Creating the Twentieth Century

by Vaclav Smil

  MacKeand and Cross tried to settle this uncertainty by using the best possible information in order to reconstruct Marconi’s complete system and then to model its most likely performance. They concluded that the transmission centered on 475 and 540 kHz, that its total power at Poldhu reached about 1 kW and at St. John’s was about 50 pW, and that it is likely that Marconi received high-frequency wide band signals, “spurious components of the spark transmitter output, propagated across the Atlantic by sky waves near the maximum usable frequency” (MacKeand and Cross 1995:29). In contrast, Belrose (1995, 2001) found it difficult to believe that Poldhu signals could have been heard on Signal Hill because the broadcast took place during the day (hence their heavy attenuation) and during a sunspot minimum period, and as the untuned receiver used by Marconi had no means of amplification.

  Definitive demonstration of long-range wireless transmission took place in February 1902, when Marconi and a group of engineers sailed from Southampton to New York on the Philadelphia fitted with a four-part aerial: coherer-tape reception was up to 2,500 km from Poldhu, and audio signal reached as far as 3,360 km (Marconi 1909; McClure 1902). Although Marconi was completely surprised by the difference between the maximum daylight reception (1,125 km) and the nighttime maxima (he was not aware of the ionospheric reflection of radio waves during the night), he was now confident that as soon as he set up an American station similar to that at Poldhu, he would be able to transmit and receive easily across the Atlantic. The Canadian government provided financial assistance to set up a high-power station at Glace Bay, Nova Scotia, and the exchange of messages with Poldhu began in December 1902. In 1907 an enlarged Glace Bay station and a new European station at Clifden in Ireland were used for the first commercial signaling across the Atlantic (Marconi 1909; Marconi Corporation 2003).

  FIGURE 5.20. The basic circuit of Marconi’s Poldhu transmitter contained an alternator (A), chokes (C1 and C2), low-frequency (T1) and high-frequency (T2 and T3) transformers, spark gaps (G1 and G2), capacitors (P1 and P2), and telegraph key (K). Transmitted power at low frequencies had the highest density at about 0.6 MHz (bottom left); at high frequencies, at about 3 MHz (bottom right). Based on Fleming’s original drawing in Bondyopadhyay (1993) and on Aitken (1976) and Belrose (1995).

  Marconi’s first major customer was the British Royal Navy, which made a substantial purchase of radio sets in 1900, and wireless communication was used (unsuccessfully) for the first time during the Boer War (1899–1902) and shortly after that routinely during the Russo-Japanese War of 1904–1905. Commercially much more important was Marconi’s contract with the Lloyds of London, the world’s premier shipping insurer, which committed itself in 1901 to 14 years of exclusive use of Marconi’s wireless telegraph system. But the diffusion of wireless telegraphy in merchant shipping was not a rapid affair. Only in 1914—two years after the signals sent on the night on April 14, 1912, by the senior radio operator Jack Phillips from the Titanic (he drowned, and several nearby ships did not have their receivers on) called for help for the sinking passenger liner—did an international conference mandate the presence of wireless on all ships carrying more than 50 people (Pocock 1995).

  Marconi’s aggressive advances had other, and much more experienced, engineers playing catch-up. Lodge filed his syntony (two circuits tuned to the same frequency) patent in 1897 (Lodge 1908). Adolf Slaby (1849–1913), the first Professor of Electro-technology at Berlin’s Technische Hochschule, was present at Marconi’s May 1897 experiments, and after his return to Germany he and Georg Wilhelm Alexander von Arco (1869–1940), working for Allgemeine Elektrizitätsgesselschaft (AEG), began to patent various improvements to receivers and antennas. So did, independently, Karl Ferdinand Braun (1850–1918), working for Siemens & Halske in Strassburg. After the two rival companies joined to set up Gesselschaft für drahtlose Telegraphie (much better known as Telefunken), those inventions provided the foundation for Germany’s leading role in the early development of wireless broadcasting (Telefunken 2003).

