A new approach to impressing sound on radio waves created less static
Early in 1924, 34-year-old Edwin Armstrong returned to Columbia University, the scene 11 years earlier of his breakthrough invention of the regenerative circuit, while only a sophomore. His device had amplified radio waves a thousandfold and made radio practical. This time he had set his sights on eliminating static from radio, a problem most felt was insoluble. “Static, like the poor, will always be with us,” declared the chief engineer of AT&T.
Armstrong tackled the problem with extraordinary persistence, logging 15-hour days and seven-day weeks, broken only by lunches of a sandwich and a glass of milk. Sometimes it took several months to set up an experiment that involved as many as a hundred vacuum tubes. His hard work paid off handsomely with the invention of an entirely new radio system: frequency modulation, or FM.
Modulation refers to the way in which voice and music information is impressed on a radio wave. In amplitude modulation (AM), which was the industry standard, the information signal varies the amplitude of the wave. What Armstrong suggested was a method of modulation that would vary the wave’s frequency. Using an analogy of sea waves, AM imparts a signal by changing wave heights; FM does so by varying the spacing between wave crests. Noise affects wave height, or amplitude, much more than frequency, making FM less vulnerable to interference. But FM would require new transmitters, receivers, and a fairly wide channel spacing—up to 200 kilohertz—which was not readily available in the already crowded 500 to 1600 kilohertz AM band. The inventor proposed VHF (very high frequency) allocations, a spectrum where plenty of room was available and the promise of improved high fidelity would be fulfilled.
Challenging the multibillion-dollar radio industry, which had a considerable investment in medium-band AM, seemed nearly insurmountable. But with help from his old friend David Sarnoff, later president of RCA, Armstrong moved his equipment in March 1934 to the top of the Empire State Building for definitive broadcasting tests. Receiving sites were set up first at Westhampton Beach on Long Island, New York, and then at Haddonfield, New Jersey.
What the experimenters showed was a truly substantial improvement in the signal-to-noise ratio with the new technique. An FM signal twice as strong as a noise pulse would suppress the pulse; in AM a signal had to be 100 times as strong. Also, FM displayed a capture effect—that is, if two stations on the same frequency arrived at the receiving antenna with different signal strengths, the system would grab the stronger one rather than pick up both at once. The capture effect, together with the fact that VHF signals cannot be received farther than about 50 to 70 miles from a transmitter, suggested that FM stations in not-too-distant cities could operate on the same channel. The advantage of FM resided not simply in its high fidelity, with which AM could compete, but in a combination of effects, the most significant of which was its spectacular ability to suppress atmospheric and internal electronic noise.
On July 18, 1939, Armstrong began broadcasting from the first FM station, W2XMN, which he had built entirely with his own money in Alpine, New Jersey. That year General Electric began manufacturing—under Armstrong’s license—the first commercially available FM radios. By the end of the year the five-year-old Federal Communications Commission had received 150 applications for permits to establish FM stations.
Frequency modulation gradually became the standard of high-fidelity broadcasting all over the world, thanks in part to developments in the high-fidelity industry and innovative regulatory procedures. These developments led to a dramatic increase in the number of FM stations in the United States, from 990 in 1961 to nearly 12,000 in the present day. And with the proliferation of digital receivers, which produce sound quality far clearer than analog radios and equal to CDs, Armstrong’s dream of realism in the transmission of sound has been fulfilled.