Opinion
No. 9376.
June 24, 1943.
Appeal from the District Court of the United States for the Southern District of Ohio, Eastern Division; Robert R. Nevin, Judge.
Action by International Steel Wool Corporation against the Williams Company for patent infringement, wherein defendant filed a counterclaim which was subsequently voluntarily dismissed. From a judgment dismissing the bill of complaint, the plaintiff appeals.
Affirmed.
Marston Allen and Erastus S. Allen, both of Cincinnati, Ohio (Allen Allen, of Cincinnati, Ohio, and Paul A. Staley, of Springfield, Ohio, on the brief), for appellant.
Albert L. Ely, of Cleveland, Ohio (John M. Cole, of Springfield, Ohio, Ely Frye, of Cleveland, Ohio, and Cole Hodge, of Springfield, Ohio, on the brief), for appellee.
Before ALLEN, HAMILTON, and MARTIN, Circuit Judges.
This appeal arises out of a suit charging infringement of United States patent 1,907,453, issued to appellant as assignee of William A. Steinbart, who filed the patent application. The patent is for a method of and a machine for making steel wool, article claims 17, 23 and 29, and method claim 34 being in suit. A counterclaim filed by appellee for infringement of Robbins patent, 1,584,145, for improvements in a machine for cutting steel wool, was voluntarily dismissed with prejudice. The District Court held the claims in suit invalid and not infringed, and dismissed the bill.
The parties to this suit were also parties to an interference in the United States Patent Office (Robbins v. Steinbart). The appellee is the owner of the Robbins patent, supra, which came into interference with the Steinbart application after the Robbins patent was granted. In Robbins v. Steinbart, 57 F.2d 378, 19 C.C.P.A., Patents, 1069, the Court of Customs and Patent Appeals sustained the Board of Patent Appeals in awarding priority to Steinbart on claim 23, stating that while the court might have arrived at a different conclusion if it were an initial question, nevertheless the court was not convinced that the lower tribunals were manifestly wrong.
Appellant contends that since the appellee was a party to the interference proceedings and in this case filed a counterclaim for infringement of the Robbins patent, holding itself out as being the owner of a valid patent for the same invention, it cannot in good faith assert the defense of invalidity. We think this contention has no merit. However inconsistent appellee's former attempt to procure a patent may be with its present contention of invalidity of the Steinbart patent for want of invention, the Supreme Court has long recognized that such inconsistency affords no basis for an estoppel and does not preclude the court from relieving the alleged infringer and the public from the asserted monopoly when there is no invention. Paramount-Publix Corp. v. American Tri-Ergon Corp., 294 U.S. 464, 477, 55 S.Ct. 449, 79 L.Ed. 997. This is because the defense of want of patentable invention in a patent operates not merely to exonerate the defendant, but to relieve the public from an asserted monopoly, and the court cannot be prevented from so declaring by the fact that the defendant had ineffectually sought to secure the monopoly for himself. Haughey v. Lee, 151 U.S. 282, 285, 14 S.Ct. 331, 38 L.Ed. 162; Kellogg Switchboard Supply Co. v. Michigan Bell Telephone Co., 6 Cir., 99 F.2d 203.
The early machines for cutting steel wool comprised a cutting table with reversible drums for winding and rewinding wire, and double-edged knives held in tilting holders arranged to be reversely tipped as the wire moved to and fro. In these machines the series of knives operated on only one strand of wire. Later machines, such as those used in 1914 at the plant of the Ridgely Trimmer Company, Springfield, Ohio, included idler pulleys placed at the ends of the table around which the wire looped so that two strands were cut by the gang of knives. Even so, the necessity for reversing the wire reduced the speed and efficiency of the operation. Steinbart declared it to be one of his principal objects to solve this problem. He stated that his device was intended to provide "a method and apparatus for making metal wool wherein a length of wire may be fed and guided through an apparatus in a unidirectional way and reduced to attenuated condition by a single run through the apparatus, thus enabling a much greater quantity of wool to be made in a given length of time than by other methods and apparatus and also obviating the necessity of numerous reversals in the direction of travel of the wire from which the wool is formed."
