Appeal from the May 23, 1983, judgment of the District Court for the Southern District of Texas, sitting without a jury and holding invalid claims 1, 2, and 4 of appellant’s (Lindemann’s) U.S. Patent No. 3,945,-315 issued March 23, 1976 and entitled “Hydraulic Scrap Shearing Machine”. We reverse and remand.
BACKGROUND
The Patent
United States Patent No. 3,945,315 (’315) issued March 23, 1976 on an application filed April 16, 1975. Peter Dahlem and Hubert Milles are named co-inventors and Lindemann is listed as the assignee. The ’315 patent claims a priority filing date, under 35 U.S.C. § 119, of May 13, 1974, based on West German application 2423003.
Hydraulic scrap shears, the subject matter of the ’315 patent, are a principal tool of the scrap metal industry. The shears are large, often weighing several hundred tons, and are designed to cut scrap metal into smaller, uniform pieces for recycling.
There are two basic types of metal processed in the shears: “peddler’s scrap” and “rigidly massive scrap”.
Peddler’s scrap consists of light to medium gauge metal objects, such as light tubing, automobile bodies, and window frames. It makes up a large percentage of the available scrap and is comparatively easy to process.
Rigidly massive scrap consists of heavy gauge metal objects, such as boilers, oil tanks, and railroad cars. Because of thickness or internal reinforcements, massive scrap objects are difficult to process. Traditionally, massive scrap had been processed in very large, tremendously powerful shears, or had been pretreated, e.g., with oxyacetylene torches, to reduce its size or weaken its internal reinforcements. Either approach was costly and time-consuming. Many scrap dealers handled peddler’s scrap exclusively.
The Invention
The ’315 patent contains five claims. Claim 1, the only independent claim, is written in Jepson form:
1. In a hydraulic scrap-shearing machine comprising an open feed channel having two opposing side walls, scrap shears at one end of said feed channel and having a mouth narrower than the normal width of said feed channel between said side walls, hydraulic means for moving at least one of said side walls towards the other of said side walls whereby scrap placed in said feed channel can be squashed to a final width no greater than the width of said mouth of said scrap shears, and a feeder ram for pushing scrap along said feed channel into said mouth of said scrap shears, the improvement consisting of said movable one of said side walls being divided into two longitudinal portions of different lengths, and said hydraulic means comprising a main hydraulic ram having a working face forming the longer portion of said movable side wall, and an auxiliary hydraulic ram having a working face forming the shorter portion of said movable side wall just upstream of said mouth of said scrap shears, said auxiliary hydraulic ram being capable of operation independently of said main hydraulic ram.
The claimed structure is shown in Figure 2 of the ’315 patent:
*1456 [[Image here]]
In operation, the combined rams (17, 19) advance into the feed channel (9), crushing and compacting the scrap (12) against the other, non-movable sidewall (14). With peddler’s scrap, the two rams move the entire distance together. However, when the channel contains rigidly massive scrap, such as shown at (12), the two rams are quickly brought to a standstill by the scrap’s resistence to crushing. The auxiliary ram (19) is then moved forward independently of the main ram (17). The auxiliary ram, having a smaller working surface than the combined rams, is capable of applying a greater crushing force to the scrap. The auxiliary ram cracks and buckles the scrap directly in front of it to crush the leading end of the scrap so it can be pushed through the mouth of the shears. That action also propagates that effect to an adjacent area (H) of the scrap. The structural integrity of the scrap is thus overcome by the auxiliary ram, thereby reducing the resistance of the portion of the scrap in contact with the main ram, allowing both rams to continue forward to crush the scrap to a width less than that of the shear mouth. The feeder ram (11) then pushes the crushed scrap through the mouth of the shear and under the shear blades (at 5) and clamp (at 6). The clamp holds the crushed scrap in place during cutting.
The claimed invention allows one machine of moderate size to process both peddler’s and rigidly massive scrap, and to do so quickly, inexpensively, and without the need for pre-treating massive scrap. Unchallenged testimony described crushing accomplished in minutes of scrap that would have required hours to crush in earlier larger machines and that could not have been crushed without pretreatment.
District Court Proceedings
On October 5, 1980, Lindemann sued appellees (collectively “Amhoist”) for infringement of claims 1, 2, and 4 of the ’315 patent. Amhoist asserted non-infringement and counterclaimed for a declaratory judgment that the ’315 patent is invalid.
