IEEE
You are not logged in, please sign in to edit > Log in / create account  

Oral-History:Harold B. Law

From GHN

Revision as of 05:10, 10 August 2009 by WikiSysop (Talk | contribs)
Jump to: navigation, search

Contents

About Harold B. Law

Harold B. Law received a Ph.D. in nuclear physics from Ohio State University in 1941. His doctorate was one of the first granted as a result of work with the high-state cyclotron. His early interest in electron multipliers led him to a job, in 1941, with RCA in Camden, New Jersey, working on the development of camera tubes. Law moved to RCA's Princeton laboratory and worked with Al Rose on the development of the image orthicon.

In the interview, Law offers detailed descriptions of the technological developments and breakthroughs related to his work at RCA on the image orthicon, the vidicon camera tube, and the photo-deposition of phosphors. There is an extended discussion of the complicated technical difficulties involved in the development and production of color tubes. Law also discusses recent developments in color-tube technology, such as the black matrix, focusing tubes, in-line tubes, and feedback tubes. The interview concludes with Law's discussion of possible future developments in television display.


About the Interview

HAROLD B. LAW: An Interview Conducted by Mark Heyer and Al Pinsky, IEEE History Center, July 15, 1975

Interview # 028 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:

Harold B. Law, an oral history conducted in 1975 by Mark Heyer and Al Pinsky, IEEE History Center, New Brunswick, NJ, USA.


Interview

INTERVIEW: Harold B. Law INTERVIEWED BY: Mark Heyer and Al Pinsky PLACE: Princeton, New Jersey DATE: July 15, 1975

Education and RCA

Heyer:

Please tell us when you came to RCA, and about your schooling.

Law:

I came to RCA directly out of the graduate school at Ohio State University. I spent all my graduate time at Ohio State obtaining a master's and then finally in 1941 a Ph.D. in nuclear physics. I recall that George Martin was then working with Dr. Zworykin in Camden. I wrote to them indicating that I had worked in the past on electron multipliers, which they were interested in at the time. They invited me to come to Camden for an interview, and eventually I wound up working there with Dr. Zworykin on camera pick-up tubes.

Heyer:

How about your undergraduate work before you went to Ohio State?

Law:

My undergraduate work was done at Kent State University. In fact, I had lived in Kent, Ohio since I was in pre-school. I was born in Iowa. My grandfather lived in Iowa, and I spent some time on his farm in the summer times and have fond recollections of that place. I consider Kent as my hometown because I grew up and went to high school and college there.

Electron Multipliers

Heyer:

That was in 1941, the year you began at RCA?

Law:

Yes.

Heyer:

Was Jan Rajchman here at that time?

Law:

Yes.

Heyer:

Was he also interested in electron multipliers?

Law:

Yes. I have known him for many years. Relatively few people, John Rudy was another, were working on the same area as I was at that time. He has retired now.

Heyer:

He was also at RCA?

Law:

Yes, at RCA. There was also Gardner Creeger, who later left RCA.

Heyer:

Paul Weimer was from Ohio State University too?

Law:

Yes, a year behind me. He came to RCA the following year. Actually, I knew his wife because at Ohio State she was doing graduate work in nuclear physics. I became acquainted with her shortly before I left. I knew both of them before I came to RCA.

Heyer:

When you came, did you know specifically that you would be working on electron multipliers?

Law:

Yes, that's why they were interested in me — I had done some work along that line. Camera tubes are rather closely allied to the principles and techniques of electron multipliers, so I found a niche there.

Heyer:

Was there a sort of an overt interest in television at that point?

Law:

I was pretty green when I came to RCA. I had never had any practical experience to speak of, and no electronic experience at all. I hardly knew what an ohmmeter was at the time, but I was propelled right into glass blowers, pumping stations, getters, and the whole bit all at once. It took me a while to get oriented.

Heyer:

So your particular interest prior to coming to RCA was in electron behavior?

