Discussion of "A History of Some Foundations of Modern Radio-Electronics Technology" Page 1254

Toward the end of Section II of the paper,¹⁰ there is quoted a claim from a Hammond patent of 1932–1936¹¹ which is said to be “the statement of the Proximity Fuse principle.” This is erroneous, for the claim is for a torpedo, meaning in water, from which the energy radiated must be compressional or sound waves, the only kind disclosed in the patent, whereas the proximity fuse employs radio waves.

THE TRIODE TUBE (SECTION III)

The technological revolution that has resulted from the electron tube requires that reports of its onset be rendered correctly. The report given in the paper is from the standpoint of the Hammond group, and while this is a welcome contribution, it is by nature limited and one-sided, and contains some errors. Hence, additional evidence is offered as follows.

The second paragraph quotes our old friend Robert Marriott as having written that de Forest’s Audion “was used to some extent as early as 1906 . . . .” The quotation is correct, but the assumption that it referred to the triode rather than the diode, as of 1906, is in error. It was the two-element tube that was publicly christened that year with the name Audion in de Forest’s AIEE paper of October 20, 1906, devoted to the diode and entitled simply, “The Audion.” Incidentally, Pupin amusingly said of the name: “It is a mongrel. It is a Latin word with a Greek ending!” And he expressed dissatisfaction with de Forest’s inability to explain its modus operandi.¹² In his recent autobiography, de Forest refers to his detector tube as “a carbon filament surrounded by a platinum plate,”¹³ which he used in receiving at 42 Broadway, New York, in 1906.

The grid triode appears to have been invented toward the very end of that year or the beginning of 1907. It was filed upon January 29, 1907 and issued as a patent February 18, 1907, No. 879,532, entitled “Improvement in Oscillation Detectors.” (There was no time lost in those days!) Just ahead of it de Forest had devised another form of triode, one in which the control electrode was a second plate located on the side of the filament opposite to the anode. He sought to make of it a telephone amplifier; but being unable to obtain amplification, contented himself with filing a speculative patent application October 25, 1906. It was issued January 15, 1907, as No. 841,387, entitled “Device for Amplifying Feeble Electrical Currents.” Interestingly enough, the patent drawing shows, connected in series with the input control electrode, a condenser such as became familiar in the grid audion detector. But whereas in the grid tube it enhanced the detector action, it could only do harm in the amplifier, causing it to block. This probably was one reason for the failure of the two-plate amplifier, the other being the lesser electrostatic control of the filament-anode electron path. So, it fell to the grid form of tube to lead the way.

The earliest public disclosure known to have been made of the grid audion detector was in a lecture de Forest gave on March 14, 1907, on “The Wireless Transmission of Intelligence,” before the Brooklyn Institute of Arts and Sciences. It was about that time that the grid audion began to come into some use, “but apparently in very small numbers,” as Marriott said. Among those small users, to the writer’s knowledge, were a few amateurs who learned where the bulbs could be procured, managed to save up the required $5, and then suffered the filament burn-out soon to follow! The writer, with two other amateurs, Austen M. Curtis and Francis A. Hart, had attended de Forest’s Brooklyn lecture and there made the acquaintance of “Doc” and his young assistant John V. L. Hogan. The grid audion detector continued for some years solely as a detector without getting very far, as the authors of the paper indicate, for it was relatively expensive and only moderately more sensitive than the simple crystal-contact detector.

Hammond and Purington are to be congratulated for having at long last pulled aside the curtain of secrecy and revealed Fritz Lowenstein’s contribution in making de Forest’s grid audion detector into an amplifier and also an oscillator. But they do not tell the whole story and give the impression of Hammond’s role having been more than it was. It was not actually “as a consultant of the Hammond Laboratory,” that “Lowenstein on May 11, 1911, undertook in New York the development of the three element ‘ion controller’ . . . .” The date is that of a letter-contract¹⁴ which defined the Hammond-Lowenstein relations, addressed by Hammond to Lowenstein and signed by both, which runs as follows:

May 11, 1911.
Frederick Lowenstein, Esq.,
115 Nassau Street,
New York City.

Dear Mr. Lowenstein:

I am writing you a letter to confirm the points taken up in our conversation this morning at the Hotel Belmont, with Mr. John Hays Hammond and Mr. George Clark. These points, as I recollect, are as follows:

You are to work and develop two separate inventions—one, my own automatic wireless selective system, and the other, your controller.

