In Re Vamco MacHine and Tool, Inc

RICH, Circuit Judge. This appeal is from the U.S. Patent and Trademark Office (PTO) Board of Appeals (board) decision of April 12,1984, affirming the examiner’s rejection, in a reexamination proceeding, of all claims of patent No. Re. 29,795 issued to Vamco Machine and Tool, Inc. (Vamco), assignee of the inventor, Harry Eyberger. The patent title is “Self-Contained Feed Roll for Power Punch Presses.” We affirm. How This Case Got Here The original Eyberger patent, No. 3,483,-782, issued December 16, 1969, with three claims. After the discovery of some new prior art (apparently the V & 0 feeder hereinafter discussed), Vamco applied for reissue of the patent with extensively amended claims on Dec. 16, 1977, and the patent was reissued on Oct. 10, 1978, with claims 1-9. The original and reissue file histories are not before us. Vamco brought suit on the reissue patent for infringement by F.J. Littell Machine Company in the District Court for the Northern District of Illinois, Eastern Division. We copy the following paragraphs from Vamco’s statement in the PTO under 37 C.F.R. § 1.510(b) in the reexamination petition: On January 28, 1982 Judge Bua issued an order (Exhibit 1) in which the Defendant Littell was ordered to make a diligent search for all prior art and file the same with the Court. Plaintiff, Vamco, was ordered within 90 days of January 28, 1982 to request reexamination of United States Patent Re 29,795. Plaintiff was also ordered to advise the United States Patent Office of all the prior art filed in Judge Bua’s Court and served on Plaintiff, Vamco, by Defendant, Littell. The Defendant on February 25, 1982 submitted a Prior Art Statement Pursuant To The Order of January 27, 1982 (Exhibit 2). Judge Bua’s order, Exhibit 1, which states that it is “for the purpose of avoiding piecemeal proceedings and litigation, for the purpose of conserving judicial resources, and for the purpose of utilizing the recently enacted laws of the United States” (obviously referring to reexamination proceedings under 35 U.S.C. Chapter 30, which added §§ 301-307 on December 12, 1980), also granted plaintiff’s motion for volun *1566 tary dismissal without prejudice and granted plaintiff leave to file a motion to vacate the order within thirty days after the PTO completed reexamination, saying: 6. Upon Plaintiff’s filing of the motion to vacate ... this matter will be reinstated before this court, this matter will retain its original docket number and this matter will proceed on the complaint originally filed by Plaintiff in this action. Pursuant to the order, defendant Littell filed its prior art statement February 25, 1982, listing nineteen U.S. prior art patents, and Vamco obediently told the PTO about them while arguing no substantial new issue of patentability was raised. The request for reexamination was granted on June 8, 1982, the PTO finding that a substantial new question of patentability was raised, and took its course through several examiner’s letters and responses and the filing of several affidavits until a final rejection resulted on December 16, 1982. Following that, there was a personal interview with the examiner attended by Vamco’s attorney and Mr. J.P. Gentile, who is an affiant and an officer of Vamco (a family-owned company), and Edward S. Paris, another affiant employed by Vamco. Five more affidavits were thereafter filed, appeal to the board was taken, and the examiner filed his Answer. Vamco’s attorney then realized he could overcome one of the examiner’s principal points by a clarifying amendment to the claims, got the examiner to agree to it in a telephone interview, and the examiner wrapped it up with a Supplemental Answer. Claims 1-9, all the claims in the patent, were on appeal to the board, which affirmed, and the same are on appeal to us. The references relied on had by then been reduced to four, in support of the sole ground of rejection which is obviousness from the prior art, 35 U.S.C. § 103. The references are: "V & O Heavy Duty Roller Gear Feeds,” a catalog of V & 0 Press Company, Inc. (hereinafter V & 0) Sweet U.S. patent 363,776 May 24, 1887 La Ganke et al. (La Ganke) 1,408,894 Mar. 7, 1922 Wittek 1,796,417 Mar. 17, 1931 Our appellate jurisdiction in this reexamination proceeding is provided by 35 U.S.C. § 306, 35 U.S.C. § 141, and 28 U.S.C. § 1295(a)(4)(A). Issue The sole issue before us is whether the board erred in holding the invention, as defined in claims 1-9, construed in the light of the disclosure in the specification of Reissue patent No. 29,795, would have been obvious within the meaning of § 103 from the disclosures of the prior art references relied on to reject them. A subissue is whether commercial success and other “secondary considerations” were established for Eyberger’s invention. The Invention Disclosed in the Patent No issue has been raised with respect to what is disclosed, but the precise nature of the disclosure is of considerable significance, especially with relation to appellant Vamco’s heavy reliance on the so-called “secondary considerations” which must be considered in passing on obviousness vel non. We therefore point out certain aspects of the disclosure which