Transistor History 101 by Bob McGarrah

[This article was originally published in two parts in Transistor Radio Collector magazine, July/August and September/October, 1997. I update it periodically as I research new material.]

The purpose of this article is to provide some background for those who are curious about how the transistor emerged from being a research laboratory experiment to a commercial marketplace reality.

Although the transistor officially celebrated its 50th anniversary in 1997 its origins go back farther than most of us generally realize. In September, 1939 two Bell Labs research scientists discovered that there were both positive (p-type) and negative (n-type) regions in a material called silicon. This research led to the recognition that semiconducting materials might have useful applications. In December, 1939 a young Bell Labs scientist named William Shockley first conceptualized a dramatic new application for semiconducting materials:

  • "It has occurred to me that an amplifier using semiconductors rather than a vacuum is in principle possible."
    •      Unfortunately, World War II interrupted further work in this area so it wasn't until 1945 that a semiconductor research group, headed by Dr. Shockley, was established at Bell Labs.

    The first experimental semiconductor device was actually made of another material called germanium. A crude looking collection of wires, insulators, and germanium was demonstrated on Dec. 23, 1947, at Bell Labs by William Shockley and his research team of John Bardeen and Walter Brittain.

    The name transistor (a combination of TRANSconductance  varISTOR) was suggested by another Bell Labs engineer. To read the rather interesting Bell Labs memo related to naming the transistor, click here.  Subsequently in 1956, Shockley, Bardeen and Brittain won a Nobel prize for their work.

    The point contact transistor patent (shown below) was applied for on June 17, 1948 by John Bardeen and Walter Brattain.

    William Shockley had wanted to be listed as the sole inventor for the point contact transistor. When this didn't happen he didn't waste any time filing a patent (shown below) for the junction transistor only nine days later on June 26, 1948.

    The first public demonstration of the transistor was at a press conference in New York City on that same day (June 26, 1948), however it was not until five days later that a rather obscure reference to the event appeared in print. It was buried in the "News Of Radio" column on page 46 of the New York Times:

  • "A device called a transistor, which has several applications in radio where a vacuum tube ordinarily is employed, was demonstrated for the first time yesterday at Bell Telephone Laboratories... In the shape of a small metal cylinder about a half-inch long, the transistor contains no vacuum, grid, plate or glass envelope to keep the air away. Its action is instantaneous, there being no warm-up delay since no heat is developed as in a vacuum tube. The working parts of the device consist solely of two fine wires that run down to a pinhead of solid semi-conductive material soldered to a metal base. The substance on the metal base amplified the current carried to it by one wire and the other wire carries away the amplified current."
    •      This first device was called a point-contact transistor since the wires were directly embedded into the semiconducting material. Point-contact transistors with consistent operating characteristics proved to be difficult to manufacture so research continued.

    My search into the history of transistors was aided greatly by the fortunate acquisition of several periodicals from the time. In particular I was able to review almost every issue of Radio-Electronics, Radio & Television News, and Electronics Engineering from 1948 through 1953 and beyond.

    Radio-Electronics first announced the transistor in September, 1948 and Hugo Gernsback followed up with one of his typical "crystal ball" type editorials in December, 1948. Interestingly, he got most of it right. Although he correctly believed the electron tube would remain dominant in high power applications he did predict such miniature devices as pen size radios, pocket recording instruments, personal military receivers, proximity fuses, and onboard airplane radar. He even wrote at great length about how performers would be able to wear transmitters which would eliminate unsightly microphone cords on stage. Remember, this was in 1948 and the transistor was less than a year old! Mr. Gernsbach seems to have been something of a visionary. His only failure was speculation about "atomic" transistors with point contacts made of radioactive material which would allow them to be self powered!

