Appeals from the District Court of the United States for the Northern District of Illinois, Eastern Division; William H. Holly, Judge.
Before MAJOR and TREANOR, Circuit Judges, and LINDLEY, District Judge.
The Johnson Laboratories, Inc., brought suit against the Meissner Manufacturing Company, for infringement of Claim 38 of Polydoroff Patent No. 1,982,690 and Claim 5 of Crossley et all., Patent No. 1,978,568, both assigned to Johnson. The court held invalid Claim 5 of the Crossley patent and valid and infringed Claim 38 of Polydoroff. Each party appealed from that part of the decree against it. The Ferrocart Corporation of America brought suit for a declaratory judgment under Section 274d of the Judicial Code, 28 U.S.C.A. § 400, to determine the questions of validity and infringement of seven of Johnson's patents, including those involved in the first suit, and charged unfair competition by Johnson Laboratories, Inc., and Aladdin Industries, Inc. The two defendants thereupon filed their counterclaim against Ferrocart for infringement of four patents and for unfair competition. Ferrocart appeals from a decree declaring valid and infringed Polydoroff Patent No. 1,982,690, involved in the first mentioned suit, and holding that it had unfairly competed with Johnson and Aladdin. The latter appeal from that portion of the decree holding Crossley et al., Patent No. 1,978,568 included in the first mentioned suit, and Polydoroff Patent No. 1,982,689 invalid and decreeing that the two defendants had unfairly competed with Ferrocart.
Ferrocart maintains its office in New York and is a corporation of that state. It imports merchandise from Germany and exploits certain inventions in the radio art and sells to Meissner Manufacturing Company certain cores for use in high-frequency inductance devices. These two corporations were represented in the two suits by the same counsel and their interests are identical. The same is true of Johnson and Aladdin. The various appeals were argued as one and this opinion will dispose of the issues in each. It will be observed that two of the patents are involved in each of the two suits and that the decree of the court was the same with regard thereto in each cause, the first Polydoroff patent being held valid and infringed and the Crossley patent invalid. The only other patent now involved is the second Polydoroff, which, in the second suit, was held invalid.
Polydoroff Patent No. 1,982,690, was applied for August 26, 1929, and issued on divisional application December 19, 1933. Claim 38 only is involved. It covers "a high-frequency inductance device for use in resonant circuits, including at least one low-loss winding and a compressed comminuted magnetic core having insulated particles small enough to pass through a screen having 300 meshes to the inch, the increase in the effective inductance of said winding due to said core being substantially greater than the increase in the effective resistance of said winding due to said core." Such a device, as known to the electrical world, is, in ordinary language, one consisting of a number of turns of wire, whereby the resulting coil possesses the property of electrical inductance. If it is wound upon a non-magnetic core, it is commonly known as an "air-core inductance"; if upon a magnetic core, of iron or other metal, as a "magnetic core inductance." A resonant selective radio circuit, in which the patentee said his device was useful, is one tuned to distinguish between broadcasting stations with various adjacent frequencies. By the tuning, the receiver may obtain reception at one frequency and exclude broadcasters transmitting on adjacent but different frequencies.
The difficulty encountered by many workers in the art of such devices arose largely from the fact that all magnetic core devices inherently possess a quality not desirable in such a circuit, namely: electrical resistance, - inducing losses and tending to destroy selectivity of tuning. In practice, therefore, an inductance device adapted to the frequencies encountered in radio broadcasting will not work satisfactorily if the resistance quality is so great as to defeat the purpose. In other words, the ratio of inductance to resistance must be a useful practical one, - the inductance comparatively large and the resistance proportionately small. It had long been recognized that a magnetic core in a coil serves to increase the inductance but, due to the fact that socalled eddy currents circulate within the core and absorb energy, it likewise increases the resistance, and it seems to have been largely accepted that the additional resistance due to the use of a magnetic core increases rapidly with increased frequency, though the increased inductance due to such a core is substantially independent of frequency. Consequently many in the art believed that such a magnetic core, which improves inductance at low frequency, will improve it to a lesser degree at a higher frequency and, if the frequency be increased sufficiently, becomes positively harmful. The problem, then, of makers of such devices, to be placed in radio receiving sets, was to produce, under the frequencies controlling in broadcasting, the highest possible practical inductance and the lowest possible interfering resistance.
