Opinion
Floyd H. Crews, New York City, for the plaintiffs. Darbys&sDarby, New York City, Edmund H. Parry, Jr., Washington, D. C., and Morris Relson, New York City, were on the briefs.
T. Hayward Brown, Washington, D. C., with whom was Assistant Attorney General H. G. Morison, for the defendant. Gordon C. Biddle, Washington, D. C., was on the brief. The Court, upon the evidence, the report of Commissioner Hayner H. Gordon, and the briefs and arguments of counsel, makes the following
Special Findings of Fact
1. These are patent suits alleging infringement of two United States patents as follows: United States patent to F. B. MacLaren No. 1,959,540, issued May 22, 1934, and United States patent to Julius Gourgues Aceves No. 2,113,263, issued April 5, 1938. These patents will be hereinafter referred to as the MacLaren and Aceves patents. The title of the MacLaren patent is 'Method and Means for Increasing the Power Output of a Thermionic Tube,' and the specification thereof states that the invention relates to thermionic tubes and especially such tubes used to produce amplification of audio frequency currents. The title of the Aceves patent is 'Electrical Transmission System' and it is stated in the specification that the invention relates to transmission systems and particularly to systems employing electron discharge devices as repeaters or amplifiers.
Copies of the two patents in suit, plaintiffs' exhibits 1 and 2 respectively, and copies of the Patent Office file wrappers and contents thereof of the two applications which materialized into the two patents in suit, defendant's exhibits 24 and 25, are made a part of this finding by reference.
2. In suit No. 47608 the plaintiff, Remco Products, Inc., seeks compensation for the alleged unauthorized use and manufacture of inventions covered by the enumerated patents for a period beginning six years prior to the filing of the suit on February 17, 1947, to January 15, 1946, when plaintiff transferred the patents to Remco Electronic, Inc. Suit No. 47189 is for the period beginning January 15, 1946, when the plaintiff in that case, Remco Electronic, Inc., became owner of the MacLaren and Aceves patents, this period extending to the date of filing of this case, which was July 26, 1946.
Inasmuch as both of these suits relate to the same subject-matter and involve the same facts, the testimony was incorporated in one record and the two suits were consolidated by the court on September 22, 1947, upon motion.
Title
3. During the pendency of the MacLaren application in the Patent Office MacLaren assigned his entire right, title, and interest therein to Harrison Osborn, Thomas I. Sheridan, and Charles R. Coulter. This assignment was dated December 2, 1929, and recorded in the Patent Office at Liber Z 142, pages 665-667.
Also, during the pendency of the MacLaren application, Osborn, Sheridan, and Coulter assigned their entire right, title, and interest in the MacLaren patent to Revelation Patents Holding Company of New York, New York, a corporation of Delaware. This assignment was dated March 28, 1930, and is recorded in the Patent Office at Liber K 143, pages 466 and 467.
4. Prior to the filing of the application for the Aceves patent in suit, the entire right, title, and interest in the patent was assigned by Aceves to Revelation Patents Holding Company of New York, New York, a corporation of Delaware. That assignment was dated April 17, 1931, and recorded in the Patent Office at Liber I 148, pages 605-606. The document is also recorded in the United States Patent Office at Liber P 213, pages 529-530. A second assignment confirmatory of this earlier assignment was signed by Julius G. Aceves, dated November 19, 1947, and is recorded in the Patent Office at Liber P 213, at pages 531-532. A certificate of acknowledgment of the Notary Public who notarized the original Aceves assignment certifying that he did notarize that assignment on that date was signed by the notary, H. Krieger, on November 7, 1947, and is recorded in the Patent Office at Liber P 213, at page 533.
5. A certificate of the change of name of Revelation Patents Holding Company to Remco Products Inc., a corporation of Delaware, as of September 7, 1945, was recorded in the Patent Office at Liber U 204 on pages 274-280, inclusive.
On January 15, 1946, Remco Products, Inc., a Delaware corporation, assigned its entire right, title, and interest in both patents in suit to Remco Electronic, Inc., a New York corporation. This assignment was recorded in the Patent Office at Liber P 205, pages 39-42.
6. The plaintiff Remco Products, Inc., owned both patents in suit from the dates of their issuance to January 15, 1946.
The plaintiff Remco Electronic, Inc., owned both patents in suit from January 15, 1946, to the date of the trial.
Electronic Amplification
7. Amplification consists in producing a relatively strong voltage or power variation in the output circuit of an electronic tube which bears a continuous proportionate relationship to a relatively weak variation impressed upon the input circuit of the tube.
The term 'amplification' is defined in the American Standard Definitions of Electrical Terms, published by the American Institute of Electrical Engineers, August 12, 1941, as follows: 'An amplifier is a device for increasing the power associated with a phenomenon without appreciably altering its quality through control by the amplifier input of a larger amount of power supplied by the local source to the amplifier output.'
A conventional amplifier circuit is reproduced herewith. As shown, the electronic tube consists of an evacuated glass envelope, indicated by a circle, which encloses three elements. The bottom element, known as a cathode, is a filament connected to a source of heating current, not shown. A second electrode known as a plate or anode is shown at the top of the tube, and located in the space between the cathode and the plate is a third electrode consisting of a metallic mesh or screen known as a grid. Such a tube is customarily referred to as a triode.
In actual construction, a tube of this type is frequently made with a vertical filament with the grid consisting of a spiral of fine wire with widely spaced turns surrounding the filament. The plate is made of a concentric tube, which in turn surrounds the grid.
(Image Omitted)
Conventional amplifier circuit.
The input, which may consist of a microphone or phonograph pickup or photoelectric cell and which produces a relatively weak voltage variation, is connected to the grid and cathode of the tube by means of a transformer. The output circuit of the tube is connected between the plate and the cathode and includes a source of power such as a battery B usually having a high voltage potential of one hundred to several hundred volts. This output circuit is connected through an output transformer to the output device which may consist of a loud speaker.
8. When the filament or cathode of the tube is heated, electrons are emitted therefrom and are attracted by the plate, flowing through the open spaces in the grid. This electron stream provides a path for a flow of energy in the plate circuit or output circuit of the tube from the high voltage battery B. If the grid, however, is maintained a few volts negative with respect to the cathode by means of a small battery C (ordinarily 2 to 4 volts), it tends to repel the emitted electrons, and consequently the flow of plate current will be decreased. If it is made sufficiently negative, or beyond cutoff the plate current will be reduced to zero; that is, the flow of electrons through the grid to the plate will be shut off completely. The grid therefore acts like a valve in controlling the flow of electrons from the cathode to the plate, thus controlling the flow of current in the output circuit.
The relatively weak variations in voltage that are applied to the grid-cathode circuit from the input device or microphone cause the grid in normal operation to become more or less negative in voltage with respect to the cathode, and the grid thus acting as a valve will produce a faithful reproduction of such variations in the relatively strong plate current flow from the battery B to the associated output circuit of the tube, thus producing an enlarged facsimile of the input signal. In conventional operation of such an amplifying system the voltage of the battery C must be so selected and the input circuit must be of such impedance that the voltage variations impressed upon the grid circuit by the input device will never become sufficiently negative to entirely cut off the flow of electrons. If so, accurate reproduction of the input variations would then no longer take place in the output circuit and distortion would occur.
Ordinarily, no current flows in the grid circuit, the grid merely having impressed thereon a fluctuating negative voltage. If voltage variations are so large in character that the grid at times becomes positive with respect to the cathode, the grid itself will attract electrons and a current flow will take place in the grid circuit each time the grid becomes positive. Should such current flow take place through a relatively high impedance in the input circuit, it will cause a potential drop in the input circuit superimposed upon the signal voltage. When this occurs, accurate reproductions of the input voltage variations will not take place on the grid, and distortion will again occur.
To obtain distortionless amplification, the grid must operate within narrow limits, not becoming too negative and not becoming positive. The degree of distortionless amplification of a single tube is thus limited, and while amplifying tubes may be connected in cascade, the output of the first tube feeding the input of a second tube, etc., distortion may still occur because of the added complexity of the circuits and the difficulties of connecting the output of one tube to the input circuit of the next to transmit equally the entire range of frequencies occurring in music or speech.
Both of the patents in suit are addressed to the problem of permitting the grid to operate both in the negative and positive regions and still give undistorted amplification with a consequent increase in the power output of the tube.
