Opinion
No. 92-1356.
November 23, 1993.
Dennis J. Mondolino, Hopgood, Calimafde, Kalil, Blaustein Judlowe, of New York City, argued for plaintiff-appellant. With him on the brief was Lee A. Goldberg.
Granger Cook, Jr., Cook, Egan, McFarron Manzo, Ltd., of Chicago, IL, argued for defendant-appellee. With him on the brief were Stephen B. Heller and Ted K. Ringsred. Of counsel were Thomas W. Buckman and Neal C. Johnson, Illinois Tool Works, Inc., of Glenview, IL.
Appeal from the United States District Court for the Southern District of New York.
The Rawlplug Company ("Rawlplug") claims infringement of two of its patents by Illinois Tool Works' ("ITW") line of masonry anchors manufactured and sold under the names of New Striker and Hammer Loc. Both the patents in suit and ITW's alleged infringing anchors relate to a single piece, pre-shaped anchoring device for use in masonry.
After a bench trial, the district court, in a written opinion and order, entered judgment in favor of the Defendant ITW and dismissed the Complaint. Rawlplug appeals, claiming that the district court erred in both interpreting the patent claims and in limiting their scope. For the reasons that follow, we reverse and remand.
BACKGROUND
This patent dispute concerns certain anchors designed to secure fixtures such as roofing material to masonry surfaces. The appellant Rawlplug manufactures and sells various fasteners, including masonry anchors, for use in the construction industry. In the early 1980's Rawlplug's Vice-President for Manufacturing and Engineering, Louis Giannuzzi, conceived a design for a simple yet efficient masonry anchor. The design of this anchor can be generally described as consisting of a shank formed of resilient material such as heat tempered steel, which is shaped during manufacture to include at least one bend along its longitudinal axis. This metal shank has a memory such that, when forcibly deformed — e.g., pounded into a hole — it seeks to recover its original bent shape. As it returns to this original shape, the anchor exerts outward pressure on the walls of the hole into which it has been driven. Rawlplug's patents on this device, which are set out in detail below, are the subject of this suit.
Through its Buildex and Ramset/Redhead Divisions, appellee ITW is also engaged in the manufacture and sale of masonry anchors. ITW manufactures two masonry anchors under the tradenames "New Striker" and "Hammer-Loc." For purposes of this litigation the parties have stipulated that these two anchors are essentially the same. Hence, we refer to them collectively under the title "New Striker."
Rawlplug filed suit in the United States District Court for the Southern District of New York against ITW arguing that ITW's New Striker anchor infringed two of Rawlplug's patents, specifically U.S. Patent Nos. 4,828,445 ("the '445 patent") and 4,963,062 ("the '062 patent"). Rawlplug sought damages for willful infringement. ITW denied infringement.
The trial court's jurisdiction was based on 28 U.S.C. § 1338(a).
Rawlplug is the exclusive licensee of the '445 and '062 patents, herein called the "Rawlplug patents".
After a four-day bench trial, the district court concluded in a written opinion that the New Striker did not infringe the claims asserted by Rawlplug, either literally or under the doctrine of equivalents. Rawlplug Co. v. Illinois Tool Works, 23 USPQ2d 1054, 1992 WL 91703 (S.D.N.Y. 1992). Rawlplug has appealed from this judgment.
I. The Patents in Suit
In its suit against ITW, Rawlplug charges that the ITW New Striker infringed claims 1-4 of the '445 patent and claims 1-3 and 25 of the '062 patent. The language of the patents in suit is straightforward.
A. The '445 Patent
Claim 1 of patent '445 is the focus of our attention, for claims 2, 3, and 4 are all dependent on claim 1 and add additional limitations not relevant to this appeal. Claim 1, which is fully set forth in Appendix I, states the following:
The pertinent part of claim 1 reads:
. . . said shank being pre-shaped to have at least one undulation which deviates from the longitudinal axis passing through the head and shank to create an arcuate bend therein having a peak which normally lies at a position well outside the periphery of the drilled hole and having an upper and lower base, . . . . . . .
