Apr 21, 2015

Japan’s First Transistor

I read the following blog post last week. I’ve met David Manners a few times during my marketing days and always looked forward to his visits. Quite often, his posts lead me to do a little more research, as is the case today. He wrote:

David Manners, 4/17/2015

The Man Who Made Japan’s First Transistor

Tadashi Sasaki went to America in 1947 to investigate ways of improving the vacuum tube technology used by his firm Kobe Kogyo.He visited Western Electric’s tube factories and had access to Bell Labs’ scientists one of whom was John Bardeen.Just before Christmas 1947 Bardeen hinted to Sasaki that they’d found something interesting.In 1951 Bardeen was more forthcoming, telling Sasaki about transistors and giving him some single-crystal germanium.In 1953, Kobe Kogyo became the first company in Japan to make a transistor.

I’m well aware of the post-war situation in Japan, the invention of the transistor at Bell Labs and the eventual licensing of the transistor technology. I’m also burdened with the Japanese stereotypes that tended to subdue individuals pursuing crazy ideas. So something about David’s post seemed to be missing.

Tadashi Sasaki was born in a small fishing port on Japan's western coast but grew up in Taiwan, then a Japanese colony. His father, a former samurai from the garrison at Hamada castle, was a teacher there. Sasaki studied electrical engineering at Kyoto University. During World War II, he was assigned to an aircraft maker called Kawanishi Machine Works, based in Kobe, where he did research on vacuum tubes for use in telephones, wireless, and radar.

Kawanishi separated its vacuum tube business as a new company called Kobe Kogyo. In post-war Japan, a top allied priority was the reconstruction of the telephone system, so many doors were opened. As you know, vacuum tubes were key components in telephone switching. In 1947, Sasaki was sent to study modern methods of tube production at Western Electric's factory in Allentown, Pennsylvania. When he didn't understand something, Sasaki was permitted access to the Bell Telephone Laboratories in nearby Murray Hill – none other than John Bardeen.

Timing is everything. Shortly before Christmas 1947, just as he was about to return to Japan, Sasaki had a meeting with Bardeen. Normally the most mild-mannered, soft-spoken of men, Bardeen seemed uncharacteristically excited. He told his Japanese visitor that he and his coworkers had discovered a most interesting new phenomenon but he was not at liberty to say what it was.

In September 1951, Sasaki happened to be in the U.S again, meeting with RCA on subminiature receiving tubes. Sasaki and a colleague, Arizumi Tetsuya, heard about the transistor symposium Bell Labs for its licensees. They took a train to Murray Hill and found Bardeen, who told them all about the transistor. Kobe Kogyo signed a licensing agreement with RCA, which had a cross-licensing arrangement with Western Electric. In early 1953, ahead of all other companies in Japan, Kobe Kogyo became the first company in Japan to bring the transistor to production. The company's first transistorized product was a car radio it supplied to Toyota, which in 1955 was just beginning to produce passenger cars in earnest.

I had never heard of Kobe Kogyo. It was destined to become one of those companies that disappeared. They failed to invest heavily in transistors, since they had a vacuum tube business to defend. Matsushita, linked to Toyota, pushed them out of the car radio market. Eventually, the banks stepped in and forced an acquisition by Fujitsu in 1963 – you may be familiar with Fujitsu-Ten, that’s what happened to Kobe Kogyo. Taking the blame for the company’s failure, Sasaki resigned from the company's board. His only reasonable option was to return to Kyoto University as a professor.

This is all well within the stereotype that I’ve been taught over the years. But what happened next wasn't.

Returning from the United States in December 1963, he stopped for a layover in Honolulu. He had a chance meeting with Saeki Akira, senior executive director of a consumer electrical appliance maker called Hayakawa Electric Industry. Hayakawa had purchased vacuum tubes from Kobe Kogyo in the past. Over a meal on the terrace of his hotel at Waikiki Beach, Saeki invited Sasaki to join Hayakawa as head of the company's newly formed industrial equipment division. It was an exceptionally brave decision for him to accept. It was more characteristic of American entrepreneurs than my Japanese stereotype. Instead of taking a comfortable window office, he chose to jump from the frying pan into the fire – from a bankrupt firm to one that was in bad financial shape. Hayakawa was struggling so badly that it was believed it was about to be acquired by Hitachi.

He couldn't resist the challenge. "Everybody wants to join an elite company," he said, "because they want to have it easy. But I'd rather try and rebuild a company that has collapsed." A former colleague once said of him, "He has tremendous drive, and he's extremely interested in getting new technology applied in new product areas… Dr. Sasaki loves to talk, he loves to give ideas, he loves to encourage people to get into new areas – he's a tremendous man." This is not the type of description I expected.

At Hayakawa, his energy soon won him the nickname "Doctor Rocket." "Once he takes off, he's unstoppable," said Wada Tomio, a researcher who worked under Sasaki at Hayakawa for many years.

Hayakawa himself, the founder, was also an energetic entrepreneur. In 1915, his firm manufactured a mechanical pencil, dubbed in English, for export purposes, the Every-Ready Sharp Pencil. To this day, mechanical pencils (which are commonplace in Japan) are commonly known by the generic name shya-pu penshiru.

