Oral-History:Roger Webster

From ETHW

About Roger Webster

Born in 1920, Roger Webster developed interests in electronics and ham radio as a youth. After receiving a University of California electrical engineering degree in 1943, Webster conducted war-related research at Harvard's Radio Research Laboratory and at Caltech, where his work on an altitude sensor for bomb detonation contributed to the Manhattan Project. After World War II, Webster took employment with the Stancil-Hoffman Corporation, a Hollywood, California, manufacturer of tape recorders. He began working at Texas Instruments (then General Instruments Incorporated) in October 1951. During his first three years with the corporation, Webster worked on aspects of the magnetic airborne detector, a device used by the Navy to detect submarines. His radio technology experience on the Manhattan Project led to Webster's assignment to TI's Regency radio project in 1954.

The bulk of this interview concerns Webster's work on the Regency radio. Webster details the early stages of this program, including his work to produce a practical transistorized radio design. He recalls the challenges of working with germanium transistors, resulting in the use of silicon transistors to develop a successful five-transistor radio. The high cost of silicon necessitated a return to germanium transistors and use of a six-transistor radio model. Innovations including the autodyne converter ultimately allowed further design streamlining, and TI transistor production technology proved essential to the success of the Regency radio.

Webster describes TI's rationale for collaborating with I.D.E.A. to manufacture and market radios. After the entry of I.D.E.A., Webster worked mainly on the crystal-growing process and on transistor evaluation. Following the successful development of the four-transistor radio model, Webster shifted his focus to other radio projects, including circuit improvement and car radio development. He later worked on transistor production for TI's contract with IBM and designed a transistorized television set.

In the conclusion of this interview, Webster considers the significance of the Regency radio, detailing its role in stimulating the semiconductor business. The technological success of TI's transistor development for the radio led to a new focus on transistor contracts with radio manufacturers and with IBM. Webster feels that the I.D.E.A. company did not pursue the full marketing potential of the Regency radio.

For further discussion of the Regency radio and other transistor radio products, see Willis Adcock Oral History, Gordon K. Teal Oral History, Norman B. Krim Oral History (1984), and Susumo Yoshida Oral History.

About the Interview

ROGER WEBSTER: An Interview Conducted by Michael Wolff, IEEE History Center, 5 March 1985

Interview #461 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, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. 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:

Roger Webster, an oral history conducted in 1985 by Michael Wolff, IEEE History Center, Piscataway, NJ, USA.

Interview

Interview: Roger Webster

Interviewer: Michael Wolff

Date: 5 March 1985

Place: Texas Instruments, Semiconductor Division in Dallas, Texas

Texas Instruments career

Wolff:

We're talking about the Regency radio. What is your position here now?

Webster:

I'm responsible for some special millimeter-wave microwave diode design and manufacture in the Semiconductor Division of Texas Instruments.

Wolff:

When did you come to TI?

Webster:

I came to TI in October of 1951.

Childhood electronics and radio interests

Wolff:

Tell me a little about your background and how you got into electronics.

Webster:

I got into electronics very early in life as a ham radio operator when I was 13 or 14 years old.

Wolff:

Where was that?

Webster:

That was in Long Beach, California.

Wolff:

When were you born?

Webster:

I was born in 1920 in California. I'm a native Californian who immigrated to Texas.

Wolff:

When did you come to Texas?

Webster:

I came to Texas in 1951 when I went to work for TI. Being a ham radio operator as a boy really sparked a lifelong interest in electronics and everything of an electrical nature.

Education and academic employment; Manhattan Project

Webster:

I graduated from the University of California with a degree in electrical engineering in 1943. And after graduated I worked at the Radio Research Laboratory at Harvard during the war. During the latter part of the war I went back to California and worked at Caltech for about a year. Caltech had part of a program that was in the Manhattan Project, and I worked there for about a year and a half.

Wolff:

You were involved in the Manhattan Project?

Webster:

Yes.

