Full opinion text
LEVIN, District Judge (sitting by designation). This infringement action by The Cold Metal Process Company against Republic Steel Corporation was commenced on August 14, 1942 and involves United States Letters Patent Nos. 1,744,016 and 1,779,-195 (hereinafter respectively referred to as ’016 and ’195), both of which patents have expired since the filing of this suit. Cold Metal is an Ohio corporation and Republic is a New Jersey corporation doing business in Ohio. The alleged acts of infringement are said to have been committed within the Northern District of Ohio. On December 28, 1945, subsequent to the commencement of this suit, Cold Metal, owner of the patents, assigned them to The Union National Bank of Youngstown, Ohio, as Trustee, which then joined as a party plaintiff to the action. ' • Republic has interposed numerous defenses asserting (a) the invalidity of the patents, for- lack of novelty and invention as well as for other reasons, and (b) the absence of infringement, both because the accused mills do not practice the art disclosed in the patents and because Republic enjoys an immunity under a certain license agreement between Cold Metal and the party from .whom Republic purchased the mills, United Engineering & Foundry Company. After disposing of numerous preliminary motions, an order of reference was made to William L. West, Esq., of Cleve- .¿and, Ohio, as special master, to receive testimony and other evidence and for an advisory report only, the scope of the inquiry on reference being restricted by an order pursuant to plaintiffs’ motion to such of the accused mills as are hereinafter discussed. s During the proceedings before the special master a daily copy of the transcript was submitted to me which indicated that the parties received a fair and complete hearing. I was also in receipt of copies of the briefs submitted to the master, and I have before me his comprehensive and helpful report. A personal inspection of the accused mills and of certain prior art mills was made in the company of counsel. At the completion of this inspection, several days of oral argument were had on the objections of the two sides to the findings of the special master, and I now conclude that, except for certain claims which I shall hereinafter discuss, the plaintiffs must prevail on the issue of validity of the two patents. As to the issue of infringement, I find that certain of the accused mills, as will later appear in detail, came within the scope of the valid claims of the disputed patents, but plaintiffs are precluded from recovery by virtue of the license agreement. On January 14, 1930, ’016 issued to Cold Metal as assignee of an application, Serial No. 648,761, which had been filed on June 30, 1923 by Abram P. Steckel. ’195 issued on October 21, 1930 to Cold Metal as assignee of another of Steckel’s applications, Serial No. 412,742 filed on December 9, 1929. Although the ’195 application was filed considerably after the application upon which ’016 issued, it is said to have been a division of said application and is claimed to be entitled to the same filing date. Both of these applications disclosed a mill for rolling at high speeds thin metal in strip or sheet form and commercial tin plate. Thin sheet metal is used for a variety of purposes: automobile bodies, fenders, hoods, refrigerators, metal furniture, utensils, etc. It is characterized by a high ratio of width to thickness and has a high surface finish. In the process of rolling steel, a short, thick bar of the metal is successively passed between rollers for flattening, smoothening, and elongating into a thin sheet or strip. The difficulties of rolling steel into thin sheets or strips in this fashion increase as the reduction in thickness progresses, and they become so pronounced when the ratio of width to thickness' is in the neighborhood of 400 to 1 that this proportion is accepted as the dividing line between “ordinary” sheets and “thin” sheets or strips. Such thin strips range in thickness from .1379" to .0123". Tin plate, so-called, is sheet metal with an especially high ratio of width to thickness, generally being rolled to a thickness of between .010" and .0079". It is then surfaced with tin and used in the manufacture of cans and similar products. Prior to the 1923 disclosures of Steckel, there were three common techniques for producing sheet or strip metal: hot pack rolling, hot strip rolling and cold strip rolling. Hot pack rolling was the least efficient of the three methods and is described at length in two previous cases: Cold Metal Process Co. v. United Engineering & Foundry Co., D.C., 3 F.Supp. 120, 121-122 and Cold Metal Process Co. v. Carnegie-Illinois Steel Corp., 3 Cir., 108 F.2d 322, 324-328. Quite briefly, in this process a workman would feed by hand a hot bar of steel between the rolls of a 2-high mill, that is, a mill stand in which one roll is arranged vertically over the other. The steel bar was then picked up manually by a workman on the other side of the mill stand and handed back to the first workman over the top of the mill so that it might be passed through again in the same manner. Frequent reheating was required, after which two pieces would be placed on top of each other to be passed through the rolls simultaneously. The plies were thus increased as the rolling and reheating continued until the metal was of the desired thickness. It was impossible to get satisfactory accuracy of gauge by such a technique; scrap losses were very • high; and the size of the finished product was limited by the capacity of the men to handle excessive weights. It was a slow and laborious process, totally inappropriate to the large industrial demand then existing for thin metal sheets. Hot strip rolling was also carried out on 2-high mill stands. The reduction of the metal was carried out usually on a continuous mill wherein several mill stands were arranged in a row, one after the other, so that the piece being rolled would pass through the successive mill stands at a high speed, being in two or more stands simultaneously during the course of its progress through the mill. The motive power was applied to the necks of the rolls to move the bar through the mill series, and, as the strip issued from the last mill stand, it was gathered into a coil. The best reduction that could be achieved on such an operation was to a ratio of 200 to 1, which did not satisfy the requirements for thin metal. In cold strip rolling the metal is fed into the rolls at room temperature, and extreme pressures are applied to the metal through the rolls. The principal power for moving the cold strip through the rolls was likewise furnished by motive power connected to the necks of the rolls, and the strip was gathered into a coil as it issued from the roll. However, in some of these cold rolling mills, a motor was connected to the reel around which the strip was wound when it came out of the mill, and a tension was maintained on the strip sufficient to keep it taut and straight. By repeating this operation the steel can be reduced to the requisite thinner gauges, and a highly finished surface on the metal can be obtained, such as was not possible in the hot strip rolling process. However, because of the magnitude of the pressures required to reduce the metal in cold strip rolling, it was not feasible to use roller-type bearings on the necks of the roll bodies because there was insufficient space