BIRTH AND DEVELOPMENT OF FARNSWORTH TELEVISION IDEA

Mr. FARNSWORTH. It also gave me a theme for research which has continued throughout the years as a guiding light, or as a direction for research and development, namely the elimination of all moving parts from television equipment. That idea I had fairly well established in 1921, when I was 13 years old, so that the moment I discovered tools, out of textbooks I mean, which would enable television to be done without moving parts, the invention seemed almost simultaneous, as a matter of fact simultaneously with the discovery that there was an electron and a photoelectric effect.

In 1922 when I was a freshman in high school I made the first invention, my first big invention in television, and it consisted of a means for producing an electric counterpart of an optical image. At that time it was a daydream, a daydream only. I had no facilities for doing research, I had no money to buy equipment, all I had was access to a very modest school library, but my sum total of equipment which I had for forming any definite practical idea as to the problems in television consisted of a static generator of a physics laboratory and an old Braun tube.

Nevertheless, this daydream, as you might term it, had the basis for perhaps the most important invention and certainly the earliest invention in the electronic field, namely, that of a tube for electrically transmitting a picture without employing any moving parts.

In the 2 years following 1922, that is 1923 and 1924, I continued to do research in libraries and with any type of electrical equipment that I had to work with, with the idea of evolving a complete television system free from all mechanical inertia.

My family at that time moved from Idaho, where I was attending high school, to the town of Provo, Utah, where I had slightly more laboratory equipment at my disposal and continued to develop the previous notion of television without moving parts, so that in 1924 I had evolved what is essentially the present system of electronic television. Again I had no money and no suitable laboratory facilities to reduce this theory to practice. As a matter of fact it has taken 15 to 17 years to make that a practical reality, but I didn't know how long it was going to take then, very fortunately.

In 1926 I was in Salt Lake City looking for anything I could do either to continue my schooling-my father had died and left the responsibility of the family to my mother and myself, and I was hunting for work, when I met two California businessmen to whom I disclosed my hopes and dreams of this television idea, who agreed to put up a sum of $8,000 to see if it was worth anything. I decided that was the proper time to get married, being 19 and quite old, so I got married and moved to Los Angeles, where we set up a laboratory, such as it was, to explore at least as well as we could with $8,000, the possibility of this television idea.

After just about 2 months, 22 months, as a matter of fact, of intensive work, we had used up all of the $8,000 and we had taken the idea to the California Institute of Technology and to experts wherever we could find some who would listen, and I convinced these early backers of mine, one of whom is Mr. Everson, who is here with me today, that at least the idea had some merit, but we had the basis for a nice patent, perhaps, but no substantial experimental evidence yet.

Mr. Everson's step then was to interest-and it sounds quite easy further financial backing in San Francisco, which he did very effectively and a group of San Francisco businessmen decided that it might be worth while to take a flyer on this television idea. As one of the men put it, it was a darned crazy idea but somebody ought to put some money in it. So they did agree to put up $12,000 more to see a little more what it was worth.

At that time, which was in October of 1926, we established the Crocker Research Laboratories in San Francisco for the purpose of continuing research on this idea which was essentially to take all moving parts out of television. Twelve thousand dollars sounds like it should be enough to find out what the idea was worth, but after 18 months we had spent 60 thousand, and without listing the problems in detail, I think it can be understood that it is very much as though someone with a considerable amount of knowledge-or it doesn't matter really to what extent the knowledge runs-is suddenly cast on a desert island, removed from all tools, and given the job of building a steam engine. That means building the tools to build the tools to build the tools to build the steam engine, and our problem of making a laboratory was, in the early part of our work, by far the greatest problem. Also, the state of the photoelectric art and of electric optics at that time was not far enough advanced to carry out properly the basic conception of the electron image-scanning, as we have called it.

In 1926 and 1927 the photoelectric materials that we had were almost scientific playthings. The photoelectric material available were the haloid of alkali metals, particularly sodium and potassium, and the construction of photoelectric cells required an amount of knowledge and art and technique far beyond that available to me, and so I proceeded to get all the help from scientific institutions I could. I pestered the people of the University of California and Stanford and California Tech and anybody who would give me information or sell information. But to make a long story short, in the latter part of 1927 we demonstrated a television transmission which used apparatus that did not employ a single moving part.

