Oral-History:Ki Dong Kang

About Ki Dong Kang

Ki Dong Kang was born in 1934, and grew up in Japanese-occupied Korea. The son of a civil engineer, Kang attended Seoul National University in electrical engineering. Kang later came to the United States, first attending graduate school at the University of Minnesota and then Ohio State, where he received his Masters and PhD. While at Ohio State, Kang was asked to take part in the new semiconductor lab working on device fabrication and diffusion, influencing Kang’s later work in the semiconductor industry. After graduating, Kang joined Motorola in 1962 in the semiconductor division, and he was unofficially part of opening a Korean assembly operation for Motorola. Kang then worked for Stuart Warner Corporation, where he contributed to the assembly operation, helped set-up a plant in Taiwan, and had projects like the seat-belt interlock system. While at Stuart Warner, Kang became interested in starting his own semiconductor operation, eventually starting Integrated Circuit International, Incorporated (ICII) making watch chips. Kang eventually sold ICII to Samsung and was employed by the company a few years before leaving Samsung and the semiconductor industry.

In this interview, Kang discusses his experiences in the semiconductor industry and Korea. He talks about his life in Japanese-occupied Korea, early education and the directions Korean technical education has taken since his training there, as well as his work at Ohio State. Building assembly plants in Korea and Taiwan is also covered, along with the various issues that went along with opening the plants. Kang also talks about the difficulties of opening his own semiconductor operation, particularly gathering the necessary money. He discusses the working environments at the semiconductor lab, Motorola, Stuart Warner and Samsung, along with learning about management and semiconductors in general at the Ohio State lab.

About the Interview

KI DONG KANG: An Interview Conducted by Andrew Goldstein, IEEE History Center, 13 June 1996

Interview #270 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Ki Dong Kang, an oral history conducted in 1996 by Andrew Goldstein, IEEE History Center, New Brunswick, NJ, USA.

Interview

Interview: Ki Dong Kang

Interviewer: Andrew Goldstein

Date: 13 June 1996

Location: Saratoga, California

Background and Education

Goldstein:

Dr. Kang, can you start by telling me about your early education and early career?

Kang:

Actually I grew up in Japan.

Goldstein:

When were you born?

Kang:

I lived in Japanese-occupied Korea for about forty years, so during World War II we were part of Japan. Since my father worked for the Japanese government, our family was treated better than the average Korean family.

Goldstein:

When were you born? What year?

Kang:

1934. My family did not mix with Koreans, but we mixed more with the Japanese as part of the occupation, so as a small kid I was full-time Japanese. When World War II ended we became Korean. It was very hard to adjust to a new environment, because everybody in Korea hated the Japanese because of the occupation. And because of my father's working for the Japanese government, our family was [a] target of their hate. So as a small kid it was tough to overcome that problem. That only lasted a short period, but after that the Korean War started.

Because my father was actually a civil engineer, he built lot of dams for agriculture purposes, so people started recognizing his contribution for the Japanese government. He was actually drafted by the U.S. - what did they call it? A temporary government. He started his government job again in Korea. I went to grade school and high school and college at Seoul National University. I came first to [the] University of Minnesota.

Goldstein:

You graduated from the National University of Seoul, and then came to the University of Minnesota for post-graduate work?

Kang:

Yes. The first few quarters I didn’t take any technical things, then I transferred to Ohio State. And there I was in graduate school, and I was very fortunate there. Some of the courses I took I was doing fairly well in, so my advisor recommended that I go all the way to PhD courses. So I stayed there.

Goldstein:

If you weren’t taking technical courses at Minnesota what were you taking?

Kang:

Language.

Goldstein:

That’s right. What made you want to come to school in the U.S.?

Kang:

When?

Goldstein:

Why did you want to come to school in the States?

Kang:

Well there was nothing I could do over there. I graduated National University of Seoul, Seoul National University. At that time its faculties' standard wasn't really there. Well, I didn’t really know, since I have no way to compare it.

Goldstein:

What year was that?

Kang:

I think I came here in 1958 or '59.

Goldstein:

What was your degree in at National University in Seoul?

Kang:

A Bachelor of Science.

Goldstein:

Were you specializing in engineering or chemistry?

Kang:

Electrical engineering.

Goldstein:

It sounds like you were interested in material sciences?

Kang:

No. At Seoul National University there were two departments. One is electrical engineering, that’s basically power engineering. In the old days there was no radio engineering. Then they set up another department called, I think, communication engineering. It’s more than any radio or wires type of engineering. I felt it more up to date, but I graduated with a degree in it. Electrical engineering is basically power engineering.

Because of I have this problem [indicates physical problem], I was not drafted in the army, since I cannot bend any more than this. Everybody went to join the army, yet I was left behind. So I decided to come to U.S. for the study. One of the professors at Seoul University also recommended me to go to the United States. He didn't specify a particular school, but he knew somebody at the University of Minnesota, since at that time the University of Minnesota had some faculty training program with Seoul University.

So there was a tie. Not at a student's level, but at the faculty level. Somehow I was recommended and I went to Minnesota. The reason I transferred to Ohio State was that a very close friend of mine was there, but I didn’t have any close friends at the University of Minnesota. In a foreign country I’d rather be near to somebody close to me. So I transferred to Columbus, Ohio.

Semiconductor Lab

First, a funny story. The Ohio State EE department is a five story structure, and basically first level and second level are lectures and office facilities and the higher levels are labs. Because it’s close to Dayton, Ohio, lots of Air Force officers came to the school to get credit. Somehow it’s well known for a non-linear synthesis of control system, for instance, a missile guidance system.

