Opinion
Nos. 4750, 4844.
October 3, 1932. Rehearing Denied October 28, 1932.
Appeal from the District Court of the United States for the District of New Jersey; William Clark, Judge.
Patent infringement suit by Steel Tubes, Incorporated, against the General Tube Company. From the decree holding patents valid and infringed, but limiting scope of infringement, both parties appeal.
Modified and affirmed.
Drury W. Cooper, Dean S. Edmonds, and Ernest D. Given, all of New York City, for Steel Tubes, Inc.
Fred A. Klein, of New York City, Joseph G.M. Browne, of Brooklyn, N.Y., and Briesen Schrenk, of New York City (Henry C. Quigley, Jr., of New York City, of counsel), for General Tube Co.
Before WOOLLEY, DAVIS, and THOMPSON, Circuit Judges.
In this suit for infringement of claims 4, 5, 10, 14, 17, and 19 of the Johnston patent, No. 1,388,434, for a new and improved method and apparatus for butt-welding thingauge tubing, and of claims 3, 5, 6, and 9 of the Johnston patent, No. 1,435,306, for the product of the process, the District Court held both patents valid but limited their application to operations conducted at a speed in excess of thirty feet per minute. Both the plaintiff, Steel Tubes, Inc., and the defendant, General Tube Company, appealed to this court. The plaintiff's appeal relates only to that part of the decree which limits infringement of both product and process to a speed in excess of thirty feet per minute. The defendant's appeal goes to the whole case, questioning both validity and infringement.
The process, and product thereof, of the patents in suit relate to the manufacture of steel tubing by folding a flat strip of steel into the form of a tube, having a lengthwise seam and then welding the tube in a particular manner by the use of electricity.
The patents in suit have been sustained and held infringed in three separate suits brought by the plaintiff or its predecessor. Elyria Iron Steel Company v. Mohegan Tube Company, Inc., 7 F.2d 827 (C.C.A. 2); Steel Tubes, Inc., v. Greenpoint Metallic Bed Company, 37 F.2d 172 (D.C., E.D.N.Y.); Steel Tubes, Inc., v. S. Jackson Tube Company, Inc., 42 F.2d 760 (D.C., E.D.N.Y.). We are in accord with the conclusions reached in those cases. It would be useless, therefore, for us to go into a detailed discussion and repeat here what has been well said there.
Claims 4, 5, 10, 14, 17, and 19 of the process patent are in issue. Claim 19 is typical and is as follows: "The method of electrically butt-welding thin walled tubing which consists in applying welding pressure to a transverse, narrow zone of the tube stock intimately to close the seam cleft and applying impulsive welding current to traverse said zone, all in such continuous and rapid progression and with such pressure and current density as to produce a substantially continuous weld exhibiting in the seam a recurrent welding effect synchronizing with the impulses of the welding current."
In the Mohegan Case, the Circuit Court of Appeals for the Second Circuit compared the Johnston process with the process described in the Parpart patent (No. 658,741), which admittedly represented the highest development of the art of butt-welding thinwalled tubing by the electric resistance method before Johnston. We know of no better way to a ready understanding of the Johnston process.
Parpart caused the folded steel strip to be fed on a compression roll by copper electrode rolls, embracing the seam of the tubing, through which an electric current flowed to the seams of the stock. By a series of condensed shots of the electric current, the metal at the edge of the seams and adjacent thereto is heated to such a degree that the pressure exerted by the conducting rolls and the compression rolls (which are preferably offset from one-eighth to one-fourth of an inch from the compression rolls) combine "to make a definite and material upset on each of the two meeting edge portions of the tube, the construction and adaptation of all parts being such as to insure such upsetting formation to a substantial degree." That is, the edges are pushed together to secure the weld beyond the damaged edges through which the series of shots of the heavy electric current must pass. Consequently, such a weld must result in metal that is pushed aside on the upset forming a considerable "burr" at the seam both on the outside and inside of the tube. There is no need to repeat the disadvantage of such a product.
In addition to the waste that resulted from the manufacture of the Parpart tubing and the costly process of removing the burr, it appears that the machines on which Parpart was practiced commercially could not be operated at a speed greater than twenty feet per minute because the weld required the metal to heat uniformly for some disstance back from the edges of the blanks.
Parpart used a certain correlation of the elements of speed, current, and pressure. The speed was such as to allow sufficient time for the electrical current to heat the metal to a fusion temperature for some distance back from the seam edges. Pressure was such as to insure a uniform "body-weld" back of the seam edges so that the burnt, damaged metal at the seam edges would be pushed out and extended.
Johnston's process was evolved to obtain a product that could be manufactured with speed and less waste and without the heavy burr. What Johnston did was to employ a new correlation of the variable elements of speed, current, and pressure. In the specification of his patent No. 1,435,306, he describes this as follows: "In practice I have found that for welding one-inch tubing of say .025 inch thickness, utilizing 60 cycle alternating current to supply the primary circuit of the transformer, the apparent welding current may be about 12,000 amperes at an apparent voltage of approximately 1.5 volts or a little more, as measured across the insulated gap of the roller electrodes at the nearest accessible point to their lines of contact with the tube being welded, and the speed of progression of the tube-stock may be over 70 feet per minute, to state a modest rate, which under proper current changes, can be considerably exceeded." That is, the changes are susceptible of calculation. Johnston conceived the idea that there was too much pressure and heat. The heat extended far back from the edges. By just bringing the edges together with little pressure and only enough heat to heat the edges to a welding temperature, greater speed was obtained; the tube was better welded with practically no burr or upset. No waste resulted. The seam was welded by single impulses or shots of the alternating current. The characteristics of the product is a "stitch seam." "The resultant of this correlation of forces and elements was and is the stitch seam; a weld line which looks as if it had been sewed, that is, the `recurrent' weld of the patent in suit. If any two of the following elements be known, the other can be computed, viz.: Speed per minute in feet, frequency of cycles of alternate current, and number of stitches per foot."