Law:

Well, I was interested in nuclear physics. I worked with the high-state cyclotron. I had one of the first degrees that was granted as a result of the work with the cyclotron. Dr. Pool was my sponsor, a very research gadgeteer-oriented man. We hit it off very well. I had an interest in real experimental work.

Heyer:

You certainly came to the right place.

Law:

Yes, this is what made me so happy. The work involved really getting your hands dirty. You made your own parts in those days and did what you wanted.

Image Orthicon

Heyer:

A couple of years later you came over here? [Princeton.]

Law:

When I was hired at RCA in Camden, they mentioned that a new laboratory was being constructed at Princeton and that in about a year we would move. I knew that we would set up a new laboratory here. As it happens, the groups of people from Harrison and Camden more or less combined, forming another group under Dr. Rose to continue the work on camera tubes. We started on the development of the image orthicon, which had been initially developed by Dr. Rose in Harrison. We wanted to make it more sensitive. I was involved in turning it into a practical device.

Heyer:

Into a producible component?

Law:

We had to work out some of the details. I was working on the target end of the tube. At that point we needed a very thin piece of glass, of a special type, mounted very close to a very fine mesh screen. When I entered the picture Al Rose was using some sort of fine fly-screen, but it was so coarse that it was interfering with the picture. Also, he was using slivers of thin glass that he could manage to fabricate, but they were too thick. My contribution was to learn how to make the glass about two-tenths of a millimeter thick. I did this by first blowing a bubble out of the glass and then sealing the glass to a metal ring by heating it, so that the glass softened and pulled taut by surface tension. The mesh was another problem. I worked for a long time trying to find a way of making a very fine mesh. I started out by doing it by photographic means, trying to work from rulings and so on. Finally, I discovered a way almost by accident. I couldn't get a fine enough ruling for my photographic work, so I went to the Max Levy Company in Philadelphia, who make rulings for the photo-engraving trade. They made me a ruling on glass that had about 500 lines to the inch. They did this by a diamond ruling which was slide etched. My object was to fill this with some lamp black or ink and then produce a contact print or negative that I could work with. One day I wondered if I could actually use this directly, so I sputtered some metal on the surface. The glass consisted of little islands because the lines were etched into the glass. My intention was to try to plate the screen into the glass. In order to avoid plating across the top, I took a piece of moist rubber and rubbed the surface, and sure enough the metal that had been sputtered came off. I dumped the thing in a plating bath and the wires plated. I plated them for a few moments, and then I was able to strip the screen off. I had a 500-mesh screen with about 60% transmission.

It was exactly what I wanted and it just came on a hunch. There is one other thing to this story, which I think is interesting. When you get a piece of very delicate copper like that, it isn't going to be flat, and we needed to have it very flat to lie within about two mils from the glass. I decided to work with the mesh to try to flatten it by mounting it on a ring and welding it, but it always left some unevenness. It would always stretch a little bit here and there. One day I decided that I would try to anneal the screen, so I put it in a vacuum after it was welded to the ring. I heated it up and the screen tightened like a piece of glass. It just pulled taut. The plated material is not very dense, and it was coalescing and becoming more dense. In the process it was pulling taut. It acted like a piece of glass and came out beautifully flat. Another sidelight is that after I had disclosed this in the Washington patent department, they refused to grant a patent because they said, "Anybody knows if you heat up metal it will expand and not contract." So Joel Johnson and I went down to the patent department in Washington with a hot plate, and I rigged up a special demonstration where I used some gold leaf foil and some silver leaf. It turns out that with just hot plate temperature you can start with a gold leaf or silver leaf that is not very tight at all with lots of crinkles in it, and when you put it on the hot plate it will tighten up just like the mesh did. So the patent examiner said, "Ok. You get the patent."

Heyer:

The next thing to talk about is the vidicon.

Law:

Dr. Weimer worked on the multiplier for the image orthicon. Eventually we got together a tube with all of the parts, which was very sensitive. I recall we went to NBC to put on a demonstration, where Ben Drower lit a match and was televised using just the light of the match. It was quite exciting and interesting after having worked on it to go and participate in a demonstration of that kind.