On account of the development of my wireless selective system, I agree to advance to you the sum of $1,000.00 for your personal supervision, advice and services in the designing and construction of my invention. . . .

In the second proposition, to develop your controller, I shall advance you the sum of $1,000.00 for the purpose of constructing such apparatus as may be necessary for experimentation, with a view to perfecting your controller. . . .

(The term “controller,” or “ion controller,” was Fritz’s alias for the name “Audion” which de Forest had bestowed upon Fleming’s tube upon adding to it the “B” battery.) The rest of the agreement gives Hammond an option to take a 50 per cent interest in the controller invention (which is understood not to have been exercised).

From the above it is evident that Lowenstein was a consultant to Hammond for the latter’s selective radio system for torpedo control, but not in respect to his own ion controller. Hammond’s relation to Lowenstein in respect to the ion controller was that of financial backer, or business partner. This is borne out by the fact that in applying for a patent Lowenstein used his own patent attorney; and when he later sold the patent to the American Telephone and Telegraph Company the payment, of the famous $150,000, was to him. Fritz had tried the grid audion as an amplifier before becoming associated with Hammond. This is indicated in his letter to Hammond of September 19, 1911, quoted in part in the paper, wherein Lowenstein speaks of his “efforts on reproducing the telephone tests of last winter . . . ,”¹⁵ the winter of 1909–1910. What these tests were, fortunately, has been told by Fritz himself in another document, in an affidavit submitted to the Patent Examiner in the course of prosecuting his patent application—that which became the famous negative grid, or “C” battery, Patent No. 1,231,764 of 1912–1917. The affidavit, sworn to on September 8, 1915, is to be found in the Patent Office record of that patent. It describes how interest was first aroused by an experience in 1906 wherein Lowenstein observed the sensitivity to electrostatic influence of a Cooper-Hewitt mercury arc rectifier. It then goes on to say:

In the winter of 1909–1910 I did some work as a consulting expert to the President of the Radio Telephone Company of Newark, N. J., and in their shop and at that time I constructed and assembled an operative telephone system of which the receiving end was arranged substantially as shown in the drawings of my above application, except that the grid 18 was simply connected to one side of the filament 16, i.e., at the same potential as the filament.

Accompanying the affidavit were Exhibits A, B, and C. The latter, a reproduction of the patent drawing, shows a grid Audion provided with audio input and output transformers connected as an amplifier in the receiving leg of a standard telephone subscriber instrument.

The affidavit continues:

In 1911 I did further work on the invention in the way of perfecting details thereof.

(i.e., when backed by Hammond). The affidavit then reveals how Fritz came to substitute for the grid condenser of the audion detector the negative “C” battery which made of it a class A amplifier, which negative grid condition proved to be the invention of the patent.

The apparatus employed in the 1911 tests were constructed with the aid of Benjamin F. Miessner (now at Purdue, Indiana), and George H. Scherff, of New York City.

In a talk in 1911 with Louis Engelhorn at the Chicago end of the line, and Harris Hammond and myself at the New York end, we employed a condenser in the grid circuit and found that the talk over the line was poor, the current through the controller tending to choke or stop. I found that this could be remedied readily by touching the grid binding post repeatedly to discharge the grid and open up the talk. This gave me the idea of connecting the grid to a point ultranegative in potential relative to the filament. Such a connection was actually made in the latter part of 1911 and was found to be a very substantial improvement.

These, then, were the steps by which Lowenstein converted de Forest’s grid audion detector to the magic amplifier it became:

¹⁰ Footnote k, p. 1197.
¹¹ Footnote 1, p. 1197, reference 21.
¹² L. de Forest, “The audion,” Proc. AIEE, vol. 25, pp. 719–747; October 20, 1906. Discussion, pp. 863–873.
¹³ de Forest, op. cit., p. 210.
¹⁴ This letter-contract has kindly been made available to the writer by the authors. It was part of a report by E. S. Purington, “Early History of Selective Receiving Systems with Special Reference to . . . Cases of RCA vs Splitdorf Co., and RCA vs George H. Walker Co.,” p. 22; February 19, 1938.
¹⁵ Footnote 1, p. 1198.