we deem significant. We note in passing that Harry Eyberger, the named inventor, on the record before us, is a mere name. Nothing is disclosed about him or about the making of his invention. The dramatis personae in this case are officers and employees of Vamco, an outside consulting engineer, a professor, and a customer who have submitted affidavits on Vamco’s behalf. The invention relates to an accessory for power punch presses which stamp out metal parts, large and small, from sheet metal which has to be fed between the stamping or punching dies and is often fed from rolls. At the top of such a press is a continuously rotating shaft which can be used as a source of power to run a strip-feeding machinism. What Eyberger’s patent says he invented is, in broad terms as set forth in the “Abstract”: A self-contained unit to be attached to a power punch press to provide a feed for stock being fed to the press. The unit contains a direct drive from the punch press crankshaft to feed rolls which are *1567 located adjacent to the table of the press. A non-slip, positive, drive is provided____ In stating what had been wrong with prior feeders, he said many of them “contained clutches and brakes, which did not permit positive drive ... clutches wore or brakes failed causing spoilage of pieces and damage to tooling.” His invention, he said, “provides positive, direct drive between the crankshaft of the punch press and the feed rolls. There is no slippage, clutches, brakes, or other friction-type mechanism which can cause loss of timing or costly repairs.” Another aspect was a “novel release means which permits the feed stock ... to be released from the feed rolls during the pressing ----” Notably missing from his list of stated “objects” of his invention is any mention of providing a more accurate or more precise or speedier setting or adjustment of the amount by which the stock is fed to the press, which has become the central theme of Vamco’s arguments for patentability, the significance of which will appear later. Below is Fig. 1 of the patent, which is “a front elevational view of the feed roll unit of the present invention with a portion of the housing removed and showing a portion of the punch press in phantom view.” *1568 The side of the phantom press is at 12, its crankshaft is 14, driving the feed unit through belts 20 and 24. The power-driven feed roll is 40 and 50 is the releasable idler feed roll. Attention is directed to the roller chains 34 and 42 which connect sprockets (after the manner of bicycle drives) 28, 30, and 38 which will be better understood from Fig. 3, below, which is a “diagrammatic perspective representation of the working portions of the feed roll mechanism ... in semi-schematic form.” (We have added “S” to the stock, “30a” and “30b” to the compound sprocket and also the brackets with references to elements named in claim 1. In the schematic, the sprocket teeth have been omitted.) *1569 We can now describe the parts of the feeder unit and their relationship and operation. At the top, is the crankshaft 14 of the punch press equipped with a pulley 16, which drives, through cleated belts 20 and 24 and pulley 18, pulley 22 on an “index drive 26.” This is a commercially available right-angle drive which changes the continuous rotary motion of shaft 26a into step-by-step incremental rotary motion in shaft 26b so as to feed the stock to the press intermittently. Output sprocket 28 drives unit 30 which consists of the two sprockets 30a and 30b connected together, through drive chain 34 connected to 30a. Sprocket 30b is connected by chain 42 to feed roll 40’s drive sprocket 38. Stock S being fed is pressed against roll 40 by idler roll 50 but can be released during stamping by electro-pneumatic controls 52, 58, 66, and 60 which need not be explained, though Eyberger seemed to regard them as an important part of what he invented and claimed in claim 3. The upper die or punch of the press is 12. With respect to the arguments in this case, the significant disclosure is the double sprocket 30 shown in Figs. 1 and 3 and what the specification of the patent says about it and its closely associated parts. The first statement reads: The slotted hole 32a [see Fig. 1 at the bottom] permits adjustment of the idler pully [sic; sprocket] in various positions to accommodate various size chains and pullies to thereby adjust the feed stroke *1570 of the feed roll unit in a manner to be more fully hereinafter explained. The latter explanation is under the heading “OPERATION” and is: It may be seen that by placing various combinations of sprockets 28, 30 and 38, of different sizes, the amount of lineal motion imparted to the feed stock through power driven feed roll 40 may be varied for any single rotation of crankshaft 14. Thus, the exact length of feed desired may be controlled by changing the size of the sprockets. The indexing increment can also be increased or decreased by the use of various gear trains (not shown) in place of chains 34 and 42 and sprockets 28, 30 and 38. When the size of idler sprocket 30 is changed it is positioned along the slotted hole 32a of the support 32 so that drive chain 42 is in proper tension. The chain tensioner