These appeals arise out of patent infringement actions between the same parties, involving two patents for improvements of diaphragm type carburetors of different design and construction. They were argued together.

The District Court referred both actions to a Special Master who heard the evidence, considered them separately, wrote opinions and made findings of fact and conclusions of law in each case. These findings of fact and conclusions of law were approved and confirmed by the District Court.

In Appeal No. 15,439, the District Court held that Claims 1 and 2 of Phillips Patent No. 2,733,902 were invalid and not infringed by the accused carburetors of defendant. Plaintiff appealed only from the judgment declaring the patent invalid.

In Appeal No. 15,440, the District Court held that Claims 4 to 7, 9 to 12 and 14 to 17 of Phillips Patent No. 2,841,372 were valid and infringed. Defendant appealed.

Phillips, the inventor of both patented carburetors, assigned the patents to the plaintiff.

Diaphragm carburetors were developed to permit the operation of small two cycle combustion engines at extreme angles from vertical or inverted positions without leakage of the fuel. This could not be accomplished with the float type carburetor. The development extended over a period of many years. Phillips was not the first to design such a carburetor although he unquestionably contributed significant engineering advances in the field. The carburetors were used primarily for the operation of chain saws although they could be used to power lawn mowers and boats.

Carburetors of the general type disclosed in Claims 1 and 2 of the patent were not new.1 These claims define an improvement, namely, the combination of valve means (shown in the specification as a mechanical check valve) in the main fuel orifice of the carburetor to prevent or interrupt, when the engine is idling, the back flow of air from the mixing passage of the carburetor through the main fuel orifice into the idling chamber. This flow of air is called back-bleeding.

The problem of back-bleeding or reverse flowing of air does not arise when the engine is operating at high speed with the throttle valve open. It occurs when the throttle valve is closed and the engine is idling and supplied with fuel through an idling orifice which is separated from and located some distance away from the main fuel orifice. The air pressure at the main fuel orifice is greater than at the idling orifice and the air flows back from the mixing passage through the main fuel orifice into passage ways leading to the idling orifice. This mixes too much air with the idling fuel making the mixture so lean that the engine will stall. To stop or interrupt this reverse flow of air Phillips provides valve means. The patent drawing (Fig. 4) discloses a ball check valve located in the main fuel orifice.

When the engine is operating with the throttle open, fuel is sucked from the fuel chamber through the main fuel nozzle. This suction lifts the ball valve off its seat and permits the free flow of fuel through the main fuel nozzle into the mixing passage. When the throttle is closed and the engine is idling, there is a suction in the reverse direction through the main fuel nozzle into the fuel chamber. The ball check valve is sucked onto its seat thus preventing the back flow of air into the fuel chamber.

Phillips' application for patent had "rough going" in the Patent Office. In the first action, the Examiner cited the following references against it:

  Wirth et al    2,345,168  (1944)
  Wirth          2,445,097  (1948)
  Brunner        2,568,987  (1951)
  Sloane         2,569,782  (1951)
  Carlson et al  2,339,320  (1944)

The Examiner rejected all the claims of the patent as unpatentable over Wirth et al "which shows the diaphragm actuated fuel inlet valve for the chamber 11, the air chamber 75, and the main and idling fuel systems." The Examiner was of the view that the arrangement of Wirth for controlling the inlet valve is the full equivalent of Phillips' arrangement and it "teaches the use of a check valve in the idling system to prevent back flow there-through." He noted that Brunner and Sloane also teach the same.

Phillips filed an amendment. The Examiner again rejected all of his claims as unpatentable combinations citing as additional references:

  Rubatto (Italy)    311,708  (1933)
  Gistucci et al   2,156,115  (1939)
  Clesse (France)    985,460  (1951)

The Examiner thought that Gistucci showed a fuel regulator structure similar to Phillips' while Rubatto showed the main and idling fuel system together with a throttle check valve for closing the main discharge nozzle while the idling parts are discharging. Gistucci et al provided a spring that opened the fuel inlet valve that is biased to closed position by another spring. The Examiner stated it would not be invention to add the fuel systems of Rubatto to the regulator of Gistucci et al. He was of the opinion that the lever and the diaphragm arrangement of Gistucci et al was the full equivalent of Phillips. In view of the teachings of Wirth et al, the combination of Rubatto and Gistucci et al was considered obvious to one skilled in the art.

Clesse taught the use of a spring acting on a lever and diaphragm for controlling a fuel inlet valve biased to open position in a carburetor.

Phillips filed a second amendment and the Examiner for the third time rejected all of the claims.

A third amendment was filed which finally resulted in an allowance of six claims.

