Hazeltine Corporation v. AH Grebe & Co.

21 F.2d 643 (1927)

HAZELTINE CORPORATION et al.
v.
A. H. GREBE & CO., Inc.

District Court, E. D. New York.

June 20, 1927.

Pennie, Davis, Marvin & Edmonds, of New York City (William H. Davis, Willis H. Taylor, Jr., and R. Morton Adams, all of New York City, of counsel), for plaintiffs.

Walter H. Pumphrey, of New York City (Thomas G. Haight, of Jersey City, N. J., of counsel), for defendant.

MOSCOWITZ, District Judge.

These two actions, brought by plaintiffs, Hazeltine Corporation (the owner of the patents) and Independent Radio Manufacturers, Inc. (the exclusive licensee), charge defendant, A. H. Grebe & Co., Inc., with infringement of the Hazeltine neutrodyne patents, No. 1,489,228, granted April 1, 1924, and No. 1,533,858, granted April 14, 1925.

Plaintiffs charge that the so-called "synchrophase" radio receiver, manufactured and sold by defendant since July, 1924, embodies the Hazeltine neutrodyne invention, and is an infringement of claim 1 of patent No. 1,489,228, and of claims 1, 2, 3, 5, 9, 12, 13, and 17 of patent No. 1,533,858.

Claim 1 of patent No. 1,489,228 reads as follows:

"An electric circuit arrangement for neutralizing capacity coupling between the grid and plate circuits of an audion, due to the capacity between the grid and plate electrodes, comprising a coil connected between one of these electrodes and the filament system and an auxiliary coil and a neutralizing capacity connected in series between the other of these electrodes and the filament system, said auxiliary coil being coupled electromagnetically to the first coil and a co-efficient of coupling substantially equal to unity and having a ratio of turns thereto equal to the ratio of the coupling capacity to the neutralizing capacity."

Claim 1 of patent No. 1,533,858 reads as follows:

"The method of neutralizing capacity coupling between the grid and plate circuits of an audion having a transformer in the plate circuit which consists in capacitively coupling the grid of said audion and a secondary of said transformer to cause equal capacity currents to flow to and from the grid whereby such current is prevented from flowing between the grid and the filament system."

Claim 2, patent No. 1,533,858:

"An electric circuit arrangement for neutralizing capacity coupling between the grid and plate circuits of an audion having a transformer in the plate circuit thereof, comprising means for capacitively coupling the grid of said audion and a secondary of said transformer whereby equal capacity currents due to a variation in the potential of the plate of said audion are caused to flow to and from the grid thus preventing such current from flowing between the grid and filament system."

Claim 3, patent No. 1,533,858:

"An electric circuit arrangement for neutralizing capacity coupling between the grid and plate circuits of an audion due to the capacity between the grid and plate electrodes, comprising a coil connected between the plate and the filament system, and an auxiliary coil and a neutralizing capacity connected in series between the grid and the filament system, said auxiliary coil being closely coupled electromagnetically to the first coil and having a ratio of turns thereto equal to the ratio of the coupling capacity to the neutralizing capacity."

Claim 5, patent No. 1,533,858:

"An electric circuit arrangement for neutralizing capacity coupling between the grid and plate circuits of an audion due to the capacity between the grid and plate electrodes, comprising a coil connected between the plate and the filament system, and an auxiliary coil and a neutralizing capacity connected in series between the grid and the filament system, said coils and said neutralizing capacity being so proportioned that variations in plate potential cause equal currents to flow through the coupling capacity and through the neutralizing capacity and prevent such current from flowing between the grid and the filament system."

Claim 9, patent No. 1,533,858:

"In a multistage amplifier, including in each stage an audion and an output transformer having a coil electromagnetically coupled to the primary of said transformer, and a capacity in each stage connected in series with said coil between the grid and the filament system of the audion in that stage, said filament system comprising all points having substantially the same alternating current potential as the filament."