sprocket 36 [Fig. 1] is then adjusted to properly tension drive chain 34 so that a direct drive is provided between the index drive 26 and the power driven feed roll 40. The foregoing is the totality of what the patent has to say on the subject of adjusting the feed and it will be noted that such adjustment involves removing chains from sprockets, removing sprockets from shafts, replacing them with others, and reinstalling and tensioning the chains. It will be further observed that though passing mention is made of the possibility of using “various gear trains,” none is shown and none is described, yet, as will appear, this case is argued primarily on the basis of apparatus using gear trains, particularly with respect to “secondary considerations.” In concluding the specification and summing up what the invention achieves, all the inventor chose to emphasize was that he had provided “a self-contained feed roll unit” which could be “readily installed,” which had a “positive drive” and in which “the force on the feed rolls may be released periodically during the pressing operation.” The Invention Claimed Of the nine claims, only claim 1 is independent. Although the PTO rejections of various claims rely on various combinations of prior art references, it will suffice to discuss claim 1 in its final, reissued, and subsequently amended form. It reads: A self-contained feed roll unit for a power punch press having a crankshaft rotating in timed relation to the pressing portion of the operating cycle of said punch press, said feed roll unit comprising: (a) a power driven feed roll; (b) an idler feed roll rotatably supported on a movable frame and normally urged toward said power driven feed roll, said idler feed roll being periodically moved away from said power driven feed roll by feed roll release means to free stock being fed to said power press during the pressing portion of said power press operating cycle; (c) an index drive having an input shaft to receive continuous, uniform rotary motion and an output shaft transmitting non-continuous, incremental rotary motion; (d) first positive drive means drivingly connecting said index drive input shaft to said power press crankshaft; and (e) second positive drive means including a first drive member connected to said index drive output shaft, a driven member connected to said power driven feed roll, a compound idler member having different diameters drivingly connected to said first • drive member and said driven member to thereby drivingly connect said index drive output shaft to said power driven feed roll, said second positive drive means being adjustable to vary the ratio of the angular velocity of said index drive output shaft relative to said power driven feed roll. [Emphasis ours.] The eight dependent claims, by their numbers, add the following additional details or *1571 limitations, stated for simplicity in our own terms: 2. element (e) in claim 1 is formed of chains and sprockets and the length of stock fed is varied by changing the size of the sprockets. 3. the “feed roll release means” of (b) in claim 1 comprises certain named compressed air, cam, solenoid valve, and pilot valve apparatus, which the specification describes in detail. (A featured aspect of the invention as described and claimed in the patent.) 4. the length of stock fed is varied by changing at least one (in (e)) of the first drive member, compound idler drive member, or the driven member. 5. the length of stock fed is varied by changing the size of the “compound idler drive member.” 6. the stock fed is varied by changing the size of the “first drive member” in (e) of claim 1. 7. the stock fed is varied by changing the size of the “driven member” in (e) of claim 1. 8. the unit of claim 1 contains means permitting the mounting of different drive, idler, and driven members so stock fed can be varied by changing their sizes. 9. the unit contains mounting means for the index drive output shaft in fixed spaced relation to the feed roll (i.e., the relation of 28 to 38, Fig. 3). It is of interest to note that most of the italicized wording of clause (e) in claim 1 and all of dependent claims 4-9 which refer back to it was added by the application to reissue the original patent filed eight years to the day after the issuance of the original patent. Clause (e) and the language in claims 4, 5, and 8 was further amended at the very end of the reexamination, more than 18% years after the original patent, to avoid a prior art patent (Hallenbeck) heavily relied on by the examiner, to specify that the “idler member” was “compound,” i.e., consisted of two sprockets “having different diameters” connected together. In the reissue patent, it was claimed simply as an “idler member.” In the original patent it was not recited in the claims at all, but only included by implication as an unnamed part of the “second positive drive means” defined as “being adjustable.” 