|Hawes Mechanical Television Archive||
|John Baird & Secret Signaling|
Where's the secret? Baird's wife says that he didn't work for the British government during World War II. Then to believe in secret signaling, you must accept certain Baird rumors. Among these rumors is a description of an actual device. Baird called his prototype by various names: "Television facsimile," "facsimile television" or just "fast facsimile." This device promised to send pages of text at the speed of TV frames. Such extremely high-speed facsimile messages might even defy detection by enemy listeners. Anyway, that was the somewhat exaggerated claim. In fact, the device is probably real. The rumor involves what it actually achieved for the Allies. Yet we're not exactly sure where the secret comes in. With his usual fanfare, Baird publicly announced this device to the press. He even published a photo of himself and the "television facsimile" scanner. The photo tells a story: Like many Baird inventions, "television facsimile" was an ungainly device. It held down a reinforced floor rather well.
Cable and Wireless. During the war, Edward Wilshaw hired Baird as a consulting technical advisor at Cable and Wireless. Wilshaw was the company chairman. The pay, £1,000 per year, was very generous. In Wilshaw's own words, this appointment was really charity. At most, the alleged "work" amounted to an option to use Baird's ideas. [Moseley, pp. 245-246; Margaret Baird, p. 149] Yet Baird offered Wilshaw the unwieldy "television facsimile" prototype. [Kamm and Baird, p. 354; McArthur and Waddell, pp. 246-247]
Similar devices. As usual, Baird's brainstorm started with established inventions by others. In 1931, German inventor Manfred von Ardenne created an electronic flying spot scanner [Abramson, pp. 163-164]. Previous flying spot scanners all the way back to Bain's 1842 image scanner had been mechanical. In the block diagram below, the left half of Box "B" is von Ardenne's contribution. Using von Ardenne's device, the Axis Powers had all they needed for sending high-speed, wideband faxes. According to McArthur and Waddell [pp. 246-249], the German armed forces deployed a shoebox-size device that could rapidly scan and send documents. The alleged device could operate at the speed of television. If one page occupied one frame, then a fax would transmit in four-tenths of a second! Whether MacArthur and Waddell are right about this audacious German transmitter is hard to say. The two authors spin a rich and compelling yarn. In contrast, the evidence and details seem rather threadbare.
By 1944, Baird announced that he had built a British version of the device. The Baird reference appears in Costigan, a reputable facsimile history. [Costigan, p. 14] Although Baird promoter Waddell is a Costigan source [Costigan, p. xi], the claim seems plausible. Anyway, von Ardenne's flying-spot scanner must have made Baird's work fairly easy. Baird also claimed a transmission speed of a page every 1/25 second. [Costigan, p. 14] We can assume that this claim is accurate. After all, since 1937, the British television system had transmitted frames at this speed.
Plowing the same field. "Television facsimile" is another so-called "invention" that is really just a second plowing of the field. For example, the American firm AT&T had invented "intermediate film" video recorders in 1925. These gadgets were the work of Frank Gray and Herbert Ives. [Abramson, pp. 78, 105-106] Both the Baird company and the German Fernseh had extensively employed intermediate film devices. [Abramson, pp. 177, 213, 214] Following form, Baird's "television facsimile" scanned film images with a CRT flying spot scanner. The scanner duplicated both intermediate film and von Ardenne's 1931 German flying spot scanner patent. By World War II, the film scanner was a proven technology. One of the earliest film scanners came from Charles Francis Jenkins. As early as 1923, Jenkins had transmitted film shadowgraphs by television. [Abramson, pp. 59-60] In the realm of electronic TV, some of the best film scanners included Philo Farnsworth's image dissector tube. This tube had a linear response that was perfect for film reproduction.
For some years, the British television system had used film scanners in telecine equipment. In fact, a BBC telecine transmission of a Mickey Mouse film was the last TV show before the transmitter shut down during World War II. Baird's announcement of "television facsimile" (Scanning still pictures with a telecine unit) came some years afterward.
Let's take a closer look at "television facsimile." What parts are original? None of them. See the block diagram above...