Claim 12, patent No. 1,533,858:

"In a multistage amplifier, including an output transformer having a primary, and an audion in each stage, means for neutralizing the capacity coupling between the grid circuit and the plate circuit of each audion, comprising a capacity connected between the grid of the audion in each stage and a coil coupled to the primary of said output transformer of that stage."

Claim 13, patent No. 1,533,858:

"In a multistage audion amplifier, including an audion in each stage, means for neutralizing the capacity coupling between the grid and the plate circuit of each audion comprising a capacity connected between the grid of each audion and a point in the plate circuit of said audion of opposite alternating current polarity to that of the plate."

Claim 17, patent No. 1,533,858:

"A multistage audion amplifier, comprising an audion in each stage, means for neutralizing the capacity coupling between the grid and plate circuits of each audion, and means for preventing at all frequencies substantial coupling between any two stages thereof, except for the conductive coupling on which the amplifying action of the audion depends."

The defendant contends that claim 1 of patent 1,489,228 is invalid because it embodies the principles of the Rice patent, 1,334,118, and the Armstrong patent, 890,593. The subject-matter of the first patent in suit is a special application of the auxiliary neutralizing circuit to neutralizing the capacity coupling between the grid circuit and the plate circuit of an audion, characterized by the fact that the neutralizing coil is coupled to the coil in the neutralized circuit with a coefficient of coupling substantially equal to unity.

The Rice patent, 1,334,118, has for its object the avoidance of the undesired production of oscillatory currents between grid and plate circuit. In order to accomplish this result, Rice proposed to neutralize the electromagnetic coupling by a second electromagnetic coupling in the opposite direction. Rice never had the idea of complete neutralization. Rice's purpose was to prevent oscillations in a regenerative amplifying receiver, whereas Hazeltine's idea was to eliminate all the regenerative or feedback effect between the plate and the grid circuit of an audion. In other words, Rice intended to avoid oscillation, but his receiver still remains a regenerative receiver, giving inferior selectivity and tone quality, and producing howls and squeals during the process of adjustment from one broadcasting station to another. Hazeltine, by providing close coupling between the coils in unequal turns, produced permanent neutralization for all frequency, a result that neither Rice nor the prior art disclosed.

Judge Thacher in Hazeltine Corporation et al. v. Electric Service Engineering Corporation, 18 F.(2d) 662, decided: "Rice employed a fixed ratio of equal capacities and equal turns, and arranged his coils with loose coupling. Hazeltine provided close coupling between the coils and unequal turns, with capacities in the ratio stated, and attained permanent neutralization for all frequencies, a result never attained before, and one which had, as the evidence discloses, an astounding effect upon the entire industry. * * * Rice and Hazeltine were not very far apart, but the difference between them is the difference between success and failure."

The defendant's witness Morecroft testified that the nearest approach that can be found to Hazeltine is by taking the disclosure of Rice and using it with the close coupling of Armstrong. "The validity of a patent is not to be defeated by simply showing that all the elements of a patent are old, where they do not show all of the elements of the claim in combination, and this the defendant has not shown. To find in the prior art each element in isolation is not to anticipate the work of a patentee, who by an inventive act first evolves a new combination of these elements which by their conjoined functions produce a new result." Benthall Machinery Co. v. National Machine Corp. (D. C.) 222 F. 918; Western Electric Co. v. Millheim (C. C.) 88 F. 505.

Professor Hazeltine, in discussing the reason for the parasitic oscillation in the Rice patent, stated "that, due to the lack of close coupling between those two halves of coil 4 (see Fig. 1), each current will produce in its half a certain voltage. It will produce in the other half a lower voltage. The voltage in any one-half due to its current is not annulled by the presence of the other current, and there will remain a residual voltage between the end of the coil and the tap. That residual voltage transfers energy from the plate circuit into the grid circuit, and, if it is sufficient, as it may well be in a practical case, will produce oscillations. The remedy for that which is in no way suggested by Rice is the close coupling between two halves of coil 4. With such close coupling the voltage induced in one half by the current in the other half would exactly balance the voltage due to the current in the same half. These two voltages balancing one another, if they occur separately, would actually result in zero voltage, and then we will have no feedback of energy and no production of oscillation."