Heyer:

Where were they built? In Harrison?

Law:

I am not sure whether they initially went to Lancaster or not, but my first real recollection is that Lancaster did the mass production of these tubes. They were used by the television industry for a long time. Of course, the orthicon had been used prior to that, but this was of much more interest.

Photo-Deposition of Phosphors

Heyer:

After the development of the image orthicon, what was next?

Law:

There was a period there where our group, which consisted at the time of Al Rose, Paul Weimer, and Stan Fong, became interested in the vidicon type of camera tube, where you used cadmium sulfide and attempted to make a very small, compact, cheap type of pickup tube. I worked with the fellows for a while in the late 1940s. I first became interested in television at that time, particularly thinking about color, because Dr. Epstein and his group, Ted Nichol and so on, were on the floor above me and were making some experiments settling line screens with color phosphors in very preliminary attempts to get ideas on how we might be able to make some kinescopes for color. I began to do a few experiments related to color kinescopes, even though I was supposed to be working on pickup tubes. I did the ground work for what later was a rather important process, the photo-deposition of phosphors. It was not my main job, but it was a side issue that I was interested in. I didn't realize at the time that it might become important. When the real need for a concentrated program came about in the latter part of 1949, I made a shift away from Al Rose's group to Epstein's group for the kinescope work. There were several projects which were undertaken to try to make a color tube that would be of use to us in trying to demonstrate potential for a compatible color system. I took on a particular project that I was interested in. There were about four or five other groups that also started projects. This was when the primary work was done on the present shadow mask tube.

Heyer:

As I understand it, a compatible system had been demonstrated using [unintelligible] mirrors.

Law:

Yes. That was very useful for doing circuit work and trying to make a compatible system insofar as the circuitry would actually present a color picture if you had the proper kind of display. The only display they had at the time, or one of the most prominent ones, used three color tubes in which superposition was made optically by means of [unintelligible] mirrors, but this had limitations because of its bulk. I think they did make up some receivers of this type, but I have heard it said that they looked more like piano boxes than anything else. They were so big that they were obviously out of the question for the home.

Heyer:

I suppose you heard at that time that CBS had a mechanical system that they were putting together.

Law:

We were certainly hoping that we could come up with something that would be more appropriate for the future. It didn't look, to any of us, like something that had a potential for continued development where you could have a larger picture and free yourself of some of its limitations. Any fast-moving object appeared as discrete images because of the rotating disk. It just didn't seem like a final system.

Heyer:

But there was obviously a great awareness that color was something to be pursued.

Law:

Right.

Heyer:

Do you want to say some more about the work that you did with photo-deposition? Were you working with Leverenz on that? He was the phosphor man.

Law:

No, actually this was a Friday-afternoon type of experiment. Nobody was really asking for this sort of thing; I was just puttering around. I didn't think of photo-deposition as being something useful right away. It was just an interesting technique which I thought would be useful when we had developed the rest of the system far enough so that people would begin asking us for single-color tubes. At that time, we didn't have any way of making them, and I was just looking for some techniques so that we would already have some ideas when we needed to proceed. The available schemes at the time weren't very attractive, but I do recall looking at this shadow mask principle, which had already been invented of course. It seemed to me really to have the potential of something if you could only build it. I thought it would really provide a good display. I began to think about how to build a tube where you would have a mask and have little phosphor elements at the back of each hole in the mask. It didn't take long to occur to me that light and electrons could be interchanged, and then it all rather fell rapidly into place like a jigsaw puzzle.

Dr. Weimer had a demountable system going at the time — he was still working on pick-up tubes — and I prevailed upon him to let me put a small sample screen, a sheet of metal with regular holes in it, in front of a solid phosphor screen. Then I used an electron gun to scan this system. When I looked at the phosphor screen, I saw very sharp edged images of the holes in the plate, and I knew right then that the whole thing was going to work because there was not a fuzzy image. I would be able to put the two dots very close together without interference. The whole thing fell into place. It was then a very exciting time trying to devise a way of making a sample that would demonstrate what I had indications would really work. Everyone was working on their own projects so nobody was really paying too much attention to each other, except Ed Harold, who was overseeing the whole business. He would come around every once in a while to see how things were going. In those days, the only technique that I considered using was to settle the dot screen through the mask itself. That was the way the first shadow mask tube was made. I settled the phosphor through the little holes and then moved the screen and settled through another set, and so on. It was a time-consuming process and didn't work very well.