The Patents in Suit
9. MacLaren patent.--MacLaren initiates the detailed description of his invention in his specification by referring in some detail to the facts which have been set forth in the two preceding findings. He states that the secondary of the input transformer is necessarily composed of a large number of turns of fine wire and forms a high resistance in the input or the grid-cathode circuit, and that whenever current flows in the grid-cathode circuit due to the grid operating in the positive region, there will be a potential drop across the secondary of the transformer. This will prevent the grid potential from varying in accordance with the incoming signal, resulting in a consequent distortion in the output circuit of the tube. He also refers to the use of a grid bias so that the grid will always operate in the negative region as a means of eliminating distortion but with a consequent limiting of the power output of the amplifying tube.
10. The MacLaren patent in suit then lists in the specification three objects. The first object stated is to--'provide a means for increasing the range of power output of a thermionic tube.'
The second object stated is to--'provide a method and a means for compensating for the ordinary distortion produced when the grid of an amplifying tube is positive so that the grid may swing either negative or positive and still produce true amplification at the output of the tube.'
The third object stated is to--'provide an input circuit for an audio frequency thermionic tube which will not have a potential drop therein to cause distortion.'
11. In order to compensate for the distortion that occurs when the grid of an amplifying tube operates in the positive region, MacLaren teaches the use of a circuit in which there is inserted what he calls a 'compensating tube' between the high impedance secondary of the input transformer and the input circuit of an amplifying tube.
'Compensation: * * * That which constitutes, or is regarded as, an equivalent or recompense; that which makes good the lack of variation of something else * * *.' [Webster's New International Dictionary, 1934.]
Fig. 1 of the MacLaren patent, which is reproduced herewith, shows the compensating tube 22, the amplifying tube 17, and the associated electrical circuits connected therewith. The input transformer has one end connected to the grid 23 of the compensating tube, and its other end connected through the battery 18 to the plate 25 of the compensating tube. Also connected as a second return path from the lower end of the secondary of the input transformer is a circuit extending to the cathode 20 of the amplifying tube, thence through the interelectrode space to the grid 21 of the amplifying tube and back to the cathode 24 of the compensating tube.
(Image Omitted)
Fig. 1 of MacLaren patent.
The plate circuit or direct current circuit for the compensating tube is from the battery 18 through plate 25; thence through the electron stream of this tube to the cathode 24 and through a direct metallic connection therefrom to the grid 21 of the amplifying tube; across the inter-electrode space from this grid to the cathode 20 of the amplifying tube and back to the battery. The output circuit of the amplifying tube is conventional in character and, as shown, consists of an output transformer connected to the plate through the battery 18 to the cathode 20 of the amplifying tube.
12. When the cathodes of the tubes are heated by conventional means, not shown, and the circuit is thus rendered operative, but is in a quiescent stage, the connections of battery 18 are such that electrical energy will flow across the electron stream from
the plate to the cathode of the compensating tube and thence through the direct metallic connection 24' to the grid 21 of the amplifying tube, which is positive with respect to the cathode 20 of the amplifying tube (and thus attracts electrons) and back to the battery. In order for this energy flow to take place it is essential that the grid 21 of the amplifying tube be at all times positive with respect to the cathode 20 of this same tube. Otherwise, there would be no electron flow in the first tube. This is contrary to MacLaren's theory of operation which refers to the grid 21 at times being negative. In this respect MacLaren's theory is erroneous.
13. When a signal voltage variation is applied to the grid of the compensating tube 22, which is maintained in the negative region by virtue of battery 18, this will cause variation in the electron stream existing between the cathode and plate of this tube. This will in turn cause a variation in the energy flow through the plate-cathode circuit of this tube, and the direct connection of the cathode of the compensating tube to the grid of the amplifying tube will cause this grid to become more or less positive with respect to the cathode of the amplifying tube. This, in turn, controls the electron stream between the cathode and plate of the amplifying tube, thus controlling its output. If the particular signal that is coming through the system causes the grid of the amplifying tube to become more positive with respect to the cathode of the same tube (a situation which in a conventional amplification circuit would cause distortion), the increased flow of electrons from the cathode of the amplifying tube to its grid will in turn produce a greater electron flow between the cathode and plate of the compensating tube because of the direct metallic connection 24'.
In other words, voltage variations between the grid and cathode of the compensating tube are controlled not alone by the variations of the input signal coming from the transformer high impedance output winding but also by the load resistance which in this case is the grid-cathode space 21-20 of the amplifying tube. This will in turn represent an increase in a positive sense (the grid becoming less negative) in the voltage between the grid and cathode of the compensating tube. In other words, the output circuit from the cathode of the compensating tube when it needs more current has its hand on the throttle, so to speak, and can cause the additional current, which it needs, to flow so as to maintain the voltage existent between the grid 21 and 20 of the amplifying tube at a proper value to overcome distortion. Stating it crudely, the amplifying tube, if it had a brain, would think that the input signal being fed to it was coming from a low impedance source instead of from a high impedance source. This accomplishes a reduction in distortion.
14. One of the requisites of the circuit disclosed by MacLaren is the direct metallic connection 24' which provides for direct current flow across the cathode-plate space of the compensating tube to be connected in series with the grid-cathode space of the amplifying tube so that the electron streams in these two inter-electrode spaces are connected in series. This connection 24' is specified in claim 14 of this patent (the claim in suit) which is quoted below, as 'a connection of substantially zero impedance.' Unless this connection is a direct metallic one MacLaren's system would fail to operate, as no current flow could take place in the plate-cathode space of the compensating tube.
Since the issuance of the patents in suit, a triode tube and associated circuits in which the input is connected to plate and grid, and the output is taken from the cathode thereof, and in which the cathode follows the grid potential up and down, have become known in the art as cathode followers. In such systems the variation in output voltage at the cathode is in phase, or in step, with the voltage variations impressed upon the grid. In the conventional amplifier where the output is taken from the plate circuit, when the grid becomes more positive the plate becomes more negative, and the output circuit may therefore be considered as being 180° out of phase with the input circuit.
15. Claim 14 of the MacLaren patent is in issue. This claim reads as follows: '14. A system of the character described comprising an amplifying triode, a compensating triode, a source of anode potential for both triodes, a connection of substantially zero impedance from the anode of the compensating triode to a positive terminal of said potential source, a connection of substantially zero impedance from the cathode of the compensating triode to the grid of the amplifying triode, means for impressing the signals to be amplified on the grid of the compensating triode, and circuit connections for causing the signal variations on the grid of said compensating triode to be in substantially the same phase as the signal variations on the grid of the amplifying triode.'
The Aceves Patent
16. The Aceves patent deals with certain relationship between resistance and inductance. These characteristics of electrical circuit may be defined as follows:
(a) Resistance is that quality of an electrical circuit by virtue of which it opposes the passage of an electrical current there through. It is analogous to the frictional resistance tending to oppose the flow of water through a pipe. (b) Inductance is that property of an electrical circuit by virtue of which it tends to resist a change in the rate of current flow. It is due to electromagnetic effects in the circuit and may be approximately defined as 'electrical inertia.' Inductance is only manifest in an electrical circuit where the current flow therein is an alternating or a varying current, and the inductive effect increases in a circuit having inductance, with the frequency of such alternation or rate of variation in current flow.
17. Among the objects of the invention stated in the Aceves application is the provision of a repeater system which is capable of transmitting current variations of larger amplitude without appreciable distortion; to provide a repeater employing a plurality of electron discharge devices in cascade wherein one of said devices is capable of being operated with a positive grid bias and is provided with means for limiting the normal plate current though said device; to provide an amplifier system employing a plurality of cascaded electron discharge devices wherein the grid of one device is connected directly to the cathode of the preceding device, together with means for maintaining said grid and cathode at substantially ground potential without substantially affecting the amplifying powers of the system; to provide an amplifying stage employing an electron discharge device, the input circuit of said device being provided with another electron discharge device which compensates for the grid loss of the first mentioned amplifier device, together with means for limiting the normal plate current of the said device to a minimum value.
18. More specifically, the Aceves patent is directed to an improvement of the MacLaren circuit. Aceves describes his improvement by disclosing in Fig. 1 of his drawings the MacLaren circuit, and discussing its disadvantages in his specification and then showing in Fig. 2 the changes that he makes therein. These two figures of the Aceves patent are reproduced, 93 F.Supp. 246-247.
FIG. 1
(Image Omitted)
FIG. 2
(Image Omitted)
Figs. 1 and 2 of Aceves patent.
