The First Transistor in Space: Little-Known Aspects of the Space Race
On November 3, 1957, the Soviet Union launched the Second Artificial Earth Sputnik with the first living "passenger" - the dog Laika. The sensational success and secrecy of projects still leave behind the scenes the outstanding achievements of specialists in the field of radio-electronic equipment, whose participation in the space race is practically not considered by domestic literature, which is completely unfair.
What are we talking about?
- Where is the log?
- The devil knows, they say, on the satellite, the macaque scratches.
Translation:
- Where is Captain Derevianko?
- I don’t know, but they say that it works through a closed communication channel and monitors American tests of the Mk-48 torpedo prototype.From the invention of the transistor by Shockley, Bardeen, and Brattain in the late 1940s to the start of the space age, transistors have changed quite dramatically. Point transistors were replaced by planar ones, planar transistors were replaced by alloy ones, and so on, until they were all replaced by planar ones. Silicon transistors replaced germanium transistors, although not immediately. The first silicon transistor was made in 1954 by Texas Instruments, and, looking ahead, the transistors of this particular company were used in the first American satellites.
Rice. 3 Bardeen, Shockley and Brattain at Bell Labs
Rice. 4 Alloy transistor. A square plate is a base, on one side an emitter bead is welded to it, on the other, a collector bead (from Wikipedia)Semiconductor production in the USSR began in 1947 with a line for the production of germanium detectors for radar, exported from Germany. The development was carried out by a group led by A. V. Krasilov at NII-160 (now - JSC NPP Istok named after Shokin). S. G. Madoyan - a graduate of the Moscow Institute of Chemical Technology - in 1948-1949. developed a layout of the first point germanium transistor in the USSR,. The first lab sample worked for no more than an hour, and then required a new setting.
Rice. 5 Alexander Viktorovich Krasilov
Rice. 6 Susanna Gukasovna Madoyan. 1950
Rice. 7 Vadim Evgenievich Lashkarev
Rice. 8 Academician Axel Ivanovich BergIn 1950, the subject of transistors appeared in the TsNII-108 MO (now JSC "Central Scientific Research Radio Engineering Institute named after academician A. I. Berg"), the Physical Institute of the Academy of Sciences, the Leningrad Institute of Physics and Technology and other organizations. The first point transistors were made by V.E. Lashkarev in the laboratory at the Institute of Physics of the Academy of Sciences of the Ukrainian SSR. Due to the secrecy of research, often in that period different scientific groups did almost the same thing, received similar results and made discoveries independently of each other. This situation continued until November 1952, when a special issue of the American journal Proceedings of IRE (now Proceedings of IEEE) was published, completely devoted to transistors. At the beginning of 1953, Deputy Minister of Defense Academician A. I. Berg prepared a letter to the Central Committee of the CPSU on the development of work on transistors, and in May, the Minister of Communications Industry, M. G. Pervukhin, held a meeting in the Kremlin devoted to semiconductors, at which a decision was made to organize Specialized Research Institute of Semiconductor Electronics (NII-35, now NPP Pulsar). The laboratory of A.V. was transferred to Pulsar. Krasilov, in which they created the first prototype of a germanium planar (“layered”) transistor in the USSR. This development formed the basis of serial devices P1-P3 (1955) and their modifications.
Rice. 9 The first germanium and silicon Soviet transistors
The first silicon alloy transistors appeared in the USSR in 1956 (P104-P106), then in 1956-1957. - germanium P401-P-403 (30-120 MHz), as well as P418 (500 MHz). As you can see, by the launch of the first artificial satellite of the Earth in the USSR, there was the production of both germanium and silicon transistors, although even in the 1960s, the percentage of yield of suitable silicon transistors was only 19.3%. According to data, in 1957, the Soviet industry produced 2.7 million transistors (for comparison, in the USA, the production of transistors this year amounted to 28 million pieces, and the number of different types reached 600). The first germanium transistors operated in the temperature range up to +85 o C and their characteristics were unstable, which turned away both the military and the political leadership of the USSR from transistors.
Transistors and military
A popular story among "transistor builders" is that transistors became widespread due to the resourcefulness of inventors who said that the transistor could not be used for "special applications" and the shortsightedness of the military. Apparently, this story has a real basis.