    Another of the first references to this exciting new discovery in the field of semiconductors was published in the October, 1948 issue of R&TN. It was titled "A Crystal That Amplifies - details of experiments with a double-contact germanium type crystal having transconductance". Essentially this article describes how to make a point contact transistor using a chip of germanium from a 1N34 diode and cats whiskers fashioned from tube heater filament wire. An interesting sidebar reads as follows:

  • "Bell Telephone Laboratories recent announcement of a new electronic component - the "Transistor" has started many "back of the shop" experiments. Like any other new design, complete technical details on the product are withheld for a time. In view of this we believe it to be appropriate to publish this report of an independent experiment that was, however, suggested by the announcement of the "Transistor". - Editor"
    •      In the December, 1948 R&TN further details were given on the new device.
  • "Additional technical details on the "Transistor" have been recently released by Bell Telephone Laboratories. The term transistor was derived from the fact that the unit is essentially a resistor which can amplify electrical signals as they are transferred through it from input to output terminals. The "Transistor" is, in reality, the electrical equivalent of a vacuum tube amplifier. However, there the similarity ceases. It has no vacuum, no filament, no glass tube. It is composed of cold solid substances."
    •      The first picture I have found of a transistor accompanies the article. It is a cut-away model of a Bell Labs point contact device in the early cartridge or cylinder style case with the collector and emitter leads (also the first time I've seen them identified as such) emerging from one end and the base being connected to the case.

    The first reference I have found for a commercially available transistor appeared in the January, 1950 issue of R-E. Amazingly it was a construction article for building a three transistor radio using CK703's as shown in figure 4 below. The article also described plans for constructing transistor sockets since they were not yet available commercially.

    Interestingly there was no reference as to the source or price of the CK703. As shown a little further below the CK703 was almost $20 each so the cost of transistors alone would have made this set an unobtainable dream for most hobbyists.

    In the May, 1950 issue of R-E was an article discussing an oscillator circuit which was designed using a Raytheon CK703 point contact transistor or "crystal triode". The June, 1950 issue of R-E carried another article with circuit applications for the CK703. The device measured about 5/8" long and 1/4" in diameter. A paper label bearing the Raytheon logo and type number was attached for identification.

    I am indebted to a friend, Allan Pellnat, who shared a considerable amount of information regarding the commercial availability and pricing of early transistors. Allan provided an image of the first transistor advertisment he has found. It was in the 1950-51 isssue of the Radio Master catalog. The solitary listing was for the Raytheon CK703 and it was priced at $18.00. For a reality check, an $18.00 transistor in 1951 would be around $110.00 today after taking inflation into account. Because of manufacturing difficulties at this early stage of development (and obviously its high cost), the CK703 was not produced or used in significant numbers. Most likely it remained a laboratory curiousity.

    The next transistor product announcement I could find appeared in the September, 1950 issue of R-E. It was for the GE type SX-4A and Z-2 point contact devices which were built to "joint Army-Navy specifications". Again this appears to be earlier than the general Bell Labs licensing announcement but since the application was apparently military in nature that may be the explanation.

    As transistor research progressed a new variation was developed. The August, 1951 issue of R-E reported that Bell Labs had announced on July 5th a junction type transistor. Two advantages of junction transistors were reported to be a significantly lower noise figure coupled with higher gain. The article concluded with the news that development of the point contact transistor had reached such a stable level that they would be put to trial use in Bell Telephone System equipment early the following year (1952).

    Hugo Gernsbach wrote another interesting editorial in R-E in December, 1951 in which he praised the advantages of the transistor. Although they had many advantages over vacuum tubes he noted that several problems would preclude transistors from replacing tubes for many years. For instance the transistor at that time was still limited in its high frequency response, and many "bugs" remained which prevented its mass production. Perhaps the most significant item was the transistor's economic disadvantage. He reported that the devices cost about $18.00 each at that time. Based on the information reported above it can be safely assumed he was referring to the Raytheon CK703. He ended the editorial with the following bizarre observation:

  • "What also in the writers opinion, will hold back the transistor is its present name. The public never has taken kindly to it and probably never will. The word "crystal", from the Greek "Krystallos" is the basis for a popular name. I would seriously suggest that, particularly for public consumption, a new name should be adopted. I advance the term "crystron" (crys=crystal; 'tron=electron). This is what the device is -- an electronically operating crystal. The adoption of this designating name now will do away with a great deal of confusion in the future."
    •      My personal observation is that we are fortunate that Mr. Gernsback's suggestion seems to have been ignored by those to whom it was directed. Just imagine a shirt pocket radio with shiny chrome grill proudly announcing "SIX CRYSTRONS" in bold letters!

    In 1952 Bell Labs began licensing its new invention to interested companies for a fee of $25,000. Among the first to take advantage of this were General Electric, IBM, Raytheon, a new company called Texas Instruments, and a small Japanese company that eventually became Sony Electronics.

