Oral-History:George F. Smith
About George F. Smith
Smith was born in Franklin, Indiana, in 1922. He went to Caltech and received his BS in 1944. After a stint in the Army working in electronic intelligence, and a year after the war at Engineering Research Associates in St. Paul, he returned to Caltech for graduate school. His thesis advisor was Byron Nichols; his thesis covered the measurement of the thermionic emission properties of a single crystal tungsten (a subject which became irrelevant once transistors replaced vacuum tubes). After he received his PhD, Smith went to Hughes Aircraft Company in 1952. He first worked in the storage tube department, where he made measurements of the secondary emission properties. He then rose to management positions at the Hughes Research Laboratories—he was in charge first of Exploratory Studies, then of many other departments. Most notably, he managed the department which included Ted Maiman, who had just produced the world’s first laser. Smith helped foster applications of Maiman’s laser, including the laser rangefinder, the laser target identification equipment, and other items. Smith also did considerable work on device physics and the development of devices. He supervised Robert Bower, who developed the ion-implanted self-aligned gate MOSFET device. Smith was also involved with work on electron beam lithography, the gallium arsenide field effect transistor, liquid crystals for projection displays and ion engines for NASA. He retired in 1987.
About the Interview
GEORGE F. SMITH:An Interview Conducted by David Morton, IEEE History Center, 4 June 2000
Interview #398 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc., and Rutgers, The State University of New Jersey
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, Rutgers - the State University, 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:
George F. Smith, an oral history conducted in 2000 by David Morton, IEEE History Center, Rutgers University, New Brunswick, NJ, USA.
Interview
Interview: George F. Smith
Interviewer: David Morton
Date: 4 June 2000
Place: Los Angeles, California
Childhood, family, and educational background
Morton:
Where and when were you born?
Smith:
I was born in Franklin, Indiana on May 9th, 1922.
Morton:
Was that in the industrial part of Indiana?
Smith:
No, it is a small town about 20 miles south of Indianapolis. My father was a professor of physics at Franklin College. I had several bouts of pneumonia before I was six years old and the doctors advised that my family to move to a drier climate. My father picked up and moved the entire family to New Mexico on my account and got a job as head of the physics department in the college part of the New Mexico Military Institute.
Morton:
What year did your family move to New Mexico?
Smith:
It was 1929 or 1930. I grew up Roswell, New Mexico. I had to make a choice between engineering and physics at one point in my schooling. In today’s world I would call myself an applied physicist. I went to Caltech. They have established an applied physics organization that reflects the sort of things in which I have been involved, such as electron device physics. I started my education at the New Mexico Military Institute, where I was able to go under favorable conditions that I suspect included no tuition, though I don’t know that for a fact. It was and is a high school and junior college. At one point during the early war years they tried to make it a four-year college, but they gave up and returned to the junior college status. That’s what it is today and that’s what it was when I went there. After taking my freshman year in college at New Mexico Military Institute I was fortunate to get a scholarship at Caltech. I’m sure my father could not have afforded to pay the tuition at Caltech with his teacher’s salary.
Undergraduate studies at Caltech; military service
Morton:
When did you matriculate there?
Smith:
I graduated from high school in 1940 and I enrolled at Caltech in the fall of ‘41. As you know, 1941 was an historic year.
Morton:
That was probably an interesting first semester.
Smith:
Pearl Harbor came along on December the 7th and we were at war.
Morton:
Did any of your friends take off to volunteer?
Smith:
Both the Army and the Navy decided to launch programs. The Army had an ASTP program and the Navy had a B-12 program. Naval people who signed up for the B-12 program were able to continue with their education and finish their bachelor’s degrees. The Army reneged on its program and put people into active duty at the end of the first year I think. I didn’t trust either program and decided to play the odds to stay deferred while I continued my education. That worked fine until the spring of 1944 when the Defense Department decided to cancel all the deferments and draft people. I was told I would be drafted in July of 1944.
Caltech was very gracious. We were on a semester system at Caltech and going to school year round at the time. I was one semester short of finishing everything required for the bachelor’s degree by the end of June in 1944, and Caltech gave me a degree. Having the degree in hand, I applied for a direct commission. I was drafted on the 19th of July and went to the Great Lakes Naval Training Center for boot camp. I had the foresight to take the radio technician exams and qualified to be a candidate for schooling in radio technician work. When I went to boot camp I was a part of that group, along with a lot of other academic people who had similar ideas and many of whom had been teaching or going to school with deferments.