As previously set forth in Finding 14, with reference to the MacLaren patent, one of the requisites of MacLaren's circuit is the direct metallic connection from the cathode of the first tube to the grid of the second tube through which circuit flows the direct current which produces the electron stream in the first tube. With this particular circuit arrangement, the grid 5 of the second tube shown in Fig. 1 (MacLaren's circuit) of this patent at all times has to operate in the positive region with respect to its associated cathode 6. Should the grid 5 become negative with respect to its cathode the circuit would cease to operate because there would then no longer be a current flow through the circuit to establish the electron stream in the first tube.
With the grid 5 of the second tube positive at all times there is therefore a relatively high normal plate current flowing at all times between the plate 11 and the cathode 6 of the second tube, even during the period when the circuit is quiescent and not transmitting signals. This high normal plate current performs no useful work and tends merely to heat the elements of the tube. As there is a practical limit to the power output of any tube, a high normal plate current when the tube is in a quiescent state also will limit the useful power in the signal variations that are superimposed on the plate current that feeds the output circuit.
19. After discussing in detail in his specification these drawbacks pertaining to the MacLaren circuit, which have been set forth in the previous finding, Aceves teaches a remedy therefor which is exemplified in Fig. 2 of his patent by the use of a reactance coil indicated by the reference numeral 16. The specification emphasizes the low ohmic resistance of this coil, stating that it is 'of extremely low ohmic resistance' and elsewhere in the specification that the ohmic resistance of coil 16 is 'vanishingly small.' The specification further indicates that this device 16 preferably possesses large inductance, the specification stating that--'Preferably the inductance of coil 16 is so large that for the lowest impressed frequency its reactive impedance is greater than the plate-cathode impedance of the device 1. Under these conditions, therefore the reactive voltage developed by the passage of the signal variations through the coil 16 will be applied directly between the grid 5 and the cathode 6, and will produce corresponding changes in the output circuit of the device 2.'
With the system in a quiescent state and the coil 16 possessing negligible ohmic resistance, the potential drop across the coil 16 will be negligible so far as a steady direct current flow is concerned, and the grid 5 of the second tube and its cathode 6 will remain only slightly positive or substantially at ground or zero potential. The direct current flow from the cathode of the first tube (MacLaren's compensating tube) which maintains the electron stream in that tube, no longer has to pass through the grid cathode space of the second tube but can return to the battery through coil 16. Coil 16 thus enables the grid 5 of the amplifying tube to remain at substantially zero potential instead of having to operate in the positive region, as in the MacLaren circuit, and therefore reduces or limits the normal anode current of the second tube when no signals are being transmitted.
20. When, however, voltage variations due to a signal input are impressed upon the grid 9 of the first tube and are reproduced on the cathode thereof, these signal variations, which comprise a plurality of various frequencies, set up voltage variations across the terminals of coil 16 because its inductance tends to resist current fluctuations. The voltage variations are impressed upon the grid 5 of the second tube by the direct circuit connecting the cathode 3 of the first tube with this grid. In other words, the coil 16 and its associated circuit to ground function as a filter to permit the steady direct current to return via the ground without passing through the grid-cathode space of the second tube but cause the signal voltage variations to be impressed upon the grid 5 of the second tube. With the grid 5 of the amplifier tube at approximately normal ground or zero potential, signal voltage variations impressed thereon may therefore swing both in the positive and negative regions without interrupting the electron stream in the first tube.
21. The Aceves specification states--'Thus a suitable source of steady potential may be inserted between the coil 16 and the ground, although satisfactory operation is secured without such external potential, and the circuit of Fig. 2 may be arranged to provide any frequency-gain characteristic required for the system as a whole.' To follow these instructions would defeat the main purpose of the patent. If such source of steady potential or bias voltage were so connected as to make the grid 5 of the amplifying tube negative with respect to its cathode, in an attempt to further reduce the normal current flow through the amplifying tube, this would result in making the grid of the first tube, or compensating tube, positive with respect to its cathode and therefore introduce distortion at this point. In order to prevent this it would be necessary to disconnect the grid circuit of the first tube from its ground connection and instead to connect it to the negative pole of such a battery. Such a rearrangement of circuits is not taught nor suggested by Aceves.
If, on the other hand, the source of potential were so connected as to make the grid 5 of the amplifying tube more positive with respect to its cathode, the normal current flow through the amplifying tube would be increased and no longer limited to a minimum value. This is again contrary to the results Aceves attempts to obtain.
The patent further states
'For example, the coil 16 may be provided with any other form of reactance or resistance in shunt and/or in series therewith in order to obtain different voltage drops for the various frequencies, or if desired, these additional reactances may be used for the purpose of controlling the phase angles between the various impressed frequencies. 'For example, a resistance connected in parallel with the coil 16 provides a simple and efficient method of obtaining uniform gain at all audio frequencies. It will be understood, of course, that the circuit of Fig. 2 may be used for the transmission of any range of frequencies, for example, if the system is designed to transmit audio frequencies then the coil 16 will be designed to have an extremely low ohmic resistance, but a high reactance to the audio frequencies. Similarly, if the system is to be used for the transmission of radio frequencies, the coil 16 will likewise have very low resistance, but will have high reactance to the radio frequencies.'
As the inductive effect of the coil 16 will be greater with the higher frequencies involved in the signal transmitted, a resistance in parallel with this coil will tend to even up the effect of the coil 16 upon the multiple frequencies present in the signal. A resistance in series with or in lieu of the coil 16 will provide a voltage drop between the cathode and grid of the amplifying tube and will result in making the grid positive with respect to the cathode. Any ohmic resistance introduced between the point 17 and the ground, unless the same is in parallel with the coil 16 and its negligible ohmic resistance, will tend to make the grid 5 of the amplifying tube normally positive with respect to its cathode and will cause an increased flow of the normal plate current in the amplifying tube.
Other than the addition of the inductance coil 16 of negligible resistance, which provides the means for limiting the normal anode current of the amplifying tube, and for permitting the grid of this tube to operate both in the negative and positive region, the Aceves circuit and its operation are the same as the MacLaren circuit.
22. Claims 6, 22, 23, and 27 of the Aceves patent are in suit. These claims read as follows: '6. In an amplifying system the combination of a first electron discharge device, a second electron discharge device, the cathode of said first device being directly connected to the grid of said second device, and the cathode of said second device being connected in a direct current circuit to the grid of said first device, means for limiting the normal anode currents of said second device, and circuit arrangements for impressing signal variations on the grid of the first device so that variations in potential between the grid of the first device and the cathode of the second device are in phase with the variations in potential between the grid and cathode of the second device.'
'22. In a transmission system of the type employing two sets of electron discharge elements each set comprising a cathode, anode and control electrode, a direct connection between the cathode of the first set and the control electrode of the second, means for limiting the normal anode current of the second set when no signals are being transmitted, said means consisting of a direct current path between the grid and cathode of the second set, an anode potential supply, input means for impressing signal variations on the control electrode of the first set, said input means making a direct current path between the last mentioned control electrode and a negative pole of the said anode potential supply.
'23. A combination according to claim 22 in which said means for limiting the normal anode current comprises a resistance.'
'27. In a transmission system of the type employing two sets of electron discharge elements each set comprising a cathode, anode, and control electrode, a direct connection between the cathode of the first set and the control electrode of the second, means for limiting the normal anode current of the second set when no signals are being transmitted, said means consisting of a direct current path between the grid and cathode of the second set, an input circuit arrangement for impressing signal variations between the control electrode of the first set and the cathode of the second set, the variations in potential being repeated across the control electrode and cathode of the second set approximately in phase with the said impressed signal.'
23. At the time of the filing of the amendment (February 6, 1934) in the Aceves application inserting claims 22 and 23, the word 'comprise' had the dual meaning of include or consisting of. While the Aceves patent indicates that a resistance may be used in shunt or in series with the inductance coil 16 (see Finding 21), it does not teach that the coil 16 may be replaced by one having only ohmic resistance. To substitute ohmic resistance in lieu of the inductance coil 16 would do away with any filter effect of this coil as set forth in Finding 20 and would make the grid of the amplifying tube positive with respect to its cathode and would be contrary to the results Aceves desired.
'Comprise--* * * 2. To comprehend or include; to contain or cover compendiously, or as constituent parts; sum up; cover. 3. a. To consist or be made up of; * * *.' [Webster's International Dictionary, 1934.]
When the phraseology of claim 23 is interpreted in the light of the disclosure contained in the specification, the word 'comprises' used therein has the meaning 'including'.