Not only does the testimony of Professor Hazeltine distinguish the Rice and Hazeltine patents, but Professor Pupin, of Columbia University, in an affidavit attached to the file wrapper of the Hazeltine patent, 1,489,228, stated the difference between the Rice and Hazeltine patents as follows:

"From a superficial point of view it looks as if both Hazeltine and Rice used the same instrumentalities, but fundamentally they do not. Where is the difference? That is the question. In answer to this question it is well to refer to Fig. 1 of the Hazeltine application and Fig. 1 of the Rice patent. In Hazeltine's figure it will be seen that the whole system of electrical conductors is divided into two parts by an axis of electrical

symmetry, consisting of the conductor, in which are located the impedance Z, the energizing batteries, and the lead to the junction of the coils L1 and L2; on one side of this axis of symmetry we have the coil L1 and the inherent or natural capacity C1, and on the other side we have coil L2 and the added capacity of C2. Correspondingly in the Rice figure, the secondary system is also divided into two parts by an axis of electrical symmetry, consisting of the conductor, in which are located the tuned circuits 6, 7, the telephone, the energizing batteries and the lead to the mid point of coil 4; on one side of this axis of symmetry we have the upper half of coil 4 and the inherent or natural capacity 12, and on the other side we have the lower half of coil 4 and the added capacity 13.

"This is as far as the superficial similarity goes. Now comes the radical difference. In the Rice patent the electromagnetic coupling between the two halves of the coil 4 is not even mentioned. In the Hazeltine application the coupling is stated to be `closed' and is indicated in the mathematical proportions of the application to be one hundred per cent. Hence the neutralization of a capacity coupling and the consequent elimination of feedback action aimed at by both inventors will be produced by Hazeltine alone. Rice secures only partial neutralization when he produces any at all. Whereas with the `close' coupling of the Hazeltine arrangement the added capacity C2 can be adjusted once for all to give neutralization which is not disturbed by adjustments of the circuits.

"Since Rice eliminates the feedback action only partially, when he eliminates it at all, it is obvious that oscillation of the vacuum tube will cease to be prevented as soon as the amplifying power passes beyond a certain definite lower limit; whereas Hazeltine, with complete neutralization of capacity coupling, may carry the amplifying power to any practicable extent without producing feedback action. Hazeltine also shows an arrangement not present in the Rice device, which is apparent from even a most superficial examination — Fig. 2 of the application. In this figure we have a coupling capacity between the grid and a coil electromagnetically coupled to a coil in the plate circuit — an arrangement which under certain conditions offers substantial advantages which is not even suggested in the Rice patent."

Professor Morecroft, writing in the "Radio Broadcast" of March, 1924, stated that the credit of neutralizing the tendency of high frequency amplifiers to oscillate rightly belonged to Hazeltine. Professor Morecroft continued:

"This type of set (referring to Hazeltine) is now on the market under the name he has chosen for it, the neutrodyne. This new set has been much in favor because it requires no fussy adjustment of rheostats or regeneration to make it perform at its best. A simple chart of condenser settings, once obtained for all the stations within the range of the set, enables one to tune in on them at any time and get them, if they are sending; no hair-splitting adjustment is necessary before the operator feels that the set is doing all it can to bring the desired signal."

Morecroft, according to his own articles, realized the vast importance of Hazeltine's invention to radio. The invention of Hazeltine completely revolutionized the radio industry, by eliminating the bothersome noises and squeals.

The Independent Radio Manufacturers, Inc., entered into an agreement with Hazeltine whereby they became the exclusive licensees under the Hazeltine patents and had the exclusive rights to use the words "neutrodyne," "neutroformer," and "neutrodon" for electric circuits, radio apparatus, transformers, and radio condensers. Between January 1, 1924, and May 21, 1926, the Independent Radio Manufacturers paid to the Hazeltine Corporation $1,329,000, representing 5 per cent. of the manufacturers' selling price of neutrodyne sets, which aggregated $26,000,000; $55,000,000 was paid by the ultimate consumer for radio sets manufactured by the Hazeltine licensees.