between the roll necks of the two roll bodies to accommodate roller-type bearings of adequate size and sturdiness to withstand such pressures. In the 2-high mills which were in general use it was necessary that the diameter of the roll necks be approximately two-thirds of the diameter of ’ the roll bodies to prevent the necks from being broken off under the heavy loads characteristic of this type of rolling process. The only space that remained for fitting-in an adequate roller bearing was between the periphery of the roll neck and the periphery of the roll body, since a bearing assembly that •extended beyond the periphery of the roll body would interfere with the bearing assembly on the companion roll body. Thus, plain brass or babbit bearings were used, causing the roll necks to overheat. As the pressure was increased, the oil film used for lubricating these bearings would be broken, making the frictional resistances even higher and more variable. This frictional heat at the roll necks would spread to the roll bodies, causing uncontrollable expansions and producing buckling and riffling in the metal strip, destroying its commercial value, and resulting in frequent scrapping of the entire strip. These cold rolling mills had to be operated at low speeds to minimize the distortions caused by this heating up of the roll necks and the roll bodies, 100 to 165 feet per minute being the maximum speeds for the cold rolling of thin sheets. Furthermore, the metal became harder and less ductile in proportion to the amount of reduction that had been made. This tendency made it necessary to remove the strip and to subject it to annealing in the middle of a rolling so as to permit the grain structure to be reformed. This ' combination of slow speeds, high scrap losses, and the necessity for intermediate annealing accounted for the high cost of producing thin metal strip in the pre-Steckel cold rolling mills. Despite the inadequacy of these techniques for supplying the insistent demand for less expensive thin metal strip, until Steckel conceived his invention in 1922 and reduced it to practice shortly thereafter, the industry appeared to be helpless over a period of many years to overcome this technological block so as to be able to cold roll thin strip at high speeds, without intermediate annealing, with accuracy of gauge, and with a minimum of scrap loss through buckling and riffling. Essentially, Steckel’s conception consisted of the use of a 4-high mill, that is, a mill in which four rolls are arranged vertically one above the other. The metal is passed between the two interior rolls, which are relatively small compared to the rolls above and below them. In fact, the working rolls are so small that they would buckle and give way when subjected to the pressures necessary for reducing the metal unless supported by the heavier backing rolls. By using this arrangement of rolls, diminution in the necessary pressure on the metal is achieved and the frictional drag on the backing rolls is reduced so that it is possible to use roll necks on the backing rolls of smaller diameter than is possible in the 2-high mills. More space is thus provided between the periphery of the roll neck on the backing roll and the periphery of the backing roll to enable roller bearings of increased sturdiness to be attached to the backing roll necks so as to support the high pressures experienced in the operation of the mill. The use of such roller bearings in combination with these other elements of the mill results in the reduction of frictional drag on the backing rolls and permits the mills to be operated at speeds far in excess of those possible in the prior art mills, without experiencing the overheating of the backing rolls and the buckling and riffling of the metal strip, which was characteristic of the prior art mills when operated at such speeds. Steckel’s invention revolutionized the manufacture .of thin metal sheets or strips. His mills operated at speeds greatly in excess of the maximum speeds possible in the prior art mills. By the spring of 1923 he had succeeded in rolling material to tin plate gauge 6" wide and commercially flat at speeds of 900 feet per minute while it appears that speeds for the production of comparable material on the prior art mills ranged between only 100 and 165 feet per minute. Today these patented mills are capable of speeds far in excess of these early improvements. Intermediate annealings were eliminated entirely for the low and high carbon steels and for some alloys and were materially reduced in the rolling of other alloys. Thus, orders may now be rolled in a day which required several weeks to complete on the old mills. There is evidence that as early as 1930 the Steckel mills made possible savings in cost of $19.46 per ton on low carbon steel and as much as $145.00 per ton on stainless steel. Metal sheets can be produced which are 30 to 40% thinner and many times wider than was previously possible and can be made in coils, hundreds of feet long, as compared with former maximum lengths of 8 to 13 feet. This dramatic advance in the art coincided with an enormous rise in the demand for metal sheets, especially stimulated by the rapid increase in the activity of the automobile industry. After the customary aloofness which seems to greet most inventions, approximately 85% of the steel industry came, during the lives of the patents, to give tribute to Steckel’s invention, by accepting licenses and paying royalties. About $29;000,000 has been received under these patents, either as royalties or in settlement for past infringements. The essential combination of elements which comprises the mill itself is present in both patents. However, there is one conspicuous difference between them. In ’016 there is present, in addition to the concept of the mill itself, the concept of driving such mill “entirely,” “principally,” “largely” or “substantially” by tension on the metal strip as distinguished from applying the motive power solely to the necks of the working rolls, ’195 does not disclose this tension feature or otherwise reveal a technique for driving the mill, but both apparatus and method claims are addressed solely to the concept of the mill itself. Another less significant difference is that ’195 applies to both hot and cold rolling while ’016 is limited to a cold rolling process. Nothwithstanding these differences, all of the elements of '195 are inherently present in ’016 and, as to both patents, it will simplify our consideration of whether the prior art so anticipated or foreshadowed Steckel’s concept as to deprive it of the necessary qualities of novelty and invention if the issues of validity and infringement are discussed first with respect to ’195. In this manner it will not be necessary to reiterate so much of this discussion of the prior art as bears on those elements of ’195 which are equally present in ’016. ’195 '195 contains seventeen claims, all of which are here in suit. In his proposed findings, the special master concluded that all of these claims were invalid. Although I concur with his findings with respect to the five method claims, I must disagree with him as to the twelve apparatus claims. Reproduced here are claims Nos. 4, 7 and 15 as representative of apparatus claims Nos. 3, 4, 6, 7, 8 and 11 through 17: “4. A mill for rolling material of substantially uniform thickness, comprising working rolls provided with backing rolls of larger