462 F. 2d 1115 - Shatterproof Glass Corporation v. Guardian Glass Company

Shatterproof Glass Corporation instituted this action against Guardian Glass Company, Inc. charging Guardian with patent infringement and the theft of alleged trade secrets. Both parties are in the glass bending business, and the patent and alleged trade secrets relate to molds or fixtures which these manufacturers employ in bending glass to form automobile windshields. Shatterproof complained that Guardian infringed Claims 1, 3, and 7 of the Jendrisak patent-in-suit 3,103,430 and appropriated 43 trade secrets. Following a twelve day trial, the district court found Claims 1, 3, and 7 invalid as obvious and determined that Shatterproof possessed no trade secrets. The court further concluded that Claims 1 and 3 were infringed if valid and that Claim 7, if valid, was not infringed.

On appeal, Shatterproof does not contest the finding that Claim 7 is obvious from the prior art. Appellant also abandons its claim to the alleged trade secrets. Thus we consider only whether Claims 1 and 3 were properly declared invalid.

Shatterproof and Guardian are manufacturers of replacement windshields for automobiles and do not make the original equipment. Pittsburgh Plate Glass, Libbey-Owens-Ford, and Ford Motor Company supply the original equipment to automobile manufacturers, while the parties to this action supply replacements to repair shops which replace broken windshields. Originally, the manufacture of windshields was simple. All that was needed were two flat panes of glass. One pane served as the left side of a windshield, and the other pane served as the right side; a divider strip separated the two panes. Beginning in the early 1950's, new concepts in windshield design were introduced. The single piece of curved glass became the prevailing style. Then in 1954, Ford Motor Company introduced the larger, heavier, more deeply bent wraparound or panoramic windshield. These new designs required a technique for bending glass and to meet this need glass bending molds were developed.

The bending process is simple in theory. A flat pane of glass is placed on a mold. The mold then travels along a conveyor through a long furnace. While moving through the furnace, the heatsoftened glass sags and conforms to the shape of the supporting mold. A modern laminated windshield is produced by placing a plastic sheet bonding agent between two flat panes of glass and passing this combination through the furnace.

Early molds for the slightly curved windshields consisted of only one piece. Later, however, the wraparound windshields required more sophisticated fixtures. These molds consisted of three parts: a center section and two end sections. During the bending process the mold would move from an open to a closed position. Unbent flat glass would be supported horizontally on the mold in its open position. During heating, the glass initially would sag at the middle and begin to drop into the center section. Then, the end sections would swing upward to the closed position of the mold and bend the glass. The sections would be connected with and their movement coordinated by means of various hinges, pivots, and links. Some center sections were movable; some were not. Different molds used varying numbers of support points for the flat glass. Some early multiple section molds were "thrust" molds in that the fixture employed a pushing force against the end edges of the flat glass to create a buckling effect in the softened glass and accelerate the bending. Later molds discontinued thrust forces and relied solely upon the force of gravity to produce the necessary sag in the softened glass. The simplified sketch, demonstrates the basic structure of a modern glass bending mold.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

The continuing changes in mold design and structure were due not only to the manufacturers' natural tendency to improve performance but also to the need to maintain pace with windshield style changes. A new shape in a windshield might require a new mold. Whenever an automobile manufacturer proposed a new style in a windshield, mold designers at the original equipment fabricators would determine the feasibility of making a mold which would produce the desired style. If the proposed windshield could be molded, the automobile makers would be supplied. Thereafter, the original equipment manufacturers would provide the replacement suppliers, such as Shatterproof and Guardian, with samples of the windshield to be used. From these samples, Shatterproof and Guardian would then prepare molds to produce the replacement parts.

There emerges from the record, therefore, a picture of a crowded art not precisely scientific in its refinement. The testimony of Joseph Jendrisak, inventor of the patent-in-suit, indicates that making a mold to produce the desired style of windshield is in part a matter of trial and error. If a test mold did not produce the windshield, changes in design would be made. The changes would involve such modifications as relocating hinges and pivots or adjusting the weight of the three mold parts.1

At trial, hinges, pivots, mold weights, and points of support were among the bases utilized to distinguish the patent-in-suit from the prior art. Shatterproof sought to prove that the structure of its mold differed significantly from that of other molds. Appellant also sought to show that its mold operated differently to produce a different result from other fixtures. The district court found no patentable distinctions in the structure of Shatterproof's mold and found it obvious from the prior art. Further, the court found that a mold which had been placed in prior public use by an original equipment manufacturer demonstrated the same basic structure found in the patent-in-suit and worked in the same way to produce the same result.