24. The MacLaren circuit was originated in an attempt to produce audio-amplifying equipment suitable for use in motion picture theaters showing talking movies and in an effort to reduce the cost of the amplifying equipment available at that time. Plaintiffs and their predecessors invented and patented various types of electronic tubes in which the electrodes of the two tubes used in the MacLaren circuit were incorporated in one glass envelope, making, in effect, a single tube.
Plaintiffs and their predecessors have issued numerous licenses to various companies in the radio industry. These patents have been recognized by the industry and are not so-called 'paper patents.' However, these licenses include the use of the various tube patents as well as the MacLaren and Aceves patents here in suit. None of the license contracts introduced in evidence granted a license under less than seven patents. It does not appear that any license has been granted under the MacLaren and Aceves patents alone.
25. No evidence has been presented to show either conception or reduction to practice of the MacLaren and Aceves inventions prior to the filing dates of the applications which matured into these two patents in suit.
The Alleged Infringing Structure
26. Within a period of six years prior to the filing of the petitions in the present cases, radio sets known as SCR-584-A and SCR-584-B have been used or manufactured for or by the Government without the license or consent of the plaintiffs. These sets are described in War Department technical manual TM 11-1524 which is in evidence as plaintiffs' exhibit 9. Both of these radio sets are radar equipment, and so far as the present issue is concerned, are similar in operation and function.
27. In the radar systems herein involved a powerful high frequency radio transmitter is periodically energized to send out a short burst of radio frequency energy into space. This is transmitted on a directional antenna provided with a reflector, the radio frequency beam thus being pencil-like in character and somewhat similar to the beam of light from a conventional searchlight. Like a searchlight, the radar antenna and its associated reflector are movable on a vertical pivot and horizontal trunnions so that the radio beam may be pointed in any desired direction. The antenna and its reflector may be operated either manually or automatically by electric motor mechanism for scanning any given portion of the sky.
If during the scanning operation the radar beam encounters some object such as an airplane, a portion of the radio energy is reflected back to the radar unit as an echo and is picked up by the reflector and its associated antenna. The echo signal is fed into a radio receiver and amplifier, the output of which is connected to a cathode ray tube with a screen somewhat similar in size and character to a television screen. Each time the periodic energy pulse is sent out by the transmitter a spot of light appears at a fixed point on the screen. This may be termed the zero spot. When the echo beam is received from the target and passes through the receiving and amplifying system, a second spot of light appears on the screen. In the intervening time, the cathode ray beam which produces the indicating spots on the screen of the cathode ray tube, has moved across the screen at a constant time rate. As the velocity of travel of the radar beam and the returning echo is also at a fixed rate, the distance between the fixed zero spot and the echo spot on the receiving screen of the radar set gives an exact measurement of the distance to the target. For this purpose the screen has a scale calibrated in yards. The direction in which the antenna and its reflector are pointed at the time echoes from a target are received gives the location of the target.
28. The sequence of events that take place in a radar system involves extremely short and exact time intervals. The velocity of the radar beam to and from a target is 186,000 miles per second. This is approximately 10 to 11 microseconds per mile for the transmission of a radio signal out to the target and back again. The transmitting pulses sent out by the transmitter in the defendant's structures in issue have a duration of eight-tenths of a microsecond. These pulses are repeated at a periodic rate of 1,707 per second. After such pulse has been transmitted the transmitter is quiescent and the receiving system awaits the reception of an echo for a period of 586 microseconds.
One-millionth of a second.
All these various time intervals are precisely controlled by what a layman would call a master clock. In the sets here in issue this consists of a crystal-controlled oscillator circuit which produces a master timing frequency of 81.95 kilocycles (81,950 cycles per second). This master frequency is further modified into various other multiple time intervals by electronic circuits, which are termed 'multivibrators.' A mechanical analogy is found in a watch in which the hour, minute and second hands are driven in multiple relationship by a single balance wheel.
A quartz crystal incorporated in an electronic oscillating circuit has the property of accurately controlling the oscillations of that circuit. The rate of oscillation is dependent upon the thickness to which the crystal is ground and certain other characteristics of the crystal.
The electronic circuits associated with the control of the various time intervals not only function to produce the pulses at properly spaced time intervals but, in addition, determine the duration of the various pulses and also their contour or wave shape.
29. The present issue relates to that portion of the electronic timing circuits that receives two different pulses from the multivibrator circuits and combines them (both as to contour and time) to form the pulse which periodically operates the transmitter and certain other components of the system every 586 microseconds.
There has been introduced in evidence as plaintiffs' exhibit 10 a circuit diagram illustrating the circuits here involved, and the circuit diagram illustrated herewith and entitled 'Defendant's alleged infringing circuits' has been taken from this exhibit.
(Image Omitted)
Defendant's alleged infringing circuits. As shown in this diagram, two electronic tubes are used in this portion of the circuit. The left-hand tube, identified by the legend 'V613B,' is a conventional triode having a plate, grid and cathode. This tube is so connected as to receive its input signal between the grid and plate and to deliver a positive output signal from its cathode, thus functioning as a cathode follower, the voltage produced on the cathode following the voltage of the grid up and down and in phase therewith. The plate of this tube is connected to a D. C. source of 250 volts.
The right-hand tube, identified by the reference character V-614, is a tube having five elements, viz., a plate, a cathode, and three grids. In the present instance we are only concerned with the two lower grids designated 4 and 6. The plate circuit of this tube is supplied with plate current through a resistance from a voltage source of 250 volts. The cathode 3 of the first tube or cathode follower is connected to one terminal of a blocking or isolating condenser C 659 of .01 microfarads, the other terminal of which is connected to the second grid 6 of the right-hand tube. This condenser presents an infinite impedance to direct current flow and thus permits a negative bias voltage of 20 volts to be applied to this grid through the resistor R689.
The application of this negative voltage to grid 6 of the tube V-614 enables this grid to be biased beyond the point of cutoff so that normally no current will flow in the plate circuit. Similarly, the lower grid 4, which is also isolated from its input circuit by the condenser C 657 has a normal negative bias of 9 volts applied to it through the resistor R 690. This negative voltage is sufficient to bias this grid beyond the point of cutoff so that no current will flow through the plate circuit of the tube. That is to say, if the grids 4 and 6 are contemplated as mechanical gates within the tube, the negative bias voltages applied to and normally maintained on them are sufficient to keep each tightly closed and not permit any current flow in the plate circuit of the tube.
30. Every 586 microseconds a pulse of approximately square wave shape and 6 microseconds in duration, is fed into the grid of the first or cathode follower tube from the multivibrator system. This produces a strong positive pulse of similar shape on the cathode of this tube. At this repetition frequency, which is 1.707 kilocycles, the isolating condenser presents an impedance of 9,350 ohms and therefore permits the recurring square wave pulses to pass there-through and to the second grid 6 of the second tube. On each pulse, grid 6 swings from the 20-volt negative bias to approximately 40 to 100 volts positive.
In terms of the previous mechanical analogy of a gate, grid 6 is suddenly swung from a tightly closed position to a wide-open position every 586 microseconds for a period of 6 microseconds. The approximate qualitative shape of these trigger gate pulses is shown in the top line of the figure herewith reproduced from the Service Manual, plaintiffs' exhibit 9, which figure is entitled 'Trigger gate and selector wave-forms', as shown on page 21.
(Image Omitted)
Trigger gate and selector waveforms.
The lower grid, designated 4, of the right-hand tube is continuously receiving an input from the multivibrator system which has a contour approximately that of a sine wave. These are produced in the multivibrator system at a rate of 81.95 kilocycles which spaces individual pulses 12.2 microseconds apart. This input is shown in the second line of the above figure, the individual pulses being designated thereon as trigger pips. The voltage value of the narrow curved portion of the peaks is of sufficient value to overcome to 9 volts negative bias of grid 4 of the second tube. This action is shown in the second line of the above diagram.
As the bias on each of the grids 4 and 6 is sufficient to independently hold the tube in a nonconductive state, no current will flow in the plate circuit until both biases are simultaneously overcome. Again referring to the mechanical analogy, either gate 4 or 6 can be opened but as long as the other gate remains closed no plate current can flow through the tube. However, every forty-eighth trigger pip is fed to grid 4 at the same time that a positive trigger gate pulse occurs on grid 6. When this occurs and in the brief interval of time (approximately 1 to 2 microseconds) that the top curved portion of the sine wave forming the trigger pips rises above the cutoff value of 9 volts, the tube V-614 swings from a negative cutoff with zero current flow and heavily conducts, producing in its plate circuit a negative going pulse of approximately 1 to 2 microseconds having a steep or approximately vertical wave front. Such pulse is illustrated in the bottom line of the above-referred-to diagram of wave forms. This is the pulse that triggers or energizes the transmitter and other associated devices.
