Opinion
October 14, 1924.
Charles Neave and Stephen H. Philbin, both of New York City, for plaintiff Westinghouse Electric Mfg. Co.
L.F.H. Betts, of New York City, for plaintiff Radio Corporation of America.
Charles J. Holland, of New York City, for defendant.
In Equity. Suit by the Westinghouse Electric Manufacturing Company and the Radio Corporation of America against Morris Taub, trading as the Atlantic Pacific Radio Company. On motion for preliminary injunction. Granted.
Defendant has put upon the market and sold a superheterodyne radio receiving set, which plaintiffs claim to be an infringement of Fessenden patents, Nos. 1,050,441 and 1,050,728, and of the Armstrong patent, No. 1,113,149. Each of these patents has been adjudicated and sustained. They are now owned or controlled by plaintiffs.
The two Fessenden patents relate to heterodyning; one being for the method, and the other for the apparatus. Concededly defendant's receiving set is not in form the apparatus shown in the Fessenden drawings. Nevertheless infringement of both patents is charged. The Armstrong patent relates to an arrangement of audion circuits on which oscillating current energy is transferred from the output or plate circuit of an audion to its input or grid circuit, to sustain the oscillations in the grid circuit. See Armstrong et al. v. De Forest (C.C.A.) 280 F. 596.
Heterodyning, as applied to radio, is the method by which is brought about the production of signals by beats whose frequency is the difference between that of a received current and that of locally produced oscillations. See Kintner v. Atlantic Communication Co. (D.C.) 241 F. 956.
Defendant's alleged infringing device, when in operation, receives a wave length of say 1,000,000 radio frequencies per second. Before such wave is "detected," it is subjected to the opposition, or, to use the nomenclature of Fessenden, the "interaction," of locally produced audion oscillations of say 950,000 or 1,050,000 frequencies per second. The difference between the number of received frequencies and those locally produced results in a beat frequency of 50,000 cycles. This frequency is then passed through appropriate circuits, whereby it is amplified and detected in such manner that the signals carried by it may be heard through the use of telephone receivers. The means which defendant employs to amplify or regenerate the beat frequencies are claimed to infringe the Armstrong patent.
In the Kintner Case, where an oscillating audion was held to infringe upon the method of generating local oscillations as shown by Fessenden, Judge Mayer said: "The beat system in acoustics was old and well known, but Fessenden was * * * the first to apply this principle to signaling in the radio art. * * * He made, in the best sense, a new contribution to the knowledge of the time; for nowhere and by no one had there been even a suggestion of the applicability of the `beats' principle to radio."
The court regarded the Fessenden invention as being of a high order, and entitled to a broad scope of equivalents. That a similar view was held by the Circuit Court of Appeals is indicated by what was said in International Signal Co. v. Vreeland Apparatus Co., 278 F. 468. The division in the court, which there occurred, was not founded on any difference of opinion as to the merit of the Fessenden patents, and the scope to be accorded them, for as to that the judges were in accord, and even the majority felt that the Vreeland patent, if permitted to stand, would infringe Fessenden. The division arose as to whether, in the presence of the Fessenden patent, the patent to Vreeland, which doubtless improved Fessenden, should be permitted to stand.
So broad and comprehensive is the scope given to the Fessenden patent in the Kintner and Vreeland Cases, that it can no longer be persuasively argued, as is here attempted, that the disclosures of Fessenden can only cover a specific device for utilizing the heterodyne principle in a particular way. While it is true that his apparatus is inoperative for voice reception, and is suitable only for code, and that to accomplish the end sought to be served he attached a diaphragm to a coil, and set up a magnetic field between such coil and another, and that the interaction between the constancy of one coil and the variations induced in the other produced an actual physical motion of one coil towards or away from the other, and thus impressed a resultant upon the diaphragm, there is no doubt that he employed the "beat" principle.
