Reginald A. Fessenden was a seminal figure in the development of wireless telephony technology, the first important North American inventor to experiment with the wireless telephone, the immediate precursor of radio. Born in Canada, he undertook his most important inventive work in the United States, including the development of continuous-wave transmission and heterodyne principals (he was granted 229 U.S. patents from 1891 to 1936), and he conducted what was probably the world's first broadcast.

Origins

Fessenden was born in 1866 in what would become the Canadian province of Quebec, the first of four sons of an Episcopalian minister. His parents supported his drive for education, and he excelled in school, especially in math and science, earning an invitation from his father's alma mater, Bishop's College, to teach math and languages, although because of doing so, he never completed his own degree work. While teaching, he became increasingly fascinated with the scientific journals of the time, focusing especially on developments in electricity. He continued his teaching at a secondary education institute in Bermuda for another year or so.

There were several key turning points in Fessenden's professional life, and the first came in 1886 when he decided to leave secondary (high school) teaching and become involved more directly in the field of electrical engineering, in which he was largely self-taught. He began as a field tester with Thomas Edison's company, which was then wiring the streets of New York City. Within a year he was working in power engineering at Edison's New Jersey laboratory, a post he held for nearly three years, from 1887 to 1890. Here he learned by observation the importance of patents and the scientific method-bur apparently not the importance of the market or of the process of successful innovation. He was increasingly attracted to the study of Hertzean waves, reading laboratory journals in his spare time. Financial problems at the labs led to layoffs of many workers, including Fessenden. His practical experience in electricity continued as an electrical assistant at the Westinghouse subsidiary, the United States Company, in 1890. Just a year later, in 1891, he joined the Stanley Company of Pittsfield, Massachusetts.

Though he lacked academic credentials, Fessenden had by the early 1890s already published in respected journals. Based on his record, he was named a professor of electrical engineering at Purdue University in 1892, an institution then striving to develop a reputation in this field. After a year at Purdue, in 1893 Fessenden accepted an offer to move to the University of Western Pennsylvania (later the University of Pittsburgh) to occupy a new chair of electrical engineering. Half of his salary was paid by Westinghouse, which hoped to make use of Fessenden for its own research needs (he undertook incandescent lamp work for the company). While at Western, Fessenden increasingly concentrated his own research on wireless telegraphy technology and applications.

Building on the work of Nikola Tesla and other pioneers, Fessenden by about 1900 had come to a fundamental conclusion: for effective wireless voice transmission, the generation of continuous wireless waves was required, not the dot-dash-friendly spark-gap technology that was then the Marconi standard for wireless telegraphy. This new system would require development of new means of both signal transmission and detection. Fessenden focused first on developing a new type of wireless detector to replace the crude Branly coherer (then widely used in wireless telegraphy, as it was a central piece of the Marconi system which dominated wireless).

Fessenden's second important career change came in 1900 when he was persuaded to resign the relative comfort and security of his tenured university post with the enticement of government support for his expanding research. He began what would be his most historically important decade (1900-11) with a contract to work as a special agent with the U.S. Weather Bureau, then a part of the U.S. Department of Agriculture. The bureau wanted him to develop wireless applications for weather forecasting and dissemination of weather reports. In April 1900 he began work for the Bureau at Cobb Island, Maryland, south of Washington, D.C.

By the middle of the year, he had sent the human voice by wireless to stations a mile apart at Rock Point, Maryland, though the signals were very noisy using the spark-gap technology of the period. For his bureau-supported work, he developed further transmitters at Cape Hatteras, Roanoke Island, and Cape Henry-and sent voice and music signals among those stations as well. For a while, the bureau's support of his experimentation seemed to be all that he expected, but arguments soon arose over rights to his prior patents as well as rights to patents growing out of bureau-supported work. Finally, in August 1902, Fessenden left the bureau's employ (before they let him go), the immediate cause being his interest in forming a commercial company to exploit his work in wireless telephony. The negative feelings that ended the Weather Bureau contract offered a bad omen for his future working relationships.

Improving Wireless Systems

An invaluable Fessenden contribution in this period was his discovery of the heterodyne principal (drawing on his Greek language training, the word refers to the mixing of two forces) in 1901. Not well understood at first-because the inventor was probably a decade ahead of what technology could accomplish-the principal would eventually lead to great improvements in both receiver sensitivity and static control. This allowed sending and receiving of signals from the same antenna with no interference between the two functions. The heterodyne principal has remained central to radio ever since.