Wolff:

What did you work on for the Manhattan Project?

Webster:

We worked on the altitude sensor that was part of the control mechanism that was to detonate the bomb.

Stancil-Hoffman Corporation

Wolff:

I see. And then from that you came to TI?

Webster:

No. One of the people I met there set up his own business after the war was over, first as a distributor for Brush tape recorders. Then he wanted to get into the manufacture of specialty tape recorders, particularly for the movie and broadcast industry. He talked me into coming to work for him. At that time it was called the Stancil-Hoffman Corporation. We made all kinds of specialty tape recorders at that time from miniature battery-operated recorders. I don't know whether you remember Lowell Thomas' journey to Tibet somewhere in the late '40s. I think it was Tibet. He broke his leg on the journey. He took several of our tape recorders with him, and they flew back the tapes of his description of his journey over there.

Wolff:

Where was the company located?

Webster:

It was located in Hollywood. I worked there for about four years.

General Instruments Inc./ Texas Instruments Inc.

Webster:

I had married a Texas gal, and she talked me into trying to find a job in Texas. She was homesick. We came down here and I interviewed with what was then called General Instruments Incorporated.

Wolff:

Or Geophysical Instruments.

Webster:

No. For a brief period of time Texas Instruments Incorporated was known as General Instruments Incorporated. That was the name of the company when I came to work for it. General Instruments Corporation sued them over the infringement of the copyrighted name, and within six months or so after I came to work here that they changed their name to Texas Instruments Incorporated. The geophysical part had been set up as a wholly owned subsidiary.

Wolff:

I see. Who hired you at TI and what did you start off doing?

Webster:

I was hired by Bob Olson who was chief engineer at the time. I remember very well the letter I got offering me the position. It came from Mark Shepherd, who was Bob Olson's administrative assistant at that time.

Magnetic Airborne Detector Program

Wolff:

Let's go from what you started doing there to the assignment to design the breadboards for what became the Regency radio. What did you start off doing there and how did you make that transition?

Webster:

Perhaps because there was some remote connection between magnetic tape recorders and the magnetic airborne detector – at least they both had the word magnetic in them – I was assigned to the Magnetic Airborne Detector Program when I came to Texas Instruments. I worked on various adjuncts of that for three years or so.

Wolff:

Was that for oil prospecting?

Webster:

No. The magnetic airborne detector was something that the Navy used for submarine detection. It was carried in a stinger on the back end of a Lockheed P2V I think. It was a great big long stinger that they built onto the tail end of it. There have been some rather striking photographs made of that thing sticking way out behind the tail. The magnetometer was towed behind the aircraft in that housing. That program was headed by Paul Davis. I worked for Paul for several years when I first came to Texas Instruments. Prior to coming to TI Paul had been involved with a small local radio manufacturer. TI got interested in promoting transistors in a portable radio, which particularly Pat Haggerty saw as an opening wedge. They knew of Paul's background in this area and called him in to work on that. I had ham radio experience building receivers and transmitters and things as a kid and I had worked on RF projects at Radio Research Laboratory.

Webster:

When I worked on the altitude sensor on the Manhattan Project I was one of the few with RF experience at TI at that time, so I got assigned to the Regency radio program. I think that was early in 1954.

Family background; childhood radio interest

Wolff:

Okay. Before we go back to that, tell me a little about your parents. What did they do?

Webster:

My father worked for the Security First National Bank in Los Angeles for most of his life. He was a branch manager in one of the many hundreds of branches they had. My mother was a housewife. She was a native of West Virginia and my father was born in Lebo, Kansas. They met in the Imperial Valley in California in the very early 1900s.

Wolff:

Your interest in ham radio does not seem to have been inspired by anything that your father did.

Webster:


Audio File
MP3 Audio
(461 - webster - clip 1.mp3)


No. It was a neighbor who worked for the phone company. He was a young man at that time. In fact I have remained lifelong friends with him even though I haven't seen him in thirty or forty years. We still talk on the phone fairly often. He was interested in anything of an electrical or electronic nature. He helped me get started by building a crystal set, which is how I got started.