I was most interested in that kind of thing, control systems, so I took a lot of courses in that direction, and also the circuit theories, AC theories. One day my advisor called me and asked me would I have any interest in joining [a] new lab called semiconductor? A semiconductor lab. I didn’t have any knowledge as to what a semiconductor really was at that time, I only knew it was a new thing. I knew there were many changes at that time.

Goldstein:

This is around 1960 by now right?

Kang:

Yes, I think ’59.

Goldstein:

Were semiconductor diodes very common at this time?

Kang:

Yes. But we were still learning a lot from the vacuum tubes, vacuum [Unintelligible Passage] characteristics and load co-op, that sort of thing. Of course a few changes, yes. But because my advisor was so strongly recommending this I couldn’t turn it down. I said that I would join the new lab. After I made that decision or commitment, everybody said, “You made the wrong decision; you’ll never graduate. We don’t have even a lab yet. How can you print your thesis?” We didn’t have a faculty trained in semiconductors. I didn’t know.

But anyway, I joined it. After that point I had a graduate assistantship that was basically grading papers. As soon as I joined the semiconductor lab, they officially opened the lab, and I think I became the first or second employee there, and suddenly I was given my own desk in a very affluent lecture room; there was nothing else there. From that day on I got promoted. As far as money was concerned, it was a better position and no work, yet. I saw it as a starting point.

Suddenly they were putting lots of money into a new lab. And my advisor got some research project from Bell Labs. No, not Bell Labs. He went to Bell Labs one summer and he got the project through Bell Labs and actually the monies came from Fort New Jersey at...

Goldstein:

Fort Monmouth?

Kang:

Fort Monmouth, New Jersey, yes. At the army signal supply agency over there. We got some diffusion work, it was silicon diffusion. I didn’t know anything about diffusion. So that was a starting point.

Goldstein:

This was the first time. Who was your advisor?

Kang:

M. O. Thurston.

Goldstein:

The objective was device fabrication?

Kang:

Yes, there’s many ways to fabricate with silicon devices. And the new technology was so-called arrow injunction, but the arrow is not very controllable, so in weather we can diffuse at a high temperature, then a diffusion process is more controllable, at least that's what we all thought. That’s the starting point of the diffusion technology.

How can you control the geometry? That’s the other question. In geometry use, silicon naturally formed silicon oxide in an oxidation environment using oxide as a masking layer. You could cut the oxide in a certain part of the silicon then create diffusion in a certain pattern. That was the major objective. How can we diffuse it through the area where we want diffusion, how can we control the area where we don’t want diffusion? It’s how effective these oxide layers are at other things.

We had to set up everything that they funded us, and that they are capsules, [Inaudible Passage] is more capsules about this big. Their concept was that everything’s new. This was no EE. First of all we were dealing with silicon; that’s a kind of metal, all right. Its work was closer to the metallurgical department rather than an EE department. Diffusion work.

One day I went to the metallurgical department library, and I tried to just scan the dissertation topics. Wow, I found all sorts of diffusion work done twenty, thirty years ago! I thought this was brand new in the EE department. Basically I went to the metallurgical department to see how they conducted, how they analyzed diffusions. All these theories were there. One of the interesting topics I came up with was the solid state diffusion kind of thing I was doing at that time, and the author of that particular dissertation was now a faculty member in that department. I visited with him and discussed things, and found that he somehow knew my advisor, too. Of course they all knew each other there, they were all in the same school.

When it came to a dissertation or thesis, we published so many papers. Of course we had quarterly reports, annual reports, semi-annual reports. Automatically my advisor said, “Why don’t you re-type that couple of quarter reports? That’s good enough for your Master's thesis.” I didn’t know that it was that easy - it’s already there! Also, because these were the reports for the government, all the graphs or charts are all done by professionals. I just borrowed that in my dissertation. Of course I originated it. So my thesis looks a hundred times better than those of the other guys. They are typed by themselves but we had a secretary. Somehow I got the Master's very easily.

For a Ph.D. at that time I had about half a dozen people working for me, of course all students. I went to IBM or two or three places presenting our work. Not my dissertation, but our contract work. We went to Dayton, Ohio, and a couple of places. Now that I knew more, I knew how to present my paper. When it came to the oral exam, that was the toughest part of the Ph.D. The written part was relatively easy; but the oral part, it’s where they will ask you questions until you really get in trouble. But I did very well, I thought I did very well presenting my dissertation.

The advice of all the EE department professors was that this was a foreign topic. This was not electrical engineering anymore, you know. This was all meteorological or chemical, not even physics, since I was dealing with the fabrication side of the semiconductor, not the theory. Theory is all Shockley and Bardeen and all these guys did it. But mine was how to create new devices, so this was more like chemistry or even like a sort of physics. I really got through very easily, in a way. I got the Master's and Ph.D. in three years.

Management and Joining Motorola

Then we come toward getting a job. After I got the Ph.D. in the semiconductor field, I accumulated lots of coursework in the known semiconductor areas: controls or design and that sort of thing. I was debating which way I wanted to go. It’s fabricating semiconductors high temperature treatment, chemical treatment, bonding, its application is electrical engineering, no question there, but fabrication itself is not electrical engineering. Yet I had done that part. I interviewed at General Electric in New York somewhere.

Goldstein:

In Schenectady.