The patent in suit discloses a machine consisting of a base on which are mounted two vertical parallel pulleys about which a steel wire is wrapped in a plurality of parallel convolutions. The pulleys are grooved to receive the wire and are driven by a motor located in the base of the machine and connected to the shafts of the vertical pulleys. 560 knives are disclosed, each knife operating upon a single strand of wire which is supported beneath the knives by blocks. The shavings from the wire (in the patent denominated "shavings" or "fibrous shavings"), constitute the cuttings of steel wool.
At the sides of the machine are located parallel guideways in which are supported vertical boards upon which the knives are mounted. The knives were directed in the original application to be reciprocated during the operation of the machine for the purpose of crosscutting the wire, and the issued patent retains this feature in a number of claims. Steinbart joined the end of one wire to the next by a coupling and provided lever mechanism to raise the knives to permit the coupling to pass under the knives in succession. By an interlocking connection between all the levers on each board, as each knife is raised or lowered, all of the other knives on the same board are raised or lowered simultaneously. The District Court found that as the lever mechanism is carried by the reciprocating boards and the wire is moving in a fixed path, any certain coaction between the coupling and the lever mechanism would be impossible, with the result that even if the coupling could pass the blocks, it would strike some, if not all of the knives, either breaking the knives or stopping the machine. No provision was made for adjusting the knives to compensate for the gradual reduction of the wire. The District Court held that the patent is inoperative for the purpose of cutting wool, and concluded that the only portions of the Steinbart machine which are operative are the vertical pulleys and their power-driving means.
The Court of Customs and Patent Appeals in Robbins v. Steinbart, supra, where it was also contended that the Steinbart machine was inoperative, declared: "It seems obvious that the knives would not function as intended; that the drums would not rotate properly; and that the machine had other serious defects which rendered it, as the Board held, an impracticable construction." (Page 380 of 57 F.2d.) However, that court, declaring that the knife mounting was "no part of the invention of the involved claims," decided that the features of the device responsible for its alleged inoperativeness could have been corrected without the exercise of inventive skill.
Appellant attacks the finding of the District Court on the issue of inoperativeness, contending that the appellee is bound by the decision of the Court of Customs and Patent Appeals within the doctrine of Morgan v. Daniels, 123 U.S. 120, 14 S.Ct. 772, 38 L. Ed. 657. It does not claim that the Steinbart machine as originally disclosed is not inoperative, but states that since its defects can "be corrected by the use of standard equipment in wool cutting," it cannot be said to be inoperative.
We think that the ruling in Morgan v. Daniels to the effect that the decision of the patent tribunals upon priority of invention is controlling upon that question of fact in any subsequent case between the same parties is not in point here, where the issue is not priority, but validity of invention. Also we consider that the findings of the two courts upon the issue are entirely consistent. Both of them find the cutting mechanism of Steinbart to be completely inoperative. The District Court does not hold the patent invalid because of inoperativeness, nor does it state that the defective features could not have been cured by exercise of ordinary mechanical skill. We regard the finding of the District Court upon this issue as limiting the scope of the inquiry upon article claims 23 and 29 to the invention residing in the parallel pulleys and their power-driving means, as embodied in a machine for the cutting of steel wool. That this is a correct limitation is virtually conceded by the appellant, for one of its experts stated that the gist of the claimed invention resides in the feature that each loop of wire is independently driven by the application of power to the sheaves or drums, thus taking away the strain from the wire. He said:
"The dominant mechanical feature or principle of the Steinbart disclosure lies in simultaneously applying unidirectional driving power individually to each of a series of parallel convolutions independently of each other convolution of the series, and limiting the pulling strain imposed on each convolution to a fractional portion only of the total cutter resistance imposed upon the entire wire by subjecting each convolution to the cutting action and resistance of a relatively small number of cutters, far less than the number of cutters necessary to reduce the wire from its initial condition to an attenuated scrap condition, as distinguished from the prior art practice of subjecting the wire to a cumulative resistance of all the actual cutters by the application of power at a single point on the wire."
Claims 23 and 29 of the patent in suit read as follows:
"23. In a machine for manufacturing steel wool, a table, cutting tools arranged adjacent thereto, means to guide a plurality of parallel strands formed from a single wire in engagement with said tools over said table, and means to drive said guiding means so that each loop of wire will be independently driven by said guiding means whereby the driving strands will be applied to each loop independently against the resistance of the tools engaging the strand of wire of that loop."