My direct commission came through sometime in August or September while I was at boot camp and I had the pleasure of being able to walk back to the barracks as one of the very few commissioned officers at the Great Lakes Training Center. When I walked into the barracks someone hollered, “Attention!” The next thing I heard was, “Ah, it’s only Smith.” I was assigned to Washington and the commission put me into the organization that eventually did electronic intelligence work.
Morton:
Let’s back up a little. You mentioned you took a radio technician’s test. What did you study in college? How did you learn about electronics?
Smith:
I had a typical Caltech undergraduate education, fairly broad and general with concentrations in physics, chemistry, mathematics and the like. I had gotten into Ham radio in high school, so I understood a little bit about building amateur radio equipment. I got my Ham operator’s license in 1937.
Morton:
Do you remember your call number?
Smith:
Yes. It was W5GSD. I still have an amateur license, K6BYV, Baker-Young-Victor. For a period of time it was required that one get on the air to renew a license something like every five years. I always did whatever it took to renew my license. Now I have a ten-year license and they no longer require getting on the air to renew it. I will probably renew the license when it runs out again. At any rate, I had some amateur radio experience. In high school I went out to the local radio station in my spare time where the engineer in charge of the station would oversee the construction of handmade radios. I built a one-tube radio or two. That was long before the Heath Kit era came along, although I did build several pieces of Heath Kit equipment later on.
The day in which one can build state-of-the-art radio equipment has long since passed. Today one can buy something Japanese-built for a pittance, but back in the day when I got into amateur radio it was a pretty big deal. One could easily build equipment that would be topnotch gear. I had that exposure to that technology, and the fact that my father was a physics professor also enhanced my interest. Exactly what branch of science might have been a problem to be resolved later on, but I was sure I was going to get into either physics or electrical engineering at a pretty young age.
Back to my commission, I was assigned to a group in the Navy that had succeeded in cracking the German enigma code not long before. It had rooms full of equipment spinning all the time. They duplicated the code machine many times over and they were spinning wheels looking for language statistics that would be typical of German. Once they found the language statistics that corresponded to German they would back up to the position that message came through and print out the [inaudible word] language. This Navy organization was a predecessor to the National Security Agency (NSA). The NSA grew out of the organization to which I was assigned, but it probably had other roots as well.
Engineering Research Associates
Smith:
I never doubted that I would go on to do postgraduate work, but before I got around to that I spent a year in St. Paul working for the Engineering Research Associates, which was organized by the Reserve Captain Engstrom, who was in charge of the operation I was assigned to in Washington. The Captain that ran that operation realized that none of his decent engineers would likely want to stay in the Navy or go into civil service, so he had the foresight to organize a little company and get funding for it.
Morton:
Did Engstrom become one of the principals in the company?
Smith:
Yes, he was in charge of the company. He was not the Engstrom from RCA, by the way. He had been a mathematics professor before getting pressed into duty in the Navy. The company was organized in ’46 or ’47. One of the benefits of the company was that each founding engineer was given an opportunity to buy stock in the company. I bought some stock, but at the level of an assistant research engineer or some similar position I was not allowed to buy very much. Eventually Remington-Rand bought out the company and then Remington-Rand merged with Sperry and became Sperry-Rand. The operation in which we were involved became the nucleus for the Univac division of Sperry-Rand. I returned to Caltech for graduate school, and I believe I spent my first summer in graduate school back at Engineering Research Associates. The value of the stock I got in Sperry-Rand increased in value by a factor of something like 300. Unfortunately I didn’t have very much of it. I think I had $100.
Morton:
Not bad in those days.
Smith:
That’s true.
Morton:
Do you have any sense of why that company was in Minneapolis?
Smith:
I think it was in Minneapolis because the source of the money was in that area. The company was actually in St. Paul, but St. Paul and Minneapolis are twin cities. We were near the border of the Mississippi River and Minneapolis was on the other side of the river.
Morton:
What was the source of money?
