Raytheon was founded by Laurence Marshall, Vannevar Bush and Charles Smith in 1922 and was originally known as the American Appliance Company. It was formed to develop a refrigerator with no moving parts. This was unsuccessful and the founders elected to begin making gas filled tubes designed for "battery eliminators" that enabled early vacuum tube radio sets to run without expensive high voltage “B” batteries. The tube was launched late in 1925 with the claim “Revolutionizing the B-Eliminator.” The new direction was celebrated with a new company name: Raytheon and it concentrated on vacuum tubes. [Stokes 1982 Photo courtesy of Joe Knight] In 1928 Laurence Marshall became the company president and remained in this position until 1948.

Percy Spencer became a key figure through his innovations in tube design and production. He redesigned the B rectifier which became the BH and the BA enabling higher operating voltages and introduced a line of transmitting tubes. His innovations in magnetron design and production created a platform for Raytheon's success in radar during the Second World War.

In 1938 Percy Spencer also designed a range of miniature tubes intended for portable applications such as hearing aids. Ruggedized versiona of these were used in proximity fuses that revolutionised artillery during the War. Raytheon’s experience in the hearing aid market and understanding of its needs was a key factor in their aggressive response to the opportunity created by the transistor in 1948.

In 1942 Raytheon built its first radar set, the SG, widely deployed by the Allied navies in 1943. The SO, a smaller model, appeared shortly afterwards. Production reached a maximum by 1944 and at its war time peak the company employed 18,000 workers; but employment had dropped to 3,500 by 1946.

So Marshall diversified the company into consumer products: he purchased the Belmont Radio Company and the Russell Electric Company that produced phonograph parts. He established a network of microwave sites that he intended would support television and FM radio networks.

Raytheon found peacetime uses for wartime technologies. It had been noticed that microwaves generated by radar sets caused heating; an observation that led Raytheon to invent the microwave oven which was sold under the Radarange brand.

In 1946 the company merged with another major war time contractor facing diminishing revenues: the Submarine Signal Company which specialised in sonar. The architect of the merger, Charles Adams, was invited to join the board of the expanded company and two years later replaced Marshall as president. Marshal left the company in 1950.

The military is imbedded in Raytheon’s DNA. In 1948 Raytheon began a guided missile programme and in 1950 it successfully demonstrated that its Lark missile guidance system could destroy an aircraft in flight. Development of a succession of air to air and ground to air missiles was accelerated by the Korean War.

Raytheon grew thanks to the exceptionally entrepreneurial leadership of Laurence Marshall, the risks he was prepared to take and the opportunities that arose out of the Second World War, the Korean War and the Cold War. Its core technology evolved from tubes to magnetrons and radar then to missile guidance systems.

Raytheon was never a semiconductors company: its histories scarcely mention them. Radar crystal detectors were an important element in the development of radar but Raytheon will be remembered for its ability to devise the means to mass produce the magnetron and to build battle ready radars and guided missiles.

The Semiconductor Era

The point-contact transistor was first developed and demonstrated at Bell Laboratories in December 1947 representing the first fragile entrant of the semiconductor age. In the ensuing months further work was carried out in support of a patent application which, when in place, would enable Bell to announce its breakthrough to the World.

In a carefully orchestrated programme, Bell first disclosed its invention to a gathering of representatives from the Army, Navy and Air Force on 23^rd June 1948. On the 30^th of June a similar presentation was made to the mainstream and technical media. This was calculated to make an impact: reporters were invited to listen to the commentary on earphones fed by a transistor amplifier. Then they heard the first public demonstration of a fully transistor radio tuned to local stations.

Bell’s press release simply stated: “An amazingly simple device, capable of performing efficiently nearly all the functions of an ordinary vacuum tube, was demonstrated for the first time yesterday at Bell Telephone Laboratories where it was invented.”

