had to be tapped with a vibrator to decohere the filings.

Fessenden replaced the coherer with what he called a barretter, a very thin piece of wire. A radio wave induced a current in the wire, heated it, and increased its resistance. This “hot-wire” barretter, which took form during 1901, was no more sensitive than the coherer, but because it lacked the coherer’s on-or-off nature, it could reproduce speech much more efficiently.

Long-distance wireless telephony, however, required a more sensitive detector. He found an answer by accident in 1902 when he broke a barretter while cleaning it in nitric acid. Surprisingly, the broken wire worked much better than a whole one. He then designed a detector incorporating two extremely fine platinum wires whose ends were dipped into a pool of acid.

Though it required frequent maintenance, Fessenden’s “liquid barretter” proved successful, soon enabling him to broadcast musical notes between Roanoke Island and Hatteras. As word of his accomplishments spread, various U.S. and Mexican government agencies began placing orders for his apparatus.

We hope you enjoyed this essay.

Please support this 70-year tradition of trusted historical writing and the volunteers that sustain it with a donation to American Heritage.


Stay informed - subscribe to our newsletter.
The subscriber's email address.