    Although amateur radio operators used transistors to design experimental radio circuits as early as 1950, this was the exception and not the rule. During that early period some Bell Labs engineers made a wrist radio "in their spare time" as a gift for cartoonist Chester Gould, the creator of cartoon character Dick Tracy, who had used such a fictional device in the comics for many years. The radio case measured 1 1/2" x 2" x 3/4", however the external components included a 1" lapel speaker, the battery, and an antenna wire. Radio-Electronics (R-E) magazine reported in September, 1952:

  • "The little wrist radio is, of course, an experiment, and is not likely to be manufactured in the near future. It is important because it shows what transistors may be able to do at a not-too-distant day."
    •      Fantasy had become reality with the help of four point-contact germanium transistors.

    By mid 1952 research and development had come a long way. Dr. Lee De Forest, the inventor of the triode vacuum tube in 1906, wrote an article for R-E in July, 1952 about the future of transistors. The majority of his predictions, as viewed from our vantage point in time, have been met. Many of the advantages he listed are things that we take for granted today; that transistors would likely replace tubes in applications requiring long life, low power, and small size. Specifically mentioned were hearing ai ds, proximity fuses (for bombs), and self guiding missiles. It must be remembered that transistor development was seen as a priority by the military during the cold war build up in the early 1950's. This undoubtedly hastened their acceptance for use in all areas of electronics.

    In terms of consumer goods, perhaps the following observation by Dr. De Forest is the most pertinent for today, considering the vast numbers of personal computers.

  • "I have in mind especially that great marvel in electronic calculating machines, the IBM selective-sequence electronic calculator, which now occupies three sides of a gigantic hall on Madison Avenue, New York City, and comprises among its labyrinthian circuits and relays 6,000 electron tubes, occupying a tube rack some 150 square feet in size. Barring the space required in the rear for the essential conductors, this tube area could now be reduced to one-fiftieth of the present required board area, with corresponding reduction in cathode heating requirements."
    •      A quick calculation shows that this would be an area of only 3 square feet which is still large on a personal scale although that was not what he was envisioning. He couldn't foresee the subsequent development of large scale integrated circuitry which would reduce the physical volume to the size of today's hand held calculators.

    In the area of personal entertainment devices he correctly predicted:

  • "The match-case size radio receiver and the pocket-size television chassis will make possible radical price reductions."
    •      And finally, a prophetic observation:
  • "Would God that the cultural improvements in television programs were as certain as are the engineering improvements hereinabove visioned!"
    •      It is somewhat ironic to note that even though we do indeed have the pocket television sets that Dr. De Forest predicted, many of us still wonder whether there is really much worth watching.

    Also in the July, 1952 issue of R-E was an article on transistor production. It noted that most transistors being made were of the point-contact type. Western Electric was the only manufacturer producing junction devices, and then only on an experimental basis at the rate of less than 100 per month.

    Popular Science magazine published what may be the first ever transistor radio construction article entitled "Now You Can Build a Transistor Radio", in its October, 1952 issue. It is also fascinating in that it is the only article of this type that I have seen which uses point contact transistors! The author states that the Western Electric A-1698 devices could be purchased for $6.50 apiece from WECO in Allentown, PA. Based on other prices at that time this appears to be a real bargain. In reality WECO was probably just selling "factory seconds" which didn't meet the high standards required for use in the Bell Telephone System. The three transistor circuit was very simple and used one transistor as the detector. The other two provided audio amplification. If you didn't want to wait for the proper Cinch transistor sockets you could build your own using automobile fuse holders and spring brass to make clips for the two leads. The article cautioned, "Don't solder leads directly to the (transistor) prongs. The heat will ruin the crystal".

    A true milestone in transistor history occurred during the month of October, 1952 when the Bell Telephone System placed the very first transistorized equipment in the world into commercial service in Englewood, New Jersey. Point contact transistors were used to replace vacuum tubes in the oscillator circuits which generated signals that allowed 10,000 customers to personally dial long distance calls without the assistance of an operator. Other equipment soon followed and included rural telephone carrier amplifiers and headset amplifiers for operators.

    The major manufacturers in late 1952 were, in order of monthly production levels: Western Electric - 6,000, Raytheon - 1,000, General Electric - 800, and RCA - 400. Delivery times were from 4 to 8 weeks and Raytheon, GE, and RCA were projecting that sample lots of junction type transistors would be available by December, 1952.