The Jendrisak patent-in-suit was issued on September 10, 1963. The application had been filed on August 6, 1958, as a continuation-in-part of a patent application filed March 1, 1956. The patent relates to a glass bending apparatus consisting of a center section and two symmetrical end sections which are hinged to the ends of the center section. The mold supports a flat glass pane on its marginal edges at six-points in a horizontal plane, two points being at the tip ends of the end sections and four points being at the four corners of the center section.

Patentability depends upon the satisfaction of three express statutory requirements: novelty, utility, and nonobviousness. 35 U.S.C. Secs. 101-103. We are here primarily concerned with the last of these criteria.2 In reviewing the determination of the district court, we are mindful of the teaching of Graham v. John Deere Co., 383 U.S. 1, 17, 86 S. Ct. 684, 694, 15 L.Ed.2d 545 (1966):

While the ultimate question of patent validity is one of law, A. & P. Tea Co. v. Supermarket Equipment Corp., supra, 340 U.S. at 155, 71 S.Ct. at 131, the Sec. 103 condition, which is but one of three conditions, each of which must be satisfied, lends itself to several basic factual inquiries. Under Sec. 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this back-ground, the obviousness or nonobviousness of the subject matter is determined.

Thus we first consider the district court's factual findings on the content of the prior art and the improvements on that art contributed by the patent-in-suit. These findings may be disapproved only if clearly erroneous. Rule 52(a), Federal Rules of Civil Procedure; Westwood Chemical, Inc. v. Owens-Corning Fiberglas Corp., 445 F.2d 911 (6th Cir. 1971). Then we consider the correctness of the district court's legal conclusion of obviousness.

One of the most important prior art references relied upon by Guardian and the district court was an unpatented mold utilized by Ford Motor Company. Guardian maintained that the mold was a proper prior reference because it had been in public use more than a year prior to the date of the Jendrisak application for patent. 35 U.S.C. Sec. 102(b).3 Shatterproof contends in its brief that "there is no finding that the Ford mold was in public use more than a year before the filing date of the patent-in-suit. . . ." On the contrary, one of three such findings by the district court is that "[i]n the month of July, 1954 (which is about 20 months prior to the filing of the patent-in-suit) the Ford mold was put into commercial operation." 322 F.Supp. 854, 858 (1970). Such a use was public. Electric Storage Battery Co. v. Shimadzu, 307 U.S. 5, 59 S.Ct. 675, 83 L.Ed. 1071 (1939); FMC Corporation v. F. E. Myers & Bro. Co., 384 F.2d 4 (6th Cir. 1967). Ample evidence supports that finding. This objection to the consideration of the Ford mold is without merit.

*****

* * *

In a further attempt to remove the Ford mold from view, Shatterproof stated at oral argument that Guardian did not give the statutorily required notice that the Ford mold was to be used as prior art. 35 U.S.C. Sec. 282.4 But Shatterproof points to no objection in the record to the use of that art; it did not raise the question at the hearing on its motion for a new trial; it did not brief the issue on appeal. The purpose of the Sec. 282 notice is to prevent surprise. The testimony of Jendrisak reveals that over a year prior to trial he was supplied with drawings of the Ford mold, and examined that mold. His attorney was sent photographs. The pre-trial order listed as one issue to be litigated: "[w]hether the bending fixtures of Defendant Guardian Glass Company, Inc. were based upon designs of the Ford Motor Company or upon the design of Plaintiff." The right to this notice may be waived, and Shatterproof's afterthought comes too late. The mere breach of Sec. 282 does not necessarily require reversal. Absent the required notice, the trial court may still allow proof of the prior art where fairness is not sacrificed, and in this regard, "much discretion is committed to the trial judge." Thermo King Corporation v. White's Trucking Service, Inc., 292 F.2d 668, 677 (5th Cir. 1961). The unpatented Ford mold was properly considered as part of the prior art.