The Special Master considered references which were not cited by the Examiner, namely, Bracke 2,680,605 and the McCulloch 3-25 chain saw.

The application for the Bracke patent was filed on October 20, 1950 and it is undisputed that the patent is a valid reference. The Bracke application for patent was pending in the Patent Office at the same time of Phillips' application which resulted in '902. No interference proceeding was instituted. The Patent Examiner apparently did not know of the Bracke patent as he most certainly would have cited it as a reference.

Bracke is similar in many respects to Phillips '902 as defined in Claims 1 and 2. Bracke asserts that his carburetor operates in all positions and that one of its objects was to provide new and improved means for preventing back-bleeding. The Special Master found that Bracke has the same general arrangement of parts as specified in these claims and employs a mechanical check valve to prevent back-bleeding which is a full equivalent of the check valve used by Phillips in his main fuel orifice. He compared the disclosure in Bracke with Claim 1 of Phillips and with one exception found that they both provided for substantially the same elements. The exception was that in Phillips the fuel inlet valve operates to close the supply duct against pressure of the fuel while in Bracke the fuel valve operates in the direction of fuel to close the duct. He stated that both types of valves were old, well known in the prior art and are considered as equivalents. Mr. Phillips, in his testimony, admitted that the use of a needle valve with the point of the needle opposed to the delivery of fuel is conventional. Valves moving against pressure to close an inlet were shown in Armstrong Patent 2,724,584, Conover 2,713,854, Ensign 2,653,228 and Phillips earlier Patent 2,387,676. Claim 2 of '902 provided for a lever pivotally supported in the chamber and in constant engagement with the diaphragm. The use of levers of different types to transmit force is old. A pivoted lever was shown in Armstrong, Ensign above-mentioned and Ensign Reissue 22,151. None of these references were cited by the Patent Office.

Bracke used a spring biased stem connected with the valve which provided the force of a lever although it was not pivoted. Nor do we think that patentable invention was shown by Phillips locating his inlet valve duct at the side of the structure instead of in the center of the diaphragm as provided by Bracke.

The Special Master was of the view that Bracke anticipated Phillips. In our judgment, Phillips was obvious in view of Bracke to a person skilled in the art. There is no invention.

Tillotson claims that the Special Master and District Court erred in holding that the check valve to prevent back-bleeding was the crux of Phillips' invention. We find no error in this conclusion. In our judgment, the device to prevent back-bleeding was the principal part of the invention without which it would never have been successful.

Tillotson claims that Bracke was a paper patent and that carburetors embodying its features were never manufactured and sold. This contention is irrelevant on the question whether its disclosures constituted anticipation. Coats Loaders & Stackers, Inc. v. Henderson, 233 F.2d 915 (CA6, 1956). Such a patent does not indicate lack of invention. Edward Valves, Inc. v. Cameron Iron Works, Inc., 286 F.2d 933 (CA5, 1961). Failure to use a patent does not affect its validity. Special Equipment Co. v. Coe, 324 U.S. 370, 65 S. Ct. 741, 89 L. Ed. 1006 (1945).

70,777 of McCulloch chain saws, equipped with his carburetors, were sold prior to May 1, 1952. It was conceded that the McCulloch carburetor was prior art.

The Special Master compared Phillips' construction with McCulloch and found that many similarities existed in the elements of the two carburetors. Tillotson contended that the McCulloch check valve was introduced to prevent dribbling of fuel, whereas Phillips was to prevent back-bleeding of air.

Demonstrations were conducted in Court which revealed that the McCulloch check valve did in fact interrupt the flow of air through the main fuel orifice from the mixing passage into the idling passage when the idling orifice was delivering fuel into the mixing passage.

It was not material what purpose motivated McCulloch in using the check valve so long as it operated as it did to prevent back-bleeding.

McCulloch had a hook instead of a lever which engaged the fuel control valve. Bracke used a stem. We have shown, in considering Bracke, that the use of a lever was well known in the art.

McCulloch had no spring engaging the hook in a direction to close the fuel valve as provided in Phillips. It certainly does not constitute invention to install a spring which was also known in the art and is shown in Ensign, Armstrong and Bracke patents.

The Special Master found that in the McCulloch carburetor the elements were so arranged that the ball was held in its seat by forces exerted by the resilience of the diaphragm calibrated with relation to the length of the hook and the hydrostatic pressure of the fuel on the diaphragm. The spring used by Phillips was not necessary in McCulloch.

With the prior art cited by the Patent Examiner against the Phillips application for patent, it barely passed muster in the Patent Office.