diameter, the backing rolls having necks of sufficient size to withstand the rolling pressure, anti-friction bearings for said backing roll necks, the diameter of the backing roll body relative to the diameter of the working roll being such as to permit of using anti-friction bearings of sufficient size to withstand the rolling pressure, said anti-friction bearings being of a character to withstand operation at speeds which are high relative to those ordinarily employed, a working roll having a neck extending between the anti-friction bearings of the backing rolls, and means for supplying rolling power through the working roll neck. “7. A rolling mill for material of substantially uniform thickness, the mill being of such width that the available space in a two-high mill of the same width would prevent the use of roller bearings of the necessary load carrying capacity, comprising large backing rolls with necks, intermediate working rolls, means for driving the working rolls, and anti-friction bearings for the backing rolls, the anti-friction bearings having outer races encircling the necks and being of such character as to withstand speeds which are high relative to those ordinarily employed, the intermediate working rolls being of sufficiently small diameter to reduce the spreading force in the mill to such point that the anti-friction bearings employed require a smaller space than those which would be required in a two-high mill of the same width, whereby the outer races of the bearings for the backing rolls are sufficiently spaced apart to allow room for the driving means. “15. A mill for rolling sheet or strip like material of substantially uniform thickness comprising a working roll of substantially uniform diameter in the body portion provided with a backing roll of larger diameter and anti-friction mounting for said backing roll, the backing roll being of sufficient diameter to permit of using an anti-friction mounting of a character to withstand the rolling pressures encountered and to operate for extended periods of time at speeds which are on the order of multiples of speeds commonly employed in the rolling of such' material.” It is conceded that the combination disclosed in these apparatus claims is made up of old elements. For example, the Blake & Johnson 3-high mill, which consisted of one small roll betwéen two larger backing rolls, the diameters of which were twice that of the smaller roll, was in commercial operation prior to 1922, and contained many' of the elements to be found in the Steckel mill. The diameters of the necks of the larger rolls were greater than the diameters of the bodies of the smaller rolls and about three-fourths the diameters of the bodies of the larger rolls and were mounted in plain bronze, or babbit bearings and lubricated with oil or grease. However, the speeds of these mills ranged from only 100 to 360 feet per minute in the.rolling of thin metal strip and they could achieve reductions of only 50% in high carbon and alloy steels and only 70 and 78% in low carbon steels without intermediate annealing. In addition to their inability to approach the performance of the Steckel mills, it is also clear that the Blake & Johnson mills do not foreshadow the Steckel disclosures since they lack at least one vital feature which is disclosed in ’195 — roller bearings on the necks of the larger rolls to provide an anti-friction mounting for the backing rolls. Although Steckel did not use the explicit term, “roller bearings,” in all of the apparatus claims, he did use two other expressions — “anti-friction bearings” and “anti-friction mounting” — which were used interchangeably with roller bearings in such a manner as to indicate that he meant “roller bearings” thereby. It appears that “anti-friction bearings” had the specific meaning in the art prior to and at the time of the filing of Steckel’s application of roller bearings. “Ahti-frictiori mounting,” on the other hand, was an expression of Steckel’s own devising. Where an expression is used which does not have an explicit meaning in the art but by which the inventor intends to convey an explicit meaning, it is permissible tó consult the specifications to ascertain how the inventor defines such expression. From these specifications it becomes clear that Steckel had in mind roller bearings on the necks of the backing rolls as the sole means of providing an anti-friction mounting for the backing rolls when he spoke of “anti-friction bearings” and “anti-friction mounting,” and the three phrases which he used— “roller bearings,” “anti-friction bearings” and “anti-friction mounting”— were all to have that one meaning. Therefore, absent such roller bearings in proper combination with the other elements of the mill, the Blake & Johnson mills did not anticipate or foreshadow Steckel’s .conception. Nor did the Keating (Bridgeport) mills which were installed about 1910 anticipate Steckel’s conception. They are called cluster mills because they contain smaller work rolls each of which is supported by two larger backing rolls, the mill consisting of six rolls in all. The two work rolls were 3" in diameter and 7" in length, and the backing rolls were 5" in diameter with roll necks 3" in diameter. The backing rolls were mounted on full-complement, pin or quill-type bearings. This type of bearing' consisted of slender cylindrical rods which were attached to the moving roll neck by a common case. As many rods as possible were fitted into the surrounding case, each rod being free within the case to move around and against one another. This type of mounting is to be distinguished from the caged roller bearing, in which each bearing is separated from another by its separate cage. The pin-type bearings were unsatisfactory in their performance. They often became dislocated in the case, jamming the bearing and resulting in bearing' failure, and the constant jostling against the other bearings caused flacking and breaking. The bearings would overheat and this Would be reflected in buckles and wrinkles in the strip. It was often necessary to replace individual bearings daily and it was the practice to change each backing roll, together with its bearing, every week. Even if such pin or quill-type bearings could be classified as “roller bearings,” they did not fulfill their function in the environment of a cluster mill as they were able to do in a mill which embodied Steckel’s combination. This is borne out by the poor performance record of the Keating mills. Their maximum speed was about 165 feet per minute, and they were able to make the extreme reductions only with difficulty, and even then intermediate annealing was required. In such cluster mills there was not enough space between the periphery of the roll necks and the periphery of the roll bodies for the fitting-in of roller bearings that were adequate to withstand the pressures and to carry the loads characteristic of such mills. ■ Having failed to use roller bearings in such combination with the other elements of the mill as to enable them to perform their role effectively, the Keating mills fell significantly short of Steckel’s invention. Defendant next objects to the apparatus claims of ’195. It asserts that they are indefinité and functional because they contain such language as: “sufficient size to withstand the rolling pressure,” “being of a character to withstand operation at speeds which are high,” and of “sufficient diameter to permit of using