Seeking to develop the transmission side of his new wireless system, Fessenden conceived of the "alternator" to obtain the continuous alternating current waves he needed. After Westinghouse turned down his order, in 1901 Fessenden turned to General Electric (GE) to build his first alternator, a contract supervised by GE's brilliant Charles Steinmetz. Less capable than he sought, the resulting device was tested later that year, though it was only delivered to him in March 1903. The 10,000-cycle machine owed some of its design and principals to power-generation equipment. But Fessenden sought a 100,000-cycle machine that could serve as a transmitter on its own. With Fessenden's order to GE for such a device-which really pushed the margins of what could then be accomplished-a new player entered the story when engineer Ernst Alexanderson became involved. Though he and Fessenden would argue over a key aspect of the device (whether its core should be wooden or iron-Fessenden holding out for the former), they otherwise worked well together, and by 1906 an 80,000-cycle machine was delivered to the inventor's Brant Rock, Massachusetts, transmitter site, just south of Boston. It would soon be utilized in a pioneering experiment.

In an era when there was no means of amplifying often weak wireless signals, Fessenden by 1903 had developed a new means of detecting signals for his new system, a replacement for the then-standard iron-filing-based coherer. Fessenden's "liquid barretter" or electrolytic detector was a complex device that nonetheless proved faster and more reliable and used less energy than the coherer; in addition, operators could use headphones to hear the signals rather than reading them on a tape. The barretter was soon widely adopted (Fessenden's patent rights being largely ignored by both other inventors-chiefly Lee de Forest-and users, especially the U.S. Navy). It became something of a standard in the field until it was replaced by vacuum tubes just before World War I.

The National Electric Signaling Company

At this creative peak of his radio work, Fessenden entered into agreement with two Pittsburgh investors to financially support his continuing invention and testing process. In November 1902 bankers Hay Walker Jr. and Thomas H. Given agreed to fund the formation and operation of the National Electric Signaling Company (NESCO) in order to develop and market a wireless system that they then hoped to sell. Unlike many other fledgling wireless companies of the time, there was no public stock sale-Given and Walker provided all the financial support, a few thousand dollars at first and eventually a total of nearly $2.5 million (nearly $50 million at 2003 values). Fessenden was general manager. Almost immediately, tensions, and eventually out-and-out battles, developed between the investors (who knew virtually nothing about radio technology but understandably sought a return on their funds) and the inventor (who held the technical knowledge but was becoming increasingly difficult to work with). From the beginning, NESCO focused on constantly changing short-term strategies as the partners bickered and fought. There was never an agreed-upon strategic plan for the firm, along the lines that Marconi was so successfully demonstrating at the same period. For example, Fessenden wanted to sell equipment, whereas his backers wanted to develop and market a complete wireless system.

NESCO transmitting stations were built in Washington, New York (Brooklyn), and Philadelphia (in nearby Collingswood, New Jersey) that were designed to provide overland communication in competition with existing wired systems. Despite considerable investment, wireless was still developing and often proved unworkable; not surprisingly, few customers materialized. Under a 1904 contract with GE, NESCO developed further transmitters in Lynn, Massachusetts, and Schenectady, New York, but the contract was cancelled in mid-1905 because of poor performance between transmitters. Faced with this failure, late in 1904, a shift in company priorities saw the partners seeking to develop a transatlantic radio service in competition with Marconi. At the same time, and against Fessenden's advice and pleas, NESCO stopped selling equipment. Transmitting stations were built in Scotland and in Brant Rock, Massachusetts, south of Boston. But they provided only sporadic communications-and almost none in the summer because of atmospheric static. By May 1906 Given and Walker had invested over half a million dollars, with precious little to show for it. That December a storm destroyed the Scottish station, effectively ending NESCO's transatlantic dreams. That, in turn, set the stage for one of Fessenden's most famous accomplishments.