The creators of the first transistor could not know everything that it would be capable of, but the Bell Labs administration understood that the significance of this discovery was enormous, and did everything in its power to make the discovery known in scientific circles. A big press conference was scheduled for June 30, 1948, to announce the discovery. But before showing the transistor to the public, it had to be shown to the military. It was hoped that the military would not classify this development, but it was clear that they could do it. On June 23, Ralph Bown showed a transistor to a group of officers. He showed a crystal with wires and how it could amplify an electrical signal more efficiently than a massive vacuum tube. He also told them that exactly the same demonstration was going to be held in a week, without formally asking their permission to do so. The military discussed the issue among themselves after the demonstration, but in the end, none of them spoke in favor of classifying this topic. Either because of their own short-sightedness, or in the form of additional protection against encroachments by the military, it was stated that “the transistor is expected to be used primarily in hearing aids for the deaf” (“it is expected that the transistor will be used mainly in hearing aids for the deaf). As a result, the press conference went off without a hitch. The New York Times ran a note about the transistor on page 46 in the Radio News section, after "a lengthy note on the resumption of reporting by a certain incomparable Ms. Brooks."
In early September 1951, Bell Labs held a symposia in Murray Hill, New Jersey, during which engineers explained, in fairly general terms, how to make point transistors and talked about the current progress with alloy transistors. At the same time, nothing was said about the specific manufacturing process and about military applications. The first symposium was attended by over 300 people (mostly military), each of whom paid a fee of $25,000 (twenty-five thousand dollars in 1951). Many firms wanted to make their own transistors rather than buy them, and many succeeded. Philips even made the transistor without attending these seminars, using only information from American newspapers. It should be noted that AT&T neither encouraged nor prevented other firms from making transistors.
In 1951, there were only four American companies making transistors for commercial applications: Texas Instruments, International Business Machines (IBM), Hewlett-Packard, and Motorola. They got licenses for the same $25,000 with low royalties. They were invited to a second symposium in April 1952, where the secrets of making transistors were fully revealed. By 1952 there were eight manufacturers, by 1953 there were fifteen, and by 1956 there were at least twenty-six germanium transistor companies with revenues of over $14 million a year. At the same time, the US military was the main consumer of transistors. In 1952, semiconductor manufacturers from Bell Labs signed over $5 million in military contracts. The share of research funding (R & D, Research and Development) from the military from 1953 to 1955 grew to 50%.
With all this, the future of semiconductors for the military remained unclear, because. the transistor was "noisy" compared to tubes, it could withstand smaller loads, could be damaged by sudden power surges, its characteristics were unstable over the temperature range, and the frequency range was relatively narrow. The situation was aggravated by a large spread of parameters between the two transistors. The price of transistors was also high: the first samples cost $20, by 1953 they fell to $8, while the lamps cost about $1. Fairchild Semiconductor's silicon mesa transistors were sold by IBM in quantities of 100 for $150 each in 1958 - while germanium transistors cost less than $5 at the time. In the mid-1960s, these same transistors began to cost less than 10 cents apiece.
What about hearing aids? They actually appeared in the USA in 1952-1953, and this was the first non-military use of the transistor. AT&T issued free licenses for use in hearing aids in memory of Alexander Bell's work with the deaf.
Unfortunately, this story has a little-known sad continuation, which concerns the Soviet Union. Professor Ya.A. Fedotov (author of one of the first monographs on transistors in 1955) in 1994 in the article "Electronics sends SOS!" mentions the "murderous" verdict that was pronounced at one of the meetings of the USSR Council of Ministers in 1956: "The transistor will never enter into serious equipment. The only promising area of application for it is hearing aids ... ". Familiar expressions, right? Fedotov writes: "This distrust of the transistor and the craving for the old tube technology was explained by a lack of understanding of the new situation in electronics." And this is a year before the launch of the first satellite! Thus, everything that the American "transistor builders" avoided and successfully avoided fell on domestic ones: secrecy, lack of centralization, misunderstanding of the prospects by the top political leadership of the USSR. Obviously, under such conditions, the transistors had little chance of getting on board.
If not a transistor, then what?