an anti-friction mounting of a character to withstand the rolling pressures encountered.” Although when these phrases are taken out of the context of the entire claim they may give that impression, if the claim is considered as a whole it clearly appears that Steckel explicitly identifies his invention — the combination of work rolls, backing rolls of larger diameter, and roller bearings, all proportioned in the manner conceived by Steckel to permit effective utilization of each element. With the insight thüs provided, the manner of translating it into a working model was a question of mechanics and would' be manifest to anyone skilled in the art. Indeed, that very information is to be found in the specifications where it properly belongs. However, I do hold that Republic’s objection to the five method claims of ’195, Nos. 1, 2, 5, 9 and 10 is well taken. Claim No. 1 is representative of such method claims: “In the method of rolling thin sheet or strip material in a mill having working rolls, and backing rolls, of' larger diameter, for the working rolls, the steps consisting in limiting the rise in temperature of the backing roll necks to such a point that variations in the contour of the backing rolls are controlled within such limits as to substantially prevent buckling or wrinkling of the thin material passed between the working rolls.” It is to be noted that Steckel does not specify roller bearings as an element of the combination to limit uncontrolled rises in temperature of the backing roll necks and the backing rolls. Instead, he makes a generic claim to any and all steps that may limit such rises in temperature, which is so broad as to include any process practiced in the prior art, such as flood cooling and specializing lubricants and lubricating systems, as well as any process that mechanical skill or inventive genius might discover for the future. Clearly, these method claims fail to point out particularly and to claim distinctly an identifiable invention; they overclaim the alleged invention; they do not properly define the limits of the sought-for monopoly. In these respects the method claims do not satisfy the statutory standards. 35 U. S.C.A. § 33; Graver Tank & Mfg. Co., Inc. v. Linde Air Products Co., 336 U.S. 271, 69 S.Ct. 535, 93 L.Ed. 672; Standard Oil Co. of California v. Tide Water Associated Oil Co., 3 Cir., 154 F.2d 579. They are not to be salvaged by an attempt to limit them to the methods disclosed in the specifications. The claims measure the invention, and when there is no ambiguity in their language, resort cannot be had to the specifications to restrict them. Unlike Steckel’s use of the expressions, “anti-friction bearings” and “anti-friction mounting,” to which he attributed a special meaning and in which case it was permissible to seek definition in the specifications, the method claims speak in plain language and offer no problem of understanding what Steckel intended. On September 23, 1932 Cold Metal filed a disclaimer in the Patent Office which eliminated from method claims 1, 2, 5, 9 and 10 the method of “limiting the rise in temperature of the backing roll necks by limiting the mill speeds to those employed prior to my invention for rolling such material.” It also disclaimed from the scope of the apparatus claims 6, 11, 12, 14 and 16 “any mill having backing roll bearings of a character to limit the mill speeds to those employed prior to my invention for rolling such material.” Presumably, Cold Metal believed that parts of certain of the patent claims were so broad as to include areas of the prior art. Under such circumstances, the patentee, his heirs or assigns is entitled to make a disclaimer of such parts, providing the original claims were made through inadvertence, accident, or mistake and that the disclaimer is filed without unreasonable delay or neglect. 35 U.S.C.A. §§ 65, 71. Republic contends that not only were these claims intentionally made broad and then disclaimed after unreasonable neglect and delay but that the effect of the disclaimer was to introduce a new element into the scope of the claims. That the allegedly overbroad claims were made inadvertently is demonstrated by a consideration of the essence of the Steckel conception, which is a mill to roll thin metal at high speeds. Since rolling at speeds materially in excess' of the speeds characteristic of the prior art was the significant contribution of this conception, it is obvious that Steckel did not deliberately seek to claim apparatus or methods which would limit the speed to those employed prior to his invention. As to the allegation of unreasonable delay in filing the disclaimer, it appears that the '195 application was filed December 9, 1929; that the patent issued on October 21, 1930; and that the disclaimer was filed on September 23, 1932 as soon as Cold Metal became aware of the problem. There is no indication of unreasonable delay or neglect in this sequence of events. Finally, there is no merit in the position that the disclaimer extended the claims. It did no more than to delete a distinct and separate matter without modifying that which it left standing. This precise question was previously raised and similarly disposed of by Judge Wilkin in E. W. Bliss Co. v. Cold Metal Process Co., D.C.N.D.Ohio 1942, 47 F.Supp. 897. To understand the force of Republic's next contention, it must be recalled that '195 is a divisional patent of '016. Although the history of its prosecution is presented in some detail in the subsequent discussion of the license defense, it is enough now to say that the claims which were eventually embodied in '195 were not introduced into the original Steckel application until May of 1928— about five years after the '016 application was filed. However, the Patent Office determined that there was a basis of disclosure of these new claims in the 1923 original Steckel application and accorded to them the filing date of the original Steckel application. Later, at the instance of the Patent Office, these claims were transferred to a divisional application upon which '195 issued in October of 1930, still possessed of the advantage of the 1923 filing date. Republic now argues that there was no basis of support for both the '195 apparatus and method claims in the 1923 application and that when they were finally introduced into that application in May of 1928, the mills and methods therein disclosed had been in public use for more than two years, the period of the statutory bar in effect at that time. I must reject this argument with respect to the apparatus claims. The combination which was disclosed by the 1923 application consists of essentially the same elements as are to be found in the apparatus claims of ’195 — relatively small work rolls, larger backing rolls, and roller bearings on the necks of the backing rolls. True, the 1923 application also disclosed the tension subject matter, but the presence of this additional element does not detract from the support that was given by the other disclosures of the 1923 application to the claims for the identical mill which appear in ’195. Any elaborations of language used in ’195 to describe this mill are not material but are features inherently present in the original Steckel conception. However, with respect to the method claims of ’195, which have heretofore been held to be invalid for overbroadness and indefiniteness, I find no adequate disclosure in the original 1923 Steckel