Feature by Dr Bruce Taylor HB9ANY

l E-mail: [email protected]

A The Transistor
t Christmas 1938,
working in their small
rented garage in Palo
Alto, California, two

Revolution (Part 1)
enterprising young men
called Bill Hewlett and Dave Packard
finished designing a novel wide-range
Wien bridge VFO. They took pictures of
the instrument sitting on the mantelpiece
in their house, made 25 sales brochures
and sent them to potential customers.
Thus began the electronics company Dr Bruce Taylor HB9ANY describes the invention of the
that by 1995 employed over 100,000
people worldwide and generated annual
tiny device that changed the course of radio history.
sales of $31 billion. The oscillator used
five thermionic valves, the active devices
that had been the mainstay of wireless
communications for over 25 years.
But less than a decade after HP’s first
product went on sale, two engineers working
on the other side of the continent at Murray
Hill, New Jersey, made an invention that was
destined to eclipse the valve and change
wireless and electronics forever. On Decem-
ber 23rd 1947, John Bardeen and Walter
Brattain at Bell Telephone Laboratories (the
research arm of AT&T) succeeded in making
the device that set in motion a technological
revolution beyond their wildest dreams. It
consisted of two gold contacts pressed on
a pinhead of semi-conductive material on a
metallic base.
The regular News of Radio item in the
1948 New York Times was far from being
a blockbuster column. Relegated to page
46, a short article in the edition of July 1st
reported that CBS would be starting two
new shows for the summer season, “Mr
Tutt” and “Our Miss Brooks”, and that “Waltz
Time” would be broadcast for a full hour on
three successive Fridays. But right at the
end, after another unexciting story about
the broadcasting of road traffic reports, the
article mentioned that Bell Labs had dem-
onstrated a small metal cylinder that could
“create and send radio waves” but contained
“no vacuum, grid, plate, or glass envelope
to keep the air away”. It could amplify and The point contacts of the first transistor were created by a razorblade slit in gold foil wrapped around the
oscillate and had been named a “Transistor”. edge of a triangular plastic wedge. (Bell Labs)
The name had been chosen by internal
ballot among Bell Labs executives and germanium transistor, which he called a Origins
research staff. Semiconductor Triode and Transitron, while working for CFS Westing- The roots of the invention were much
Surface States Triode were considered fairly house near Paris. By mid-1949 many of older. The rectifying properties of crystals
good but unwieldy, and Transistor came out them were in use as amplifiers in the French had been discovered by Karl Braun in
well ahead of Crystal Triode, Solid Triode and telephone system. At this time European 1874, before wireless existed, and the
Iotatron. Little did the apathetic NYT reporter industry was still recovering from the devas- cat’s whisker detectors that became popu-
realise that he had witnessed the first public tation of war but research at UK companies lar in the early 1900s were semiconductor
demonstration of an invention that would such as BTH, GEC and STC was not far diodes in all but the name. Nor was the
spawn a world-changing technology. behind the US and their first products were concept of a three-electrode solid-state
Quite independently, Herbert Mataré in named Crystal Valve and Germanium Triode amplifier a new one. As far back as 1926,
June 1948 also invented the point-contact as well as Crystal Triode. the German-American engineer Julius