Was there an alternative to transistors? After all, we repeat, not any device can be put on board, but only with the required reliability characteristics. An alternative appeared in the late 1940s, i.e. almost simultaneously with transistors, in the form of rod radio tubes. Due to the secrecy of the topic, it is quite difficult to trace the history of the invention and development of this type of radio tubes, and often you have to be content with information from Internet forums.
June 1946. The Council of Ministers of the USSR instructs Plant 617 (in the near future - the Union Scientific Research Institute No. 617 (NII-617) with an experimental plant of the State Committee of the Council of Ministers of the USSR on Radio Electronics) in Novosibirsk to develop subminiature and extra-strong lamps for on-board computer systems of aviation technology. V.N. was appointed to supervise the work. Avdeeva.
Rice. 10 Valentin Nikolaevich AvdeevValentin Nikolaevich Avdeev was born on May 16, 1915 in the town of Kotelnich, Vyatka province. After receiving his primary education, he worked at the Svetlana plant (now PJSC Svetlana) in Leningrad. He graduated from the factory technical school, then studied at the All-Union Correspondence Institute of Technical Education in 1934-1938. In 1941, he was sent for an internship in the USA for six months (at the Radio Corporation of America, RCA factories) to study the production of radio tubes. When the Great Patriotic War began, he was evacuated to Novosibirsk together with the plant staff. There he worked first as a site foreman, from 1942 - the chief engineer of the plant, from 1943 - deputy head of the laboratory. Subminiature radio tubes were developed by the Design Bureau of Plant 617 by 1947, and secret production began in 1948. Since 1949, the work "Molecule" has been opened to create subminiature lamps with increased vibration resistance. On the basis of laboratory No. 1, NII-617 is being created, with Avdeev appointed as its director.
Rod tubes were practically free from the shortcomings inherent in "ordinary" tubes and, unlike transistors of that time, could operate over a full temperature range. A series of radio tubes has been created: 1Zh17B, 1Zh18B, 1Zh24B, 1Zh29B and 1P24B. In 1960, an article was published in the Radio magazine devoted to the principles of operation of rod radio tubes, in which the advantages of this type were noted, and the cut-off frequency was declared - more than 200 MHz, which more than met the requirements for the frequency of receiving radio signals from the first artificial Earth satellite (see . ).
Rice. 11 Comparison of "conventional" and rod radio tubes from an article in the journal "Radio"For the creation of rod radio tubes V.N. Avdeev was elected a corresponding member of the USSR Academy of Sciences in 1958 (the same year that S.P. Korolev was elected a full member). This despite the fact that V.N. Avdeev never defended dissertations - neither a candidate's, nor even a doctoral one.
The authors of an article in the Radio magazine complain: “Several years ago, when semiconductor devices appeared, some radio specialists were inclined to immediately “bury” the vacuum tube. The lamp, which for decades brought radio electronics one triumph after another, suddenly revealed many shortcomings ... An electronic lamp, compared with a semiconductor triode, undoubtedly has a number of shortcomings, but the remarkable advantages of the lamp are well known ... ". And they add: “Unfortunately, it must be noted that the issue of the scale of application, and therefore the production of rod lamps, is not resolved quickly enough, despite the fact that these lamps have existed for many years and have been highly appreciated.” In these words - a clear distrust of the "newfangled" transistors.
Rod radio tubes were used not only in space and aviation - on their basis, radio stations were created for the special forces of the GRU and the KGB of the USSR (R-353 "Proton"), a portable VHF radio station R-126, a complex of radio stations "MARS" for the Ministry of Internal Affairs, etc. .
Transistors in the first satellites
Soviet Army's RED STAR:
Uncle Sam thought of launching a Sputnik into the sky.
He announced it to the whole world, not two days but two years in advance.
The boastful and rich uncle called his Sputnik Vanguard.