Kang:

That’s the government division. I interviewed in three places. One is in New York, and the other one, government division, is Poughkeepsie, New York. That’s their semiconductor work. The third one is in Phoenix, Arizona. That’s Motorola. Because of my allergies, I have a bad case of hay fever; the ragweed kills me. During the summer it was real bad, and doctors told me to go somewhere where there’s no ragweed. The obvious choice was Phoenix, Arizona - the desert. Also when I interviewed, it was the middle of the winter, and I thought, wow, it’s great! I got offers from all three places. Motorola's offer was not the highest money-wise, but we liked it there, so we joined there.

Goldstein:

What year was that?

Kang:

That’s ‘62 I think. Another thing: when I joined the semiconductor lab in Ohio State I didn’t know anything about organization, how to manage it. And Ohio states that its lab belongs to the research foundation. The research foundation is kind of a management organization. They set up at the labs in the many locations on campus. And with a new lab, they sometimes don’t have many management skills, but they send very experienced secretaries. Well, the position is officially secretary, but basically the secretary knows the entire management repertoire: how to hire, how to do paperwork, and do all of these things.

Actually I learned a lot at that semiconductor lab in Ohio State since it was a start-up situation, like purchase requisitions. If you start requisitions, then your supervisor has to sign it, it is just procedure. Now I’m very familiar with this. I didn’t pay too much attention and when we are hiring people for a new lab, then you had to write down the procedures first. I guess in a way we were on our own. If you were going to buy equipment, then this was the way we were going to buy it. Firstly, who is going to approve it, how much you can spend at the lower level? We have to have some rules for any organization. That I learned there.

I joined the Motorola organization. Arnie Lesk was my supervisor.

Goldstein:

How do you spell that?

Kang:

Arnold. I.A. Lesk. I is for Israel and A is for Arnold. Lesk, L-e-s-k. I think he came from GE and he was the lab director. So he was at the same stage as I had been in Ohio State, organizing a research lab in the Motorola semiconductor division. I’m one of the first Ph.D.'s who joined his organization. I faced the same problems, but now I knew how to run or how to organize things. They thought I was the genius, fresh from school with a Ph.D. Most of the Ph.D.'s didn’t have any experience in organization. I already knew how to run a small organization, and how to initiate certain requisitions; if you don’t have it, we have to make a certain rules.

Goldstein:

They’d let you set up the system?

Kang:

Not set up the system. They just really respected me, since I already had the knowledge. In some areas I had more experience than my boss.

I enjoyed the work there. Since they studied so many small products, they were difficult to control, so for my first job they asked me to re-organize or consolidate all the products in a few families. They had many twin numbers and many different transistor shapes, and we had to come up with new theories, such as SCR1, SCR2. Actually it wasn’t given to me by production people.

Then as soon as I started that, we started seeing more problems. Transistors or wafers we were making were not reproducible, or sometimes they came out very good, sometimes bad, sometimes we’d scrap the entire lot. We started realizing that somehow our surface treatment was not reproducible.

Goldstein:

Was Motorola making these transistors for use in their own products, or were they selling them to other companies?

Kang:

They were number two in the entire world in producing semiconductor products at that time. Texas Instruments was number one. The Japanese were one of the largest customers. I had a fringe benefit at that time. Since Japan was buying so much of our semiconductor product, they were coming to the plant practically every day. And one August they wanted a tour.

Marketing people were so tired of these people. Not only that, they didn’t speak Japanese, and not all the Japanese guests coming in spoke good English. Obviously they knew I spoke Japanese, so they often asked me to entertain them and give them a tour. So that became a rule now. It was on company time but my boss, he didn’t mind it since that customer was very important to us. And our marketing people, they liked me to do it so that they could rest.

I got to know lots of Japanese people there, and they brought cameras. One time we objected, but later we said that they could do whatever they wanted because we thought they would never catch up to us. The technology gap was so huge at the time. Also, after they saw Motorola's operation, all these people, the facilities, the businessmen, the engineers, they learned how the Japanese could catch up to us in technology. I think this is an American trait, in part: we are too naive. We believed that they would never learn how we made the semiconductors. Being very open-minded, that was the American way of doing things. No more. But at that time we were so generous.

I enjoyed that as we came up with new products, a couple of government projects. I stayed there about seven years.

Korean Assembly Operation

At one time I went to Korea to do a plant survey for Motorola's Korean assembly operation.

Goldstein:

Did Motorola have plants in Korea?

Kang:

Yes, Motorola tried to set up some overseas assembly plants. And as I happened to be a Korean citizen, and at that time the general manager was Les Hogan, L-e-s Hogan, and he didn’t know anybody in Korea, so he asked me whether I would go to Korea before he arrived there. I was just to find out what was available and where to buy land, who to contact and that sort of thing, basically ground work.

I went there and stayed for a couple of months. I enjoyed that. I opened up the opportunity to set up the Korean operation, but officially I could not be named as the representative to Korea because at the time I was still a Korean citizen. I was also head of one government project that was finally coming very soon, so I couldn't stay away for too long. So I came back and they replaced me with George Needham. He was the first president of Motorola's Korean operation.

Goldstein:

You couldn’t be the representative in Korea because of your citizenship?

Kang:

Yes, it’s just a conflict of interest. How can Korean citizens be sent to Korea to represent an American company? It can’t be done. Your representative should be your own national.

Goldstein:

But surely since you were an employee of Motorola...?