For the reasons just stated, it was undertaken to limit the Fessenden patents in the Vreeland Case. The court refused to do so, and in his dissenting opinion, Judge Manton called attention to Vreeland's admission that there is no distinction between the fields and currents or between superposition or combination of currents and the interaction of fields. Such an admission, made long before the present litigation arose, and in connection with a principle here involved, is helpful in considering the weight to be given to defendant's contention that there is substantial difference between the production of a beat current through the intermingling of two currents of different frequencies, and the "interaction" of magnetic fields of two coils, one of which is operated by a current of frequencies differing from that which operates the other.
According to defendant his superheterodyne receiving set is one in which "the high frequency waves coming from a broadcast station are brought in through the antenna and through the first tube. The second tube produces an independent train of waves of high frequency. The plate current of the first tube and the waves produced by the second tube produce a third wave, which is sent through the so-called filter system and the amplifier system, consisting of three tubes and three transformers, around which has been connected a neutrodon to prevent the natural regeneration which is unavoidably present in most radio circuits employing radio vacuum tubes. The filter system determines the frequency of the waves which pass through the amplifier."
In considering this description of the operation of defendant's device, it is to be recalled that one of the primary objects of Fessenden was "to utilize the interaction of forces produced by a continuously maintained stream of oscillations as they are received, with forces produced at the receiving station by oscillations practically continually produced by a local source." Surely, this object is attained by defendant. It is also the fact, I think, that defendant's third wave, which is made available for filtration, is the resultant of the "interaction," the "superposition," or the "combination" of the incoming wave with that which is locally produced. Unless such resultant wave is of a frequency capable of passing through the filter system, the receiving set, as I understand it, will not operate. From this I should say that, while the filter system determines the frequency of the waves to be amplified, the local oscillator is used to determine the frequency of the wave that will pass the filter.
If this conclusion be sound, the operation of defendant's set squarely reads upon claim 2 of the Fessenden patent, No. 1,050,728, which is in these words:
"In the art of signaling, the method which consists in making an indication by the interaction of received impulses of sustained frequency and amplitude with impulses of neighboring frequency generated by a constantly acting local source of energy at the receiving station."
There is also a reading on claim 3 of Fessenden patent, 1,050,441, which is:
"A signaling system having in combination at a receiving station, a receiver and a constantly operating frequency determining element having a frequency differing from that of the received impulses to such an extent as to cause beats to be formed at the station on the receipt of transmitted impulses."
While the mechanical means disclosed by Fessenden and those used by defendant are different, and although voice reception was not possible through the use of the devices shown in the drawings of Fessenden, I find myself unable to say, taking into account the natural development of the radio art, that there is any essential difference between the methods and means of signaling disclosed by the patent and those employed by defendant.
Alleged Infringement of the Armstrong Patent.
In Armstrong v. De Forest Radio Telephone Telegraph Co., supra, the court, at page 596, said that "the patent covers the use of the feed back circuit, and applies even though the claims are in terms limited to a receiver." It was also said that "the inventor is entitled to the benefit of all uses to which his invention can be put, no matter whether he conceives the idea of the use or not."
Claim 9 of the Armstrong patent employs this language:
"An audion wireless receiving system having a wing circuit interlinked with a resonant grid circuit upon which the received oscillations are impressed, and an inductance through which the current in the wing flows, the grid circuit including connections for making effective upon that circuit the potential variations resulting from a change of current in the wing circuit."
Defendant's receiving set is claimed to infringe Armstrong in these particulars:
(1) Through the control of the amount of regeneration to be produced in the wing or plate circuit of the audion tube D1 by means of the variometer I²;
(2) In the circuits including the coils I3 I4 in the local oscillator system; and
(3) In the circuits including the condenser C4.
Now, the dial by which the variometer in the first instance of alleged infringement is varied is marked by the word "regeneration," and defendant's advertisements, in describing the superiority of his device, employ this explanation:
"(9) Tuning plate of first tube gives additional regenerative short wave radio frequency amplification."