The CHAIRMAN. You said that you pestered everybody who would give you information or sell you information. How much information did you have to buy?

Mr. FARNSWORTH. Most of it.

The CHAIRMAN. How did you buy it?

Mr. FARNSWORTH. Through the funds made available to me through this group of bankers.

The CHAIRMAN. And what type of information was sold?

Mr. FARNSWORTH. The technique of forming electron surfaces, the experience necessary to blow glass and evacuate tubes and sensitize photoelectric surfaces in vacuum, and purification of the alkali metals, the electrical circuits necessary for amplification, and so much similar material that it practically covers the entire field of physics and optics.

The CHAIRMAN. Were you actually buying the information or the preparation of the information, the service of conveying it to you? Mr. FARNSWORTH. The services of the scientists or technicians who gave it.

The CHAIRMAN. In other words, it was information already available.

Mr. FARNSWORTH. It was information known at that time but not known by me.

Mr. PATTERSON. Mr. Farnsworth, was this the first transmission of a purely electronic television image?

Mr. FARNSWORTH. Yes; it was definitely the first transmission of an electronic image, and as a matter of fact I made an effort at that time, but failed, to be the first to transmit a television image. I could have known then that would be impossible. C. Francis Jenkins had transmitted television images prior to that. Earlier that year a demonstration was made by the Bell Laboratories of a mechanical television system.

Mr. PATTERSON. What year was that?

Mr. FARNSWORTH. This was 1927.

Mr. PATTERSON. In New York City?

Mr. FARNSWORTH. The Bell system transmission was from New York to Washington. Our transmission was simply a laboratory demonstration that an image could be converted electrically, entirely electrophonically or electrically, into the required signals and reconverted into an image.

Mr. PATTERSON. From New York to Washington?

Mr. FARNSWORTH. Was the mechanical demonstration.

Mr. PATTERSON. Didn't they have public demonstrations in New York City, too, at about the same time?

Mr. FARNSWORTH. At about the same time; yes.

It required about 1 year, then, to convert, to build a minimum amount of laboratory technic and to reduce an idea conceived in 1922 to a practical result in 1927. Incidentally, the first image transmitted was about a 60-line image of a dollar sign. That seemed to climax the work, when we could get real money and see the sign of real money, so at that time our company was incorporated and we continued to do work to perfect this idea and to perfect an organization and laboratory capable of eventually making something practical out of a laboratory toy.

About a year later, in 1928, we had a television transmitting tube as sensitive as then theoretically possible. I hate to say what that sensitivity was, and by sensitivity I mean the amount of light that it was necessary to project on to the tube in the image in order to get a useful result. The notion was obviously impractical for televising a subject because the amount of light on the subject would have caused it to blow up and burn up immediately, whether that was a subject or object.

But, also to just gloss over the immediate years of the ensuing years, the sensitivity then, which was theoretically possible was agreed by all the consultants that I consulted were as approximately 100,000 times less than is available today with the same tube and with substantially no change in the tube other than improved methods of making it.

Since 1927 one of the major problems has always been to obtain sufficient money to continue the experiments. There has never been any substantial revenue, or almost no revenue, coming in, and it speaks well for the original backers of this invention that they have now spent greatly in excess of a million dollars without any revenue for a development which has taken 13 years. It will be 13 years in May since it started.

In 1928 and 1929 we began to get recognition for our work and other engineers and inventors agreed that it was a difficult problem to work out, but would be the ultimate way television would be accomplished, and in 1929 Philco, the Philadelphia Storage Battery Co., took a license under patents which we then had issued, and under the promise of what our future developments would be.

But to go back to the point we had reached where we had the maximum theoretical sensitivity of our transmitting tube, it has been my experience that whenever a stone wall was encountered where possibility of scaling it seems hopeless, there is a grand opportunity for a good invention, and it happened in this case to call for one of the most important developments that we have made, namely that of the principle of electron multiplication. What we needed was more electrons for a given amount of light.

It may seem a little radical to expect that electrons could reproduce themselves, but this is in effect what they did. We produced an electron stream and had it arranged that it was capable of producing offspring at the rate of 5 per litter, 5 to 10 per litter, and then to take the offspring, and they have the fortunate property of being born mature so that in about less than one-billionth of a second they can produce each child can produce, with no question of sex involved, either 5 to 10 more in less than a billionth of a second.

