Opinion
No. 3604.
July 24, 1930.
Cooper, Kerr Dunham, of New York City (Drury W. Cooper and Dean S. Edmonds, both of New York City, F.O. Richey, of Cleveland, Ohio, and Ernest D. Given, of New York City, of counsel), for plaintiff.
Fish, Richardson Neave, of Boston, Mass. (J.L. Stackpole and H.L. Kirkpatrick, both of Boston, Mass., of counsel), for defendant.
In Equity. Patent infringement suit by Steel Tubes, Inc., against the S. Jackson Tube Company, Inc.
Decree for plaintiff.
This is a patent infringement suit in which the defendant is charged with the infringement of United States letters patent No. 1,388,434, granted August 23, 1921, to G.V. Johnston, for a method and apparatus for butt welding thin gage tubing, letters patent No. 1,435,306, granted to the same inventor for butt welded thin walled tubing, and letters patent No. 1,611, 875, to H. Belmont, December 28, 1926, for a welding apparatus.
The first two patents have been before this court and before the Circuit Court of Appeals for the Second Circuit [ 7 F.2d 827], and, the validity of the claims in suit having been sustained, the defendant herein does not raise the issue of validity, if these claims be limited to a butt welding process and product.
The Belmont patent has not been the subject of prior adjudication, and its validity is contested. Noninfringement is urged both as to the Belmont patent and the two Johnston patents.
Claims 4, 5, 10, 16, 17, and 19 of the Johnston process patents, claims 3, 5, 6, and 9 of the Johnston product patent, and claims 1, 2, 3, and 4 of the Belmont patent are in issue.
Inasmuch as the prior art in Steel and Tubes, Inc., v. Greenpoint Metallic Bed Co., Inc. (D.C.) 37 F.2d 172, relating to the process and apparatus for butt welding thin gage tubing and the resulting product, was rather fully reviewed, and inasmuch as I still hold to the views which I there expressed, no useful purpose will be served by a review of that art herein; but, for the purposes of explanation, it will be adopted as though specifically set forth in this opinion.
The question of infringement presented in this suit is somewhat different from that urged in the former litigations of the Johnston patents because of the difference in the apparatus employed by this defendant. This difference arises out of the fact that in defendant's apparatus one of the two electrode rollers may be moved by its adjusting screw to a lower position than the companion electrode roller.
I do not understand that, as operated by the defendant in the manufacture of tubing, for example, shown by Exhibit No. 5, one electrode was lower than the other. On the contrary, I understand that the two electrodes were adjusted to the same horizontal level. So that process and the product must be held to infringe.
The larger and more important question, since the early use hereinabove referred to involved but 100,000 pounds of steel strip, concerns the process employed by the defendant when the two electrodes are not on the same horizontal level; and indeed the essential controversy herein revolves about that condition.
The defendant contends that by lowering one of the electrodes it practices, not a butt welding, but a lap welding, process, and that the resulting product is not a butt welded but a lap welded tube. The defendant argues that in so doing it avoids infringement of both patents, because Johnston urged as a condition of successful operation of his process that the edges of the seam must be brought into accurate registry. On the other hand, the plaintiff urges that, whereas it may be conceded that Johnston sought a machine and process which would bring the edges of the tube in accurate alignment, nevertheless an intentional deviation from the tubing of Johnston, predicated upon an endeavor to avoid infringement, and with no attendant beneficial result, will not avoid infringement, when in all substantial respects all elements of the process are employed in substantially the same way, producing the same result.
It is true, as defendant contends, that Johnston does stress accurate registry. In his process patent he says, in ascribing reasons for failures in the art of butt welding tubing electrically, among other things: "Improper compressing, whereby portions of the edges were forced into lap relation."
He continues: "I finally discovered that successful results are obtained by so applying and controlling the current that the extreme edges only — practically the edge surfaces only — of the thin stock are brought to a temperature approximating the fusing point (and possibly reaching that point) while the contiguous portions are not heated high enough to render them soft and mushy."