*****

* * *

With this threshold matter aside, we may examine the substance of the prior art. In the interest of clarity and economy we juxtapose the factual inquiries into the state of the art and the improvements thereon of the claimed invention. Much time was spent in the court below analyzing the alleged distinctive features of the patent-in-suit. Claim 1 of the patent describes the essential features of the invention.5 There are two end sections pivoted to and swinging about a center section on parallel horizontal transverse axes. The center section is supported by the end sections for vertical movement therewith. The glass is initially horizontally supported at six points, of which two are at the tips of the end sections and four are between those sections. Claim 3 is dependent on claim 1 and requires that the four intermediate high support points be at the ends of the center section.6

At trial, much testimony and argument centered on the "six point support." After eliciting from Jendrisak the statement that he thought that he had invented "[t]he six-point support," Guardian sought to prove a lack of novelty and obviousness. Shatterproof contends that Jendrisak merely used an abbreviated designation for his invention. In considering this aspect of the patent, the district court noted that two copending patents of Jendrisak disclosed everything of significance in the patent-in-suit except for the six point support.7 This feature, however, was old in the art. The Ford mold used the same glass supporting technique. Three patented molds, while distinguishable as to other features, displayed the six-point support. Thus this element of Jendrisak's mold was not novel. The district court also noted that judging from the prior art, as the glass became heavier and larger it was an obvious step to increase the points of support to six. But the court did not confine its inquiry to an examination of the six-point support characterization of the invention. Other asserted distinctions of the patent-in-suit were considered. In reviewing the factual findings of the district court only essential portions of the evidence are referred to here, and a more complete summary may be found in the lower court's opinion at 322 F.Supp. 854 (1970).

One alleged difference between the Shatterproof and Ford molds is that the Ford mold supposedly does not provide horizontal support as required by Claim 1. The Ford mold has a slight tilt of about three-eighths of an inch in a two foot span. The district court noted that "both Mr. Jendrisak and his counsel admitted that a mold with such a slight tilt would infringe Claim 1, that it would still be 'horizontal', which we think is the practical and intended meaning of the word." Id., at 860. The evidence indicates that the glass was supported in a horizontal plane to keep it from slipping off the mold while traveling through the furnace. The district court did not erroneously conclude that the Ford mold similarly supported the glass in a horizontal plane as required by Claim 1. Any modification to make the Shatterproof mold more nearly horizontal would be an obvious step.

Another suggested distinction between the Jendrisak patent and the Ford mold is that the patent-in-suit allegedly requires extreme lightness in mold weight. But the claims state no such requirement. Both the Guardian mold and the Shatterproof mold were made of fabricated strips whereas the Ford mold was a cast mold. A Guardian vice president, who had previously worked for Ford, testified that the choice of construction was dictated by considerations of cost and projected length of use. The district court found that this unclaimed aspect of the Guardian and Shatterproof molds was old and in the prior art. This finding was not erroneous.

A third alleged distinction between the Shatterproof mold and the Ford mold is that the Ford end sections do not swing about parallel axes as required by Claim 1. The Ford axes are slightly non-parallel. The designer of the Ford mold testified that the orientation of the axes was determined by the shape of the windshield to be bent and had no effect on the operation of the mold. The district court, in our view, was correct in finding no basis for patentability predicated upon the parallelism of the axes.

Shatterproof next contends that the Ford fixture is a thrust mold whereas its mold is a gravity mold. As stated above, a thrust mold utilizes compressive forces on the ends of the flat panes of glass to assist the natural gravity-in-duced sag of the heat-softened glass. Shatterproof argues that the Ford mold employs a stop which exerts pressure on the glass. The district court found from the evidence that the stop engaged only the top of the glass to prevent an upward curl of the glass as the mold reached its closed position. Thus the stop could not create a thrust. Appropriate here is mention of Claim 7 of the Jendrisak patent.8 That Claim relates to a "hold down device" found obvious by the district court. The specifications of the patent demonstrate that this device performs the same function as the stop. Testimony of the Ford mold designer emphasized that the Ford mold was a gravity mold and utilized no thrust pressures. Thus the district court correctly found that the Ford fixture was not a thrust mold.

A final attempt at distinguishing the Ford mold concerns the mounting of the end sections. In the Ford mold, one end section was mounted on a fixed pivot, and the other section was mounted on a swinging link. In the Jendrisak patent-in-suit, both end sections are supported on swinging links. However, a prior art Jendrisak patent 2,551,607, issued five years before the filing date of the patent-in-suit, disclosed symmetrical links supporting the end sections of a glass bending mold. The evidence indicated that the prior art Jendrisak patent and the Ford mold teach and reasonably suggest the end section mounting of the patent-in-suit.

Consequently the Jendrisak mold shows no significant structural improvement over the prior art which would support a conclusion of patentability. The invention taken as a whole would be reasonably suggested to a person of ordinary skill in the art. However, Shatterproof contends that its invention has been misunderstood. Appellant maintains that its mold differs from the Ford mold in that it operates in a different way to produce a different result. This conclusion relies upon the total effect created by the suggested distinctions in the Shatterproof mold. It is argued that the patented mold is not obvious because it solves a problem which existed in the art. See Application of Conover, 304 F.2d 680, 49 C.C.P.A. 1205 (1962).

