The Irresistible Transistor
Fifty years ago this month, a man embraced his inner hobbyist and gave thousands of engineers their first transistor
Is it possible to love a transistor? Certainly what Jack Ward feels for the Raytheon CK722, the first transistor sold to the general public, goes beyond casual affection. He's collected thousands of early transistor specimens, including dozens of CK722s. His stately yellow Victorian home on a quiet, tree-lined street in Brookline, Mass., has a basement crammed with enough code oscillators, Geiger counters, radios, hand-wrought circuit boards, transistorized hearing aids, subminiature vacuum tubes, diodes, resistors, and capacitors to make any collector of vintage electronic gear drool. He's written one book about the CK722 and has started another about early transistor history at RCA. When he's not working as associate director of quality for the Bedford, Mass., facility of gene-chip maker Affymetrix Inc., he's busy maintaining his virtual Transistor Museum on the Web and is widely acknowledged by fellow collectors as a techno-anthropologist par excellence.
“My wife's very supportive, and my younger two children think it's fairly amusing, and probably not a bad way to have a mid-life crisis," says Ward of his family's reaction to his passionate pursuit of transistor history. Far from thinking that his dad's a square, Ward's oldest son, Nick, who is pursuing a B.A. in physics, is learning a lot from his old man. “Nick can't believe how fast technology changes and that the people I talk to have changed the world," adds Ward, who as curator of the online museum has shifted his focus from collecting early transistors to collecting oral histories from the engineers who sparked the Semiconductor Era.
With this ad, Raytheon Co. introduced the CK722 and slightly better CK721 (less noise, higher gain) in 1953.Photo: Jack Ward
For Ward and the CK722, it was love at first sight. The year was 1959: Fidel Castro had just taken Cuba, John F. Kennedy was campaigning for U.S. president, Buddy Holly was flying around on what would be his last tour, and Texas Instruments and Fairchild Semiconductor had both filed patent applications for something called an integrated circuit. Recalling himself as a boy of 10 marching into his local radio distributor and plunking down his allowance for his first transistor, Ward [ see photo] taps into the same wonder that gripped him when he laid eyes on the CK722, which Raytheon Co. (Lexington, Mass.) made available to hobbyists through RadioShack stores starting in March 1953.
“They were probably only a couple bucks at the time, but just the excitement of actually owning one of these was intense. The package is quite spectacular, you know, the actual shape of the device and the color," he says. “The blue ones, for instance, the iridescent blue color is just gorgeous."
With his new transistor, Ward built a radio, just a simple tuned circuit with a germanium diode to detect a signal and a CK722 as an audio amplifier. “I turned it on in my room at night after lights out, and listened to rock and roll or a baseball game," he says wistfully. “For sheer excitement, I can't think of a parallel with another thing in technology. I'm tempted to say the PC, but that doesn't quite capture it. You see, it's different than that."
Love potion No. 722
Ward wasn't the only boy smitten. Tens of thousands of CK722s were sold between 1953 and the mid-1960s. The irresistible transistor cast a spell over even die-hard vacuum tube enthusiasts like Terry Hosking. By the ripe old age of 12, Hosking, now a senior application and design engineer with SB Electronics Inc. (Barre, Vt.), had concluded that vacuum tubes were the only way to go.
“I told some of my relatives that I didn't think that transistors were going to amount to much," Hosking told me. “A few weeks later, I got a care package from them with a blue CK722 and a Sylvania 2N35 transistor and a couple books that showed how to hook them up. I was amazed to find that the transistor radio I built would pick up the local stations without an external antenna and ground like I had to use with the tube radio."
Transistors weren't just sensitive devices, they were the mysterious oracles of a new age—“Just a little solid block of black plastic with three thin wires sticking out," says Tom Lee, associate professor of electrical engineering at Stanford University. Lee started fooling around with transistors when he was only five. At that time, in the mid-1960s, RadioShack sold “blister packs" of five transistors for a dollar. “They were the only transistors a kid could easily obtain with saved-up pocket change," he says. “The CK722 is the first recollection I have of that transistor type, indeed, of any transistor type at all. The things seemed magical."