Again: "If the temperature be raised too slowly or maintained at high degree too long the edges of the blank are heated and softened for a considerable distance laterally from each edge, thus tending to produce detrimental sloughing and overlapping of the thus softened edges. * * *"
And: "It being found that if the edges are softened for a substantial distance back into the body of the blank the upsetting will be distributed laterally and the liability of one edge buckling out of register with the other and lapping instead of properly upsetting, greatly increased."
The inventor also speaks of the synchronization of the roller electrodes, "so that there cannot be any considerable bodily movement of one roller relatively to the other, either due to working stresses or to warping under the heat of the current; and the electrode rollers being maintained in substantially accurate and continuous register with the seam-cleft."
Insistence likewise is urged by Johnston "that the forming rolls which transform the flat steel ribbon into the tubular blank be so organized that the edges of the blank are made to register accurately and are substantially without undulations or irregularities when they reach and pass through the welding throat."
Another passage: "Describing the welding rolls, it will be observed that they are of twin construction, co-axially arranged, and bolted together so as to prevent relative movement, either axially or radially, the two rolls practically operating as a single roll divided down the middle so as to prevent short circuiting of the welding current."
On the other hand, whatever may be the references in the specification to the desirability of accurate registry of the seam edges, I do not find that any of the claims in suit is thus limited; and I think the explanation is that Johnston was seeking an improvement in continuous tube welding specifically of the Parpart process.
In Elyria Iron Steel Co. v. Mohegan Tube Co. (C.C.A.) 7 F.2d 827, 829, Judge Hough referred to the limitations of the Parpart machine and process: "The Parpart machine made tube with so much waste, made it so slowly, and usually left the weakest part of the tube, not in the weld, but alongside of it."
To meet the failures of the Parpart process, Johnston urged the co-ordination or correlation of stock-feeding rate, current control, and pressure application. It seems, therefore, that, when Johnston referred to the fact that "portions of the edges" were forced into lateral relation, in describing a defect of the prior art, he had in mind the improper pressures of the Parpart process. The Parpart process of compression of the stock was among those causes that prevented the operators of the Parpart machine from producing tube commercially at more than about twenty feet per minute.
If defendant, therefore, had reverted to the Parpart apparatus and process, and adopted the same means of compression, obviously defendant would avoid infringement. There is no such reversion, however, by the defendant to the Parpart means of compression of the stock.
Nor should too much emphasis be placed either on the term "butt welded," as used in the Johnston specification. The welding art knew roughly lap welding, lock welding, and butt welding. I think it reasonable to infer that Johnston used the term "butt welding" to distinguish it from the other two welding processes. I understand lap welding is in effect the complete overlapping of one member of the product to be welded over the other. Any process in which the edges substantially abut each other, in my judgment, under the Johnston classification, would be an example of butt welding.
Now, coming to the defendant's apparatus and process, we find a forming machine connected to the welding unit, and, after the welding unit, sizing rolls. The strip is fed to the forming machine and passes through the forming rolls until within about two inches of the center line of the copper electrodes. As part of the welding machine, the defendant employs a mandrel or plug over which the tube progresses. At about two inches in advance of the center line of the copper electrode, there is mounted on this mandrel or carriage a roller which presses up on the open seam of the tube. At that point the open seam is approximately 1/16 to 3/32 of an inch.
As one looks in the direction of the movement of the tube, the right hand electrode is adjusted lower than the left hand electrode by approximately the thickness of the metal sheet of which the tube is formed. This vertical adjustment of the rolls, together with the shape of the welding throat causes the opening at the top of the supporting roll to converge to a V at a point approximately three-quarters of an inch in advance of the center line of the electrodes.
At this point of convergence of the V, the right side of the tube is lower than the left side, and thus the tube passes to the welding point. The tube as thus welded passes along a distance of about seven inches from the center line of the copper electrodes, at which point the shoulders of the seam are crushed down by internal and external crushing rolls. Thence the tube passes through the final sizing and straightening rolls.