  In the United States, Tesla filed his two key radio patent applications (U.S. Patents 645,576 and 649,621) in 1897. He specified the apparatus that will produce “a current of excessively-high potential” to be transmitted through the air, and he also recognized that the receiving coils may be moveable, “as, for instance, when they are carried by a vessel floating in the air or by a ship at sea” (Tesla 1900:2; figure 5.21). As these filings preceded by more three years Marconi’s U.S. patent application for an apparatus for wireless telegraphy (U.S. Patent 773,772 filed on November 10, 1900), Tesla’s priority seemed secure: his reaction to the news of Marconi’s trans-Atlantic broadcast was to note that that achievement was based on using 17 of his patents. But in 1904 the U.S. Patent Office reversed itself and recognized Marconi as the inventor. This only aggravated Tesla’s already precarious financial position, but worse was to come in 1909 when Marconi but not Tesla got the Nobel Prize in physics.

  This bitter story had its unexpected ending in 1943 when, a few months after Tesla’s death, the U.S. Supreme Court finally ruled in his favor (Cheney 1981). But this was no righting of intellectual wrongs, merely a way for the court to avoid a decision regarding Marconi Co. suit against the U.S. government for using its patents. Marconi’s success is a perfect illustration of the fact that no fundamental innovations are needed to become a much respected innovator. Being first to package, and slightly improve, what is readily available, being aggressive in subsequent dealings, and making alliances with powerful users can take an entrepreneur and his company a lot further than coming up with a brilliant new idea. The preceding sentence, describing Marconi’s moves, also sums up what William Gates and his Microsoft Corporation did with Windows for the IBM personal computer. The fact that Marconi was not a great technical innovator is best illustrated by the fact that he considered Morse signals quite adequate for the shipping business and that he, unlike Tesla and Fessenden, did not foresee the multifaceted development of radio and broadcasting industry (Belrose 1995; Cheney 1981).

  Spark generators, favored by Marconi, could only transmit the Morse code. The first continuous wave signals that could be modulated by audio frequencies for the transmission of voice or music were produced by arc transmitters and high-frequency (HF) alternators. Valdemar Poulsen designed the first effective arc transmitter in 1902, and the largest devices were patterned on Kristian Birkeland and Samuel Eyde’s process, which was commercialized in the same year in order to produce nitrogen oxides as feedstocks for the synthesis of inorganic fertilizers. These transmitters became eventually truly giant: Telefunken’s Malabar in the Dutch East Indies, commissioned in 1923, had input of 3.6 MW and a 20-t electromagnet to quench the arc. Europe’s largest arc transmitter, completed in 1918 in Bordeaux, had eight 250-m masts and covered an area of nearly 50 ha.

  FIGURE 5.21. Tesla filed his application for an apparatus for transmission of electrical energy on September 2, 1897, and received U.S. Patent 649,621 in May 1900. The transmitter consists of a suitable source of electric current (G), transformer (A and C being, respectively, high-tension secondary and lower voltage primary coils), conductor (B), and terminal (D), “preferably of large surface, formed or maintained by such means as balloon at an elevation suitable for the purposes of transmission.” In the receiving station, the signal is led from the elevated terminal (D’) via a conductor (B’) to the transformer, whose coils are reversed, with A being the primary, and the secondary circuit contains “lamps (L), motors (M), or other devices for utilizing the current.”

  Tesla, already famous because of his electric motor and AC inventions, built the first low-radio-frequency alternator (operating at 30 kHz) in 1899, but two men whose work made the first radio broadcasts possible—a Canadian, Reginald Aubrey Fessenden (1866–1932), and a Swede, Ernst Frederik Werner Alexanderson (1878–1975)—are well known only to students of radio history. Fessenden (figure 5.22) was first to conceive and Alexanderson first to build, and then to perfect, HF alternators that coul
d produce continuous wave signals. Fessenden’s early accomplishments included the work in Thomas Edison laboratory and professorships of electrical engineering at Purdue and Pittsburgh University. In 1900 he began to work for the U.S. Weather Bureau and pursued his idea that HF well above the voice band should make wireless telephony possible.

