This is an appeal and a cross-appeal from an action brought in the United States District Court by plaintiff1 for infringement of United States Letters Patent No. 2,832,848 by defendant. In a prior appeal2 this court reversed the entry of a summary judgment by the district court. After a supplemental complaint had been filed and answered, the case was tried and the district court entered a judgment holding the patent valid but not infringed. No issues here involved were present in the prior appeal. Plaintiff appeals from the district court's holding that the patent was not infringed. Defendant appeals from the district court's holding that the patent was valid in law.

Jurisdiction in the district court was found upon Sections 1338(a) and 1400(b) of Title 28 United States Code. This court has jurisdiction to review the judgment entered by the district court by virtue of Sections 1291 and 1294 of Title 28 United States Code.

The facts are the same for both appeals. The patented device is an electrical signal amplifier which is especially useful in instrumentation systems in which relatively weak signals representing physical phenomena such as temperature, pressure, stress, and vibration must be strengthened, i.e., amplified, for application to output equipment for measuring and recording the phenomena (Ex. 1, R. 409).3

In order to measure each of the desired phenomena, a number of small measuring devices known as transducers may be attached to various parts of the engine being tested. Common types of transducers include thermocouples (for the measurement of temperature) and stress gauges (for the measurement of stress and vibration). The transducers function to generate a small electrical impulse or signal which varies in a manner corresponding to the action of the phenomena being measured. Since the signals emitted by the transducers are so weak as to be unusable directly, an electrical signal amplifier is used to receive the signal from the transducer and to provide a magnified signal which is suitable for application to output equipment such as a recorder, meter, or a computing system.

The patented device here in question is directed to an electrical signal amplifier which has unique capabilities for the amplification of signals in an instrumentation system such as described above. Such an instrumentation presents several problems:

(1) The signals emitted by a transducer generally include a direct current4 component with a relatively constant value and an alternating current component which varies with time. Both signal components must be amplified to produce an output signal faithfully representing the phenomena being measured. Consequently, the ability to amplify direct and alternating signals components is an essential characteristic; the first of three characteristics of the patented amplifier.5

(2) Where signals are simultaneously emitted from a number of separate transducers and individually amplified, each of the individual amplifiers should provide an output signal which faithfully represents but one of the signals emitted by the transducers without interaction with the other of the signals.

Where interconnections exist for the passage of signals between the amplifiers, interaction occurs between the signals. (An interconnection between amplifiers may be made through a 'common ground.') The matter is further complicated by spurious signals, which may be generated within these interconnections, appearing in the output amplifier.

To the extent that spurious signals appear at the output of an amplifier, the output signal from the amplifier does not truly represent the phenomena being measured. To overcome this problem, the patented amplifier has an input circuit which is (conductively) isolated from the output circuit, i.e., no electrical connection is established for the conduction of current between the transducers which may be connected to the input circuit and whatever equipment may be connected to the output of the amplifier. This 'conductive isolation' is the second of three characteristics of the patented amplifier (R. 61, 103, 169-170).

(3) Where it is necessary to amplify electrical signals with a high degree of accuracy, the output signal from the amplifier must be a faithful replica of the input signal. To achieve the required degree of fidelity in the output signal, it is common to employ a 'negative feedback.' With negative feedback, a portion of the output signal is fed back into the input circuit in a direction which produces a partial cancellation of the input signal. This practically eliminates any distortion of the signal produced by the amplifier. But where a connection is made between an output circuit and an input circuit the result is that the connection defeats and militates against the isolation of the input circuit from the output circuit-- as described in problem (2) above. The third characteristic of the patented amplifier allows the negative feedback without loss of the other characteristics mentioned above.

Therefore, the patented amplifier for the first time provides the following simultaneously:

(1) Amplification of direct current signals;

(2) Isolation of the input circuit from the output circuit; and

(3) Negative feedback from the output to the input circuit.

The patent in question describes and claims an electrical signal amplifier having all three of the above described characteristics. The structure defined in the patent claims is divided into ten elements, although some of the claims are subcombination claims and do not include all ten. The patent (Ex. 1, R. 407) contains a total of eleven claims.6

Electrical signal amplifiers substantially identical to the amplifier described in the patent have been successfully manufactured and sold by the plaintiff since the early part of 1957. On July 3, 1957, after plaintiff had placed its amplifiers on sale, defendant purchased one of plaintiff's amplifiers for investigation by defendant's engineers (Ex. 11, R. 433; Ex. 12, R. 435). A second, improved, amplifier was subsequently purchased by defendant through a third party (Ex. 13, R. 437; Ex. 14, R. 439; Ex. 18, R. 373).