Regency radio program

Wolff:

Good. Coming back to your working for Davis on the Magnetic Airborne Detection Program and then it was decided that they wanted to build a miniature radio. Do you recall when you were first told about it and who told you, what the nature of the assignment was and what they asked you to do?

Webster:

It would have been early in 1954, perhaps early spring. A rough breadboard had already been built by someone, but it was a first kind of feasibility attempt. I don't remember who built it or why they didn't stay on the program. However it had quite a few transistors in it and since the transistors were going to be the dominant cost mechanism in the thing it had to be greatly simplified. I guess that first breadboard demonstrated that TI had some transistors that could be made to work. What had to be done was to come up with something of a practical nature. It was not really a practical design.

Wolff:

Was that your assignment?

Webster:

Yes.

Wolff:

Do you remember who told you to report to Davis?

Webster:

That was Paul Davis.

Wolff:

And he said that he wanted you to take this design and make it practical.

Webster:

More or less.

Wolff:

Or improve it I guess.

Webster:

We never really used the design that was done because it was a demonstration and it did work and that was about it.

Wolff:

Do you recall whose inspiration it was to do this in the first place?

Webster:

As far as I know it was Pat Haggerty. Buddy Harris can fill you in better on that.

Wolff:

I'll ask him about that. Your assignment was basically to make a transistorized radio design.

Webster:

That's right.

Wolff:

I understand that you designed two breadboards – one with silicon and one with germanium transistors.

Webster:

That's right. We struggled with the germanium transistors, and it was difficult to get an operating model that had sufficient sensitivity. My recollection is that their goal at that time was no more than five transistors in it, and we were having a very difficult time getting a transistor radio that would perform in a satisfactory way with five transistors. I think that first breadboard had eight or nine transistors in it.

Wolff:

You mean the earlier one?

Webster:

Yes, that first demonstration model. I don't remember for sure, but that's my recollection that it had on the order of eight or nine transistors in it.

Wolff:

It's my understanding that you were able to come up with a design that had six transistors.

Webster:

That's correct. The way the silicon transistor got involved in it was that we were having a very difficult time making the germanium transistors perform satisfactorily with six transistors. I got hold of some silicon transistors that were being developed at that time, and because of the much narrower base region on the silicon transistors their high-frequency gain was much higher than the available germanium transistors at that time. We were able to make a five-transistor radio with silicon that played up a storm. Everyone was quite impressed with that, but that would not cut the mustard because of the cost of manufacture of the silicon transistors.

Wolff:

What was the cost? Do you remember?

Webster:

Didn’t they sell for perhaps a hundred dollars apiece when they first came out or something of that nature? I remember Pat Haggerty said when he came and looked at it, "Well Roger, maybe someday silicon, but not now." That was in 1954. We just set the silicon transistor radio aside and went back to germanium.

Wolff:

About how long did it take you to do that six-transistor design? Do you remember? It must have been a matter of months.

Webster:

It was no more than a month or two. Of course we had the advantage of knowing something about radios, plus the breadboard that was already there.

Wolff:

Was the design largely yours?

Webster:

The design was largely mine, yes, but I had help and others contributed to it.

I.D.E.A. radio manufacturing

Wolff:

Okay. When did you first hear that the I.D.E.A. people were going to come into the act?

Webster:

I don't recall specifically, but it was fairly early. It was not long after we had a working model of the germanium transistor radio. We finally succeeded in making a six-transistor model that worked very well, and it was probably at about that time.

Wolff:

When you first hear that I.D.E.A. was going to be involved? Did anyone say why they were bringing in these people?

Webster:

Yes. It was because I.D.E.A. already had a line of radios that they were marketing and they had both manufacturing expertise and a distributor network set up. TI did not have the manufacturing expertise in the radio field and had no type of a distribution setup. That is my recollection of why they contacted I.D.E.A. to manufacture the radios.