application to support claims of such broad scope. In any event, I do not find that the operation commencing in January 1926 of the United mill by the Rome Brass & Copper Company at Rome, N. Y., which Republic urges as evidence of a prior public use, was of a character to disqualify the subsequent application by Cold Metal. Much of the early operation of this mill was experimental, of doubtful success, and certainly not sufficiently widespread to constitute a public use. Therefore, I find the twelve apparatus claims, Nos. 3, 4, 6, 7, 8 and 11 through 17 unimpeachable and that ’195 is valid with respect to them. However, as stated earlier with respect to method claims 1, 2, 5, 9 and 10 of ’195, I find they are invalid and are not to be given effect in that patent. Having found that the apparatus claims of ’195 are valid, I must now determine if the following mills designated in the order of reference infringed such claims: Name or Number of Mill District Where Located Type of Mill No. 44 Warren 42" single-stand reversing cold mill No. 40 Warren 38" single-stand reversing cold mill No. 1 Tandem Warren 20" 3-stand tandem cold mill Niles Tandem Warren 42" 4-stand tandem cold mill 34" Reversing Masselon 34" single-stand reversing cold mill 2-stand Tandem Masselon 20" 2-stand cold mill 20" Reversing Masselon 20" single-stand reversing cold mill 54" Tandem Cleveland 54" 4-stand tandem cold mill 98" Hot Cleveland 98" 10-stand continuous hot mill As to both ’195 and ’016, the individual accused mills will be compared against the valid claims of the respective patents, without crediting defendant’s claim that such mills come within the cover of the license agreement. Having settled the question of infringement subject to that hypothesis, the license agreement will be subsequently discussed to ascertain if such mills as are found to infringe are thereby relieved of such charge. All of the foregoing mills, except the No. 1 Tandem at Warren, the 54" Tandem at Cleveland, and the 98" Hot at Cleveland, respond in all details to the combination disclosed in '195. They are 4-high mills, having small work rolls, larger backing rolls, and roller bearings on the necks of the backing rolls and are so proportioned as to permit the use of roller bearings of adequate sturdiness to withstand the loads characteristically experienced in the rolling of thin metal strip. Republic has used these mills to roll thin metal strip or sheet and to obtain a commercial product that is substantially free of buckles and wrinkles. Infringement of the apparatus claims is not avoided because Republic also uses flood cooling as a means of controlling roll pass contour on all of these mills, except the 2-stand Tandem at Central. I, therefore, find that these six mills are so constructed and were so operated during the patent period as to infringe the apparatus claims. However, on the No. 1 Tandem, the 54" Tandem,, and the 98” hot mills, a Morgoil bearing instead of a roller bearing was utilized during the patent period. The Morgoil bearing is structurally related to the plain brass or babbit bearings which were characteristic of the prior art mills, but it operates on a hydro-dynamic oil film principle. Instead of the bearings being open and exposed, as was characteristic of the prior art bearings, they are sealed and the lubricant is supplied to the surfaces by a circulating system, in large quantities and under pressure. This results in the maintenance of an oil film between the moving parts, even under heavy loads and at high speeds, and in the exclusion of dirt, water, and other foreign matter. These bearings are fundamentally far superior to the plain •open-type sliding bearing, and exhibit ■coefficients of friction which compare favorably with the best roller bearings. The Morgoil bearings were not introduced into the art until the early 1930s, long after Steckel’s conception of his mill and even subsequent to the issuance •of ’195. Plaintiffs contend that the Morgoil bearings respond to the disclosures in the apparatus claims calling for anti-friction bearings and anti-friction mounting. To accept plaintiff’s argument that these expressions were intended to embrace any bearing, present and future, without being limited and defined by that specific disclosure of roller bearings found in the claims and specifications, would make the apparatus claims so broad and inchoate that they would fall victim to the same defect which invalidated the method claims. Plaintiffs cannot, for the purpose of defending the validity of the apparatus claims, have the benefit of the limited meaning which the inventor intended these expressions to have, and then for the purpose of charging an infringement, assert a broader meaning so as to include within their scope another type of bearing, independently developed after the issuance of the patent through the skill and ingenuity of other people. I, therefore, find that the three mills using the Morgoil bearings do not infringe the apparatus claims of ’195. ’016 Of the twenty-six claims contained in ’016, only the following, Nos. 5, 11, 12 and 21 through 26, are here alleged to be infringed: “5. The process of cold rolling strip, comprising passing it through a plurality of stands of rolls and maintaining a tension on the strip between the roll stands, substantially as described. “11. A continuous mill for rolling metal strip comprising a plurality of stands of rolls, the several stands having at least one relatively small reducing roll and an anti-friction backing roll therefor, the several stands being arranged to successively engage the material and maintain it under tension between stands, a substantial part of the power required for driving the several stands being supplied by such tension. “12. A continuous mill for the rolling of metal strip comprising a plurality of stands of rolls having relatively small reducing rolls and relatively larger backing rolls therefor with anti-friction mounting, a delivery reel, and means for driving the reel and at least some of the stands of rolls, the reel being adapted to supply a substantial part of the power, through tension on the delivered strip, required for the last stand of rolls, said last stand of rolls being arranged to supply, through tension on the strip, a substantial part of the power required for operating the next to the last stand of rolls, and so on, whereby a tension is maintained on the strip between the delivery reel and the last stand of rolls and between the several stands of rolls. “21. The process of cold rolling thin strip metal, comprising pulling a strip of thin metal past a relatively small reducing roll having an anti-friction backing roll. “22. The process of cold rolling thin strip metal, comprising pulling a strip past a relatively small reducing. roll having an anti-friction backing roll, with a tension sufficient to furnish the principal driving, power for the roll. “23. The process of cold rolling thin metal strips, comprising pulling the strip past a relatively small reducing roll provided with an anti-friction backing roll and maintaining a sufficient tension on the delivered strip to insure its delivery in an unwrinkled state. “24. A mill for rolling strip-like material having at least one relatively small reducing roll having an anti-friction backing roll and means for engaging the issuing material in such manner that the same may be maintained under tension, the mill being arranged to be driven