    Production limitations were still hampering the widespread use of transistors and their high price remained a major barrier. The first transistors offered for sale by Allied Radio appeared on page 55 of their 1953 catalog. Both were point contact devices, the Raytheon CK716 (successor to the CK703) for $18.00 and the General Electric G11 and G11A for $17.40 each. Only the most dedicated and wealthy of hobbyists could hope to afford the luxury of experimenting with their own personal transistor, not to mention taking the risk of burning it up!

    Raytheon subsequently became the first manufacturer to mass-produce junction transistors. The first product announcement I have found is in the January, 1953 issue of the IRE Proceedings. It showed both the CK722 and CK721 devices and stated that the CK722 was available in production quantities, while the CK721 would only be available in limited quantities until April. Photographic date code evidence suggests that these devices were being made as early as late 1952 although these may have simply been pre-production samples. The differences between the CK721 and CK722 will be discussed later.

    A February, 1953 editorial in R-E stated that:

  • "Shortly after the advent of the transistor, handmade specimens sold around $18.00 apiece. [This appears to be a reference to the CK703 or CK716] Very recently the price has dropped to around $8.00. This naturally, is but a beginning. It is certain that transistors, when finally mass-produced, will sell at a lower price than present-day vacuum tubes."
    •      This was a price drop of over 50% in about two years. A February, 1953 article by Rufus Turner in Radio and Television News entitled "Care Of Transistors", had a sidebar stating that the price of the Raytheon CK722 was $7.60.  Click here to see the very first ad for the CK722 in the March, 1953 issue of QST.  Today this would be equivalent to about $46.00. One of the early Raytheon advertisements in R-E magazine gave the following exciting information:
  • "For the first time in history, Germanium Junction Transistors are commercially available. Raytheon Junction Transistors, types CK721 and CK722 can now be obtained for your experimental and developmental use. Here's another first for Raytheon! Leaders in the development and production of Electron Tubes and Germanium Products, Raytheon now leads the way in production of this important new electronic development. For price and delivery information of Raytheon Germanium Junction Transistors, write, phone, or wire your Raytheon tube distributor."
    •      The first commercially available consumer product to use transistors was the hearing aid, a device which benefited greatly from miniaturization and long battery life. Compared to the existing vacuum tube models of that era, transistorized hearing aids represented a quantum leap into the future. It is also interesting to note that Bell Labs waived the patent royalties for transistorized hearing aids in honor of Alexander Graham Bell who had been a life-long advocate for the hearing impaired.

    R-E reported in March, 1953 that completely transistorized hearing aids had been offered to the public for the first time. One of the pioneering companies was named Maico and they offered a three transistor unit called the "Transist-Ear" which was later described in some detail in the December, 1953 issue of R-E. Its dimensions were only 2 5/8" x 1 7/8" x 13/16" and it weighed only two ounces. Power was supplied by a 1.5 volt mercury cell which had an estimated life of 70 - 90 hours. The transistors were special (read expensive) Raytheon CK718 germanium types made especially for the hearing aid industry. Due to the wide variation of characteristics in early mass-produced junction transistors, the biasing resistors in the common emitter circuit design had to be matched to each transistor. This essentially amounted to each hearing aid being a custom-made device! The article noted:

  • "These individual transistors vary considerably in gain and noise as well as in impedance. As part of the quality control program established for the manufacture of these hearing aids, each transistor is checked in the laboratory and marked as a first-stage, second-stage, or output-stage transistor."
    •      I have a Zenith Royal T hearing aid (Zenith’s first transistorized consumer product) which shows quite plainly that the transistors were hand selected. The three devices are Raytheon CK718's and all are dated "339" or the last week of September, 1953. One is marked with a white paint dot on top and one has a yellow dot of paint on the side. The last is simply unmarked. The original price was $125.00. This would be equivalent to about $750.00 today.

    Despite these difficulties, the advantages in size, expected useful life and minimal power requirements clinched the transistor's superiority over vacuum tubes.

  • "The estimated life of transistors is about 100,000 hours, which would be equivalent to 20 years of use; compare this with the 5,000 hour life of a vacuum tube. (The switching of a transistor on and off may have some effect on its useful life.)"
    •      In May of 1953, QST magazine featured a Raytheon ad offering free copies of articles on "How To Transistorize Your Audio Amplifier" and "Building A Transistor Receiver".