If the Examiner had known of Bracke and McCulloch, it is doubtful that the Patent would have been issued.

The presumption of validity which attends the issuance of a patent is weakened where relevant prior art was not considered by the Patent Office. Kennatrack Corp. v. Stanley Works, 314 F.2d 164 (CA7, 1963); Aluminum Company of America v. Sperry Products, Inc., 285 F.2d 911 (CA6, 1960); Gibson-Stewart Co. v. Wm. Bros. Boiler & Mfg. Co., 264 F.2d 776 (CA6, 1959).

We see no invention in Phillips '902 over the disclosures in McCulloch.

We find no basis for overturning the findings of fact adopted by the Special Master, which were approved and confirmed by the District Court.

The accused carburetors did not use a mechanical valve in the main fuel orifice to interrupt the flow of air, but provide a capillary seal that was covered by Phillips Patent '372 that we hold is valid and infringed in Appeal No. 15,440.

Textron claims that its capillary fuel carburetors did not infringe Phillips '902 patent. Tillotson, the owner of Phillips '902 patent contends that there was infringement.

Claim 2 of Phillips '902 patent provided in part:

"* * * Valve means arranged to interrupt air flow through the main orifice from the mixing passage into said channel, when said secondary orifices are delivering fuel into the mixing passage."

The specification and drawings of '902 provided mechanical means, namely, a ball check valve in the main fuel orifice. They did not disclose a capillary fuel plug or valve.

Claim 2 must be read in the light of the specification and drawings. It may not be given a construction broader than the teachings of the patent. Cincinnati Milling Machine Co. v. Turchan, 208 F.2d 222 (CA6, 1953); Blanc v. Curtis, 119 F.2d 395 (CA6, 1941). In view of the crowded art, the patent must be limited to the precise structure disclosed and claimed. Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1, 67 S. Ct. 6, 91 L. Ed. 3 (1946); Kennatrack Corp. v. Stanley Works, supra. Adopting this construction, valve means in Claim 2 included only mechanical valves.

The capillary seal concept is not an equivalent of the valve means provided for in Phillips '902 patent, but is a new element. The accused carburetors did not infringe '902 patent.

Tillotson marked all of the carburetors which it manufactured and sold "Patent No. 2,733,902," whether the valve means was mechanical or capillary.

The District Court held that there was a mismarking insofar as the capillary valve carburetors were concerned, but that it was unintentional. An injunction was issued restraining Tillotson from so marking its capillary valve carburetors in the future. We think this was proper.

This patent is related to a diaphragm carburetor adaptable for internal combustion engines of the two-cycle type used to furnish power for chain saws, lawnmowers, small boats and other similar contrivances where the engine is sometimes tilted to extreme angles. It will operate in all positions and at all speeds.

The basic concept of this carburetor was disclosed in '902 where a ball check valve or other mechanical means was provided to eliminate back-bleeding of air from the mixing passage into the fuel chamber when the engine idled or was running slowly. This so-called back-bleeding causes the engine to run lean or to stall.

'372 introduced for the first time a significant improvement over '902 by rearranging and reconstructing the parts of the carburetor so that capillary action and surface tension of the fuel in the passages were utilized to establish an effective liquid seal which operated as a check valve and prevented the back-bleeding of air from the mixing passage into the fuel chamber. This eliminated the necessity of a ball check valve.

Appellant's statement as to the phenomena involved appears in part in footnote

.

The capillary seal concept had many advantages over the ball type check valve. There was no wear or need for replacement. There was no difficulty caused by dirt getting under the ball and holding it from its seat. The need for a conventional nozzle or outlet extending into the venturi, which was required with a ball type valve, was eliminated.

This was a new carburetor. There was nothing like it in the prior art. The invention was accompanied by commercial success. A total of 1,899,352 of the capillary seal type carburetors had been sold up to the time the present patent litigation was commenced.

The claim of appellant was not that '372 carburetor lacked invention, novelty or utility. Rather, it contended that "the essence of the invention is recited in a vague and confusing passage setting forth a distance in terms of a force in a manner which is unintelligible without resort to explanations not embodied in the specification and beyond the knowledge of those skilled in the art at the time the application was filed," that the claims call for a certain desirable result without recitation of sufficient structure to produce the result; that the diaphragm cannot be located except by independent experiments and that the distance factor stated in the claims was irrelevant.