. . . said shank to be driven axially into the drilled hole whereby when the shank is so driven this force acts to straighten out the bend exerting substantial outward pressure against the wall of the drilled hole at a first level at which the peak contacts the wall of the drilled hole at one side thereof and at a second and third levels at which the upper and lower bases contact the wall of the drilled hole on the other side thereof. . . .
(a) one piece, pre-shaped anchor;
(b) containing at least one indentation in the shank;
(c) which indentation creates a bend resulting in a peak on one side of the shank and upper and lower base on the other. This configuration can be described as arcuate, i.e., bent or curved in the form of a bow;
(d) the anchor's shank is of a constant diameter that is slightly less than the pre-drilled masonry hole;
(e) the shank is made of a resilient material whose "memory" is such that after it is driven into the masonry hole, it seeks to regain its original shape.
When the resilient shank seeks to regain its original shape, it creates an outward pressure on at least three points — the peak and the two base points. In essence then, claim 1 teaches that the pressure of driving the anchor into the masonry hole tends to straighten the shank without damaging the hole, and that after the shank has been driven, its resiliency causes the anchor to regain its original shape.
The language of the '445 patent also reveals that the claim is not limited in the number of undulations or bends. Claim 1 requires "at least one undulation" or bend, but the language covers an anchor with more. Secondly, the shank cannot chisel its own original shape into the pre-drilled masonry hole. If such occurs, the memory of the shank is not actuated, and the anchor will not hold.
B. The '062 Patent
Rawlplug's other patent in suit, U.S. Patent No. 4,963,062, is similar but adds various head configurations not germane to the present appeal. The '062 patent does require that the resilient material of the shank have a "non-abrading bearing surface" so that the anchor, when hammered into the hole, will not "score, chisel or otherwise mutilate the hole." Claim 1 of this patent (on which claims 2 and 3 depend) reads in pertinent part:
. . . said masonry being of a material which if scored, chiseled or otherwise mutilated by the anchor would alter the shape of the hole wall and thereby weaken the holding power of the anchor. . . . . . . .
. . . said shank being formed of resilient material and having a non-abrading bearing surface, so that when it is forcibly driven into said tightly fitting hole it will not score, chisel or otherwise mutilate the hole wall, and said undulation will be forcibly altered by reaction with the wall of the hole, and in seeking to recover its original shape, will frictionally secure the anchor in place.
(emphasis added). Claim 25, like claim 1, contains this "non-abrading" language. Claim 25, however, does not require that the anchor have at least one undulation.
Claim 25 of the '062 patent reads in pertinent part:
. . . said masonry being of a material which if scored, chiseled or otherwise mutilated by the anchor would alter the shape of the hole wall and thereby weaken the holding power of the anchor; said anchor compromising a drivable element and an elongated pre-shaped shank formed of resilient material having a non-abrading bearing surface and having a cross-sectional dimension which at no point there-along substantially exceeds that of the drilled hole, said pre-shaped shank having a first portion extending from the first portion and displaced from said longitudinal axis, so that when the shank is forcibly driven into said tightly fitting hole it will not score, chisel or otherwise mutilate the hole wall, and said second portion will be forcibly altered by reaction with the wall of the hole to bring it toward alignment with said longitudinal axis and thereby frictionally secure the anchor in place.
Rawlplug manufactured a product called the "Rawlplug Spike" in accordance with the disclosures of these two patents. The spike has enjoyed remarkable success in the industry as a substantial improvement over prior fastening methods.
II. The Alleged Infringing ITW New Striker
At trial, ITW never contested that it set out to engineer around the Rawlplug patents. A civil engineer at ITW named Nilsen purchased several Rawlplug Spike anchors. After testing those anchors and reviewing the Rawlplug patents, Mr. Nilsen designed the ITW anchors. Mr. Nilsen's anchors were:
(a) pre-shaped with upper and lower portions and a transition center portion with a bend;
(b) uniform in diameter throughout its length;
(c) made of heat-treated steel to give it a resiliency or spring temper.