Later, on the lookout for promising new products, he discovered an early crystal radio set that had been imported from the U.S. One mark of a true entrepreneur is a refusal to be put off by lack of knowledge. Despite an almost complete ignorance of the principles of radio – or of electricity for that matter – Hayakawa decided to make radio sets. This was in 1924, the year before radio broadcasts were scheduled to begin in Japan. Hayakawa Electrical Industries would be the first Japanese company to make radios, which they sold under the name Sharp, the old brand from mechanical pencil days.

This spirit and enthusiasm for new, yet often unrelated, ideas is shared by Sasaki. Years before joining Hayakawa, one of Kobe Kogyo's picture tube customers was a Los Angeles-based television assembler called Packard Bell. This firm had entered Chapter 11 owing the Japanese firm a lot of money for parts already shipped. Sasaki went to L.A. to see what he could salvage.

Instead of forcing them to sell assets, he suggested product ideas to generate sales. One of the things he suggested was a wireless remote-controlled garage door opener. It was easy to produce, and it became a big hit product. When he joined Hayakawa, Sasaki remembered this asked himself what sort of product could pull the ailing firm out of trouble? His conclusion: an electronic calculator.

The idea of a calculator came from an obscure English company called Sumlock Computometer. The “Amita” was based on vacuum tubes that Kobe Kogyo manufactured. Perhaps Sasaki was at the right place at the right time, or more to my entrepreneurial stereotype, he knew a good idea when he saw it and matched it with the right technology when it finally arrived.

These were desktop calculators, some with CRT displays – not what we used in college. The leaps from vacuum tubes to integrated circuits to portable pocket calculators took vision. Around 1965, two men far apart geographically (yet similar in their strategic thinking) independently come up with the idea of the calculator as something more than an office tool. One was Sasaki. The other was Patrick Haggerty, chairman of Texas Instruments.

TI was one of the first companies to put transistors into production and it was clear to Haggerty that they needed a product to demonstrate its usefulness. Therefore, he launched an R&D program to build a pocket transistor radio. Eventually, they partnered with a small company to manufacture radios for them. That was their first step into the consumer realm. The next step was into calculators.

TI could conceivably also have been a major player in consumer electronics. In 1967, the R&D team prototyped the world's first handheld calculator. They had integrated a desktop machine into just four chips and a product slightly larger than the transistor radio. Then, as Jerry Merryman, one of Kilby's collaborators on the project recalled, things started to go wrong: "What they did was they assembled a marketing task force.” They essentially concluded there was very little market.

Conventional wisdom in Japan was similar, but Sasaki pursued his vision. He was convinced that calculators needed to be truly portable and that MOS (Metal Oxide Semiconductors) technology was the key to integration and low cost. At the time, the three biggest Japanese chip makers were NEC, Hitachi and Mitsubishi. None of them were interested, they all viewed LSI MOS-ICs as untried technology, bound to fail. A logical opinion at the time; it was extremely difficult to produce reliable MOS components. Sasaki continued to aggressively pursue his vision; he was desperately searching for a company which could supply MOS-ICs.

Sasaki flew to the U.S., where he visited Fairchild and most other semiconductor companies from Silicon Valley to the East Coast. "We're too busy making chips for the Air Force, we don't have any spare capacity for you." (If you recall your Fairchild history, you know how important the military market was in the early days.)

His last visit was at Autonetics, the electronics arm of the giant aerospace conglomerate North American Rockwell in Los Angeles. If you've seen Jim William’s artwork, you’ll recognize Autonetics as a manufacturer of defense products, too. Sasaki offered an order worth tens of millions of dollars. But profit margins would be low. He thought he had convinced Fred Eyestone, who headed Autonetics. But once again the answer was a polite, "Sorry, we'd like to help you, but we're fully booked."

He went to LAX airport to fly home. Then, just at boarding time, he was paged: "Would Dr. Sasaki please come to the information desk?" There he found a message that a helicopter was waiting to fly him back to Autonetics. Timing is everything.
Sharp QT-8D, under the hood

In September 1969, Autonetics shipped its first 25,000 MOS chips. The following year, Sharp produced one million calculators equipped with Autonetics/Rockwell MOS-ICs.

Today, Sharp is still making calculators. Rockwell is no longer making calculator chips. Toshiba replaced Rockwell and soon migrated to new C-MOS (Complementary MOS) technology. CMOS calculators would be truly portable. Sharp finally chose to make their own chips.

As an aside, Sasaki felt an obligation to Robert Noyce for licensing the planar patent. Noyce in turn, asked for advice after he left Fairchild. Sasaki shared his thoughts on a single-chip calculator, while in parallel had invested 40 million yen into a company run by his old Kyoto University friend, called Busicom. As you are now predicting, Busicom contacted Intel in 1969 to develop ICs resulting in the single microprocessor known as the 4004. I’m sure you know the rest of that particular story.

I recommend reading Bob Johnstone’s book. It certainly reversed my stereotype. Here was a story about an engineer with boundless energy, creativity, charisma and the drive to break with convention to follow his vision.

http://www.barnesandnoble.com/w/we-were-burning-bob-johnstone/1111985584?ean=9780465091188 “We Were Burning: Japanese Entrepreneurs and the Forging of the Electronic Age”, By Bob Johnstone