Kang:

No, it was a developing country. You don’t understand? If you are a citizen of another nation, they cannot touch you. If you are their own citizen, they can do anything they want to you

Korea, at the time, was not a democratic country. One general had one department. On the surface it looked like it was a democracy, but in certain areas it’s entirely different.

Goldstein:

Do you know why Motorola wanted to open plants in Korea?

Kang: The Korean people are very diligent, dedicated, and there was an excess of educated manpower available. Goldstein: They could get good labor? Kang: Quite good labor, and it’s not too far. Also, the U.S. has [a] large army stationed there, so in that sense it was more secure than in other countries. I don’t know all the details, but they had looked at Taiwan, Hong Kong and the Philippines. At that time we were not serious about looking at Indonesia or Thailand as they were way behind. Korea was the right choice at that time, although it lack of communications and transportation was terrible. But transportation was no big deal since the finished product didn’t weigh that much as compared to other products.

That was the new venture entirely from my point of view. I had concentrated on lab work or developing new products. Over there, I was basically talking to the people. In all my work in U. S., my Korean language didn’t mean anything. Suddenly it became the whole thing. I came back in I think ‘67, somewhere in that period. Goldstein: What people were you talking to? Were you talking to people to buy property from, or government officials? Kang: Yes, government officials, basically concerning import, export problems. Korea was at that time a free-trade zone, and there were some investment credits involved. Gradually they implemented all these things after we moved in. But the government was eager to have the investment come in from the United States. Goldstein: Korea has no indigenous industry to protect? Kang: Its government was very helpful. They wanted to succeed in bringing in American industries, especially those that were high-tech. Goldstein: In some countries a foreign plant comes in and is allowed to operate in the country, but they have to share the technology. Was there any demand from Korea to Motorola to share its technology so that Korea’s industry could grow on its own? Kang: Well, this was the tail end of their assembly operation. There was a lot of discussion. I don’t know if there [was] such a demand when they were actually signing the investment papers. I have no idea, since I was not involved in that part. I only looked for the best-priced land to locate the plant, and found out which government officials they needed to contact without wasting much time. Goldstein: All right, well so you say they were eager to have you come in, what sort of things did they demand? What sort of things did they want? Kang: The Koreans? Goldstein: What did Korea demand of Motorola? Kang: I don’t think they demanded much at that time. Any company, even a non-technical company, brings money in, and that was welcomed by the Koreans. All right, so this company was obviously high technology, and there was a potential for hiring lots of people. This was considered a golden opportunity, so there were almost no restrictions. At that time foreigners could not buy property in Korea. I think they let Motorola buy property, so that was a big shift on their initial policy. I think Motorola bought ten acres of land. I don’t know what it's worth now, but probably about a thousand times more than what they had paid at that time.

Motorola started there. And I had contributed to the potential. I interviewed lots of Korean engineers. That was when the Americans were coming in. When they first arrived, they didn’t know whom to contact. I did that part of the work. Goldstein: Did you still have contacts at Seoul National University? Kang: I closed the door. No more.

Stuart Warner Corporation

After I came back I got the very good offer from the Stuart Warner Corporation. Basically that company is located in Chicago. Goldstein: Stuart Warner? Kang: Stuart Warner. That’s a gauge manufacturer. They were building lots of gauges for automobiles. I don’t know how big that company was, but they were opening semiconductor operations here in the Silicon Valley. They wanted to keep up with new technology. They were in business only a few years, but for whatever reason they needed somebody who had more up-to-date knowledge in processing manufacturing. I interviewed for the position, and they liked me. Goldstein: Were you looking for a job or did they just contact you? Kang: I don’t know. I think that head-hunters had something to do with it. There are a lot of head-hunters. When I go to a conference there are a lot of people. I was giving many papers, not so much to chip or ray conferences, but to the Electrochemical Society. I think I gave papers in reliability at a couple of conferences. But basically I presented at the Electrochemical Society. That’s where most of the processing people came to give a new paper.

When I came to Stuart Warner I immediately noticed they were many years behind. I didn’t know that Motorola was way ahead, but it was obvious. I started contributing right away. Goldstein: They already had their plant built? Kang: Yeah, it had been up only for a few years, and one of the reasons they started operating it was they were very oriented towards DTR. Their main product was DTR logic circuits, the diode transistorizor. They were producing lots of them.

Now, their major customer was phasing out purchasing, so they had too many assembly operations here. That’s about thousand people. So as soon as I joined Stuart Warner, after a few months it became very obvious that they would have to cut this assembly operation. They were discussing that Fairchild had an operation in Hong Kong, Motorola had a Korean operation, and Stuart Warner, which was not as big as these companies, had to move the manufacturing operation somewhere off-shore. So one day I was told to look into it, because I had told them that I had done some survey work for Motorola. I visited Hong Kong, I went to Singapore and Hong Kong, to Taiwan and the Philippines.