The right-hand tube of the circuit under discussion operates as a pulse-forming tube, the output pulse of which is formed by the coincidence of two different input pulses, both of different shapes and contours, neither of which is reproduced in the output circuit. In forming the output pulse, the tube swings from a nonconductive condition (biased beyond cutoff) to a full conductive condition and back again.
31. In the defendant's circuit the first tube functions as a cathode follower in presenting a low-impedance source for the 6-microsecond positive pulses fed to the second grid 6 of the second or pulse-forming tube. In this sense there is a compensating action present in the first tube of the defendant's circuits. With respect to such compensating action, the Service Manual (plaintiffs' exhibit 9) states on page 51 thereof as follows: '(12) Trigger gate cathode follower. V613B makes the positive selector gate available to the trigger selector stage from a low-impedance source and prevents loading of V612. The positive selector gate output is fed to the screen grid of trigger selector V614.' Tube V612 (not shown above) is the source of the 6-microsecond pulse wave which is fed to the grid of the cathode follower tube.
The cathode follower tube compensates to supply an ample 6-microsecond positive pulse to the pulse-forming tube, ample in the sense that when the pulse-forming tube is called upon to form the trigger pulse for the transmitter, its plate circuit will be made fully conductive.
The second tube of defendant's circuit is not an amplifying tube, and the problem of distortionless amplification is not present. Still referring to the analogy of a gate in connection with the operation of grid 6 of this tube, this gate is alternately kicked wide open and then slammed shut by the gate tender (the 6-miscrosecond timing pulse). In contradistinction, the gate (grid) of an amplifying tube is always open (operated above cutoff) and is opened or closed more or less by the hands of a maestro in exact accordance with a musical score (the input voltage to the grid) which he faithfully follows as he manipulates the gate, sometimes toward the open position, and other times toward the closed position, but never in a completely open or closed position.
The Prior Art
32. The following patents were available to those skilled in the art prior to September 21, 1929, the filing date of the MacLaren patent, the first filing date of the two patents in suit: British patent to Smart and Tucker No. 152,190, complete accepted October 14, 1920 (defendant's exhibit 1); French patent to Montchàtre No. 565,735, Délivré November 10, 1923, and translation thereof (defendant's exhibits 2 and 2a, respectively); British patent to Massolle, et al., No. 157,744, complete accepted March 30, 1922 (defendant's exhibit 3); United States patent to Nakken No. 1,928,760, issued October 3, 1933, on an application filed May 21, 1929 (defendant's exhibit 4); United States patent to Winther No. 1,700,393, issued January 29, 1929 (defendant's exhibit 5); United States patent to Case No. 1,625,409, issued April 19, 1927 (defendant's exhibit 6); United States patent to Langmuir No. 1,273,627, issued July 23, 1918 (defendant's exhibit 7).
The above enumerated patents are made a part of this finding by reference.
33. British patent to Massolle 157,744 (defendant's exhibit 3) relates to improvements in circuits for electric amplifiers of the type in which the grid-cathode path of a succeeding tube is connected in series with the anode-cathode path of a preceding tube and has for its object the provision of means for adjusting the average potential at the grid. The specification refers to a prior patent of Massolle's, No. 157,733 (defendant's exhibit 25-A) which discloses such a circuit.
Massolle in the present specification states with respect to his prior patent that such a type of circuit will demand a large normal current in the amplifier tube, which is generally undesirable.
This is the same problem which is present in the similar circuit of MacLaren, i. e., the grid-cathode path of MacLaren's amplifier tube being connected in series with the anode-cathode path of the preceding tube, which problem Aceves attempted to deal within the Aceves patent here in issue.
Massolle in the present specification suggests applying an auxiliary potential at the grid of the amplifying tube through the use of an auxiliary resistance so that the potential of the grid may be controlled with respect to its cathode. Four different circuits are shown in which the grid potential is thus controlled. In the circuits shown in Figs. 1, 2 and 3, the input is connected to the first tube in the conventional manner, i. e., the input being fed to the grid and cathode of the first tube, and these circuits are not material to the present issue.
(Image Omitted)
Fig. 4 of Massolle patent. The circuit shown in Fig. 4 of Massolle is reproduced herewith. The input device indicated at Z is broadly termed 'a controlling device Z' which Massolle states may consist of a photoelectric cell. The output device indicated in this figure by R is a telephone receiver or other indicator.
Massolle's complete description of Fig. 4 is as follows:
'Another auxiliary arrangement, which makes it possible to feed all the tubes and also the controlling member Z from a single anode battery B, is shown in Figure 4, in which the path of the current is clearly indicated. With two tubes this arrangement works just as well, as with separate batteries for each anode circuit, provided that the useful resistance R (i. e., a telephone receiver or other indicator) be connected up, as shown in the drawing, so that it does not lie in the anode circuit of the tube I. If it were connected up at X, that is to say, in series with the tubes I and II, which are connected up in parallel to one another, then the tension between a sub1 and k sub2 would no longer be constant but would fall with the increase of current in the tube II. 'The amount of amplification in the positive region of the valve is equal to the ratio, of the grid current variations, to the simultaneous anode current variations, and this ratio may advantageously be determined by a suitable selection of tubes. Furthermore, the alterations in the distribution of more, the alterations in the distribution of potential over the anode cathode path and the grid cathode path are greatest when these paths have substantially equal resistances. It is, therefore, preferable to adapt the grid-cathode output of the succeeding tube to the anode-cathode output of the preceding tube. Under these conditions the succeeding tube is constructed for greater anode power than the preceding tube and the anode-cathode outputs of the tubes will then stand in a ratio corresponding to the intensification figure. 'Increased anode power is obtained by connecting the tubes I and II in parallel as illustrated in Figure 4 of the drawings.'
In phase therewith.
34. In the circuit shown in Fig. 4, the input circuit is not connected in the conventional manner but is connected to the plate and grid of the first tube. In this circuit as disclosed, the first tube is degenerative and will operate as a cathode follower and will tend to compensate whenever the grid of the second tube becomes positive and draws current. The two bias batteries by means of which the regulatory potential may be applied to the grids, are shown in this circuit at h sub1 and h sub2 . These batteries are connected to the respective grids by means of resistors w sub1 w sub2 .
35. An inter partes test was conducted by defendant's expert Stodola during the trial of this case, in which this circuit was tested. Voltage gain of the system shown in Fig. 4 of Massolle was tested both with the cathode follower circuit as disclosed therein and with the conventional type of grid connection.
Voltage gain was greater with the conventional type than with the cathode follower type connection. As a cathode follower gives a power gain and not a voltage gain this would be an expected result. The circuit of Fig. 4 operates as a cathode follower or compensating tube feeding into an amplifying tube, through a direct connection between the cathode of the compensating tube and the grid of the amplifier.
36. The United States patent to Nakken (defendant's exhibit 4) relates to amplification by means of electronic tubes. For convenience, the circuit diagram of Nakken is reproduced herewith.
(Image Omitted)
Nakken system.
The system disclosed in this patent contemplates the use of three tubes. The input to the first tube consists of a photoelectric cell which the Nakken specification indicates is primarily for the purpose of producing pulsating or alternating energy impulses and is not a photometer. The photoelectric cell is connected to the first tube of the Nakken amplifier system in the conventional manner, i. e., to the grid and cathode thereof. The Nakken disclosure shows the cathode of each preceding tube directly connected with the grid of the succeeding tube, so that with the exception of the first tube the grids of the other tubes will operate in the positive region of grid potential.
Photoelectric cells are and have been for a long time a conventional input device for talking motion pictures. They are actuated by a small beam of light projected through a sound track on a film as it passes an opening, thereby giving rise to a pulsating output of the photoelectric cell in accordance with the sound track, which output is fed into the initial tube of an electronic amplification system.
The last tube in the Nakken system operates as a conventional amplifier, having its output circuit connected to a loud speaker or equivalent device. Two high impedances, each consisting primarily of inductance and having relatively low ohmic resistance, provide a return path for the direct current component back to the battery and reject the fluctuating currents of the signal. These impedances, which are connected respectively between the grids and cathodes of the second and third tubes, function to limit the normal anode currents of the second and third tubes.
