In 1961, Sony constructed the Research Laboratory and the Sony Bridge in Hodogaya, Yokohama. Under its second director, Shigeo Shima, the Laboratory was renamed the Research Center in October 1969 and merged with the Research Division of the Sony head office. At that time, the number of Sony researchers increased from the initial fifty or so to three-hundred, and the Research Center was functioning as Sony's core R&D unit for basic research.
One day in 1970, a fellow engineer showed Shigeyuki Ochi, an engineer at the Research Center, a magazine and said, "Look at this article!" It was a report submitted by WS Boyle and GE Smith of Bell Laboratories in the United States announcing the development of the CCD in April of that year. A CCD is a semiconductor with a structure similar to an MOS, which stores and transfers electric signals. According to the report, two researchers invented the CCD one day after their supervisor instructed them to "think about a semiconductor with the same functions as magnetic memory." They thought up various product applications for such a device, including use in cameras and displays.
Ochi's initial response to the article was "This CCD has a very simple structure, but looks interesting. Let's ask the other engineers if they can come up with application ideas and see if anyone is interested in developing an application." He then submitted a questionnaire to the Research Center, asserting that "The development of technology is pointless unless the technology meets actual needs." Many engineers responded positively to the questionnaire.
"Well, since I was the one who started it, I might as well keep going," thought Ochi. He added the development of CCD to his list of projects related to the development of MOS devices. This was in December 1970. Depending on the voltage applied, an electric signal is transferred in a CCD and the two ends of the CCD act as the entrance and exit for the signal. Thus, a CCD can be used as a delay unit in signal transmission. Ochi, idly playing with a CCD, noted, "If you input an electric signal at this end, it's transferred and emerges through the other end." Moreover, he discovered that when a CCD is used as a photo-sensor, it converts the received light signals into electric signals. In other words, it performs as an optical signal reader. Thus, a CCD connected to a lens can act as an "electronic eye."
Ochi was delighted by this prospect, saying, "This is a very simple device. If it can be commercialized, we'll be able to make inexpensive cameras." He went on to develop several prototypes, the first of which was a CCD that captured an 8 pixel image. For an 8 pixel image, eight sets of photo detectors and transmitters receive light signals and convert them to electric signals. The more pixels a CCD has, the greater the resolution of the final image. Ochi next succeeded in 1972 to project the letter S with an 8 x 8 or 64 pixel CCD. By this time, Bell Laboratories had given up on commercializing the CCD, and other manufacturers had nearly given up as well. Ohga came to see Ochi's success and, although happy with the result, noticed that the image was very soft and out of focus. Ohga jokingly said to Ochi, "You can barely recognize the five fingers of a hand with this CCD."
As primitive as CCD technology was at this stage, it already had a strong supporter in Iwama, then deputy president. Ochi and his group were still merely playing with the CCD, but Iwama ordered them to, "seriously focus on the CCD and turn their R&D efforts into a usable product." Thereafter, the development of the CCD became a full-time project. Iwama gave the team a specific goal: "We have to produce a camera using CCDs at a price under 50,000 yen within five years. We're not competing against other electronics manufacturers in this field. Our competition is Eastman Kodak." The goal was clear, but the engineers were puzzled. In comparison with conventional tube cameras, it was obvious that a camera incorporating CCDs would be much smaller and easier to carry, while boasting more stable image processing capabilities. But the researchers asked, "Why is our competitor Eastman Kodak instead of electronics manufacturers?" In truth, no one quite understood why Iwama had stated this goal.
Iwama had worked for Sony in the United States from 1971 through 1973. Upon returning to Japan, he became a deputy president of Sony and the director of the Research Center. He had always attended the monthly CCD progress report meetings held at the center, always inquiring on the progress with interest. But there was a reason behind his sudden serious interest in the commercialization of CCDs.