TV is fax. Television is, more or less, high-speed, low-resolution facsimile. Both technologies have the same roots. [Costigan, p. 1] For this reason alone, "television facsimile" isn't an original idea. Undoubtedly "television facsimile" independently occurred to many people who were familiar with communications technology. In proof of that point, here is a visionary passage from Alfred Dinsdale. Dinsdale was the former editor of Television magazine. He published this passage in 1932, some 12 years before Baird demonstrated "television facsimile." [Dinsdale, pp. 236-237]
There is another potential field for television in connection with the world-wide dissemination of visual news, or the simultaneous publication at very remote points of a newspaper, each remote edition being an exact duplicate in every particular of the original or central edition. To achieve this it would be necessary to send facsimile copies of each page, en bloc, to the remote points. Phototelegraphy can go part way towards the accomplishment of this, but it cannot yet deal with so large a surface as the opened page of a newspaper, nor can it deal with it in sufficient detail, nor sufficiently rapidly. A combination of the techniques of phototelegraphy and television would seem to be required, a kind of speeded up phototelegraphy, or a slowed down form of television, giving the amount of detail which is at present possible with the best systems of phototelegraphy.
Differences between TV and fax. Despite their similarity, television and facsimile have some noteworthy differences. (Here, we refer to practical devices of the past several decades. We ignore potential and experimental devices.) The main differences are these: Television writes white to a black picture tube or reproducer. Typical facsimiles write black to a white screen or sheet of paper. (The two systems also use the opposite color primaries: RGB for TV vs. CMY for fax.) Television has a narrow contrast range. For example, monochrome TV pictures consist of white on gray or gray on black. Facsimile, with its broader contrast range, can reproduce black on white. Television, even broadband television, has very limited resolution. For example, a 25-inch NTSC screen in perfect adjustment can reproduce about 12 dots per inch. Typical facsimile equipment can reproduce hundreds of dots per inch. Many of these differences between television and facsimile challenge any combination of the two technologies.
Requires wide bandwidth. When you consider the speed of document transmission, Baird's "television facsimile" system seems very stealthy. Yet Baird's device is a sword that cuts both ways. First, the necessary signal bandwidth would be very broad, several megahertz wide. If this signal appeared on the shortwaves, enemy receivers on many frequencies could detect it. Bottom line: The signal probably isn't very stealthy. And the notion that Nazis wouldn't recognize video signals is absurd. After all the Germans televised the 1936 Olympics!
Noise & Fading. Noise, fading and ghosting on many shortwave frequencies is severe. If this signal appears on VHF frequencies (or above), then the signal can only travel for a short distance. As the carrier frequency rises, the signal becomes a line-of-sight transmission. Higher carriers tend to bounce off ever smaller bits of terrain, eventually including leaves. VHF signals are line-of-sight communications. Under normal conditions, the limit for VHF frequencies is at best 70 miles (113 kilometers). For good reception over this distance, a spy needs an excellent antenna in an unobstructed, high place. Each of these facts strips more of the alleged "stealth potential" from Baird's "television facsimile."
Reduce Transmission Speed. Baird could have worked around problems of wideband transmission. For example, he could have transmitted documents at a very slow speed. This is how slow-scan television works. Unfortunately, then the transmission time extends back to normal facsimile times. Long-duration signals return us to the problem of triangulation by the enemy. If the modulation mode is AM, typical of World War II transmissions, fading damages received pictures. Also, slow-scan reception requires some means of storage. For example, a P7 phosphor retains a useful image for six to eight seconds. In the photos that we've seen, Baird uses a normal-persistence phosphor and rephotographs the CRT at the receiver. The receiving station rephotography uses a large, unwieldy 16mm kinescope (movie) recorder. In a photo of the apparatus, the kinescope recorder is nearly as tall as Baird is. The machine looks like a theatrical projector. This isn't the sort of thing that a French spy would carry in his backpack. Apparently Baird's recorder doesn't use microfilm, but rather normal 16mm film. If the CRT only reproduces 405 lines, then normal film would serve well.
Reduce resolution. Baird could have speeded up the transmission by reducing the image resolution. For example, Copthorne Macdonald's post-war slow-scan system sent 120-line frames in eight seconds. The 120-line picture is adequate for sketches and snapshots, but not so good for document reproduction. Content compression is another possibility. Vestigial sideband (VSB) transmission could reduce the bandwidth by say, 35 percent. While impressive, this amount of compresssion still leaves a broadband signal. Digital compression is another matter. Yet during the war, such technology was in its infancy. Even if Baird had understood digital techniques, he'd have required some form of digital storage. For example, USB memory sticks, RAM chips, core memory, or punched tape. Except for the punched tape, these inventions were in the future. Baird also could have stored pictures in a bank of several thousand capacitors. We haven't seen any evidence that he considered the capacitors.