  FIGURE 5.22. In contrast to Tesla and Marconi, only radio experts and historians of invention now know the name of Reginald Fessenden, a Canadian working in the United States, who made the world’s first radio broadcast in December 1906. Photograph courtesy of the North Carolina State Archives.

  His first success came on December 23, 1900, when he transmitted a couple of sentences— “One-two-three-four, is it snowing where you are Mr. Thiessen? If it is, would you telegraph back to me?”—over a distance of 1.6 km between two 15-m masts built on Cobb Island in the Potomac River in Maryland (Raby 1970). After he left the bureau, he set up his broadcasting hut and antenna at Black-mans Point on Brant Rock in Massachusetts and its trans-Atlantic counterpart at Machrihanish in Scotland, so his challenge was to span a distance longer than did Marconi’s wireless telegraphy experiments. His eventual success was made possible by a new 50 kHz alternator that was built by General Electric according to the designs of Ernst Alexanderson, a young Swedish engineer who was hired by GE in 1902 (Nilsson 2000). Alexander-son’s machine had the periphery of its tapered disk rotating at 1,100 km/h (i.e., the speed of sound) and yet wobbling less than 0.75 mm.

  This was an inherently wasteful, and hence costly, way to broadcast: due to high heat losses in windings and armature, and to high frictional losses arising from up to 20,000 rpm, efficiency of smaller units was less than 30%. Moreover, the technique was limited to relatively low frequencies of no more than 100 kHz, compared to 1.5 MHz for spark generators. But it was the first means of generating true continuous sine waves, and it could compete well with spark generators at longer wavelengths and lower frequencies that were favored at that time. After WWI, vacuum-tube transmitters began replacing all of the pioneering transmitting devices, but some spark generators remained in operation until after WWII, long after arc transmitters and most of the alternators were gone.

  The first long-distance broadcast using the Alexanderson’s alternator took place accidentally in November 1906, when an operator at Machrihanish clearly overheard the instructions relayed from Fessenden’s Brant Rock headquarters to a nearby test station at Plymouth. Encouraged, Fessenden prepared the world’s first programmed radio broadcast: on Christmas Eve of 1906 he gave a short speech and then played a phonographic recording of Handel’s Largo followed by his own short violin solo of Gounod’s O holy night, a reading from the Bible, and wishes of merry Christmas (Fessenden 1940). This broadcast was heard by radio operators on ships of the U.S. Navy along the Atlantic coast and on the vessels of the U.S. Fruit Co. as far away as the Caribbean, and its variant was repeated on New Year’s Eve. For Fessenden, this was a personal triumph because a widespread belief, attested by Fleming’s 1906 book, held that an abrupt impulse is a necessary for wireless transmissions and that HF generators could not to the job.

  Fessenden also developed heterodyning—a way to transfer a broadcast signal from its carrier to an intermediate frequency in the receiver in order to avoid retuning the receiver when changing channels—and amplitude modulation (AM). AM frequencies, now used in the range of 540 to 1,600 kHz, remained the best way to broadcast voice and music until the diffusion of frequency modulation (FM, now in the band between 88 and 108 MHz) that was invented by Edwin Armstrong (1890–1954). But these admirable achievements required further development and modification in order to make broadcasting a commercial reality (White 2003). Contributions by John Fleming, Edwin Armstrong, Lee De Forest, and others were essential in that respect. The year of the first broadcast brought another milestone in the history of radio because of Lee De Forest’s (1873–1961) invention of triode or, as he called it, the Audion tube.

  In 1904 Fleming invented the diode (a device based on Edison’s unexploited effect), which was essentially an incandescent light bulb with an added electrode and could be used as a sensitive detector of Hertzian waves as well as a converter of AC to DC (figure 5.23). But, as he later recalled in his biography,

  Sad to say, it did not occur to me to place the metal plate and the zig-zag wire in the same bulb and use an electron charge, positive or negative, on the wire to control the electron current to the plate. Lee de Forest, who had been following my work very closely, appreciated the advantage to be so gained, and made a valve in which a metal plate was fixed near a carbon filament in an exhausted bulb and placed a zig-zag wire, called a grid, between the plate and the filament. (Fleming 1934:144)

  The added grid electrode, interposed between the incandescing filament and the cold plate, acted as a modulator of the flowing current and made the Audion the first highly sensitive practical amplifier (figure 5.23). In 1912, when Armstrong introduced the first acceptable AM receiver, de Forest began using a series of his Audions in order to amplify HF signals, and he also discovered that feeding part of the output from the triode back into its grid produces a self-regenerating oscillation in the circuit that can be used, when fed to an antenna, for broadcasting speech and music.

