This is an appeal from a judgment of the District Court of Kansas, holding invalid for lack of invention and for anticipation, appellant's Patent 2,211,173 for improvement in pipe coupling, referred to as a tool joint. In so holding, the trial court did not reach or decide appellant's claim that appellee Hughes Tool Company's 'Counter Bore Weld Tool Joints' infringed his patent. On appeal, our first consideration is validity of the patent in view of the prior art. Or, otherwise stated, whether the trial court's judgment is clearly erroneous.

The tool joints involved here are used for bringing thirty-foot sections of drill pipe into threaded engagement to form a 'drill string' or 'drill stem' in the drilling of oil wells. When thus assembled on a rotary drilling rig, the drill stem serves as a drive shaft for turning a rotary bit and as a conduit to circulate 'drilling mud' under pressure to the rotary bit at the bottom of the hole and back on the outside of the drill stem to the surface. When the rotary bit becomes worn, it is brought to the surface by pulling and breaking the drill pipe connections at every second or third joint. When a new bit is attached to the drill collar on the last section, it is returned by remaking the connections as the pipe is extended to the bottom of the hole.

The tool joints are ordinarily composed of two members, adapted to be threaded together by a coarse threaded connection. The opposite ends of the two members are internally tapered and threaded to be brought into threaded engagement with complementary externally tapered and threaded pipe to form a 'bucked-on' threaded union of the drill pipe section and the tool joint member. The tool joint members customarily extend or overlay the drill pipe beyond the area of the threaded engagement to reinforce the tool joints against the expansive stress set up therein when the adjacent ends of the drill pipe are tightly screwed into them. The overlap or extension is usually of greater internal diameter than the external diameter of the drill pipe immediately adjacent the threaded area, so that the only engagement of the tool point and the pipe is along the threaded area. As a consequence, the expansive stress against the tool joint and the complementary compressive stress against the drill pipe abruptly terminate at the last threaded engagement, causing a concentration of the stress at the notch formed by the last engaged thread on the drill pipe. The twisting, bending, flexing and vibration of the pipe caused by rotating many hundred and sometimes several thousand feet of drill pipe in a hole larger than the pipe, many times results in costly failure or fracture of the drill stem at the last threaded engagement of the pipe and tool joint.

This problem has been the subject of considerable discussion in the industry and has provoked repeated efforts at solution, with varying degrees of success. As early as 1928, the industry was applying conventional or multiple bead metal weld on the drill pipe immediately adjacent the distal end of the tool joint. This metal band or weld prevented most washouts, but the space between the pipe and the tool joint where it overlapped the pipe, was not wide enough to allow the penetration of the weld to or near enough the last threaded engagement to relieve the concentration of the compressive stress at that point. And, although last thread failures were reduced, the problem was not solved. Moreover, the application of the multiple welded bead on the pipe tended to crystalize it and cause breakage at that point.

The Reed Roller Bit Company devised what is called its 'Shrink-Grip' tool joint. It is only vaguely described in the record, but apparently embodies the principle of shrinking the tool joint shoulder over the pipe to reinforce it beyond the threaded engagement. The Pittsburgh special drill pipe, also directed to the last thread failure, provided a precisioned machined drill pipe and tool joint shoulder, designed to fit tightly together from the threaded area to the end of the tool joint shoulder, giving rigid support to the drill pipe immediately adjacent the last thread. While these devices tended to solve the problem, and enjoyed commercial success, they were said to be highly expensive and impractical.

It was in this setting and in the early part of 1936 that patentee Shaffer, an experienced machinist and rotary driller, conceived the idea, later embodied in his patent, of counterboring the unthreaded overlapping shoulder of the tool joint down to the last threaded engagement to provide a suitable space into which the metal weld would penetrate, reinforcing the last threaded engagement, and making the pipe and tool joint integral from the last threaded engagement to the end of the tool joint, thus eliminating the fracture-causing concentrated stress at the point of the last engaged thread.

About that time, the patentee prepared a paper on 'Improved Conventional Type Tool Joint for All Makes of API Drill Pipe,' explaining his proposed improvement. He attached a drawing depicting his conception, and comparing it with the conventional or multiple beaded weld. Another paper termed 'Proof of Conception' (later lost and not produced in the trial or on the record), was prepared by the patentee to be signed by all parties to whom he should explain his improvement. In the Spring of 1936, he exhibited his conception and explained it to two representatives of the Hughes Tool Company, who signed his proof of conception and offered to make the tool joints in accordance with his specifications. Sometime thereafter, he caused several tool joints to be counterbored and welded in accordance with his specifications, and tested them for about a year on the end of a drill stem, experiencing no failures.