The polarities of the battery for the photoelectric cell are such that, under normal operating conditions, the grid of the first tube has a negative bias with respect to its cathode, which is the normal operating situation for a high impedance source such as a photoelectric cell. The output of the first tube is applied to the grid of the second tube. The grid of the second tube operates in the positive region and will, under certain circuit conditions, tend to introduce distortion in the system. The output of the second tube is taken between its cathode and plate and this is applied to the grid-cathode space of the third or amplifier tube. This second tube and its associate circuits form a cathode follower system in which the output is in phase with the input and while there will be no compensation for any distortion introduced into the system at the input of the second tube, the second tube will act to compensate for any distortion due to the operation of the grid of the third tube in the positive region. In other words, while the system will not compensate for distortion introduced into the system in front of the second tube, the second tube will prevent further distortion and compensate for positive current flow in the grid of the third tube.
37. In a comparative inter partes test of the MacLaren, the Aceves, and the Nakken circuits made by defendant's expert, and in which a record player was used as a source of input, there was a noticeable reduction in distortion and noticeable increase in volume between Aceves as compared to MacLaren. There was no appreciable change in either distortion or volume as between the Nakken circuit and the Aceves circuit. This test was made by listening to a loud speaker connected to the output and no comparative measurements were made of the degree of distortion.
Infringement and Validity
a. The MacLaren Patent
38. For convenience, claim 14 of the MacLaren patent in issue is paraphrased as follows, the pertinent elements being italicized:
A system of the character described comprising---- (1) an amplifying triode, (2) a compensating triode, (3) a source of anode potential for both triodes, (4) a connection of substantially zero impedance from the anode of the compensating triode to a positive terminal of said potential source. (5) a connection of substantially zero impedance from the cathode of the compensating triode to the grid of the amplifying triode, (6) means for impressing the signals to be amplified on the grid of the compensating triode, and (7) circuit connections for causing the signal variations on the grid of said compensating triode to be in substantially the same phase as the signal variations on the grid of the amplifying triode.
In the Government circuit under consideration (see Finding 29) the purpose of the second tube is not amplification. Instead of this tube being an amplifying triode, the proper designation therefor is 'a pulse-forming pentode.' This tube instead of having the function of amplification without distortion produces distortion in that it receives two different pulse-forming waves and combines them into a single output pulse, the shape of which is unlike either of the input signals. The defendant's system does not possess 'an amplifying triode.'
39. In the MacLaren circuit there exists a direct connection of substantially zero impedance from the cathode of the first tube to the grid of the second tube. This is for the purpose of providing a path for both the signal and the anode-cathode space current of the first tube without drop or distortion to the grid of the second tube.
In the Government structure the circuit from the cathode of the first tube to one of the grids of the second tube is not direct and has impedance. This circuit is interrupted by a blocking condenser which is essential to the operation of the Government structure as it enables a negative bias of 20 volts to be applied to the grid of the second tube in order to entirely cut off the operation of that tube between pulses. The presence of this blocking condenser and its associated resistor would tend to introduce distortion into an amplifying tube if its grid were driven by such circuit, and if such blocking condenser were inserted in the corresponding circuit of the MacLaren patent, the MacLaren system would be inoperative. The impedance of this condenser in the Government structure instead of being substantially zero is infinite to direct current flow and is 9,350 ohms at the pulse frequency employed in this circuit.
Claim 14 of the MacLaren patent is not infringed.
40. The British patent to Massolle (Findings 33-35) discloses in Fig. 4 an amplifying system involving two tubes in which the grid-cathode path of the amplifying tube is connected in series with the anode-cathode path of the preceding tube. The first tube is shown to be so connected as to operate as a cathode follower and will tend to compensate whenever the grid of the second tube becomes positive and draws current. The connection from the cathode of the first tube to the grid of the second or amplifying tube is disclosed as being of substantially zero impedance.
The terminology of claim 14 of the MacLaren patent is applicable to the circuit disclosed in Fig. 4 of Massolle. This statement also applies to the second and third tubes and their associated circuits disclosed in the Nakken patent (Findings 36, 37). The use of these circuits would infringe claim 14 of the MacLaren patent. Their disclosure, being prior, invalidates claim 14 of MacLaren.
b. The Aceves Patent
41. The circuit of the MacLaren patent in suit necessitates that the entire direct current from the anode-cathode space of the first tube has to flow through the grid cathode space of the second tube so that the grid of the second tube will always operate well over into the positive region, which in turn demands a relatively heavy flow of anode current in the second tube even under quiescent conditions.
The Aceves patent in suit was directed to cure this condition by inserting an inductance coil of vanishing small resistance between the circuit extending from the cathode of the first tube to the grid of the second tube and the return circuit to the battery. Such inductance coil formed a return path for the direct current flow but because of its inductance rejected the fluctuating signal input which was fed to the grid of the second tube. This inductance coil because of its 'vanishingly small resistance' enabled the grid of the second tube to operate near zero potential instead of considerably in the positive region, and thereby formed a means for limiting the normal anode currents of the second tube. Such inductance coil is the 'means' referred to in claim 6 in suit of the Aceves patent which is herewith paraphrased with pertinent portions italicized:
6. In an amplifying system the combination of
(1) a first electron discharge device, (2) a second electron discharge device, the cathode of said first device being directly connected to the grid of said second device, and the cathode of said second device being connected in a direct current circuit to the grid of said first device, and (3) means for limiting the normal anode currents of said second device, and (4) circuit arrangements for impressing signal variations on the grid of the first device so that variations in potential between the grid of the first device and the cathode of the second device are in phase with the variations in potential between the grid and cathode of the second device.
As stated in Finding 39, with reference to the MacLaren patent, the Government structure does not have any direct connection between the cathode of the first tube and the grid of the second tube, this circuit being interrupted by a blocking condenser.
The means for limiting the normal anode current in the second tube of the Government structure differs materially from the Aceves disclosure, is not the equivalent thereof, and is not taught thereby. In the Government structure this is accomplished through the use of the blocking condenser and an associated circuit which enables a negative bias voltage of 20 volts to be applied to the grid of the second tube. The application of this amount of negative voltage to the grid of the tube biases the same beyond cutoff and limits the normal anode current of the second tube to zero. The 'means' set forth in item 3 of the above paraphrased claim is limited to circuits taught or disclosed by the Aceves patent and does not comprehend any and all circuits which may be designed for this purpose. Aceves does not teach the use of a blocking condenser.
Claim 6 of the Aceves patent is not infringed.
42. Claim 22 of the Aceves patent also includes as an element thereof the direct connection between the cathode of the first set and the control electrode of the second. In addition, this claim sets forth in slightly more detail the means for limiting the normal anode current of the second tube, the claim stating with respect to this means--'said means consisting of a direct current path between the grid and cathode of the second set * * *.'
In the Government structure the only direct current path existing between the grid and cathode of the second set consists of a circuit which includes four resistors comprising a total of 98,200 ohms resistance. This is not the 'vanishingly small resistance' specified in the teachings of Aceves, nor does such a resistance contribute to limiting the normal anode current of the second tube. As previously stated, this is accomplished by the use of the blocking condenser and the consequent application of negative bias voltage to the grid.
Claim 22 is not infringed.
43. Claim 23 of the Aceves patent in suit differs from claim 22 only in the addition of the following phraseology: 'A combination according to claim 22 in which said means for limiting the normal anode current comprises a resistance.'
Findings 19-21 describe in detail the operation of the Aceves inductance coil to limit the normal flow of anode current in the second tube. Finding 21 refers to a statement in the Aceves specification indicating the fact that such inductance coil may have a resistance in shunt or in series therewith in order to obtain different voltage drops for various frequencies. In the Aceves circuit the inductance coil is an essential adjunct for the purposes set forth, i. e., to limit the anode current in the second tube. Resistance alone would not perform this function for (a) it would not separate the varying signal current which goes to the grid of the second tube from the direct current path back to the negative pole of the anode battery, and (b) it would tend to make the grid of the second tube of Aceves more positive instead of less positive. The word 'comprises' as used in the above-quoted phrase has the meaning of 'including.' (See Finding 23.)
Claim 23 as modified by this added clause is not infringed for the reasons stated in connection with claim 22. The means for limiting the normal anode current used by the Government is the blocking condenser and the biasing voltage applied to the grid of the second tube and not a resistance.
44. Claim 27 of Aceves differs only slightly from the previously considered claims. This claim also specifies as one element thereof--'a direct connection between the cathode of the first set and the control electrode of the second.' In addition, this claim specifies the means for limiting the normal anode current of the second tube as 'consisting of a direct current path between the grid and cathode of the second set.' This is the same phraseology that occurs in claim 22, and for the reasons set forth in connection therewith this claim is not infringed.