The machine used by the defendant was purchased in April, 1924, from the Thompson Lock Welding Company. It was operated during that year and at various times thereafter up to the early part of 1928. Operations were then discontinued, but were resumed in October, 1929. Plaintiff's Exhibit No. 5 represents the kind of tube made on the machine prior to 1928.
After purchasing the machine, defendant made certain changes in it. At first defendant found the plug or carriage too close to the copper electrodes, and therefore increased the diameter of the plug from about one and a quarter to one and a half inches. Some other minor changes were made with the plug and the rollers. The copper rolls themselves were modified. At first they circumscribed about one-half the circumference of the tube to be welded. Thereafter the defendant reduced its copper rolls so as to cover about one-third. Also a flat spot was put on the top of the rolls of about one-eighth of an inch, so as to force one edge of the tube under the other before welding. These changes were made during the latter part of 1926 and the early part of 1927.
Mr. Reed, the general manager of the defendant company, testified as to some other changes. He said when the machine was delivered the tube was very nearly circumscribed. The defendant then separated the supporting rolls "a little, in accordance with the height, and also put a flat spot on those." Also the lubricating system was defective. The witness said that the electrodes would stick and would "continually duck the edges of the tube under."
Hence the question becomes this: Did these changes in the machine evidence such a departure from the teaching of Johnston as to enable the defendant to avoid infringement of the Johnston patents? That the changes made indicated a defective machine is clear, or rather it is clear that, as delivered, the Thomson machine was imperfect and perhaps caused a larger scrap loss than sound commercial practice justified.
But it is not clear that this commercial failure should be attributed to the fact that the electrodes were on the same horizontal plan. Rather it appears to me that what the defendant did with the Thomson machine was to make those adjustments which are so frequently necessary in machine operation to bring about the best results. Lowering the right-hand electrode slightly — a matter of a few thousandths of an inch — can hardly be said to be a radical departure from the teaching of the Johnston process patent. Certainly it was not a reversion to Parpart, and, since the defendant's practice shows that correlation of time or speed of movement, heat both as to kind to be applied and area to be heated, and contact of butts as to pressure, which the Court of Appeals, in the case of Elyria Iron Steel Company v. Mohegan Tube Company, said were, in addition to accuracy of butt contact, the essential and novel thoughts involved in Johnston's theory, I believe the process claims are infringed.
To state that there is accuracy of butt contact effected by the defendant in its present operation of the Thomson machine, despite the fact that the lowering of one electrode does prevent a perfect abutting relationship, does seem to involve a contradiction; but in point of fact there is no such contradiction, for, within the fair scope of the Johnston invention, any process embodying all of the Johnston teachings but which does not attain the perfection which the inventor sought as his ideal, should not escape the charge of infringement.
As bearing on the scope of the invention, it is interesting to read, in an affidavit of Prof. Elihu Thomson filed in this proceeding, that he states: "In lap welding there is no burr or anything that corresponds to a burr, and the plates are simply squashed down one upon the other, so that the thickness at the joint is less than the combined thickness of the two layers of metal."
Applying the test of that definition to the defendant's product, we find that defendant's tubes, as they passed the welding point, certainly show a burr, and moreover, after the de-burring operation, the thickness at the joint is not less than the combined thickness of the two layers of metal. So it must follow that Prof. Thomson himself would be compelled to describe the defendant's process, not as lap welding, but rather as butt welding, for he adds: "In butt welding there is the characteristic extrusion of metal into a burr," etc.
Moreover, effort to avoid infringement of these process claims on the ground that there was not accurate registry of the edges of the tube was urged also in the Mohegan Case, but did not save the defendant from infringement.