Stuart Warner at that time didn’t have much money to invest directly. But they liked to grab local financiers to set up the operations so that we could go in and do our work. We decided to build a plant in central Taiwan, about one hundred fifty miles south of Taipei. I went there a few times and set up my part of the assembly operation. There are about three or four different ways of doing assembly operation, but mine was brand new. I went there many times to make sure that the assembly technology transfer was complete. Goldstein: This is the assembly of chips or devices that would then go back to America for assembly into gauges? Kang: Right. But as far as the semiconductor product is concerned, that was the final assembly. Previously its order of encapsulation was in a metal can or through hermetic sealing, so that the inside was capped with nitrogen gases. That was too expensive, so the newer trend at that time was plastic molding. But plastic molding interferes with the surface. My specialty was how to protect the surface from degradation or passivation. At the time, that was one of the key areas most of the manufacturers didn’t have good control over. Motorola had it. I was one of the contributors. Goldstein: Were you able to use that technology for Stuart Warner? Kang: Yes, sure. Goldstein: It wasn’t proprietary for Motorola? Kang: No. Another counterpart was Fairchild. Andy Grove was the top man at Intel. He did lots of surface work too. He’s well known. But Motorola didn’t have anybody waving flags. We were just quiet and did the job. But anyway, Motorola's product was as good as Fairchild's. One time we actually invited the engineer, I was the manager of the department, and my engineer gave a paper on how to create the devices. There we were permitted to say that it is de-ionized water that makes the surface stable. Since then everybody says de-ionized water, de-ionized water, de-ionized water! The entire world copied it.

All right, so that was the "de-i water", but what really de-i water was doing was leaching sodium iron out of the oxide. We didn’t say that much; we just said de-i water is the trick. Soon, all these people had identified and published it, that "de-i water" was leaching sodium. Sodium had been degrading semiconductor device[s] for twenty years. So at least now everybody knew how to make the device or the surface more stable. That, from the processing point of view, was a big breakthrough.

At Stuart Warner, I was not very happy since the management was ten, twenty years behind Motorola's management. Motorola was very aggressive; they were growing every day. Over at Stuart Warner, the top man was actually the finance officer. The accountant became the top man. He could not make any decisions by himself, and was always checking with Chicago headquarters. You cannot run a semiconductor business, a very aggressive technological company, in such a manner. If this were a trucking company, it would have been all right.

And well, I came up with a new product. I was in new processing. We had demonstrated that we could do certain things, and the first product we demonstrated was the seat-belt interlock system. Our product was the first one in Silicon Valley to be delivered to the automobile division. At that time, without buckling up, you couldn't start the car. We were the first ones who delivered working products. But that didn’t impress the management, or somehow they didn’t back us up any further. I became fed up.

Setting-Up a Semiconductor Operation

I was still making the Taiwan trip for the assembly operation. One of the Taiwan businessmen was asking me whether he could set up another assembly operation. I said I don’t know much about assembly. I am really involved with the front end, and not the tail end, of the process. He was asking how much it was going to cost, and so on. I made a rough estimate on the equipment and what it took to get started. At that time it was only a few million, as I recall it.

On my way back, I met one of the business people in Korea; I knew him from a long time ago. He said, "Well what are you up to?" I said I was going to propose this kind of thing to China or Taiwan. He looked at it and said, "Why don’t we do it?" See, the semiconductor was going to cost a lot of money if you ask somebody else to do it for you. But if you do it yourself it’s a different story.

Goldstein:

He was with what company?

Kang:

It's the KEMP, Korea Engineering and Manufacturing Company, that was the organization at that time. He was very interested in technologies. The reason I knew him very well, was that I had started a ham radio operation in Korea when I was in college. I had organized it. The organization was KARL, Korean Amateur Radio League. We had the organizational meeting at my home with about [a] dozen people. And since I was a young kid and college student I couldn't be the officer, so I brought him into it.

His name is Chu Han Kim. He was some kind of technical vice-president of something at one of the radio stations, the Christian radio station. He was well- known. So I brought him to KARL, the Korean Amateur Radio League, as an officer. I knew him very well, and now I saw him again. Well, I saw him from time to time, but this time he had a business operation. Obviously, they made some money. What they had done was that they set up some microwave network throughout Korea for the air force or some military installations.

With that connection he not only knew technology but had made a fortune. I gathered, at the time, that with this kind of money they could back it up. But that was quite different from setting up a semiconductor operation. I thought, this is something I can do by myself, I don’t need anybody else. I’ve done it at Motorola on a smaller scale, I did it all by myself with technicians at Stuart Warner. So at least I can set up a basic unit. From there we can expand.

Goldstein:

All the money was coming from Kim?

Kang:

Yes, from his organization.

Goldstein:

How much money was that?

Kang:

I think it was about five or six million, something like that. Before that there were another couple of talks. When I was at Stuart Warner, when I came up here, we were planning with one of the marketing managers from Motorola, he was no longer with Motorola. We got together and we were discussing whether we can set up a small operation. And these talks didn’t go too far since nobody among us had enough money. We knew we had to bring in some money men. But at least we were thinking of some ventures.

At some point before this came the camcord deal. Park Shung, the father-in-law of one of my Korean high school classmates, was a strong man in the Korean government. He was the president, and an army general. One night he took over power, became the new president, and he was a real money man. You couldn't find anybody bigger than him. So this classmate of mine, or rather his father-in-law, was the biggest money man in Korea. This classmate of mine was living in Boston. I think he came to San Francisco and we met.

All right, so we discussed. Well, the only thing I really knew was semiconductors. I was so fed up with Stuart Warner. I dreaded facing my general manager and saying, “Look, we have this knowledge now. We are capable of running this and this. Let’s go in a more aggressive way.” He was scratching his head.

Goldstein:

Was it obvious how Stuart Warner would apply the things you wanted to do in their gauge business, or was it a new business?

Kang:

No, in a new business.

Goldstein:

Were you suggesting that Stuart Warner get into things like watches, or the like?

Kang:

Yes. Watch chips, this was a new technology. Watch chips could not be made with existing technology. This was CMOS technology or component MOS technology. Low power, that’s new; you didn’t have that before I came in but today we have it. That’s cool.