45. The terminology of all four claims of the Aceves patent in issue is applicable to the amplification system disclosed in the second and third tubes and associated circuit of the Nakken patent. Nakken suggests an inductance coil of low resistance and high impedance for the identical purpose employed by Aceves and connected in the same manner. These claims are invalid in view of the Nakken disclosure.
The British patent to Massolle discloses a circuit having a compensating tube of the cathode follower type driving an amplifier tube. In Massolle's circuit the grid potentials are controlled by bias batteries and resistors. If any of the claims of the Aceves patent in suit were interpreted with sufficient scope to be infringed by the Government structure, they would likewise be applicable to the circuit disclosed in the Massolle patent and would therefore be invalid.
46. Claim 14 of the MacLaren patent and claims 6, 22, 23, and 27 of the Aceves patent are not infringed and are invalid in view of the prior art. Before JONES, Chief Judge, and LITTLETON, WHITAKER, MADDEN and HOWELL, Judges.
JONES, Chief Judge.
Plaintiffs seek to recover for the alleged infringement of two patents, referred to herein as the MacLaren patent and the Aceves patent, both of which relate to electronic tubes and circuits. Both of these patents are directed to the avoidance of distortion in electronic amplification, and plaintiffs allege that both patents are infringed by the manufacture by or for and use by the United States of certain electronic circuits utilized in connection with radar systems.
The defendant contends that, in so far as the claims in issue here are involved, neither patent in suit is infringed; that neither patent discloses novelty over the prior art; that the alleged infringing circuits are essentially like those of the prior art and, hence, if any claims of the patents were construed as infringed they would be anticipated.
The essential facts established by the record in this case are fully set forth in the findings and except in certain controverted issues hereinafter discussed, it is unnecessary to refer to them in detail.
In order to present an intelligent discussion of this matter it is necessary to revert to some of the basic principles of amplification.
A conventional amplifier consists of a glass tube enclosing three elements. One of these is a heated filament termed a 'cathode.' The second is a small plate. When these two elements are connected by an external circuit which includes a source of electrical energy of substantial voltage, such as a battery of 100 to 200 volts, with the positive pole of the battery connected to the plate and the negative to the filament or cathode, a flow of electrons takes place from the same to the plate and provides a path inside the tube for current flow from the battery between the plate and cathode and through the external circuit.
The third element in the tube is the heart and soul of this contrivance and consists of a metallic screen, termed a grid, interposed between the cathode and plate. If this screen is connected to an external circuit wich makes it a few volts negative with respect to the cathode, it exercises a powerful influence upon the electron stream passing through it, and may, if sufficiently negative, even interrupt this stream entirely. Thus the grid becomes in effect the 'spigot' or 'gate' for controlling the amount of flow of the electron stream from the cathode to the plate of even cutting it off completely.
Starting with the 'spigot' closed, i. e., sufficient negative voltage on the grid to completely block the electron stream, it may next be contemplated that it is slowly turned toward an open position, i. e., by decreasing the negative voltage. As this occurs, the intensity of the electron stream will correspondingly increase, with a consequent increase in the flow of electrical energy in the external circuit connected to the plate and filament. If we next assume that the negative voltage has been reduced to zero, the 'spigot' is still not fully open. To open it completely and still further increase the electron stream we must pass through the zero point and apply an increasing positive voltage to the grid.
This brings us to an underlying and important fact in the present issue. Whenever the grid becomes positive it also attracts electrons which cause a conflicting flow of electricity back into whatever is supplying the control voltage to the grid.
For this reason it is essential in a conventional amplifier that the input voltage waves or variations, which find their origin in such devices as a microphone, phonograph pickup or photoelectric cell, operate within a relatively narrow negative range. As these voltages vary in accordance with the sound waves of the voice or music being reproduced, they must at no time impress upon the grid a sufficiently high negative voltage to cut off the operation of the tube nor a positive voltage which will cause the conflicting flow of electricity back into the input circuit. This is necessary in order to avoid distortion and to produce in the output circuit of an amplifier a magnified facsimile of the input. The 'spigot' or 'gate' to which we have referred by way of analogy, must neither be completely closed nor opened to wide.
It was to this problem that MacLaren's efforts were directed. He desired to increase the power output of an amplifier tube by giving the 'spigot' a greater range of opening, thus providing for a greater electron flow in the tube. MacLaren states his purpose so succinctly that we quote at this point the second object of his invention from his patent: 'provide a method and a means for compensating for the ordinary distortion produced when the grid of an amplifying tube is positive so that the grid may swing either negative or positive and still produce true amplification at the output of the tube.' MacLaren attempted a solution by employing a circuit between the signal source and the grid of his amplifier tube and which included what he calls a compensating tube. The intended purpose of this was to automatically compensate for the conflicting flow of electricity back into the input system whenever the grid of the amplifier tube became positive.
Without going into the details of this circuit and its operation, which are already set out in Findings 9-14, inclusive, we refer to one important feature. The cathode of the compensating tube is directly connected to the grid of the amplifying tube by means of a continuous metallic wire which MacLaren refers to as 'a circuit of substantially zero impedance,' which, in plain, every day terms, as well as in technical parlance, means that there is substantially nothing to impede current flow in this circuit. It is by means of this circuit that the electrical energy, which causes the electron flow in the compensating tube, is supplied from the grid of the amplifying tube.
It was here that MacLaren stumbled. While he developed a system in which compensation for distortion took place whenever the grid of the amplifier was operated in the positive region, his system would not function with the grid operating in both the negative and the positive regions which he indicated as the object of his invention. When the grid in the amplifying tube becomes negative it no longer receives electrons from the cathode, and there is then no source of electrical energy for energizing his compensating tube. In using the MacLaren circuit it is therefore always necessary to keep the grid of the amplifying tube at all times positive. Still referring to our analogy of the 'spigot,' it must at all times be rather fully opened whether the input circuit is quiescent or is receiving a series of sound-produced impulses to be amplified. This results in a relatively heavy continuous flow of electrons and current in the amplifying tube, which tends to overheat the tube.
Aceves then stepped into the picture. Seeing these defects in MacLaren's circuit, he attempted to remedy them. Findings 16 to 21 detail what changes Aceves suggested in the MacLaren circuit, and which became the subject-matter of his patent. Aceves' though was to connect to a common point of the entire system or to 'ground' MacLaren's circuit extending between the cathode of the compensating tube and the grid of the amplifying tube through a device known in electrical parlance as a 'filter.' This filter consists of a coil possessing large inductance but of extremely low ohmic resistance. The operation of this device has been covered in detail in the findings and it is sufficient to state here that such device will provide a ready path to ground for direct current flow while rejecting the varying voltage waves which find their origin in the sound waves at the input of the system. This device thus forces the varying electrical energy due to the sound waves to pass from the cathode of the compensating tube through the direct connection to the grid of the amplifying tube and at the same time permits the grid of the amplifying tube to assume an approximate zero voltage with respect to its cathode which is also grounded. Aceves by permitting the grid of his amplifier tube to operate at or near zero potential enables the varying voltage impulses coming into it to swing both in the positive and negative regions, thus correcting MacLaren's mistake and at the same time limiting the detrimental effects of a relatively heavy normal current flow in his amplifying tube.
Aceves makes a suggestion in his specification that a source of steady potential could be inserted between his inductance coil and the ground. To do this would vitiate the objects sought by Aceves. This matter is discussed in detail in Finding 21, to which plaintiffs have taken exception. There is ample foundation in the testimony for this finding.
Aceves also suggests that a resistance may be used in shunt or series with his inductance or filter coil. This does not teach nor suggest that an ohmic resistance may be used instead of the filter coil. To do so would do away with any filter effect and would make the grid of the amplifying tube positive, which would be directly contrary to the use of the vanishingly small resistance which Aceves suggested in order to correct MacLaren's mistake and to limit the normal flow of current in the amplifier tube.
Summarizing both the MacLaren and Aceves patents, they are both directed to avoidance of distortion in an amplifying system. They both employ a direct connection between the cathode of a compensating tube and the grid of an amplifying tube to control the electron flow in the latter. Aceves adds to this contribution the teaching that the grid of an amplifying tube may be operated at or near a normal zero potential and that normal current flow may be limited in the amplifying tube by means of a filter of extremely low ohmic resistance inserted between the direct connecting wire and the ground to which the cathode of the amplifying tube is also directly connected.