At the time of the first purchase, defendant (who then manufactured amplifiers of another type) had not achieved a finished working model having the desirable characteristics of plaintiff's amplifier. With respect to the characteristics of the patented amplifier in providing simultaneously for the amplification of direct current signals, isolation of the input circuit from the output circuit, and negative feeback to the input circuit, testimony and evidence indicate that both the patented and the accused amplifiers are alike. The description of the operation of the patented amplifier would apply equally well to a diagram of the accused amplifier published by defendant. It is admitted that each of the claims of the patent read on the infringing device.

The district court's judgment held plaintiff's patent to be valid in law but not infringed. The court recognized the characteristics of the patented amplifier but stated:

'There are present in the accused device elements which are not present in the patented device and which achieve an entirely different result. These are the presence of the output amplifier, the difference in filters in the feedback and capacitator capable of a high degree of AC common mode rejection.'

'Here admittedly the added elements not only constitute a different combination but achieve a different result. So we have both dissimilarity of means and dissimilarity of result.'

The following dissimilarities were found by the district court: (1) The patented amplifier uses a capacitor 66, the accused amplifier does not; (2) The patented amplifier uses an LC filter, the accused amplifier uses an RC filter; and (3) The accused amplifier uses an output amplifier which is not shown nor used in the patented amplifier.

We first consider defendant's cross-appeal as to the validity of the patent. Defendant's specification of errors relied upon reads as follows:

(1) The district court erred in holding:

'That the subject matter of patent in suit would not have been obvious at the time the invention was made to the person having ordinary skill in the art to which such subject matter pertains.' (Conclusion of Law II.)

(2) The district court erred in holding:

'That the patent in suit and each of the claims thereof is good and valid law.' (Conclusion of Law III.)

(3) The district court erred in not finding that the patent in suit was an aggregate of well-known old expedients.

(4) The district court erred in not finding that the patent in suit is invalid and void as each of the claims thereof fails to particularly point out and distinctly claim the subject matter thereof as required by Title 35 U.S.C. 112.

(5) The district court erred in adjudging:

'2. That the patent in suit and each of the claims thereof is good and valid in law.' (Para. 2, Judgment.)

(6) The district court erred in including in its judgment a holding that the patent in suit was valid in view of the holding of noninfringement.

Defendant's primary contention is 'that the alleged invention was readily available once one was willing to spend this kind of money for one amplifier.' (Op. Br. 5.) It argues that 'as the device therefore incorporates an alternating current amplifier it obviously makes pertinent all of the prior art relating to alternating current amplifiers.' (Op. Br. 6.) This is undoubtedly true. But the fact that the steam engine incorporates all of the prior art of making steam did not render it any less an invention. We believe plaintiff's device measures up to the standard of an invention, and the district court was correct in holding the patent to be good and valid in law.

A patent was issued to plaintiff by the Patent Office. A presumption of validity, which can only be overcome by clear and convincing proof, arises from the issuance of a patent. This presumption is based upon the expertness of the Patent Office.7

Defendant makes an exhaustive effort to combine prior art references to support the argument that the device here in question was actually readily available if one were willing to spend the money. But defendant presents few more pertinent references to prior art than were considered by the Patent Office.8 And the presumption of a patent's validity becomes all the stronger when the Patent Office has considered the most pertinent references before issuing the patent.9

We note that the district court found that the prior art contained no reference capable of providing the simultaneous characteristics found in the patented amplifier (Finding IX, R. 27); that the subject matter of the patent would not have been obvious, at the time the invention was made, to a person having ordinary skill in the art (Conclusion II, R. 27); and that the patent and each of its claims is good and valid in law (Conclusion III, R. 27). The presumption of a patent's validity, which arises from its issuance by the Patent Office, is strengthened by the finding of validity by the district court.10

In the face of these findings and presumptions, defendant urges that the provisions of the patent's simultaneous characteristics were clearly within the skill of one familar with the art as soon as the commercial need for an instrument with these characteristics arose. Yet an expert witness was familiar with the need for such an instrument as early as 1943, and testified that there was nothing available prior to plaintiff's invention which would fulfill this need. And defendant, who conducted an extensive business in the electronic amplifier field and had expended a quarter of a million dollars in research and development, did not achieve a finished working model until after it had purchased one of plaintiff's patented amplifiers.11 Nor had anyone else solved the problem despite the increased need for such an instrument for use in data processing systems in 1953.