Goldstein:

Were you suggesting that Stuart Warner really change their whole business, or expand their business?

Kang:

Going into main business. That’s what I was saying, not just me, two times. Well, the boss didn’t have enough appreciation; he didn’t want to stick his neck out since he didn’t know about technology and where it was heading. All right, so it was not only me. My co-workers and the production manager felt the same way. At about the time we were discussing it I met Harry Cho.

Goldstein:

Who is Cho? Where did you meet him?

Kang:

I think San Francisco, or in Sunnyvale. He’s living in Boston.

Goldstein:

He’s the one who’s related to Park. What was the connection there?

Kang:

Harry Cho was my classmate.

Goldstein:

Cho was your classmate and he’s related to Park?

Kang:

Yes. At first I didn’t know it, but he was so enthusiastic about this venture's potential that later I understood why. He was not a technical person but he could control money, or put money in. He could run it, I didn't mind. He said, "OK, I will send somebody to Sunnyvale, we can get started. Why don’t we do some ground work?"

Goldstein:

Did you want to make the plant in the U.S., not in Korea?

Kang:

I wanted it basically in the U.S., because everything I knew was here. Korea was too far, but there were advantages. Maybe we could produce the front end here and the tail end over there, just like the big guys. If we set up the big plant in Korea, then the money could be available here. Not the other way around, since the Korean foreign exchange was so tight.

If it’s a big corporation, only one-tenth of the operation could be brought here. In comes a new fellow named Baeon Won Kim. I think he now runs one of the big companies in Korea. Samsung is part of that conglomerate in Korea and he’s tied to them. But anyway, he stayed with us for a few months. He’s actually a structural engineer, and very sharp. We made some print drawings, and did some preliminary work, mostly technical since nobody had the money to spend on anything else.

Then what happened was that Cho’s father-in-law was pushed out from his powerful position and he was put in jail. He finally gave up all his possessions and came to Boston to recuperate. His money source evaporated overnight. So Won Kim just left. I didn’t know where he went but it was obvious that he went back to Korea. That was one venture.

Goldstein:

Had you already left from Stuart Warner?

Kang:

No.

Goldstein:

So you still had your job at Stuart Warner at least?

Kang:

Yes, right. I could not do this kind of known technical work at Stuart Warner. But I knew that there was a conflict, that company time couldn't be wasted on my own work. Then Mr. Kim from the Kim Corporation came here and expressed the desire to make a commitment. It was all right.

Goldstein:

Did you think about trying to find other sources of capital? Was there an opportunity to do that maybe in Korea?

Kang:

I didn’t have that many contacts. Whatever came my way I just grabbed. I didn’t go out and seek it. If I did that my story might have had a different ending. I already knew them. They knew me, I knew them and it happened naturally, not like here where you are looking all over the place for venture capital. It wasn't like, interview him and interview this guy and present it here, and pick the best one. I didn’t do that. Actually I didn’t know that this is what I could have done. I didn’t have that much knowledge on where to go and seek money. So far it was great, but as soon as I landed in Korea there was all sorts of red tape. Now, I had been expecting a certain degree of red tape. But this went too far. I literally went to a certain office, pounded on the table, and they fought this. I know that certain things have to be dealt with under the table. I was willing to do that. Otherwise the venture would never have succeeded. It was all a one man show there, in terms of equipment and training people. It took almost a year and a half. The rest of the story is as I told you: when we arrived we brought all the equipment. I planned, installed...

Goldstein:

What happened after Park was deposed and the deal fell through?

Kang:

First of all, the plan couldn't be implemented as had been originally planned. There were many reasons. One was that the oil crisis made everything a mess. Equipment delivery was no longer the way I had originally planned. No one could promise that plumbing fixtures, or for instance plastic wares, would be available when you wanted them. I had to go out and buy products at twice or three times the prices originally planned, since I could not wait six months for them. Shipping was canceled, so I had to find another ship. Prices a lot higher. We just used up all the money. And not only money, but stretching. Also, the local chief bureaucrat delayed things.

Goldstein:

Who were your customers for the chips?

Kang:

At that time we were talking watch chips. Our customers were everywhere: Europe, Japan, the U.S.

Goldstein:

You sold to all the different watch makers?

Kang:

Right. They came and waited in your lobby! It was amazing. In the semiconductor business, if you hit the right product, demand can be found all over the world. You don’t have to do any marketing. For the first module of its production you have to have certain volume capabilities.

All I had demonstrated was the first module. If customer wanted ten, I could only provide one. So we had to expand. Firstly, we could not expand since the investors didn’t trust whether we could build or not. So they reserve their money and wait for when it can be done. Once it’s demonstrated it can be done, then you can put in the money and expand. That’s the business. Stuart Warner failed at it. They knew that this market was there but they didn’t have any money. This oil crisis made the entire world crumble for two, three years. That was the period when we were coming out with the products.

Integrated Circuit International, Incorporated

Goldstein:

Were you doing all this while still at Stuart Warner?

Kang:

No, no. I had quit Stuart Warner a long time ago. I had resigned, and I had about a week's gap with nothing to do. At Stuart Warner I didn’t see much of a future unless they put in the new production module. So I thought I had made the right decision and I spent about a one week period readjusting my way of living.

Goldstein:

What was the name of your new company?

Kang:

The company's name in the U.S. was ICII, International- Integrated Circuit International, Incorporated. I had an office in Sunnyvale.

Goldstein:

Where was the manufacturing being done?