The defendant's product, when one examines Plaintiff's Exhibits 5 or 6, 104, 105, 117, is butt welded; the metal has been heated back but a small distance from the joint, the burr is very small, the recurrent stitch effect is visible, in most cases even to the naked eye, and the thickness of the tube wall is practically that of the strip from which the tube is formed. So much is this so that Mr. Ober, defendant's expert witness, was unable to distinguish the tubing as having been made on defendant's or plaintiff's machine.
Hence I conclude that the product claims are likewise infringed.
Belmont patent No. 1,611,875. The object of this invention was to provide an apparatus for "smoothing out the weld that will be adapted for use with the so-called continuous welding machines such as are commonly used for welding the longitudinal seam of tubing formed from strip metal."
The inventor describes a device capable of operating on the inside as well as on the outside of the tube, immediately after the welding operation and while the metal is still hot and comparatively soft.
This device consists in the main, in combination with the welding apparatus, of a carriage or mandrel inside of the tube to be welded; the mandrel being suitably supported and having rolls for engaging the surface of the tube opposite the weld, supporting rolls for the tube opposed to the former rolls, a roller on the carriage for operating on the weld so as to smooth down the interior of the burr of the tube, and a roller on the exterior of the tube opposed to the interior roller, which operates on the exterior burr of the tube.
It seems to be conceded that, if the Belmont patent is valid, it is infringed.
Claim 2, which may be considered typical, reads: "2. In apparatus of the class described, the combination of means for heating the longitudinal edges of a tube to be welded, means for forcing said edges together to produce a weld, a carriage arranged inside the tube and having rotatable supporting means engaging the surface of the tube opposite the weld, supporting means for the tube opposed to said rotatable supporting means, a roller on said carriage for operating on the weld to smooth down the interior burr, a roll on the exterior of the tube opposed to said roller and operating on the weld to smooth down the exterior burr, and means for holding said carriage in the desired position."
In this claim, "in an apparatus of the class described, the combination of means for heating the longitudinal edges of a tube to be welded" refers to the rotary electrodes; "means for forcing said edges together to produce a weld" are the electrode and rollers below them, the electrode and rollers together forming the throat through which the tube passes; "a carriage arranged inside the tube and having rotatable supporting means engaging the surface of the tube opposite the weld" refers to the mandrel and the lower roller or rollers of the mandrel; "supporting means for the tube opposed to said rotatable supporting means" refers to the external rollers bearing up against the bottom; "a roller on said carriage for operating on the weld to smooth down the interior burr" refers to the upper roller of the mandrel; "a roll on the exterior of the tube opposed to said roller and operating on the weld to smooth down the exterior burr" is, of course, exterior to the mandrel and tube; "and means for holding said carriage in the desired position" are means shown whereby the long mandrel is held in adjustable position with reference to the particular point at which the rollers of the carriage are to engage with the external and internal rollers that bear on the tube.
The defendant contends that the Belmont patent is to be limited to butt welding, and that it was old in the art to roll down the burr in the butt welding process. It is further urged by the defendant that it practices a lap welded process in which no burr is extruded, and that in consequence the rolling down apparatus which it uses is for the purpose of rolling down the edges of the tube while hot and forging them, as had been done in the old art.
Since I have expressed the opinion that the process practiced by the defendant is not lap welding but essentially butt welding, and also that the defendant's tube passing the welding point has a burr, I shall consider in connection with the Belmont patent only the question as to what the prior art shows and whether invention was involved.
The defendant places considerable reliance on patent No. 324,610, issued to Strong. This patent was cited by the Patent Office on the Belmont application. It relates to "a mechanism for use in welding longitudinal seams of heavy iron tubes (particularly such tubes as are designed for steam and water boilers and tanks * * *) and which as now commonly constructed have their seams closed by riveting."
A sheet of iron, after having been bent to the form of a tube with its edges lapped, and after having these lapped edges heated, is placed around a horn or horizontal support. The tube is then grasped by a pair of clamps which are operated by steam or hydraulic power and are arranged to press downward upon the tube at each side of the seam to hold the lapped edges securely in position. Then the lapped and heated edges are acted upon by one or more rolls which are mounted in a suitable head and which are driven backward and forward so as to press and weld the edges of the sheet together. I do not see how this apparatus could be used for continuous rolling operations, and I find nothing in it to suggest the Belmont invention.