In February of 1938, Shaffer and the Reed Roller Bit Company entered into a contract, by the terms of which Reed was granted a nonexclusive license to the appellant's improvement, and began the manufacture and sale of the improved tool joint. On June 6, 1938 Shaffer applied for a patent. After prolonged proceedings in the patent office, during which a total of seventeen claims were urged, and after three complete rejections, the patent was granted with three claims, the first of which is typical and fairly illustrative: 'In a pipe joint comprising complementary tapered pin and box members in tight threaded engagement with each other, whereby the box member exerts a compressive stress on said pin member along the zone of such threaded engagement, a ring of weld metal surrounding and joined to said pin and box members and extending from the end of said zone of threaded engagement, said ring tightly engaging said pin member to exert a compressive stress thereon for preventing the localization of vibration at the end thread beginning at the end of said zone of said threaded engagement, and with said box member to produce a continuous zone of compressive stress in said pin member bridging the end of said threaded engagement and avoiding a too abrupt change in compressive stress on said pin member at the end of said threaded engagement.'

Hughes Tool Company commenced the manufacture and sale of the accused device in the latter part of 1937, describing it as 'Hughes Counter Bore Weld Tool Joint,' and offering it as a 'practical solution to the problem of the failures of the pipe at the last engaged thread.' About the same time, it applied for a patent on the device, in which it called attention to last thread failures, and proposed to avoid the concentrated compressive stress at the last threaded engagement by laying a bond of weld material in a 'channel between the tool joint end and the adjacent wall of the pipe, by cutting a flaring bevel on the end of the tool joint member.'

While the description and claims are textually different, they describe substantially the same principle disclosed in the patent in suit. The drawings describe the same mechanical arrangement, with only a slightly differently shaped counterbore. The application was finally rejected April 13, 1939, and Hughes has continued to manufacture and sell the accused device, with knowledge of the patentee's claim of infringement.

The patent in suit was granted and the claims allowed over patents to Mixer, Spencer, Brantly, and Dillon. In considering patentability, the trial court described and treated these patents, as well as all others cited in the prior art. It pointed out that the patents to Mixer (220,407- 1879), Spencer (British 280,644- 1926), Kane (1,966,248- 1934), and Dunn (2,206,166- 1940) were all in the art of constructing pipe joints designed to conduct and hold fluids or gases, and while they disclosed various forms of sealing and welding threaded joints, they did not recognize or deal with the problem of a tool joint designed to act as a drive shaft for the transmission of power, as well as a conduit for fluids, such as a tool joint involved in the patent in suit.

The patent to Brantly (2,073,093- 1937) related to a tool joint somewhat similar to the one in suit. But, as the trial court suggested, the sole object of Brantly's teachings was to 'eliminate the notch effect in the threaded area of the drill pipe by employing curved or sine wave threads rather than the customary V-shaped thread.' Conventional type welding on the pipe adjacent the end of the tool joint was optional, and did not purport to fill the thread relief space normally provided.

Dillon (2,087,185- 1937), also relating to tool joints similar to the one in suit, shows the sealing of the joint against washout by the use of a metal packing gasket and the mechanical introduction of low melting point metal between the threads of the tool joint and pipe, forming a seal or bond to prevent leakage through the threads and vibration or 'creeping in the thread joints.' It provided for a conventional weld at the ends of the joints to lock the pipe sections to the joints, if desired.

The patent to Flagg (2,145,168- 1939) relating to similar tool joints, disclosed the threaded surfaces of the drill pipe and tool joint bonded together by heating and coating them with a low melting point metal, and engaging the threads while the metal was still in molten form. No actual weld or counterbore was contemplated or claimed.

The court treated Brantly, Dillon and Flagg as strictly within the prior art, with the appropriate comment, however, to the effect that none of them disclosed a metal weld in a special counterbore to relieve the compressive stress at the last engaged thread of the joint. The court went on to point out that Flagg and Dillion disclosed 'primarily a sealing of joints against washouts by leading or a similar process,' while Brantly proposed to prevent last thread failures by a combination of sine wave threads and the conventional weld. The court could not read any of these patents upon the one in suit.

Of the other four patents not strictly within the prior art, the court considered Spencer and Kane as disclosing a weld laid around a simple bevel or undercut shoulder, which did not teach a counterbore consciously designed to extend substantially to the last engaged thread for the purpose of strengthening that area.

The court recognized the Dunn Patent as teaching a weld laid in a gap between two raised pipe shoulders, and Mixer as disclosing a simple soldered or leaded sleeve joint, but observed that neither of these patents dealt with the problem with which the patent in suit was designed to cope.