Smith:
I don’t remember. It was some person or some small company that had some money available. I didn’t concern myself about things like that at the time. It was a fun place to work because the engineers in the organization were able to choose their hours. A group of us there chose our hours in such a fashion that we could get in eighteen holes of golf after work on Wednesdays. I had taken up golf when I was in the Navy in Washington D.C. because we had free access to a very nice country club and very minimal rates to play. A group of fellow engineers and officers golfed on Wednesdays and we played tournaments for trophies on Saturdays. Most of the married fellows – they were all men at that time as women had not yet penetrated the organization – had left their wives in the Washington area and were planning to bring them out once they got settled. At the beginning we had a lot of camaraderie, playing golf and going out for dinner. It was a nice experience.
Morton:
It sounds like it.
Graduate studies at Caltech
Smith:
Although I knew it was desirable to go to a different graduate school than where one does one’s undergraduate work, Caltech was the only school I knew well and I applied to Caltech for graduate work. I went back to Caltech full time in the fall of ’47.
Morton:
Did you work with a specific person there?
Smith:
My thesis advisor was Byron H. Nichols. He was at the University of Michigan and had accepted a position at Caltech, thinking he might like to live in California. I knew generally what the thesis work was going to be from conversations with him, but he didn’t come until I had already gotten started on my thesis work and left before I finished. He stayed at Caltech two or three years at the most.
Morton:
What was the topic of your thesis?
Smith:
My thesis had to do with the measurement of the thermionic emission properties of a single crystal tungsten. This was back in the days before semiconductors took over the world and tubes were still important.
Morton:
It seems like tungsten was already important.
Smith:
Tungsten was a widely used cathode material. Tungsten has the interesting property that it is inclined to grow into crystalline form. Usually a wire of tungsten will form a lot of different crystals oriented in different directions. However there was one production of tungsten wire that happened to grow into single crystals rather easily, and I found that if I used that wire I could grow a single crystal tungsten. I grew a single crystal tungsten and constructed a vacuum tube that allowed me to measure emission in different directions from that single crystal of tungsten.
Morton:
Did it make any difference?
Smith:
Yes. There was much more emission in some directions than others. Essentially the work function was much lower in some directions.
Morton:
Do you know if that resulted in any commercial applications?
Smith:
In 1952 I went to work for Hughes Aircraft Company after finishing my graduate work at Caltech and getting a Ph.D. By that time William Shockley, John Bardeen and Walter Brattain had invented the transistor and the world of vacuum tubes essentially disappeared.
Morton:
You were at Caltech during some crucial years. Was semiconductor physics very quickly incorporated into the curriculum? What was the reaction at that time?
Smith:
I didn’t study much semiconductor physics at Caltech. It had not yet taken over the world of electronics. Robert Noyce and Jack Kilby’s work on the integrated circuit came a little after I left Caltech.
Morton:
Yes. That would have been a few years later.
Smith:
They were the ones that really invented the integrated circuit, and that was the instigator for the revolution.
Morton:
Do you think anyone was doing on semiconductor work when you were there, or do you think that wasn’t for a few more years?
Smith:
It’s pure speculation on my part, but I suspect that there were activities going on in trying to develop some connector devices. I don’t remember the date when Shockley, Brattain and Bardeen did their work.
Morton:
I think the original transistor work was in ’47.
Smith:
Was it that early?
Morton:
They may not have announced it until ’48. It would be easy to understand if people weren’t paying too much attention to it.
Smith:
I’m quite sure that there was a fair amount of emphasis on that area of physics at Bell Labs in those days. How much there was elsewhere I don’t know. I don’t know when Texas Instruments got started.
Hughes Aircraft Company employment; storage tube development
Morton:
You mentioned that you went to work for Hughes right out of school.
Smith:
Yes. I interviewed at RCA, General Electric, Sandia Corporation and Los Alamos, the latter two being the only work that would be reasonably high tech in New Mexico. I had an attraction to New Mexico, but had no hesitation to go to work for Hughes. I was recruited by Ruben Mettler. He left Hughes shortly thereafter, joined Ramo and Wooldridge, and eventually became CEO of TRW. I knew Ruben Mettler mainly as a result of the fact that he and I were housemothers at two different student houses at the same time. Caltech doesn’t have fraternities. Ruben Mettler got his Ph.D. a little before I did. We have remained good friends all these years.