The Hazeltine invention was a great advance made in the art, an advance from the prior art that received the approval of the Patent Office and is entitled to the presumption of validity. "Knowledge after the event is always easy, and problems once solved present no difficulties — indeed, may be represented as never having had any — and expert witnesses may be brought forward to show that the new thing which seems to have eluded the search of the world was always ready at hand and easy to be seen by a merely skillful attention. But the law has other tests of invention than subtle conjectures of what might have been seen and yet was not. It regards a change as evidence of novelty, the acceptance and utility of change as a further evidence even as demonstration." Diamond Rubber Co. of New York v. Consolidated Rubber Tire Co., 220 U.S. 428, 31 S. Ct. 444, 55 L. Ed. 527.

Professor Morecroft, in order to prove that the Rice circuit was similar to Hazeltine, demonstrated a radio receiver with an amplifier made by combining two audion tube circuits, arranged essentially as in Rice, Fig. 1. According to Mr. Hazeltine, the circuit was not arranged essentially as in Rice, but was arranged so as to prevent parasitic oscillation.

Professor Morecroft used Rice split coil 4 and the neutralizing condenser 13 in the grid circuit, but did not include in the circuit of his apparatus the tuning elements illustrated in the plate circuit of the Rice patent. The tuning element in the plate circuit of the Rice patent consists of coil 6 with a middle tap arranged in parallel with a condenser, so that the plate circuit may be tuned in accordance with the characteristic idea and mode of operation of the Armstrong regenerative receiver. Professor Morecroft put in the plate circuit of his apparatus a single coil 3, with no tuning condenser and no middle tap, and the coil "was made up physically with a very much smaller number of turns than would be used in such a coil as 6 of the Rice patent." Professor Morecroft modified the natural and proper arrangement of Rice, in that he distributed the turns of his coil 3 under the entire length of the coil on the next grid circuit with which coil 3 is associated.

Hazeltine, in order to show that this arrangement was for the purpose of preventing oscillation, removed the load and oscillation occurred. The tests of Professor Morecroft showed that, in order to prevent parasitic oscillations in the Rice arrangement, specific modifications had to be made in the Rice arrangement.

The Armstrong patent, 890,593, has for its object the prevention of the disturbing effects of static induction in telegraph and telephone wires. This object is accomplished by maintaining the catenary which is employed in the suspension of trolley wires at a potential approximately equal, but opposite, to that of the trolley wire. This patent fails to disclose any of the benefits of neutralization disclosed in the Hazeltine patent.

The Hazeltine patent, 1,533,858, relates to the neutralization of capacity coupling between the grid and plate circuit of an audion, which capacity coupling results in undesirable reactions of the plate circuit on the grid circuit. Capacity coupling between grid and plate circuit sometimes produces oscillation. To prevent these oscillations Hazeltine provided an auxiliary circuit, electromagnetically coupled to one of two original audion circuits, called a first circuit, and capacitively coupled to the other or second circuit. If there is any disturbance in the second circuit, the disturbance, because of the capacity coupling, will flow in the first and the auxiliary circuit. The electromagnetic coupling is so arranged between the first and auxiliary circuit that the magnetic effect of these two circuits will neutralize each other and will result in no voltage across the first circuit.

The defendant contends that the Hazeltine patent is invalid because it embodies the principles of the Hartley patent, 1,183,875, and the Colpitt reissue patent, 14,380. Hartley had for his purpose the elimination of the regenerative feedback. He effected his purpose by "introducing still another electromotive force into the input circuit and so adjusting it as to amplitude and phase that it shall annul the effect of the electromotive force introduced by the first-mentioned unavoidable coupling inside the audion itself. He utilized the built-up radio frequency energy in the plate circuit by taking a portion of it, reversing it in direction, and applying this reversed energy to the grid circuit, so as to oppose the regeneration or amplifying energy fed back through the tube. His apparatus comprised the auxiliary circuit, including the coil shown in Figures I and II. Thereby radio frequency energy is taken from the plate circuit, reversed in direction, and transferred to the grid circuit by means of a magnetic (inductive transformer) coupling between coils 15 and 16 shown in Figures I and II. The energy thus fed back opposes the energy fed through the tube by means of plate 13 and grid 12." Radio Corporation of America et al. v. Twentieth Century Radio Corporation (decided by Circuit Court of Appeals, Second Circuit, May 2, 1927) 19 F.(2d) 290.