Wolff:

Did you have any particular difficulties in doing your breadboard before they brought in I.D.EA.?

Webster:

Yes. The main problem we had was that the germanium RF transistors available at that time were low gain and the sets had very low sensitivity.

Wolff:

About what was the gain?

Webster:

Probably at the IF frequency which we chose at 255 because of the gain problem on some of those early transistors I'm sure we had difficulty getting as much as 20 or 22 dB gain.

Wolff:

The radio was not very powerful.

Webster:

It wasn't very sensitive. It would pick up the most powerful local stations and that was about it.

Wolff:

Okay. However you were able to get a working radio.

Webster:

Yes, we had a working radio by selection of transistors with gain. We culled through lots of transistors and picked those that had sufficient gain and got a pretty good working model with six transistors in the radio. Later with additional work we got that down to five transistors. That was the best that we did.

Four-transistor radio; autodyne converter

Webster:

One of the great things that Dick Koch did was to develop that autodyne converter. That's what gave it the final configuration of four transistors.

Wolff:

Koch's recollection is that he and his boss came down to TI around the July Fourth weekend, probably July 2nd, and met with you and Pat Haggerty. His recollection was that you showed them your breadboard with six transistors. He said that TI wanted something portable and very small that would fit in a shirt pocket.

Webster:

That was Pat Haggerty. Interestingly enough, at that time Emerson had manufactured a small battery-operated vacuum tube radio. We bought that and gutted it and used the tuning capacitor. That's probably about all we used from that, but we built our little six-transistor radio in that Emerson case. It made a very presentable little package.

Wolff:

You used that tuning condenser?

Webster:

Yes.

Wolff:

This is what you showed the I.D.E.A. people?

Webster:

I believe that's what we showed them. I can't be positive, but I think we did. I know we built it, and it would have been about that time.

Wolff:

Okay. Koch recalls that you showed them the schematic and he said right away, "Okay, let's see if we can replace the detector with a diode."

Webster:

Yes.

Wolff:

Dick says that you and he went down to your lab and he took it out and replaced it with a diode and the performance was the same. "So we went back to the room and said, 'Now we have five transistors'."

Webster:

That could well be, and frankly I've forgotten.

Wolff:

That's all right. His recollection was that then they took it back and came up with a converter.

Webster:

Dick came up with the autodyne converter. That's correct.

Wolff:

I guess you got to know Dick Koch pretty well.

Webster:

Yes. There were a lot of dealings with Dick at that time.

Wolff:

What was he like?

Webster:

I always thought he was quite a brilliant individual. In a way, after I got to know him, I was surprised that he stayed with an outfit like Regency. Eventually he left them of course. I met him a few years ago and we had quite a talk. I understand he still does consulting for I.D.E.A.

TI development strategy, collaborations with Regency

Wolff:

Okay. Is this correct to your recollection that TI was considering building a radio to market themselves but they figured it would cost them about hundred dollars at retail and they wanted something much cheaper?

Webster:

I don't know about the cost because I never got involved with that part of it, but yes, TI did consider building it. I am sure of that. There were quite a few discussions about that. There were some discussions to which I was a party. I think cost was a factor. Other factors were the lack of experience in this field plus the fact that TI at that time had no consumer-type contacts in the electronic industry and would have had to establish some kind of distribution setup. I think that would have been time-consuming and TI was driven by a desire to get something going that year. I don't think they felt they could do that by themselves. Buddy was involved a lot more at that level than I and can fill you in much better on that part of it than I.

Wolff:

Okay. Tell me about Mark Shepherd and Pat Haggerty's role in all of this. What do you recall?

Webster:

I recall Haggerty very clearly because he was down there virtually on a daily basis wanting to know how it was going and what we were doing.

Wolff:

Was this even before you brought in I.D.E.A.?