Unlike the patents in suit, the ITW anchors had unidirectional notches on the shank. These notches form serrations that extend above the surface of the anchor to the approximate thickness of a human hair. This feature was designed to enhance the functional engagement between the shank and the cylindrical wall of the pre-drilled masonry hole. Because these notches are very thin, they abrade the masonry wall only slightly. These notches apparently increase the holding or gripping strength of the anchor.
More importantly, however, the ITW anchors, when struck into masonry holes, function in a similar fashion as set forth in claim 1 in Rawlplug's '445 patent. The upper and lower portions of the anchor (above and below the center bend) tend to straighten to conform to the masonry hole. After insertion, the anchor tends to regain its original shape, giving four points of pressure on the masonry walls. Because of the notches, the forced insertion of the anchor tends to abrade the masonry wall to a small degree.
III. The District Court Opinion
The district court found that although ITW knew of the Rawlplug patents and its Rawlplug Spike and intended to design around them, the ITW New Striker and Hammer Loc products did not literally infringe either the '445 patent or the '062 patent. First, with respect to the '445 patent, the district court focused on the patent's requirement of an arcuate bend. The court emphasized that the '445 patent was awarded despite its similarities to Japanese Patent Publication 55-122918 (the "Katou patent") on the grounds that unlike the Katou anchor, which has a linear offset shank, Giannuzzi's application described an arcuate bend. The district court concluded that "arcuate" meant both shaped as a bow and having an upper and lower portion of the shank (or bow) that was in the same line or axis. Finding neither in the uninstalled ITW products, the court found no literal infringement of either Rawlplug patent.
Second, the court held that the '062 patent had not been literally infringed by the New Striker because the New Striker's serrations preclude it from having the "non-abrading" surface described in claims 1 and 25 of the '062 patent. The court specifically rejected Rawlplug's claim that the term "nonabrading" could mean that the anchor might slightly abrade.
Finally, the court found that the New Striker did not infringe either the '445 patent or the '062 patent under the doctrine of equivalents. It determined that the New Striker's holding power does not result from a three point pressure system centered around a peak and two bases, but instead results from a combination of a four point pressure system and a mechanical interlock between the serrations and the masonry wall.
IV. Appellant's Arguments
In its appeal, Rawlplug contends that the district court erred as a matter of law in holding that ITW's anchors did not infringe claims 1-4 of Rawlplug's '445 patent and claims 1 and 25 of Rawlplug's '062 patent. It makes three arguments to this end. First, it asserts that the trial court erred when it interpreted the '445 patent claims as encompassing only those anchors whose holding power is derived solely from three points of frictional contact between the wall of the masonry hole and the anchor and in which the levels of contact center around a single peak and two bases lying in the same longitudinal axis. In support of this argument, Rawlplug places particular emphasis on the phrase "at least one undulation" in claim 1 of the '445 patent. Rawlplug suggests that the district court improperly relied on the commercial embodiment of the '445 patent — which relies on three pressure points — rather than the claim language, in concluding that the patent required only one undulation in which the upper and lower axis of the bend must be in the same longitudinal axis.
Rawlplug's second argument is that the district court erred as a matter of law by interpreting the '062 patent term "non-abrading" as requiring zero abrasion. Rawlplug argues that this term should be read in context with the other phrases in the claim and with reference to the patent history. Ultimately, Rawlplug believes that the '062 patent was literally infringed by the New Striker despite the New Striker's slightly abrasive quality.
Its third contention is that the court incorrectly limited the scope of the claims of both the '445 and '062 patents based on a combination of elements selected from two unrelated prior art patents — the Katou and Carroll patents.