largely by such tension on the delivered material. “25. A mill for rolling strip-like material having relatively small reducing roll with anti-friction backing rolls and means for engaging the issuing material in such manner that the same may be maintained under tension, the mill being arranged to be driven largely by such tension on the delivered material. “26. A mill for rolling strip-like material having relatively small reducing rolls with anti-friction backing rolls and a reel for engaging the issuing material in such manner that the same may be maintained under tension, the mill being arranged to be driven largely by such tension on the delivered material.” There is present in all of the foregoing claims, except Claim No. 5, the same combination of elements which is present in the ’195 mill. Just as he did in the apparatus claims of ’195, to describe the mounting for the larger backing rolls, Steekel uses two expressions of his own devising — “anti-friction backing roll” and “anti-friction mounting”— which had no technical meaning in the art. Reference to the specifications, however, is once more permissible to give specific definition to these terms. There Steekel discloses two express techniques for creating an “anti-friction backing roll” and an “anti-friction mounting:” (1) supplying roller bearings for the necks of the backing rolls, as disclosed in ’195; and (2) making the backing rolls of very large diameter compared with their shafts, the backing rolls thus resembling wheels, the turning moment of each wheel (backing roll) on its shaft (roll neck) being exerted through a radius so much larger than the bearing shaft (roll neck) radius that it results in an anti-friction mounting for the backing roll and, therefore, an anti-friction backing for the small work rolls. Thus defined, the claims are not in either the foregoing or in any other respect overbroad but are sufficiently definite to enable anyone skilled in the art to construct and practice Steckel’s invention and to determine the limits of the subject matter covered thereby. My conclusion, arrived at with respect to ’195, as to the failure of any of the prior art to anticipate or foreshadow this combination of elements is equally applicable to the identical combination as it appears in ’016. As was stated earlier, in none of the prior art are these elements brought into combination in the new and useful manner pioneered by Steekel, and it is in the novelty of this combination and in the successful results which it produced that the element of invention resides. Republic has argued that the Blake & Johnson Lancaster mills anticipate ’016 at least with respect to the second technique for achieving an anti-friction mounting for the backing rolls — the wheel-like backing roll — if not with respect to the roller bearing technique. However, in the Blake & Johnson 3-high mill the large rolls are 12" in diameter and have necks 9" in diameter which are mounted in plain sliding bearings. This proportion of 12" to 9"- between the diameter of the backing roll and the diameter of its neck does not respond to Steckel’s conception of a wheel-like backing roll in which the disparity between backing roll diameter and the diameter of the backing-roll neck (or shaft) would have to be far more exaggerated to benefit appreciably from the leverage principle. However, Claim No. 5 does not disclose this distinctive combination of Steckel. It merely involves the process of cold rolling metal strip by passing it through a continuous or tandem mill and maintaining a tension on the strip between the respective mill stands. Since there is no mention of the bringing together of the elements of smaller working rolls, larger backing rolls and a specific type of anti-friction mounting for the backing rolls, there is nothing in this claim to distinguish the mills referred to therein from the prior art mills. The operation of 2-high tandem mills in such a way as to maintain a tension on the strip between the mill stands was well-known in the prior art, and such mills were to be found at Superior Steel Corporation, West Leechburg Steel Company, Trumbull Steel Company and at the Stanley Company. I, therefore, find Claim No. 5 to be invalid, having been anticipated by the prior art. However, in all the other aforementioned claims of ’016, there is present not only the distinctive combination which comprises the mill itself, characteristic of the claims of both ’195 and '016, but an entirely new and different combination as well, one in which the element of tension is to be used instead of or in combination with power applied to the work rolls for the purpose of driving the mills. Apparatus Claims Nos. 11 and 12 specify that a “substantial” part of the power required for driving the several stands of a continuous mill be supplied by tension on the strip, either between stands or between the last stand and the reel. Apparatus Claims Nos. 24, 25 and 26 specify that the mill is to be driven “largely” by tension on the strip. Process Claim No. 22 specifies that the “principal” driving power for the mill is to be furnished by tension on the strip. Process Claims Nos. 21 and 23 disclose an all-pull mill in which the strip is to be moved entirely by tension. Republic has challenged the validity of the foregoing tension disclosures on the ground that they are vague and indefinite. It is true that, with the exception of Nos. 21 and 23, the all-pull mills, these claims do not disclose the precise amount of tension to be used in conjunction with the power supplied to the work rolls. Indeed, it would be impossible to express these values in any more restricted fashion since the amount of tension to be applied to the strip together with the amount of power to be applied to the work rolls are values which must vary with each particular mill. The efficiency of its operation, the peculiar properties of the metal that is then being worked, the degree of reduction that is then being attempted, the roll pressure being applied, and the speed of operation sought to be achieved are all factors which must be balanced in the specific operations. But it is clear that these claims do set forth the guiding rules with sufficient clarity and certainty so as to permit anyone skilled in the art to practice the teaching of such disclosures and to be informed of the boundaries of the claimed invention. Eibel Process Company v. Minnesota and Ontario Paper Co., 261 U.S. 45, 43 S.Ct. 322, 67 L.Ed. 523. For example, in determining what Steckel intended by “substantial” tension with respect to any given operation, it would be necessary to take all of the aforementioned factors into consideration as well as the amount of tension that was characteristic of the prior art under such conditions. Although tension in varying degrees had been used in prior art mills, it was so used only for the purpose of winding up the strip and guiding it through the rolls. In no case had it ever been used in a combination, such as Steckel disclosed, where it was to play a material role in driving a mill engaged in the cold rolling of high ratio metal strip. Therefore, in construing “substantial” it must be assumed that Steckel in-. tended that forward tension would be' used in quantities at least in excess of those needed merely to fulfill' the prior art function. It is in this conception of forward tension as playing a material role in the driving of the mill, in combination with the power