Webster:

Yes. Before, during and after. There was still a lot of work that went on after I.D.E.A. was brought into the project. The big problem then was to get a really suitable transistor, because all the transistors we had at that time were still pretty well hand-selected transistors.

Wolff:

Haggerty was deeply involved and concerned and down there all the time monitoring it?

Webster:

Yes, yes.

TI transistor production technology

Wolff:

In the Regency TR-1 brochure it said, "This accomplishment has been made possible by close cooperation between Regency and TI whose unique transistor production techniques made possible mass production at a reasonable cost." What were these unique production techniques?

Webster:

That was the double-dope grown-junction germanium transistor. Gordon Teal pretty well brought that technology to TI. At the time that was probably about the only practical way of getting sufficient gain in quantity at the radio frequencies. There had been radios made with alloy-junction transistors but the control of the base on the alloy junction was so poor that it was much more a matter of luck than anything else when they got a device that would work. Whereas with the grown-junction transistor the control of the base width was a lot more technique-oriented than it was with the alloy junction at that time and the techniques were under better control just by virtue of the nature in which it was made. Time, temperature and pull rate pretty much determined the base width of the transistor.

Wolff:

What do you mean when you say it was more technique-oriented?

Webster:

I should have said the technique was more controllable.

Wolff:

For production.

Webster:

Yes.

Wolff:

Was this production under the direction of Mark Shepherd?

Webster:

Yes.

Wolff:

I was told that he's too busy to be interviewed but I should send some written questions and I would get a response. What do you think would be good to ask him about that would be revealing?

Webster:


Audio File
MP3 Audio
(461 - webster - clip 2.mp3)


The trials and tribulations of pulling the crystal with which they could fabricate suitable transistors. After we got the original model of the radio working, the next big problem was to pull crystals, because that part of the process was what virtually controlled the final performance of the transistors. At that time we set up a fairly quick feedback scheme whereby we could fabricate and measure transistors. We were looking for transistors that were 262 Kc and would have an average gain of 30 dB. I will never forget the guy that was growing crystals. I've forgotten his name now. He came in one day with a Bob Pritchard article, which in essence showed that the maximum achievable gain on a transistor built like that at that frequency was something like 31 dB. He threw that article down on the table mad as a wet hen and said, "Here it's been proven that you can't get more than 31 dB." They may not be the right number. I've forgotten the number now. He said, "You're asking me to come within 1 dB of that. You're out of your mind." He was very incensed about it. Needless to say, the final models exceeded that goal. Particularly the later models of the radio had reasonable sensitivity. Another thing I remember about that time was struggling to get adequate sensitivity. I remember at that time Frank Horak asked me what the big problem was and I said, "Really there is not enough gain in this set, to my way of thinking, to give adequate performance." He said, "Well, how much gain does it have?" and I said, "Well, right now it's got about 100 dB typically in overall gain." And he said, "How much do you need?" and I said, "I'd say at least 110 dB." He said, "Just like engineer. Ten percent more." I looked at him. I didn't explain the difference between 10% and 10 dB, but I don't think he ever understood.

Crystal-growing and pulling

Wolff:

After this meeting and I.D.E.A. went off to refine the design and start building the radio what was your involvement during that period? They came out with the radio in the late fall of '54. What were you doing at that time?

Webster:

The only thing I did at that time was to help them define what they needed to do in the crystal-growing process to get a device. My role was primarily to set up measurement techniques and I worked in the screen room with the radios.

Wolff:

When you say "help them" do you mean the TI people?

Webster:

Right. The original work was done in the original TI building, but the crystal pulling part of it was in what was called "the bowling alley," which was a rented facility just about a half block north of the original facility at 6000 Lemon Avenue. We moved up there to be close to the crystal pullers so that they could pull crystals, get them cut, get bars mounted in headers, get them to us and get instantaneous feedback. We set up screen rooms and measurement facilities. That was after I.D.E.A. came in and started their development of it. Then I did very little more on the circuit development. That was mostly Dick Koch after July. Most of that was his work.