"29. In a machine for manufacturing steel wool, a support, cutting tools arranged adjacent thereto, means about which a plurality of parallel strands of wire formed from a single wire pass, power-operated means for driving at least a part of said means whereby the driving strains will be applied independently to each loop of wire, and a plurality of cutting tools to engage the loops of wire."
The District Court held that these claims present an obvious mechanical expedient applied to the steel wool machines of the prior art, and we agree with this conclusion. So far as they present the application of power to wire-guiding pulleys, which, as stated by appellant's expert, is the dominant feature of the disclosure, they lack patentable invention.
It was old in the art to increase production from a single strand acted on by a single knife by guiding wire over grooved idler pulleys in two loops so that each knife would act upon two wires instead of one. Devices disclosing this feature were used at the Ridgely Trimmer plant in 1914. Schonitzer, 1,549,807, granted August 4, 1925, upon an application filed February 18, 1922, for a metal wool making machine, disclosed in the drawings three convolutions acted upon by the cutters. The wire was wrapped around the reels and power applied to one of the drums was transmitted through the wire and distributed to the several convolutions, thus, in the words of the specifications, insuring "a feeding pull upon such strands directly in the direction of their lengths. * * *" The claimed invention of Steinbart consisted of applying power to the idler pulleys so that each loop was independently driven, the driving strains were distributed and the loops could be greatly multiplied. We agree with the District Court that this was an obvious conception. Once the desirability of distributing the strain of the tool load was established, the mechanical skill of those familiar with engineering problems in the wire-working trades would naturally make use of known methods for solving the problem. We think that what was shown in the application of power to the wire-guiding pulleys was merely the accepted skill of the calling. It was obvious that by driving the pulleys the force to propel the wire would be distributed along the wire rather than concentrated at its end and the number of loops and cutting operations which could be performed at any one time on the wire could be increased to any desirable extent. Galvin, appellants' consultant and former general manager, stated that in 1920 it was obvious to him that if power was applied to the sheaves or drums, the number of loops of wire could be increased, and conceded that having conceived that idea, the building up of any number of loops is merely a matter of multiplication.
In addition, the conception of distributing power along a wire in the steel wool machine so as to relieve the wire of strain exerted by a number of knives, and the propulsion of the wire in one direction eliminating reversals in the direction of the travel of the wire had already been presented in the five Graf-Buchler patents, France, 321,028, May 10, 1902; Germany, 147,758, May 13, 1902; Switzerland, 28,187, May 8, 1903; United States, 761,585, May 31, 1904. These patents cover machines for the making of "stahlspanen," which technical dictionaries in evidence translate as "steel wool." They were not cited before the Patent Office. The advantage claimed to be secured from the Graf-Buchler patents, which are substantially the same, is significantly stated in the French patent to be "the considerable yield or a great rapidity of the work, and on the other hand the little stress on the material worked." The design of this patent presents a circular plate mounted on a shaft by which it may be placed in rotation, the plate carrying the steel wire on its circumference, a series of rollers periodically spaced which push against the circumference of the plate by springs or otherwise, and press the steel wire against the plate with such force that "in spite of the resistance created by the work to which it is subjected," the wire is carried along by adherence while the plate is rotated. Over a portion of the circumference of the plate there are distributed numerous cutters, each one of which cuts one shaving from the steel wire, so positioned as to cut shavings in such manner that there remains from the steel wire "only a very thin portion when leaving the machine." A similar statement in the Swiss patent is that "Only a very small residue of wire remains when it leaves the machine."
Appellant concedes in its brief that Graf "shows a cutting operation in which the strain of the pulleys is distributed along the wire," and seeks to avoid the inevitable conclusion that the Graf-Buchler patents anticipated the conception of distributing power along the wire in a steel wool machine so as to relieve the wire of pulling strain by asserting that the Graf-Buchler machine cut "shavings" and not steel wool. This is disproved by the mechanical dictionaries in evidence, and by the Steinbart patent which refers to the steel wool as shavings.