Ramo and Wooldridge had a lot to do with getting Hughes started as a first rate electronics company. When they found that Howard Hughes was not willing to give them any ownership in the company they decided to leave and found their own, which they did with TRW. TRW proved them right. They were wise to leave if they wanted to make a lot of money.
Morton:
What were you hired to do at Hughes?
Smith:
The first job I had at Hughes was in the storage tube department. The other department in that organization was the traveling wave tube department. Both of these activities were initiated by Andrew Hythe [spelling?], who was the director of what became the Research Laboratories – the position I eventually had. Andrew Hythe was interested in traveling wave tubes and display devices. It was felt that the kinds of systems in radar equipment that Hughes was interested in pursuing could use both of those devices. One of the objectives we had in mind was to build a bright display. The existing displays on radar sets used a long persistence phosphor. They were generally faint and faded out rapidly. What we had in mind was to build a storage tube on which an image could be written that would persist and be bright so that a fighter pilot would be able to quickly glance down his display and back up at the view outside without being blinded by the change of light intensity.
Morton:
That’s interesting. I knew that a lot of people were working on storage tubes, but mainly in the context of computing where some people were thinking about memory type devices. Was Hughes interested in that kind of thing at all?
Smith:
Our initial projects were with developing direct view storage tubes for radar displays in mind. I personally made measurements of the secondary emission properties. The way we did this was by depositing insulating material on a wire mesh, using the charge that would be laid down on that insulating material. This was to control the flow of a flooding beam of electrons that would illuminate the entire area of the device and provide a constant bright picture. The flooding beam was controlled by the charge pattern laid down on the insulating layer controlled the amount of electron density that could pass through the screen and onto the final electrode with the storage tube. My first assignment was to try to pick an ideal insulating material to be charged by secondary emission. This was a bit tied back with my thermionic emission experience at Caltech. I ended up essentially developing the material that was subsequently used in the factory to build storage tubes.
Morton:
And manufacturing equipment.
Smith:
Actually the first material that we used was a powdered material that was sprayed on. This may be more detail than you are interested in, but it was a stabled situation where the flooding beam over the entire area could maintain two different potentials. If it were charged up sufficiently it would keep it charged up; and if it were not charged up it wouldn’t affect it, so that the electrons that came through and onto the viewing screen, that phosphor, could be controlled in such a way that an image could be stored essentially permanently. An image could be made bright by acceleration. Applying a few thousand volts at the phosphor screen – which was separated from the storage screen by a fraction of an inch – could keep a picture indefinitely.
Managing research laboratories at Hughes; the first laser
Morton:
What did you go on to do at Hughes after that?
Smith:
I was presented with the option of broadening out and was given the chance to take on the management of a department that called Exploratory Studies. It was a department in which we could do a variety of kinds of projects limited only by our initiative and imagination. I decided that would be a better career path for me than sticking with the storage tubes. The people who stuck with the storage tubes ended up down at the factory as an advanced development organization. I stayed at the Research Laboratories and went on to a broader spectrum of activities. Over subsequent years I headed up several different departments at the Research Laboratories, probably the most noble of which was the department that had just produced the world’s first laser.
Morton:
This must have been in the late fifties.
Smith:
By the time I was moving from one department to another we were out at Malibu. The way things worked at the Culver City plant, unless we had a high priority project we got squeezed for space. It became very clear to those of us involved in the management of the Research Laboratories that we needed our own home and one that could be maintained in spite of the pressures of the hot projects. Thus it was decided to move the Research Laboratories out to Malibu. That move was made in 1960. After thinking about it a while and buying a piece of land, I found that it was not that hard to commute to Malibu against the traffic. In 1960 Malibu did not have a junior high or high school, and all the students to come in to Santa Monica for their schooling. We had two kids about to enter junior high, so we decided to stay here and I started commuting. It’s twenty miles to the laboratories now. The laboratories are adjacent to Pepperdine on Malibu Canyon Road.
Morton:
You mentioned the laser demonstration. Was that in 1960?
Smith:
Yes. It was unfortunate that we were making the move just about the time that Ted Maiman was getting ready to build the first laser. Our move disrupted his research.