Transistor evaluation; production and pricing

Wolff:

Most of what you did was in helping them pull the transistors?

Webster:

Helping to define what made a good and acceptable transistor. I was not directly involved in the manufacture of the transistors.

Wolff:

I see.

Webster:

I was then involved in evaluating transistors and telling them whether or not they were good enough. If you had asked me ten years ago I could have told you the number of the crystal that they pulled that was the first really good one and everybody said, "Hallelujah, we finally know how to do it." Then they made 300 more before they drew another one like that. You know how that goes.

Wolff:

In this little write-up by Ross Smith he said, "Producing enough quality transistors in order to reduce costs was a major problem for TI at that time."

Webster:

Yes.

Wolff:

He says, "The original purchase order from Regency to TI was for 100,000 kits of four transistors each at a price of $10 for each kit."

Webster:

My recollection was that the first kits were priced at $12.50, but I won't guarantee that. You might check that out with Buddy. I'm pretty sure Buddy was directly involved in all those negotiations. Of course the price came down and down and down as time passed.

Transistor and crystal researchers, breakthroughs

Wolff:

During this time when you were defining what made good transistors and evaluating them the biggest problem was in the crystal-pulling part of it. Do you remember any particular discovery or technique that solved that problem?

Webster:

No.

Wolff:

Was Teal involved in that?

Webster:

Gordon may have been involved as an advisor, but as far as I know he was not directly involved in that. They were pretty much head-over-heels in silicon transistor development, and he was very much a part of that at about that same time. Additionally, Gordon was involved in the development of the pure silicon factory too. Gordon was not, to my knowledge, very closely aligned with that, but I could be mistaken. I had virtually no contact with Gordon at that time.

Wolff:

It was just his original transistor concept.

Webster:

Right.

Wolff:

I think I interrupted you. I asked if there was any particular breakthrough that it possible to pull transistors.

Webster:

Not that I remember. There were a great many crystals pulled in which they varied the time and rate of pull and the timing of the dropping of the doping pellets into the melt and that sort of thing in order to try and find a combination that would give a narrow base width and a fairly low base resistance. That was one of the problems with the grown-junction transistor. With the narrow base and the contact only on one side they had quite a high base resistance. If you got a high cutoff frequency the base resistance also went up and the two tended to counteract each other. A lot of time and effort was spent in trying to develop a combination of pull rates, times and temperatures that could be used to get reproducible results. However we reproduced a lot of them that did not work well enough. Getting the combination that would work well enough was the big problem.

Wolff:

It sounds like it was trial and error.

Webster:

As best I recall, it was trial and error then. You'll get a lot better insight into the problems by talking to someone who was directly involved. I believe that Frank Horak was the guy who, through much of that time, was trying to grow the crystals. He left TI sometime after that and went out to someplace on the west coast. I haven't seen or talked to Frank in twenty years or more. There were others involved in crystal pulling too, and Buddy can probably fill you in on that. Frank was the guy I dealt with mostly because he was actually out there peering in the puller and twisting the knobs and things like that.

Wolff:

Haggerty talks about the TI team of Boyd Cornelison, Elmer Wolf, Roger Webster and Jim Nygard.

Webster:

Jim Nygard and I worked on the transistor radio a lot.

Wolff:

Was this on the design of the original breadboards or was this in the semiconductor operation?

Webster:

Jim Nygard was in the screen room and worked an awful lot on that. I don't believe that Jim was involved when we made that first breadboard. I believe that when we were working on the original breadboard he was still a student at A&M. I believe he had come in that summer as a technician. I'm almost sure that at least for part of that time Jim worked on it as a technician and then stayed. He would have been involved fairly early because Texas schools get out fairly early.

Wolff:

Is he still at TI?