The name was beautiful and quite chic
But it turned out to be pshik.From a selection of Time magazine about the reaction of the world media to the unsuccessful launch of the American Vanguard satellite on December 16, 1957. VANGUARD'S AFTERMATH: JEERS AND TEARS Monday, Dec. 16, 1957
Our country launched not only the first artificial satellite of the Earth (and then launched the first man into space), but after the first satellite, 2 full-fledged space laboratories were launched within 7 months - Sputnik-2 with Laika and Sputnik-3, with the help of the equipment of which, in particular, the Zamli's natural radiation belts were discovered. The American first satellite Explorer 1 was 3 months ahead of Sputnik-3, but in terms of its “functional” characteristics it was closer to Sputnik-1, and in weight it was almost 4 times less than it. The launch of Sputnik-1 caused the well-deserved respect of scientists, bewilderment and even fear of the inhabitants in the West, general joy and triumph in the USSR, and a storm of emotions among politicians. I will cite only two characteristic statements of Soviet and American politicians (I quote from ). First Secretary of the Central Committee of the CPSU N.S. Khrushchev: “It seems that the name Avangard reflected the confidence of the Americans that it was their satellite that would be the first in the world. But ... our Soviet satellite became the first, it was he who was at the forefront ... ". US Senator and future President Lyndon Johnson: "I don't believe this generation of Americans is willing to put up with having to fall asleep every night by the light of a communist moon." No wonder the space race has become fierce.
For clarity, the table below shows the launch dates and the main weight and size characteristics of the first artificial Earth satellites.
Launch date Name Country Dimensions Weight, kg 04.10.1957 Sputnik-1 USSR ~58 cm (without antennas) 83,6 03.11.1957 Sputnik-2 USSR 2 m x 4 m 508 01.02.1958 Explorer 1 USA about 1 m long 21,5 17.03.1958 Vanguard-I USA 16.3 cm (without antennas) 1,474 26.03.1958 Explorer 3 USA about 2 m long 13,97 15.05.1958 Sputnik-3 USSR 1.73 m x 3.57 m 1327 Echoes of the fierceness of the race are heard now. So, in 2015 (No. 138), the National Geographic Russia magazine published a short, but very notable for its unprofessional engagement note, “Satellite Avangard-1: Still at the forefront.” I quote it in full: “Size of a melon and weighing about a kilogram, Vanguard-1 became the first solar-powered satellite and an important step for the United States in the space race. In an attempt to catch up with the Soviet Union, which launched Sputnik 1 and Sputnik 2 in 1957, the United States sent Avangard 1 into orbit on March 17, 1958. Khrushchev called it pejoratively "grapefruit". However, the larger Sputniks deorbited and burned up on re-entry in 1958, while Avangard 1 is still flying. It stopped transmitting data in 1964 when the last photocells failed. But the device holds the title of the oldest artificial satellite in orbit and is predicted to stay there for about 240 more years. With all due respect to National Geographic and the American developers of Avangard-I, I think that comments are unnecessary here.
Let's get back to transistors. As we have already noted, some authors argued that transistors had already appeared on Sputnik-1, and they even cited the type of transistor - P401,. The site also cites this statement, although it makes a reservation that the use of rod radio tubes is more likely. For a long time on various forums, various enthusiasts tried to understand what was happening, but it was almost impossible to figure it out until the publication of the report on Sputnik-1 by Russian Space Systems OJSC (formerly NII-885). I don’t have the text of this publication, but it is quoted in the Radio magazine (No. 4, 2013), they also give a diagram of the transmitter of the first artificial satellite of the Earth:
Rice. 12 Scheme of the main transmitter of Sputnik-1 at 20 MHzThere is not a single transistor on the circuit, but there are 2P19B rod radio tubes. It turns out that those who believe that the first transistors appeared only in the American Explorer 1 are right?
Rice. 13 William Pickering, James Van Allen and Wernher von Braun demonstrate a full-scale model of the Explorer 1 satellite at a press conference in Washington following confirmation of the satellite's launch into orbit.