The record also demonstrates that the precise means which Steinbart adopted to carry out this conception was old and was identical with the means employed in the wire-drawing art to solve an identical problem. In continuous wire-drawing the wire is passed through a series of dies in order to reduce its cross-section. In such an operation the sum total of the resistance of the drawing dies is in excess of the load that the wire emerging from the last die can support. As the wire is reduced it becomes weaker, and a pull on the end of the wire strong enough to pull it through all the dies will break the wire. In this art, therefore, the identical problem claimed to be solved by Steinbart was presented, and presented while the steel wool art was in its infancy, namely, the problem of distributing the load to the different strands or convolutions. One solution of the problem in wire drawing was attained by wrapping the wire in a number of convolutions about a pair of parallel drums and applying a separate tool to each convolution. This practice exactly conforms to the provision in claim 23 of Steinbart "means to drive said guiding means so that each loop of wire will be independently driven by said guiding means whereby the driving strains will be applied to each loop independently against the resistence of the tools engaging the strand of wire of that loop." The following patents none of which was cited before the Patent Office, illustrate this development: Bolton, 397,272 (Feb. 5, 1889); Burnes, 427,150 (May 6, 1890); Lautzenheiser, 549,322 (Nov. 5, 1895); O'Donnell, 560,278 (May 19, 1896); O'Donnell, 725,472 (April 14, 1903).
The Lautzenheiser wire-drawing machine disclosed two power-driven wire-guiding pulleys. This machine was manufactured and sold by the Waterbury-Farrel Company, of Waterbury, Connecticut, some 377 of the machines having been sold from 1900 to 1939, and one of them being introduced in evidence in this case. The adoption of this idea from the wire-drawing art plainly did not constitute invention of the arts are analogous.
Appellant contends however that the steel-wool and wire-drawing arts are not analogous. Analogy depends not merely upon similarity of purpose, but upon similarity of elements. A.J. Deer Co., Inc., v. United States Slicing Machine Co., 7 Cir., 21 F.2d 812, 813. Invention does not consist in the mere conception of applying an old method or device to a new use, if the new use is so analogous to the old that the thought of adapting the device and applying it to the new use would occur to a person skilled in the art and seeking to devise means to perform the desired function. In such case even though changes or modifications are essential to the practical application of the method or device to the new use, invention is not involved. Concrete Appliances Co. v. Gomery, 269 U.S. 177, 181, 46 S.Ct. 42, 70 L.Ed. 222; Vandenburgh v. Truscon Steel Co., 261 U.S. 6, 15, 43 S.Ct. 331, 67 L.Ed. 507; Lovell Mfg. Co. v. Cary, 147 U.S. 623, 13 S.Ct. 472, 37 L. Ed. 307.
Here we think the similarity of elements exists. The arts both deal with wire and with the reductio of wire, the steel-wool art developing subsequently to the wire-drawing art, and closely connected with it. Those skilled in the steel-wool art would certainly be as apt to be acquainted with the devices of the wire drawing art as concrete mixers with methods of conveying grain or ore by gravity to the points desired. Concrete Appliances Co. v. Gomery, supra. This record shows that the steel wool experts actually had an intimate knowledge of the development of the wire-drawing art. We think these fields are closer than the fields of metal-working and re-enforcing concrete held in Vandenburgh v. Truscon Steel Co. to be kindred, and the art of restoring coil springs in furniture to the art of restoring springs in clock bells and marine clocks. Lovell Mfg. Co. v. Cary, supra. More must be done than to utilize the skill of the art in bringing old tools into a new combination Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 89, 62 S.Ct. 371, 86 L.Ed. 58. This is the situation presented here. Whatever differences of operation exist in the two arts, they do not affect the immediate problem and its solution, which is precisely the same in the two arts, namely, the reduction of the aggregate tool load by the application of power at spaced periods, thus securing a continuous unidirectional operation and increased production. Clearly the application of this identical principle in the wire-drawing art prior to Steinbart strongly supports the conclusion that claims 23 and 29 are void for lack of patentable novelty.
Claims 17 and 34 read as follows:
"17. In a machine for making metal wool, a plurality of feeding and guiding rolls arranged in parallel relation, said rolls acting to feed and guide a wire in a series of convolutions, means other than the wire for rotating said rolls at the same speed, and cutters acting upon said convolutions of wire during their travel about said members to cut fibrous shavings therefrom, said cutters being of sufficient number to gradually reduce the wire to attenuated form during a single continuous travel of the wire past the cutters."