1 The patent will expire December 16, 1986. We find it significant that the claims have been twice altered over the years to keep up with Vamco’s commercial interests. (See discussion, infra, on commercial success, etc., of the “Eyberger invention.”) Rejection of the Claims for Obviousness This appeal is from the decision of the board; but its decision was simply to affirm the examiner’s final rejection, which was that all claims except claim 2 would have been prima facie obvious in view of the V & 0 Catalog taken with La Ganke and Sweet, and claim 2 for the same reason when Wittek was considered, the references listed at the beginning of this opinion. Our own review of the references leads us to conclude that the PTO position respecting prima facie obviousness was fully justified. The V & 0 Catalog discloses in words and pictures a power press roll feed driven from the press crankshaft by a chain drive and a roller gear (“indexing”) unit that imparts intermittent motion to the feed. We quote further: On the output shaft of the roller gear unit and the input shaft of the roll feed *1572 are a pair of pick off gears that transmit the power to the feed. These pick off gears provide a fast and low cost method of changing feed lengths. By using a standard diameter lower feed roll, feed increments of .001" can be provided. A time consuming change over of costly lower feed rolls is not necessary. Because of this type of construction, friction clutches, brakes with water cooling devices and reciprocating motions are eliminated. Among listed “features” is an adjustable, cam-operated idler roll 2 release operated by the press slide. Because appellant pressed the point that the V & 0 feed has only two gears, whereas its feeder has four, the examiner cited Sweet which discloses “gearing for changing speed,” for its increase or decrease, and for application “to any kind of machinery.” Sweet is cited for its disclosure of a movable and changeable combination gear interposed between the driving and driven gears in a manner similar to Eyberger’s combination sprocket 30a-30b. In Sweet, the support for the shaft for the combination gear is slotted, in a manner similar to Eyberger’s slotted support 32a for his idler sprocket shaft. Below are figures from Sweet showing three different gear combinations, the combination idler gears being G-H, G-H', or G'-H. Gear B on the left is the power gear and gear C on the right is the driven gear. The slotted mounting for shaft F of the combination gear is self-evident. *1573 Wittek discloses another power punch press stock feed with a pair of feed rolls wherein there is a chain-and-sprocket drive for the lower feed roll powered from the press main shaft. It is cited for the statement: Since the sprockets 13 and 29 are removable, they may readily be changed to vary the speed of the die mechanism [which carries the feed rolls] and thus cause a more rapid or slower movement of the stock 33 through the die set. Compare claim 2, supra. La Ganke also discloses a roll feed for punch presses wherein the upper or idler feed roll is moveable. He provides mechanism, including cams acting on the idler roll shaft, to release the pressure on the stock while the pressing is in progress, as recited in clause (b) of claim 1, supra. V & O and La Ganke taken together disclose punch press feeders having everything shown by Eyberger except the compound idler sprocket defined in clause (e) of claim 1, which is worded broadly enough to cover compound gears. The examiner relied on Sweet to show the compound gear speed adjustment which falls within clause (e). It is clear to us that one of ordinary skill in machine design would have had no difficulty in designing the Sweet gear arrangement into the V & 0 feeder should greater versatility in speed, or length of stock fed, be desired. With no more than that change necessary to arrive at the invention claimed in the Eyberger reissue patent, a rejection of the claims for obviousness was clearly warranted. Vamco’s “Secondary Considerations” Arguments Of course, “evidence rising out of the so-called ‘secondary considerations’ must always when present be considered en route to a determination of obviousness.” Stratoflex Inc. v. Aeroquip Corp., 713 F.2d 1530, 1538, 218 USPQ 871, 879 (CAFC 1983). As we believe was done in the PTO, we shall fully consider the evidence, giving each piece of evidence its appropriate weight, in answer to Vamco’s contention that the board disregarded this evidence. As indicated earlier, appellant submitted several affidavits for the purpose of showing commercial success, the filling of a long-felt need in the art, copying by competitors, advantages of the invention, and in order to submit opinions of “experts” that “the subject matter described and claimed in the Eyberger patent” would not have been obvious to one of ordinary skill in the art in 1967, at the time the invention was made, which was taken to be the year the application for that patent was filed. Vamco also made of record an article on its feeders from “Metal Stamping.” The affidavits and the article were considered in detail by the examiner and in the opinion of the board. They found them unpersuasive but for somewhat different reasons. The examiner said in his Answer on the appeal to the board, inter alia, “it is emphasized that it is clear that these affidavits and article relate to the specific commercial embodiment which is simply not disclosed in Requestor’s [the one who requests reexamination] patent.” The