supplied to the work rolls, that Steckel’s teaching makes its first departure from and advance over the prior art. In saying that forward tension would constitute the “principal” driving power, Steckel, enlarging on his conception, indicated that forward tension would contribute more to the driving of the mill than the power applied to the work rolls. To ascertain what Steckel meant by a mill “largely” driven by forward tension it must be assumed that he intended to denote a value, separate and independent of the values represented by “substantially” and “principally.” Therefore, I construe “largely” to mean that forward tension must make more than just a material contribution but it need not be greater than 50%, as required by “principally.” It is a value intermediate between “substantially” and “principally.” Republic contends that even if these disclosures of driving the mill “substantially” and “largely” by tension are not to be held invalid for vagueness and indefiniteness, they are invalid because they are broader than and are not supported by the 1923 application. These claims of “substantially” and “largely” were first introduced into the 1923 application by way of amendment on October 17, 1929. At the time of this later amendment, Claims Nos. 11 and 12, disclosing the use of “substantial” tension on continuous mills were added to the original application. At this same time, Claims .Nos. 29 through 31 of a separate application (serial number 200,227, filed June 20, 1927 and maturing in Patent 1,744,018) were later transferred to the original Steckel, ’016 application on December 19, 1929, as Claims Nos. 24 through 26, and were amended by substituting the word “largely” for “principally.” Similarly, the specifications were amended at that date so as to "read “largely or entirely” with respect to driving the mill instead of “principally or entirely.” Republic insists that before these amendments were effected in the fall of 1929, the original Steckel application merely disclosed that the mill would be driven “principally” or “entirely” by tension and that the effect of the subsequent amendments was to broaden the claims of ’016 so as to cause them to en-, compass the use of lesser amounts of tension than originally contemplated by Steckel. Lacking a basis in the original application for such broader claims, Republic argues that they are now void because they were inserted more than two years after the public use of the subject matter included therein. I must reject the foregoing argument of Republic. In connection with one of the embodiments shown in the specifications of the original 1923 application (Fig. 2), the following disclosure is made regarding the amount of tension to be used in the mill: “While a certain amount of power may be supplied to the rolls, it is desirable that a large part of the work be done by pulling on the strip itself, since this holds the strip taut at all times and makes the product perfectly straight.” Clearly, by this disclosure Steckel established in his original application a foundation for broader claims with respect to tension than driving the mill “principally” or “entirely” by tension, a foundation which is adequate to support his later claims with respect to driving the mill “largely” or “substantially” by tension. In any event, the proof tendered by Republic is not adequate to support its position that there was á public use of these disclosures more than two years prior to their insertion in the 1923 application. Republic has also interposed several objections to these claims on the' grounds of indefiniteness and functionality. These, objections are not well taken. Method Claims Nos. 21 through 23 do not describe the method merely iii terms of the function and effect of the operation of the apparatus described in the apparatus claims. Although both apparatus and method claims describe essentially the same type of mill, in the apparatus claims the mill is driven either “substantially” (Nos. 11 and 12) or “largely” (Nos. 24, 25 and 26) by tension while in the method claims it is driven either “principally” (No. 22) or “entirely” (Nos. 21 and 23) by tension. Also, Republic contends, as it did in objecting to the ’195 claims, that the use of the expression “thin metal strip” is too uncertain. However, the meaning of this expression as used in the claims was well understood by persons skilled in the art, even though Steckel did not prescribe precise gauges or thicknesses. Further, Republic contends that certain expressions, such as “arranged,” “adapted,” “sufficient tension,” etc. are vague and indefinite. But I am satisfied that, insofar as they are material, any person skilled in the art would have no difficulty in understanding their meaning and applying their teaching upon reading them in the light of the specifications and the other portions of these claims. Finally, it was not necessary for Steckel to disclose or define a reel structure to be used on a part-pull, part-drive mill. It was well known in the art prior to Steckel that some means would have to be provided for synchronizing the speed of the reel with that of the driven rolls, and such reels were then available. Republic maintains that the product claims are invalid because they are not supported by the original application and were not inserted into such application until more than two years after Steckel’s own prior use and sale of the product. Although these product claims are not here in issue, Republic argues that because Cold Metal permitted the patent to issue with knowledge of their invalidity and because it failed to disclaim them immediately or within a reasonable time, the entire patent is invalidated under the authority of Marconi Wireless Telegraph Company of America v. United States, 320 U.S. 1, 63 S.Ct. 1393, 87 L.Ed. 1731. However, in Marconi it was apparent to the patentee that the original disclosure was inadequate while in the instant case, there having been no adjudication to that effect and the Patent Office having allowed the amendments upon its determination that there was adequate disclosure in the original application, Cold Metal cannot be charged as a matter of law with actual or constructive knowledge of the invalidity of the product claims. Cold Metal was not, therefore, under a duty to disclaim the product claims, and its failure to do so does not affect the validity of the other claims of ’016. I, therefore, concur with the findings of the special master that Claims Nos. 11, 12, 21, 22, 23, 24, 25 and 26 of ’016 are valid and effective, and that Claim No. 5 is invalid. We must now consider plaintiffs’ contention that all of the accused mills, with the exception of the 98" Hot at Cleveland, infringe some or all of the foregoing claims. Since the use of forward tension in' prescribed proportions for the purpose of driving the mill is perhaps the most distinctive feature of the combination disclosed in ’016, before being able to ascertain whether the accused mills infringe any of the ’016 claims, it will be necessary to determine to what extent these mills were operated so as to respond to the tension disclosures of such claims. In July and August of 1949, an inspection of the accused mills was made by both parties pursuant to an order of Court. In the course of this inspection both parties had their experts apply special instruments to the relevant parts of the mills to record the power relationships. The values measured were