A three day trial was conducted on June 21-23, 1982. On May 23, 1983, the district court entered FINDINGS OF FACT AND CONCLUSIONS OF LAW, the introduction of which stated:
After hearing all the evidence the Court concludes that the patent is invalid. Plaintiff simply incorporated two admittedly well-known metal compression features in the same machine and sought to gain a, monopoly in the use of knowledge *1457 that had previously existed in the public domain. The Court finds and concludes that the claimed invention of the Plaintiff does not meet the statutory or constitutional requirements established for patent protection. Specifically, the machine was an obvious aggregation of prior art which produced no new or synergistic result. It failed materially to promote the progress of science and the useful arts.
The district court entered 60 findings and 20 conclusions indicating its view that the ’315 patent is invalid under 35 U.S.C. § 102(b), 35 U.S.C. § 103, and 35 U.S.C. § 112.
On May 24, 1983 the district court entered judgment declaring the ’315 patent invalid. The judgment is silent respecting infringement, though the district court had stated from the bench at end of trial:
Well, if the ’315 patent is valid, I think the proof is clear that it has been infringed and it is pretty clear that it was done with knowledge, conscious knowledge to the point of willful infringement. 1
Issues
I. Whether the district court erred in finding the inventions set forth in claims 1, 2, and 4 anticipated by U.S. Patent 3,763,-770 (’770) under 35 U.S.C. § 102(b).
II. Whether the district court erred in concluding that the inventions set forth in claims 1, 2, and 4 would have been obvious under 35 U.S.C. § 103.
III. Whether the district court erred in concluding that the ’315 patent specification was non-enabling under 35 U.S.C. § 112.
IV. Whether this court on remand should order entry of a judgment that claims 1, 2, and 4 were infringed by Am-hoist.
OPINION
Of the district court’s 60 findings, 57 were those submitted by Amhoist before trial. The source of findings does not render the “clearly erroneous” standard of Fed.R.Civ.P. 52(a) any less applicable or binding.
Rosemount, Inc. v. Beckman Instruments, Inc.,
Having written them, Amhoist argues strenuously for retention of the findings behind the shield of the “clearly erroneous” rule, and repeatedly reminds us of our duty to review the findings favorably and of the burden resting on the appellant. *1458 However salutary, the rules governing review do not envision an appellate court shirking its duty to reverse an appealed judgment that is clearly based on legal error and unsupported by evidence in the record.
We review judgments, not the rhetoric in opinions. Nonetheless, the language in an opinion, or in a set of findings and conclusions, may indicate that numerous harmful errors of law produced an erroneous conclusion, and that the decisional approach of the district court led to a judgment not supported in law by the facts of record. That happened here.
I. Anticipation
Anticipation is a factual determination, reviewable under the “clearly erroneous” standard.
Carman Industries Inc. v. Wahl and Vibra Screw Inc.,
Anticipation requires the presence in a single prior art reference disclosure of each and every element of the claimed invention, arranged as in the claim.
Connell v. Sears, Roebuck & Co.,
Lindemann contends the district court’s finding on anticipation is clearly erroneous and we agree.
The finding of anticipation rested on a series of mistakes. The two gags of the ’770 patent do not correspond to “said sidewall being divided into two portions of different lengths”. The gags are beyond the end of the wall and constitute no part of a feed channel sidewall as claimed. The court found the ’770 patent’s magazine corresponded to the claimed “open feed channel having two opposing walls”, but the “movable” wall of the magazine is movable only to adjust the magazine’s width and not, as the claim requires, to crush scrap. Moreover, the findings that the magazine is the feed channel and that the gags are parts of a sidewall of the channel contradict each other. Nor does the shear anvil of the ’770 patent, as the court stated, correspond to the “opposite sidewall” of the claim. Nor do the cylinder assemblies of the ’770 patent move one sidewall of a feed channel toward the other as the claims require. Nor are the ’770 patent’s cylinder and gag (equated by the court to the claimed auxiliary ram) located “just upstream of said mouth”. They are within the shear area and are thus downstream from where a mouth narrower than the feed channel would be if the ’770 patent disclosed such a mouth, which it does not. Similarly, the other cylinder and gag of the ’770 patent do not form a “longer portion of said movable sidewall”. Nor can the channel that receives rod cuttings after shearing be equated, as did the district court, with the shear mouth claimed. 2
The ’770 patent discloses an entirely different device, composed of parts distinct from those of the claimed invention, and operating in a different way to process different material differently. Thus there is presented here no possible question of
*1459
anticipation by equivalents.