Webster:

No, he retired last fall but he lives here in Dallas. You might want to talk to him. He stayed on the project after I went on to other things and worked with Regency after that time much more closely than I. From the time that Dick came up with the four-transistor model I began to drop out of the Regency work and started work on some other radio developments for improved circuits and the possibility of car radios and that sort of thing. Jim stayed on and worked very closely with Regency people. At least in the original part of that it is also my recollection that neither Elmer Wolf nor Boyd was involved until somewhat later. It may have been Boyd who took over that crystal pulling and finally made it a practical thing. That was after I had more or less gotten away from it.

Wolff:

Do you have a phone number for Nygard?

Webster:

We can get you one. Jim or Buddy could tell you when Boyd and Elmer came in on it, but they were not involved in the early phases. In fact I had not even met Boyd at that time, but later I worked closely with him.

Significance of the Regency radio

Wolff:

Is there anything else I should have asked but didn't? What do you think the significance was of the whole thing?

Webster:

Of course I'm prejudiced, but it had a very significant impact on the semiconductor business. It really put semiconductors into the consumer market. That was the start of it. It's not to say that it would not have happened, because obviously it would have happened sooner or later, but that's the history of most things like that. The people who did it first are generally the people with vision. In that sense I think you have to credit Pat Haggerty with the vision to push it and push it hard at that time.

Wolff:

They made the radio for a few years and then it sort of petered out as far as I understand.

Webster:

I think that was basically the conservative attitude and approach that I.D.E.A. took. I think they were basically a very conservative and small company, and I was always disappointed that they didn't really capitalize on what seemed to me a unique position for them and push it. Perhaps the lack of financial backing or something else might have influenced their decisions. I don't know. However after this one big push into the market, you're right, they seemed to withdraw. TI had a pusher in Pat Haggerty. Haggerty at about that time was going after IBM. As a transistor manufacturer that proved to be a far bigger and more important than the radio.

TI transistor sales and contracts

Wolff:

Are you saying that TI's big effort after this radio was to sell transistors to IBM?

Webster:

That came up very shortly. They started with IBM within a year or so after that.

Wolff:

I know the radio was given to Watson. He ordered a couple hundred models to show his executives that they ought to put transistors into computers. Are you saying that IBM started buying those transistors from TI?

Webster:

Not those transistors. Those were not switching transistors. No. The transistors that TI eventually made for IBM were designed by IBM.

Wolff:

Are you saying that making the radio led to your first major contract, which was selling transistors to IBM?

Webster:

Indirectly it did, yes. To me, that was Pat Haggerty looking for opportunities. I think he saw that as a big opportunity and I'm sure Watson was looking at solid state as a way around the problems with the vacuum tubes.

Wolff:

Yes. The radio opened the market for TI's transistors?

Webster:

Absolutely. TI maintained a substantial radio and RF type development in semiconductor devices. I was involved in that directly and for the rest of my career at TI. I still am indirectly involved, and it was a result of that. There were many radio designs made here, some in which I was involved and some I was not. There was a radio design section that contacted all of the radio manufacturers – Warwick, Emerson, Zenith, you name it. We talked to all of them.

Wolff:

Did you sell them transistors for radios?

Webster:

Yes.

Wolff:

Did you make radios for them?

Webster:

No. TI never made any radios. It was always a support effort for radio manufacturers. None of them at that time knew anything about semiconductors. They needed a circuit built in developing suitable radio designs, and there was a substantial effort that TI maintained for many years.

Transistorized TV set design

Webster:

Pat Haggerty also felt that there would be a real opportunity in semiconductors in television. That was one of the big jobs to which I was assigned was to develop a transistorized television set, and in 1957 or '58, I built a little portable transistorized TV set. We bought a 9-inch vacuum tube set manufactured by GE, gutted it, took the picture tube to one of the local picture tube rebuilding houses and got them to put in a special 12-volt 150 mil filament gun on it, now working with Boyd Cornelison and Elmer Wolf. They were designing some germanium mesa transistors that had a much higher frequency capability than grown junctions did for these things. I designed a little portable battery-operated TV set and carried it all over the country in 1958.