24 Practical Wireless October 2018

 physicist who had been engaged on anti-
submarine warfare research during WW2.
When he learned of Ohl’s discovery, Shock-
ley immediately postulated that it might be
possible to make a solid-state amplifier by
applying an electric field across a p-n junc-
tion but, initially, all attempts to vary the con-
ductivity with the control field failed. Without
any tools to see what was happening inside
the crystals at the subatomic level, progress
was dependent on intuition and trial-and-
error. By January 1946 the group admitted
that they were “groping in the dark”.
However, the failures led Bardeen to
postulate a theory of surface states in semi-
To the chagrin of Walter Brattain (right) and John conductors and to continue experimenta-
Bardeen (glasses), Bill Shockley insisted on sitting tion. By late 1947, Bardeen and Brattain had
at their workplace for this posed Bell Labs publicity switched from silicon to n-type germanium
Research Director Ralph Bown introduces the first photo. (Bell Labs) and felt that they were getting close to suc-
point-contact transistor to the press. (Bell Labs) cess. Then Brattain tried a configuration in
old pipe-cleaner knife and these detectors which a pair of very closely spaced point
Lilienfeld, who patented a forerunner to famously outperformed those that were not contacts were created by using a razor blade
the modern electrolytic capacitor, filed optimised in this way. He refused to allow to cut a minute gap in the gold foil wrapped
a patent for an FET-like device that was anyone to clean the dirty knife, which had around the edges of a small triangle of plas-
granted in 1930. Since materials of the pu- acquired “exactly the right momentum for tic. The contacts were pressed into the sur-
rity required weren’t available at the time, the job”! face of the germanium by a spring fashioned
it is unlikely that Lilienfeld succeeded in This war effort also gave rise to the next from a paper clip. By December 16th the
making a working transistor but his claim important breakthrough. While working on team had achieved significant power gain
was strong enough to prevent Bell Labs radar detectors in 1940, Bell Labs chemist and on December 23rd they demonstrated
from patenting the field-effect approach Russell Ohl made the fortuitous discovery speech amplification to Bell Labs manage-
18 years later. of a rectifying junction at a defect in a bar of ment. An “entirely new thing in the world”
The theoretical foundations for tran- silicon and noted its photovoltaic behaviour. had been created.
sistor operation were laid in 1931, when Ohl had been bitten by the radio bug when
Cambridge University mathematician Alan using spark transmitters on 150m during Patenting
Wilson formulated the quantum mechanical WW1 and had built a superhet as early as Everyone at Bell Labs was familiar with the
theory of conduction in semiconductors. He 1921. He coined the terms n-type, for mate- legend that their company originated be-
correctly attributed their properties to the rial containing some atoms of phosphorus, cause Alexander Bell had beaten Elisha
presence of impurity atoms in the crystals, antimony or arsenic (in which conduction Gray in a race to the patent office. So, the
opening the way for less empirical work on is by electrons), and p-type, for material transistor invention was classified as Bell
solid-state devices. During WW2, Wilson containing some atoms of boron, aluminium Labs Confidential until it was better un-
worked on radio communications for the or gallium (in which conduction is by holes), derstood and patent protection had been
secret Special Operations Executive (the and the p-n junction was born. Initially this applied for. But in whose name? Shockley
‘Ministry of Ungentlemanly Warfare’) and discovery wasn’t disclosed outside Bell was Bardeen and Brattain’s supervisor,
later on the UK project to develop the atomic Labs, and Ohl was instructed to cut any and was disgruntled that his subordinates
bomb. He was knighted in 1961. chance p-n junctions out of silicon that was had made the breakthrough without his
With the development of the high-power sent to his British counterparts. active participation. Bell Labs’ lawyers
10cm cavity magnetron by John Randall advised that Shockley’s own work was
and Harry Boot at Birmingham University, Bell Labs overshadowed by that of Lilienfeld, and
high resolution microwave radar became Work on semiconductors at Bell Labs had that Bardeen and Brattain were the actual
feasible if a reliable detector could be found. begun before the war, and energetic boss inventors. Consequently, they refused to
After Bell Labs were unsuccessful in devel- Mervin Kelly initiated a new unified solid- put his name on the patent for the point-
oping thermionic valves for this very short state programme in June 1945, as soon contact transistor.
wavelength, attention was turned once again as the men who had been assigned to Professional jealousy and bruised ego
to cat’s whisker crystal diodes. In Britain, the military work began to return to the lab’s spurred Shockley into a frenzy of independ-
Telecommunications Research Establish- new ‘Idea Factory’ at Murray Hill. The aim ent work, which he initially kept secret from
ment (TRE) developed an aluminium-doped was specifically to devise an alternative to the rest of the group, thus starting to alienate
silicon cartridge that was manufactured by thermionic valve telephone amplifiers and them. By the end of January 1948 he had
BTH and GEC. While at TRE, the eccentric the initial funding of $417,000 was billed come up with a theoretical transistor design
Bristol physicist Herbert Skinner improvised to AT&T. that worked quite differently from that of
a technique for finding a sweet spot for As manager of the programme, Kelly ap- Bardeen and Brattain, being composed of a
the contact by tapping the crystals with an pointed William Shockley, a London-born sandwich of p-type germanium between two