The big breakthrough — as far as actually getting a good screen — occurred when Norman Freedman in Harrison said, "Why don't we do this by silk screening instead of settling?" He was able to make a sample and it was obviously a very much superior way of depositing the phosphor. So I did this type of deposition to some extent. There was a problem however in that you could not make a contact print of the phosphor from a pattern and still have the electron beam hit the spots. In other words, the mask had to be spaced a distance from the phosphor plate. The scheme that seemed to me to be at least something possible was to put a piece of film in the position where you wanted the phosphor screen to be, space the mask from it, then use the light optics to make an exposure on the photographic film. This gave you black dots when you developed the film, and it was this negative photograph that was usable to make a silk screen. In effect, you were printing the phosphor screen indirectly, as a second step. Later we got ultraviolet sources strong enough to use the photo-deposition of phosphor to combine these two steps so that you could make a direct process rather than an indirect printing process. Right from the beginning, when production was on the curved screen tube, this became a necessity. Up until the time we used a curved screen, RCA used indirect photographic method on flat surfaces, but it was proving to be a very costly and difficult process. I was exceedingly happy when we were able to go to the curved screen and combine the photographic deposition with that as a direct process. This was really a breakthrough that was actually sparked by CBS. They did it first. We had all of the tools at our command, but we didn't do it first.

Advantages of Curved Screens

Heyer:

Would the curved screen make it easier to fabricate the raster or give you a better picture by itself? Why did you curve the screen?

Law:

There are two reasons. One is that the picture looks better when it's on the end of a tube rather than looking through a window at something inside a tube. The other was that we were finding it very difficult to put an internal flat piece of glass in the tube and get a practical pump-down because the glass inside would not allow itself to be heated up or cooled down very rapidly without breaking. On a standard kinescope the glass is opened to the air and it follows temperatures pretty well, but if you have a piece of glass inside of a tube like that, the primary way it can cool is by radiation and it takes a very long time to heat up and cool down. In addition to that, it was very expensive to use a piece of film to go through all of the development process and then make silk screens for each tube and print it. I would say, from both counts, that the end product was more desirable with the phosphor on the face plate, and it was a great cost-saving process, which made it a very natural thing to do.

Heyer:

When did these tubes start to come out?

Law:

We made these internal face plate tubes beginning about 1952 or 1953. It was later that the curved face plate tubes arrived, around 1955. We abandoned the flat internal face when we went to the curved screen type.

Color TV

Heyer:

By then color TV was on the way?

Law:

Well, it seemed everyone was forecasting it was right around the corner, but it did take an awfully long time to catch on to the point where we could produce quantities of the tubes and look forward to lower-cost production.

Heyer:

Before you were making any money on it?

Law:

Yes, before the volume could be large enough. Of course, there was the question of programs. It was the chicken and the egg business — people didn't want to buy it until there were programs. Also, the construction of the tube was recognized as a very difficult process and expensive, and so people were waiting until it got perfected. You heard any number of times through the years that surely you can't expect commercial products to be that complicated and that this kind of a tube is just too complicated and too hard to make. But the people at Lancaster, and other companies too, did a marvelous job in working out the processes, engineering the thing, and working to extremely close tolerances. Eventually everybody came to realize that it just didn't seem like anything else was going to appear, and that although we were making slow progress, maybe it would be a commercial product after all.

Heyer:

It's turned out to be.