While Hartley's invention discloses that oscillations may be prevented by transferring from the plate to the grid circuit by a path around the tube an electromotive force in a direction to oppose the regenerative electromotive force through the tube, it fails to disclose a neutralizing condenser, or any means for neutralizing wholly within the plate circuit.

The Colpitt reissue patent, 14,380, has for its object the simplification of the mechanism and the strengthening of the apparatus required for the production of the modulated oscillation. The statement in the Colpitt patent is to the effect that the device used may be a repeater of the audion type, with the output circuit connected with the input circuit, which, with the proper adjustment of the devices in the circuit will result in the production of sustained frequency oscillations. If the potential in the input circuit is varied, that is, by connecting a transmitter or other device for sending a message in circuit with one of the electrodes, for example, the grid of the audion, then the sustained oscillations are modulated in accordance with the messages sent. The arrangement of a coil in series with a capacity in an audion plate circuit produces regenerative feedback and not neutralization. There is no suggestion in this patent that would disclose the idea of complete and permanent neutralization within the plate circuit, as claimed in the Hazeltine patent.

In a broadcast radio receiver the objectionable disturbances arise in the plate circuit, find their way in the grid circuit, where they are reamplified, repeated again into the plate circuit, and so on progressively in accordance with the regenerative principle, until squeals and howls are produced. Hazeltine's patent, 1,533,858, aware of this defect, neutralized the disturbances within the plate circuit without allowing them to get in the grid circuit at all. To accomplish this result, Hazeltine provided an auxiliary circuit electromagnetically coupled to one of two original audion circuits, which is the first circuit, and capacitively coupled to the second circuit. If the disturbing voltage exists in the second circuit, the capacity coupling will cause currents to flow in the first circuit and in the auxiliary circuit. Between the auxiliary circuit and the first circuit the electromagnetic coupling is so arranged that the magnetic effects of these two currents will neutralize each other, and so will result in no voltage across the first circuit.

In the prior art audion receivers, the troublesome capacity coupling between the grid and plate circuit through which disturbances arising in the plate circuit were able to pass into the grid circuit, resided in the audion itself. The plate of the audion and the grid of the audion present extended surfaces capable of holding an electric charge like the two plates of an electrical condenser. Where an electrical condenser is interposed in an electrical circuit carrying an alternating current, energy can be transferred across the gap between the two condenser surfaces; that is, between the plate and the grid. Even though there is no physical connection between the two surfaces, the alternating electrical potential, as it increases and decreases on the plate, will induce another alternating electrical potential, which similarly increases and decreases on the grid. By this electrical action across between the plate and grid the objectionable disturbances pass from the plate circuit into the grid circuit.

Because of the effect of this inherent capacity coupling residing within the audion itself, Hazeltine set out to neutralize the plate circuit. In chart 4, which is a reproduction

of Fig. 6 of the patent and the defendant's circuit, the capacity coupling is represented by a dotted line connection including the two plates of a condenser marked C1." The two plates of the condenser are the grid and the plate. C1 represents the inherent capacity coupling to be neutralized. Hazeltin's plate circuit neutralization patent was intended to cause an equal and opposite capacity current to flow in the plate circuit away from the grid, so that the disturbing capacity current was diverted within the plate circuit and rendered harmless. Hazeltine accomplished the result by connecting or coupling the grid through another capacity, just like the capacity to be neutralized, to a coil of the electromagnetic transformer, which exists in the plate circuit of a radio frequency amplifier as one of the necessary elements thereof. C2 represents the added neutralizing capacity, and is represented in the defendant's apparatus as a small neutralizing condenser. The neutralizing coil in the transformer is represented by L2 in the chart and in the defendant's apparatus it is a coil closely coupled to the plate circuit transformer coil by being interwound therewith. With the Hazeltine arrangement the disturbance does not as in the old radio receiving sets, flow from the circuit plate into the grid circuit, and there set up amplification, resulting in howls and whistles, but away from the grid around through the wire which leads to condenser C2 through that condenser and through coil L2 in the plate circuit. Without condenser C2 and auxiliary coil L2, complete neutralization within the plate circuit would be impossible. The defendant's apparatus copies these two features of Hazeltine for the purpose of complete neutralization.