    Raytheon announced a transistor application contest with $10,000 in prizes in the June, 1953 issues of both CQ and R-E magazine. 1st prize was $5,000, 2nd $2,000, 3rd $1,000, 4th $500, 5th $300, and 6th to 17th $100. The contest closed at midnight on August 31, 1953.

  • "Here's all you have to do! Simply build equipment that uses one or more Raytheon CK722 Transistors which you have purchased from your local Raytheon Tube Distributor. Then send a photograph of the unit you have built, a completed official entry blank and a 500 word minimum, 1000 word maximum constructional article on the equipment. All entries will be carefully judged and prizes will be awarded on the basis of ingenuity of application."
    •      The contest winners were announced in an April, 1954 news item in R-E.
  • "First prize of $5,000 in the nationwide Raytheon transistor application contest, was won by Robert T. Bayne of Los Angeles. The winning entry was an audio frequency meter using two CK722 transistors and four CK705 germanium diodes. Second prize of $2,000 was awarded to Peter G. Sulzer of Kensington, Maryland for his transistorized audio frequency and voltage standard. Third and fourth prizes were taken by G. F. Montgomery of Bethesda, Maryland and Lt. Robert Perkins, a Navy dental officer, respectively. Montgomery's entry was a general purpose a.c.-d.c. voltmeter using transistors, while Lt. Perkins' entry was a Vitalometer, a device for measuring the condition of tooth pulp, thus indicating the possibility of decay. Mr. Sulzer is already known to our readers, and they will see an article by Montgomery in one of our next few issues. Presentation to the top four prize winners was made by John A. Hickey of the Raytheon tube division, in ceremonies held in Philadelphia, Washington, and Los Angeles."
    •      The CK722 finally debuted in the 1954 Allied Radio catalog at only $4.50. It's higher quality brother, the CK721, was $12.50. Only six other transistor types were offered. The Raytheon CK716 point contact remained at $18.00 as did the G.E. G11 and G11A at $17.40. RCA's first series of four transistors also appeared: the 2N32 (P-C) at $15.40, the 2N33 (P-C) at $23.00, the 2N34 (P-N-P) at $13.40, and the 2N35 (N-P-N) at $18.40. It is easy to see why the CK722 became the darling of economically minded electronics hobbyists. It instantly dominated almost all the experimenter/hobbyist articles and continued to do so for many years.

    I have examples of at least three case variations which were made over the production life of the CK722. Despite the differences which will be described below, they are all based on a distinctively shaped small rectangular block about 1/4" wide x 1/8" deep x 1/2" high. The three wire leads are evenly spaced and the collector is identified by a red dot on the side of the case.

    The first CK722's (from late 1952 through at least part of 1954) were made of a black plastic with sharply defined edges and a clear glass base. These early units were not hermetically sealed in a metal case, but Frank Dukat of Raytheon remembers that the plastic encapsulation was "airtight" since moisture and contamination would cause failure. Moisture was a greater problem for grown junction transistors as opposed to the alloy junctions which Raytheon made. The lettering was white and the production date can be determined from a 3 digit code printed on the case. The format is YWW where Y=last digit of year and WW=week of the year. For the black units the date is printed on the back of the case. I have a "3-32" which would be 32nd week of 1953 (August) and a "416" which would be the 16th week of 1954 (April). Several R-E construction articles also support this format and clearly show two CK722's with date codes of 250 and 334. Another article shows two CK718's with codes of 245. My earliest junction transistor is a Raytheon CK721 dated 309 or the ninth week of 1953.

    The first variation of the CK722 was being made by 1955 and it had a beautiful deep blue painted metal case with a slightly more rounded appearance on the edges. It was also hermetically sealed and had white lettering. These units have the date code printed perpendicular to the leads just above the case bottom on the same side as the type number and Raytheon name. Otherwise the format is the same.

    The third and final CK722 variation I am aware of is almost identical to the blue jobs except that the case appears to be unpainted aluminum and has black lettering. I have examples of these from 1960 and 1963. Sometime after 1960 the date code format changed to YYWW. My 1963 device is marked "6352" which is the last week of 1963. Frank Dukat remembers that the final Raytheon CK722's were made in late 1964 or early 1965 when they closed their Lewiston, Maine plant and ceased manufacturing germanium devices.