The Webster patent, No. 950,163, discloses a metal tube forming machine. The inventor sought to make metal tubes rapidly without marring the metal and with a strong but nevertheless inconspicuous joint. The joint disclosed is a locked seam. The interlocking of the flanges and the closing of the seam formed by them is accomplished by means of "the mandrel, G, terminates between the compression rollers, K, K¹, in a tapered enlargement, g, of a diameter equal to the internal diameter of the tube to be formed. In vertical alinement with the points at which the compression rollers, K, K¹ contact the tubing, mandrel, G, has a vertical bore, g¹, within which friction rollers, g², g³, are journaled upon transverse pins, g 4, g 5. The rollers, g², g³, are shaped to conform to the contour of the inside of the tube. Between the roller, g², and the compression wheel, K, the seam formed by the interlocking flanges, a, a¹, is flattened and compressed. The rollers prevent the metal's being drawn, or marred, by the action of the mandrel and the compression rollers. The tube is then guided and fed to the end of the machine by sets of rollers, M, and N."
The mandrel or the outside roller must always be provided with a groove. If the rib is formed on the exterior of the tube, the groove is formed in the outside roller. If the rib is formed on the interior surface of the tube, the mandrel must be grooved to receive the rib. Thus it follows that the mandrel and outside rollers of the Webster patent could not be utilized in the Belmont device.
Finally, it may be stated that there are no means shown in the Webster patent for heating or welding. As I read it, it is the closest of the prior art patents to the Belmont device, but I do not find that any of the Belmont claims in suit reads upon the disclosure of the Webster patent.
British patent No. 14,290 of 1885 to Robertson. In this apparatus, as in the Strong construction, the rolls are actuated by being rolled over the article to be drawn out. While structurally it may be conceded that Robertson shows a guide and rolls, there is no suggestion of the specific object that Belmont had in mind, to wit, of rolling down a burr on the interior and exterior of a continuous moving tube.
Hawes, United States patent No. 1,209,704. This patent discloses a method and apparatus for electric welding, but relates particularly to the manufacture of tubes of comparatively short length. In addition to other differences which patently exist between the Hawes construction and that of Belmont, it is seen that Hawes uses a hammering device. He says: "To further assist in thus smoothing down the freshly welded joint on the article, and also to insure the thoroughness of the weld itself, a hammering device is provided, such device being disposed closely adjacent to the inner face of the electrode, as shown in Fig. 2."
Moreover, in Hawes, the article is welded by means of a moving electrode. It is hard to find any element of similarity between Hawes and Belmont.
Taylor patent, No. 1,511,849, describes a method of welding tubing in which the edges are overlapped. In this process, welding is performed by advancing a welding roll along the outer part of the lap joint. The inventor states: "After the completion of the lap weld 11, the tube 5 is drawn between two rolls 12 and 13, as shown in Fig. 4."
So that not only is there no continuous operation — no unified process of welding and rolling — but there is no such combination of rollers and mandrel or plug shown, as is disclosed in the Belmont patent.
In addition to these prior patents, the defendant offered testimony as to a prior use by the Empire Art Metal Company. The difficulty with this use is that those who worked in the shop and who are supposed to be familiar with the operation of the machine described contradict one another in respect to the accuracy of the drawing offered by the defendant as illustrating the machine which it is claimed was operated by that company. Certainly it is not clear, beyond a reasonable doubt, that the mandrel and rod supporting means, as shown on the drawing, were embodied in the machine itself. Because of the failure of the defendant to sustain the heavy burden of proof which rests upon him who seeks to establish a prior use, I am obliged to find that the Empire Art Metal Company's use fails for lack of adequate proof.
Plaintiff may have a decree.