In the 1950s, when Sony introduced the transistor radio, the company boasted that it was the world's number one manufacturer of semiconductors. Iwama was spearheading Sony's R&D efforts at that time. Under Iwama's supervision, Leona Esaki, who later became a Nobel laureate, developed the Esaki Diode. It was at this time that the transistor television was born (see Part I, Chapter 4). When Iwama returned to Japan from his assignment in the United States, Sony's semiconductor development team was stagnating. It had withdrawn from the fiercely competitive race to develop MOS devices for electronic calculators. Iwama had the distinct impression that "Sony's semiconductor team was dead." This was why Iwama pushed Ochi's team. Iwama's aim was to revive Sony's semiconductor development efforts through the CCD. He encouraged them in their efforts saying, "The Research Center must have confidence in itself." Iwama had actually seen the Bell Labs CCD with his own eyes, and he truly believed in the future of CCDs. "The CCD is a perfect vehicle for reviving Sony's semiconductor business. The day will come when CCDs will contribute to Sony's commercial line of products," thought Iwama. The reason for singling out Eastman Kodak as the competition was borne of an ambition to outperform the then prosperous photographic materials industry and to present the single-unit CCD camera and VCR as a "challenge from the electronics industry."
Thus, in November 1973, the engineers who were working on the CCD were transferred from the Atsugi factory to the Research Center, and development of the CCD as an "electronic eye" began in earnest.
The move of personnel and equipment to the Research Center marked the beginning of the protracted struggle to commercialize the CCD.
The main issue addressed by the researchers was the improvement of the sensitivity and resolution of CCDs, both of which increase in accordance with the number of pixels. At the same time, it was crucial to solve the problem of black specks and white scratches appearing in the images. These two flaws were caused by the presence of fine particles of dust and contamination by heavy metals. Solving these flaws as well as eliminating other factors that resulted in a poor image quality was easier said than done.
Except for Sony, almost all competitors vying to develop CCDs had quit. Many companies had been basing their development of CCDs on established in-house MOS technology, since devices such as CCDs and MOS have similar structures. Because Sony had withdrawn from the development of semiconductors for use in electronic calculators, the main application for MOS technology, Sony had no MOS technology available on which to base its CCD development. In other words, Sony had to start from scratch. Sony's competitors had the luxury of retreating to MOS when they decided that CCDs were too difficult to make. Sony, however, had no such technology to rely on. This, in effect, influenced Sony to continue with the development of CCDs. "Perseverance will bear fruit," was the spirit. In 1977, another company offered to collaborate with Sony in the development of CCDs. However, this company gave up after a year and redirected its efforts to MOS. Through it all, Ochi and the engineers believed that "CCDs will replace tubes in the future," and they continued to hope for success.
Despite their confidence, Ochi and his team felt pressure to abandon their efforts on several occasions because of the expense incurred for the CCD development. In fact, at times the team was referred to as "expensive, good-for-nothings," and the members had to endure criticism from their colleagues. But Iwama never pulled the plug on them, visiting their laboratory at least once a month. At times, he expressed his dismay at seeing the same, out of focus image on the screen marred with black and white lines. "Why can't you improve the image?" he would often ask. At the same time, Iwama never failed to give Ochi and his team guidance and encouragement, saying to them, "Don't give up."
Iwama was not a very talkative person. He thought things over before speaking, and his succinct statements always impressed those who heard them. Each time Iwama would visit, the entire team would consider his words carefully. Ochi used to describe him as the "Guru speaking a few words of wisdom." His disciples, so to speak, would put their heads together later to decipher the meaning behind them." Iwama was not receptive to vague answers or overly qualified explanations from his engineers. These engineers learned to present him with concise and concrete reports. Whether or not Iwama was purposely doing this, it was good training for the engineers.
