Japanese Technology - The New Push for Technical Leadership
December 13, 1965 Electronics Magazine Article

December 13, 1965 Electronics [Table of Contents] Wax nostalgic about and learn from the history of early electronics. See articles from Electronics, published 1930 - 1988. All copyrights hereby acknowledged.

If the December 13, 1965, edition of Electronics magazine had a theme, it was undoubtedly reporting on the current state of Japanese technology. The cover photo is a shot of microwave antennas dominating Tokyo's skyline. Japanese technology companies worked hard to overcome the largely undeserved negative connotation that a "Made in Japan" label carried in the day. Do you remember the scene in the movie "Back to the Future Part III" where Doc, having traveled back in time from the 1950s, is trying to fix his DeLorean time travel car and discovers a burnt out circuit and claims, "No wonder this circuit failed, it says made in Japan?" Marty, who travelled from the 1980s, replied, "What do you mean doc, all the best stuff is made in Japan." There is no denying the Japanese people succeeded at their goal.

I plan to eventually post all of the articles reporting on Japan's efforts to gain a leadership position in the electronics world in the 1960s : Westernizing Japan | Japanese Technology - The New Push for Technical Leadership

Japanese Technology - The New Push for Technical Leadership

No longer content with improving on U.S. developments, Japanese engineers are now stressing research. One reason is the increased cost of labor

By Lewis H. Young, Editor

In 1960, when this magazine last took an in-depth look at electronics in Japan [Electronics, May 17, 1960, pp. 53 to 100], we found an industry built almost entirely on cheap labor. Research was meager. The principal products were low-priced transistors, components and radios. The best customers were bargain stores.

Today's vibrant Japanese industry is growing as rapidly as any in that country. High-speed production lines spew forth television sets, radios and tape recorders, and workers apply the private brand names of American electronics manufacturers as well as such American retail giants as Sears, Roebuck and Co. and Montgomery Ward & Co. Silicon transistors are slowly appearing, and engineers are installing facilities to produce integrated circuits.

But the biggest change is the new accent on research. At company after company, engineers talk about developing new concepts, rather than improving on American technology. Money, long scarce, is beginning to arrive from surprising places: The Bell Telephone Laboratories, for instance, is financing work at the University of Tokyo on time-division electronic switching for telephone exchanges.

A Glamor Industry

Electronics' importance to the Japanese economy is far greater than one would guess from its annual sales of $2.5 billion to $3.0 billion. The technology offers Japan everything she needs in order to attain her national goals:

• Economic growth. The electronics industry is expanding nearly 15% a year, and that's just a start. With relatively little capital necessary, compared with steel or chemicals, electronics companies are springing up all over the country and existing concerns are constantly adding facilities.

• Exports. Despite a year-long recession, exports of semiconductors, components, radios, television sets, tape recorders and microwave equipment continue to climb, bolstering the country's balance of international payments. Record sales have been reported by the Sony Corp., which exports 63% of its production; also by the Nippon Electric Co. and Sanyo Electric Co., other big exporters.

• Long-range potential. Although the United States is Japan's best customer, with Western Europe a poor second, the new countries of Africa and Asia have gargantuan appetites for communications equipment and consumer products, which are Japanese specialties. These countries still lack money to buy such goods, but the Japanese are confident that they have the inside track for the time when the demand materializes, Japan has already sold solid state microwave equipment to Pakistan, Indonesia and India, and maintains good relations with many of these new countries.

• Prestige. Because electronics is an advanced technology, the Japanese think it can help them attain the reputation of being an advanced country. After shutting themselves off from the rest of the world for nearly 300 years, until 1850, the Japanese became known as copiers when they rushed to catch up. The desire to be considered advanced - which approaches the dimensions of an obsession in some quarters - is one reason the Japanese grasp any new technological idea that comes along, even if they see no immediate application or benefit. An example is the way Japanese companies plunged into color television in 1960 - and, in many cases, were burned.

• Productivity. Electronics technology offers the capability of developing the automatic control equipment necessary to keep Japanese industries competitive with those in other parts of the world.