    Frank also recalled the following interesting information. As time went on and technology allowed transistors to became smaller, Raytheon still had to maintain the originally registered case size for the CK722 and its close family. The later transistors were actually much smaller and were simply inserted into dummy cases of the correct larger size! According to Frank if you cut one of the later ones open you will find a much smaller one intact inside. I have verified this to be true.

    The September, 1953 issue of R-E carried a construction article for a simple hearing aid housed in a metal cigarette case. It was based on commercial products but used the popular CK722 transistors instead of CK718's or CK721's which were a higher grade and thus more expensive and less available to the private experimenter. One compromise was the necessity of using a 15 volt battery instead of a 1.5 volt one. Another caution was:

  • "The four fixed resistors must be selected by experiment for the individual transistors. There is enough normal variation in CK722 characteristics to necessitate this picking process."
    •      The article concluded by pointing out that the project was not intended to be the ultimate in sub miniature, low power transistorized hearing aids, but was meant to be:
  • "an answer to a challenge to develop the smallest practical instrument that can be built from parts obtainable, which would use the readily available CK722 transistor, and which we might reasonably expect a radioman to duplicate with ordinary tools."
    •      The Raytheon Manufacturing Company was also honored in late 1953 by the Society of Hearing Aid Audiologists, who presented them with their 1953 engineering award for achievement in transistor design and production.

    An interesting discussion of the production variations in small signal PNP junction transistors was written by F. M. Dukat of the Raytheon Manufacturing Company in the September, 1953 issue of Radio-Electronic Engineering magazine.

  • "First of all, it might be well to acknowledge that at the present state of transistor development no manufacturing techniques are known which can control transistor parameters within close limits. Selection after manufacturing is necessary. Therefore if transistors are to be used commercially today, a knowledge of the extent of their variability is very important in order to minimize this variability by suitable circuit design."
    •      He further clarifies the difference between the CK721 and CK722.
  • "Two Raytheon junction types are now available to the military and general trade: the CK721, which can be considered a relatively high grade unit; and the CK722, a medium grade unit which is useful in many applications. The basic difference between the CK721 and the CK722 is one of current amplification, the CK721 having an alpha of about .95 or greater while the alpha of the CK722 runs from about .85 to .95."
    •      The article concludes:
  • "The foregoing analysis may be summarized as follows:

    •  

     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     

    1. Presently available transistors are quite variable in characteristics, and although more is being learned daily about control, units must still be selected after manufacture.

    2. Some form of d.c. compensation is almost mandatory to control operating points.

    3. Although the highest gains are available in the grounded emitter amplifier connection, the spread of gain in the grounded base amplifier is less.

    4. Designs for transistor amplifiers for production purposes would do well to sacrifice appreciable stage gain by suitable feedback in order to reduce unit to unit variability."

         Frank recalled his 1953 article for me in 1999. All the CK718, CK721, and CK722 transistors came from the same production run. The alloy process consisted of fusing indium spheres to a tiny chip of germanium. The goal was to place the spheres very close together but obviously not to touch. The distance between the spheres was the base width and this determined the current gain in the device. The resulting batches of transistors had just about as many shorts as they could tolerate. The units which were not shorted had a very wide range of current gains, voltage breakdowns, and leakage currents. The highest quality transistors were labeled CK718 and were set aside for hearing aid use. The CK721's were selected next from those units which had a high enough current gain. The CK722 was what was left after all the other selections had been made. Even after 10 to 12 years the alloy process was still not very well controlled. Fortunately the bulk of Raytheon transistor production was  used for computer switches whose specifications were easily met. What was left over went mainly into portable radios. Raytheon's largest customer was General Electric.

    Further clarification of the transistor's importance was provided in the November, 1953 issue of R&T News.

  • "Washington celebrated an unusual birthday a short while ago; that of the transistor. [It is interesting to note that the "official" birthdate referred to in this article seems to have been the July, 1948 public announcement date of the device.] Five years ago this mighty mite was born, and today it has become quite a factor in the civilian world, and particularly on the military front. In countless defense projects, this unique device has served to expedite equipment construction, improve performance of the most involved systems, and shrink the size of complex gear. To the military, the transistor has been one of the most significant developments since the invention of the vacuum tube. Thus, on this fifth birthday, it was hailed and applauded in the Pentagon, at numerous depots throughout the country, and at many plants now engaged in producing transistorized devices for defense. To the transistor, a roaring salute on this happy occasion."
    •      An ad in the May, 1954 issue of R-E offered "NPN grown junction" transistors "manufactured by Germanium Products Corp. under Western Electric Co. license" for $3.95 each. This would be equivalent to about $24.00 today.