However, even Iwama's courage and endurance were tested during the pursuit of CCDs. He often had to fight a nagging feeling of insecurity during the long development stage. In view of the time and investment it required this project would not turn a quick profit in the near future. Morita and other top management members were anxious about taking on such a project and wondered when the investment would finally begin to bear fruit. Iwama constantly said to Ohga, who was in a position to approve budgeting, "CCD development requires a lot of money and I can't assure you that we'll get returns on this investment some time this century." At one point, Iwama surprised Ochi by saying, "The heart of semiconductors is in crystallized silicon. Why don't we buy a crystal manufacturing company?" Indeed, the major raw material used in CCDs is crystallized silicon, and the purity of this crystal determines the quality of the image. Yet Ochi had never thought about the prospects of buying a whole company simply to gain access to high quality crystallized silicon. This comment led Ochi to believe that Iwama was truly expecting to wait until the turn of the century to collect on the investment. At the same time, Iwama's endurance, passion and willingness to undertake such a large-scale endeavor mesmerized him.
Iwama's enduring dedication combined with Ochi and his engineers' efforts yielded slow but steady improvements. The engineers were able to increase the number of pixels from 2,000 to 8,000 and to 70,000. Eventually in 1978, they were able to increase the number to 120,000. In its quest to produce high quality crystals, Sony invented the MCZ crystallization method. As a result, the problem of dust contamination in the pixels was overcome. Dust from people, machines and other origins were removed and countered. Finally, a perfect, clear image was achieved. The team took a photograph of it and presented it to Iwama. "You finally did it," was all he said.
In 1978, the project was transferred once again. This time it was for the Atsugi plant to start production. Test manufacturing of the 120,000 pixel CCDs began that same year, and in 1979 the new CCD was commercialized as the "ICX008." Total investment until that date amounted to 20 billion yen.
Following tests for mass production, a prototype CCD camera production line was established in Atsugi in November 1978. Finally in January 1980, six years and three months after Iwama first gave orders to begin work on the CCD; the world's first CCD camera was produced. The XC-1 was mounted to an All Nippon Airways (ANA) jumbo jet and used to project images from the cockpit at departure and landing times into the cabins, acting as the "eye" of ANA's Sky Vision system.
However, the production line's yield of viable CCD chips at this stage was poor and only one in several hundred was usable. It took twelve months to manufacture fifty-two CCD chips necessary to manufacture twenty-six cameras. As a result, each CCD was priced at 317,000 yen. Such a CCD would never realize the 50,000 yen target price that Iwama had specified.
"We can't call this a yield rate. It's more like an occurrence rate of usable chips," complained those overseeing the CCD production. The vicious "enemies" undermining their efforts were superfine dust particles that measured less than several microns and were undetectable by human eye. Clean rooms and dust proof attire reminiscent of space suits were introduced at all development and production sites. The battle against small dust was vigorous; people, machines, and every other thinkable source of dust were identified and dealt with.
After all the measures were applied, the yield rate improved enough to implement full mass production from 1983 at the Kokubu Semiconductor plant in Kagoshima.
The day finally came for CCDs to be used in mass quantities. In January 1985, the CCD-V8, an 8 mm camcorder (i.e. single-unit VCR and camera) with a 250,000 pixel CCD chip (the ICX018) as its "eye" was introduced worldwide. This product marked the beginning of a new business age using CCDs (see Part II, Chapter 1).
Unfortunately, by then Iwama, who would have been happier and prouder than anyone else to witness this milestone, had passed away. It was in 1982, just as preparations began for the mass production of CCDs under the supervision of Masahiro Takahashi, that Iwama became seriously ill. From his hospital bed, Iwama repeatedly said, "I'll definitely go and see the CCD plant," but this never came to pass. Iwama died in August of the same year.
Ohga succeeded Iwama as president and assisted by Takahashi, oversaw mass production of CCDs and the establishment of Sony's CCD business. From the first production lot, Ohga took a CCD chip and brought it to Iwama's gravesite. He affixed the chip to the back of Iwama's tombstone saying, "Iwama-san, we've finally succeeded in the mass production of the CCDs you promoted." When Sony employees visited Iwama's grave on the seventh anniversary of his death, the chip was still there. It was as if Iwama was tightly holding the chip in his hand.