DISCUSSION
As noted above, the Appellant is challenging the district court's interpretation of its patents as a matter of law. We review the court's decision in this regard de novo since claim interpretation is an issue of law. Intervet America v. Kee-Vet Laboratories, 887 F.2d 1050, 1053 (Fed. Cir. 1989); Mannesmann Demag Corp. v. Engineered Metal Prod. Co., 793 F.2d 1279, 1282 (Fed. Cir. 1986); Envirotech Corp. v. Al George, Inc., 730 F.2d 753, 758 (Fed. Cir. 1984). We conclude that the district court incorrectly interpreted the claims, and thus clearly erred in finding no literal infringement.
As recognized by the district court, a determination of infringement requires a two-step process: interpreting the claims of the patent and then a comparison of the properly interpreted claims with the accused device. Graver Tank Mfg. Co. v. Linde Air Prod. Co., 339 U.S. 605, 70 S.Ct. 854, 94 L.Ed. 1097 (1950); Amstar Corp. v. Envirotech Corp., 730 F.2d 1476, 1481 (Fed. Cir.), cert. denied, 469 U.S. 924, 105 S.Ct. 306, 83 L.Ed.2d 240 (1984). Claim construction takes into account the specification of the patent, the prosecution history, and the prior art. Intervet America, 887 F.2d at 1053.
II. Literal Infringement: The '445 Patent
An examination of ITW's New Striker reveals that its form is deceptively different before and after installation. For example, the New Striker, before installation, appears to have only a single bend. There are offset notches on either side of the shank before the bend, and the shank does not curve back after the bend to form the shape of a bow. The district court apparently considered the shape before installation. Yet the claims of the Rawlplug '445 or '062 patents are specifically crafted to claim an innovative device after installation. These claims clearly reflect that it is the fastening configuration in masonry that is protected. Thus, we look to the product as installed, since the installed product is what is detailed in the claim and the claim language is controlling. E.I. DuPont De Nemours Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433 (Fed. Cir.) (citing Autogiro Co. v. United States, 384 F.2d 391, 395-96, 181 Ct.Cl. 55 (1967)), cert. denied, 488 U.S. 986, 109 S.Ct. 542, 102 L.Ed.2d 572 (1988).
The New Striker when installed appears below:
When installed the Hammer Loc appears as follows:
Both of these exhibits were in evidence at trial, and neither was controverted by ITW. A careful examination reveals that each enters the predrilled masonry hole on an angle. Where that occurs, it is clear that there are one or more undulations in the shank taking on an arcuate shape. In other words, the ITW products, when installed, have undulations resulting in peaks and basis that are bow shaped. Equally important, the peaks and bases extend and rest against the wall of the masonry hole, creating the binding and fastening effect. Finally, the binding and holding strength is derived from a resilient material in the shank that seeks, after being driven into the masonry, to regain its original shape.
The difference between the ITW products and the Rawlplug patent claims (if there is one) is that the arcuate bend in the uninstalled ITW products is not as clearly defined before installation. The undulation and peak become clear on installation because the ITW products position themselves within the hole on an angle.
Apart from the clear language of the patent claims, we also consider the prosecution history, the prior art, and any expert testimony on the meaning of the claims. Fonar Corp. v. Johnson Johnson, 821 F.2d 627, 631-32 (Fed. Cir. 1987), cert. denied, 484 U.S. 1027, 108 S.Ct. 751, 98 L.Ed.2d 764 (1988). As indicated in the district court's opinion, Rawlplug was required to distinguish its claims from those of the Katou patent before its patent issued. The district court found that Rawlplug's inventor was required to assert his anchor had a wavy undulation, in contrast to the Katou anchor, which had a straight offset portion, in order to obtain the '445 patent. The offset portion of the Katou anchor had a straight form without a peak. The district court was persuaded that the ITW anchors more closely resembled Katou and were thus distinguishable from Rawlplug's '445 patent. We disagree. Again, the district court's analysis was based on the ITW anchors before installation. After installation as discussed above, the ITW anchor, which enters the masonry hole at an angle, clearly shows the undulations and arcuate bend. ITW's own expert, Hasan, testified to such.