the forces involved at the respective points rather than the power ratings of the respective motors. Republic objects to the use of the readings taken at this July-August, 1949 inspection on the grounds that since ’016 expired on January 14, 1947, readings taken in July-August, 1949 would not be probative of the power relationships in existence during the patent period. Republic, however, has not been able to show that there was any material variance in the operation of the accused mills between the patent period and the time when the inspection was made, and I accept the voluminous evidence of plaintiffs to the effect that there was no such variance. On each of the accused mills, Republic employs a substantial braking power on the reel which is feeding the strip into the mill, as well as a forward power on the reel which is collecting the strip as it issues from the mill. In most instances this back tension exerted on the strip by the reel which is feeding the strip into the mill is considerably in excess of the forward tension exerted by the reel collecting the strip. The practice of using a substantial braking power on the reel feeding the strip into the mill is not described in the ’016 claims or speeifications, and the essential conflict between' the parties as to the proper formula to be used for evaluating the proportion of force supplied by forward tension and the proportion supplied by the work roll motors revolves around the treatment to be accorded to this element of back tension. It is clear that the total forward impetus supplied to the mill comes from only two sources: the power applied through forward tension on the strip and the power applied through the work rolls. Since the value that is being measured is the proportionate contribution of each source towards this forward impetus, it follows that this value must be found by comparing the forward thrust supplied by each such element to the combined forward thrust as was determined from the July-August, 1949 inspection. By substituting, in the following formula, the readings taken during that inspection, the percentage of total forward power supplied by forward tension will be given: % of Forward Forward Tension_ Power Supplied (Forward Tension) plus by Forward Tension :----------- (Force Supplied by Work Rolls) But, the plaintiffs have argued that such a formula is not appropriate when substantial back tension is used on a given mill. It is their contention that since such back tension is neutralized exclusively by the forward power applied at the work rolls, the large power values found at the work rolls in defendant’s mills are to be accounted for by defendant’s use of rear tension. Since rear tension is not part of the Steckel disclosure and since it has been voluntarily added to defendant’s mills, the plaintiffs say that the amount of such back tension must first be deducted from the work-rolls power before the proportionate contribution of forward tension can accurately be measured. Otherwise, it is held that by the judicious use of rear tension, an infringer might so build up the force value at the work rolls as to be able to conceal the contribution made to the operation of the mill by forward tension and thus to avoid the burdens of his infringement. Plaintiffs’ position would be represented by a formula such as this: % of Forward Power Forward Tension_ Supplied by Forward (Forward Tension) plus Tension = =(Force Supplied by Work Rolls Minus Rear Tension) I find no justification for the arbitrary assertion that the resistance to forward motion which is supplied by rear tension is exclusively neutralized by the forward power applied at the work rolls. Rear tension is but another of the resistant forces, such as friction, resistance of the metal to deformation, bearing losses, etc., which the combined forward-tension and work-roll forces must overcome to make the mill operate. One might just as plausibly say that such rear tension is completely overcome by the forward-tension force and should be deducted from that value before the respective forces are compared. This would result in a distortion prejudicial to the plaintiffs, just as the formula recommended by them produces a distortion prejudicial to the defendant. It is not our purpose to determine to what extent the rear tension is neutralized by each of the respective forces. It is sufficient to realize that both forces cooperate in overcoming the drag of rear tension as they cooperate in overcoming other resistant forces in the mill, and that to measure the contribution of each to this task, its value must be measured against the total forward thrust applied to the mill. Before applying the appropriate formula to the data obtained from the July-August, 1949 inspection, it must be noted that the observations then made on the 2-Stand Tandem were inadequate to permit satisfactory computations showing the power distributions in this mill. Cold Metal has sought to sustain its burden of showing infringement in the operation of this mill by offering certain informal, supplementary data and by the oral testimony of two of its employees as to the way in which this mill was operated. I cannot accept such slight evidence as probative of the power values in the light of the technical accuracy and precision which has been recognized as indispensable to their proper measurement with respect to the other seven mills. Of the seven mills for which there are adequate data, it appears that none were operated entirely by tension on the strip as is called for by Claims Nos. 21 and 23. ■ Similarly, in none of the mills did tension supply the “principal” (or more than 50% of the) driving power as is required by Claim No. 22. Each of the four single-stand mills responded to the tension features of Claims Nos. 24, 25 and 26, which require that the mill be driven “largely” by tension, since each showed tension values at least up to 30% of the total driving power in each mill. Similarly, each of the three tandem mills responded to the tension features of Claims Nos. 24 and 25, since each showed tension values at least up to 35% of the total driving power in each mill. However, only the No. 1 Tandem responded to the tension features of Claim No. 26. This claim applies only to the last stand of a tandem mill since it specifies the reel as the element for “largely” driving the mill. In the No. 1 Tandem there were tension values of at least 25% on the last stand. But in the Niles Tandem and in the 54" Tandem the respective maximum values on the last stand did not exceed 9.6% and 16.4%, which falls short of “largely” driving the mill by tension. Each of the tandem mills responded to the tension features of Claims Nos. 11 and 12, which apply only to continuous mills and which call for the mill, including the last stand, being driven “substantially” by tension. As appeared from the discussion of the accused mills with respect to infringement of ’195, they respond in all their other details to the combination disclosed by Steckel in ’016, except that the No. 1 Tandem and the 54" Tandem must be relieved of the infringement charge under ’016 just as they were under ’195. These, it will be remembered, are two of the three Republic mills which utilize the so-called Morgoil bearing on their backing rolls. In ’016 Steckel disclosed two specific techniques for achieving an anti-friction mounting — roller bearings and wheel-like backing rolls. I have already found that the Morg