See Tate Engineering, Inc. v. United States,
On the unchallenged evidence of record, we are left with a “definite and firm conviction” that the district court’s finding of anticipation was mistaken and therefore clearly erroneous. That part of its judgment relating to invalidity under 35 U.S.C. § 102(b) must therefore be reversed.
II. Obviousness
A. Presumption of Validity
Guided by remarks found in then applicable court opinions, the district court: (1) viewed the statutory presumption of validity, 35 U.S.C. § 282, as “vanished” or “severely weakened” when Amhoist introduced prior art not cited by the examiner; (2) reduced the required burden of proof, in light of that introduction, to a “mere preponderance” 3 ; and (3) implicitly required Lindemann to prove that the uncited art had been considered by the PTO.
(1) Courts are not, of course, at liberty to repeal a statute, or to legislate conditions diminishing its effect. Hence the statutory presumption cannot “vanish” or be “weakened” and the statutorily assigned burden of proof cannot be shifted.
Stratoflex Inc. v. Aeroquip Corp.,
(2) The burden upon the challenger of validity under 35 U.S.C. § 282 is to introduce evidence of facts establishing invalidity (thus overcoming the presumption).
American Hoist & Derrick Company v. Sowa & Sons, Inc., 725
F.2d 1350 (Fed.Cir.1984). That evidence, if it is to carry the day, must be clear and convincing.
Radio Corp. v. Radio Laboratories,
(3) Similarly, the parties have devoted much unnecessary argument to the question of whether Lindemann is entitled to a presumption that the examiner had considered the uncited art because it is found in the classes and subclasses searched by the examiner (and because, as Lindemann says, the examiner had cited that art in examining an earlier application). Authorities are cited on both sides. 4
Because the touchstone is whether the uncited art is sufficiently more relevant than that cited to serve as evidence of obviousness, argument respecting
*1460
a presumption based on the uncited art’s classification is pointless. The argument here, moreover, appears to have led to the erroneous view that Lindemann bore the burden of proving that the uncited art had been considered. To the extent that the examiner’s consideration of uncited art is material, the burden is on the challenger to show that “that prior art had
not
been considered.”
Richdel Inc. v. Sunspool Corp.,
B. Scope and Content of the Prior Art 6
“The scope of the prior art has been defined as that ‘reasonably pertinent to the particular problem with which the inventor was involved’.”
Stratoflex, Inc. v. Aeroquip Corp.,
Lindemann attempts too much in arguing that waste compactors are non-analogous. Though the problems differ, both parties manufacture both products and both are exhibited at the same trade shows. Art that is analogous may or may not render a claimed invention obvious. As indicated below, it does not do so here.
The content of the prior art discussed in Amhoist’s brief is that disclosed in the ’770 patent (discussed above) and in British Patent No. 1,230,014 (’014). 7
The ’014 patent discloses a compactor for particulate waste, e.g., garbage. The loose waste is pressed into the wide mouth of a funnel by a circular plate. The smaller end of the funnel communicates with a container to receive the compacted waste. A small finger-like ram is coaxial with, and normally moves with, the plate. When the material fills the funnel so tightly that the plate can add no more, the separately operable small ram can be advanced ahead of the main ram and into the waste material. The small ram has a diameter smaller than that of the funnel outlet. When the small ram has pressed a core of waste material through the funnel outlet, the remaining waste material is loosened and additional waste material may then be pressed into the funnel by the plate and ram working together.
In a conclusion of law, the district court stated that it had considered the facts in light of the inquiries mandated by
Graham v. John Deere & Co.,
C. Commercial Success.
The district court improperly discounted the weight due the evidence of commercial success because that success occurred abroad. A showing of commercial success of a claimed invention, wherever such success occurs, is relevant in resolving the issue of non-obviousness.
Weather Engineering Corp. v. United States,
The evidence at trial showed that the claimed invention accounted for 30% of Lindemann’s total sales worldwide for a total sales price of over $20,000,000 (30 machines at approximately $667,000 each). The district court correctly stated that commercial success cannot by
itself
establish nonobviousness. However, having concluded that the claimed invention would have been obvious from the prior art, the court looked only to see whether the showing of commercial success was so overwhelming as to overcome that conclusion. That was error. All evidence must be considered
before
a conclusion on obviousness is reached.