Postcard-size. Don Lancaster is an expert on text transmission by video. Among other accomplishments, he's designed several types of television terminals. (See Lancaster's classic The Cheap Video Cookbook.) According to Lancaster, an NTSC TV (525-line) frame can legibly reproduce up to 40 characters across by 24 lines. (In some cases, 32 lines of characters are possible. See pp. 26-27.) Consider that a typical typewritten page is 80 characters across by about 60 lines. Then an NTSC frame can reproduce about a quarter of a typewritten page, postcard size. Of course, during World War II, the British TV system didn't produce 525 lines. This system only had 405 lines. A 405-line frame with four-by-three proportions could store only some 32 by 18 lines. That's about a fifth of a page, smaller than a standard postcard. Add into the bargain that during transmision, analog TV pictures lose one-third of transmitted data. This loss is the significance of the term "essential area." A picture's essential area is its center two-thirds. [Zettl, pp. 223-225.] The sync signals remove about another 10 percent of the usable picture area. Here's what Zettl has to say about displaying type on a TV screen...
The amount of information that can be simultaneously projected on the television screen is limited. You should confine your copy to no more than ten words.
20-meter band. According to Baird's son Malcolm, Baird once thought about reducing the line count to 180 lines. [Kamm & Baird, p. 355; Malcolm Baird, email] Reducing the line count would narrow the necessary signal bandwidth. The reasoning was that a lower line count would allow "television fax" transmissions on the HF bands.[Malcolm Baird, email] Baird wanted to transmit "television facsimile" on the popular 20-meter band (14 MHz). Of course, a narrowband picture poses drawbacks, too. A 180-line picture can't legibly carry as many characters as can a 405-line picture. But would a 180-line signal fit on the 20-meter band? Consider today's 20-meter ham band: It's about 350 kHz across. Let's look at bandwidth vs. what some TV picture standards can store. The table below assumes CRT scanning and a doubly interlaced picture. Typical scanning generators can't interlace 180 or 120-line pictures. The nearest electronic standards are 175 and 125 lines. For this table, 180 and 120 are close enough...
Note: DSB = AM, double sideband modulation. VSB = AM, vestigial sideband modulation.
What about bandwidth compression? What if if Baird just didn't transmit during picture whites? We know this method as run-length encoding (RLC). [Costigan, pp. 144-146] With RLC, the picture would consist of pulses with dead air in between. Assuming that the receiver could maintain sync, the picture might still go through. Maintaining sync would be very difficult, though. Between pulses, noise would intrude on the image. There's no way to predict the gray level of this noise. The resulting picture would be a casualty. That is, allowing noise between pulses would severely reduce the average signal-to-noise ratio. I haven't heard any evidence that Baird applied compression to his "television facsimile" signals.
What about encryption? Baird would have known about several ways to encrypt a video tranmission. Such ideas occur to anyone familiar with television science. Encryption means include variable scanning rates, sync suppression, out-of-sequence scanning and signal inversion. In the 1980s, various pay television systems used some of these very techniques. (See the television decryption books of Graf and Sheets.) There are many more ways to encrypt video information. Unfortunately, the Nazis were quite advanced in television. The enemy could have figured out how to counteract most typical analog means of encryption.
No technical value. Apparently Baird never found a solution to his bandwidth problem. Cable and Wireless engineers attempted to explain this problem to Baird. Yet he insisted on testing his system on the shortwaves. During wartime, Cable and Wireless simply couldn't commit its transmitters to experimental use. The company refused Baird's request and in fact never found a use for Baird's unwieldy contraption. In September 1944, Sir Edward Wilshaw, Baird's manager, reported on Baird's service to Cable and Wireless. Wilshaw's report reads that Baird produced nothing of technical value for the company. Afterward, Cable and Wireless served Baird with his three-month notice of leave. [Kamm and Baird, pp. 354-357. Burns, pp. 383-384.]
No production. "Television facsimile" never went into production. The only evidence that the contraption even worked is one blurry, low-contrast photo. This photo is supposedly a reproduction off the "television facsimile" scanner. Whether the photo came from the transmitter or receiver is unclear. The photo itself is dubious, because it depicts Wilshaw and not a text page.[Kamm & Baird, plate 57] Further, Baird's press conference didn't exhibit a working model of this purported "invention."[Kamm & Baird, p. 356] Undoubtedly the device's many compromises and marginal performance were factors in Baird's avoidance of a working display. Instead, Baird posed before the presumed kinescope recorder. We can only wonder if "television facsimile" worked at all. If it did, then for reasons that we've explained, it couldn't have worked well. For these reasons, "television facsimile" was at best a partial prototype. It never progressed further, and never found a use.