October 2018 Practical Wireless 25

 The Transistor Revolution

n-type regions. After it was discovered that

minority carriers could indeed traverse the
bulk semiconductor, the concept appeared
feasible and he successfully patented this
bipolar ‘junction transistor’ in his own name.
It would eventually supersede the point-
contact type but at the time no-one knew
how such a device could be fabricated and it
was dubbed a ‘persistor’ because it seemed
that much persistence would be required to
make it.
Meanwhile, with patent protection
secured and after Naval Research Labora-
tory staff had withdrawn a claim that they
had already made the same thing, Bell Labs William Pfann (left) with an early zone refining Shockley describes the concept of the npn junction
announced the invention of the point-contact furnace. (Bell Labs) transistor. (Bell Labs)
transistor at a press conference on June
30th 1948. Research Director Ralph Bown applications, including CV versions for to displace a small molten region slowly
illustrated the structure of the new device military use. Mullard’s Reference Manual of along its length. This causes the impurities
with a giant cutaway model. The attendees Transistor Circuits provided a useful starting to be swept to the end of the crystal, which
were given headphones to hear the device guide for many amateur designers but high is eventually cut off. For silicon crystals, a
amplify and oscillate and to listen to a broad- frequency transistors were still rare and the variation of this technique called float zone-
cast received on a radio set that used tran- only transmitter described was a 5-transistor refining was devised in which contamina-
sistors instead of valves. After the unveiling, unit for the emergency frequency of 500kHz. tion is avoided by maintaining the molten
the public showed only lukewarm interest portion in place by surface tension, without
but Bell Labs was besieged by requests for Crystals a containment vessel. A similar process was
sample devices from the electronics industry While the first point-contact transistors also developed independently by Siemens in
and the armed forces. By the summer of used polycrystalline germanium, the key West Germany.
1949 the lab had fabricated 4000 working to the manufacture of junction transistors To produce the npn or pnp sandwich
germanium transistors. is the production of extremely pure single of Shockley’s junction transistor, Morgan
Also in that year, the US Justice Depart- crystals that can be doped very precisely Sparks invented a double-doping method of
ment filed a new antitrust suit against the with p-type and n-type impurities. Chance changing the relative impurity concentrations
Bell System. In view of this, as soon as the often plays a vital role in discovery. In while the crystal was being drawn from the
military agreed in late 1951 that transistor 1915, the Polish chemist Jan Czochralski melt, producing the first grown-junction ger-
technology need not be classified, AT&T absent-mindedly dipped his pen into a manium transistor in April 1950. After further
made manufacturing licences available crucible of molten tin instead of his ink- refinement, Bell Labs announced this inven-
without restriction to interested companies in well. When he lifted the pen, he found that tion in July 1951. Meanwhile, the trio that
NATO countries and Japan for the relatively it drew out a filament of tin that proved to had made the initial invention had broken up.
modest fee of $25,000 (about $260,000 in be a single crystal. Frustrated by Shockley’s management style,
today’s money). Licensees, government Bell Labs chemical engineer Gordon Brattain refused to work with him again,
labs and university researchers were given Teal advocated developing this technique while Bardeen left Bell Labs altogether. This
sample transistors. They were also invited for growing large single crystals by draw- was a foretaste of the problems that would
to a technology symposium and a two-day ing a seed slowly from a crucible of molten later lead to the downfall of Shockley’s busi-
transistor manufacturing plant tour at West- germanium but he received no support ness enterprise.
ern Electric that was attended by 100 rep- from Shockley or management for this
resentatives from 40 companies, including work. Working evenings and weekends, he Silicon
BTH, Ericsson, GEC, Philips, Siemens and pursued the idea unofficially, shutting his ap- For transistor manufacture, silicon has
Telefunken from Europe. When published in paratus away in storage during the day since inherent advantages over germanium
book form, a revised edition of the symposi- there was no room for it in the metallurgy because of its greater abundance, lower
um proceedings called Transistor Technology lab. By late 1949 he was able to make very leakage current and higher operating tem-
was soon dubbed Ma Bell’s Cookbook. In pure single germanium crystals and the huge perature range. But the material is more
1955 AT&T relinquished its original transis- improvement compared with polycrystal- difficult to purify and process and, owing
tor patents to stave off forced divestiture, line material became evident to all. For the to the lower charge carrier mobility, a sili-
although Western Electric hung on to those production of wafers for modern VLSI micro- con transistor must have a much narrower
covering key manufacturing processes. circuits, the Czochralski/Teal process is still base region for high frequency operation.
In the UK, Mullard started production the basis of the principal method of growing It wasn’t until 1961 that, with gold dop-
of the point-contact OC50 and OC51 in silicon boules, which today may be up to 2m ing and epitaxy, a silicon transistor (the
1952, before launching a range of junction long and 450mm in diameter. 2N709) exceeded germanium speed.
transistors the following year. Subsequently, At the same time, William Pfann invented During WW2 great strides had been made
over ten British manufacturers produced the zone-refining method of purifying a in the art of purifying the material, and by
hundreds of transistor types for different crystal ingot by using induction heating coils 1945 silicon of 99.999% purity was available.