The defendant's structures alleged to infringe are radar sets. Their operation is described in Findings 26 to 31, inclusive. It is necessary only to state here that they operate on the principle of receiving the echo of a directed radio wave like a searchlight beam which is sent out into space by a powerful transmitter, and which when it hits an object such as an airplane, bounces back and is picked up by a receiver adjacent the transmitter. The duration of time between the radio wave sent out by the transmitter and its arrival at the receiver is measured by electronic devices, and this time interval gives the distance to the target.
The portion of the radar system alleged to infringe and with which we are concerned, is a pulse-forming device or electronic switch which turns on the transmitter for a minute interval of time approximating one-millionth of a second and then shuts it off while waiting 586 millionths of a second for an echo to come back. This operation is repeated periodically at the rate of 1,707 times per second.
This device consists of an electronic tube in which two grids are located between the cathode and the plate. (While the tube has three grids, we are here only concerned with the operation of two of them.) Each of these grids normally has a sufficient negative voltage applied thereto to entirely prevent any electron stream from flowing between the cathode and the plate, and thereby flow of electrical energy in the plate and output circuit of the tube is prevented.
Referring to our previous analogy of a 'spigot' or 'gate,' both 'spigots' are completely shut and as they are both located between the cathode and plate the opening of either one alone will not permit an electron stream to flow. Both gates or 'spigots' must be opened before the electrons can flow.
Both grids are supplied with periodic waves or pulses which cause them to change their voltage from negative to positive, thus periodically opening the 'spigots' or 'gates' from a completely closed to a wide open position. These pulses are different both in periodicity and character and find their origin in a device which for our present purposes is sufficient to call a master electronic clock. This is all discussed in detail in Findings 29 and 30, the pulse shapes being illustrated in Finding 30. It is solely necessary to emphasize here that only at certain intervals do the incoming pulses coincide and only then do the grids permit the flow of energy. When this happens the pulse occurring in the output of this tube is different in shape and character from either of the incoming pulses. It is this pulse that energizes the transmitter.
A comparison of the accused structure with the patents in suit is next in order. The upper grid in the defendant's pulseforming tube receives its controlling pulses from a circuit connected to the cathode of a preceding tube, which in turn is periodically operated by the master electronic clock. This is for the purpose of applying a strong burst of positive potential at the appropriate time. As far as feeding the grid of a succeeding tube from the cathode of a preceding tube is concerned, the Government structure and the patents in suit are similar. Here, however, similarity ceases.
Both patents in suit utilize a direct connection for this purpose. Instead of using a direct connection or a path of zero impedance the Government structure interrupts this circuit between the cathode of the first tube and the grid of the second tube by inserting therein what is known as a blocking condenser. This is a device which will permit the periodic pulses of relatively high frequency to flow from the cathode of the preceding tube to the grid of the second tube although at the frequency involved it presents an impedance of 9,350 ohms. It will also block the flow of any direct current in this circuit by offering infinite impedance thereto. It is by virtue of this blocking condenser that it is possible to isolate and to normally apply to this grid of the pulse-forming tube a normal negative voltage of 20 volts, thus keeping the 'spigot' tightly closed until a positive pulse comes through. The Government device would become inoperative were this circuit not interrupted by the blocking condenser.
We next take up the claims in suit. It is too well settled in patent cases to require citation, that the terminology of the claims of a patent must be interpreted in the light of the disclosure of the specification by what structures are therein described and how they function. Claim 14 of the MacLaren patent, the only claim of that patent in suit, specifies the circuit that we have just discussed as 'a connection of substantially zero impedance.' Such a circuit is not present in the Government structure.
Claim 14 of the MacLaren patent also includes an an element 'an amplifying triode.' We do not think that the pulseforming tube of the Government structure amplifies in the sense that it is used in either of the patents in suit or by those skilled in the art. It is not intended to produce in its output circuit an enlarged facsimile of the signal introduced into the input of the tube. Instead, it is a switch operated at the proper times by the combined pulses sent to its grids by the master clock to turn on and off the transmitter of the radar set.
As set out in Findings 38 and 39, claim 14 is not infringed.
Claims 6, 22, 23, and 27 of the Aceves patent are in suit. These claims all specify in slightly varying phraseology the 'direct connection' between the cathode of a first tube and the grid of a second tube, the circuit that we have discussed supra. As already explained, the defendant's device does not have this circuit and these claims are not infringed.
Other limitations occur in this group of Aceves' claim which point to noninfringement. These deal with the limitation of the normal flow of the plate-cathode current in the second tube of Aceves'. For this he basically depends upon his inductive filter of 'vanishingly small resistance' providing a direct current path from grid to cathode of his second tube.
In the Government device this normal current is not only limited but reduced to zero by the high negative voltage applied to the grid by virtue of the blocking condenser, an entirely different structure.
These distinctions are set out in Findings 41-44, inclusive, and except to state we think these findings are supported by the record, it is unnecessary to reiterate them here.
The two prior art patents pertinent to the question of validity are referred to in detail in Finding 33 to 37. These are the Massolle and the Nakken patents. The Massolle patent states that it 'relates to improvements in the arrangement of circuits for electric amplifiers,' and the patent to Nakken states that the object 'is to improve amplification of feeble energy manifestations through the medium of the thermionic triode amplifiers.'
Not only are Massolle and Nakken directed to a solution of the same basic problem that MacLaren and Aceves attempted to solve but they did it in the same way. In each of these prior art disclosures the circuits shown and described teach the feeding of a signal input to the grid of an amplifying tube from the cathode of a preceding tube by means of a direct connected metallic circuit. The Nakken patent, in addition, both describes and shows the grounding ot connection of this circuit and its return to the cathode of the amplifying tube by means of a coil having high impedance and relatively low ohmic resistance, a device which is similar to what we have previously referred to as the Aceves filter and by means of which he attempted to cure the difficulties present in the MacLaren patent.
Both of these prior disclosures show a photoelectric cell as an input device and telephones connected to the output of the amplifying tube.
Plaintiffs raise several curious contentions with respect to the disclosures of these patents. One of these is that the problems of MacLaren and Aceves do not arise in the amplification of the output of photoelectric cells because of their rather weak pickup characteristics. We see no merit to this. This, even if true, would only result in a weaker amplified signal which is a matter only of degree. The function of the circuits is the same. In addition, neither of the patents in suit is limited to the use of any particular type of imput device and the monopoly expressed by the claims in suit would not exclude the use of signals originating in a photoelectric cell.
Plaintiffs also urge that if the circuit of the Massolle patent here relied upon has any bearing on MacLaren and Aceves, it is only because of an accidental disclosure. The Massolle circuit is no more accidental than the disclosures of MacLaren and Aceves. These are all patents directed to the improved amplification of electrical signals. One skilled in the art at the time of the Massolle disclosure could clearly reproduce the Massolle circuit from the disclosure. While it is true that Massolle did not completely explain how his circuits operated, he did not have to do so. In fact, when MacLaren attempted to explain the operation of his circuit he made a mistake which Aceves later corrected. The prior art circuits and their operation with the direct connection between cathode and a subsequent grid are more analogous to the circuits of the patents in suit than is the circuit of the Government device.
As indicated in Findings 40, 45, and 46, claim 14 of MacLaren and claims 6, 22, 23, and 27 of Aceves are invalid in view of the prior art.
Plaintiffs further urge that a presumption of patentability should be based on the fact that numerous licenses have been issued in connection with the patents in suit and this points to commercial success. It is true that when invention is in doubt, acceptance of the same may be thrown in the scale, but this is only when one must weigh carefully the teachings of the prior art as compared with the claims in issue. In the present instance we think that the teachings of the prior art are clear and we are in agreement with In re Thayer, 143 F.2d 996, 1000, 31 C.C.P.A., Patents, 1224, in which it is aptly stated--'* * * However, commercial success is only one element to be considered in determining the patentability of claims, and when, as in the instant case, the prior art either discloses or plainly suggests the structure and the process defined by an applicant's claims, commercial success is not of great importance.' Moreover, none of these licenses were limited to the patents in suit and none granted a license under less than seven patents. Acceptance and usage of an article depend upon many factors, and in the present instance we do not feel that these licenses materially affect the issue of patentability.
For these reasons, we are of the opinion that claim 14 of MacLaren and claims 6, 22, 23, and 27 of Aceves are not infringed and are invalid.
The petitions are accordingly dismissed.
It is so ordered.
HOWELL, MADDEN, WHITAKER, and LITTLETON, Judges, concur.