Rice. 14 George Ludwig With Explorer 1 BackupThis question was asked directly to George Ludwig, the designer of the Explorer 1 systems. He replied that he had indeed thought so before, but then he investigated this issue in more detail and found out that although the Soviets did not use transistors in Sputnik 1, they did use them in one of the instruments of Sputnik 2, launched in November 1957. Ludwig laments: "Of course they (the Soviets) had much more capacity and their carriers could output vacuum tubes and the batteries they needed." At the same time, he emphasizes that Explorer 1 became the first satellite, the equipment of which was completely on transistors (recall that there were no devices like rod radio tubes in the United States at that time). The interview curator refers to a 2001 publication which states: “Sputnik 2 was a true scientific platform containing various electronic components. In addition to the radio transmitter and cabin for Laika, it had solar ultraviolet and X-ray detectors, and cosmic ray instruments were mounted on the rocket body. And further: “Two identical detectors in the cosmic ray experiment functioned as recorders of scintillations due to charged particles. The pulses were counted by a semiconductor (based on triodes) circuit...“. Unfortunately, the article does not contain a link to the source of this information. Alas, in foreign literature, it happens that Sputnik-2 and Sputnik-3 are confused (for example, this happened in, although there is no confusion in an earlier article by one of the co-authors).
So in what Soviet apparatus were transistors first used? Only Sputnik-3 is known for certain. Sputnik 2 launched just a month after Sputnik 1 - what is the probability of transistors getting on board, in any capacity? To be honest, it is small, taking into account not only the attitude towards transistors in the leadership of the USSR, but also other considerations. As noted earlier, germanium transistors (namely, they were mainly produced by the Soviet industry and enough was known about them to judge reliability) are unstable in the temperature range, and where temperatures above +85 o C are needed, they were not used. On the other hand, American germanium transistors suffered from the same diseases, but in Explorer 1, according to Ludwig, they were used along with silicon, since germanium had a lower base-emitter voltage (0.2 V versus 0.5 V for silicon), therefore, in a number of circuits with a supply voltage of 2.8 V, they were used.
The first transistor radios
Excuse me, but where does the mention of the transistor P401 next to the Sputnik come from? In fact, given the recommended Sputnik frequency of 40 MHz and the fact that the cutoff frequency of the P401 was 30 MHz, it is difficult to imagine this transistor as a candidate for installation on board. The reason why this transistor is mentioned in the context of Sputnik may be comical. Remember the remark that in everyday life they confuse a transistor and a transistor radio? So, in 1957, the Voronezh Radio Plant began to produce the Sputnik radio receiver, the diagram of which is presented below.
Rice. 15 Scheme of the Sputnik radio receiver (1957)In the circuit, you can easily find P401, and P402, and other transistors. The first samples were produced in April 1957, 5 months before the launch of Sputnik-1. The body was made of dried pine, impregnated with an alcohol solution of cellulose, and covered with decorative plastic.
Rice. 16 Transistor radio receiver "Sputnik"Dimensions - 185x125x49 mm, weight with batteries - 950 g. There was a solar battery on the upper edge of the case! The cost of the device was 514 rubles - it was approximately the average salary of a worker at that time.
So, due to the lack of data on the Sputniks, there was confusion with the Sputniks.
And what follows from this?
Next year, 2017, we (Russia and the whole world) will celebrate the 60th anniversary of the launch of the First and Second artificial Earth satellites. I would like to turn to the management of Russian Space Systems JSC with a proposal to publish a report on the Sputnik-2 and Sputnik-3 systems by this time, since it is obvious that this is of great historical importance not only for the space industry, but also for electronic industry in Russia, which is alive, no matter what.
The superiority of Soviet space technology over the American one unwittingly played against the development of domestic transistors, since there were suitable radio tubes that made it possible to solve existing problems without worrying about saving dimensions and weight in the way that the Americans had to do. As a result, looking back, we see how advanced NASA's robotic space systems are now actively engaged in the exploration of the solar system (Mars, Jupiter, Saturn, Pluto...). The European Space Agency (ESA), which is actively involved in small-sized satellites (micro- and nanosatellites), does not lag behind. It is unlikely that a person will master the solar system in the coming decades, but the human mind can do this with the “hands” of automatic devices that have the necessary “intelligence”. After the decline of 1990-2000, despite certain successes of domestic developers, Russia is in dire need of its own microcircuits capable of solving computational problems of the modern level or even the level of tomorrow (after all, space projects have been planned for several years) and at the same time possessing the necessary radiation resistance and fault tolerance. And the problem here is not so much the existing technological backwardness, but the lack of a common understanding of the appearance of such computing systems and, consequently, the lack of not only an electronic component base, but also reliable and efficient software. You can't repeat the mistakes of the past - you need to learn from them.