26 Practical Wireless October 2018

 were initially reluctant to change over from
thermionic valves, which were cheaper
than transistors at the time. So, in October
1954 Texas Instruments and Industrial
Development Engineering Associates (IDEA)
in Indianapolis teamed up to launch the
famous $49.95 Regency TR-1. This pocket-
sized MW AM set used four npn germanium
junction transistors in a superhet circuit
with one local oscillator/mixer, two 262kHz
IF stages and one audio amplifier. (The low
Gordon Teal (left) and Morgan Sparks developed the IF was chosen to increase the gain). Before
crystal growing and doping technologies that made teenagers adopted it as a personal music
the junction transistor possible. (Bell Labs) player, the radio was marketed as a ‘security
item’ since the USSR was now an atomic
power. In spite of its somewhat mediocre
performance, the radio was a success and
almost 100,000 sets were sold in the first 12
months. But having underestimated the cost
of making the radio, TI hardly broke even
on the venture and after that the company
concentrated on making semiconductors for
In 1954 the 4-transistor Regency TR-1 was the other manufacturers to incorporate in their
world’s first commercial transistor radio. Over products.
100,000 sets were made. (Joe Haupt) One of the enterprises that emulated the
TR-1 was a small Japanese company called
TI then dominated the market for silicon Tokyo Tsushin Kogyo (Totsuko). With little
transistors for several years. The little more information than a Bell Lab’s licence
company was suddenly in the big league. and Ma Bell’s Cookbook, they started
In the UK, the TRE electronics engineer manufacturing point-contact and then
Geoffrey Dummer, a specialist in radar junction transistors, obtaining a reasonable
synthetic trainers, presented a fundamental yield of about 5%. After a false start with
paper in 1952 describing the concept of the a portable radio whose plastic case was
The first germanium junction transistor was integrated circuit. But British semiconductor found to come apart in the summer heat,
fabricated by Morgan Sparks in 1950. (Bell Labs) manufacturers were unwilling to take the they launched their TR-55 model in August
risk of investing in the idea and it would be 1955. This 5-transistor set was Japan’s first
Further improvements by DuPont allowed six years before radio engineer Jack Kilby transistor radio and it spearheaded the entire
the first grown-junction silicon transistor to (W9GTY) of TI was awarded a patent for Japanese consumer electronics industry.
be fabricated at Bell Labs in January 1954 the first crude IC. He demonstrated that For the product launch, Totsuko’s owners
but in spite of this success the development transistors, diodes, resistors and capacitors changed their tongue-twisting brand name
wasn’t followed up commercially by Western could be fabricated in a single chip, although to one that could more easily be pronounced
Electric, the manufacturing side of AT&T. his hybrid design required them to be in the West. They called it Sony. The name
Meanwhile Teal had left Bell Labs to join interconnected by flying bonding wires. Bell became well known outside Japan after
a fledgling start-up company that would Labs researchers also missed the integrated discerning thieves who broke into a New
become called Texas Instruments. After circuit and Western Electric had to enter York warehouse ignored other makes to steal
telling Teal repeatedly “This business is cross-licensing agreements to make them. only the 4000 Sony radios stored there.
not for you. We don’t think you can do it”, In 1953 Philco developed electrolytically
Bell Labs finally agreed to accord a licence Consumer Radio machined surface-barrier germanium
to TI in 1952. But instead of making only In addition to hearing aids, it seemed transistors that could operate up to 60MHz
germanium transistors Teal established a natural that portable radio receivers would and manufacturing licences were granted
group to work on grown-junction silicon be among the first commercially made to other companies including Sprague
transistors and 150 good ones had been consumer products to employ transistors. and a division of Plessey. In 1955 these
made by the time of the IRE National Demonstration sets were shown at the transistors were used in the first commercial
Conference in May 1954. At the conference, 1952 Components Exhibition in Paris transistor car radio and in 1958 improved
speaker after speaker reported that the and at the 1953 Dusseldorf Radio Fair, versions were orbited in the 108MHz beacon
development of the silicon transistor would where the Intermetall Corporation that transmitter carried by Explorer 1, the first US
take several years so that Teal created a had been founded by Mataré on his return satellite.
sensation when he revealed that he had to Germany demonstrated a prototype Next month I will conclude by describing
a handful of them in his pocket and that receiver that used four hand-assembled the first uses of transistors in amateur radio
commercial production at TI was beginning point-contact transistors. and more recent developments in transistor
immediately. With no effective competition, But the established radio manufacturers technology.