Specifically, the anchor in Katou has an angle cut in the shank to create an area of overlap. As Figure 6 of the Katou patent graphically demonstrates, the anchor sits straight up in the hole and does not tilt. This is the result of an overlap area that forms the bulge with greater cross-sectional difference, which the Patent and Trademark Office used to distinguish over the Rawlplug claims.
Consequently, the Katou patent is easily distinguishable both from the Rawlplug claims and ITW's infringing anchors. The anchor in Katou has an angle cut in the shank to create an area of overlap. As Figure 6 of the Katou patent (reproduced below) demonstrates, the anchor sits upright in the hole.
Above the overlap area, the shank contacts only one side of the wall; below the overlap, the shank contacts the other. In contrast, ITW's anchor tilts and creates an undulating shank without an overlap area or bulge. All of this was explained by Rawlplug's expert witness, Holmes. The New Striker does not have a linear offset like Katou but arcuate bends that are the basis of the fastening strength of the anchor.
The district court also found that the Rawlplug patents were not literally infringed because the axis of the upper and lower portion of the ITW anchors were not in the same line. We can find no support for imposing such a requirement on the '445 patent in the record. No expert testified to such a requirement, and more importantly the claim language of the '445 and '062 patents do not contain such a restriction. By including such a requirement, the district court improperly limited the scope of the Rawlplug patent claims.
We conclude that the district court committed reversible error in finding no literal infringement of claim 1 of the '445 Rawlplug patent by ITW's New Striker and Hammer Loc products. We reverse.
III. Literal Infringement: The '062 Patent
ITW fares no better with the '062 patent. As noted above, the district court determined that the serrations on the New Striker took that anchor out of the '062 patent because that patent calls for a non-abrading shank. This court finds that a different result obtains when the term "non-abrading" is read in the context of the other language of the patent and in light of the prior art against which the non-abrasion language was added.
We note that we are required to compare the alleged infringing product with the claims of the patent not with the patentee's product. Intervet American, Inc. v. Kee-Vet Laboratories, 887 F.2d 1050, 1055 (Fed. Cir. 1989).
Claim 1 states in pertinent part:
. . . said shank being formed of resilient material and having a non-abrading bearing surface, so that when it is forcibly driven into said tightly fitting hole it will not score, chisel or otherwise mutilate the hole wall, and said undulation will be forcibly altered by reaction with the wall of the hole, and in seeking to recover its original shape, will frictionally secure the anchor in place.
(emphasis added). We read the phrase "so that it will not score, chisel or otherwise mutilate the hole," as modifying and illuminating the term "non-abrading" in the context of the design of the shank. The phrase "frictionally secure" suggests that the anchor was to operate not only by the force of the shank pressing outward against the walls of the masonry hole, but also by the operation of friction between the shank and the walls. The inclusion of "frictionally" suggests that at least some minimal abrasion was contemplated, but the qualifying language above that phrase indicates that the friction or abrasion would not be such as to score, chisel or mutilate the hole. Moreover, we note that the claim states only that the anchor was to be non-abrading "when it is forcibly driven into said tightly fitting hole," not that it will not abrade at all when finally in place.
This last reading is consistent with the patent's purpose as recited in the patent specification's "Background of Invention" section. This section describes that one "serious drawback" of Rawlplug's "Rawl Drive" anchor is the bulge created by the split and expanded half sections, which has a "maximum transverse dimension which is necessarily significantly greater than the diameter of the hole drilled in the masonry." Because the bulge is significantly greater than the diameter of the hole, "when the Rawl Drive anchor is driven into a hole driven in masonry, it may score and mutilate this hole and thereby weaken the holding power of the anchor." The summary of the advantages of the '062 patent invention provides that "[t]he maximum cross-sectional dimension of the shank is not substantially greater than the diameter of the hole, whereby when the shank is driven therein it will not chisel or otherwise mutilate the hole wall." This discussion suggests that the patent was designed to overcome the insertion abrasion that had been a problem with earlier devices. This discussion jibes with the statement in the claim that the anchor has "a non-abrading bearing surface, so that when it is forcibly driven into said tightly fitting hole it will not score, chisel or otherwise mutilate the hole wall." The specification also provides that upon full insertion, the anchor will have a "gripping action." The reference to gripping is entirely consistent with the claim language "will frictionally secure." Thus in a number of respects, the patent specification supports our interpretation of claim language. See E.I. DuPont De Nemours Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1433 (Fed. Cir.) ("It is entirely proper to use the specification to interpret what the patentee meant by a word or phrase in the claim."), cert. denied, 488 U.S. 986, 109 S.Ct. 542, 102 L.Ed.2d 572 (1988).