And messy in a way tinkerers love. Junior engineers constructing projects out of transistors and circuit boards had to hone basic shop skills: measuring, cutting, drilling, and assembly. “Of course, the most important skill to master was soldering," says Bob McGarrah, now staff system planning engineer at Central Illinois Light Co. (Peoria, Ill.).
What madeleines were to Proust, solder is to McGarrah. “The unique smell of the hot flux still brings back happy memories," he says, one of which is a of small audio amplifier that he discovered had an impedance high enough not to draw a dial tone when connected to a telephone line. A huge fan of the TV spy drama “The Man from U.N.C.L.E.," young McGarrah used the amplifier to practice his surveillance skills by listening in on family members' phone calls.
Connecting on the Internet
Like old high school chums who reunite on Classmates.com and realize that they shared a crush on the same girl way back when, Hosking, Lee, Ward, McGarrah, and dozens of others linked up on the online auction site eBay in the late 1990s and began swapping stories along with vintage transistors.
“Old transistor collectors tend to be a small, close-knit bunch," says McGarrah, who runs his own transistor history Web site. “The power of the Internet to bring together such a narrowly focused group of hobbyists is amazing."
Transistor Family Tree
(From left) Raytheon subminiature vacuum tubes such as the CK549DX dominated the market for hearing-aid tubes throughout the 1940s. In 1948, the CK703 point-contact transistor went into production, but proved too unstable to unseat tubes. The more robust CK718 alloy junction transistor debuted in late 1952 and rapidly superseded tubes in hearing aids. CK718s that didn't have the noise and gain characteristics for hearing aids were relabeled CK721 or CK722, both introduced in 1953. The black CK722 was replaced by the blue metal style in 1955. The silver metal case followed in the late 1950s and sold through 1964. Finally, smaller, better-performing transistors like the half-size CK784 were introduced throughout the late 1950s.Photo: Jack Ward
Ward concurs: “Without the Internet, none of this interaction would really be possible." Inspired by the online communities he saw sprouting up around vacuum tubes, Ward decided to use the Internet to research the history of early transistor radios. He soon became more interested in radio components than the radios themselves, “in how these little devices were developed, and what a profound impact they had on society."
In 1999, Ward scratched an itch to write and took as his subject his first transistor. He began working on The Story of the CK722, and put up the http://www.ck722.com Web site. Here he posted the fruits of his research—pictures of the CK722 and other early Raytheon transistors, charting the 722 through its three case colors, silver, black, and that iridescent blue [see “Transistor Family Tree,"]. He also posted pictures of old ads, circuit schematics, packages, and devices people made with the CK722 that they sent him for his growing collection.
Deep into his yearlong project, Ward attended the Cambridge-based Massachusetts Institute of Technology's (MIT's) monthly hamfest, a flea market for the “geek to the max," as he puts it. He made known to fellow collectors his eagerness to find out more about the origins of his favorite transistor. “Someone mentioned that they thought Raytheon had a historian," Ward remembers. “So I called up Raytheon, and sure enough, there was one. A gentleman named Norman Krim."
It turns out that 89-year-old Norm Krim is not only Raytheon's archivist, he's a living link to the roots of the electronics industry. He's also the father of the CK722.
Late one chilly night this past October, as he and I sat in the kitchen of his Newton, Mass., home sipping green tea and munching on roasted almonds, Krim spun his story. Having been a student of Raytheon founder Vannevar Bush at MIT, Krim took a job with his mentor's company as an engineer in the receiving tube division in 1935. By 1938, Krim had developed subminiature tubes for hearing aids.
The expertise gained in that work earned six patents on the subminiature tubes found in the proximity fuses used in U.S. Army artillery and antiaircraft shells, credited by some historians with turning around the Battle of the Bulge in late 1944. After the war, under Krim's steady hand, Raytheon's receiving tube division dominated the market for hearing-aid tubes with a 90 percent share through the 1940s.