October 2018 Practical Wireless 27

 Feature by Dr Bruce Taylor HB9ANY
l E-mail: [email protected]

W The Transistor
ireless World first
reported the inven-
tion of the transistor
in October 1948.

Revolution (Part 2)
Using little more than
a microscope, an Avometer and a pulse
source for point-contact forming, enter-
prising amateurs were soon making their
own transistors by replacing the single
cat’s whisker of selected germanium
diodes by two. By 1950 enthusiasts in the
UK were building simple receivers using Dr Bruce Taylor HB9ANY concludes his look at the history
commercially-available Raytheon point-
contact transistors that were primarily
of the transistor, relating how the invention impacted
made for hearing aids. amateur radio and moving on to modern developments
Radio amateurs also started ex-
perimenting with transmitters as soon as in integrated circuitry and microprocessors.
suitable transistors became available at
accessible prices. The February 1953 is-
sue of QST described how RCA manager
K2AH even made a 146MHz CW QSO with
W2UK over 25 miles with a power input of
24mW to a single selected experimental
point-contact germanium T165/6 transistor.
In the UK, a miniature 3.5MHz transistor
transmitter by G5CV aroused great inter-
est when operated at the 1953 Amateur
Radio Exhibition. A simple receiver using
two GEC point-contact transistors was
described in the January 1954 issue of
Wireless World, while a topband (160m)
transmitter by G3IEE using a Mullard OC50
featured in the RSGB Bulletin for March.
At this time manufacturers such as Philips
made ‘experimental transistors’ available
to amateurs at low prices. These were
devices that failed to meet the full profes-
sional specifications and would otherwise
have been scrap.
In mid-1956 IDEA launched the
Regency ATC-1, a simple two-transistor
mobile converter that tuned the five HF
amateur bands. One germanium npn tran-
sistor acted as oscillator/mixer, with output
at 1230kHz for a broadcast receiver, and The 1956 Regency ATC-1 HF band converter was one of the first commercial transistor products for radio
a second pnp one as Q-multiplier/BFO. It amateurs. (W8ZR)
was sold for $79.50.
In September of that year W1OGU, a 10-transistor kit set used three Mullard IT-2232 was phased out in 1990.
technician with Raytheon, achieved the germanium AF115s as RF amplifier, local It took several years for many of the tra-
first transatlantic QSO with a transistor oscillator and mixer, with four OC45s for ditional shortwave receiver manufacturers
transmitter, working OZ7BO in Copenha- the BFO and three IF amplifiers coupled by to change from valve to transistor designs
gen and G3AAM in Birmingham on 14MHz. 455kHz piezoelectric ‘transfilters’. After re- and not all did so successfully. Eddystone’s
His crystal-controlled rig used one 2N113 placing the audio output stage by a higher first solid-state communications receiver
germanium alloy junction transistor as power amplifier, I used the compact set was the S960; virtually a model S940
7MHz oscillator, driving a second as dou- as a mobile receiver for several years but with 12 transistors in place of valves. Its
bler/output stage with 78mW input. its performance didn’t match that of good performance was inferior to its parent and
One of the first shortwave transis- valve receivers of that period. Starting with it was dropped two years later. The more
tor receivers for radio amateurs to be the Model IM-30 in 1961, Heathkit also compact EC10 that was introduced in 1963
made in the UK was the Heathkit GC-1U produced a series of rudimentary transistor was more successful and over 6,000 were
Mohican, which was sold in 1961 for £38 testers and curve tracers that evolved with made, followed by around 10,000 of the
15/- (around £800 in today’s money). The the device technology until the final Model Mk2 version with an S-meter and fine-