Kang:

This was basically the planning office. We couldn't bring the money out of Korea. And the money available here was not enough, so we thought, "Why didn’t we ship the wafer fab over there?"

Goldstein:

Was establishing a plant there that much different than it had been when you did it for Motorola seven years before?

Kang:

Of course. Motorola had a host of expertise available: structures here, plumbing here, air conditioning. But now there was no expertise, I had to do it all by myself. For the air condition units I had to go to Trane. I had to learn all the hookup all by myself. I did the air conditioning training.

Goldstein:

I meant in your relations with the government, or with the landlords, were those negotiations different?

Kang:

No, I didn't do local negotiation. Industry itself the government was promoting. That’s not much of a problem. But the paperwork! The red tape! Well, we could not do everything by the book. For instance, certain things we shipped yesterday are still in the customer warehouse. Officials are still going through the papers and saying, "Okay, you didn’t fill out this line, and we needed this certain date of origin, you have put down three but one is missing..." This sort of thing.

If we packed, for instance, ten different kinds of equipment in one box, everything was okay but for the one unit here you don’t have the right certificate, or you don’t have the right invoice. It was a challenge, but we did it. I think I’m really proud of what I achieved. We were able to produce the chips at the same quality we produced here. There was only about a one year gap.

Goldstein:

But if the country was trying to promote industry, why do you think there was so much red tape? Why didn’t they try to cut it out? Was there corruption at low levels?

Kang:

I think in developing countries there's basically corruption from top to bottom. But it’s like that here, certain things are not politically right so you are not to say it. The same thing applies over there: you may feel it, but you don’t say it. So those are business practices.

Goldstein:

Do you think that has been changing as industries become more mature in Korea? Does that change?

Kang:

I don’t know. I think they have changed a lot. They made lots of money as a nation. I think they work very hard. I admire that they’re hard working people. But politics is behind industry. Just in the last ten, twenty years there has been a lot of corruption in money talks. Personally I don’t have any positive feelings about it. But perhaps if I had made a lot of money I may not have said this.

Goldstein:

You were telling me before that you had these problems, like with supplies, et cetera.

Samsung

You then sold the business to Samsung?

Kang:

Yes, but we came to the point that we could not close the door. We needed additional financing to do organizing. We were the only place that held any interest in high tech for Samsung at that time. They had money and so we talked. All the talking was done by Korean financiers. And their conclusion was that Samsung was willing to buy including Dr. Kang, at [a] certain price. The price was basically, I think, so the financiers got their money back. Not that they were going to profit. They invested, so at least they like to get their money back since Samsung’s willing to pay that.

So at that price they included me, and what they promised is ten percent of the operation. At the end everything was signed, but they just ignored me because they knew there was nothing I could do. I thought that this was a big organization, and it was one of the biggest in Korea. I thought they were going to keep their promises, but they saw the organization as more [Unintelligible Passage] of organization. I hope they changed quite a bit, but that’s the way I felt.

Goldstein:

Did Samsung have a semiconductor production operation already when all this began? Was this their first?

Kang:

They didn’t have anything at that time. Nothing. Nobody had seen even a single wafer.

Goldstein:

By buying this company, were they expanding into that business?

Kang:

Yes, and buying about a hundred trained personnel there.

Goldstein:

How long were you with Samsung?

Kang:

I don't know, about two, three years. Enough to have a manufacturing base. My entire life was there. After that, I lost interest in the business of semiconductors. Well, not lost interest, but it’s that it turned into a nightmare.

Goldstein:

Did you have a contract with Samsung? Were you able to just quit from there?

Kang:

Yes, I just quit.

Goldstein:

You could do that?

Kang:

There was so much conflict there. I gave everything. Well, I don’t know. In a way they created the environment in which you are no longer able to function. All right, so what can you do? Everything is like that, and for your survival, get out. So I got out.

Goldstein:

After you left Samsung and you didn’t want to work in...?

Kang:

Basically they got the semiconductor operation for nothing, in a way. Otherwise they may have had to pay hundreds or [a] hundred million dollars to reach that particular stage.

Goldstein:

They got you from the negotiations. How about the other scientists and researchers, were they the ones that you had hired?

Kang:

Well, they were fresh from school. Actually I devoted myself, trained myself, stayed practically 24 hours a day.

Goldstein:

Were these people from Korea?

Kang:

Well, I brought a couple technicians from there. And one plumber, but they are very skilled plumbers. And one technician that I had in Ohio.

Korean Technical Education and Production

Goldstein:

How was the technical education in Korea at that time?

Kang:

It was good. At least I could communicate with them. The technicians were engineers. Over here, even when fresh from school, we had to train them from the beginning. Particularly in the semiconductor fabrication field. Nobody had had previous experience outside of school. So comparing these engineers I trained in my lab in Phoenix, Arizona, or at Stuart Warner here in Sunnyvale, to Koreans, the Koreans are better since I handpicked them out of [a] larger pool. Koreans are not necessarily any brighter or any better at it. It's that I handpicked the best ones out of a larger pool, compared to here. There is a small pool here, I have no choice but to use whoever comes in.

Goldstein:

Were there many good universities, all strong?

Kang:

I went to Seoul National University and picked the best, the top two or three. I didn’t need that many, only about a half-dozen good, sharp engineers who could follow the instructions and who were willing to learn.

Goldstein:

I have the feeling that these people may have been the first generation of semiconductor engineers in Korea. Is that correct, or were there people who were trained in the ‘60s or even earlier?