When, however, the trial court considered all of the teachings of the prior art in their totality and in their proper relationship to the claims of the patent in suit, its studied judgment was that Shaffer 'merely applied mechanical skill when he caused tool joints, being used by him in the oil fields, to be counterbored so that a weld might be laid in the fashion shown by this record and in this alleged patent.'

The court also found anticipation in the aggregate of the relevant elements of the prior art in view of the revelations in Reed's 'Shrink-Grip' and Pittsburgh's precisioned machined tool joints.

The appellant patentee finds no fault with the trial court's description of the prior art. He does, however, earnestly contend that the court erred in not according invention to his device over the prior art. Our attention is called to the costly and baffling problem of last thread breakage and the diligent, but unsatisfactory efforts of those schooled in the art to devise a mechanically and economically acceptable remedy. In that connection, it is said that none of the cited prior art purports to teach the critical concept of laying a bond of welded material at or sufficiently close to the last engaged thread to reinforce it against the critically destructive compressive stress. This idea, the patentee insists, is a new combination of old elements, whereby a new and useful result is obtained, amounting to invention, for which he is entitled to a monopoly. See Harris v. National Machine Works, Inc., 10 Cir., 171 F.2d 85. And then, we are reminded of the ready acceptance of the patent in suit on the market with its attendant commercial success. As proof of its practical utility and public acceptance, it is shown that Hughes almost immediately adopted and exploited the idea in the accused device, and that it also met with commercial success.

It would seem that a patent for an improvement which fills a long felt need, and which those schooled in the art have not been able to devise, and which meets with acceptance in the market, ought to have the benefit of the doubt when challenged by those who now seek to pirate the advancement, and excuse it on the ground that any ordinary mechanic should have thought of it. Cf. National Slug Rejectors, Inc., v. A.B.T. Mfg. Corp., 7 Cir., 164 F.2d 333; Russell v. J. P. Seeburg Corp., 7 Cir., 123 F.2d 509. But the contract we construe is between the patentee and the public, not the patentee and Hughes, and our first inquiry is whether the patentee's conception is a 'substantial innovation' for which society is truly indebted to the patentee. Sinclair & Carroll Co., Inc. v. Interchemical Corp., 325 U.S. 327, 65 S. Ct. 1143, 89 L. Ed. 1644.

Commercial success from market demand is an objective test which sometimes tips the scales in favor of patentability in a doubtful case. Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275, 64 S. Ct. 593, 88 L. Ed. 721. But no amount of commercial success can bridge the gap between mechanical skill and invention. Dow Chemical Co. v. Halliburton Co., 324 U.S. 320, 330, 65 S. Ct. 647, 89 L. Ed. 973; Jungersen v. Ostby & Barton Co., 335 U.S. 560, 567, 69 S. Ct. 269, 93 L. Ed. 235.

'The design of the patent laws is to reward those who make some substantial discovery or invention, which adds to our knowledge and makes a step in advance in the useful arts.' Atlantic Works v. Brady, 107 U.S. 192, 200, 2 S. Ct. 225, 231, 27 L. Ed. 438. The device must not only be 'new and useful,' it must also amount to 'invention or discovery,' Thompson v. Boisselier, 114 U.S. 1, 11, 5 S. Ct. 1042, 1047, 29 L. Ed. 76; and 'perfection of workmanship, however much it may increase the convenience, extend the use, or diminish expense, is not patentable.' Reckendorfer v. Faber, 92 U.S. 347, 356-357, 23 L. Ed. 719; Cuno Engineering Corp. v. Automatic Devices Corp., 314 U.S. 84, 91, 62 S. Ct. 37, 86 L. Ed. 58. And to make sure that a monopoly is not granted for mere perfection of workmanship, the courts closely scrutinize claims to combinations for improvements in a crowded art. Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1, 10, 67 S. Ct. 6, 91 L. Ed. 3. Hollywood-Maxwell Co. v. Streets of Tulsa, 10 Cir., 183 F.2d 261.

The prior art, and apparently the whole industry, diagnosed the cause of last thread breakage, and recognized the efficacy of getting the weld to the threaded area. This is evidenced by the conventional weld fully illustrated by the patentee as belonging to the prior art. The only advance in the prior art is the patentee's idea of counterboring the tool joint shoulder so that the metal weld could be laid to the threaded area, making the joint integral.

We agree with the trial court that this does not amount to invention. Having decided that the patent lacks invention, we have no occasion to decide the subsidiary questions of anticipation or infringement.

The judgment is affirmed.