Other Video Methods for "Secret Signaling"
The Germans had a high-speed, narrowband video signaling device. In 1929, Rudolf Hell invented his Hellschreiber (bright writer). This formidable electromechanical communications device sends messages one character at a time. The Hellscreiber transmits and reproduces fonts by a scanning process. The operator types messages onto a standard keyboard. With the original apparatus, letters feed in, or print out on a long paper tape. The Nazis could secure Hellscreiber communications by using the encryption means that we mentioned earlier. In fact, the Hellscreiber was most everything that makes "secret signaling" effective. German war communications produced much Hellschreiber traffic. In 1929 and for a few years thereafter, John Baird tested a Hellschreiber-like system that he called "telewriter." Baird's British patent 324,029 covers this device. [Kamm and Baird, pp. 133-135 and plate 33. Burns, pp. 190-192.]
Rtty. Apparently there's no record of the Baird device's use during World War II. Baird's system was both a copy of Hell's method, and a redundant machine. Radioteletype ("rtty"), a well-established and capable technology, filled the same role. In World War II, rtty used the Baudot code to send alphanumeric characters over the airwaves. A rtty message alternates between two audio notes, a modulation method that we call frequency-shift keying. Baird's device wouldn't have added any advantage in security. The Allies knew many ways of obscuring and encrypting rtty transmissions. For example, the use of non-Baudot character encoding, or non-standard frequency shifts. One-time pads and ciphers were also very effective means of securing wartime rtty communications.
Sync. RCA also came up with a Hellschreiber-like machine. This system took the name "tape facsimile." One version of the RCA machine produced synchronized lettering. Synchronized machines were a decided improvement over earlier tape facsimile or Hellschreiber machines. On earlier systems, letters tend to drift upward or downward. Hell and RCA worked around the drift by reproducing duplicate lines of type. Either the top line or the bottom line would always be readable. As one drifted out of view, the other drifted in. By adding sync, RCA at last introduced stable text. [Goldsmith, pp. 264-293]
RCA tape fax message. Left: Asynchronous type. Right: Synchronous type. [Shore, p. 283]
RCA's manufacturing leadership. For many years, RCA actually manufactured and sold radio facsimile equipment. In contrast, Baird's facsimile products failed to attract a market. No factory ever produced them. Baird's fax products never entered the production engineering stage. These products might not have been manufacturable at a reasonable price. As with phonovision, publicity seems to be the purpose of "television facsimile." Baird often seems to value recognition more than wealth. The "television facsimile" story is a case in point.
Fate of Ultrafax. Baird promoters can take solace in this fact: Ultrafax and "television facsimile" had one indisputable common feature. Like "television fax," Ultrafax bombed in the market. [Light, p. 367]
A perceived similarity between RCA's and Baird's work doesn't prove that RCA's work depended on Baird's. The alleged priority doesn't guarantee causality. About Ultrafax, very little technical data is available. [Jones, p. 190] Yet we can make certain reasonable judgments about alleged similarities between "television facsimile" and Ultrafax. First, both devices include kinescope recorders. The term "kinescope recorder" derives from the kinescope. RCA's Vladimir Zworykin and not Baird invented the kinescope. Zworykin's kinescope is an improvement on the Braun display tube. In the diagram above, the two CRTs are probably both kinescopes. The inclusion of a kinescope means this: That Baird's "television facsimile" depends on RCA's work, not the other way around.
Patents. Several other judgments about "television facsimile" vs. Ultrafax involve patents. The RCA and Kodak patents might not duplicate Baird's method, or might avoid it. The Ultrafax patents might only apply to a specific design. A design patent is likely, because other inventors had already patented most of the technology. Again, these devices already existed: Single-frame animation cameras, intermediate film, telecine and kinescope recordings. These earlier inventions cover the scope of "television facsimile." In fact, Baird's device and Ultrafax both incorporate these earlier inventions. You'd have trouble finding a new cog anywhere in the mechanism. Repatenting the earlier devices would be either impossible or fraudulent. Maybe that's why Cable & Wireless never took out a patent on Baird's "television facsimile."
Recordak and PPPI. Electronics magazine traces the idea of document storage to two previous technologies. Neither of these technologies came from John Baird. One of the earlier technologies was Recordak. During World War II, Recordak reduced the weight of V-mail by storing it on 16mm film. Near the delivery site, a processor drew a projection proof from the reduced Recordak film. The postal carrier delivered this normal-size proof. The second earlier technology was PPPI (photographic plan position indicator). PPPI was a radar-related film storage system. The rapid processing of Ultrafax films derives from PPPI. Ultrafax films move through hot processing baths at eight feet per minute. Drying requires only 25 seconds. [Ultrafax, p. 79; Evans, p. 249; Hills, p. 38]
Differences. There were some differences between Ultrafax and "television facsimile"...