Claim 3 adds to claim 2 that the capacity coupling to be neutralized is that coupling which is due to the capacity between the plate and grid electrodes. Claim 3 states that the neutralizing coil may be any coil connected in the plate circuit provided that the coil is "closely coupled electromagnetically to the first coil and having a ratio of turns thereto equal to the ratio of coupling capacity to the neutralizing capacity." In the defendant's apparatus the neutralizing coil is a secondary coil of the plate circuit transformer, and it is closely coupled to the primary transformer coil. Claim 3 continues: "A ratio of turns thereto equal to the ratio of coupling capacity to the neutralizing capacity." In the defendant's receiver the ratio of turns of the neutralizing coil to the number of turns of the primary of the transformer is equal to the ratio of the coupling capacity to the neutralizing capacity.

Claim 5 sets forth "an electric circuit arrangement for neutralizing capacity coupling * * * between the grid and plate electrodes comprising a coil connected between the plate and filament system." The coil in the plaintiff's apparatus is L1 of chart 4, and is one of the two interleaved coils that Hazeltine found when he dissembled the middle ratio frequency transformer of the defendant's set. Claim 5 proceeds: "An auxiliary coil and a neutralizing capacity connected in series between the grid and filament system." L2 in Fig. 6 of the patent and in chart 4 is the auxiliary coil, and is one of the two interleaved coils in the defendant's receiver. C2 is the neutralizing capacity of the plaintiff's device, and is the small condenser with the mica insulation in the defendant's device.

Claim 5 continues: "Said coils and said neutralizing capacity being so proportioned that variations in the plate potential cause equal currents to flow through the coupling capacity and through the neutralizing capacity, and prevent such currents from flowing between the grid and the filament system." The proportioning is done in adjustment of neutralizing capacity. The defendant used simplest proportions, 1 to 1.

Claims 9, 12, 13, and 17 define a multistage amplifier. According to Professor Hazeltine's definition, a multistage amplifier is consecutive vacuum tubes or audions between which are connected the coupling transformer. If there was one vacuum tube ahead of the detector, it would have a one-stage amplifier. If there were two such vacuum tubes ahead of the detector, then it is a two-stage amplifier and any number of stages from two up would be a multistage amplifier.

Claim 9 states: "In a multistage amplifier, including in each stage an audion and an output transformer having a coil electromagnetically coupled to the primary of said transformer." In the defendant's device the output transformer associated with the vacuum tube is the binocular transformer, and the primary is one of the interleaved coils and the coil electromagnetically coupled to the primary is the other one of the interleaved coils.

Claim 12 calls for "in a multistage amplifier, including an output transformer having a primary and an audion in each stage." In chart 5. L1 is the primary and is associated with one of the vacuum tubes or audions. Claim 12 continues: "Means for neutralizing the capacity coupling between the grid circuit and the plate circuit of each audion, comprising a capacity connected between the grid of the audion in each stage and a coil coupled to the primary of said output transformer of that stage." The capacity is C2, which connects the grid of the vacuum tube directly to L2, which coil is coupled to the primary of the output transformer, the primary being represented by L1.

Claim 13 is as follows:

"In a multistage audion amplifier, including an audion in each stage, means for neutralizing the capacity coupling between the grid and the plate circuit of each audion comprising a capacity connected between the grid of each audion and a point in the plate circuit of said audion of opposite alternating current polarity to that of the plate."