  FIGURE 5.23. Detail of John A. Fleming’s patent drawing (U.S. Patent 803,684 filed on April 10, 1905) for the diode (left), an “instrument for converting alternating electric currents into continuous currents.” A glass bulb (a) contains a carbon filament (b) operating at 6–8 V and 2–4 A and connected to leads (e and f) by platinum wires. An aluminum cylinder (c), suspended by platinum wires (d), is open at the top and bottom, and it surrounds the filament without touching it. Lee De Forest’s diagram (right) of “a wireless telegraph receiving system comprising an oscillation detector constructed and connected in accordance with the present invention,” namely, his triode (U.S. Patent 841,387), shows an evacuated glass vessel (D) containing three conducting members: a metallic electrode (F) connected in series to a source of electricity that can heat it to incandescence; a conductor (b) made of platinum and, interposed between these two members, a grid-shaped platinum wire (a). Both drawings are available at http://www.uspto.gov.

  Consequently, the triode could be adapted for reception, amplification, and transmission. For more than half a century, all electronics—radios, televisions, and the first computers—depended on an increasing variety of vacuum tubes: tetrodes and pentodes, containing four and five electrodes in a high vacuum, and cathode ray tubes (for TV screen and display monitors) were eventually added to diodes and triodes. Domination of vacuum tubes ended only when transistors, superior amplifiers that operate on an entirely different principle, began taking over during the 1950s.

  By 1913 the combination of better generators, antennas, amplifiers, and receivers provided a good foundation for commercial development of radio broadcasting, but large-scale diffusion of the new communication medium took place only after WWI (White 2003). Several reasons explain this delay. To begin with, only limited numbers of radio amateurs and military and ship operators had requisite transmitters and receivers. Public access to radio was delayed by restrictions imposed on the use of airwaves during WWI and by the necessity to wear headsets, a requirement that made the earliest reception a peculiar hobby rather than an easily accessible source of news and entertainment. That is why radio’s expansive years came between the two world wars as transmission distances increased, numbers of radio stations multiplied, and every aspect of reception was greatly improved thanks to new tuning circuits, capacitors, microphones, oscillators, and loudspeakers. Radio’s second great transformation came during the 1950s with the deployment of transistors that made it inexpensive and easily portable.

  6

  A New Civilization

  Every now and again something happens—no doubt it’s ultimately traceable to changes in industrial technique, though the connection isn’t always obvious—and the whole spirit and tempo of life changes, and pe
ople acquire a new outlook which reflects itself in their political behaviour, their manners, their architecture, their literature and everything else…And though, of course, those black lines across the page of history are an illusion, there are times when the transition is quite rapid, sometimes rapid enough for it to be possible to give it a fairly accurate date.

  George Orwell in a BBC broadcast on March 10, 1942

  Orwell’s observations make a perfect epigraph to fortify this book’s arguments about the unequaled pre-WWI technical saltation. Their discovery was serendipitous: I began reading Orwell’s wartime essays a few months before I was to start this chapter. As soon as I came across the quoted passage, I felt as if this book had been distilled, 60 years before it was written, into a paragraph. Besides capturing the very intent and essence of this book, by asserting matter-of-factly that great civilizational shifts are ultimately traceable to technical changes, Orwell also believed that some of these momentous transformations could be rather accurately dated. And although these might be seen as only inconsequential stylistic preferences, I was pleased that Orwell used three specific terms that are not commonly encountered in writings about epochal shifts. Rather than using the inappropriate, but now generally accepted, term technology, he wrote correctly about technique (a perceptive reader might have noticed that this is the only instance when technology appears in this book).