The plaintiff contends that when Johnston entered the field in 1919, the art of electrical resistance welding had been well developed and widely practiced. Among the numerous patents cited, Thomson, Ries, and Rietzal, Gorton, Baehr, and Guay are urged as anticipations.
The Thomson patents, Nos. 347,140, 347,141, and 444,928 were considered by the Supreme Court in Thomson Spot Welder Company v. Ford Motor Company, 265 U.S. 445, 44 S. Ct. 533, 68 L. Ed. 1098, wherein it was held that the Thomson patents anticipated the Harmatta patent, No. 1,046,066 (1912), for spot-welding of metal plates. Thomson's patents describe a process of welding wires, bars, the formation of elongated, longitudinal joints by heating with an electric current of low voltage and high amperage and simultaneously applying pressure. He states that a continuous, intermittent, or direct current may be employed and that it is preferable to confine the heating to the surface of the weld.
The Ries patent, 611,222 (1898), apparently describes an impractical method of manufacturing metal tubing. The weld is to be formed by "heat generated by springing an are between the meeting edges and the terminal of an electric circuit or the are may be formed adjacent to the meeting edges between two terminals of the electric current, or the current may be passed directly through the formed tube, so that the higher resistance offered by the meeting edges will cause them to be heated to the desired welding temperature." At any rate, Ries seeks a "beaded seam" formed after heating, closing the seams with the "spring pressure" caused by the tube passing through the forming dies.
Rietzal, No. 666,157 (1901), Gorton, 681,694 (1901), Baehr, No. 747,841 (1903), and Guay, No. 1,022,025 (1912), merely describe structural improvements to existing electrical welding apparatus.
There is no reason to borrow confusion on the issue of validity. The matter appears quite simple. In certain phases, the art of welding metal by the application of heat generated by electricity was well developed. Patents in the field taught the necessity of using a heavy current of low voltage to heat metal to a welding temperature, confining the heat to the point of the weld and insuring uniformity of the application of the heat with simultaneous or immediate application of pressure to secure the weld. Johnston knew the art, and with his knowledge unquestionably he taught something new.
The commercial development of electrically welded metal tubing had advanced no nearer than Parpart. Edge-welding had been suggested; likewise, the type of current used and method of applying pressure. But we do not find an approach to Johnston's combination of the elements of heat, pressure, and speed resulting in a commercially successful product in spite of the number of years of experiment between Parpart's invention and that of Johnston. The fact that Johnston immediately attained success, that his patent has often been involved in litigation, and that during a long period in which the Parpart machine was operated by those skilled in the art of welding without conceiving the process which the defendant admits can be practiced on Parpart's machine by simple reorganization, leaves small doubt that Johnston's useful and practical disclosure involves invention.
The defendant contends that the Johnston patent is invalid because of his failure to file a disclaimer within the proper period of time after claims 1, 2, and 22 of Johnston, No. 1,388,434 were held invalid in Steel Tubes, Inc., v. Greenpoint Metallic Bed Co., Inc. (D.C.) 37 F.2d 172. On January 31, 1930, the court entered an interlocutory decree holding the claims invalid; the plaintiff filed a disclaimer on February 24, 1930; three days later the court entered the final decree; and the plaintiff took no further action in regard to the invalid claims. The defendant contends that the plaintiff should have filed a second disclaimer within 30 days of the final decree. Ensten v. Simon Ascher Company (C.C.A.) 38 F.2d 71, judgment affirmed in 282 U.S. 445, 51 S. Ct. 207, 75 L. Ed. 453. The plaintiff disclaimed as required by the law. Defendant's argument has no merit.
Coming to the question of infringement, the evidence shows that Parpart's machines may be and have been changed by skilled mechanics who have had Johnston's specifications before them to operate under Johnston's process and to produce his product. It also shows that the defendant purchased two of the Parpart machines. As stated above, the characteristics of Parpart's tubing are easily distinguished from the plaintiff's product even by the inexpert, by the heavy burr of the body-weld before its removal and the homogeneous seam afterwards. The tubing manufactured by Johnston's process has a small, lengthwise burr which has recurrent stitches representing each "shot" of the alternating current. If the teachings of Parpart are followed this stitch seam cannot be produced because he required a plurality of the impulses of the current to heat the metal contiguous to the seam edges to order to effect his body-weld. In this case, the plaintiff has introduced specimens of defendant's tubing which cannot be distinguished from Johnston's product. The conclusion to be drawn is obvious and there is considerably more evidence to support it in the history of defendant's machines and testimony of the witnesses.
We are of the opinion that the learned trial judge correctly held that the patents were valid and infringed, but we are unable to agree that they should be limited to speeds of operation in excess of thirty feet per minute, for we do not understand that the invention is dependent on high speed for success. The speed of the Johnston process is one of its chief merits but it is a mere economical result. We have endeavored to explain that Johnston obtains his peculiar edge-weld by properly applying the variable factors of heating current, the length of time of application of the current, and pressure. Each of these factors is variable and, while independent in themselves, are interdependent in the practice of the invention. It follows that the patents are valid at all speeds and infringed.
The decree, as modified, is affirmed, with costs.