Stratoflex, Inc. v. Aeroquip Corp.,
D. Unexpected Results
The district court ignored the unexpected or surprising results achieved by the claimed invention. Though no requirement for such results is present in the statute, 35 U.S.C. § 103,
Chore-Time Equipment, Inc. v. Cumberland Corp.,
Neither the district court nor Amhoist’s brief on appeal has a word to say about the unexpected results asserted by Lindemann, namely, the rapid crushing of rigidly massive scrap in a moderate sized scrap shear without pretreatment. That the claimed inventions achieve those results is unchallenged. Neither the district court nor Am-hoist suggest anything in any piece of prior art, or in the prior art as a whole, that would lead one skilled in the art to expect achievement of such results.
The record is clear that no earlier shears of any size, and no prior art device of any type could economically process rigidly massive scrap without pretreatment. Unchallenged testimony of experts was characterized by surprise and amazement that the claimed invention was able to accomplish that feat. That it could do so in minutes, and with a moderate sized structure, were further sources of surprise. That those skilled in the art had previously believed pretreatment of rigidly massive scrap was required was also uncontradicted.
*1462
It is further clear from the uncontradicted evidence that the claimed invention achieved new and unexpected results nowhere suggested in the prior art, and that the district court overlooked the effect of that achievement in reaching its determination of obviousness. In so doing, the district court erroneously focussed its inquiry “solely on the product created, rather than on the obviousness or nonobviousness of its creation”.
General Motors Corp. v. U.S. International Trade Commission,
The district court viewed the claimed invention as merely the “aggregation” of two different sized rams. Finding the first in one place in the prior art and the second in another place, the district court entered this conclusion:
Plaintiff simply put the two features in the same machine and connected them as was necessary depending on whether the scrap was small or large. It used a known connection idea. The ’315 machine possessed one known feature to operate in a known way to produce a known result to deal with the first scrap situation and another known feature operating in a known manner to produce a known result to deal with the second. Clearly, this was an obvious solution using already appreciated or obvious features to solve the problem of how to develop a machine that could handle both types of scrap most economically.
The ’315 patent specifically stated that it disclosed and claimed a combination of features previously used in two separate devices. That fact alone is not fatal to patentability. The claimed invention must be considered as a whole, and the question is whether there is something in the prior art as a whole to suggest the desirability, and thus the obviousness, of making the combination.
In re Imperato,
Nothing in the references alone or together suggests the claimed invention as a solution to the problem of crushing rigidly massive scrap. There was nothing whatever of record, therefore, to support the district court’s statement that the claimed machine possessed “another known procedure operating in a known manner to produce a known result” or its conclusion that Lindemann “knew ... that a small sidewall ram could most economically process large scrap”.
The ’014 patent deals only with soft, easily compactible, particulate material. Though that patent discloses a two-ram structure and the principle that loose material when too tightly compacted can be loosened by injection of a thin ram into the material, the claims here are not drawn to the mere concept of two differently sized rams, or to the known principles governing the effects of large and small rams (or to the propagation of force principle discussed at trial). That the claimed invention may employ known principles does not in itself establish that the invention would have been obvious. Most inventions do. Nothing in the ’014 patent would suggest that rigidly massive scrap could be rapidly and economically crushed and sheared without pretreatment.
The ’770 patent, as above indicated, deals only with holding brittle material within a shear by compression. Nothing in the ’770 patent suggests that making the crushing wall of a metal scrap shear in two independently operable parts, with a smaller part adjacent the mouth of the shears, would enable the crushing of massively rigid scrap without pretreatment.
Nothing, moreover, in the ’014 or ’770 patents adds anything to the prior art considered by the examiner. As above indicated, the ’315 specification itself recognized the separate presence in the prior art of feed channels with one solid moveable crushing wall and of feed channels with a small ram in one of two fixed sidewalls. The examiner cited as “of interest” the Pioch patent which, like the ’014 patent, *1463 disclosed two independently operable pushers in a waste compactor.
Applying the standard of Rule 52(a), Fed.R.Civ.P., we are persuaded that the findings underlying the district court’s conclusion of obviousness are clearly erroneous. Further, that conclusion resulted from errors of law in interpreting the claims and in consideration and application of the prior art. That part of the appealed judgment relating to 35 U.S.C. § 103 must therefore be reversed.