Morton:
Do you think it would have otherwise been a little bit sooner?
Smith:
It might have been a little bit sooner, but I don’t think we slowed him down very much. He was very dedicated and was working very hard. At an International Quantum Electronics Conference Art Schawlow had predicted that a laser made out of pink ruby would never work.
Morton:
Which is of course what Maiman did.
Smith:
And that’s what Maiman did. He was successful because he didn’t believe a measurement that had been made back at Westinghouse by another investigator who had concluded that the quantum efficiency for the laser transition was very low. He measured it for himself and found that it was really 70 percent, not 1 percent. That made all the difference in the world as far as the practicality of the ruby laser was concerned. Art Schawlow had gone on record at this International Quantum Electronics Conference saying that it would not be possible, insuring that he got a good patent.
Morton:
I guess you can’t be right about everything.
Smith:
You can’t be right about everything. Ted Maiman had already been working on ruby masers and had a lot of experience with the material. He figured that the quantum efficiency was probably high. And he was right.
Hughes research and business strategies; interactions with University and corporate labs
Morton:
Let’s back up again. It would be really helpful if you would give some sense of the dynamic between Hughes and places doing similar research, such as Bell Labs and the various universities in California – Stanford and so on. It is clear that ideas were being circulated around.
Smith:
Yes, they were.
Morton:
I’d like to get a sense of how that was possible, particularly in a commercial environment and in a company like Hughes that didn’t have a special position like AT&T. What was the big picture at Hughes? Were they just trying out random things or was there a big scheme behind it all?
Smith:
The big scheme at Hughes, if there was one, had to do with the development of military systems. Hughes Aircraft Company – which has always been a misnomer because it really has always been electronics – decided early on not to bother changing the name because that would confuse the customer. The true big picture was to do innovation. Ramo and Wooldridge managed to collect quite a number of good innovators, and they were determined to build superior military systems – missiles, radars, fire control systems and the like. That was long before satellites came on the scene. Direct TV is now the biggest project at Hughes Electronics. Hughes is now in the process of selling off the manufacturing of satellites to Boeing. I don’t know how long it’s going to take, but eventually Hughes will no longer be making things but will be providing satellite services. There’s a premium on earnings, now that Howard Hughes is dead and General Motors owns the company, and moderately short-term earnings. I think that is unfortunate in many respects. For many years we were able to not worry about earnings and take the long view. That is no longer possible with Hughes Electronics. The defense business has already been sold off to Raytheon, the manufacturing of spacecraft is being sold off to Boeing, and Hughes is turning into a different company altogether.
Morton:
It sounds like it. I didn’t realize that was happening.
Smith:
Research Laboratories is now called HRL Incorporated and is jointly owned by Raytheon and Hughes Electronics. Half of the members of their board of directors are from Raytheon and the other half from Hughes – which is now a subsidiary of General Motors. I consider that to be an unstable situation and I don’t think it’s going to last. On the other hand, it’s gone better and lasted longer than I expected thus far. You were asking what was the motivating force behind the original thrust of the Hughes Aircraft Company. The original thrust of Hughes Aircraft was to develop and polish the systems engineering end of things and build military systems.
Morton:
Do they have a sort of “not invented here”?
Smith:
The “not invented here” syndrome it just about everywhere I guess. I’m quite sure that there was a lot of communication back and forth between the various investigators who were pursuing the quantum electronics field.
What I am about to say is something that I don’t think ought to come into a formal transcript.
Morton:
Do you want me to push the button?
Smith:
I don’t mind. I’m just saying I will edit it out.
Morton:
You can edit it out later. Whatever you want.
Smith:
Ted Maiman was very possessive of the laser idea and didn’t want to share it with anybody. I was running an exploratory applications oriented department at the time and he didn’t even want to give me a laser to play with when I was trying to build a laser rangefinder. I did however manage to get one from his technician.
Applications for lasers; Q-switch laser and argon ion laser development
Morton:
That’s an interesting story. It leads me to a question I want to ask. What little I know about the history of the laser comes from the Bell Labs side of things. One never gets a sense from reading what those people say in their memoirs that Bell Labs had any application in particular in mind for the laser. However it sounds like you had had an application ready to go.