Confronted with these statements — which are in the nature of admissions — a failure to find that regeneration takes place is most difficult, particularly when an expert such as Hogan positively declares that it does occur, and in infringement of Armstrong.
It is no doubt true, as said by defendant, that the variometer is a tuning instrumentality, and for such reason it is probable that its primary purpose is not for the regeneration of current. These circumstances, however, do not disprove that regeneration does not take place, nor that it is not used. But frankly the affidavits of the experts leave me in some doubt as to just how the regeneration in this particular circuit takes place, and, if it does occur, as to whether it is fairly within the claims of the Armstrong patent. Hogan admits that in the circuit of D there are actually present at the same time currents of three different frequencies. Clarkson, defendant's expert, says that under such circumstances a feed back will not occur, due to the fact that there is no resonance between them.
Because of this conflict of opinion, I shall, for the present, withhold from plaintiffs the relief they ask upon this alleged infringement of defendant's variometer circuit.
The Oscillator Tube Circuit.
Defendant admits the use of a feed back circuit composed of the oscillator tube, the variable condenser C3, the coils I3 and I4 and the B battery. The defense is that the feed back is not covered by the Armstrong patent, and that any claim he may have to an oscillating circuit is embodied in an application made to the patent office some time after the application for the patent in suit. From this it is argued that no adjudication has been had which sustains any patent rights in Armstrong for a feed back oscillator. Such contention would need to be seriously regarded were it not that the Circuit Court of Appeals for this circuit in the Armstrong-De Forest litigation, 280 F. 596, gave a broad interpretation to the feed back invention, using this language:
"It is essentially useful as a radio receiver, and has proven to be a wonderful improvement on the old audion receiver. The appellant uses the feed back circuits in a radio transmitter, and the patent covers the use of the feed back circuit, and applies even though the claims are in terms limited to a receiver."
Being bound by the interpretation given to the patent in suit by the Appellate Court of this circuit, I do not see that I can do otherwise than hold that defendant's oscillator unit infringes upon Armstrong. Unless and until the decision of the Circuit Court of Appeals is retracted or modified, or unless there be a contrary decision by the Supreme Court, I can give no heed to the recent decision of the Court of Appeals of the District of Columbia which awarded priority in the invention of the oscillator to De Forest.
There remains the question as to whether defendant's neutrodon condenser circuit C4 infringes. Here plaintiff says that by means of the variable condenser C4, as well as by the relation of the system's coils and wiring, the inherent coupling in the audion through the capacity relations between the plate and grid is supplemented, so that additional radio frequency energy is transferred from the plate circuits to the grid circuits of the tubes RA1, RA2, and RA3.
Defendant claims that the condenser C4 is nothing but a neutrodon, and is for the express purpose of preventing feed back or regeneration within the tubes of the intermediate frequency amplifier. The argument is that, if oscillation of these tubes be permitted, they will fail to amplify and would thus prevent the transmission of signals to the loud speaker or head phones. Although admitting that a neutrodon may be used for preventing a feed back of oscillation from a plate to a grid circuit, when utilized with reversed coupling coil, Hogan asserts that no such hook-up is present in defendant's set. The definite statement is made that the neutrodon, as used by defendant, cannot prevent feed back.
The purpose it actually serves is open to some question, but this much is certain: There is a feed back of energy from the plate circuit of the amplifier tube R3 to the grid circuit of amplifier tube RA1. The circuit is one in which the operative features of the receiving set require some amplification, and in its production I am reasonably certain the defendant employs the method shown by Armstrong.
The plaintiffs may have an injunction as to the matters upon which the decision is in their favor. In other respects the prayer for relief is denied. If the issuance of a restraining order should hereafter be held to be erroneous, the defendant will have been prevented from selling receiving sets which, presumably, yield him a considerable profit. Therefore the injunction bond should be substantial, and I will fix it at $10,000.