The alleged vague and confusing passage of which appellant complains appears in italics in Claim 6 of the patent which is a representative claim and is set forth at length in footnote3

Appellant relies on 35 U.S.C. § 112 (1952).4

There is nothing vague or confusing about this claim to a person skilled in the carburetor art. It must be interpreted in the light of the specification and drawings. The test is whether the disclosures make the invention clear to a mechanic skilled in the art so that he could construct the carburetor and use it. Schriber-Schroth Co. v. Cleveland Trust Co., 311 U.S. 211, 61 S. Ct. 235, 85 L. Ed. 132 (1940).

Whether the patent properly describes the invention so as to enable any person skilled in the art to make and use it must be determined by consideration of the specification, drawings, claims and the other evidence in the case.

The Special Master appointed by the Court heard the evidence. It consisted of oral testimony, written documents and demonstrations of physical models. Phillips, the inventor, who was a mechanical engineer, testified in behalf of plaintiff. Kenneth Charles Schneider, who was also an engineer, testified for the defendant. Mr. Schneider was Vice President of Brown Engine Products, Inc. This company made the accused carburetors and sold them to the defendant.

The Special Master, in his decision, had this to say with respect to defendant's evidence:

"Defendant has not offered any significant evidence that a person skilled in the carburetor art could not, from the data given in the patent, construct a carburetor having the seal strength greater than the suction required to aspirate fuel out of the idling orifice, nor that he could not construct a carburetor from this data having a seal (32) strength sufficient to support a column of water of approximately 1" in height.

"Defendant failed to adduce evidence or testimony that with the information formation given in the patent it would be impossible to make a carburetor wherein the capillary effect of the fuel would not maintain the fuel in the passage against a pressure differential equivalent to that pressure required to support a column of water about 1" in height."

In his findings of fact Nos. 39 to 54, the Special Master considered in great detail the specification, drawings and the evidence. He found that the design, dimensions and proportions of the fuel passages and the location of the diaphragm with reference to the fuel delivery orifice are proportioned so that a capillary plug of liquid is rendered available as a seal to prevent back-bleeding; that the patent illustrates in the drawings and the specification describes a construction whereby the seal is obtained; that the patent teaches the proportioning of the constructional elements and the positioning of the fuel discharge system and that the patent adequately teaches one skilled in the carburetor art how to make a carburetor embodying the features of the invention.

These findings of fact of the Special Master, adopted by the District Judge, are entitled to great weight and may not be disturbed unless clearly wrong. Patrol Valve Co. v. Robertshaw-Fulton Controls Co., 210 F.2d 146 (CA 6, 1954); Research Products Co., et al. v. Tretolite Co., 106 F.2d 530 (CA9, 1939).

The patent was presumed valid. The burden of proof to establish invalidity was on the party asserting it. 35 U.S.C. § 282 (1952)

The findings of fact of the Special Master, confirmed by the District Judge, are not clearly wrong. We find no basis for holding that the invention was inadequately or improperly described. Aluminum Co. of America v. Sperry Products, Inc., 285 F.2d 911, 915 (CA6, 1960).

We think the claims define a structural limitation and not merely a result. The structural limitation was ascertainable by manometer measurements, which are familiar to a person skilled in the art.

Nor do we believe that the location of the diaphragm in the carburetor requires experimentation of the type condemned by the patent law. A specific example of the carburetor was given in the specification. The fact that some testing may be required in calibrating carburetors for different engines does not render the claims indefinite where adequate guideposts were furnished as here. Minerals Separation Limited v. Hyde, 242 U.S. 261, 37 S. Ct. 82, 61 L. Ed. 286 (1916); Toledo Rex Spray Co. v. California Spray Chemical Co., 268 F. 201 (CA 6, 1920). It was not claimed by the appellee that the capillary seal carburetor will work on all engines. It has worked on engines of the type described in the patent.

It is further contended that the distance factor is irrelevant in achieving the object of the invention. The patent specifications provide otherwise. Plaintiff's expert, Phillips, in his testimony established relevancy. Mr. Phillips testified that the capillary seal may form at any one of three different places in the carburetor, the uppermost being the main fuel orifice. This was the starting point of the measurement. The other places were in the metering passage. The Special Master determined this factual issue adversely to appellant. We see no reason to disturb the finding.

Claims 4 to 7 of the patent are valid. Claims 9 to 12 and 14 to 17 were dependent upon the validity of Claims 4 to 7 and are, therefore, valid.

The findings of the Special Master with respect to infringement are not seriously questioned by appellant. The accused carburetors were of the capillary seal type embodied in Phillips' patent '372. In its brief filed in Appeal No. 15,439, Textron stated that it "takes issue with that judgment (in Appeal No. 15,440) only on the ground that the capillary seal claims of the '372 patent are invalid under 35 U.S.C. § 112." id. p. 29.

The judgment of the District Court is affirmed in each appeal.