The CCD development war began in 1986, when NEC's CCD was used in a JVC camera. The development team for Sony's 8 mm camcorder took on the challenge of developing a smaller unit with even higher image quality. In order to achieve this goal, the team of engineers requested even smaller and better quality CCDs than had previously been used. Once the 250,000 pixel CCD chip was commercialized, the next target was a 380,000 pixel chip. Just then, an unexpected event occurred. Despite clean room facilities, dust was found throughout the mass production line of the semiconductor plant.
The yield rate did not improve, and Sony was forced to delay the launch of the CCD-V90 high-resolution 8 mm camcorder by two months, just as it was to incorporate a 380,000 pixel chip. This was a time of hardship for Takahashi, then senior general manager of the Semiconductor Business Group, and his CCD team. As the competitors' cameras were enjoying increasing sales, the delay threatened to cause a devastating blow to Sony's 8 mm products. It was the first time Sony had delayed the launch of a new product by two months. The press criticized Sony with such headlines as, "Dust Kills Electronic Eye" and "The Latest Model Surrenders to Dust." Furthermore, Sony had begun marketing the CCD chips themselves as well as CCD cameras on an OEM (Original Equipment Manufacturer) basis. The 1987 "CCD Dust Rampage" affected Sony's OEM customers as well.
Once a quality problem affects an OEM customer, it requires a lot of time and effort to regain that customer's trust. Ochi and his team did have a very difficult time selling the CCD chips in 1983 when they first began OEM sales. They visited various camera manufacturers, but were turned away due to a poor track record. Sony had reduced the volume of semiconductors shipped to its customers to meet a surge in internal demand. Ochi and his team started by apologizing for this. The marketing of CCD chips was literally built on the team's sweat and tears.
Therefore, Ochi and his engineers spent this time in distress, saying, "We never want to inconvenience our OEM customers again." Luckily, Kanoi (then senior general manager of the Video Business Group) and Morio (then Kanoi's deputy who was responsible for the development of 8 mm camcorders) were sympathetic to the efforts of the CCD sales team. Like the sales team, Kanoi and Morio were also eager to get their hands on as many CCD chips as possible. The CCD-V90 model was being advertised in the media and causing a lot of demand in the market. Kanoi and Morio told the team, "Don't worry about our internal requirement for these chips. Sacrifice us and tend to the needs of our OEM customers first." Ochi and his engineers were relieved.
Sony was not the only one that suffered from dust problem. Sony's competitors were experiencing the same problem. The two major manufacturers of CCD chips also found their production stymied.
Takahashi and those overseeing CCD production worked day and night through their summer vacation. Everyone worked to eliminate one source of dust after another. An amulet from the Kagoshima Shrine was placed in front of the CCD tester to bring everyone good luck. And suddenly, one day in August, it was as if their prayers had been answered--the yield rate improved. They all cried, "Let's celebrate!" and they rejoiced that night. Sony's CCD manufacturing operations were finally revitalized.
No one has ever forgotten the struggles after that, which some said was the worst ever. The gains made that year became the basis for enabling Sony to become one of the few semiconductor manufacturers enjoying a constantly high yield.
Overcoming setbacks, the CCD business gradually grew to become one of the key elements of Sony's semiconductor business.
In the 1980s, CCD chips rapidly became smaller and denser. Within Sony, challenges such as, "We have a two-thirds inch chip, so let's make a half-inch chip," and, "We have achieved 250,000 pixels per inch, so let's make it 380,000," were frequently heard. In the early 1990s, a chip one-third of an inch in size was indeed introduced, and a high-resolution chip of two million pixels per inch for high definition television was also realized.
New technology to support CCD chips was also developed. On a CCD chip, the pixels are neatly aligned. Each pixel receives optical signals, converts these signals to electric signals, and stores them. However, the light hitting the wiring between the pixels is wasted. "How can we ensure that this light is not wasted? If this light was also converted, then the image quality could be improved without increasing the number of pixels," said Ochi. He and his team tried applying curved shape lens on top of the pixels to collect the wasted light into the pixels. The world's smallest lenses, with a diameter of only seven microns, were dubbed "On-Chip Lenses." Small as the lenses were, they doubled the CCD chips' light sensitivity. On-Chip Lenses were developed in June 1989.