Tough Nut to Crack

"To compete in world markets, Japan will have to become a highly automated, high-efficiency producer," says Masahiro Shimizu, president of Hokush in Electric Works, Ltd., a producer of instruments, process controllers and computers for automation. Shimizu and other progressive executives recognize the nutcracker in which Japan's electronics industry is being squeezed.

From the south - Taiwan and Hong Kong - comes the threat of cheap labor - the same asset that Japan once used against the United States. From the east comes pressure by the superior technology of the United States. But the Japanese figure to be a tough nut to crack.

Salaries in Japan have risen sharply in the past five years - about 10% a year, with a 13% jump in 1965. A production worker starts at nearly $168 a month if you add fringe benefits; that's hardly competitive with the $15 a month paid in Taiwan or $30 a month in Hong Kong. And these increases will probably continue.

Electronics technology in the United States is still superior to Japan's, but the Japanese have been able to shorten development schedules by using the results of U. S. experiments and omitting the procedure that U. S. engineers had found to be unproductive. Evaluating the current status of Japanese electronics, Ichiro Isaka, chief engineer of the Electronic Industry Association of Japan, says: "In consumer products, Japan is number one in the world because of its high production rates and low costs. Our microwave equipment competes on even terms with U. S. products. But we are way behind in the study of integrated circuits."

Being second to the U. S. in technology is more than a matter of embarrassment to the Japanese. It is expensive, because Japanese companies must pay royalties to the U. S. owners of patents the Japanese want to use.

Fujitsu, Ltd., is an exception because it will not sign a licensing agreement with a foreign company. This producer of computers, numerical control for machine' tools, components and semiconductors prefers to develop its own devices and procedures even though it may enter a market late as a result. More typical, however, is the Nippon Electric Co., which has a long list of licensing arrangements: with Honeywell Inc., for data-processing equipment; with the International Telephone and Telegraph Corp. for communications equipment; with the Western Electric Co. for telephone equipment; with Varian Associates for microwave tubes and linear accelerators; and with the Fairchild Instrument & Camera Co. and General Electric Co. for semiconductor processes.

Electronics makes the Japanese appear advanced.

An American who has lived in Japan for many years noted another change recently. He said: "When I first came to Japan in 1954, Japanese consumer products were made badly. The only saving grace was that you could get them repaired. Today, Japanese appliances are beautifully made, but when something goes wrong there's nobody to fix them." Most good servicemen have given up repair work for employment in factories where they work shorter hours and earn more pay.

But clearly the biggest change is the new accent on research. One of the greatest incentives to this approach is the desire to become independent of U. S. patents.

More Development Than Research

The biggest deterrent to effective research in Japan is management's inability to evaluate the importance of such studies. Sanai Mito, managing director of the Central Research Laboratory at the Hayakawa Electric Co. explains: "Now companies realize how important research is, but we are just half-way. Management is too hasty. Though they claim they understand research, they want quick results. To get quick results you have to go into development work, not research."

As a result, the accomplishments of Japanese laboratories are heavier on development than on research. At Matsushita, Tetsujiro Nakao, senior managing director, puts the case strongly: "Development is important, but if we ignore fundamental research we will be in trouble." Yet the Central Research Laboratory recently completed the design of a home video tape recorder, a project more akin to an engineering department than a research facility. Nako says his company spends about 3.8% of sales income on research and development; last year's sales totaled about $616 million. "At least 15% of the total R&D budget is earmarked for basic research," he added proudly.

The growth of research facilities and the formation of new ones in the past five years, clearly shows Japan's intent. The list of expanded facilities and new ones is imposing:

• Just four years ago Hayakawa, which manufactures Sharp-brand tv sets, tape recorders and other consumer products, organized its first central research laboratory, even though the company itself started in 1923.

• Sanyo Electric Co., another appliance maker, officially recognized the importance of research when it chartered a central laboratory in December, 1961, after a small group had operated as a technical department for three years at the company's headquarters. The laboratory staff has increased to 230 people and is expected to expand to 300 as the proper personnel is found.