Plaintiff became aware of the need for an amplifier having the characteristics of his patented amplifier in 1952. In 1956 he left his employment to go into business for himself. Plaintiff's business expanded from a one-man operation with a net worth of $4,000.00 to a business employing thirty-five persons and with annual sales of approximately $500,000.00. There thus exists indication of commercial success and: '* * * the commercial success of (a) device * * * though not conclusive, tends to prove the originality and utility of the product.' Hayes Spray Gun Co. v. E. C. Brown Co., 9 Cir., 1961, 291 F.2d 319 at 322.

In Ry-Lock v. Sears Roebuck & Co., 9 Cir., 1955, 227 F.2d 615, cert. denied 350 U.S. 987, 76 S. Ct. 474, 100 L. Ed. 854, this court said:

'This invention, made up by combination of elements, in a manner which was sufficiently new and novel to measure up to the accepted standards of invention, was not, in the language of Himes v. Chadwick, 9 Cir., 199 F.2d 100, 'a mere aggregation of a number of old parts.' Hence, a finding which * * * picks out one element in one prior patent and another element in another prior patent as a demonstration of anticipation, is manifestly insufficient to overcome the presumption arising from the issuance of the patent, a presumption reemphasized by the existing Act. 35 U.S.C.A. 282.' (Id., p. 618.)

As to the obviousness of an invention, when viewed with the wisdom of hindsight, this court has said:

'It is of no significance that 'viewed after the event, the means * * * adopted seem simple and such as should have been obvious to those who worked in the field. but this is not enough to negative invention.' Goodyear Tire & Rubber Co. v. Ray-O-Vac Co., 321 U.S. 275 (64 S. Ct. 593, 88 L. Ed. 721) * * *. 'Now that it has succeeded, it may seem very plain to anyone that he could have done it as well. This is often the case with inventions of the greatest merit.' Carnegie Steel Co. v. Cambria Iron Co., 185 U.S. 403, 446, (22 S. Ct. 698, 46 L. Ed. 715, 968). See Patterson-Ballagh Corp. v. Moss, supra (9 Cir., 201 F.2d 403).' National Sponge Cushion Co. v. Rubber Corp., 9 Cir., 1961, 286 F.2d 731, 735.

We cannot say that the district court's findings and conclusion that the patent here in question was valid was error nor without substantial evidence to support it. And defendant fails to present clear and convincing evidence to overcome the presumptions of validity. We therefore affirm the district court's holding that the patent was good and valid in law.12

We pass to the issue of infringement. Plaintiff sets forth seventeen points (Op. Br. 21-23). All of the errors specified go to the part of the district court's judgment which held that defendant had not infringed plaintiff's patent. We here summarize them as presenting the one question: Whether the district court committed any error in holding defendant not to have infringed plaintiff's patent.

It is well settled that the claims of the patent are the means by which an infringement is to be tested. At the least, this should be one of the first tests to be applied. In Walker on Patents13 it is stated:

'The patented invention is not everything disclosed within the 'four corners' of the patent; for many ideas or concepts are disclosed and suggested by the specifications and drawings which are not protected by the patent. Strictly speaking, infringement of a patent is an erroneous phrase; what is infringed are the claims of the patent which 'measure the invention' and define precisely what the invention is, and the limits beyond which one cannot pass without infringing; therefore it is to the claims of the patent to which one must look to determine whether there is an infringement (citations).'

More recently, the Supreme Court restated this test when it said that 'the claims * * * in the patent are the sole measure of the grant * * *.' Aro Manufacturing Co., Inc. v. Convertible Top Replacement Co., Inc., 1961, 365 U.S. 336 at 337, 81 S. Ct. 599, 601, 5 L. Ed. 2d 592.

This being the test, we believe the district court was clearly erroneous when it precluded the application of this test during the trial. The district court made no application of the claims to the accused amplifier; no comparison between the claims and the accused amplifier was made. And defendant's counsel conceded that each of the claims of the patent read upon the accused amplifier (R. 325-326) and argued that the major issue of the case was validity. The district court declined to permit plaintiff's expert witness to apply the claim language to a diagram of the accused device's structure (R. 182), although the witness' previous testimony had demonstrated the presence of the ten claimed elements. This we believe to have been clear error.

Instead of applying the above test, the district court held that there had not been an infringement because of three dissimilarities between the plaintiff's and defendant's amplifiers; and because a different result was achieved.