Smith:
One of the first things I did at Hughes was to take a look at the various possible applications of lasers to figure out which ones would suit the company best. By far and away the most obvious was the laser rangefinder. Building laser rangefinders, laser target identification equipment and the like turned into a big deal – a $100 million operation in the factory.
Morton:
Was that device a long way from manufacturing at the time?
Smith:
It was a long way from manufacturing at the time. Basically this is because when Ted Maiman made his first laser the Q-switch laser did not yet exist. The Q-switch laser is essential to the practical implementation of a laser rangefinder. The normal ruby laser produces not one big pulse but a string of smaller pulses. It burps many times as it is producing its laser action. That is not a practical way to build a laser rangefinder. I ran the activity to build the first laser rangefinder and I did that with a scheme in which we would record the pulse that went out with all of its burps and wiggles and correlate that with the pulse that came back with all of its burps and wiggles. That’s not a practical way to build a device. We had to record both of these pulses – outgoing and returning – and then slide them back and forth until we got a good match. That didn’t give the needed information in a hurry. It did however demonstrate the feasibility of the idea.
The chromaticity, the monochromatic nature of the laser light, allowed operation in broad daylight. That’s an advantage of course. Ordinary sunlight is not going to hinder operation. However for a practical device the Q-switch invention was needed. We made that too. There are a whole string of inventions that came out of our laboratories. I did not personally have anything to do with the development of the first laser. Ted Mayman worked for a different department manager and really pretty much did it all on his own. He built the first laser by himself. I was responsible for the activities we conducted out of Malibu that had to do with the development of the Q-switch laser, the argon ion laser and a whole spectrum of laser activities. It was kind of interesting. Everybody wanted to jump on the bandwagon and work on lasers. We didn’t have any difficulty persuading people to join the fray.
Morton:
What were the reasons for developing these other kinds of lasers?
Smith:
From the beginning it was pretty clear that the ruby laser would have fairly limited applications. In fact, the laser rangefinders being manufactured today are yag lasers. They are pulse lasers and use Q-switching, but they are not ruby lasers. I think the idea was that different kinds of lasers stood a better chance of being able to accomplish different functions. And among other things, it’s hard to make a CW ruby laser. We built one once and it’s not a very practical device. However some of gas lasers are very good at CW operation. The semiconductor lasers have probably had more impact on mankind in general than any of the others. We were not very involved in semiconductor lasers.
Communications was big, and even now communications is one of the biggest applications. The semiconductor laser is probably the best one to combine with fiber optics for transoceanic competition with the satellites. There are other CW lasers that are used for eye surgery. I’m not an expert on that subject, but I believe they are solid-state lasers and pulsed lasers. That’s my impression anyway.
Device physics research and development
Morton:
What were some of the other big device technologies in which you were involved in those years?
Smith:
The majority of the work at the Hughes Research Laboratories when I was there was in device physics and the development of devices, although today there is a much stronger emphasis on software, programming, and that sort of thing. We did pioneering work on understanding ion implantation and building equipment to do ion implantation on devices. In fact, a guy who was working for me developed the self-aligned gate MOS device. I don’t know whether you are acquainted with that technology, but essentially what the self-aligned gate MOS device did was to obsolete bipolar technology for transistors. It is much simpler to build than the bipolar device, and the MOS device technology works as fast and with as little loss as the bipolar. Before that time the parasitic affects in the MOS device had limited the speed with which they would operate so that a bipolar device was needed for speed. Today ion implantation is used to line up precisely with the dopant. Using the gate as a mask, the doped area can be precisely aligned to be immediately adjacent to the gate. Thereby the parasitic capacitants can be reduced dramatically and make the device much faster.
Morton:
Who worked on that?
Smith:
Robert (Bob) Bower. He won a National Inventors Hall of Fame Award for inventing the ion implanted gate self-aligned MOS device. This invention is being incorporated into virtually all modern integrated circuits built today.
Morton:
When was this invention?
Smith:
I think it was in the late sixties.
Morton:
Why did it take so long for it to become incorporated into integrated circuits?
Smith:
It has been incorporated for some time now. Exactly when it was generally acknowledged and the bipolar technology became passé I don’t know.