• The central research laboratory at Hitachi, Ltd., has increased from 600 to 1,400 people over the past five years - and is still growing. The present goal is a staff of 1,500.

• In 1960, the research facility at Mitsubishi Electric had 600 to 700 employees; today the number is 1,500.

Management is still too hasty; can't wait for research.

Even though Japanese companies have increased their expenditures for R&D dramatically, there is still a serious shortage of funds for that purpose. The Japanese recognize that they must compete with the United States and bewail the giant appropriations for military and space research in America. By contrast, Japan's space program is budgeted around $7 million for 1965 - and that is nearly double last year's figure.

To stretch R&D funds, the government and companies have evolved practices designed to minimize duplication of effort. A lot of fundamental research is performed at government laboratories, and the results are available to all companies.

When word of a truly significant development reaches Japan, the first research is likely to be done at a government laboratory. In 1955, development of numerical control for machine tools started at the Government Mechanical Laboratory near Tokyo. The first Japanese-designed computer was built in prototype at the Electrotechnical laboratory.

Many microwave developments started at the Electrical Communications Laboratory of the Nippon Telegraph and Telephone Public Corp., which the government owns. Much broadcasting equipment has been designed at the research laboratories of Nippon Hoso Kyokai, the government-controlled Japan Broadcasting Co.

When developments are completed at a government laboratory the results are given or sold to Japanese electronics companies. Almost every Japanese computer company owes its technical start in data processing to the development of the Mark IV computer, the first machine built at the government's Electrotechnical laboratory. A few months ago, the Electrical Communications Laboratory gave the solid state design for a 15-gigacycle microwave repeater, which had been designed and tested at the laboratory, to the Nippon Electric Co. to manufacture for the telephone company.

When a government lab charges a royalty, the fee is often microscopic. Toshiba paid only $1,500 for the design of a two-tube color camera, developed by the research laboratory of the Japan Broadcasting Co., for televising the Olympics held in Tokyo last autumn. The payment included technical help in starting production.

Universities also help the Japanese to get more research per dollar. Although government professors are prohibited from receiving payment for outside activities, many faculty members secretly ignore the ban and work for private companies as consultants, buttressing the technical effort.

There is a legal way to use the universities too; it's called a kenkyusei, or research student. A company can send a graduate engineer to a university for one or two years to perform research under university supervision and take some courses. The cost is $300 per year plus the student's full salary. This year, Tokyo University registered about 30 kenkyusei.

One complaint heard often about Japanese educational practices is that too much money is spent on elementary and secondary schools and not enough on colleges and universities. Japan can boast of a literacy rate of nearly 100%, even though the language is exceedingly complex with nearly 2,500 characters (compared with English's 26).

In contrast, Japan's colleges have unattractive, rundown physical plants that are overflowing with students. Even in new facilities, such as those built to replace the bombed-out electrical engineering building at Osaka University, the structures are stark and unadorned, with unpainted concrete walls on the inside.

Changes in Research

Because the facilities are so bad, most of the research carried out at universities is theoretical, requiring a minimum of equipment. At the University of Tokyo, for example, sitting in a ramshackle frame building, Prof. Takashi Isobe studies correlation techniques and pattern recognition. His most recent work has been to develop a new method of measuring the dynamic characteristics of a control system.

Across the campus, in a slightly sturdier building, Prof. Jin-Ichi Nagumo studies self-organizing systems and learning machines, and develops simple devices for medical electronics, such as a direct-coupled pacemaker.

But conditions are changing. More money for experimental equipment is becoming available. Some of it comes from surprising places. The Bell Telephone Laboratories, for example, is financing work on time-division electronic switching at the University of Tokyo. Money is coming from Japanese companies too, as more of them are sending their R&D staffs back to college to keep up with American technology, because so many faculty members have gone to the United States for graduate study.