"34. The method of making metal wool consisting in causing a length of wire to travel at all times in a unidirectional way from a point of supply to a point of discharge in the form of a series of convolutions by applying a feeding action directly to the convolutions, and applying to the convolutions a series of cutting operations sufficient to reduce the wire to an attenuated condition by a single unidirectional travel of the wire."
While claim 17 is an apparatus claim and claim 34 is for a process, each of these claims covers the reduction of the wire to attenuated form during a "single unidirectional travel" or a "single continuous travel" of the wire past the cutters.
We agree with the District Court that these claims are invalid. At the time of Steinbart's application the conception of reducing the wire to attenuated form without reversal of the wire by a single continuous operation in a steel wool machine had already been presented in the various Graf-Buchler patents, supra. Moreover, the claims are void for indefiniteness. They do not specify whether a full round wire is to be employed, what grade of steel wool is to be produced, nor the degree of attenuation to be achieved. The number of cutting operations is not defined, nor can it be inferred from the described factors which must determine the number of cutting operations.
Appellant contends with great vigor that the entire art shifted after the publication of Steinbart's patent from previous expensive processes and machinery used for generations to machinery operating in principle as the patentee suggested, and claims that this, the reduction in cost of manufacture, and the production of highly superior steel wool is evidence strongly probative of invention, which was ignored by the trial court. As already shown, machinery had operated in principle as the patentee suggested long prior to the application for his patent, and the shift in process on this record is not shown to be due to the knowledge of the claimed outstanding success of his first machine. As a matter of fact no such machine as is shown and described in Steinbart was ever built or used, although for a number of years a machine in use at appellant's plant in Springfield, Ohio, was called a Steinbart machine. This machine fails to disclose many features of the Steinbart patent, such as the trip mechanism to permit the passage of the coupling of the wire and the means for raising and lowering the supports on which the cutters are carried. Also it embodies many of the features of the Galvin patent, 1,977,053. As testified by Galvin, these features include the manner in which the knives are adjusted with the pivot, the wire guide, the rails bearing on the flat surface of the wire, the slides arranged along either side of the wire loops, the oil pad, the drive mechanism, and the takeoff, all of these being identical with the Galvin device. The principal features common to Steinbart and to appellant's machine are the vertical power-driven pulleys about which the wire is passed. The results of the operation of appellant's machine so glowingly asserted cannot be ascribed to Steinbart.
Nor is it shown that the development of the highly superior types and widely varying grades of steel wool and steel wool products introduced in evidence, the cutting of expense and the increased efficiency of operation, were in any way due to Steinbart. The production of the ribbon wool, which is greatly emphasized, is shown to be made possible largely by the simple expedient of looping up the strands of steel wool upon hooks or arms which are no part of the patented mechanism, but merely an efficient arrangement for preventing the wool from falling into a mass on the floor. Galvin conceded this in effect. So far as the possibility of making different grades of wool is concerned, this was at least foreshadowed in Graf-Buchler, supra. The circular introduced in evidence, advertising the Graf-Buchler steel wool, described different grades of steel wool. One of the pictures of the circular shows a narrow wire called "lang-gezogen," or long-drawn out. These long-drawn out strands are further described in the circular as "faden," or threads. It is an easy non-inventive step from "threads" to "ribbons" of steel wool, one that could confidently be expected from the usual mechanical skill of the art. The conclusion of the District Court that any commercial success of the International machines is not attributable to the Steinbart patent is amply sustained by the record.
The District Court found that appellee's machine does not infringe. It consists of an installation of eleven identical steel wool machines arranged in line without any physical connection except a common drive-shaft. The District Court, which saw all of the important machines in operation, found that each of appellee's machines "comprised a pair of idler pulleys at the ends of the cutting table over which the wire passes in fifteen parallel convolutions. Each knife on a machine cuts all fifteen convolutions at one time. The wire is propelled by the so-called `clover-leaf drive' located above the cutting table on each machine. The wire passes from one machine to the next and then to the next and so on. Each machine performs the same operation upon the wire and the wire is progressively reduced by the sum of the operations of all of the machines." Hence it concluded that appellee's means and method of operation are separate and distinct from those disclosed by Steinbart. In view of our decision as to invalidity we do not discuss this feature of the case.
The decree is affirmed.