It is insisted that Polydoroff was the first to solve the problem and in so doing achieved invention. His claim is limited first of all to a "high frequency" inductance device to be used in "resonant circuits." He provides in his combination, first, a low-loss winding coil; second, a compressed comminuted magnetic core having insulated particles, and, finally, the requirement that the particles shall pass through a screen having 300 meshes to the inch. This combination, he said, brought about an increase in the effective inductance of the coil winding, due to the magnetic core, substantially greater than the corresponding increase in the effective resistance of the winding, due to the core. He recognized that successful attempts had been made to employ magnetic cores but asserted that the losses from increased resistance arising in their utilization had made impractical their use in high frequency radio transformers. He said that by using powdered-iron cores it was possible to decrease the size of the windings and thereby minimize the size of the transformers, thus saving cost of space. He asserted that for the most satisfactory results "for frequencies between 1500 k.c. and 1000 k.c." i.e., between 1,500,000 and 1,000,000 cycles, the builder should use iron "reduced by hydrogen," consisting of particles which would fall freely through a screen of 300 meshes to the square inch. He recognized that coarser particles might be used for frequencies below 1,000 k.c. He stated that the fineness of the particles and, to some extent, the degree of compression determined the resistance of the coil and core combination.
It was well known prior to Polydoroff that losses in a ferromagnetic or iron core could be reduced by subdivision of the material into thin sheets. But Johnson insists that this prior use was confined to "low frequencies" namely: those of from 20 to 60 cycles. It was well known also that comminuting or powdering the core material, producing still greater subdivision of material, which could be compressed, or molded into proper form with or without insulation, to form a compacted mass, was similarly useful but, again Johnson says, only for low frequency inductance devices. Such cores had been used for many years for telephone coils, but it is claimed that Polydoroff was the first to use them in combination with a low-loss coil to bring about an efficient device for use in resonant circuits at "high-frequencies." Many writers had previously taught that magnetic cores were improper at high frequencies; and it is insisted that it was novel for him to combine a comminuted core and a low-loss coil in a device effcient and practical in a resonant high frequency circuit.
Unfortunately the term "high frequen cy" had no settled, definite meaning. Its connotation is relative in character, and there is nothing in Polydoroff's patent to disclose what he intended thereby except his reference to 550 k.c. as an example of lower frequencies. Where he thought "low-frequency" ended and "high-frequency" began is not clear. In the Crossley patent in suit the patentee described 550 k.c. as being within upper high frequency circuits. Johnson offered evidence of one authority to the effect that high radio frequencies include those from 100,000 to 1,000,000 cycles. This divergency is even wider if we heed other authorities. Ferrocart's devices operated at 456 k.c. In Hazeltine Corporation v. Radio Corporation of America, D.C., 52 F.2d 504, the court approved a division between high and low frequencies at the point of 20,000 cycles or 20 k.c., holding those below that point low frequencies and those above high frequencies. It becomes necessary, therefore, to determine whether the prior art discloses or suggests the use of such cores with particles of the suggested size at any frequency within contemplation by Polydoroff when he made his application.
Before examining the specific frequencies at which prior art workers performed their acts, however, it should be observed that it was well known that as the frequency of the circuit increases the size of the particles used in the magnetic core must be reduced to maintain the same efficiency in the core and coil combination. The higher the frequency the finer the particles should be. This was taught by Speed and Elmen in 1921, and appears in various references of the prior art. Indeed, Polydoroff himself stated that coarser particles of magnetic material in the core could be used for frequencies below 1,000 k.c., that is, particles coarser than those passing through a 300 mesh screen.
Stone's No. 767,977, in 1904, in discussing cores used by him in a device for wireless telegraphy, said that they might be made of finely-divided soft iron. He suggested the use of finely-comminuted paramagnetic material, embedded in a dielectric matrix, used as a means to enhance the inductance of a coil without introducing the effects of variable permeability or hysteresis losses.