examiner had earlier pointed out in his Answer that “the commercial embodiment utilized by Requestor and to which specific embodiment many advantages have [been] urged throughout the [Requestor’s] Brief bears little or no resemblance to the drawing of the patent in issue [Re. 29,795].” To shed light on what he was talking about, reproduced below is one picture from the “Metal Stamping” article showing Vamco’s gear arrangement, allegedly part of the “second positive drive means” of claim 1, clause (e), used in place of Eyberger’s sprocket and chain structure shown in Fig. 1, supra. It looks different and it is different, as the examiner pointed out. *1574 The main theme of Vamco’s arguments here and in the PTO is in essence that the invention “described and claimed in the Eyberger patent” enables the user of its feeder “to vary the stock feed length from .052 inches to 6.000 inches in increments of .0005 inches” by simply changing the set of gears shown above and that there are no “blind spots” in such adjustment, that is to say, there is no dimension, to within at least .001 inch of feed, which cannot be achieved. It is further said that this can be done in “a few minutes.” The Eyberger patent does not disclose or even hint at any such possibilities or even list such accurate feed length changes among the objects of the invention. It states, at most, only that “the exact length of the feed desired may be controlled by changing the size of the 3 sprockets,” but does not say how. The board opinion also considered the affidavits at length but primarily took the position that no nexus had been established between the commercial success of the Vamco feeders and the subject matter claimed by Eyberger. In substance, we agree with both the examiner and the board but for somewhat different reasons, which we shall express as we review the affidavits. The Affidavits Patrick J. Gentile, an officer of Vamco, begins by saying his company produces a geared cam feed that is “covered by the claims of Reissue Patent No. 29,795.” That is but a layman’s opinion on a legal question, and is no indication that Vamco’s commercial feeder is disclosed in the patent in such fashion that those skilled in the art would be taught how to make Vamco’s commercial feeder. Gentile then repeats and amplifies the statement, which we made above, about micro feed adjustment possibilities, which, as we also pointed out, are not taught by the patent. He states in conclusory fashion that the geared cam feed of Vamco filled a long felt need in the industry because feed “can be accurately changed in a matter of minutes by merely changing the gear train,” a statement clearly not applicable to Eyberger’s sprocket and chain structure which would be more cumbersome to disassemble and reassemble and might also involve substitution of different chains, as the patent clearly states. He then says Vamco’s feeders are market leaders and that sales have increased from 6 in 1973 to 300 in 1979 but it is not shown how Eyberger’s disclosures had anything to do with that. As developed later, there are in fact several other reasons for the market success. The first Paris affidavit is now irrelevant due to the withdrawal of a rejection since it only discusses the Hallenbeck reference, no longer relied on and not of record. The same is true for the second Paris affidavit which presents calculations to show that Vamco’s feeder would have fewer blind spots than either Hallenbeck or the V & O feeder modified by the Hallenbeck teaching. But even so, Eyberger does not even suggest the existence of blind spots and certainly teaches no way to eliminate them. Professor Erdlac of the University of Pittsburgh Department of Mechanical Engineering, obviously well qualified in his field, repeats in substance the entire specification of the Eyberger patent, most of which is in the prior art, and critical passages from the claims such as clause (e), and analyzes all the references relied on. He then gives his opinion that “the subject matter described and claimed in the Eyberger patent” would not have been obvious in .1967 over the disclosures of the prior art, *1575 which opinion is not at all surprising. We respect his opinion, for what it is worth, but the question is one of law which we must decide and involves matters other than those he investigated and reports on. Our problem with his lay opinion on patent-ability, however, is that we have no idea what he means by “the subject matter described and claimed.” Does he mean Eyberger’s sprocket and chain feeder or does he include the Vamco gear machine as marketed, which he seems to include within the term “claimed”? We too can read this patent and understand what it describes. What it claims, however, in terms of the number of different permutations and combinations of sprockets, or of gears, substituted for sprockets with the omission of chains, especially as appellants and Professor Erdlac construe them, is vastly more than the patent describes or teaches, or even suggests. Clearly, the most important segment of the Erdlac affidavit is his computations showing what an “extremely fine change” in feed length can be produced by changing the number of teeth on the compound sprockets 30a and 30b and the sprockets 28 and 38. From his calculations, Professor Erdlac concludes that “it should be clear that the system of compound gears described and claimed in the Eyberger patent will result in extremely fine increments of advance.” (Emphasis ours.) We do not dispute his mathematical demonstrations. However, Eyberger’s patent, originally or after reissue, never described any such possibility, or any system of gears, and not until eight years after it was issued did it claim, in claim 1, clause (e), anything more than a “second positive drive means” which was adjustable to vary feed roll speed. 