According to chart 5, C2, the capacity, connects at its left-hand side directly with the grid and connects at its right-hand side with the terminal of L2, which is essentially a part of the plate circuit. L2 is connected with L1, which is the primary of the transformer, and which is directly between the plate and the filament system. In order to secure the polarity, as stated in the claim, the lower terminal of L2 is of opposite polarity to the upper terminal of L1; that is, the upper terminal of L1, which goes to the plate, is opposite to the lower terminal of L2, which goes to the neutralizing capacity.

Claim 17 calls for a multistage arrangement, with means for neutralizing the capacity coupling between the grid and plate circuits of an audion. Claim 17 continues: "Means for preventing at all frequencies substantial coupling between any two stages thereof except for the conductive coupling on which the amplifying action of the audion depends." In Hazeltine's arrangement this element is found by inclosing each audion circuit in a conductive screen, and in the defendant's device it is an arrangement of binocular coils which prevents electromagnetic coupling from one stage to another.

The defendant's contention, that patent 1,489,228 cannot be interpreted to include a degree of coupling which was excluded from it during the proceedings in the Patent Office, is true, and, if the defendant were to use the degree of coupling in the Rice patent, it would not infringe. But the defendant's neutralization is effected in the plate circuit, as in Hazeltine, rather than in the grid circuit, as suggested in the Rice patent. The defendant does not use the split coil 4 of Rice, but uses instead the structure of interleaved turns, for the purpose of effecting a closer coupling than the split coil admits.

The defendant's apparatus contains the elements of claim 1 of 1,489,228. Claim 1 reads: "An electric circuit arrangement for neutralizing capacity coupling between the grid and plate circuits of an audion, due to the capacity between the grid and the plate electrodes comprising a coil connected between one of these electrodes and the filament system." In the defendant's receiver the coil is one of the interleaved coils connected between the plate and filament system." Claim 1 continues: "And an auxiliary coil and a neutralizing capacity connected in series between the other of these electrodes and the filament system."

In the defendant's device that auxiliary coil in the other one of the interleaved coils connected to a small mica condenser, which is the neutralizing capacity, and the two are connected in series between the grid, which is the other electrode and filament system. Claim 1 continues: "Said auxiliary coil being coupled electromagnetically to the first coil with a coefficient of coupling substantially equal to unity and having a ratio of turns thereto equal to the ratio of coupling capacity to the neutralizing capacity." The ratio of turns is 1 to 1 in this case, and to be effective the ratio of capacities will be 1 to 1 when adjusted. In the defendant's receiver the close coupling is coupled with a degree of coupling substantially equal to unity.

The defendant winds the two coils interleaved, which is the closest possible coupling which can be obtained with coils of the defendant's type, and is effective as unity coupling would be, and is substantially equal to unity. The defendant contends that "substantially equal to unity" means as close to unity as it is physically possible to get. To give the language of the claim such an interpretation would be ignoring the word "substantially" of the claim, and at the same time would destroy the whole significance of the claim. After many conferences with the Patent Office, Hazeltine changed the wording of the claim from "said auxiliary coil being closely coupled electromagnetically to the first coil," because the examiners said: "Since Rice surely discloses some amount of coupling, applicant appears to be entitled to claim only such degree of coupling as clearly falls without the scope of that patent, said degree to be defined in unmistakable and definite terms; `closely' being not only indefinite but also insufficient."

The word "closely" was canceled from claim 1 before the words "coupled electromagnetically to the first coil," and after these words the expression, "with a co-efficient of coupling substantially equal to unity," was inserted, and the amended claim read: "Said auxiliary coil being coupled electromagnetically to the first coil with a coefficient of coupling substantially equal to unity." In view of the Patent Office's correspondence, claim 1 must be read to exclude the prior invention of Rice, and must be read so as to give the claim an interpretation which measures the real contribution of the inventor to the art.