Then Bell attended to the scientific and industrial communities. In the week prior Shockley at Bell had arranged for the publication of three short papers in the July edition of Physical Review of which the first The Transistor, A Semiconductor Triode, described the basic construction and performance of the device. [Bardeen 1948]

The following week Bell sent out hundreds of letters to manufacturers, research laboratories and the technical press inviting them to attend a seminar on the 20^th July. [Riordan 1997]

At Raytheon Bell’s letter dated 9 July 1948 was addressed to Laurence K. Marshall, then president of Raytheon, inviting him to Bell’s Murray Hill NJ laboratory to see a demonstration of "a new device called a Transistor." [Goldstein 2003]

Marshall took Norman Krim to the Bell presentation and both were inspired by what they saw. [Earls 2005] “On their return Krim prepared a report predicting that tubes would become obsolete. That was somewhat sweeping, but it was a moment of high excitement. He suggested a crash program to produce germanium semiconductors for use in the hearing aid industry.” [Scott 1974 Citing Krim] Krim had developed the Raytheon line of miniature tubes in 1938 and the company was successfully selling them for hearing aids. Here was the perfect application for the new transistor: Needing only three transistors their likely high cost was not an issue. And it was immediately clear that the transistor was a major breakthrough that would reduce battery costs and weight.

Raytheon were in mixed company. Braun and MacDonald comment “Even the technical journals mostly managed to control their enthusiasm, many waiting until late 1948 to report the event, some until 1949, and some not bothering at all. This seems strange considering the excitement of the transistor’s inventors but the explanation is probably two fold. At the time of the announcement, the transistor was little more than a laboratory curiosity. It was to become years before the device was made to do something useful.” [Braun 1982]Electronics magazine made it their cover story in their September edition with the headline “The Crystal Triode: Revolutionary Amplifier.”

And yet the vision was limited to replacing tubes and where better to replace them but in small portable appliances where even miniature tubes caused excessive drain on lightweight batteries.

Despite this Raytheon asserted “more dependable performance” in their advertising claiming seven features including the glass to metal seal at the anode end and “superior” humidity characteristics.

Early on Raytheon set an agenda for colourful packages as this set from 1952 indicates (CK705, CK707, CK708 and CK710). The colored sleeve insulated the metal body.

Encapsulation was changed in later production: the new narrower shorter tube was dipped in red epoxy as shown in these examples of CK705s from 1952 (yellow) and 1954 (red). Early on Raytheon signaled its intention to be a mainstream player and obtained RTMA registration on April 18^th 1951 for its first three diodes. [RTMA 1951] These were:

By the time Electronics Magazine published its September article on the new Crystal Triode Norman Krim and his team were well on the way to prototyping their first transistor: the CK703. “In order to do it Krim had purchased diodes from Sylvania for their germanium content...” [Scott 1974 Citing Krim] George Freedman, a metallurgist that had worked for Raytheon since 1946 was sent to Bell to learn how to produce point-contact transistors. He recalls with some pride meeting William Shockley. [Ward 2001 Photo courtesy of Andrew Wylie] The first provisional data sheet is dated November 3 1948 only five months after the Bell announcement reflecting the company’s ability to mount an urgent development programme reminiscent of the War years. [Raytheon 1948]

But in common with all early point-contact transistors this was not a robust device and Krim, retrospectively discussing the issues in 2003, noted it had to be “handmade with watchmaker precision, which precluded cost-effective mass production. And they were none too robust. The slightest shock could ruin them, which made them useless for hearing aids and just about everything else.” [Goldstein H 2003]

The first advertisement for this transistor was in the Radio Master Catalog 15^th edition for 1950-51 where the price was listed at $18.00. [McGarrah] The first reference to the CK703 in the literature was in the January 1950 edition of Radio & Television News which gave building instructions by Rufus Turner for a crystal set with a 3 transistor amplifier based on CK703s. This was a curiosity and would have cost over $60 in parts. By comparison the same magazine advertised an 8 tube AM-FM chassis for $32.95! A further article by Rufus Turner published in the Theory and Engineering section of the June 1950 edition of Radio Electronics magazine discussed eight experimental CK703 circuits.