Law:

If you go today to the factory and read their reports, the same problems exist that existed twenty years ago. These problems never get solved. There are so many critical things, both as to materials and processes, that have to be so well controlled that it seems to have recurring problems. Something gets out of control and you have to hunt it down, find out what it is, and get your production back on schedule again. It isn't easy to make the tube even today with the number of things that can go wrong, and of course the requirements that the commercial people put on a tube are continually rising. They are upgrading the product all the time. So people are having to work closer and closer to the limits. In no sense are they resting on their laurels. I think this is what competition does for you; other people make improvements, so you have got to make improvements too. You just can't get sloppy with your work.

Heyer:

Do you have any idea how many channel match tubes have been made?

Law:

I think Ed Harold would probably have this kind of figure because he has given some lectures and some talks on the history of color television.

Alternatives to the Shadow Mask

Heyer:

What are the developments you have been involved with since the shadow mask?

Law:

While the shadow mask was slowly rising in the production, we at the laboratories here were looking to see if there was a better way of making a color tube. I have been involved here at the laboratories for a number of years looking at focusing-type tubes, which would be similar to a shadow mass tube but would allow the mask to have higher transmission, and therefore much more of the electron beam would go through and give more light for the given power that you put into the system. E.L. Lawrence, of the cyclotron fame, invented a tube called the Lawrence tube, which was given a good test by a number of people. In fact he did quite a bit of work on it, but others did too. It amounted to having an electric accelerating field between the mask and the phosphor screen, so that each aperture in the mask became an electron lens and would cause the electrons going through the aperture to become smaller in cross-section as they proceeded to the phosphor screen. One could start with an aperture in the mask that is larger than before and still end up with the electron spot small enough so that you could mesh together the various small spots. There are a number of problems with this approach, which I could enumerate. One is that the contrast of such a tube is poor unless you do something about it, and that arises from the fact that these electrons, when they hit the phosphor screen, will scatter back away from the screen and find themselves in a retarding field. So they tend to fall back on the phosphor screen just like a ball thrown up into the air will curve and fall back to the earth. Then of course they hit the wrong place, the wrong phosphors, and tend to give you extraneous light. There are things you can do to avoid this. One way is to put a coating on top of the phosphor screen that will help absorb some of these unwanted electrons, but at the same time you attenuate some of those electrons that you do want. The electron gun in this tube is forced to operate at a lower voltage because you do not have the high voltage except on the phosphor screen. This means that the electron gun does not work as well. You can't get as much current, or as fine a spot, so you jeopardize the excellence of the tube. I must say though that focusing tubes are not dead today.

Hitachi in Japan is about to introduce a focusing tube, they say, in 1976. What they have done is to not try to go to the limit on this focusing action. They instead have only a small amount of accelerating voltage between the phosphor screen and the mask. They have developed coatings, that I spoke of a moment ago, that are fairly effective in reducing this unwanted light. The lower voltage gradient means that the effect of the extra electrons is not as great, so you don't have to overcome as much as you would if you had more voltage. Coupled with this development of the screen coating they have developed a gun that will produce a better spot than we have ever had before with a low-voltage gun. We have seen demonstrations of this tube. Although they have done what looks like a very good job, I still feel that the added brightness they are going to get is not going to be worth the extra cost because the tube will be quite a bit more expensive with this complication. I am skeptical that their development will actually be successful commercially. We, of course, must be aware of this system and experiment with it, in case it is more successful than I imagined it will be, so we can be prepared also to consider a tube of this type.

There have been other tubes that are alternates to the shadow mask. One of these is the so called "feedback tube." There has been a lot of work done at the RCA Laboratories, and it has continued to be something that other manufacturers have looked at it, notably Philips. It's a rather complicated system, quite a bit different from the shadow mask system, but it has its faults and its advantages. One has to answer the question, "If you have a well-developed shadow mask system, what are the chances that an entirely new system, an entirely new factory, and new processes can be successful in the face of the entrenched system?" So far the answer has been, I think in all cases, that no one has chosen to make the major investment that would be necessary to give it a try.

In-Line Tubes

Heyer:

What about the in-line tube?