To repeat what is mentioned before, the district court found the following dissimilarities: (1) The patented amplifier uses a capacitor 66, the accused amplifier does not; (2) The patented amplifier uses an LC filter, the accused amplifier uses an RC filter; and (3) The accused amplifier uses an output amplifier which is not shown nor used in the patented amplifier. How significant are they?

(1) The capacitor 66: On the schematic patent drawing (Ex. 1, R. 408; Ex. 6, R. 421) there is shown a capacitor 66 connected between one side of the input circuit and the ground. The text of the patent itself suggests that the capacitor may be eliminated (Ex. 1, R. 411, at Col. 5, lines 15 through 25). The uncontroverted testimony of an expert witness indicates that if the capacitor were omitted the patented amplifier would have the same capability of discriminating against AC common mode signals as is claimed for the accused amplifier (R. 204). Defendant's expert witness acknowledged that the patent mentions that the capacitor could be omitted if desired (R. 263).

(2) LC and RC filters: Each filter includes a capacitor but the difference is that the LC filter includes an inductance while the RC filter includes a resistor. Each has the function to smooth the output signal from the amplifier synchronous demodulator. LC and RC filters are both capable of doing the same job (R. 161). And these filters need have nothing to do with the characteristic of the amplifier's ability to isolate an input circuit from an output circuit for direct current components (R. 381-382).

(3) The output amplifier: The accused device includes an output amplifier which is not shown in the schematic arrangement of the patent. All witnesses agreed that an output ampliffier might be added to the patented amplifier if it was desired. It is primarily a power amplifying device. The three characteristics of the patented amplifier are achieved before this device is reached.

It seems clear to this court that the addition of these three items does not change the characteristics found in the plaintiff's amplifier. The defendant advertised its amplifiers as being 'superior * * * to other amplifiers currently available'14 (R. 424); that its amplifier 'has been designed to eliminate * * * shortcomings' (R. 425). At best, then, it appears that the accused amplifier is an improvement over the plaintiff's amplifier. And 'it is well established that an improver cannot appropriate the basic patent of another, and that the improver without a license is an infringer, and may be sued as such.' Aluminum Company of America v. Sperry Products, Inc., 6 Cir., 1960, 285 F.2d 911 at 923. The court, in the Alcoa case, went on to say, at page 924:

'If the infringing device performs the same function as the patentded device, it is immaterial that it also performs some other function. It is still none the less an equivalent of the patented device, (and an appropriation of the patented device,) and an appropriation of the patented invention. (Citations.)'As said by the trial judge, an accused device cannot escape infringement by merely adding features, if it otherwise adopts the basic features of the patent. (Citations.)'

'Outright and forthright duplication is a dull and very rare type of infringement. To prohibit no other would place the inventor at the mercy of verbalism and would be subordinating substance to form. * * *' Graver Tank & Mfg. Co., Inc. v. Linde Air Products Co., 1950, 339 U.S. 605 at 607, 70 S. Ct. 854 at 856.

The district court found that the 'different results' of the accused amplifier was that it achieved 'a very high degree of common mode rejection' because it does not use a capacitor 66. (Findings V, VII, R. 26-27.) This court has held that the addition of an element (e.g., the output amplifier in this case) does not avoid infringement. Hayes Spray Gun Co. v. E. C. Brown Co.,supra, at 326 of 291 F.2d. We likewise hold the subtraction of an unnecessary element (e.g., the capacitor 66) would not lead to a contrary conclusion. And defendant's own advertising (Ex. 7, R. 424, et seq.) evidences the fact that its amplifier performs the same functions (-- but better-- ) that the plaintiff's amplifier performs. 'Whether the two devices are identical in this respect (the design of an element) is not relevant for they are identical functionally.' Hayes Spray Gun Co. v. E. C. Brown Co., supra, at 325.

We hold the district court unduly narrowed the scope of plaintiff's patent and that it committed clear error in finding it not to be infringed.

Plaintiff seeks an award of appropriate costs and attorney's fees. But, as stated by this court in the Hayes Spray Gun case, at page 327, awarding attorney's fees in exceptional cases--

'* * * is a matter of discretion, and the court cannot be reversed except for abuse of discretion. * * * Hence even if the instant case can in some way be described as exceptional, the trial court cannot be reversed unless its refusal to award attorney's fees constituted an abuse of discretion. No such abuse has been shown.'

The district court's judgment with respect to its holding of noninfringement is reversed, the case remanded with direction that judgment be entered in favor of plaintiff, and that defendant be enjoined from further infringing activities, and for such further proceedings as may be necessary in view of this opinion; including such attorney's fees, if any, as the court below may allow.