In 1948 Bell had filed patents on the point-contact and junction transistor and had invented the name “transistor” but its licensing programme did not begin until 1952. Early on some companies exercised caution on the use of what could have been a trade-name and called their devices “Crystal Triodes.” Raytheon’s provisional data sheet uses this description although the above example is emblazoned “Transistor.”

The CK716 was introduced in 1952 and had the same outline as the CK703. It was intended to be more rugged and had a shock test specification of 900g. [Raytheon 1952] In fact this is not rugged at all. For example, Bell stated in an internal memorandum that its prototype transistors could withstand shocks of 20,000g. [Powers 1951] Raytheon advertising for the launch reminded readers of the pedigree of its transistors: “In 1948 the Raytheon point-contact transistor is perfected and put into production, now superceded by the improved type, CK716, currently available.”

Junction Transistors

Although Shockley invented the junction transistor concept in 1948 [Shockley 1948] it remained a curiosity until 1951. In April 1950 the first grown junction transistor was made by the double doping method invented at Bell. “This non-photogenic device did perform according to theory but had a wide base, and poor frequency response and provoked little interest. For about nine months, the efforts to improve junction transistors were practically negligible at the Laboratories.” [Shockely 1976] But Shockley was confronted by the Korean War and saw that junction transistors could be used in proximity fuses for mortar shells where miniature tubes were too bulky. Work was resumed on its double doping technology and approaches to reduce the base width of the transistors made this way. [Riordan 1997]

In the Spring of 1951 Shockley was ready to tell the World that usable junction transistors had been made and was working on a paper for Physical Review ultimately submitted in May and published in July that year. The improved transistors gave higher gains with less noise than point-contact transistors and were “exceptionally good low power amplifiers” that would also deliver output powers of several hundred milliwatts if needed. [Shockley 1951]

Norman Krim recalls rooming with Shockley while he worked on the paper. Both men were serving on a military procurement advisory board known as the Baker Committee that needed advice on military electronics. “Shockley would be proofreading a paper after dinner every night. He told me, 'I'm going to publish an article in the Physical Review, and you should remember, pick up that article.' When I got a copy of his article on junction transistors, that was it for me. The light bulb went on." [Goldstein 2003]

But there was a shock in store for Bell: they were gazumped by General Electric who produced a better prototype junction transistor than Bell: the alloy junction transistor. Both Bell and General Electric presented their work at the IRE Electron Devices closed conference at Durham in June 1951 and put the semiconductor industry on notice that there were now two approaches to a viable junction transistor. General Electric’s technology looked the best bet: they could demonstrate high frequency performance, voltages to 150v and large area versions could dissipate 8 watts. [Saby 1952]

The General Electric alloy junction transistor was produced by alloying tiny dots of indium metal on either side of a thin N-type germanium wafer. On heating and alloying the indium formed a P-type germanium layer under each dot. This created a PNP junction transistor. Leads were connected to the wafer (base) and the indium dots (collector and emitter) and the entire assembly encapsulated for protection. [Saby 1952]

Thus Krim decided that Raytheon would make alloy junction transistors. Raytheon knew the hearing aid market very well and transistors were ideally suited for this application: They would first target the hearing aid market.

Licensing

Bell had a pioneering position on both the point-contact transistor and the junction transistor and anyone producing either would infringe its patents. Additionally General Electric had patented its method for producing alloy junction transistors. In order for Raytheon to make alloy junction transistors it needed a license to both Bell’s and General Electric’s transistor patents.