III. Enablement
The district court concluded that the ’315 patent was non-enabling because it did not disclose a hydraulic and electrical system for controlling the operation of the rams.
Enablement is a legal issue.
Raytheon Co. v. Roper Corp.,
The unchallenged evidence of record establishes that hydraulic and electrical systems for metal scrap shears were well known to those skilled in the art, and that the selection and connection of the elements of such systems was simply a matter of plumbing.
Amhoist points to testimony relating to 800 man hours it expended in developing its split ram shear. It also points to the dismantling of the accused machines by its two customers, whereby the rams are operated together as one sidewall and asserts that the split ram structure of the claimed invention has thus been abandoned by those customers. 8 There is no evidence indicating that the dismantling was due to difficulty in designing a suitable hydraulic-electric control system.
It is clear that no undue experimentation was required in practicing the claimed invention.
W.L. Gore & Assoc. Inc. v. Garlock, Inc.,
The district court erred in its conclusion that the ’315 patent specification is non-enabling and that part of the appealed judgment relating to 35 U.S.C. § 112 must be reversed.
IV. Infringement
Relying on the statement made by the district court at close of trial, and on the uncontested evidence clearly establishing Amhoist’s knowledge of the ’315 patent and its conscious decision to disregard it, Lindemann requests this court to “affirm” the district court’s “decision” on infringement. Lindemann’s difficulty is that judgments, not statements, are appealed and the district court made no finding and entered no judgment on infringement.
A district court should decide validity and infringement and should enter a judgment on both issues when both are raised in the same proceeding.
Stratoflex, Inc. v. Aeroquip Corp.,
The case must be remanded for the district court to make a finding on infringement. Whether the present record supports a finding corresponding with the court’s end-of-trial statement, and whether further trial on the issue is therefore unnecessary, is for the district court to determine in the first instance. Upon any finding of infringement and entry of judgment on that finding, the district court will doubtless consider issuance of an injunction against further infringement and an accounting.
DECISION
The district court’s judgment is reversed and the ease is remanded for further proceedings consistent with this opinion.
REVERSED and REMANDED.
Notes
. The district court stated at the same time, "But I am not certain in my own mind at this point whether or not these gentlemen on the ’315 patent invented anything”. The statement reflects a misconception of the role of the courts under 35 U.S.C. § 103. The question mandated by statute is not “invention”; it is patentability. See Rich, Escaping the Tyranny of Words — Is Evolution in Legal Thinking Impossible?, 60 JPOS 71, May-June/APLA Bull. 237 (1978).
Moreover, the court’s role in relation to patentability does not require it to conclude whether something was or was not "invented”, or whether the court subjectively considers the invention “worthy” of patent protection. The court’s role is actually more simple. Under the statute, it is to determine whether the patent’s challenger carried the burden of establishing invalidity. 35 U.S.C. § 282.
See Environmental Designs, Ltd. v. Union Oil Co. of Cal.,
. Amhoist says Lindemann's Australian counsel "conceded” that the '770 patent cited by the Australian examiner was a "paper anticipation”. The assertion is meaningless. First, the '315 patent’s counterpart issued in Australia. Second, the language and laws of other countries differ substantially from those in the United States.
. The district court in a conclusion of law also stated that "under any burden of persuasion the ’315 patent is invalid because of obviousness”. As indicated in the text, we disagree.
. The district court indicated the view that "the 'Field of Search’ is exactly what it purports to be and nothing more, that ‘References Cited’ are the patents found within the field which were actually considered by the examiner and listed because he found them to be most relevant”. That view is flawed. The examiner could not determine which patents are "most relevant” without considering a number which are less relevant.
. Though the courts will give due respect to the examiner’s evaluation of prior art, they are not of course bound thereby. Patentees desiring the benefit of the examiner's evaluation of originally uncited art have available the reexamination procedures under 35 U.S.C. §§ 301-307. Those procedures were not employed in this case.
. The level of skill is not of record and is not discussed in the briefs.
. The district court additionally discussed the S-501 shear produced by Amhoist and incorporating a tapered feed channel with a single side ram about one foot from the shear mouth. Am-hoist correctly recognizes on appeal the absence of need to discuss the S-501 shear.
. The record does not reflect the rationale underlying a vigorously fought lawsuit and its accompanying expense in the light of two sales and both purchasers’ cessation of use of the invention.