    A November, 1954 R-E news item recognized the following achievement:

  • "A milestone in transistor history was reached on June 15 when Raytheon produced its one-millionth transistor. The company, which states that it has produced three times as many transistors as all other manufacturers combined, made point contact transistors available in 1948. In late 1952 it began production on germanium junction units, which comprise the large majority of all transistors made to date. So far, most of Raytheon's transistor output has gone to the hearing aid industry."
    •      In December, 1954, just in time for Christmas, the first commercially produced transistor radio reached the marketplace under the Regency brand name. It was given a model number of TR-1, used transistors manufactured by Texas Instruments and sold for $49.95. This would be equivalent to about $300 today. It is interesting to note that because of their collectibility the surviving TR-1's today exceed their original purchase price in terms of actual dollar value. The original TI press release announcements for the TR-1 are presented as attachments at the end of this article.

    A July, 1955 R-E news item quotes the lowest Raytheon transistor price as being $1.70 in large quantities, down about half from the year before. The hobbyist could also purchase single CK722 transistors through various mail order distributors. The first Lafayette Radio ad in R-E to feature transistors included the CK722 and appeared in the July, 1955 issue. They were priced at $2.45 each (or $2.25 ea for 10 or more). A price of $2.45 in 1955 would be roughly equal to $15.00 today.

    Prices continued to plunge dramatically over the next few months as production increased. An August, 1955 Lafayette ad lists the CK722 for $2.10 each (or $1.95 ea for 10 or more). The GE 2N107 was listed as NEW and was $1.25 each. The November, 1955 Lafayette ad showed both the CK722 and 2N107 at $1.25 each.

    By December, 1955 Lafayette was selling the CK722 for $.99 each. Only seven years had passed since the transistor was born in 1947 and it was now ready to break into the mainstream of American manufacturing technology.

    It is easy to forget just how close we are to transistors in our daily lives, but we are rarely more than a foot or two away from them. Although they are no longer typically used as discrete components, they hide by the millions inside tiny integrated circuits in computers, watches, radios, automobiles, and other items too numerous to mention. Our world would be a very different place today without this amazing little device. The invention of the transistor has truly had a profound effect on the course o f civilization during the past fifty years and its influence will continue into the future.

    --

    References:

    Radio-Electronics (R-E) magazine - January, 1952 to December, 1955

    QST magazine - May, 1953

    CQ magazine - May & June, 1953

    LIFE magazine - April, 1953

    Texas Instruments Internet Web site

    Lucent Technologies (former Bell Labs) Internet Web site

    Email interview with Mr. Frank M. Dukat, retired employee of Raytheon, March, 1999
     
     

    Exhibit 1

    The Original Press Release On The Regency Radio

  • Information Bulletin

    •  

     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     

    DALLAS, Texas

    October 18, 1954

    TI Supplying Transistors for First "Pocket Size" Radio

    TI's part in the production of the first commercial transistorized radio receiver will be announced beginning today in newspapers throughout the country.

    The "pocket size" radio has four TI transistors. It also uses a TI sub miniature output transformer.

    The transistors - technically known as n-p-n grown junction germanium triodes - are made in the Semiconductor Products Division. The transformer is a product of the Components Division.

    Their application to the new radio receiver accounts for a large part of the increased production and employment in the two divisions in recent weeks. The Apparatus Division has also had a hand in the tiny radio, having worked on engineering problems and the machining and fabricating of models for the plastic case.

    The receiver is being assembled and marketed by the Regency Division of Industrial Development Engineering Associates, Inc., Indianapolis. It will be available this week to the public through sales outlets in Los Angeles and New York City. When production permits, it will be sold in cities throughout the country. For competitive reasons it was decided to keep the development of the radio, including TI's part in it, "under wraps" until the unit was ready for marketing.