Kang:

They were the first generation, the very first. Actually it’s kind of a step-function type of thing. There was nothing before that, and suddenly I broke in and started way up there.

Goldstein:

Do you know how the universities launched programs in these technologies?

Kang:

I don’t know. Only one professor in all. It is not the Seoul University but the other in Korea, there is another science institute in Korea. Shun Ge Kim, I think he’s at Stanford. No, he’s from Columbia University, and he went to Fairchild. And now he’s a faculty member teaching semiconductor technology there. He’s very knowledgeable. They need faculty members, even one or two good faculty members. His students, they are bright, they can [Unintelligible Passage] that they are, makes a significant difference. I had about a half dozen good engineers; they made things different. They're very dedicated.

Goldstein:

You said that the operation in Korea was only one year behind the U.S.?

Kang:

At that time yes. In the products they made.

Goldstein:

In what way was it behind? What was the difference?

Kang:

Well, I had about a one and a half year gap between what I had done at Stuart Warner and the first product they came up with.

Goldstein:

The question is what features did the state-of-the-art have? Was it a higher density on the chips, or you know, lower power consumption?

Kang:

The power consumption. The way of manufacturing it. The device density was more like the function of the pattern. The patterns, or masks, we were buying from the U. S., so it was the same thing. The wafers we were buying from [the] U. S., so the same thing. And the equipment, I had a hundred percent of it purchased from the U.S. because that was the equipment I knew. All right, so it’s [the] same equipment. The only difference is the people running it.

Goldstein:

Maybe they couldn’t achieve yields quite as good as were made out of the U.S.?

Kang:

Regarding the yield, there is always a learning curve. So the U.S.'s yields came up [a] lot faster.

Goldstein:

If you had your company making chips and you’re a little bit behind U.S. companies, how is it that you’re able to sell them successfully? And why aren’t the U.S. companies, or the state-of-the-art companies, whether they're American or not, able to achieve overwhelming market share?

Kang:

Production capabilities.

Goldstein:

You mean they’re limited. They simply have a limit, the demand was too high for any...?

Kang:

Yes. I have an order for a hundred units, OK. I have an order of hundred units and I have only a ten unit parts' supply. I cannot do that. I have to go somewhere to get a hundred parts. So the production capability is key for semiconductors. Just demonstrating is one. The next thing you have to expand it. Samsung did that. Now they are convinced they have the manufacturing capabilities and just expanded it.

Goldstein:

What was Samsung’s main business when all this happened? When they started to get into semiconductors?

Kang:

Well, they are a very diversified company. They were doing all electronics, all the consumer electronics.

Goldstein:

It was, even at that time?

Kang:

Yes. Televisions, for example. Televisions were a major product they were pushing at the time.

Goldstein:

They were already a large multifaceted company?

Kang:

Correct. But all the semiconductors they were using were in-house, bought from Japan. So half of that can be depressed. So you have a small manufacturing capability. After I left, I felt that quite a few of the products went into their own company, since they have a big demand on semiconductors.

Semiconductors in Korea

Goldstein:

You were saying that when you were with Stuart Warner you were talking in Taiwan about setting up a plant. I wondered how that experience was, dealing with the Taiwanese, compared to when you were negotiating a plant in Korea for Motorola? Was it comparable or were there important differences?

Kang:

My work in Korea, for Motorola, was my background work. I had a Motorola-backed company, a U.S. company backing the project. All right so, the Koreans saw me not as an individual, but as part of Motorola. So in context this was on a different level. But when I went by myself for my own company it was an entirely different thing.

Goldstein:

When you started making the watch chips in Korea, were there any other important Korean manufacturers of semiconductors at that time?

Kang:

No.

Goldstein:

Do you think you were the first?

Kang:

We were the first and only one at that time. A lot of people never thought that this was produced in Korea. Korea didn’t have a semiconductor plant.

Goldstein:

So eventually Samsung took it over, and was there a time when the semiconductor manufacturer in Korea became a big business and became established?

Kang:

After I left, I didn’t pay any attention.

Goldstein:

You don’t know?

Kang:

I only saw that the same way you would see something in the trade magazines.

Education and Engineering in Korea

Goldstein:

One other thing I was curious about was that you said that when you were an undergraduate back in the ‘50s, the engineering courses were all power.

Kang:

At least in my education.

Goldstein:

You said that there were courses in communications. Do you remember how that got started?

Kang:

Well, that was a trend and wireless was where it started. It started about thirty, forty years ago, I mean, and that was in 1930s, ‘40s. But in Korea, its basic need, the domestic need, was power distribution. All they were thinking about in electrical engineering was power generation and distribution. Not so much in radio-related technology. So I didn’t get much training in this area.

Goldstein:

At the time you were in school, were there well known, national engineering celebrities? Was there a particular faculty member or textbook that you and all your fellow students knew about?

Kang:

Some people may not agree with me but Korea was sort of a primitive country when I was in school. There were professors who trained in Japan or in other countries. But all we are learning was basically introductory. I had a really hard time catching up when I came to Ohio State.

Goldstein:

Were all the students from a particular social class? Who got to go to college in Korea at that time?

Kang:

No. At least at Seoul National University they picked students based on their grades.

Goldstein:

In high schools?

Kang:

Well there are entrance exams. When I was there it was a hundred percent based on entrance exam[s]. Those who passed it then entered college. Back then most of the Korean families, regardless of their financial status, they were supposed to support their kids. They would even sell their houses to support their kids. So an education was considered of the utmost importance, and could help with success in life.