Law:

The in-line tubes are a development that have changed certain aspects of the shadow mask tube, but it is still basically a shadow mask tube. The configuration of the electron gun and the phosphor elements are geometrically changed from a delta structure to an in-line structure. We were aware, way back in the fifties, that this kind of a configuration would work well as a shadow mask system, but no one at that time was interested enough to contemplate the amount of research that would be necessary to give it a real good try because we didn't see its advantage. In later years what happened is that there was a development in deflection yokes such that it actually was possible to design an astigmatic yoke that would give you automatic convergence if you had your elements arranged in-line and an in-line electron gun. This meant that for converging the electron beams you need only move them sideways and their position, or color for a colored period, was not critical up and down. It meant that if you could converge these beams, then you could perhaps have a major advance in color tubes with a simplified type of convergence or no convergence controls at all, and still get a good color picture. This has come to fruition in the last couple of years and it appears that, particularly in the smaller size, in-line tubes are here to stay with a very great simplification in the electronics needed for converging the electron beam.

Black Matrix

Heyer:

And the black matrix?

Law:

The black matrix was introduced somewhat before the in-line tubes. It was a search for a way of getting a brighter picture with better contrast that sparked this development. The black matrix is an old idea if you look in the patent literature. Actually the feedback tubes that I spoke about earlier used black stripes between phosphor stripes, which is essentially the construction of a matrix where you surround the phosphor element by black non-light-emitting material. I think that the actual application of this to dot tubes was sparked by the need, or the possibility, of actually making quite an improvement in the light output of these tubes. It was a case of applying known principles and achieving this result. One of the inventions that was quite critical to this, which was made at RCA and used by the entire industry, is the carbon reversible printing process. The invention was made by Edith Miode at Lancaster and was very crucial to this whole system. In my opinion, Zenith, who came out with the early work or the development in this area, really was in trouble in getting the process to work even though the principle was very sound; and they were basing a commercial program on it. They were headed for serious trouble until this reversible printing process came about, which made their system feasible at that time.

Heyer:

I would imagine it would be similar to printing the phosphor dots — only then you would put something else on and remove those areas.

Law:

The way the process works is that you make the matrix first. Really, you have the capability of having light exposure through the mask, which gives you little pencils of light where the light goes through the aperture in the mask. The kind of resist that is used for printing phosphors hardens where light strikes it, and that isn't what you want. You want an opening at that point, so you need to have some sort of a reversal process. Edith's concept, which works beautifully, is to put down a photo-sensitive PBA on the surface of the glass and then make an exposure through the mask for each of the three color centers. When you develop that you have little dots of photo resist that are clear. No phosphor is involved; it's just clear dots at these points. The crucial part of this is that you now have a slurry of carbon that you put over the top of this pattern of PBA dots, you dry it, and the carbon adheres quite well to the glass. Then you go in with an oxidizing agent, in this particular case it's something like Clorox, that will attack the PBA and therefore make it possible to clear the carbon off where it is undercut because the PBA has been leached out. The carbon will actually slough off at that point. You are then left with a beautiful pattern of holes in carbon. You proceed as usual to print the phosphor, but it was the reversal part of this that really made the whole thing practical. The concept before was to use a negative resist where the resist washes away when it's exposed; but the only things that we know of that will do that are organics, which are fire hazards, hard to handle, expensive, and just not practical. So we wanted to work in a water system if at all possible.

Shortened Tubes

Heyer:

How about the shortening of the tubes widths, 110 and so forth? Were you involved with that?

Law:

Yes, this was primarily a yoke problem. As you deflect further and further with the electron beam, the ability to make the beams hit the right phosphor dots becomes more and more difficult because they are hitting it at such a glancing angle. In these systems, the two requirements are firstly that you have to hit the right color with your electron beams and secondly that the three beams must be converged. The problem gets to be very severe, the larger the deflection angle. But we can pretty well handle these problems if you actually make the phosphor dots and the holes in the mask somewhat smaller towards the edge to give yourself a little bit more leeway, a little bit more tolerance. We can go up to 110 degrees and in fact maybe even a little further. The deflection power however goes up very rapidly as you go to these larger angles. The climate today is that we must look toward cutting down on the amount of power to run one of these television sets. One of the obvious ways is to cut back on the deflection angle and save power there, so actually we are going back in many cases to 90 degree deflection instead of 110.