If you consider how fast that electronic multiplication process builds up you can see that in less than one-millionth of a second it would evolve a number of electrons. Each initiating electron will evolve 25 with 500 ciphers, probably more electrons than there are in the universe. But the problem is not to get the electrons but to control the process so that there is a definite proportionality existing between the number of initiating electrons and the final output. That principle, which is called electron multiplication, now is in universal use throughout the world. There are many different types of tubes employing the process, and it has added perhaps the most powerful way of amplifying feeble electric currents that exists.

When we divide the number of electrons given out by an optical image by the number of divisions that are necessary to show fine definition in an image, which amounts to perhaps 400,000, the apparatus begins to count electrons. Even though the electron is a mighty small unit it is not small enough, and we begin to count the electrons. Now, the electron multiplier minimizes the extent of interference produced by this process of counting electrons, and that is why it is important in television. Many inventions have come from this fundamental principle. They are employed in perhaps 100 different varieties of tubes. It constitutes one of the very important byproducts of television research.

Mr. PATTERSON. On that point, I don't mean to interrupt the continuity of your thought, but I think it is germane. In getting your increased sensitivity did you develop any devices that are helpful to the radio industry generally? If so, will you put it in the record and tell of that?

Mr. FARNSWORTH. In the field of electron multiplication the device No. 1 is simply what might be termed a multiplier electric eye. Ordinarily, photoelectric eyes are measured in millionths of an ampere per unit of light. Photomultipliers are measured in units 1 million times that big. I have a tube that I will show you that is measured

in 50 amperes per unit of light, whereas the corresponding tube available on the market previously might be measured in 30 millionths of an ampere, or over 2 million times improvement in sensitivity

As yet we don't know to what extent this principle will be important in the radio industry, certainly to a very great extent, but in just what fields it will be important it is hard to say as yet. Certainly in the very short wave region below 1 meter and below 5 meters, it is already the most powerful tool in measuring. In measurement of very feeble currents such as used in stellar photometry or in various scientific applications or in some projected military applications the tube is by far the most powerful tool available to the physicist and inventor.

A particular tube which I have will record the light of a candle 10 miles away, and it will do so also instantaneously, whereas other methods of doing it might require 5 or 6 seconds and maybe that many minutes for its measurement.

Mr. PATTERSON. Just what do you mean by 10 miles away? I'd like to have you develop that for the committee. I have spent considerable time in it and I want the committee to hear it.

The CHAIRMAN. Do you mean to imply that the committee doesn't understand what is said? [Laughter.]

Mr. PATTERSON. You win, Mr. Chairman.

Mr. FARNSWORTH. The measurement of small amounts of radiation, either visible or invisible, is usually made by the heating effect of the radiation or alternatively by the fact that radiation produces the emission of the electrons from certain suitable materials. Its usual practice when extremely small amounts of radiation are to be detected is to allow this process to continue over a long enough interval so that the accumulated effect is measurable on the most sensitive instruments we have. Now, in the photomultiplier, due to this multiplication process, although only one electron is released, a million or so, in some cases a billion billion, are available for measurement. I say that number although it is so astronomical it may not mean much. We can in effect detect one electron per second.

The CHAIRMAN. In all of this it is still necessary to have an instrument at each end, is it not, one at the end at which the image to be televised exists, and at the end at which it is to be seen.

Mr. FARNSWORTH. Yes; although it is not necessary to have a visible image at the transmitter.

The CHAIRMAN. Has the art been sufficiently developed, for example, to enable you to segregate the light of a single star, let us say, from all the others?

Mr. FARNSWORTH. Only through the use of the telescope. If we could measure down to very low intensity stars, stars that would require considerable period to photograph, they could be checked and the intensity of that star determined to a much greater degree of accuracy by direct reading instead of photography.

The CHAIRMAN. By the use of the telescope you could segregate a particular star from all the other stars.

Mr. FARNSWORTH. With a suitable eyepiece you could segregate that from the remainder of the stars and measure its intensity, and you could do so and measure perhaps thousands of stars per hour instead of a few. That only indicates the type of application that this tube is adapted to. A more common application is in talking motion pictures where the tube acts not only as the photo cell but