Professor Hazeltine in 1918 made a contract to enter the employ of the United States as consulting engineer in connection with radio matters. He was assigned to the development of radio receivers known as "Type S. E. 1420." While in the employ of the government, Hazeltine devised during November and December of 1918 a circuit for "neutralizing capacity coupling," which work was done at the radio test shop in the Navy Department at Washington. In 1919 Lt. Commander Loftin, who had been assigned to the Bureau of Engineering of the Navy at Washington, learned what Hazeltine had done. As a result of correspondence Loftin filed on August 7, 1919, an application for a patent in Hazeltine's name for a circuit for neutralizing capacity coupling. The application for said patent, which resulted in patent 1,450,080, stated:

"This application is made under the Act of March 3, 1883, chapter 143, Stat. (U. S. Statute XXII, p. 625), and the invention herein described and claimed may be used by the government of the United States or any of its officers or employees in the prosecution of work for the United States or by any person in the United States without the payment of any royalty thereon."

In June, 1920, in a letter to Lt. Loftin, Hazeltine stated that all papers bearing on the application be returned to him, as he was arranging with his own attorney to prosecute the application, because "prospective commercial developments make it very desirable that these cases be promptly acted upon." Mr. Hazeltine's personal attorney conducted the prosecution of the application, and the reference to the act of 1883 (Comp. St. § 9441) was stricken out of the application.

The defendant's contention that the requests for the benefit of the act of 1883 constituted a dedication to the public is untenable. The defendant acquired no rights under said act because the patent was not issued under it. The reference to the act of 1883 was stricken out before the patent was granted. Hazeltine Corporation et al. v. Electric Service Engineering Corporation, supra. In neither patent in suit was application made for the benefit of the act of 1883.

The claims of the Hazeltine patents are not coextensive. The subject-matter of patent 1,450,080 is an unsymmetrical arrangement of closely coupled neutralized circuit generally with maintenance of the proper ratio of turns to capacities. The subject-matter of the second patent, Hazeltine patent, No. 1,489,228, is the special application of auxiliary neutralizing circuit to audions characterized by couplings substantially equal to unity. The subject of the Hazeltine patent, No. 1,533,858, is an audion plate circuit neutralization.

Since the claims of the different patents are not coextensive, the law does not require that all be claimed in the same patent, but may at the option of the patentee be secured by different patents. Thomson-Houston Electric Co. v. Elmira & H. R. Co. (C. C. A.) 71 F. 396. The case was reopened on the plaintiff's motion on the 12th day of November, 1926, and there was introduced in evidence the British patent, 254,472, to the Western Electric Company, granted July 8, 1926. "The invention relates to electric wave translations systems, and more particularly to space discharge tube systems, wherein the production of undesired oscillations is prevented." This patent confirms the testimony of Hazeltine with reference to the Rice arrangement.

The case was again reopened on the defendant's motion and on December 2, 1926, there was introduced in evidence the Carl O. Weber patent, No. 1,605,411, granted November 2, 1926. This patent relates to the use of the audion as an amplifier, with deliberate elimination of regenerative oscillations. This patent was put forward by the defendant to support its contention of noninfringement. Weber in his patent claims a tap-off, but the defendant uses an auxiliary coil and not a tap-off. The Circuit Court of Appeals, Second Circuit, in Radio Corporation of America et al. v. Twentieth Century Radio Corporation (decided on May 2, 1927) 19 F.(2d) 290, held that the Hartley and Rice patents were valid.

I find that the close coupling of the Hazeltine patent, 1,489,228, was a decided advance in the art, an advance that produced complete and permanent neutralization, an advance that could only be secured by adding the disclosures of Armstrong trolley wire patent to the disclosure of Rice. In the Hartley patent the Circuit Court of Appeals held it was an invention for Hartley to create a path around the tube through which may be fed an electromotive force in a direction to oppose the regenerative electromotive force through the tube. There is no suggestion in the Hartley patent that in any way discloses a plate circuit neutralization suggested in Hazeltine's patent 1,533,858.

The patents are valid and the claims have been infringed by the defendant. Decree may be entered in favor of the plaintiffs with the usual order of reference. Settle decree on notice.