In its opinion, the district court correctly notes that Giannuzzi's application for this patent was rejected initially in light of prior art, namely U.S. Patent No. 1,574,790 (the "Carroll patent"). The Carroll patent was a design for a bow-shaped box handle to be used for a wooden battery box. The patent specifies that the handle be made of resilient material and have teeth in the nature of pointed spurs on its surface. When the metal handle was driven into a hole in the wood, it would straighten and then, once wedged in the hole, seek to recover its original shape, thus securing the handle through pressure and friction. This manipulation of the memory of the driven metal shape operated similarly to Giannuzzi's proposed anchor.
Unlike the anchor, however, the Carroll patent handle had metal teeth, which would help secure the handle in place by digging into the sides of the wooden hole. Giannuzzi argued to the patent examiners that his anchor did not have such teeth, and indeed such teeth would so severely abrade the walls of the masonry as to undermine the anchor's effectiveness. To distinguish his application from the Carroll patent, Giannuzzi added ( inter alia) the phrase "non-abrading . . . bearing surface," and "will not score, chisel or otherwise mutilate the hole wall." With this revision (and other changes irrelevant here), the application was granted, and the patent issued.
While the patent examiners may have relied on the non-abrading surface to distinguish the '062 patent from prior art, we must interpret that term consistently with the other language in the claim. Cf. Autogiro Co. v. United States, 384 F.2d 391, 395-96, 181 Ct.Cl. 55 (1967) (claims of the patent provide the concise formal definition of the invention and it is to these wordings that one must look to determine whether there has been infringement). As noted above, we read the claim to provide that the anchor will not abrade the walls of the masonry hole when it is forced therein to such an extent that the walls are scored, chiseled, or mutilated. We do not read the non-abrading language, as the district court did, to preclude the exertion of some friction on the walls of the hole once the metal anchor has regained its original shape. Indeed we read it to require that some friction be exerted by the notation that the anchor will be "frictionally secure" in place.
Moreover, we find that slight abrasion on insertion is contemplated in the '062 claim — for the anchor will be only slightly narrower than the hole into which it will be driven (thus, the phrase "tight fitting hole"), so that there will be a tight fit when the metal shank regains its curved form. All that is precluded by the claim is abrasion that mutilates, scores, or chisels the hole on entry.
Consequently, we find that the slight serrations found on the New Striker shank are not inconsistent with the claims of the '062 patent. These serrations are located on the convex surface of the shank and extend above the surface of the shank to a maximum of five thousandths of an inch, approximately the thickness of a human hair. The purpose of these serrations is to improve the frictional engagement between the shank and the cylindrical wall of the hole. Although (according to ITW's expert Mr. Hasan) there is some slight abrasion when the New Striker is inserted, the serrations do not score, chisel, or mutilate the masonry hole. Once in place these serrations increase the anchor's holding power by enhancing its frictional engagement. This abrasion is not inconsistent with the language of the '062 claim that the shank "frictionally secure the anchor in place." Hence, we conclude that the district court clearly erred in finding that the New Striker did not literally infringe Rawlplug's '062 patent.
Accordingly, we reverse and remand for further proceedings in accordance with this opinion.
REVERSED AND REMANDED.