    The radio receiver measures 5 x 3 x 1 1/4 inches - the smallest set commercially available - with the semiconductor devices themselves occupying less than 1/10 of a cubic inch. The "pocket size" is a significant achievement since it includes a high fidelity, high volume speaker and a single battery supply as well as all associated receiver circuit components.

    Gain at radio frequency with the germanium transistor is sufficient to permit a combined mixer-oscillator stage. Only two intermediate frequency stages are required and, following a germanium diode detector, one audio amplifier stage. Audio volume fidelity and reception range are the equal of or superior to that of the small vacuum tube-equipped portable radios.

    The introduction of this first mass production item to use the tiny transistor to replace the fragile vacuum tube leads the way for the long-predicted transistorization and miniaturization of many other mass production consumer devices. TIers can justly be proud of being the first to produce a high-gain transistor at a cost permitting its application to the high-volume commercial market.

         Exhibit 2

    Transistor Manufacturer Comments on New Radio

  • Information Bulletin

    •  

     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     
     

    Dallas, TX

    October 18, 1954

    Following the announcement today of the first commercial transistorized radio receiver, Executive Vice-President P.E. Haggerty of Texas Instruments Incorporated--leading transistor manufacturer--issued a statement of congratulations to the Regency company on its significant electronic achievement.

    In a message to Mr. Edward C. Tudor, President of Industrial Development Engineering Associates, Inc., Mr. Haggerty said, "As the transistor manufacturer, we sincerely congratulate your Regency division on being the first to produce commercially a transistorized consumer product. With the introduction of this first mass production item replacing the fragile vacuum tube with the tiny transistor, Electronics enters a new era," Mr. Haggerty continued.

    Texas Instruments--manufacturer of the new low cost, high gain transistor, which makes the first commercial transistorized radio possible--is one of the largest producers of germanium transistors and the exclusive producer of silicon transistors for high temperature applications. Its latest development, the low cost, high gain transistor used in the Regency radio, is one of the greatest practical advances yet made in transistor technology since it makes possible the long predicted transistorization and m iniaturization of mass production consumer devices.

    "We are proud to have contributed to Regency's pioneering achievement." Mr. Haggerty continued. "Production of a high gain transistor at reasonable cost has been a long-sought goal of the entire Electronics industry. The Regency company deserves special commendation for the remarkable speed with which it converted this latest transistor development into a superior, marketable product," he concluded.

    Kernel-sized substitutes for vacuum tubes in electronic amplifying, until now transistors have been limited in consumer application. The largest commercial application for transistors has been in hearing aids, in which the benefits of miniaturization, long battery life and high reliability have justified their use. Texas Instruments has been an important supplier of transistors for hearing aids, as well as for many specialized commercial and military applications.

    The new Texas Instruments low cost, high gain transistors used in the revolutionary new Regency pocket receiver are technically termed n-p-n grown junction, germanium triodes. Such a high gain is obtained with the new transistors--up to 40 decibels per stage--that only four are required by the radio.

    Even so, audio volume fidelity, and reception range are the equal of or superior to that of the small vacuum tube-equipped portable radios. Grown junction construction--wherein the germanium crystal layers of dissimilar electrical characteristics are grown into the transistor--contributes to the transistor's performance and reliability.

    The Regency radio is made possible by an efficient circuit design taking full advantage of the unique features of the new Texas Instruments low cost, high gain, high frequency transistor in combination with the advantages common to all transistors. Gain at radio frequency is sufficient with the new transistor so that one transistor is used for a combined mixer-oscillator stage. Only two intermediate frequency stages are required and, following a germanium diode detector, one audio amplifier stage.

    The first completely transistorized commercial broadcast receiver has a resulting "form factor" of 5 x 3 x 1-1/4 inches--the smallest compete set commercially available--with the semiconductor devices themselves occupying less than 1/10 of a cubic inch. The "pocket size" is considered a significant achievement, since it includes a high fidelity, high volume speaker and single battery power supply as well as all associated receiver circuit components. The subminiature output transformer is also of Texas I nstruments manufacture.

    Now that the low cost, high gain transistor has made possible the first mass production transistorized device, Texas Instruments anticipates that the resultant increase in transistor production will make economically feasible the transistorization of many more consumer products. TI predicts that the first commercial production of a low cost, high gain germanium transistor will rank in importance with the first commercial production of the high temperature, high frequency silicon transistor...also exclusive to Texas Instruments.

    Return to transistor page