Future TV Concepts

Heyer:

Interesting! Do you have any prognostications about the future other than what you said about feedback tubes?

Law:

I think an interesting question is the future of flat displays as opposed to the normal television display as we know it. In the last few years there has been renewed effort on flat-matrix-type displays. People have presented laboratory experiments where they have put television pictures on some of these flat displays. Most of them are not color now. I have always been rather negative about such types of displays. In the first place I think that there has not been a clamor for a display that would be suitable for hanging on the wall. One really hasn't faced up to the question whether the public wants this sort of thing. I can very well agree that if people had a larger picture they might like it, but I don't think that they would want it to depart very much from the format that we now have where you are viewing this on a piece of furniture that you are free to move around and that would fit into the decor of any room. My concept of the ideal next development would be where you would have a piece of furniture like the present TV console, containing record player and so on, that is a nice piece of furniture with cabinet work on both sides of it, and put the picture over the full front of it, say something like a thirty by forty inch picture, and have it mounted six inches or so off the floor with the controls maybe on the side. The depth would be about the same as now, maybe twenty inches. Many people would be interested in this as an advance over the present twenty-five inch set. I have racked my brain many times as to how do you approach this sort of thing. If you didn't require that the display be thin, then the only thing that I have come up with — and I have actually started a little bit of work in my group here — was as follows: suppose we were to make the picture in modular form, finding a way to put these modules side by side so that it would look as if it were a continuous picture.

For many years it didn't seem like one could handle the signal processing that needed to be done because each module would be showing only a portion of the picture. You would need to feed that signal in appropriately to that module but with the CCD devices, storage, and so on, this seems like it might be a solvable problem these days. I think there are ways that one can actually put together two tubes so that by carefully doing it you would not see a dividing line between the modules. I know the laboratories here are very much enmeshed in producing a flat-display device, an electron beam device, aimed at the forty by thirty inch size. They are devising several method schemes for this, but I am not very optimistic about the success of this approach because it seems to me that it will never be an economic success. In contrast to the shadow mask tube, where you are dealing with several million phosphor dots and several million holes in the mask, the processes that you go through here are really amenable to batch processing. You don't make them one by one, but you have a process that makes the whole thing at once. Then you look for perfection. You have a tremendous job in getting that perfection, but you don't have to get in and assemble each one and operate various parts of the picture with different circuit components. Thus you don't have to worry about whether the circuit attracts on this side of the picture or on the other side of the picture. The problem of making the picture as uniform as you need it is a horrendous problem. With the amount of construction and cost of this, I just can't see how it would ever be practical. Flat displays with electro-luminescent phosphor have a very much less taxing construction because they do lend themselves in many respects to a batch process. But here again the circuitry involved to get adequate light and to put color into it also seems to be extremely far off, and I just don't whether we are ever going to make it.

Heyer:

I could see that for exciting the phosphor dots on an electric luminescent phosphorous, providing the energy for each one of the dots discretely would be a real problem.

Law:

The crossed arrays in some cases allow you to apply the potential where the crossing is, but there are also some voltages developed at other points when they cross other lines. This seems to be a solvable problem. It isn't as large a problem as the ability to get each element to work properly to get the signals that you apply to these systems to actually end up giving you a uniform level of brightness without missing elements, and without striations in it where one line doesn't quite do what the other one does.

Heyer:

These are the CCD problems that they are having now?

Law:

Yes, but they have been able to make camera tubes with these CCD systems that look quite good. I am very surprised at that, and maybe there is a way of licking some of these uniformity problems. They seem to me almost unsolvable at the moment.

Heyer:

Or at least until several years down the road?

Law:

If you concede that they can do it in the case of a display, then it's the economics that bother me. I think you are unrealistic if you think that so many parts can be made and you could be able to sell something like this at a profit. We are just going to have to wait and see.