An interesting aside is that Hughes Aircraft Company was at one point a predominant supplier of passenger entertainment systems for airlines. Our friends down at Newport Beach, to which organization we wanted to transmit this technology, were very resistant to it. It was not invented there and they didn’t want to mess with it. For them it was a complication. They signed up for a fixed price contract to deliver MOS technology that they could not build because it wasn’t fast enough. We were able to show them that an ion-implanted device would be fast enough, and instantly they accepted the technology. Thus I was able to move the ion implantation technology into the Newport Beach factory and they in turn were able to satisfy their customers on a fixed price contract.
Manufacturing at Hughes; integrated circuits
Morton:
Was Hughes a major manufacturer of integrated circuits or did they manufacture things that were more of a specialty?
Smith:
I couldn’t say that they were ever a major manufacturer of integrated circuits. They had an operation that was trying to become a major supplier of integrated circuits, but it was not very successful – for a variety of reasons not necessarily related to lack of technology. At one point Hughes was a prime supplier of watch circuits for digital watches, but Texas Instruments and Japanese companies beat them hands down when they got into mass production.
Morton:
Do you remember the pulsar, one of the earliest commercially available electronic watches? It was Japanese and had very distinctive styling. There is a reissue of them now by another company. They are not as expensive as the originals but still pricey.
Smith:
The very first digital watches used electroluminescent displays. Only when they got liquid crystal displays did they stand a chance of having a decent battery life. In the beginning Hughes was very competitive in this area, thanks largely to Bob Bauer’s invention of the ion-implanted device. However they fell by the wayside in the long run. I don’t think even TI is a major supplier of watches anymore.
Collaboration between Hughes and academic labs on ion implantation
Morton:
I don’t think anyone was interested once they went down to two dollars a watch. I spoke to somebody last week up at Stanford who did some work on ion implantation. Was there any relationship with the Stanford folks on that technology? I know it was done at multiple places.
Smith:
The man who headed up the ion implantation activity at Hughes became a professor at Caltech. I don’t remember what connections he had.
Morton:
Who was that?
Smith:
Jim Maher. He subsequently went back to Cornell and most recently has gone to Arizona State. Jim Maher was a real driving force. He was interested in understanding how ion implantation worked and I’m quite sure that he shared a lot of his know-how with a lot of his contacts in the academic world.
Comparison of academic and commercial research approaches
Morton:
Was that a problem for the company, finding people who would work on problems that could be commercialized rather than those who were primarily interested in finding out how things worked? There is an obvious difference between the academic and commercial sectors.
Smith:
I don’t think that was a big problem, though there were some people who were better suited to the academic world than industry. The man who invented the ion laser is now a professor at Caltech. The man who invented the Q-switch laser is now a professor at USC. We’ve lost a number of people to the academic world, though I never regretted it all that much. I think that if a person is happier trying to understand things than trying to build things, that’s okay. He belongs in a university. I don’t have any problem with that. There are lots of different kinds of people out there. We’ve never had any great difficulty finding people who are interested in pursuing applications. It’s an honorable profession.
Morton:
It’s got to be tough for most companies to run a research organization academic enough in its atmosphere to keep those kinds of people.
Smith:
Sometimes it takes a while for a particular investigator to decide what he really wants and likes to do best. Again off the record: The man who went over to Caltech as a professor was in charge of our laser department for a while, and he was a lousy manager. But he’s a very good academic person. He is well placed over at Caltech.
Morton:
I think that’s common. I hear all the time about engineers that making the transition to management and loving it and others who never quite like it.
Smith:
Some of them are just plain not very good at it, even if they are trying.
Projects managed at Hughes, research achievements
Morton:
Getting back to this period of time when you were at the management level in the research part of Hughes, what were some other shining examples?
Smith:
One other area in which we devoted a lot of energy and spent a lot of money was electron beam lithography. We were one of the pioneers in developing methods of ultra small devices using electron beam lithography. One of the big success stories is a galley marsinide field effect transistor, which is at the core of the receivers used for satellite communications. The receiver onboard the satellite, using just one of these galley marsinide FET devices, is able to thereby provide a lot more bandwidth or gain could normally be gotten out of a conventional device. I think there is in place at the laboratories in Malibu today a little factory that specializes in producing these devices. It is successful even if it only produces a handful because only one is needed per satellite.