3 Professor Erdlac has taught us in his affidavit how to get very fine adjustments, but the Eyberger patent does not. In Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966), speaking of the plow shank there involved, the Supreme Court said at page 25, 86 S.Ct. at page 697: Petitioners’ argument basing validity, on the free-flex theory raised for the first time on appeal is reminiscent of Lincoln Engineering Co. v. Stewart-Warner Corp., 303 U.S. 545 [58 S.Ct. 662, 82 L.Ed. 1008] (1938), where the Court called such an effort “an afterthought. No such function ... is hinted at in the specifications of the patent. If this were so vital an element in the functioning of the apparatus it is strange that all mention of it was omitted.” [303 U.S. at 550, 58 S.Ct. at 665.] The Metal Stamping article (April 1976) on the Vamco feeders becomes very pertinent at this point. It sheds much light on the secret of obtaining fine adjustments and on how it is actually done in practice. The board dismissed it as “hearsay”, a “trade article which is expected to be laudatory”; but we find it of great interest on just what has made the Vamco feeders such a success. This relates in a way to the board’s point about nexus. The writer of the article first reports on the great accuracy of Vamco machines. They are handsomely finished and of great precision and made of the best materials. “The Vamco design, developed by Vamco president Pat Gentile [a former toolmaker] reduces feed length changeover ... to a matter of five minutes or less. The concept is simplicity itself ____ [A] gear train consisting of four gears has been developed. To change feed length, you change the gear train and readjust the roll release cam.” It is then pointed out that a separate set of gears is required for every feed length, but “costs under $100” and is available immediately from stock. That is to say, the user of the press decides what feed he wants and orders a set of gears from Vamco. Once he gets them, they can be installed in minutes. The story continues: *1576 Just as impressive as the design of the feed itself was the program Vamco used to develop the gear configurations required to provide feed lengths over the full range in increments of .001 inch. Since there are four gears in each set, the possible permutations run into the high millions ____ The computation of just a few of the possible combinations would have required hundreds of man hours. Vamco solved this problem by developing a computer program to do the necessary calculations. The result is a thick set of print-outs in which five gear combinations are established for every feed length. Why five? To allow for availability of gears at all times. [Emphasis ours.] To achieve the remarkable incremental adjustments in steps of .001 inch with no blind spots, which is alleged to be the great advantage of the invention “described and claimed” in the Eyberger patent, one therefore needs not only the gear machine developed by Pat Gentile and the appropriate set of gears, but also either Vamco’s computer program or the computer print-outs and must get “a different set of gears for each feed length,” according to the article. 4 The program would be essential to the user who wishes to overcome “blind spots.” It is clear that the patent contains no disclosure which enables even a reader skilled in the art to achieve the great advantages now attributed to the Eyberger invention and relied on to show unobviousness, or to account for the commercial success which is asserted to be indicative of unobviousness. Stephen Miketic’s affidavit consists of an extensive recitation of his qualifications, apparently as an expert in industrial management and plant design with long experience, and otherwise is either a copy or paraphrase of most of Professor Erdlac’s affidavit, minus the professor’s computations. He, of course, is of the opinion that the “subject matter described and claimed in the Eyberger patent would have been unobvious” in 1967 and has the same opinion as Professor Erdlac as to each of the references relied on by the PTO. He does add some color by saying that a quantum leap would have been required to get from the prior art to the feed unit of the Eyberger patent. However, he seems to equate the Eyberger patent unit of 1967 with the Vamco commercial unit of 1983, itself a quantum leap for which there is no justification. We do not find repetitive opinion affidavits of this type to be of much help in solving patentability problems. What we do find helpful is facts of which we would not otherwise be aware. Terry