Despite Japanese efforts to husband their research resources, there is still plenty of duplication. The reason is that Japanese companies tend to play follow-the-leader. Half a dozen companies, for instance, are striving to develop continuous-wave Gunn-effect oscillators for microwave applications. Also, after word spread that the aggressive Sony Corp. was building an electronic calculator, two appliance makers - the Hayakawa Electric Co. and the Yaou Electric Co. - rushed the development of similar calculators too.

The greatest research effort is going into integrated circuits and electronic switching for telephone exchanges. Still, a catalog of other Japanese research is impressive.

At the Nippon Electric Co., which many Japanese credit with conducting the best industrial research in Japan, projects range from computer developments to quantum electronics.

The NEAC L-2 computer, built at Nippon Electric's Central Research Laboratory, is the basis of the company's new model 500 commercial machine. It has a 10-megacycle clock rate and three memories: wire, core, and a read-only "eddycard" memory made of square holes in a copper plastic sandwich. In the experimental machine, Nippon Electric has built an index register of 15 words with tunnel diode circuitry to increase speed. The machine's add and subtract time is 0.5 microsecond, with fixed-point arithmetic and 1.4 microseconds with floating point; times for multiplication are 1.9 to 7.7 microseconds for fixed point and 2.4 to 7 microseconds for floating point.

Most of Nippon Electric's work with integrated circuits (see p. 90 for a survey of all integrated-circuit work in Japan) is being done at the company's semiconductor division. But the lab is trying to develop integrated circuits - double NOR logic units - for delta-modulation communication equipment.

In another project, a researcher at NEC has studied voice analysis and built a machine that accepts numbers, given orally, for dialing a telephone.

One of the Mitsubishi Electric Corp.' s top-priority research projects is the development of three-dimensional radar for Japan's defense agency. Says one Mitsubishi engineer: "It's the only development we can give the U. S."

Details are classified, and the company says only that the system uses a phased-array antenna. If it works, and the tests should be completed by next summer, Mitsubishi's system will be the most accurate radar in the world and have the longest range.

Matsushita's Wireless Research Laboratory has almost as many projects under way as the company's Central Research Laboratory. Its emphasis is on new materials and components. About a year away from commercial use is a cubic boride material for the recording heads of tape recorders. Its hardness is 1,000 on a Vickers scale, compared with 600 for ceramic materials; its permeability at 100 cycles per second is 3,000, far better than the 1,600 of ordinary head materials. Its one flaw is a higher coercive force than that found in conventional materials, so present work is aimed at reducing this force, the company says.

Closer to production is a cadmium-sulfide p-n junction for photovoltaic cells. This material is less expensive than silicon, which has been used the same way, but cadmium sulfide's efficiency when generating power is slightly lower. Current generated is 15 milliamperes per square centimeter at 0.4 volt with a load and 0.5 volt on open circuit.

At Sanyo's central laboratory, work on injection electroluminescence shows promise because Sanyo scientists see a potential for high efficiency and brightness in the single crystals of zinc telluride which they are using.

Despite the surge of recent research, Japan's greatest strength still lies in her ability to perform high-quality development of products based on American technology. Visitors from the United States continually find examples of ingenious engineering. Last month at one of Sanyo's laboratories, for example, an experimental silicon controlled rectifier was controlling a refrigerator's motor. Changing the frequency of electricity doubled the running speed of the 40-watt motor, producing the same cooling effect as with an 80-watt motor.

To assess the technical contribution of the electronics industry in Japan, the editors of Electronics have asked several Japanese experts to report on their specialties, emphasizing those developments which are peculiarly Japanese. From these reports, which appear on pages 81 to 112, an engineer can assess Japanese technology. Those segments of technology selected are:

1. Discrete semiconductor devices, which are essential ingredients in most products (see below). 2. Integrated circuits, because the Japanese are emphasizing linear devices and racing to catch up with the United States (p. 90).

3. Solid state microwave, because the Japanese have the densest networks in the world and boast that their equipment is as good as anybody's (p. 99).

4. Industrial electronics, because the Japanese are pushing hard for automation to offset rising labor costs. One report covers numerical control of machine tools (p. 106); the other, process control by computer (p. 110).

Posted August 14, 2018