MacKnight, in 1924, received a patent upon application made May 26, 1922, in which he stated that his object was to provide a core which would reduce to a minimum losses of electrical energy due to eddycurrent and hysteresis and still permit the desired density of magnetic flux to be obtained. He said that he contemplated broadly constructing the cores for all descriptions of electro-magnetic apparatus from finely divided insulated particles of iron and specifically included radio apparatus, thus apparently including devices for use in high frequencies, in resonant circuits. He commented that the higher the frequency of the current passing through the core, the finer should be the particles and said that he eliminated hysteresis by such cores. He suggested that the material might be suspended in a mobile insulating fluid but that the desired results might be accomplished by surrounding the iron particles with plastic or even solid insulation. He insisted that by employing particles 1/2000 of an inch in diameter and such as were not materially greater than 1/1000 of an inch in diameter he effected a marked saving in eddy-current losses in an apparatus employing high-frequency uurrent. 1/2000 of an inch is 12.70 microns; 1/1000 of an inch, 25.4 microns. Particles falling freely through a 300 mesh screen are 45 microns in diameter. He insisted that by his method he was able to use magnetic cores in apparatus where such cores previously had been impractical because of the great losses. It may be that MacKnight was not correct in some of his statements regarding the art but he clearly taught the thoughts we have referred to. Such affirmative teaching is applicable prior art.
Hochheim procured three patents: United States No. 1,840,286, issued January 5, 1932, upon application filed August 13, 1926; British No. 269,770 and German No. 473,480. They are in substantial accord in disclosure. In his American patent, Hochheim said that he had found it of great advantage to make metallic cores for transformers, which are to be operated with rapid and slow oscillations of current, of particles of iron extremely finely subdivided. This powder, he suggested, might be fixed with a small amount of binding or adhesive insulation substance before being molded into the desired form, and he made the assertion that cores thus prepared were superior to those prepared in a similar manner from metal previously marketed, resulting in small eddy-current losses. He prescribed their use in wireless telegraphy or any apparatus working at both slow and rapid oscillation. He said that the hysteresis losses were small and that the cores were especially suitable for use with various "rapid oscillations." He suggested particles generally the size of 0.001. millimetre but said that particles even of .00001 millimetre might be used. This is a range between a diameter of 1 and one of .01 microns. The fair construction of his specifications is the suggestion of use in the high frequency field.
In 1919 Fleming published a book describing a radio receiving set with transformers connecting a plurality of tubes and embracing comminuted iron cores.By this device were received signals forwarded by transmitters on a ship on a wave length of 600 meters or 500 k.c., - 500,000 cycles. Apparently this is within anyone's understanding of a high frequency inductance device, and, clearly, the circuit was a resonant one.
Polydoroff prescribed particles small enough to pass through a screen having 300 meshes to the inch. Translated this means 45 microns in diameter. Obviously smaller particles would pass, but, as Johnson's expert stated, the excellence of the core material is to be determined not by the smallest particle it contains but by the largest. It is the surface and volume of the large particles which "play the most important part," and Polydoroff in prescribing particles which fall through a screen having 300 meshes ot the square inch was providing for large particles, having considerable influence upon the performance of the core material. A comparison of other products shows tht Speed and Elmen in their article prescribed for two materials an 80 mesh screen and for grade "C" a 200 mesh, through which could pass particles as large as 73 microns in diameter.Hoerning prescribed 250 meshes to the square inch, permitting passage of particles with a diameter of 60 microns. MacKnight prescribed particles 12.5 microns in diameter at 100 k.c. and smaller ones with increasing frequency. Felten, in his British patent, prescribed particles of 10 microns diameter at a frequency of 1,000 k.c. Polydoroff prescribed 300 meshes to the square inch, which would permit the passage of particles 45 microns in diameter. Polydoroff in another patent in suit prescribed 400 meshes to the square inch, permitting the passage of particles 35 microns in diameter. The alleged infringing device uses particles from 1 to 5 microns in diameter; the ...