62 Practical Wireless November 2018

Transistors.indd 62 20/09/2018 09:24

 tuning control. In 1971 you could choose
between the valved Model 830 receiver
and the transistorised 1830 version with
slightly better performance. The 830 was
Eddystone’s last valve receiver and contin-
ued in production until January 1973.
National HRO receivers had been a
long-time favourite with radio amateurs,
and around 10,000 sets were used by the
Y-Station service that supplied intercepted
enemy wireless messages to Bletch-
ley Park during WW2. In 1964 National The UK version of the 10-transistor Heathkit
introduced a transistor version of the Mohican receiver was launched in 1961. (GM3NZI)
receiver that retained the classic ‘PW’ gear
drive but no longer had plug-in coils. This
transitional set had 37 germanium transis-
tors in sockets connected by conventional W1OGU achieved the first transatlantic QSO with a
point-to-point wiring, just like its valve 78mW transistor transmitter in September 1952.
predecessors.

Diffusion
The first successful method of making
a junction transistor involved creating
the base layer by dropping a tiny p-type
pellet into the n-type melt while drawing
the crystal, and then converting it back to The IM-30 was the first of a series of simple
n-type. With this double-doping process, transistor testers that could measure DC gain and
and also the improved rate-grown variant, leakage current. (Heathkit)
it was difficult to accurately produce and
connect to the very narrow base region W1OGU’s historic 14MHz transatlantic transmitter
required for high frequency performance. used two germanium 2N113 transistors. (Raytheon)
Problems also arose with the alternative
alloy junction process developed by GE the printing industry had already been
and RCA, since precise control of the used for the production of printed circuit
alloying temperature and the thickness of boards, using a photosensitive resist that
the base layer was difficult. In a batch of was exposed through an optical mask.
100 transistors, the gain could vary from The technology was readily adapted to
20 to 50dB. chemically etch precisely-dimensioned
Once again, chance played an essential windows in the oxide layer covering the
role in the development of a hugely impor- silicon wafer, through which the n and
tant new technology. While doping by gas p-type impurities could be diffused to
diffusion had been used to introduce the make a double-diffused transistor and this Early transistor receiver technology. The glass-
donor impurities into germanium crystals, important invention was announced by Bell encapsulated Mullard germanium transistors
attempts to use the process at the very Labs in June 1955. With the Cold War in are mounted on solder sockets like little valves.
high temperatures required for processing full swing, ample military funds were avail- (GM3NZI)
silicon were initially unsuccessful because able to develop these new transistors that
of damage to the wafer surface. But could operate at high temperature and high that he had no right to know.
while Bell Labs chemist Carl Frosch was frequency. All three men attended a celebration
experimenting with diffusion, the hydrogen dinner in New York but the Bardeen and
carrying the dopant impurities accidentally Nobel Brattain families travelled to Stockholm
caught fire, causing water to be produced It had been expected for several years separately from the Shockleys and only
in his diffusion chamber. Frosch discovered that the invention of the transistor merited shared the formal ceremonies there. For all
that the fine green silicon dioxide layer that a Nobel Prize, so it was no real surprise three Laureates, it was the pinnacle of their
this formed on the surface of the wafer when Bardeen and Brattain were chosen careers. For Shockley, it was the prelude to
sealed it and protected it from damage. to receive the 1956 award for Physics. It his demise.
Unlike germanium oxides, silica is strong, was less certain that Shockley would also
inert and an excellent insulator. Initially share the coveted prize but in the end his Silicon Valley
considered a problem, the oxide turned out name was included although the vote was In view of his abrasive management style,
to be a key element of reliable solid-state not unanimous. Shockley tried to find out Shockley was repeatedly passed over for
electronics. from the Swedish Royal Academy of Sci- promotion at Bell Labs. So, during 1955 he
Photolithography techniques from ences who had opposed him but was told left to launch his own enterprise and, with