As the industrial war machine was winding down, the semiconductor revolution was just revving up. At Raytheon's archives in Lexington, the morning after our late night bull session, Krim showed me a letter. Dated 9 July 1948, it was addressed to Laurence K. Marshall, then president of Raytheon, inviting him to Bell Telephone Laboratories Inc. (Murray Hill, N.J.) to see a demonstration of “a new device called a Transistor," specifically the point-contact transistor invented by Walter Brattain, John Bardeen, and William Shockley. Marshall was busy that day and tapped Krim to go in his place. What he saw shook him to his very core.
“I was worried that my success had been with tubes, and this was threatening my job," Krim recalls. “So what the hell was I going to do? I was going to get into transistors."
Krim's crash program eventually led to the introduction of the world's first commercially available transistor, the CK703 in 1948, less than six months after the Murray Hill demonstration.
But the CK703 had some problems. The germanium point-contact transistor—actually two pointed wires, 125 µm in diameter and 25-50 µm apart, in contact with the signal-amplifying semiconductor—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.
So Krim shifted gears and leveraged his division's growing expertise in semiconductor technology to make germanium diodes, which had a ready market as signal detectors in TV sets. By 1950, Raytheon was cranking out 20 000 diodes a day and Krim was promoted to vice president of the receiving tube division, where the diodes were being made.
Meanwhile, as germanium diodes and subminiature tubes poured out of Raytheon's plants, William Shockley was about to jolt the world again, this time with the junction transistor. Krim was fortunate enough to room with Shockley for over a week in the spring of 1951, while both were serving on a military procurement advisory board known as the Baker Committee.
“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." And Krim's engineers swung into action.
Their junction transistors were simple devices made of two indium dots (emitter and collector) alloyed to either side of a germanium chip. But the germanium wasn't pure enough and the initial devices failed. Later in 1951, at a symposium conducted by Bell Labs, Krim's team learned the value of zone refining: passing an RF coil over a quartz tube containing a large block of germanium crystal and melting portions of it in sequence. That got the impurities to migrate to the end of the ingot, which could then be lopped off, leaving a pure crystal behind.
Knowing that quartz tubes were key to making germanium pure enough for junction transistors, the crafty Krim cornered the market on quartz tubing. “And I did one other thing," he says with a sly smile. “There was a company in Missouri called Eagle-Picher, at the time the country's biggest zinc refiner. They threw out germanium as a byproduct of zinc refining. So I bought it all up."
But as Raytheon prepared to introduce its germanium junction transistor, dubbed the CK718, yields stayed stubbornly low. Water vapor 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.
Still, the manufacturing process wasn't perfect, and Krim was stuck with a mound of noisy, low-gain CK718s that weren't good enough for hearing aids. Faced with the prospect of destroying thousands of rejects, Krim, who a decade later as CEO of RadioShack would sell the electronic hobby retailer to leather craft store chain Tandy Corp. (Dallas), wondered: could what was scrap to a company be gold to a hobbyist? As a youth in the late 1920s, he had built a mechanical TV set. It included a radio receiver and a Bakelite disk drilled with 16 strategically placed holes to scan a Raytheon Kino neon lamp that projected the picture. Resourceful even then, he used his mother's milkshake mixer to rotate the disk and obtain an image.
“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," recalls Krim. “I loved to build experimental stuff and I just wanted the kids to have these. And nobody had ever seen a transistor."
Not even editors of the major electronics publications at the time. So in February 1953, Krim invited the editors of all the major electronics magazines, including Electronics and the now defunct Radio and TV News, to his office for a demonstration of CK718 rejects that were relabeled CK722. “Their tongues were hanging out," recollects Krim.
From the pens of those amazed editors the word spread about what the ordinary hobbyist could do with a transistor. And kids across the United States started putting together radios and oscillators and speakerphones, a few of which are now enshrined in the Transistor Museum.