Such a combination of old elements would be patentable if it performed a new function. But an alleged combination patent must be scrutinized with care. A. & P. Tea Co. v. Supermarket Corp., 340 U.S. 147, 71 S. Ct. 127, 95 L.Ed. 162 (1950). Appellant asserts that the parallel axes and two movable links on its fixture cause the center section of the mold to move in a purely vertical manner whereas Ford's mold has a lateral component in connection with the movement of the center section. During the bending process, the Ford center section moves laterally, perhaps as much as three inches, while moving vertically about five inches. Shatterproof contends that this lateral movement causes problems for Ford.

Ford did experience a peculiar problem in that its windshields would upon occasion explode in storage or in transit. Thus at the end of its production line, Ford positioned a man to hit each windshield with a rubber hammer. This procedure assured that only sound windshields were released. Apparently, the problem was not prohibitive since the Ford mold produced approximately 2,000,000 windshields.

Shatterproof and Guardian did not experience this problem, and the appellant asserts that its mold was responsible for preventing any explosions. However, the designer of the Ford molds testified that the problem was strictly an annealing problem and had nothing to do with tooling. The annealing process occurs subsequent to the bending. Shatterproof's hypothesis is not proven in the record.

The district court concluded that both molds operated on the same basic theory and stated that "there are no significant differences in the functioning of the Ford and Jendrisak molds." 322 F. Supp. at 860. We agree with this conclusion.

In challenging the validity of the patent-in-suit, Guardian was, of course, confronted with the statutory presumption of patent validity. 35 U.S.C. Sec. 282.9 But "[t]he failure of the Patent Office to consider pertinent prior art tends to weaken if not vitiate, the presumption of validity." Westwood Chemical, Inc. v. Owens-Corning Fiberglas Corp., supra, 445 F.2d, at 916. The significant prior art Ford mold was not considered by the Patent Office. Guardian has overcome the weakened presumption. And Shatterproof has not shown that its patent is more than an obvious aggregation of old elements. We therefore affirm the district court's conclusion that Claims 1 and 3 are invalid.

In its formal judgment the district court enjoined Shatterproof from asserting infringement of "Jendrisak Patent No. 3,103,430" against Guardian or its customers. Appellant challenges the injunction on two grounds. First, it is said that the injunction is overbroad since only Claims 1, 3, and 7 were litigated. Second, it is said that the order granting the injunction did not comply with Rule 65(d), Federal Rules of Civil Procedure, by setting forth the reasons for its issuance. Guardian maintains that the relief was not overly broad because Claims 2 and 4-6 embody the same subject matter found in the adjudicated claims.10 Further, Guardian asserts that the reasons for the issuance of the injunction may be found in the opinion of the district court.

By pre-trial order, the patent issues were narrowed to the question of the validity of Claims 1, 3, and 7. The district court's "holding of invalidity should only apply to the patents' claims directly in issue." Westwood Chemical, Inc. v. Owens-Corning Fiberglas Corp., 445 F.2d at 913. But under well-established principles of res judicata, Shatterproof is barred from asserting against Guardian any claim which could have been asserted in this cause of action. See e. g., Commissioner v. Sunnen, 333 U.S. 591, 68 S.Ct. 715, 92 L.Ed. 898 (1948); Morris v. Jones, 329 U.S. 545, 67 S.Ct. 451, 91 L.Ed. 488 (1947). Shatterproof does not dispute these principles but urges that the remaining claims were not litigable in this action.

The district court did not articulate its reasons for enjoining all claims of the patent and we think the action should be remanded to permit the parties to demonstrate whether Claims 2 and 4-6 of the Jendrisak patent were litigable in the present action. The court should make appropriate findings to support its conclusions in this respect. If in fact Claims 2 and 4-6 are, for example, merely restatements of the adjudicated claims or if they contain no significant variations therefrom, then they could have been litigated and, in our view, would be properly enjoinable.

We need not determine appellant's allegation under Rule 65(d), Federal Rules of Civil Procedure, since the district court will have the opportunity on remand to set forth adequate reasons for any injunction which may be entered. Nor do we consider other issues raised by appellant since we find them to be without merit.

The judgment of the district court is affirmed and the action is remanded for further proceedings relative to the scope of any injunctive relief.