With three days notice to shipowners using Fessenden equipment off the Massachusetts coast, on Christmas Eve of 1906 Fessenden offered what many consider the world's first true radio "broadcast"-a clear (not coded) voice and music transmission intended for widespread reception. Using his new and larger alternator, and with the inventor acting as "announcer," records were played, and Fessenden sang, played the violin, and even made a speech. All of this was repeated a week later on New Year's Eve. Fessenden was not seeking to broadcast as we think of the term today, but rather sought an improved means of competition with point-to-point wireline carriers. Unfortunately, his continuing pattern of claiming more for his system than it could in fact accomplish began to hurt. Further, because there was no press coverage (newspapers had not been invited to the experimental transmissions), few knew of the seeming wireless telephony breakthrough.

Radio telephony transmissions required more power and usually did not reach as far as radio telegraphy. But despite this limitation, there seemed to be obvious applications of wireless to the telephone business. In early 1907 Given and Walker made a concerted effort to interest American Telephone and Telegraph (AT&T) in taking over NESCO, and initially their efforts seemed to bear fruit. Both AT&T engineers and top company management were favorably disposed to the idea. However, a financial panic in 1907 led to a change in AT&T management as the J.P. Morgan interests took over, and a major retrenchment took place. Any thoughts of investing in a still largely unproven technology were forgotten. This failure to sell NESCO to AT&T was a major setback and led to increased disagreements between Fessenden and his backers-he again wanted to sell equipment, and they wanted to sell the whole firm.

For a time things looked up with the hiring of Colonel John Firth as the accomplished sales manager NESCO desperately needed. The company resumed equipment sales, thanks to his good relations with navy officials and others, and he also helped to mediate between Fessenden and his two Pittsburgh financial backers. At the same time, however, NESCO scaled down with layoffs and reduced experimental work.

Frustrated with the constant bickering and continued lack of a focused company strategy, Fessenden revived plans to compete with Marconi in transatlantic communication, and toward that end he set up an independent company, Fessenden Wireless Company of Canada, without informing or involving Given and Walker. This led to a break between Fessenden and the backers late in 1910 (because Given and Walker felt they owned the patents in question). In January 1911 Given and Walker dismissed Fessenden from the company. He, in turn, sued for wrongful termination. Some 15 years of litigation followed.

Later Life

NESCO was placed in voluntary receivership to protect it from the inventors' suit (Fessenden eventually won a judgment of several thousand dollars), though research continued with about a dozen engineers. What was most ironic about Fessenden 's departure and the demise of NESCO is that by 1911-12, the radio industry had begun to agree on the need to develop continuous-wave equipment for voice transmissions. No longer a part of NESCO, Fessenden was increasingly bypassed, due in no small part to his choleric temper, impatience, and maddening ego.

NESCO entered into a cross-licensing agreement with the Marconi interests after suits for patent violation had been brought. Finally, in 1917, after both Given and Walker had died, NESCO was sold to the International Radio Telegraph Company, which in 1920 fell to Westinghouse (and the patent rights of which were, in turn, transferred to the Radio Corporation of America [RCA] in 1921).

Fessenden's post-NESCO radio work was with the Submarine Signal Company of Boston. Beginning in 1912, he sought to improve underwater signal transmission and reception with what became known as the Fessenden Oscillator, a device also useful in detection of icebergs at sea; a wireless compass; and a fathometer or sonic depth finder. By the time he left Submarine Signal in 1921, however, Fessenden was largely through with radio innovation. In 1921 he received the gold medal of the Institute of Radio Engineers, and later he received the John Scott medal from Philadelphia for his development of continuous-wave wireless. In 1929 he received the Scientific American medal for his inventions concerning safety at sea. Fessenden was clearly a visionary ("technically farsighted" says historian Susan Douglas, 1987) who sought to pioneer new approaches and systems, often at the margins of practicality of existing electrical capability. But unlike Marconi, Fessenden was totally unfit for marketing and presenting a public persona that would encourage investment. Frustration had helped to make him a very difficult man with whom to deal. He was variously reported as being short-tempered, impatient, intolerant, vain, and a man who rarely listened to others (for example, when he wrongly insisted on a wooden core for the alternator, relenting only when Alexanderson demonstrated the benefit of iron). He drove away many subordinates with his harsh treatment.

As with many radio pioneers, Fessenden proved to be important at pioneering (invention)-in particular at understanding the need for continuous-wave transmission, conceiving the alternator, and defining the heterodyne principal-but a relative failure at developing commercially successful innovations from any of these.