After visiting the Raytheon archives, Norm and I drove over to Jack's house to see the Transistor Museum collection [see photos, below]. It was the first visit for both of us, and I was curious to see how Norm would react to seeing bits of the history he helped create.
(Counterclockwise, from lower left) Transistor Museum curator Jack Ward studies a one-transistor regenerative receiver built by fellow fan Terry Hosking. Next, the top two shelves display transistor radios of the 1950s and 1960s produced by Emerson, RCA, and Roland, along with perhaps the world's only Germanium Crystal Triode Evaluation box (with red and green knobs) for testing point-contact transistors. The curator's home in Brookline, Mass., is also home to the museum. In Ward's hands is a clone made by the Bulova Watch Co. of the first commercial transistor radio; the original, the Regency TR1, was developed by Regency and Texas Instrumentsfor the 1954 Christmas season. Finally, on top of another shelf are two large 1950s transistor radios: a black Raytheon T2500, which was one of the few commercial products to actually use the CK722 transistor, and a cream-colored Superex kit radio, which sits on an Arvin transistor radio.Photos: Larry Volk
Our first stop was the kitchen table, where Jack presented each of us with a single-transistor CK722 radio kit he had made especially for this occasion. It featured a gleaming silver CK722, a Raytheon CK705 germanium diode, an ancient pair of magnetic headphones, a vintage comb-tuning capacitor, a variable inductor loopstick to fine-tune reception of a station, and, appropriately, two RadioShack AAA batteries.
The delight on Norm's face erased 80 years, and for a few minutes he was that same precocious boy who'd built his own crude TV. Norm hung on Jack's every word as he showed us how he'd converted a double CD case into the kit's rudimentary circuit board. I marveled at Jack's ingenuity, but was frankly more interested in the components, the smooth twisting action of the comb tuning capacitor, the bright red of the CK705 diode, and, of course, the beguiling silver CK722, my first transistor.
Then we followed Jack down to the basement to see where the virtual Transistor Museum makes its real home.
We passed his son's matte-black Alien computer setup and an Altair computer resplendent in all of its toggle-switched glory before entering the inner sanctum. Here in this meticulously arranged room, Jack had everything he needed to make the museum run, including his server, scanner, digital camera, and broadband connection. The room was lined with shelf after shelf of plastic containers, each packed with hundreds of diodes, ICs, transistors, and other devices that Jack had bought on eBay. Norm, it's safe to say, was dumbstruck. The father of the CK722 was standing in the delivery room of the Semiconductor Era.
For an hour, Jack dazzled us with objects and stories that put everything we saw in historical context, right down to the dozens of packages for different transistors and vacuum tubes he brought out. He discussed the nuances of different transistors, identified according to year and lot number, and how, precisely, they were stamped. He displayed handmade gadgets people had sent him for his collection, including a one-transistor radio on a wooden board made by Terry Hosking, which, with its beautifully hand-wound antenna coil, looks like something you might see in a SoHo art gallery [see photo].
Jack placed tiny chips of germanium in our hands, revealing the mystery at the heart of the junction transistor. And he told Norm what a marketing genius he was for maintaining the CK722's brand identity for so long. As Raytheon got better at making transistors, they got smaller. While Norm could have housed the devices in a smaller package, that would have changed the look and the form factor of a familiar friend to hobbyists. So Norm potted the smaller transistors in the same size package and extended the CK722's brand life.
And while it might sound as if the hobbyist version of the Stockholm syndrome had set in, as I stood there listening to the stories flowing back and forth between Jack and Norm, one archivist to another, the warm fuzzies came on, my own love for the CK722 blossoming right there in Jack Ward's basement.
To Probe Further
To visit Jack Ward's Transistor Museum on the Web, see the site at http://www.transistormuseum.com.
The original CK722 site is at http://www.ck722.com.
For Bob's Virtual Transistor Museum and History Web Site, go to http://users.arczip.com/rmcgarra1.