L. Wissman is Vice President of Engineering of a manufacturer of metal-forming automatic presses. Though not appearing to be qualified as an expert on patents or patent law, he undertakes to tell what Reissue patent No. 29,795 is all about, saying he has read and understands it. He says the “feed roll unit claimed in the Vamco patent” permits accurate changes in feed length “by changing one or more members of the second positive drive means,” talking claim language, and significantly reduces “blind spots.” But he does not explain how. He says that in 1976 his company, Minster Machine Co., began buying Vamco geared cam feed roll units “which are the subject of the Claims 1 and 3-9 of U.S. patent Re. 29,795” to equip its presses, rather than making its own, and that the Vamco units reduced the blind spots in feed lengths to 0.001 inch or less, but he does not explain how. And finally for a third time he says feed lengths on Vamco units can be changed in increments of 0.001 inch or less in a matter of minutes by changing one or more members “in the second positive drive means,” in claim language again, without saying how any more explicitly. He concludes by saying “Such flexibility and accuracy was not available until the advent of the Vamco feed roll which is described and claimed in U.S. Patent No. Re. 29,795.” *1577 Wissman’s omission of claim 2 from his above enumeration of claims was well advised. It is the claim specific to what the Eyberger patent discloses, wherein the “second positive drive means” “is formed of chains and sprockets” and stock feed is varied by changing them. We fail to see, however, what claims have to do with what the Eyberger patent taught the art or with the Minster Company’s election to buy Vamco feeders. Certainly it is not buying Eyberger feeders. As the board said, no nexus has been established between them and the success of Vamco’s product, which is a combination of superbly built gear feeders and computerized replacement gear supply service. The second Gentile affidavit ends the story. All it says is that the Vamco geared cam feed discussed in the article in Metal Stamping “is a geared cam feed that is described and claimed in U.S. Patent No. Re. 29,795.” By now it seems superfluous but we will say it anyway, lest someone has not attentively read the foregoing opinion: The article does not discuss the cam feeder described in the patent and the patent does not describe a geared cam feed. What it claims 5 has nothing to do with the question of what it was that enjoyed commercial success and filled a felt need in the industry or was copied by competitors. In summary, for the foregoing reasons we find no “secondary considerations” in this case having a bearing on the legal issue of the obviousness of the invention of the Eyberger reissue patent. We therefore affirm the decision of the board which affirmed the examiner’s final rejection of claims 1-9 of Reissue patent No. 29,795. AFFIRMED. Footnotes [1] The practical effect of adding this language to (e) of claim 1 and of adding new claims 4-9, was to make it possible to argue the existence of "an inventive concept" which does not appear in the original patent, in either its specification or claims. The question is thus raised whether this was not, in effect, an enlargement of the scope of the claims, masquerading as a narrowing, contrary at least to the spirit of 35 U.S.C. § 251, last paragraph, which reads: No reissued patent shall be granted enlarging the scope of the claims of the original patent unless applied for within two years from the grant of the original patent. [Emphasis ours.] However, this question was not raised by the PTO. [2] The idler roll of the feed should not be confused with the idler sprocket which drives the lower driven feed roll, [3] The way original claim 1(e) actually stated it was: "adjustable to vary the ratio of the angular velocity of said index drive output shaft relative to said power driven feed roll," which is, simply, either speed of the first roll or feed of the stock, depending on the terminology one chooses to use. [4] The PTO was aware of these print-outs as a page from one of them was submitted by the attorney on two occasions and is of record before us. [5] Claims in patents are required by 35 U.S.C. § 112, second paragraph. The function of claims is (a) to point out what the invention is in such a way as to distinguish it from what was previously known, i.e., from the prior art; and (b) to define the scope of protection afforded by the patent. In both of those aspects, claims are not technical descriptions of the disclosed inventions but are legal documents like the descriptions of lands by metes and bounds in a deed which define the area conveyed but do not describe the land. Because of this characteristic of claims, the commercial success of a machine "claimed” may be due entirely to improvements or modifications made by others to the invention disclosed in a patent. Such success, we are holding, is not pertinent to the non-obviousness of the invention disclosed. This is, however, not a holding that advantages inherent in what is specifically disclosed in a patent are not to be considered in determining non-obviousness.