Wolff:

What was the result of that effort?

Webster:

There was no immediate result. Most of the set manufacturers at that time felt it was going to be too expensive and would not compete with the vacuum tube sets. They were interested, and several of them had small development efforts that they maintained in solid-state TV. Small groups. John Bell working at Zenith was one of them. Warwick had a small group working on solid-state TV, as did Emerson and several other set manufacturers. We visited Emerson several times. However as you know it was years before a commercial solid-state TV set was put on the market.

Wolff:

What you were doing was trying to show people applications for TI transistors.

Webster:

Yes, that is correct. Essentially the main purpose of the group that was maintained here was to try and show people how to use them and try and convince them that they should use them and in so doing develop markets for the TI transistors.

Wolff:

The TV set used germanium mesas?

Webster:

Yes.

Competitor products and transistorized radio marketing

Wolff:

I was interested in this comment by Haggerty on the 25th anniversary. He said they sold the radio too cheaply and if they had sold it for more they would have made so much more money they could have entered the consumer business earlier.

Webster:

Yes.

Wolff:

He said, "I think the likelihood is very high that we would have been the Sony of consumer electronics."

Webster:

Yes. I have heard that remark of his. I don't know.

Wolff:

Do you know when Sony came out with its transistorized radio?

Webster:

The Japanese didn't come out with transistorized radios until much later.

Wolff:

It's funny. Everyone thinks that Sony invented the transistorized radio.

Webster:

Good Lord, no. That was years before they came up with a transistorized set. The first competing set of which I was aware was made in Hong Kong. After I.D.E.A. went out of the business TI offered to its employees a set a little bit smaller than this one that was built in Hong Kong. I've still got the thing at home called an Olympic or something. I've forgotten the name of it, but it still works. To my knowledge there still were no Japanese sets at that time, and there were none until sometime in the '60s. That's my recollection. Motorola was the first one that really came out with a solid-state TV set. Isn't that right?

Wolff:

I don't know. I thought that right after the Regency radio Raytheon came out with transistor radios.

Webster:

Raytheon came out about the same time with a little portable radio that used alloy-junction transistors. It is my understanding – and I would rather not be quoted on this – that it was never a success because of the non-reproducibility at that time of the alloy-junction transistors for the RF portion of thing. As far as I know they only sold a handful of those things compared to the Regency.

Wolff:

Raytheon told me that they were selling more transistors than anybody for a while.

Webster:

That may have been so, but I don't think it was because of their radio. They sold an awful lot in things like hearing aids and whatnot.

Wolff:

Yes. I think that's what he meant. They used alloy junctions in the RF portion?

Webster:

Yes. It's my recollection that it was basically the non-reproducibility of the devices as they were made at that time. Obviously later on they learned how to control properties, but they seemed to have no end of troubles getting satisfactory devices for the set. I won't guarantee the accuracy of that statement.

Wolff:

The radio we are looking at here is the TR-1?

Webster:

Yes.

Wolff:

I saw a photograph of one in white.

Webster:

They came in a lot of different colors.

Wolff:

Does it still work?

Webster:

It did the last time I tried it. That had a 22 ½-volt battery. One of the reasons for that – or perhaps the reason for that – was the fact that the audio output transistor was also a grown-junction transistor. One of the characteristics of that was a fairly high breakdown but low current-handling capability. In order to get any power out of the thing at all they had to go to the 22 ½-volt battery. The total current consumption was quite low, but that was the only way we could get adequate power at that time with the grown-junction audio output transistor.

Wolff:

What is the brochure you have there?

Webster:

"How to operate your Regency Model TR-1."

Wolff:

I like this civil defense thing. Could we make a Xerox of the front and back and this?

Webster:

Sure.

Wolff:

Okay. This is the end of the interview with Roger Webster. Thank you very much.