November 2018 Practical Wireless 63

Transistors.indd 63 20/09/2018 09:24

 The Transistor Revolution

National launched the solid-state HRO 500 in 1964. It was built like a traditional valve receiver but used 37 Robert Noyce (right) addresses the other seven
germanium transistors and 20 diodes. (Merate Chronometer Center) members of the ‘traitorous eight’. (Fairchild)

the financial backing of Beckman Indus- pany had fallen apart. Shockley did almost
tries, Shockley Semiconductor Laboratory no further work of scientific value and went
started up in April 1956. The facility was on to pursue a racist dysgenics agenda,
located in a new industrial park that was lecturing about the inferiority of African-
being set up in the Santa Clara Valley on Americans and advocating the sterilisation
land owned by Stanford University. It was of people with an IQ under 100. He died
the start of silicon in Silicon Valley, the in 1989 without fortune but knowing that
The 33-transistor Eddystone 1830 had a JFET/ extraordinary zone of innovation and tech- he had been the intellectual driving force
MOSFET cascode RF amplifier and dual-gate nology that now accounts for one-third of behind a technological revolution and the
MOSFET first mixer. (Eddystone UG) all the venture capital investment in the catalyst of an industry that turned a quiet
US. Shockley planned that his company Californian valley into the most concen-
would “set the world on fire”. Indeed, it trated source of wealth on the planet.
did, but not at all as he envisaged.
None of Shockley’s Bell Labs col- Planar Technology
leagues wanted to join him. But because The ‘traitorous eight’ that defected en
of his reputation as a scientific genius, he masse from Shockley Lab went on to
had no difficulty recruiting very talented found Fairchild Semiconductor with a
staff, including the chemist Gordon Moore $1.38 million loan from Sherman Fairchild,
(of Moore’s Law), Swiss physicist Jean a New York playboy-inventor who was the
Hoerni and Robert Noyce, who was largest shareholder in IBM. Each of the
employed at that time by the radio and eight had to make an initial investment of
TV manufacturer Philco but was seeking $500 to purchase 100 of the company’s
research-oriented work. In the UK, Ferranti shares, a sum that Noyce had to borrow
began producing bipolar silicon diffused from his grandmother since neither he nor
This 432MHz converter was typical of the transistors in 1954. Shockley could also his parents had any savings. He had no
homebrew UHF art in 1965. It uses three TI have directed his new team to ramp up difficulty repaying her with interest, for less
germanium pnp transistors in the UHF stages and production of silicon transistors to profit than two years later his stock was worth
three Mullard OC171s for the frequency multiplier from a rapidly expanding US market. But $300,000 and his personal fortune was
and 28MHz IF amplifier. (GM3NZI) instead, he insisted that they devote their eventually to grow to over $3 billion.
efforts to developing a special four-layer The timing was fortuitous. Alarmed by
pnpn diode that was difficult to fabricate the surprise launch of Sputnik 1 in October
reliably and of limited application. As a 1957, the US Military suddenly had major
result, Shockley Semiconductor Lab never requirements for lightweight miniature
turned a profit. solid-state components for its catch-up
Shockley proved an even worse man- programme and total annual purchases
ager in his own company than at Bell Labs. soon exceeded $100 million. Fairchild’s
He constantly feuded with subordinates, first contract was to manufacture silicon
insulted his staff and refused to listen to mesa transistors for IBM’s navigational
their advice. He became paranoid about computer for the XB-70 Valkyrie bomber.
trivial incidents, recorded all phone calls, To make arrays of transistors, they cobbled
and at one point even ordered all his em- together a step-and-repeat camera using
ployees to take a lie-detector test. Genius ordinary 16mm movie camera lenses. The
Noyce made the first monolithic silicon IC in 1961. without ethics wasn’t a winning formula yield was reasonable but reliability proved
This basic flip-flop had four transistors and two and by September 1958, after a mutiny of an issue and technicians had to ‘tap test’
resistors. (Fairchild) eight dissidents lead by Noyce, the com- the transistors with pencil erasers to check