Morton:
Is this a single stage?
Smith:
Yes. It is the first stage preamplifier for a very high frequency microwave signal.
Morton:
That’s remarkable. It must have quite a lot of gain.
Smith:
It has a lot of gain and has a good gain bandwidth product that has become an essential device in the system, even though the yield can be pretty low. Not all that many can be gotten out of each wafer of material.
Morton:
Why is that?
Smith:
They’re hard to make. Anything that’s hard to make is likely to have a modest yield at best. On the other hand, only a handful is needed since one per satellite is all that is needed. These are crucial to some of the classified military satellites.
Morton:
You mentioned some devices I’m assuming were related to switching applications. Did Hughes get into microprocessors at all when they came along?
Smith:
I retired in 1987. By that time they had not gotten into that very much. They may have gotten into them since then.
Morton:
Were they manufacturing memory chips and/or other integrated circuits that could be used in computing or digital devices?
Smith:
There was not a major emphasis in that area. Another major development at Hughes Research Laboratories I would like to mention is in the field of liquid crystals for projection displays. We were able to build a liquid crystal device that could be put into a projector so that that light could be transmitted onto a big screen through the light valve. We were very successful in building situation and radar displays for command and control operations. At the time I left they had not yet succeeded in reliably being able to get fast enough response in the liquid crystal medium to show real television, but they were very good. We worked with our friends over in the Fullerton part of the operation building light valves to be used in projectors or command and control centers.
Morton:
Some of them are consumer products now, of course.
Smith:
I think progress has been made in speeding them up so they can produce live television now.
Morton:
Unless I’m mistaken, I think I have seen them advertised.
Smith:
You’re probably right.
Research and management challenges at Hughes
Morton:
What about failures? Were there any particularly vexing flops during your tenure at Hughes that you want to talk about?
Smith:
There was a period of time when diversification was the goal and we spent some time trying to figure out how to do that with the background we had. Usually we tried to diversify in different applications using some of the techniques that we had developed for military purposes. My recollection is that that was largely a wasted activity. We took a little bit of a flier at what turned out to be the technique that is used in supermarkets today to read and enter barcodes for purchases. We had some pretty good technology, but I think that to be really successful the best tack to take is to decide what field to work in and then invest in it fairly heavily. In my experience, a sort of general purpose “let’s diversify” kind of approach has been largely unsuccessful.
Let me mention another story. For a number of years we got a lot of support from NASA in trying to develop electric propulsion. It finally became a success, and there are ion engines onboard existing satellites for purposes of station keeping and the like. It took a lot longer to come to pass than any of us expected.
Morton:
Was this due to technical reasons, political or inertia?
Smith:
There were a variety of reasons it took so long. Some aspects of it that were not invented here, and there was some politics involved. For example, the first ion engines we built used mercury. There was a great worry about distributing mercury, even in minute amounts, above the atmosphere. I don’t think that was warranted, but there was this sentiment of “Don’t dare spew out mercury into the upper atmosphere.” Today’s ion engines all use xenon, which is not a heavy metal and is harmless. There are no worries about poisoning anyone with it. Xenon is very expensive, but not much is needed so it finally came to pass. I don’t fully understand all the reasons it took so long. For the most part we did that with NASA money and it wasn’t costing us out of our own pocket so it was okay to go ahead and do the work even though its payoff seemed always to be ahead of us. Part of the company that is going to be sold to Boeing is employing electric propulsion for station keeping purposes on their satellites.
Morton:
It finally paid off.
Smith:
It finally did. The advantages of course are that the ion engine has an extremely high exhaust velocity and doesn’t take much fuel. It doesn’t take much material to be discharged and give a long life, providing there is enough thrust. The amount of thrust that can be gotten from an ion engine is small.
Morton:
Yes, it’s very small.
Smith:
However it can be run continuously for quite a while. There have been suggestions that it would be a good way to do an asteroid rendezvous. Once the rocket has been launched and gone out of the earth’s gravity field the ion engines could be turned on and it could chug along. It would take months or years to get anywhere however. The ion engine is being used in a very practical way right now for station keeping.