Denial. Did John Logie Baird contribute technology to the Allies in World War II? As we've seen, Baird's wife denies that he did. Then believing that Baird provided some secret signaling scheme is a matter of either conjecture or mythology.
Purported. In 1944, Baird really announced his "television facsimile." Yet this purported "invention" actually recycled ideas such as intermediate film, kinescope recording, etc. Neither Baird nor his employer patented the unwieldy device. For this reason and the lack of originality, claims that RCA pilfered the invention are preposterous. Besides, the stealth potential of a publicly announced and well-known technology is dubious. Add to that doubt the notions that broadband signals could somehow fit onto the shortwaves. Or the claim that the enemy would miss and fail to spectrum-analyze the signals. Not surprisingly, there is no record of England's World War II use of "television facsimile." Whether such records might turn up is a matter of speculation, not fact.
Reason for skepticism. Such yarns as Baird's alleged "secret signaling" contain an extra measure of fleece. These tales are in fact a confidence game that depends on a naive, credulous, foolish public. When you check the source material, you find vaporware. With his 20 years of publicity stunts, Baird himself initiated this process of secular beatification. Now his disciples carry on. They offer a feast of conjecture. They second, third and fourth-guess history. And while denying the facts, they canonize the might-have-been. All the more reason to regard the latest "untold Baird story" with skepticism.
Abramson, Albert. The History of Television: 1880 to 1941. Jefferson, North Carolina: McFarland & Company, Inc., Publishers, 1987.
Baird, Malcolm. "Mechanical TV" Personal email (22 Sept. 2009).
Baird, Margaret, Television Baird: The Story of the Man Who Invented Television. Cape Town: Haum, 1973.
Burns, Russell. John Logie Baird. N.c.: The Institution of Engineering and Technology, 2001.
Costigan, Daniel M. Electronic Delivery of Documents. New York: Van Nostrand Reinhold, 1978.
Dinsdale, Alfred. First Principles of Television. Arno Press, 1975. (Reprint. Original publication: London: Chapman & Hall, Ltd., 1932.
Evans, Luther H. "Images from the Air: The Beginnings of Ultrafax." The Journal of Documentation, 4, no. 4, 1949, pp. 248-250.
Goldsmith, Alfred N., ed., Arthur F. Van Dyck, Charles W. Horn, Robert M. Morris, and Lee Galvin. Radio Facsimile. Volume I, October, 1938. New York: RCA Institutes Technical Press, 1938.
Graf, Rudolf F. and William Sheets. Video scrambling & descrambling for satellite & cable TV, 2nd edition. Woburn, MA: Newnes (Butterworth-Heinemann), 1998.
Hills, Lee and Timothy Sullivan. Facsimile. New York: McGraw Hill Book Company, 1949.
Jones, Charles Reed. Facsimile. New York: Rinehart, 1951.
Kamm, Antony and Malcolm Baird. John Logie Baird: A Life. Edinburgh: National Museums of Scotland, 2002.
Lancaster, Don. The Cheap Video Cookbook. Indianapolis, IN: Howard W. Sams, 1978.
Light, Jennifer S. "Facsimile: A Forgotten 'New Medium' from the 20th Century." New Media & Society, 8, no. 3, 2006, pp. 355-378.
McArthur, Tom and Peter Waddell. The Secret Life of John Logie Baird. London: Hutchinson, 1986.
Moseley, Sydney Alexander. John Baird, The Romance and Tragedy of the Pioneer of Television. London: Odhams Press, 1952.
Shore, Henry and James N. Whitaker. "Tape Facsimile Synchronizing Systems" in Radio Facsimile. Volume I, edited by Alfred N. Goldsmith, Arthur F. Van Dyck, Charles W. Horn, Robert M. Morris, and Lee Galvin, 270-283. New York: RCA Institutes Technical Press, 1938.
"Ultrafax." Electronics 22, no.1, 1949, pp.76-77.
Zettl, Herbert. Television Production Handbook, 2nd ed. Belmont, California: Wiley Publishing Company, Inc., 1968.
Go to Page: 1 2 3 4 5 Next
Copyright © 2009 by James T. Hawes. All rights reserved.
•Webmaster: James T. Hawes