64 Practical Wireless November 2018

Transistors.indd 64 20/09/2018 09:24

 The 1976 Ferranti F100L was the first 16-bit microprocessor designed and manufactured in Inexpensive SDR dongles spawned a new generation of
Europe. It used an enhanced rad-hard bipolar technology with 3.5 micron features. (Ferranti) communications receivers. (HB9ANY)

whether tiny loose particles trapped in the integrated electronics. In the UK, Plessey because of its high power consumption.
hermetic package could contaminate the was involved in integrated circuit develop- Two years later Intel introduced the com-
exposed p-n junctions. ment at an early date but the first ICs to be mercial 16-bit 3-micron 8086 microproces-
To solve this problem, Hoerni built on produced commercially in Europe were the sor and the x86 family architecture rapidly
the earlier work of Frosch to invent the Micronor devices introduced by Ferranti became ubiquitous in desktop PCs and
ground-breaking planar technology that from 1962. These ICs had a wired OR func- laptops. In due course this led to the next
protected the sensitive junctions with oxide tion and were used mainly in computers iteration in wireless technology. Although
and allowed all the contacts to be made on developed by Marconi for the Admiralty. the concept was much older, in 1995 Ste-
one side of the silicon wafer. This passiva- The second generation of this family used phen Blust coined the term Software De-
tion process reduced leakage currents and an enhanced DTL technology that was fined Radio and soon SDR hardware and
the configuration automatically created faster but less dense than TI 74 series TTL, software allowed any PC to be turned into
graded bases that reduce the charge car- which emerged as the most popular logic a high-performance transceiver capable of
rier transit time, as in drift transistors. It configuration after it appeared in 1966. handling a wide range of communications
also supported the aluminium-over-oxide Micronor II ICs powered the Argus 400 modes.
interconnection scheme that was used by computers that were used in many indus- Today there are 19 silicon
Noyce to make the first commercial mono- trial control applications, such as nuclear manufacturing facilities in the UK. A
lithic silicon integrated circuits in 1960, power stations and the Jodrell Bank radio single modern FPGA can have as many
after the problem of aligning successive telescope. In 1966, Plessey started the as 50 billion microscopic transistors and
photolithographic masks had been solved. first European production of MOS ICs at (including memory) a high-end smartphone
Patent litigation between Fairchild and TI Swindon. contains a total of around one trillion,
over the invention of the IC lasted many or ten times more than the number of
years but out of court the two companies Evolution neurons in the human brain. To visualise
entered a cross-licensing agreement with a In 1968 Moore and Noyce left Fairchild to the total number of transistors that have
net payment to Fairchild. found Integrated Electronics Corp, a name been manufactured, we can take as a
The individual components in the they abridged to the snappier Intel. In- metric the number of cells in the human
production versions of these ICs were spired by HP, they practised a non-hierar- body – about 100 trillion. Then a Fermi
isolated by reverse-biased p-n junctions, a chical management style that encouraged estimate of the total number of transistors
technique that had been patented by Kurt the exchange of ideas and innovation and made since 1947 exceeds the number
Lehovec while at Sprague Electric, who in 1971 their four-bit 10-micron 4004 CPU of human cells in the entire population of
paid him only $1 for the rights to this key chip with 2300 MOS transistors launched the UK and within a few years the number
invention. (A common practice at the time). the microcomputer age. At that time of transistors produced may equal the
Until it was overtaken by TI in 1967, Fair- few people envisaged the wide-ranging number of human cells on the planet. Even
child became the undisputed leader of the consequence of the synergy of micropro- visionaries like Bill Hewlett and Dave
semiconductor industry and in due course, cessors and wireless, and in the same Packard would have been astounded.
spin-off ‘Fairchildren’ spawned the creation year a marketing study commissioned by In the years immediately following its
of dozens of successful companies in Sili- AT&T reported that “there is no market for invention, the transistor was conceived
con Valley, such as AMD, Intersil, National mobile phones at any price”! as a replacement for the thermionic valve.
Semiconductor, Signetics and Teledyne. The first European 16-bit microproces- Indeed, it was, and a very good one but
The silicon dioxide layer also proved sor was the Ferranti 3.5-micron F100L, it also triggered revolutionary growth in
the key to neutralising the troublesome introduced in 1976. This mil-spec rad-hard communications technology and created
surface states that had thwarted attempts chip, designed in Bracknell and fabricated a world that has become interconnected
to make an insulated-gate field-effect in Manchester, had about 7000 compo- as never before. Apart from wireless itself,
transistor. John Atalla finally succeeded nents and 2m of aluminium track intercon- perhaps no other modern invention has
in making a working FET at Bell Labs in nects. It was orbited in OSCAR-9, although had a greater influence on almost every
1959, paving the way for modern MOSFET it wasn’t frequently enabled in that satellite aspect of our daily lives.

November 2018 Practical Wireless 65

Transistors.indd 65 20/09/2018 09:24