Comprehensive licensing arrangements between electronic companies were common in the USA since General Electric and Westinghouse made RCA their exclusive marketing channel for receivers and tubes in return for cross licenses to their patents in 1919. Federal Anti-trust legislation both encouraged pooling of technologies while challenging any trend towards monopolies. Companies commonly swapped rights to substantial patent estates for little or no royalty. [Tilton 1971] For example, Raytheon had such a deal with RCA as Ivan Getting, head of research at Raytheon recalls:

“Every year I had to sit down with the RCA top people and negotiate how much we owed them, and how much they owed us for their using our patents. We would host them at a dinner downtown, at a restaurant, which was famous in Boston (Lockober). After a few drinks, a good meal, a few more drinks, I would say, “Well, it's obvious that we can't go through your ten thousand patents and claims, and that you can't go through our one thousand patents and claims that you may or may not be using, so why don't we just negotiate a settlement?” And they'd say, “What do you suggest?” Since I knew what we had been doing, I'd say, “We'll pay you two million dollars, and you give us a license to use all your patents, and we'll give you a license to use all our patents, but you're not allowed to sell them to others. You have to go back to direct source.” One more drink, and they'd say okay and we'd sign a document.” [Nebeker 1995]

General Electric License

The outcome from Durham was a strong consensus amongst attendees that the alloy junction technology was the preferred approach. For example, RCA dropped its work on point-contact transistors and began an alloy junction programme. General Electric started licensing out its know-how. This started with its affiliates such as the British Thompson Houston Company and Compagnie Francaise Thompson Houston.

Norman Krim recalls that he “negotiated getting a license from General Electric to make germanium alloy transistors in the summer of 1951 by approaching Raytheon President, Mr Charles Adams, whose father was then on the Board of Directors of General Electric. Mr. Adams, Dr. Getting and Dr. Bowles, a consultant and I all flew to Clyde, NY outside of Syracuse to see their experimental germanium alloy transistor live in mid-summer 1951.” [Krim 2009]

George Freedman recalls: “Krim then made a deal with GE in Auburn New York. He gave up some Raytheon knowhow in return for allowing a delegation, of which I was one, to learn how to make alloy junction transistors.” [Ward 2001]

Adam Sheckler of General Electric confirms this writing “An odd exchange occurred when management told us to teach Raytheon the art. This was an exchange of information. Management valued our semiconductor so poorly that they exchanged it for some radar data. It is interesting that when Raytheon was asked about ‘their invention of the transistor’ (about 1998) they responded ‘we didn’t invent it, GE taught us the art.’ ’’ [Sheckler 2004]

Bell Laboratories License

Western Electric began offering licenses to the Bell Laboratories transistor patents after the Shockley junction transistor patent issued in September 1951. By agreement with the Military, these were restricted to NATO countries. Licensees paid a $25,000 upfront fee deductible against future royalties and were entitled to attend the third symposium in April 1952.

“Bell had waived all license fees for the first commercial transistor product. As a memorial to Alexander Graham Bell and to his interst in the deaf, Bell did not require royalties on transistors produced for hearing aids after 1954.”

“Most companies at the first Bell commercial transistor symposium eventually produced their own transistors under license from Bell. Other companies cared less for protocol and secured their transistor knowledge from RCA with whom Bell had a very comprehensive cross-licensing arrangement which included the transistor. There were still other companies which manufactured transistors for many years without benefit of anyone’s training or license.” [Braun and Macdonald 1982]

In 1948 Bell had supported the Raytheon point-contact programme by hosting George Freedman so it had at least an informal license. What is documented is that Raytheon had a license to the Bell transistor patents after September 1951 and attended the April 1952 symposium along with 24 other manufacturers. The Raytheon team of four was led by George Freedman. [Ward 2001] Participants received a two volume reference set entitled “Transistor Technology” that comprehensively set out practical information that would assist the new licensees set up production. It became known as Ma Bell’s Cookbook. [Bell 1952]

Raytheon’s First Junction Transistor: CK718

Armed with the Bell licenses, Ma Bell’s Cookbook and the licenses and technology transfer package from General Electric, Raytheon was ready to begin development of its first commercial junction transistor. “Two million dollars were set aside for the effort. That was a large sum for Raytheon at the time, and a clear indication of the importance attached to the project.” [Scott 1974]

Adams, then the President of the company recruited Dr Ivan Getting. His background was military research including the Radiation Laboratory at MIT and the Pentagon. Getting was made VP Engineering and Research and commenced with the company in August 1951. A new Research Division and semiconductors were important priorities. He was unimpressed with what he found: “In the fall of 1951 the status of transistor work at Raytheon was pitiful. One engineer, in the basement of an old red brick factory in Watertown, was melting germanium in a crucible, solidifying the melt into a heterogeneous mess of crystals and building point-contact devices, some of which showed amplification. The work was not scientific; there was no metallurgist or physicist involved. It was clear that Raytheon had its work cut out.”

A new building was leased and new hires made. Walter Leverton was put in charge of transistor research in the Research Division and in the Receiving Tube Division Krim appointed George Freedman to head a dedicated semiconductor team located at Raytheon’s Chapel Street building. Soon junction transistors were being produced in both divisions. Raytheon successfully bid on a Signal Corps and Navy research contracts to improve transistor production processes “and between these and the Company research money a substantial program was underway by 13 May 1952.” [Getting 1989]

The Research Division focussed on silicon transistors while the Receiving Tube Division developed germanium transistors for the hearing aid market.

Additional staff were appointed to work on germanium. Surprisingly Dr Charles Smith, one of the company founders and inventor of its first tube, joined the team. [Scott 1974] Others on the team were Herbert Starke and Lowell Pelfrey who had been responsible for the diode line. [Ward 2001]

The CK718 was launched as a transistor for hearing aids in September 1952 [Raytheon 1953]. Norman Krim who had worked with the hearing aid industry from 1938 headed the promotion of the new transistor. Samples were provided and sufficient demand established to launch production late 1952. In this early example the collector is indicated by a paint-filled drilled dimple. [Photo courtesy Terry Hosking]

News carried in the New York Times for January 26, 1953 reported that shipments of transistors had begun to more than fifteen manufacturers of hearing aids.

“But as Raytheon prepared to introduce its germanium junction transistor, dubbed the CK718, yields stayed stubbornly low. Water vapour and other environmental contamination occurring during the manufacturing process were to blame. To get around the problem, Krim's team used infant incubators as "clean boxes," so technicians wearing rubber gloves could reach in and assemble transistors while minimizing exposure to ambient conditions. Yields went up, and by the end of 1952, Raytheon released 10,000 CK718s to its commercial customers, the hearing-aid manufacturers.” [Goldstein 2003]

Writing to Jack Ward, Norman Krim confirms the comments about complaints from early customers. “It is correct that G E Gustafson VP Engineering Zenith called me soon after Feb 1953 and told me that a few of their transistorized hearing aids were failing in Texas, on the Gulf and Mexican borders. We immediately tested our transistors under very high humidity and found a few failing due to moisture getting in through our plastic moulded case and reaching the junction. We soon rectified the problem.” [Ward 1998] Frank Ducat, Transistor Application Engineering Manager at the time recalls these issues and considered it less of a problem: “Moisture never did penetrate our plastic packaging. But I recall that when the story arose about this problem I stuck some CK718 in a money vest around my body inside my clothes and wore it for a month! Nothing happened.” [Ward 1997]

In 1954 Dukat published an analysis of field failures which showed that Raytheon transistors were more reliable than tubes. It did admit to an issue with open connections which had already been addressed but the data presented showed that transistors were twice as reliable as miniature tubes. The data was used to justify plastic encapsulation although within a year Raytheon had moved at least some of its production to hermetic sealing. [Dukat 1954 courtesy Jack Ward]

The first CK718 were mounted on a glass stem. The active elements were protected with a wax coating prior to encapsulation in a black epoxy moulding. The tradition of packaging the transistors in tube style boxes of the kind used for the point-contact transistors continued. The CK718 continued until 1955 when it became obsolete on the introduction of Raytheon’s blue metal encapsulated transistors. Even then it survived on the surplus market being advertised at disposal prices as late as 1957. [McGarrah]

CK721 and CK722

Krim had been an enthusiastic hobbyist building a mechanical television set as a youth in the 1920s. Surely, he thought, there was a demand for transistors from electronic enthusiasts keen to experiment with their first transistor. "I thought, jeez, wouldn't these rejects make a hell of a good thing? So when the guys wanted to break them up, I said, you can't do that—they're worth something. I loved to build experimental stuff and I just wanted the kids to have these. And nobody had ever seen a transistor." Virtually as soon as production of the CK718 began, late in 1952, Raytheon began labelling low gain CK718s as CK722. In order to promote them Krim invited the major electronics publishers to Raytheon to see the new CK722 demonstrated in February 1953. [Goldstein 2003]

In fact all CK718, CK721 and CK722 came from the same production run. Frank Ducat confirmed: “The highest quality transistors were labelled 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.” [McGarrah 1999]The issue of quality control persisted throughout Raytheon’s brief history of transistor production. In an oral history for Jack Ward Ducat recalls that at the Lewiston plant they sorted transistors by gain and voltage breakdown into 200 lots from which customer orders were despatched. [Ward 2001]

“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.” [McGarrah 1999 Photo courtesy of Bob McGarrah]

Early Dominance

The April 1953 edition of Raytheon News reported that the company was the World’s largest producer of transistors making tens of thousands per month. It quoted Fortune Magazine (March 1953).

"Fortune Magazine calls 1953 the year of the transistor. In its March issue, Fortune told how the transistor "the pea size time bomb" is revolutionising the electronics industry. Since the beginning of the year when mass production of transistors got underway, Raytheon is leading as the biggest producer of junction transistors. Production of transistors at Raytheon is at the rate of tens of thousands per month.

"In the transistor," Fortune said, "man may hope to find a brain to match atomic energy's muscles." One of electronic’s wonder devices, transistors free devices from the limitations of vacuum tubes by providing compactness and low power consumption.

The first small commercial returns of transistors are coming in from the immediate practical application of the hearing aid trade. Over ten concerns are now manufacturing tubeless hearing aids using three Raytheon junction transistors.” [Reproduced in Earls 2005]

Millionth Transistor

Raytheon patented the process improvements it made where it could including inventions that it did not commercialise. An example of one approach patented in 1952 that it did not explore was a compact axial transistor more in the style of crystal diodes being sold at the time. [Blais 1952] This configuration was used by Hughes Aircraft for some of their point-contact transistors.

under an inert atmosphere to avoid contamination of the semiconductor surface. Tinning avoided the need for fluxes on final assembly also reducing the risk of contamination. [Parziale 1956] On introduction of the new can, Raytheon began to use its distinctive sapphire blue paint on its germanium units which it continued through to the late 1950s. At the same time it began obtaining JEDEC registration of its transistors and used 2N type numbers in parallel with the CK numbering series.

The oval outline shown above was hermitically sealed. Other metal cased transistors were resin encapsulated.

New AF Transistors

As part of the hermetically sealed range of Raytheon Blues a new audio frequency range was released by Raytheon in 1954-1955. These are indicated in the table:

These transistors were not successful as the metal tab permitted increased ingress of moisture, something that was already problematic in Raytheon’s resin encapsulated transistors. “A good plastic to metal seal is difficult to realize, and even if attained, is subject to deterioration with extended use...” [Parziale 1956] The introduction of metal cases around 1955 created the opportunity to improve heat transfer by soldering a metal tab to the emitter dot which, on final assembly, was soldered to the top of the can creating a large area heat sink that improved cooling.

Given the issues of sealing the metal tab in the CK750, Raytheon introduced the CK751 which was in a sealed metal case with a fitted external fin that enabled it to be bolted to a metal chassis. It was specified by Olympic for their four transistor portable, the Olympic 447, as a class A audio output transistor and some Emerson 868 sets when they utilised Raytheon transistors.

This is a rare transistor. This example shown is date coded for 1956. [Photo courtesy of Joe Knight] When used in Class B output stages and with the CK751 strapped to a metal chassis 500mw output was possible. The other “power transistor” available from Raytheon at the time was the 2N138A which could deliver a very modest 100mw in class B. [Raytheon 1957]